United States
Environmental Protection
Agency
&EPA
Environmental Monitoring
and Support Laboratory
P.O Box 15027
Las Vegas NV 89114
EPA-600/9-78-034
October 1978
Research and Development
Automated In Situ Water
Quality Sensor Workshop —
February 14-16, 1978
Sponsored by
the United States
Environmental Protection
Agency, the National Oceanic
and Atmospheric
Administration, and the
Interagency Working Group
on Satellite Data
Collection Systems
-------
This document is available to the public through the National Technical Information
Service, Springfield, Virginia 22161
-------
EPA-600/9-78-034
October 1978
AUTOMATED IN SITU WATER QUALITY SENSOR WORKSHOP
February 14-16, 1978
cosponsored by
U.S. Environmental Protection Agency
National Oceanic and Atmospheric Administration, and
Interagency Working Group on Satellite Data Collection Systems
prepared by
Donald T. Wruble, EPA
John D. Koutsandreas, EPA
Barbara Pijanowski, NOAA
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL MONITORING AND SUPPORT LABORATORY
LAS VEGAS, NEVADA 89114
-------
DISCLAIMER
This report has been reviewed by the Environmental Monitoring
and Support Laboratory-Las Vegas, U.S. Environmental Protection
Agency, and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommenda-
tion for use.
ii
-------
FOREWORD
Protection of the environment requires effective regulatory
actions which are based on sound technical and scientific infor-
mation. This information must include the quantitative descrip-
tion and linking of pollutant sources, transport mechanisms,
interactions, and resulting effects on man and his environment.
Because of the complexities involved, assessment of specific
pollutants in the environment requires a total systems approach
which transcends the media of air, water, and land. The Environ-
mental Monitoring and Support Laboratory-Las Vegas contributes to
the formation and enhancement of a sound monitoring data base for
exposure assessment through programs designed to:
• develop and optimize systems and strategies
for monitoring pollutants and their impact on
the environment
• demonstrate new monitoring systems and technol-
ogies by applying them to fulfill special moni-
toring needs of the Agency's operating programs
This report addresses water quality monitoring. It presents
an overview of automated water quality sensor applications, and
sensor research and development activities and needs. The mate-
rial presented is a compilation of prepared presentations and
discussions at a U.S. Federal agency workshop. The workshop was
held to provide information exchange among key representatives
from these agencies on their respective automated water quality
sensor programs, and to define potential areas of program coordi-
nation and co-planning in sensor research and development efforts
by the Federal agency community. This information can be used not
only by the Federal agencies for guidance in program planning, but
by states and local governments, universities and the private
sector as a similar guide in directing their research programs.
The material should also be useful to those interested in a gener-
al overview of Federal sector activities in water quality measure-
ments. For further information, contact the Monitoring Operations
Division.
George B . Morgan
Director
Environmental Monitoring and Support Laboratory
Las Vegas
iii
-------
ACKNOWLEDGEMENTS
The interest and dedication of all workshop participants
should be recognized by the various agencies represented. Its
success is a direct reflection of their outstanding efforts as
individuals and a group.
Particular recognition goes to Ms. Barbara Pijanowski of
NOAA, Mr. John Koutsandreas of EPA and Dr. Richard Paulson of the
USGS and Chairman (during the workshop planning phase) of the
Interagency Working Group on Satellite Data Collection Systems.
Their support and initiative were the key elements in designing
and conducting the workshop.
Special thanks are offered to Mr. Kenneth Birch of the Canada
Centre for Inland Waters for adding a special workshop highlight
by presenting a discussion of Canadian programs.
Gratitude is also expressed to the university participants
for their excellent technological support. These were:
Dr. Roger Bates
Dr. Waiter Blaedel
Dr. Khalil Mancy
Dr. F. H. Middleton
Dr. Richard Newton
Dr. Charles Whitehurst
University of Florida
University of Wisconsin
University of Michigan
University of Rhode Island
Texas A&M University
Louisiana State University
-------
TABLE OF CONTENTS
Foreword iii
Acknowledgements iv
Introduction 1
Summary 3
Conclusions and Recommendations 7
Open ing Remarks 12
Welcome - George B. Morgan 13
Keynote Address - Richard M. Dowd, Ph.D 16
Introductory Remarks - Richard W. Paulson Ph.D 21
Federal Agency Programs 23
U.S. Environmental Protection Agency 24
National Oceanic and Atmospheric Administration.... 30
U.S. Department of Energy 43
U.S. Army Corps of Engineers 47
National Aeronautics and Space Administration 49
National Bureau of Standards 83
U.S. Navy, Office of Naval Research 87
U.S. Coast Guard 90
U.S. Geological Survey 94
U.S. Department of Agriculture 98
Federal, State and Local Water Quality
Monitoring Needs 101
An EPA Regional Viewpoint 102
Canadian Programs 119
University Research Programs 134
Electrochemical Methods for On-Site
Determination of Trace Metal Ions in Natural
Water Systems (University of Wisconsin) 135
v
-------
TABLE OF CONTENTS (Continued)
Page
A Lidar Polarimeter for Water Quality
Monitoring (Texas A & M University) 142
New Water Quality Sensor Technology
Employing Automated Wet Chemicals
(University of Michigan). 150
A State View of the Need for Research in
In Situ Automated Water Quality Sensors
(Louisiana State University) ig?
Electrochemical Techniques in Water Quality
Evaluation (University of Florida) 175
Water Quality Sensor Technology Using Acoustic
Techniques (University of Rhode Island) 180
Working Panel Reports.
Measurement (Sensor) Needs
Optical Sensor Technology
Electrochemical Sensor Technology 214
Electrophysical Sensor Technology 228
Automated Wet Chemical Sensor Technology 235
Needs and Technology Integration. 243
List of Attendees 249
Workshop Agenda 257
vi
-------
INTRODUCTION
A Federal agency workshop to discuss a common interagency
need for development of automated in situ water quality sensors
was held on February 14-16, 1978, at the U.S. Environmental Pro-
tection Agency's Environmental Monitoring and Support Laboratory
in Las Vegas, Nevada.
Sponsored by the U.S. Environmental Protection Agency (EPA),
the National Oceanic and Atmospheric Administration (NOAA), and
the Interagency Working Group on Satellite Data Collection Sys-
tems, the meeting was organized for two primary purposes. One
was to focus interagency attention on the lack of adequate auto-
mated in situ devices for meeting national water quality measure-
ment needs. The other was to explore possible solutions to the
problem by identifying technologies that might be applied and
initiating interagency cooperation to consolidate required re-
search and development efforts.
The 3-day meeting was attended by approximately 40 selected
participants, including representatives of 10 Federal agencies
with water measurement programs, university scientists repre-
senting various fields of technology, and a spokesman for
Canadian water programs. Several observers from industry also
attended. Participating U.S. Federal agencies were the EPA,
NOAA, the U.S. Geological Survey (USGS), the U.S. Department of
Agriculture (USDA), the U.S. Coast Guard, the National Bureau of
Standards (NBS) , the U.S. Department of Energy (DOE), the U.S.
Army Corps of Engineers (COE), the National Aeronautics and Space
Administration (NASA), and the U.S. Navy, Office of Naval Re-
search (ONR).
Opening session remarks were presented by Mr. George B.
Morgan, Director of the hosting EPA Laboratory, and by Dr.
Richard W. Paulson, representing the Interagency Working Group on
Satellite Data Collection Systems. A keynote address was
presented by Dr. Richard M. Dowd, Scientific Policy Advisor to
the Administrator, EPA.
After a series of program presentations by agency and
academic participants, the meeting was divided into five working
panels, one to identify agency needs for automated in situ
sensors, and four to concentrate on technological areas that
might be applied to sensor development. The four technological
-------
areas of focus were: electrochemical, electrophysical, optical,
and automated wet chemistry. A sixth panel, having a repre-
sentative from each of the other five, met at the conclusion of
the Workshop to integrate the work of the other panels and to
develop overall conclusions and recommendations.
This report presents the proceedings of the Workshop. These
include opening session remarks, Federal agency program
narratives, a description of Canadian activities in this area,
representative university technical presentations, and reports of
Working Panel discussions.
-------
SUMMARY
Seven primary programmatic findings of the Workshop
participants were identified. These are summarized here.
1. There is a definite need for more emphasis on de-
velopment of automated in situ sensors for water quality
measurements in natural fresh and marine water bodies, in ground
water, and in drinking water.
2. Capabilities exist within the present state-of-the-art
to develop many of the required automated in situ sensors. Ap-
propriate programs and funding are required to encourage such de-
velopment.
3. An attractive side benefit resulting from development of
in situ sensors will be the associated technology that will
produce advanced automated laboratory analysis and reduce man-
power requirements.
4.. There is presently little duplication of research and
development efforts among Federal agencies except in specialized
areas of hydrology. There is a great need, however, for better
coordination between agencies, because many express interests in
overlapping fields.
5. In order to define common agency needs, a general sum-
mary of national water quality data requirements was prepared and
is included in the Workshop report. A more comprehensive inven-
tory is needed and should be pursued through Workshop followup
activities.
6. A steering committee should be formed to complete the
work of the Workshop and to plan a second workshop within the
next year to continue communication and facilitate interagency
coordination. EPA and NOAA should take a joint lead in this.
7. Because of the common needs cited, research and develop-
ment efforts for sensor development should be cooperative and
coordinated through lead agencies. Specific recommendations have
been included in the report.
In terms of technological aspects, the working panels on
sensor technology considered sensor state-of-art and developments
-------
in the areas of electrochemical, electrophysical, electro-
magnetic, and automated wet chemical technologies. In general,
all panels reported that technology exists today for improving
the nation's water monitoring programs, with various levels of
further development both possible and desirable. The panels felt
that some of these technologies, once fully developed, could
provide the States and other agencies with more cost-effective
monitoring systems for water quality. The panels also concluded
that selected sensors would be beneficial in laboratories when
employed as rapid screening methods, reducing the time normally
required by analytical methods. Such systems could also be
excellent candidates for technological transfer to developing
nations which may lack broad-based scientific talent and
laboratories.
Of the agencies present, it appeared that the National
Aeronautics and Space Administration has made significant strides
towards assisting the other Federal agencies in developing the
automated in situ sensor systems. A few of the automated sensor
systems developed include a coliform bacteria sensor, two
miniaturized multiparameter surveillance packages, and an
inexpensive polarographic sensor for measuring dissolved oxygen
using microelectronics.
A brief summary of the findings of each technology panel is
included below.
ELECTROCHEMICAL SENSOR TECHNOLOGY
The electrochemical panel reviewed amperometric and
potentiometric sensor technology. The amperometric sensors are a
very valuable and versatile technique for the evaluation of water
parameters. These sensors have a high capability for automated in
situ measurements. They are useful with analytical techniques as
detectors for liquid chromatography and in titration procedures,
and have a great potential for organic analysis. The operational
mode determines the response time. They are capable of detecting
electroactive parameters such as: organic compounds, metals,
halides, synthetic organic compounds, oxygen and chlorine, and
oxygen demand. The potentiometric sensors are small and
inexpensive, and can be easily adapted to automated in situ mon-
itoring applications. They will normally require sample pretreat-
ment, but yield good accuracy with frequent calibration.
Capabilities exist for detecting selected cations and anions,
including fluorides, chloride, cyanide, sulfide, calcium, cop-
pers, cadmium and potassium, and coliform bacteria, and
oxidation-reduction potential.
-------
AUTOMATED WET CHEMICAL SENSOR TECHNOLOGY
A number of sensor technologies were considered, including
ultraviolet, colorimetric, membrane electrodes, and specific ion
electrodes. These would present signal processing problems for
complete automation, but most are resolvable with microproces-
sers. A certain amount of pretreatment will always be necessary.
Though high capital cost can be expected initially, a very large
number of water quality parameters lends itself to detection and
quantification by this technology. Automated cleaning would be
necessary, often depending upon types of water. It is expected
that the automatic measurements can be extended from a few hours
to a few hundred hours with the application of available tech-
nology. Measurement of total dissolved phosphorus, nitrogen com-
pounds, and dissolved total organic carbon should be pursued.
OPTICAL SENSOR TECHNOLOGY
This panel considered a number of techniques including pas-
sive and active spectroradiometry, transmissiometry, and
polarimetry. These techniques do not require direct contact with
the water and, as a result, require less maintenance than water
contact sensors. Many of these devices use the same measurement
techniques and are dependent upon electromagnetic radiation from
water bodies. They can be made to scan large areas, and provide
synoptic measurements. The active systems are capable of
performing measurements night and day. Water parameters which
can be detected include temperature, water velocity, optical
properties, suspended particles, salinity, oil presence and
quantity, sedimentation, turbidity, and chlorophyll-a.
Spectrographic techniques could be developed to identify the more
complex pollutants such as pesticides and hazardous materials.
Differential radiometry can be used to measure chlorophyll con-
centration and turbidity. The quantification of suspended
sediments and oil presence has been demonstrated by lidar
polarimetry.
ELECTROPHYSICAL SENSOR TECHNOLOGY
Some of the most powerful tools discussed by this panel
include acoustical and radioactivity methods. The acoustical
methods can provide the thickness of oil and hazardous materials
(densities less than water), and sedimentation concentration.
Acoustical methods (passive) can provide information on
activities of marine life, such as snapper, shrimp and schools of
croakers. Radioactivity techniques include the use of neutron
activation and energy dispersive x-rays. Electromechanical
methods included the use of rotors and floats for measuring water
velocity and tides. A measurable parameter of pollutants is
-------
specific conductance and can be determined with this technology.
It has been demonstrated that solid state devices can detect thin
films of oil and other pollutants with densities less than water.
It was emphasized that macroscopic measurements are often
overlooked but are essential to understand the large picture such
as total stream flow and oil coverage.
-------
CONCLUSIONS AND RECOMMENDATIONS
Specific conclusions and recommendations developed by the
Workshop panels are listed below.
CONCLUSIONS
1. There is a great need for development of automated in
situ water quality measurement systems. As suitable arrays of
such sensor systems are not available, our ability to collect
information necessary for adequate resource management and pro-
tection is seriously limited.
2. Due to lack of suitable sensor systems, development of
other advanced measurement technology, such as satellite and air-
craft remote sensing, is also limited because adequate means for
obtaining supportive ground-truth (verification) information does
not exist.
3. Present water quality data collection practice relies
heavily on manpower-intensive methods of sample collection,
analysis and automatic instrument maintenance. This provides
neither real-time information nor adequate quantity and quality
of data.
4. Present capabilities for data transmission, storage, re-
trieval, and manipulation far exceed capabilities to make ac-
curate, reliable, real-time measurements of necessary water qual-
ity variables.
5. Automated in situ sensor systems have wide application
for water quality monitoring by many agencies in fresh waters,
ground water, and marine waters of the oceans and coastal zones.
6. Although Federal government agencies require water qual-
ity measurements for different purposes and in different en-
vironments, there are many areas where requirements overlap. How-
ever, the Workshop could identify little duplication in re-
search and development efforts, except in a few specialized areas
of hydrology.
7. All agencies represented at the Workshop, with the ex-
ception of NASA, have active water measurement programs. NASA
-------
was recognized, however, to have a substantial interest and ex-
pertise in development of water measurement technology as a re-
sult of its space and aeronautical technology programs.
8. Although high-priority measurements required by most
agencies could be identified, specifications required for design
and development of sensing systems needed to make these measure-
ments could not be generalized in terms of overall agency re-
quirements.
9. Private industry research and development capabilities
are considerable and, if a profitable market can be identified,
industry resources can be applied to develop devices needed for
automated in situ measurement.
10. Technology exists to develop many of the automated in
situ sensors required. With minimum effort, short-term develop-
ment could yield automated sensors to meet some requirements.
11. Interim stages of sensor development will frequently
result in a number of automated devices for immediate use in the
laboratory. These will reduce manpower requirements as well as
the human variability inherent in manual sampling and analytical
methods.
12. Automated wet chemistry sensors have high potential for
short-term development success. Many techniques are already in
use or have demonstrated feasibility for near-term implementa-
tion.
13. Additional high-potential technologies are acoustic and
optical techniques for suspended particulate measurements, and
atomic absorption spectrophotometry, neutron activation and x-ray
spectroscopy for metal detection.
14. Improved in-place sensor cleaning and antifouling tech-
niques are a high priority need to solve performance degrada-
tion by the presence of sand, sediment, and biological organisms.
15. In addition to point measurements with in situ sensors,
there remains a need for mesoscale measurement schemes to obtain
synoptic data over large areas.
16. Consideration of sensor development programs should
include and highlight the special need for monitoring of toxic
substances and hazardous materials, and efforts should^be
directed to meet this need whenever possible.
8
-------
RECOMMENDATIONS
1. Dialogue like that initiated with this Workshop is es-
sential and should be continued to establish national prior-
ities, accelerate technological development, and coordinate
agency efforts.
2. A similar workshop should be held approximately 6 months
after publication of the first Workshop report. Coverage in this
second workshop should be expanded to include areas inadequately
covered in the first one (biology and radioactivity).
3. A steering committee should be formed to follow through
on action items identified at the first Workshop and to plan fu-
ture coordination.
4. Efforts should be made to interest one or more profes-
sional societies (e.g., Institute of Electrical and Electronics
Engineers, Instrument Society of America) in including sessions
devoted to in situ sensor development for water quality measure-
ment in their future conference plans.
5. It is strongly recommended that agencies cooperate as
much as possible to develop automated in situ systems for their
mutual benefit through joint research and development efforts.
6. All agencies should examine their programs to identify
how they can best contribute to furthering research and de-
velopment efforts in the automated in situ water quality sensor
field.
7. The technological expertise of NASA should be applied to
address this national need for improved sensors for water mon-
itoring systems, taking advantage of its technology transfer
facilities and program. It is apparent that NASA could make sub-
stantial contributions in this area.
8. A measurement requirements survey should be formally
conducted within each agency, and results of this national inven-
tory should be published by an ad hoc Workshop followup com-
mittee.
9. An interagency briefing on the results of the first
Workshop should be delivered to interested private industry rep-
resentatives .
10. Consideration should be given to directing sensor re-
search and development so interim stages of development are
focused on automated analytical laboratory devices to reduce man-
power requirements and the human variability inherent in manual
methods.
-------
11. The agencies outlined in Table 1 should assume lead
roles in development of indicated sensors, pursuing additional
funding and support, in-house and through cooperative interagency
arrangements.
TABLE 1. RECOMMENDED LEAD AGENCIES FOR SENSOR DEVELOPMENT
Type of Measurement
Lead Agencies
Other Agencies
With Substantial
Interest
Nutrients
Metals and metal ions
Suspended particulate
characterization
Water velocity and
direction
Petroleum hydrocarbons
Hazardous and/or toxic
material
Optical properties
Quality assurance,
standards
EPA
NOAA
EPA
NOAA
EPA
NOAA
NOAA - marine water
USGS - stream
USCG, EPA,
NOAA
USCG, EPA,
NOAA
EPA, NOAA, USGS
EPA, NOAA
COE, DOE, USDA,
USGS, (USD!*)
COE, (**), DOE,
NASA, USDA, USGS
DOE, NASA, NOAA,
US DA
COE, EPA, NBS,
US DA
COE, (**), DOE,
NASA, Navy,
USGS (USDI*)
COE, (**), DOE,
NASA, USDA, USGS
NASA, Navy, NBS,
US DA
All
*USDI - U.S. Department of the Interior. Various elements
of USDI such as National Park Service, Pish and Wildlife Service,
and Bureau of Land Management may have interest.
**.
Other agencies such as the U.S. Public Health Service
and OSHA may also have substantial interest but were not re-
presented at the Workshop.
10
-------
12. EPA should accept a lead role in development of small,
automated wet chemical systems for measurement of total dissolved
phosphorus, nitrogen compounds, and dissolved and total organic
carbon.
13. Ongoing efforts involving optical measurement of sus-
pended particulates should be accelerated and expanded to include
existing expertise, particularly within NASA.
14. NOAA should establish a lead agency role to further
evaluate existing metal ion measurement systems and accelerate
development of in situ sensors.
15. All agencies should stress efforts toward development
of sensor cleaning and antifouling techniques for unattended
automatic sensors in natural water environments.
16. Consideration should also be given to design of systems
for large-scale measurements.
11
-------
OPENING REMARKS
The initial session of the Workshop consisted of opening re-
marks. As primary host, Mr. George B. Morgan, Director of the EPA
Environmental Monitoring and Support Laboratory-Las Vegas,
welcomed the Workshop participants. Dr. Richard M. Dowd,
Scientific Policy Advisor to the Administrator, EPA, gave the
keynote address to set the stage for subsequent Workshop
discussions. Final introductory remarks were offered by Dr.
Richard W. Paulson, representing the Interagency Working Group.
The texts of these statements are presented on the following
pages.
12
-------
WELCOME
by George B. Morgan, Director
Environmental Monitoring and Support Laboratory
U.S. Environmental Protection Agency
Las Vegas, NV 89114
I'm delighted to welcome you people here to this workshop. We
are honored to serve as host for, I think, a very worthwhile work-
shop, something that has been needed a long time.
Our fine city isn't always considered as one conducive to
productivity for business affairs such as this, but we know that
isn't necessarily true, particularly when we have a group of
dedicated individuals such as you gathered to address a subject of
great importance. I don't say that facetiously either, as I know
from our experience in hosting other workshops, conferences and
symposiums that groups such as this can really get down to brass
tacks and do a tremendous job despite the distractions of the
evening here.
You are invited here because you are honestly the key indi-
viduals from the Federal agencies and organizations that are
knowledgeable about in situ water quality sensors. We have
purposefully sought to keep the number of participants small for
two reasons: first, to keep the group from becoming too unwieldy
from sheer size, to the point of being counterproductive; and
secondly to encourage participation by key personnel who are
particularly qualified and interested in participating in the
discussions and meeting the Workshop objectives. .1 believe our
approach has been worthwhile. We have been gratified by the
enthusiastic response and encouragement we have generally
received.
In designing the workshop format, we solicited and received
considerable advice and suggestions from various individuals in
several agencies. The working panel chairmen, whom you will all
be meeting and working with the second half of this three-day
workshop, have already put in extra effort and time from their
regular duties to attend a day-long planning meeting several weeks
ago, and do a good deal of pre-planning homework so that we can
minimize the amount of "wheel spinning" and "direction definition"
normally encountered by such working panels. We're highly appre-
ciative of the work they have already done, and for the way they
have become involved to help design discussion and reporting
formats.
As Mr. Wruble noted when he called the workshop to order a
few minutes ago, the National Oceanic and Atmospheric Administra-
tion is a co-sponsor of this workshop. NOAA has a heavy interest
in water quality sensor applications, and a keen interest in
sensor improvement. The agency has been an avid supporter of this
workshop. We are particularly grateful to Ms. Barbara Pijanowski
13
-------
of NOAA for the major effort she has put forth in planning and
conduct of the workshop. She has spent many, many hours working
alone and with EPA staff, including a special trip here last
November for an initial planning meeting, and several meetings
back in Washington since then. Her understanding of instru-
mentation technology and familiarity with water quality programs
has been invaluable in preparing for the workshop. My personal
thanks to you, Barbara, for all of your help.
The second co-sponsor of the workshop is the Interagency
Working Group on Satellite Data Collection Systems. This Group
has planned a similar workshop on their own, last December. When
they learned of our interest in hosting a multi-agency workshop,
we agreed to pool the effort and co-sponsor this gathering here at
this EPA laboratory. Dr. Richard Paulson of the U.S. Geological
Survey, the outgoing chairman of the Working Group, and
Dr. Olin Bockes of the Department of Agriculture, the newly
elected chairman have both been supportive of the workshop.
Dr. Paulson will speak to you briefly regarding the Working
Group's interest in sensor development a few minutes later this
morning. We also had the pleasure of hosting a scheduled meeting
of the Working Group here at our Laboratory yesterday.
The workshop has been designed to address a specific area of
water quality monitoring—that of automated in situ water quality
sensors for field applications. By that we mean contact or near-
contact sensors that can operate unattended in, on or immediately
adjacent to the water body being monitored, and at the field
location where the measurement is desired, versus in a laboratory
facility, either fixed or mobile. We are not including remote
sensors, that is—those that are located at some distance, such as
in an aircraft or satellite, and measure some characteristic of an
emanation from the water to describe its condition or contents.
Naturallyf there is much more than just sensors that goes into a
system to collect water quality measurements. Data recording
gear, data transmission systems, data reduction facilities and
procedures, equipment housing and packaging hardware, deployment
and retrieval equipment, and maintenance and service requirements
are some of the other factors involved. All are important, and
all can be improved upon, as anything can. However, if we were to
try to address all of these in a workshop such as this, it would
only allow a superficial treatment of the sensor aspect. As none
of the other parts in a water monitoring system are of much use
unless the basic part, the sensor, is accurate and reliable, and
as the sensor has the greatest need for technology improvement in
the overall system package, we have chosen to devote all of the
effort of this workshop on that subject. Even then, the subject
is so complex, I am sure follow-on workshops, conferences or
meetings will be required to keep the momentum going on the
initiative of improved and more effective development efforts we
14
-------
hope to generate with this workshop. We are optimistic that you
will give the ball an extra good shove over the next several days.
It's a complex problem, but we're looking forward to your
unscrambling it and identifying a route of improvement.
As a reflection of the importance EPA places on the need for
improved sensors, science and technology, we are pleased to wel-
come Dr. Richard Dowd, Scientific Policy Advisor to the EPA
Administrator, Mr. Doug Costle, and Staff Director of EPA's
Science Advisory Board, as our workshop keynote speaker. Dr. Dowd
has arranged to add this special side trip to Las Vegas on top of
an already hectic schedule in order to be here and speak to you
this morning. After a full day of work and meetings in Washington
yesterday, Dr. Dowd took a late evening flight here last night,
and has to leave tomorrow to present another address in Colorado.
That kind of schedule can be very tiring, indeed.
Dr. Dowd graduated from Yale University and then received his
Ph.D. in physics from the University of Wisconsin. He taught at
Tufts University from 1965 to 1970, when he joined the Center for
Environment and Man at Hartford. There, two of his program areas
were:
water resource management, and
environmental affects of electric power plants.
He then joined the Department of EPA in Connecticut where he was
the Assistant Commissioner for Planning and Research. In 1975 he
joined the staff of the Congressional Budget Office where he was
responsible for analysis of issues in the areas of general
science, energy environment and natural resources.
Dr. Dowd, thank you for coining. The podium is yours.
15
-------
KEYNOTE ADDRESS
by Richard M. Dowd, Ph.D.
Scientific Policy Advisor to the Administrator
U.S. Environmental Protection Agency
Washington, D.C. 20590
I am especially pleased to be here with you today, to speak
with you and have this opportunity to contribute to your efforts
over the next few days in this area of water quality sensor re-
search and development. This is the first workshop, to my
knowledge, that brings together the Federal agencies concerned
with automated in situ water quality sensors for field ap-
plications. The success of this workshop is important to EPA
and, I am sure, to the other Federal agencies concerned with en-
vironmental measurements in aquatic systems, represented here.
And so, I want to lend my encouragement and support to you, for
what I feel will be a very worthwhile and productive effort. i
am sure that you will make it a success.
This Workshop is indeed important because the sensor itself
is often the weak link in our ability to use automated water
quality sensor systems. Rather recently, we at EPA received a
communication from a southern Senator, urging us to expand our
development of more advanced automated water quality measurement
systems, stating that in his state, "Water quality is now mon-
itored by several men carrying buckets into the marsh and taking
samples. . ." He considered it an antiquated process considering
our technical achievements in other areas. In contrast with this
"bucket technology", about a month ago I sat in my comfortable
and warm living room watching the Super Bowl, played in the Super
Dome, located not far from the area the Senator referred to. i
agree, and I am sure you agree with the Senator, that bucket-
dipping technology is antiquated in contrast with the technology
which allowed me to view that game first hand.
Why do we need sensor systems, and reliable sensors as part
of those systems? There are a number of reasons for collecting
water quality measurements. Legislative mandates alone place a
large burden of data collection upon us.
It is sobering to try to list the research and monitoring
needs associated with such recent Federal legislation as the
Federal Water Pollution Control Act Amendments, the Ocean Dumping
Act, the Deep Water Port Act of 1974, the Toxic Substances Act,
the Marine Protection Research and Substances Act of 1972, the
National Environmental Policy Act of 1969, the Coastal Zone Man-
agement Act, the Safe Drinking Water Act, and the Federal Envi-
ronmental Pesticide Control Act of 1972, to mention a few. All
have spawned a tremendous need for more advanced and sophisti-
cated technology for water quality measurement. One purpose of
this conference is to lend impetus to the research and
16
-------
development of automated sensing techniques to keep pace with the
need for such technology.
The era we are now living in necessitates that we have bet-
ter quality and quantity of measurements on which to base
legislation and policy and our judicial enforcement. No longer
can we go out and casually look at a river, conclude it is pol-
luted because it looks bad to us, and expect something to be done
about it. We must have quality measurements if some meaningful
action is to be taken to improve the water quality of the river.
As our measurements become more and more sensitive, we seem to
find more and more environmental contamination. Likewise, the
distribution of old and new contaminations seem to be more and
more widespread in aquatic systems. To address these en-
vironmental problems, we need more and better measurements.
In this context, of course, we are sometimes driven to a
"bucket dipping" technology, as our needs quickly outrun our
ability to instrument for field use. A theme I would like for
you to keep in mind during your discussions is the need for
tieing the human into whatever sensor system is used. By this, I
mean you must assess implementation possibilities in terms of
simplicity, timeliness and cost (capital versus labor), simplic-
ity of operation, timeliness of results and cost of introduction
of the technology. It is on these grounds that new technologies
are often proven good or bad.
The need for water quality measurements is increasing ex-
ponentially, while the manpower and money available to us to make
these measurements are increasing at a much lower rate or, in
many instances, not increasing at all. The only solution is
improved technology to make these measurements. We at EPA--and I
am sure we are not unique among the Federal agencies—just don't
have the manpower and money to use the bucket-dipping tech-
nique to make all the water quality measurements be valid and ac-
curate enough to stand up in court or to warrant a major ex-
penditure of funds for cleanup and control. If we did have the
money and people, would we want to do it this way? I doubt it.
Additionally, I think we have all found ourselves frequently
frustrated and hampered by our inability to make the broad
spectrum of parametric measurements so often needed to classify
or characterize water. Science and technology alike depend on
the ability to measure an enormous variety of phenomena. Without
instruments to make such measurements there can be no efficient
analysis and prediction.
Another problem area we face is limitations associated with
the episodic movement of nutrients and other pollutants require
either enormous short-term investments in manpower or advanced
sensor technology capable of making the necessary synoptic or
near-synoptic measurements. Increasing manpower to meet such
17
-------
situations is usually impossible or impractical. We are, there-
fore, highly dependent upon advances in the state of the art to
provide more rapid and more precise sampling and sensing tech-
niques capable of detecting a broad range of environmental pol-
lutants at or below environmentally significant levels.
In the past, we have tended to rely on measurements which
were either easily or dependably made or which, through extensive
labor in the analytical chemistry laboratory, yielded values
sensitive enough to be environmentally meaningful. The series of
operations involved includes sampling, storage, chemical analy-
sis, key punch and data entry, and requires prolonged time
periods incompatible with many of our information needs.
Hopefully, equipped with an armamentarium of automated sensors
tomorrow's scientists will be able to select those parameters for
measurement which can best provide the answers to our monitoring
questions in time frames required for remedial actions.
Everyone who has had experience with automated water quality
sensors under field conditions can testify to problems concerning
fouling, sensitivity, range, hysteresis, calibration, non-linear
drift, response time, precision, accuracy, and interferences.
Coupled with this is the fact that many of the sensors available
are too delicate for field conditions, if there even is a sensor
available to measure those parameters we need. It makes one
wonder if we have really advanced significantly beyond bucket-
dipping techniques.
Previously, I stated that the sensor itself is often the
weak link in automated water quality monitoring systems. The
technology needed to deploy such systems involves transportation
communication, data storage and handling, packaging and sensing.'
A society which can send a man to the moon and back certainly has
the capability to deploy and retrieve packages. The technology
which allows us to communicate with ships and men under the sea
certainly allows us to communicate with monitoring packages. The
techniques which allow us to take a census of 230 million people
in the United States and store, manipulate, and summarize the
information, certainly allow us to store and manipulate the data
from water monitoring packages. The technique which allows us to
package instruments to withstand a harsh Martian environment
certainly allows us to package instruments to withstand aquatic
environments on Earth. We certainly also have the capability for
the research and development needed to adapt and make these tech-
nologies economical for our use.
Today a new generation of powerful measuring instruments is
beginning to emerge, and we must capitalize on it. Not only do
we need new and improved sensors, but we might take a lead from a
recent advancement in chemical analysis procedures that employs a
combination of two instruments; for example, the gas chro-
ma tograph and the infrared spectrometer. The combination is a
18
-------
powerful analytical tool. We might examine this to see if some
similar approach of sensor paring can be of utility in sensor
technology improvements. Additionally, adaptation of advanced
systems such as the X-ray spectrometry package NASA used aboard
the Viking lander for determining heavy metal constituents in
Mars soil might well be possible for incorporating into automated
waterborne systems for deployment in lakes, streams and other
water bodies. If measurement limits can be lowered for use in
water systems, this could be a real boon to many of us. Another
recent technology advancement involves a small, inexpensive oil
sensor using a specific heat principle. This may answer many of
our needs for monitoring water areas susceptible to oil con-
tamination. There are other new ideas and prototypes being
pursued by government, academic and industrial groups. Each of
you here is familiar with these various effects, and these ideas
and possibilities need to be shared and explored. The "two heads
is better than one" philosophy still applies.
A fundamental goal of this workshop is to establish more
communications and interplay in the Federal agency community to
orchestrate our efforts in this technology area. This inter-
change needs to take place among the agencies as well as with
others working on trying to improve our technology, such as by
the academic institutions represented here today by members of
several leading universities.
Leadership in Federal agency program planning and technology
development will have to come from you. I am sure I tell you
nothing you aren't already aware of, but you do face a challenge
in using this workshop, and managing your1 various programs, to
more effectively deal with the overall challenge - that of
carrying out the most effective program possible to respond to
the sensor improvement needs I have outlined.
It has been noted that EPA has a particular interest in
incorporating a variety of sensors into small, reliable automated
waterborne monitoring packages. Mr. Morgan, the Laboratory
Director here, has told me that in dealing with others interested
in water measurements, he and his staff have regularly en-
countered the problem of trying to keep up with the numerous
activities and new developments in sensor research and de-
velopment. That experience, coupled with the recognition and
pursuit of an improved research and development effort in this
area by Mr. John Koutsandreas and others in the EPA Office of Re-
search and Development, led to the genesis of this Workshop.
Based on this background, I wish to charge you with the
workshop responsibilities of:
Reviewing Federal Agency water quality sensor ap-
plications and research and development activites.
19
-------
Identifying commonalities and overlaps in research and
development activities.
Identifying research and development gaps.
Identifying possible cooperative research and de-
velopment efforts, and
Developing recommendations for future research and de-
velopment directions.
In reviewing the list of participants and the agenda, I am
indeed impressed by the caliber and breadth of experience this
Workshop has been able to assemble here. Such representation is
essential to meet the charges and challenges which I have laid
before you. Congratulations are in order to the organizers of
this workshop for their excellent job. And to you, the
participants, I wish you success in your deliberations and my
sincere appreciation for your willingness to take on this jobf
vital to all of us.
And let me add one final note to your discussions - do not
forget the human element. Tailor the response to the need.
There may be situations that one can conceive where you require
such a flexible system that only "bucket dipping," backed up by
an analytical laboratory will suffice.
There are many things we can do for water quality measure-
ment in terms of reliability timeliness, costs and parameters
with automated sensors. They cannot, however, solve all the
problems, at all times.
Thank you and good luck.
20
-------
INTRODUCTORY REMARKS
by Richard W. Paulson, Ph.D.
Interagency Working Group on Satellite Data Collection Systems
I'd like to just make a few brief remarks about our Inter-
agency Working Group and why we very heartedly support this
particular workshop.
For several years a number of Federal agencies have been
testing a new technology called "Satellite Data Collection Sys-
tems," which is basically a fairly straightforward technology
frou the user's point of view, whereby one can telemeter data
from in situ sensors. The community of individuals within these
agencies is rather small.
About two years ago we formally banded together and formed
this Interagency Working Group on Satellite Data Collection Sys-
tems, and we meet about once a month. Our objective is basically
to keep each other informed on our activities in this technology.
It is a rather well-developed technology, and we all are
very optimistic that we can use it operationally here in the
coming years. What has become rather clear to us very quickly is
that the weak link in the chain is the sensor.
We have existing satellite telemetry systems in orbit now
that would allow us to put a small battery-operated radio on the
roof of this building and, several times a day, telemeter all the
data that a water quality sensor would produce through that
satellite back to some central receiving station where we could
process it and analyze it and all the rest. But really the weak
link is that sensor, and last year we began to discuss the possi-
bility of dedicating one of our monthly meetings to just this
area— water quality sensors. We were informally going to get
together and tell each what our various agencies were doing. And
concurrently John and Don were laying the groundwork for this
Workshop. And since so much of what they wanted to do was what we
wanted to do, we decided rather than holding two such meetings,
we would join forces and try to cosponsor this one.
So we look forward to a very successful meeting. We have
representatives from the Department of Agriculture, Department of
Energy, NASA, NOAA, Corps of Engineers, representatives of the
DOD and the Geological Survey, representatives of the Department
of Interior and EPA. We put out a newsletter every 2-3 months
and if you are interested in the technology, talk to me and we
can get you on the mailing list for this publication.
21
-------
I think what we are trying to do here today and in the next
few days is communicate with each other. So, in summary, we
appreciate the opportunity to join in this workshop and we're
going to do what we can to help make it a success.
Thank you!
22
-------
FEDERAL AGENCY PROGRAMS
Prior to the Workshop, each participating Federal agency was
requested to prepare a written narrative describing its programs
in automated water quality sensor applications, needs and/or re-
search and development activities. Each was asked to address
such topics as why the agency is interested in water quality
measurements (legislative requirements, etc.), what sensor ap-
plications are in use, where sensors are geographically employed,
the kind of environment(s) the sensors are employed in, para-
meters of interest, measurement sensitivities desired, present
sensor deficiencies, existing and future sensor requirements,
technology forecast (i.e, what promising or potential technology
is seen), views on interagency coordination of sensor research
and development, and the gap-filling and duplication of effort
and problem solutions they see. The narratives thus prepared are
presented on the following pages.
In addition to the written material, each agency was re-
quested to present a 20-minute oral summary of its programs as
described in the narrativebylines, and on the Workshop agenda at
the end of this report.
23
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY
by Thomas M. Murray
Office of Water and Hazardous Materials
U.S. Environmental Protection Agency
Washington, D.C. 20460
The Office of Water and Hazardous Materials is involved in
several water quality-related activities and therefore, has a
variety of related water quality data needs.
For example, we need statistically sound and defensible data
to support the development of effluent guidelines for setting
NPDES permit conditions. This involves collecting samples in and
around various industries to determine the types and amounts of
pollutants in their effluents and incorporating this information
into a technology-based review of in-plant processes. The end
product is a set of effluent guidelines for the different
industrial categories.
Whereas in the past effluent guidelines were more oriented
to such classical parameters as suspended solids, biochemical
oxygen demand and pH, we now have a need for data to support the
development of best available technology guidelines necessary to
control the discharge of the 129 toxic compounds identified in
the National Resources Defense Council (NRDC)/EPA Settlement
Agreement. We also have a need for data to determine if any
additional toxic compounds are a problem and therefore should be
controlled through the effluent guidelines process. This
agreement not only requires that the EPA set effluent limitations
for these toxic compounds, but also requires that the EPA monitor
to determine if these BAT limitations are adequate. The
requirements of this agreement have now been incorporated into
the most recent Amendments to Public Law 92-500.
On the other end of this process, once permits are in force
we need data to determine whether a facility is in compliance '
with its permit conditions. This is usually accomplished through
on-site visits or as part of an intensive survey.
We need data to conduct those analyses that are necessary to
answer legislative questions, such as: Are national water qual-
ity conditions improving or getting worse? Are abatement/control
programs working? Are State water quality standards adequate to
allow for the protection and propagation of a balanced population
of fish, shellfish, and wildlife? This requires that a well-
rounded data base which is up to date with current legislative
issues and concerns be maintained. This also involves extensive
monitoring. We have a special need for more information on toxic
substances. We need data to evaluate long term potential effects
from toxics.
24
-------
We need data for planning purposes in designated and state-
wide planning areas to support the preparation, adoption and re-
vision of water quality management plans and the establishment
and implementation of regulatory programs identified in these
plans. Under 40CFR Parts 130/131, State and areawide planners
must conduct water quality assessments, and then develop ef-
fective pollution controls based on these assessments. The major-
ity of these water quality assessments should be based upon
intensive surveys. We have a special need for more reliable
nonpoint-source related data in support of this planning process.
We also need data to support the drinking water program, es-
pecially data on ambient raw water quality and groundwater qual-
ity. We have a special need for more toxics data around water
supply intakes to determine which chemical substances are present
and in what amounts.
The EPA Regions and States, principally through the Section
106 grant process, provide most of the information needed to
satisfy these various needs. EPA Headquarters and other federal
agencies provide the remainder principally through special
studies and such long-term ambient monitoring networks as the EPA
National Water Quality Surveillance System (NWQSS) and the USGS
National Stream Quality Accounting Network (NASQAN). Each year,
the Office of Water and Hazardous Materials works with the Re-
gions and States through the annual program guidance process to
ensure that State/Regional monitoring activities reflect changing
as well as current program emphases.
In the past, most federal, Regional and State monitoring
activities conducted to support these needs were limited to col-
lecting water samples at ambient fixed stations and analyzing
them for such classical water constituents as dissolved oxygen,
pH, BOD-, suspended solids, fecal coliform and so forth.
These stations were usually visited on a routine basis, and mon-
itoring activities generally were geared to providing water qual-
ity trends, establishing water quality/land use relationships,
and supporting other very basic analyses. Monitoring for the
more exotic compounds was limited due primarily to resource and
technological constraints, although some very basic toxics
information, especially heavy metals information, was collected.
Any additional toxics information was usually gathered through
special studies.
Today, however, monitoring is changing very rapidly. Now
that NPDES permits are in fprce and construction activities are
well underway, monitoring is taking on new directions. In 1976
and 1977, a special Standing Work Group on Water Monitoring,
operating within the office of the EPA Deputy Administrator,
worked with the Regions and States to develop a Basic Water Mon-
itoring Program. This program is designed to: Redirect ambient
and effluent monitoring at the State level from a fixed station
25
-------
single discharge approach to an intensive survey approach; Iden-
tify dischargers to the State's waters and assess their water
quality impact; Define a minimum number of fixed ambient stations
for national use that will be operated at the State level within
a consistent framework; Provide a coordinated nationwide as-
sessment of selected toxic compounds; and, Ensure that data which
are collected are not only used in the dec ision-making process
but are also used to educate the public and inform the Congress
Under this program, goals have been set for the operating
level of intensive surveys by the States. The goal is to conduct
an intensive survey at least once within 5 years on every river
lake, estuary, bay or aquifer where wasteloads are allocated, or
significant water quality changes either have been identified or
are considered probable.
Under this program, goals have also been set for the oper-
ating level of ambient fixed stations selected as part of the
Basic Program. The ambient stations will be operated by the
State. Data will be aggregated nationally and will be used
primarily to determine national trends in water use areas such as
water supply, fishing/shellfishing areas, etc., and in problem
areas, land use areas and in areas where future development may
impact water quality and thus baseline trends are needed.
Finally, goals have been set for the operating level of ef-
fluent monitoring programs. Major dischargers will be inspected
annually with sampling as necessary to ensure compliance with ap-
plicable effluent limitations.
This Basic Program is intended to serve as a set of mon-
itoring guidelines under the Section 106 Appendix A regulations
and is currently being implemented by the States.
In addition to the changes brought about by the Basic Water
Monitoring Program, the EPA's monitoring emphases are also ex-
panding in the area of toxic pollutants. In response to the
NRDC/EPA Settlement Agreement and now the new Federal legislative
amendments to PL92-500 (Clean Water Act of 1977), the Office of
Water and Hazardous Materials is actively involved in collecting
samples for toxics analysis. During 1978, the EPA Regions, in
cooperation with EPA contractors, will be collecting some 2,000
samples nationwide and will analyze them for the 129 toxic com-
pounds identified in the NRDC/EPA Settlement Agreement. These
samples will be collected from effluents, ambient water, water
supply intakes, sediments and tissues. The resulting data will
then be used to support exposure/risk studies, fate and dis-
tribution assessments, and other general studies.
An additional several thousand samples will be collected
under contract in 1978 and analyzed for these toxic compounds in
support of the effluent guidelines development process. The EPA
26
-------
will also expand its toxics monitoring program to other areas
such as sludge. Toxics monitoring not only holds a high agency
priority but is also a continuing program that will expand over
time.
Finally, the Clean Water Act of 1977 requires that we expand
our monitoring activities to include not only the 129 toxics
which I have mentioned, but also additional toxics and what the
law calls nonconventional pollutants. These are constituents
that have not been designated as either toxic or conventional
(BOD, suspended solids, etc.) but must, nevertheless, be con-
trolled under BAT effluent limitations.
As you can see, monitoring especially in the toxics area, is
evolving quite rapidly. We are moving away from the routine mon-
itoring of the classical parameters at several fixed stations to
the monitoring of toxics and the conduct of intensive surveys.
If technological support is to keep pace, it must also evolve
rapidly.
One of the reasons for holding this workshop is to find ways
of strengthening current water quality sensor technology. This
is all well and good. For example, we need to strengthen the re-
liability of existing in situ water quality sensors to gather
more reliable real time information on nonpoint-source related
pollution. For our purposes, I feel that efforts must be made in
this area. But, more importantly, I strongly suggest that if we
are to keep up technologically with the monitoring process, we
must stay one step ahead and concentrate our efforts on de-
veloping new water sensors to detect toxic compounds.
We are looking for a way to reduce the high costs of toxic
analyses. For, the analytical costs involved in toxics mon-
itoring are high. Typically, it costs anywhere from $800-$!,500
to analyze one sample for the 129 toxic compounds. Multiply
these costs times the several thousand samples required to
adequately support the program and you have a very expensive
program.
We need a low cost, reliable method—and I emphasize low
cost and reliable--to detect and quantify, in the water column
and effluents, the 129 toxic compounds identified in the Set-
tlement Agreement. We need this not only to round out our data
base on these toxic compounds while saving resources in the
laboratory, but also to strengthen the regulatory arm of the
Agency.
Over the last several months we have investigated the use of
various in situ bioassays and also an immobilized enzyme tech-
nique as potential means of gathering reliable toxics data while
reducing the costs of toxic analysis. In this connection, a
water quality sensor may be available, or could be developed to
27
-------
satisfy this need. In any case, the time is ripe to start look-
ing at some of these new techniques.
Now, I am not suggesting that we immediately go out and de-
velop an in situ water quality sensor for each of the 129 toxic
compounds. But I do suggest that we develop, minimally, a sensor
or series of sensors, sensitive to the parts-per-billion (ppb)
level, that could serve as a screening device to tell us if a
particular class of toxic compounds is present or not. The
general classes under which the 129 toxic compounds are grouped
include halogenated hydrocarbons, the pesticides/insecticides
group, heavy metals, the aromatics, industrial solvents, the
phenols, the benzene toluene group and miscellaneous compounds.
A screening method such as this could result in considerable
savings in the laboratory.
It would also be desirable if this sensor or series of
sensors indicated relative concentrations or magnitudes of tox-
icity. Then, as technology evolves, perhaps these sensors could
be redesigned to be more quantitative than qualitative in nature
Water samples currently being collected by the EPA Regions
and contractors could possibly be split and made available
through the EPA Office of Research and Development to check the
performance and reliability of these sensors as they are de-
veloped .
If we had a low-cost and reliable in situ water quality
sensor designed to detect classes of toxic compounds we could:
• free up valuable resources in the laboratory
to support other high priority activities;
• develop an early warning system for industry
or field use that could help prevent or at
least allow us to respond more quickly to
accidental toxic spills;
•establish the sensor(s) in privately owned
treatment works (POTWs) to determine the
correct retention time for most effective
treatment;
• measure the impact of the volatile toxic
compounds on the receiving waters more
effectively; and,
• measure the variability and frequency of
toxic discharges more effectively.
We could do a great deal with such a sensor and the Office
of Water and Hazardous Materials is willing to cooperate with the
28
-------
EPA Office of Research and Development and the scientific com-
munity to develop it.
In conclusion, let me leave you with this one word of cau-
tion. I think the challenge of developing such a valuable in
situ water quality sensor is attractive. But, the real challenge
is to make sure that it is reliable and cost-effective. Let us,
therefore, work toward this goal.
29
-------
NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION
by Barbara Pijanowski
Office of Ocean Engineering
National Oceanic and Atmospheric Administration
Rockville, MD 20852
Unlike some of the other agencies represented at this
workshop, the National Oceanic and Atmospheric Administration is
a relative newcomer to the federal system. Along with the
Environmental Protection Agency, NOAA was created under a
presidential reorganization order. Unlike EPA, however, NOAA was
not given independent agency status, but was placed under the
Department of Commerce. NOAA is the only federal agency with
general ocean responsibilities; the purpose for its creation was
to consolidate federal civilian ocean activities so that improved
coordination would lead to more effective and efficient ocean
efforts.
NOAA has three primary goals, ocean resource management
coastal zone management, and environmental surveying, monitoring
prediction and control. The atmospheric activities of NOAA are
considerable and until the past few years, NOAA's efforts in en-
vironmental monitoring and prediction have concentrated pri-
marily on the atmosphere rather than the ocean.
NOAA's overall responsibilities related to its ocean
functions can be stated in general terms as:
• assessment of ocean resources including living
resources and minerals, but excluding hydrocarbons;
. execution of research leading to an understanding of
the naturally occurring phenomena in the marine
environment, and;
• execution of research leading to an understanding
of the distribution, fate, amd effects on marine
ecosystems, of pollution and disturbances caused
by man/ and coastal surveys for mapping, charting
and geodesy.
In order to carry out such responsibilities comprehensive
programs are required for the integration of basic research,
baseline studies and monitoring operations in estuaries, on the
continental shelf, in the deep oceans and in the Great Lakes.
NOAA's requirements for water measurements therefore span the en-
tire range from fresh water to sea water in depths ranging from
the top few centimeters at the air-sea interface to shallow bays
and the deep ocean. In addition, measurements must be made from
stationary platforms such as buoys and towers, from moving
platforms such as ships and helicopters with towed devices for
profiling apparatus, and from remote platforms such as air-
craft and satellites. NOAA has a need to obtain synoptic
30
-------
measurements over wide geographical areas in order to examine the
big picture, as well as raicroscale measurements over a few cen-
timeters in order to observe the microstructure of an individual
patch of plankton or chlorophyll.
Measurements are made at the air-sea interface, in the water
column, and in the particulate matter in the water column; in the
sediments on the bottom and in the interstitial waters of these
sediments as well as in the biological organisms that swim and
drift through the water column or crawl, burrow, or attach them-
selves to the ocean floor.
Since this workshop is concerned primarily with the measure-
ment of water quality, it would be useful at this point to dif-
ferentiate between water quality measurements and those typically
thought to be oceanographic measurements. Water quality
measurements provide an understanding of the water environment
and its ecosystems that can be used for assessing the character
of the water system and determining the effect of man-induced
disturbances. Water quality measurements are (1) those that
quantify actual pollutants such as hydrocarbons and metals; (2)
the chemical and biological parameters that characterize the en-
vironment in which the ecosystem exists such as totalorganic
carbon and nutrients; and (3) general supportive measurements
that lead to information on the distribution and dispersion of
pollutants. Supportive measurements are generally currents, tem-
perature, salinity and, at times, waves and tides.
It is important also to recognize that there is a close re-
lationship between water quality information and fishery as-
sessment information. Both kinds of information are necessary
for assessing the impact of pollution on the environment and for
understanding the natural relationships between fisheries and the
environment. Because of our present lack of knowledge concerning
these interrelationships and our lack of adequate measurement
technology for pollution studies, it is often necessary to use
biological organisms themselves and their abundance and dis-
tribution data as indicators of water quality. It is not possible
therefore, to treat fishery assessment activities independent of
water quality assessment activities.
Most of NCAA's water measurements result from requirements
of several programs which have evolved historically to carry out
functions that were inherited from NOAA's predecessors and to re-
spond to more recently legislated responsibilities such as the
Marine Protection, Research and Sanctuaries Act of 1972 and
Fisheries Conservation and Management Act of 1976.
NOAA's programs are broad in scope and require the collec-
tion, storage and utilization of data on many levels. Data are
collected through various programs by NOAA's own survey
activities and through the use of contractors. NOAA presently
31
-------
has 24 ships, 14 of which could be used for collecting
oceanographic and fisheries information; however, most of NOAA's
water quality data are collected under contract by university and
private industry groups. NOAA stores and manages marine environ-
mental data through its National Oceanographic Data Center which
is part of the Environmental Data Service. And finally, NOAA
uses the information that it collects and stores to assess and
predict environmental and resource conditions for management de-
cisions by NOAA, EPA, and the Department of Commerce.
Most of NOAA's current water measurement activities are car-
ried out through five major programs; several additional programs
are still in development phases, and some limited measurements
are made as part of other general research activities. A short
description of each of the major programs follows:
Marine Resources Monitoring Assessment and Prediction Program
(MARMAP) MARMAP is designed to provide the necessary
information for management and allocation of the nation's marine
fishery resources. The program is long term and requires the
collection and analysis of data to provide basic information on
the natural abundance, composition, location and condition of the
commercial and recreational marine fishery resources of the
United States.
Ocean Pulse
Ocean pulse is a new program whose objective is to monitor
the "health" of the ocean or the consequences of pollution on
living resources. The program is designed to provide baseline
information on physiological and biochemical parameters of
indicator species as well as basic water quality information at
specific sites in order to assess the health of the living re-
sources. The initial study area is the continental shelf off the
northeast coast of the United States.
Marine Ecosystems Analysis Program
(MESA) MESA is designed as a NOAA program to focus the
cooperative efforts of federal, state and local agencies, uni-
versities, industry and environmental organizations on the
investigation of specific marine environmental problems within
limited geographical areas. Selected coastal and offshore areas
are subjected to intensive investigation for relatively short
periods of time (5-7 years) to determine the impact of man on the
ecosystem of the areas and to gather sufficient information to be
able to predict consequences of future activities there. The
area selected for the first MESA project which began in 1972 was
the New York Bight. The Bight is adjacent to one of the most
populated and industrialized regions in the world. It is a re-
pository for wastes from over twenty million people and a host of
32
-------
major industries as well as the recipient of the nation's largest
ocean dumping operation. Additionally, the area is an important
environmental resource, with abundant commercial and recreational
fisheries. The second MESA project was started in 1975 in Puget
Sound. This area, unlike the New York Bight, has no evident
large-scale environmental problems at the present time and study
of the area is expected to provide information that will lessen
the impact of the increasing stresses associated with economic
growth and industrial expansion that are now taking place. Ad-
ditional areas may be added to the MESA program in the future.
Ocean Dumping Program
This was established as a direct result of legislated re-
sponsibilities. Under Title II of the Marine Protection, Re-
search and Sanctuaries Act of 1972, NOAA was given responsibility
for conducting research and monitoring the environmental effects
of ocean dumping. The program is a continuing one and its ob-
jectives are:
. determining the environmental effects of past and
present dumping activities
• investigating proposed new dump sites
• monitoring selected dump sites
• supportive laboratory and field studies to
investigate
pathways and fate of dumped materials
• developing dump site selection criteria and criteria
for differentiation between harmful and non-harmful
materials for dumping.
The Deep Ocean Mining Environmental Study
(DOMES) Domes is a program concerned with the environmental
problems that may arise from the deep-ocean mining of manganese
nodules. The objectives of the program are to:
• identify potential environmental problem areas and
establish priorities for research,
.obtain baseline information for potential mining
areas,
•develop and test predictive models for the effect of
ocean mining on the marine environment.
One unique aspect of this program is that this is the first op-
portunity for observation of environmental consequences of a new
industry from the beginning. Since deep ocean mining has not yet
progressed beyond experimental stages, opportunity to establish
true baseline information as well as to observe first hand the
effect of mining operations on the environment.
33
-------
The Outer Continental Shelf Environmental Assessment Program
(OCSEAP) OCSEAP as coordinated by NOAA is concerned with the
outer continental shelf of Alaska. It has as its primary ob-
jective collection of the information necessary to enable pro-
tection of the marine environment in that area in a manner that
is compatible with offshore oil and gas development. The program
is being conducted by NOAA for the Department of Interior's
Bureau of Land Management. It focuses on nine lease areas on the
Alaskan outer continental shelf. General objectives of the
program are:
•location of critical wildlife habitats that must be
protected
.prediction of effects of oil and gas spills or other
disturbances generated as a result of the oil and gas
development
• identification and development of new monitoring
techniques
. definition of environmental stresses on manmade
structures in order to reduce polluting and safety
incidents.
Hazardous Materials Response Program
Although this program has not been fully implemented within
NOAA, some efforts have been initiated during the past year. it
is designed to provide quick response to oil spills so that re-
quired data on the fate and behavior of oil can be collected in
order to improve modeling efforts for forecasting and prediction
of the movement of spilled oil in the marine environment. The
objectives of the program are to:
• provide on-scene support to Coast Guard coordinators
with respect to clean-up, containment and mitigation
of impact
• provide damage assessment
• perform research on actual spills
At the present time, this program does not have its own staff,
but draws experts from other water programs in-house as needed
for specific projects.
Although each of these water programs is run separately and
from different organizational levels within the agency, many com-
mon elements exist, particularly with respect to requirements for
data acquisition. Appendix 1 is a listing of parameters which to
one degree or another have been identified to be of interest in
NOAA programs.
34
-------
Appendix 2 provides information on measurement requirements
for nine of the most commonly measured parameters as they were
identified in a 1976 workshop (1). Although this information was
gathered from many non-NOAA sources, it is typical of all water
quality requirements. It is not surprising to find that most of
these parameters also happen to be the only ones for which direct
measuring in situ instrumentation for use in sea water is readily
available.
Although some NOAA programs have been able to provide de-
tailed requirements for a few of the most commonly measured pa-
rameters, program managers have been unable or unwilling to
document specific requirements because:
• The research nature of some projects makes it,
difficult to predict expected conditions during
the program planning stages in anything more than
general terms;
• Most programs have only minimum lead time and must
begin operations immediately with very limited budgets,
i.e. short term high visibility programs are the
rule rather than the exception. The luxury of
lengthy and expensive development programs for new
or improved measurement techniques to satisfy
requirements for better accuracy and reliability
cannot be afforded. Consequently, program managers
tend to settle for the available technology and
specify their requirements in terms of what they know
the available equipment can provide.
To add more complexity to the situation, actual measurements
for many programs are taken by activities under contract to NOAA.
Many of these contractors have their own equipment and capabil-
ities which may result in compromises between the requirements
that are actually desired by the program planners and those that
are realistically possible had their budget provided for procure-
ment or development of more advanced and reliable equipment.
In general terms, the deficiencies that persist in marine
water quality measurement technology fall into two categories:
those systems that exist and need improvement, and those systems
that do not exist at all. Under the first category, systems do
exist for the in situ and/or near real time measurement of tem-
perature, salinity/conductivity, depth, dissolved oxygen, pH,
current speed and direction, optical properties and chlorophyll
a, in marine waters. A few additional parameters can be measured
Tn fresh water because of fewer chemical interferences and dif-
ferent concentration range . Of these systems, only those that
measure temperature, salinity/conductivity and depth can be con-
sidered to be adequate, i.e., fairly reliable measurements can be
achieved with current state-of-the-art instrumentation. All of
35
-------
the other measurements can be achieved to lesser degrees of re-
liability. Current measuring instruments are not standardized
and different current measuring techniques produce different data
in dynamic environments. Dissolved oxygen and pH systems are un-
suitable for long term unattended use because of their calibra-
tion instability and frequent maintenance requirements. The
measurement of optical properties and chlorophyll is complicated
by the wide variety of devices available for measurement with
little comparability as well as problems with fouling during lonq
term unattended use. In all cases, the limiting factor in these
systems is the sensors and not the electronic or data handling
components.
As far as the need for new sensors is concerned under the
second category, a brief glance at Appendx A will illustrate that
out of that entire list, only a handful of parameters, those
listed under the first category, can be evaluated outside of a
controlled laboratory environment. In order to make these
measurements, analysis must be performed using standard wet
chemical techniques on samples that have been laboriously col-
lected at a few discrete points and have been removed from their
surrounding environment, artificially preserved, aged, and sub-
jected to numerous sources of contamination, yielding far less
than real time or synoptic information.
Sensors simply do not exist for most of the parameters of
interest in marine waters. While it is neither practical nor de-
sirable to develop in situ measurement techniques for all of the
parameters listed in Appendix 1, it is necessary to develop such
systems for the parameters that must be:
•measured frequently, or
• measured in remote or inaccessible areas or areas
where the cost of sending a man to obtain the
measurement is very high (operating costs for an
oceanographic vessel are in the neighborhood of
$4000 per day ), or
• measured precisely, subject to fewer of the
variations and uncertainties caused by individual
analysts.
Recommendations
NOAA has produced several documents that consider water
quality measurement recommendations. "Marine Pollution Mon-
itoring: Strategies for a National Program" (2) is the product of
a workshop sponsored by NOAA in 1972. At that time the con-
taminants of greatest concern were identified as heavy metals,
petroleum and halogenated hydrocarbons. Since that time ad-
ditional categories have been added (3), namely low molecular
weight halocarbons and transuranic elements.
36
-------
In addition to the requirements that have been identified
above, personal contacts and interviews with key program person-
nel in NOAA have identified several additional areas where
improved technology for in situ measurement would be desirable.
Priorities have not yet been evaluated.
Needs have been identified for:
• water clarity measurements over the entire water column
available light to biota at selected depths,
characterization of particulates in order to
distinguish between natural particles and
particles of dumped material,
ability to track dumped material
• in situ ammonia measurement in the water column and
near the bottom (to 1 Mgm/1 level)
• continuous and periodic measurement of heavy metals
and associated speciation
•measurement of total available nitrogen, i.e.,
nitrogen in forms available to the biota (unfortunately
this varies with the organisms)
• measurement of sulfides in sediment
• measurement of total organic carbon in dissolved
and particulate forms
• measurement of total biomass
• measurement of chlorophyll to map plankton patches,
(locate chlorophyll maxima and obtain time and
spatial variations)
• measurement of plankton over large areas
phytoplankton: identification of algae type
zooplankton: identification of species and size
distribution
zooplankton: particle count and volume
• reliable dissolved oxygen sensors for long term
unattended use
. closer to real time data on nutrients.
In the past, NOAA's efforts in the research and develop-
ment of in situ water quality devices have been limited to areas
of overlap with traditional oceanographic programs.
1. The Office of Marine Technology in the National
Ocean Survey has been exploring an acoustic
technique for the underway shipboard measurement
of water currents. This project was begun in FY 76
and is scheduled for completion in FY 80.
2. Initial research efforts by Dr. Donald Barrick of the
Wave Propagation Laboratory of NOAA have demonstrated
the feasibility of using shore based high frequency
radar to provide a map of surface ocean currents
37
-------
to a distance up to 50 miles from shore. The
possibility exists for extending the technique to
include directional wave measurements. The Office
of Ocean Engineering will be working with the
Wave Propagation Laboratory to carry out the
engineering development necessary to provide
a commercially available system.
3. An attempt to produce a more reliable and accurate
dissolved oxygen sensor for use as a laboratory
standard has led to considerable research on the
polarographic dissolved oxygen probe by Dr. J.A.
Llewellyn of "the University of South Florida. His
work sponsored by NOAA has produced a great deal of
information on the limitations of polarographic
techniques for dissolved oxygen measurement.
4. For the past several years, Dr. John Proni of
NOAA's Environmental Research Laboratory in Miami, has
been experimenting with towed echo sounding systems for
the acoustic measurement of suspended sediment in
water. Using a 200 KHz signal, he has been able to
track sediment and sewage sludge plumes and estimate
suspended load throughout the water column. Using 3
MHg signals, estimates of sediment load per centimeter
near the bottom and mixing rates have been obtained.
The system is still a research tool, and work is
continuing on its improvement.
Research and Development Programs
The responsibility for improving measurement technology
within NOAA falls within the Office of Ocean Engineering. This
office was formed a little more than a year ago to serve all of
NOAA as a focal point for ocean technology development. One of
the major programs within the office deals with development of
ocean instrumentation. In order to provide direction to this
program, several studies are being conducted which assess the
water quality measurement and standards needs of NOAA and the
marine community. The measurement needs assessment study is
being carried out by MAR Inc., Rockville, MD, under contract to
the Office of Ocean Engineering; preliminary reports are avail-
able now and a final report is due at the end of this month.
Several additional contracts dealing with marine water quality
standards and quality assurance programs are also underway.
Beginning this year, FY 78, the Office of Ocean Engineering
will be funding additional efforts in water quality measurement
R&D. In addition to the technology assessment programs mentioned
earlier which are needed to give direction to the program, the FY
78 program includes the following projects, all of which will be
carried out under contract to the OOE:
38
-------
1. improvement of dissolved oxygen sensors for
long term unattended use,
2. improvement of current measurements with existing
instrumentation,
3. engineering development of the HF radar system for
surface current measurement,
4. improvement of standards for conductivity and current
measurements.
The FY 79 program for which $620K has been requested in the
President's budget, includes additional efforts for:
1. development of field standards for chlorophyll and
dissolved oxygen,
2. initiation of a program for automation of olankton
classification and counting,
3. evaluation of existing oil slick sensors for use on
unattended buoys.
Projected plans for the FY 80 program will require about a
fourfold increase in the budget and includes additional efforts
for:
1. development of an ammonia measurement system for in
situ, real time use
2. development of in situ techniques for measurement of
trace elements and speciation
3. improvement of underwater light measurements
4. development of in situ measurement system for sus-
pended particle characterization.
Interagency Cooperation
NOAA's only formal cooperative efforts in water quality
measurement at the present time are with the Environmental
Protection Agency with Energy Pass-Through Funds, and the
National Bureau of Standards. Both efforts involve the
development of laboratory and field standards. The programs deal
primarily with standards and procedures leading to improved data
quality, and not toward the development of in situ
instrumentation.
39
-------
While no formal cooperative agreements with other federal
agencies exist, we have followed with interest the R&D being car-
ried out by NASA-Langley, in the measurement of chlorophyll and
coliform, and the USGS in their improvement of suspended
particle/turbidity measurements.
At the present time, within the federal system, we cannot
identify any areas of R&D overlap for development of in situ
measurement techniques. We can however identify many gaps, most
of which have been pointed out earlier in this presentation.
40
-------
References
1. "Measurement Requirements for Marine Water Quality Data,"
B. Pijanowski, Oceans '76, Sept. 1976
2. "Marine Pollution Monitoring: Strategies for a National
Program," ed. E.D. Goldberg, NOAA, Oct. 1972
3. "Assessing Potential Ocean Pollutants," Report to National
Research Council, NAS, 1975
4. "The Environmental Quality Monitoring Report," OCS Task
Team, NOAA, Feb. 1976
41
-------
APPENDIX A. PARAMETER REQUIREMENTS
CHEMICAL f AHAMETOS
rO
acidity
alkalinity
arsenic
asbestoa
too
bo FOB
calcium
carbon. Inorganic
carbon dioxide
carbonates
blcarbonates
carbon monoxide
carbon, organic
dissolved paniculate
total (TOO)
chlorides
chlor ine
chlorinity
chemical oxygen demand (COD)
cyanide
dissolved oxygen
dissolved silica (silicates)
fluoride
hardness
hydroxides
iodine
metals
(dissolved, partlculate, total)
aluminum
antimony
barium
beryllium
o cadmium
o chromium
cobalt
o copper
o lead
magnesium
manganes
o mercury
molybdenum
o nickel
potassium
selenium
silver
sodium
thorium
tin
titanium
vanadium
o sine
nitrogen
ammonia
nitrate
nitrite
urea
amino acids
total
orthophoaphate
phosphorus
phosphates
organic phosphate
total phoaphate
pH
redox potential
sulfur
sulfates
sulfride
sulflte
hydrogen sulfide
GENERAL PHYSICAL PAIaMTrns
bottom topography
color
conductivity
current direction
current speed
density depth odor
optical properties
transmission
scattering
ambient light
extlBCClaa
salinity
solids
dissolved
floating
settleable
auspended
total
suspended particulatee
particle shape
particle size distribution
total suspended load
tempereture
wave direction
nave height
«ave period
turMdlty
sound speed
light penetration
unveiling 1 dovwelllog radiance
visibility range
sediment type and deposition
cloud cover
precipitation
wind speed and direction
subbottom topography
Radioactive Materials
gross alpha
gross beta
gross gamma
gamma spectrotcopy
Amerlclnm - 241, 243
CM) Argon - 40
Barium - 140
Beryllium - 10
00 Calcium - 40
00 Carbon - 14
Cerium - 141. 144
o Ceelum - 134, 137
Cobalt - 158. 16O
Iodine - 131
Iron - 55, 59
Lead - 210
neptunium - 239, 23». 237
o Plutonium - 239, 240
Polonium - 210
00 Potassium - 40
Protschtlnium - 231
Kmslum - 226, 228
laden - 222
00 tubidlum - 87
Ruthenium - 103. 106
Strontium - 89, 90
(IT) Thorium
Trltlated Water
OO Uranium - 235, 238
BIOLOGICAL PARAMETERS
algae
macroalgae .
bacteria
benthlc organisms
mlcrebenthle organisms
collform
eplflore
eplfauna
fish demersal
pelegic
invertebrates
neck ton
plankton
phytoplankton
zooplankton
ichyoplankton
perlphyton
macrophyton
neuston
viruses
under Ice blocs
birds
assimilation ratio
biomass
ATP
BOD
chlorophyll a
chlorophyll b
chlorophyll c
heterotrophlc uptake
setting avlllty
Byirgcarbons
total ajiphatlcs
total aromatics
C,0, C . C17
Iw 11 12
1 •» * "1 aV* *1 K
j-3 14 15
c.,, C.,, c..
lo 17 19
C C C
19' 20* 21
C,-, C,-, C,
24
C«» C-., C-,
«J «D //
C.-, C?0, C
30
C31
prlstane
pfaytane
o-xylene
l-?ropyl bentene
N-propyl benzene
indan
blphenyl
dimethyl naphthalene
trlmethyl naphthalene
f luorence
phenant hrene
anthracene
l-methyl phenanchrene
fluoranthene
bens (A) anthracene
chryeene
perylene
dlbenzothiophene
benzo (A) pyrene
benzo (E) pyrene
o higher priority
<•) naturally occuring
-------
U.S. DEPARTMENT OF ENERGY
by Mary S. Hunt, Ph.D.
University of California
Berkeley, CA 94720
The Department of Energy is certainly pleased to be able to
participate in this meeting on water quality sensing because the
DOE sees water quality as a major concern. Because we have not
had a long time to prepare this talk, we have been unable to talk
to the large number of people doing analytical work with the
Department.
It seemed therefore appropriate to discuss a variety of
projects done at Lawrence Berkeley Laboratory which are represen-
tative of types of water quality sensor research supported by the
DOE. The types of research done in water quality can be broken
down into three categories:
• Basic Research
• Application of Developed Techniques
• Analysis of Instruments, Systems and Strategies
Basic Research involves investigation and development of the
most elemental parts of the technique including the detector and
the electronics. It is possible that at this time the sensor may
never touch water. Next the sensor is tested and proven in a
laboratory situation. Research involving the application of de-
veloped techniques involves taking the sensors developed in basic
research and applying them to the environmental situation. It is,
of course, always necessary to continuously analyze the instru-
mentation systems and strategies being used and developed with
respect to the water quality parameters which must be de-
termined. One must consider the pollutants and parameters them-
selves, the regulatory standards and the recommended water qual-
ity criteria as a basis for evaluating the sensors available.
Clearly one must also examine and evaluate the sensors them-
selves, and the scope and adequacy of the monitoring system and
the strategy behind it. The information gleaned from this analy-
sis should be readily available.
All three types of research are represented at LBL. The
various projects are as follows:
1. Basic Research Environmental Instrumentation
and the Instrumentation and
Environmental Research Groups
• semiconductor detector
development
43
-------
• XRF development
.anodic stripping voltammetry-
2. Applied Research OTEC Project (Ocean Thermal
Energy Conversion)
. artificial substances for
in-situ growth
• in-situ thermal gradient
thermal conductivity
. the Fish—in-situ water
quality monitor
3. Analysis Environmental Instrumentation
Survey
The Environmental Instrumentation Group has, over the last
decade, made significant advances in the use of semiconductor
detectors, particularly as applied to non-dispersive (energy
dispersive) X-ray fluorescence (XRF) spectrometry.
They have built XRF spectrometers with high resolution by
significantly improving the electronics. After developing an
instrument which worked satisfactorily under laboratory
conditions, they have automated and computerized it, and it has
been working for one and a half years with only two minor
breakdowns. Although the instrument is currently used to monitor
air particulate composition, the adaptation of the instrument to
samples from water systems, particularly suspended solids or
sediments, would not be a problem.
Another instrument, working on a wavelength dispersion
principle, has been developed to monitor sulfates. It is
expected to operate in the field for up to 3 years.
The anodic stripping project, conducted by Clem et al., is
exploring the means to enhance the sensitivity and broaden the
applicability of the technique. Already Clem is able to detect
lead in a 15-ml cell at 40 ppt (0.04 ppb); if the cell were
flow-through type Clem could extend the sensitivity to less than
10 ppt. Although his instrument is usually used in the lab, Clem
notes that it encounters no interference effects in the presence
of sewage and saline waters. Clem is currently attempting to
develop resin coated electrodes, which would extend the range of
metals he is already able to measure.
The OTEC group has developed a variety of novel techniques
for determining water quality parameters. The project involving
44
-------
artificial substances used for substrate in situ growth is an
excellent technique for determining primary and secondary growth,
ATPr and composition of algae. It has been used effectively on
the Truckee River in California. The technique has two
drawbacks: first, it requires substantial intelligent human
effort and secondly, the monitors tend to annoy fishermen or
interest little boys, and unless hidden carefully, seem to end up
in the bushes. Vandalism of one form or another is an important
problem to any on-site instrumentation. Blinkers used to mark
mussel beds monitored in San Francisco Bay are now reputed to
light many windows in Sausalito.
Most applicable with respect to this conference is "The
Fish," a multipurpose probe which has been used to monitor depth
temperatures,conductivity, pH, redox potential, sulfide ion
activity and ambient light while being towed behind a boat. It
has been used in a variety of marine and estuarine situations,
but is not limited to salt waters. In order to install it for
long periods of time it would be necessary to solve some
calibration problems and to install a cassette or transmitter.
The Environmental Instrumentation Survey Group has for the
last few years reviewed the instrumentation commercially
available for monitoring a variety of pollutants in water and the
principles behind their operation. We provide material
concerning the pollutants and their forms and characteristics,
their sources, the effects they have on the environment and human
health and the methods used to control them. We also provide
information regarding developing techniques, so that the reader
can be aware of the state of the art even if it is not currently
commercially available.
In conducting the survey, we have certainly noted the lack
of sensors available for long term in situ monitoring. We also
recognize that with the current regulations and with the
important need to understand and protect our aqueous environment,
it is essential to accurately and precisely monitor a wide
variety of water quality parameters on a real time basis at a
realistic cost. We have also noticed an unbalanced emphasis on
performing quantitative analyses for the known or suspected
pollutant instead of performing at least sporadically qualitative
analyses for unsuspected pollutants. These unknowns can be
detrimental themselves, but they also can have either a
synergistic or antagonistic relationship to other pollutants with
respect to measurements and environmental or health effects. One
of the other drawbacks that was noticed in our analysis was that
most environmental programs have ignored metallic speciation
entirely. Yet, it is well known that organic mercury is far more
toxic than is inorganic mercury, and the CR(VI) is potentially
carcinogenic, whereas CR(III) is more harmless.
45
-------
To summarize, the Department of Energy wishes to express its
major concern for water quality. Its interests are not limited
to one geographical area, but are nation- and world-wide, as is
demonstrated by the satellite program which has been mentioned
earlier in these discussions. Research is being performed on
most categories of water—inland waters—lakes and rivers,
surface and groundwater; in enclosed bays and estuaries and in
the ocean. The quality of water examined ranges from fresh
mountain streams to estuaries, to polluted rivers and sewage.
The parameters examined are as diverse as the water monitored.
Research is not limited to one area, but encompasses the range
from basic research to analysis of strategies.
The DOE clearly sees water quality as a major concern and
hopes to be able to help solve the problems which are being
considered by this workshop.
46
-------
U.S. ARMY CORPS OF ENGINEERS
by Earl E. Eiker
U.S. Corps of Engineers
Washington, D.C. 20314
The civil works program of the Corps of Engineers involves
the entire spectrum of water resources development and manage-
ment. The Corps' role traditionally has involved planning, de-
sign, construction, operation and maintenance of water resources
projects to meet numerous purposes. These purposes include flood
protection, navigation, water supply, fish and wildlife en-
hancement, hydropower, water quality control and recreation. The
national concern for the preservation and protection of the en-
vironment has resulted in legislation that makes environmental
quality an additional consideration in water resources de-
velopment. In order to address this concern, water quality data
collection programs are maintained at most Corps projects. Data
generated are used primarily in connection with project operation
and environmental impact analyses. The man hours required to
conduct the data acquisition and analysis for these programs is
great. An attractive alternative to manual collection of data is
greater utilization of in situ monitoring systems that can
rapidly, accurately and cost effectively accomplish this task.
Presently, all Corps field offices as well as research labs
are using water quality sensors to measure water quality para-
meters in the field. The distribution of these offices is
throughout the entire United States. Because of this wide
geographical distribution, there is a corresponding wide range of
environmental conditions associated with these sensor ap-
plications. Sensors have been used in marine, estuarine, river,
stream and lake environments in areas subject to large changes in
temperature, wind, precipitation and many other climatological
and hydrological conditions. In addition, the remoteness of
several locations has created additional problems relative to
servicing and maintenance.
The diversity of the Corps' water quality data collection
effort makes for a wide-range of parameters of interest. To
date, sensor applications have been limited by the availability
of equipment. Also, the intended use of the data has a large
bearing on sensitivity requirements. Some parameters of interest
and limits of accuracy (where appropriate) required are: 1) tem-
perature (0.1° C), 2) dissolved oxygen (0.5 ppm), 3) optical
properties - light transmission and scattering, 4) total and
ortho phosphate (0.1 ppm), 5) ammonia nitrogen (0.1 ppm), 6) ni-
trate nitrogen (0.1 ppm), 7) organic carbon (1 ppm), 8) pH (0.2
units), 9) eH, 10) specific conductance (5 mhos), 11) oxygen
reduction potential, 12) specific ion measurements (particularly
chloride and sulfide) and 13) total dissolved gases (percent
saturation). Along with sensitivity requirements, it should be
47
-------
noted that specification of accuracy, instrument response time
and stability are also very important sensor criteria. Often a
sensor can be very sensitive to change in parameter concentration
but inaccurate at the same time.
Presently, sensor deficiencies are related primarily to lack
of reliability over extended periods of time. This fact greatly
limits the application of available sensors. These problems are
caused by exposure to sunlight, algae growth, accumulation of
sedimentation and other debris, temperature variations, salinity
and ice cover. Also, sensors are, for the most part, fragile and
thus easily damaged during shipment and use. Calibration of
sensors under field conditions remains a serious problem.
Two research laboratories within the Corps have done a
limited amount of developmental work with water quality sensors.
The Cold Regions Research and Engineering Laboratory in Hanover"
New Hampshire, is presently conducting a study on the use of an'
airborne spectoradiometer to measure turbidity in water bodies as
well as studies to measure and monitor soil moisture under field
conditions. The Waterways Experiment Station at Vicksburg,
Mississippi, has recently completed a comparative study of
several commercially available sensor packages to determine
suitability for field application. Several Corps field offices
are also engaged in a variety of applications efforts.
Future sensor requirements are very difficult to predict.
There is a definite need to improve those sensors which are
presently available in order to increase reliability and this
should be the first priority. Work is also needed in the area of
standardization of signal output from sensor packages. The
growth of interest in sensors over the last several years should
provide the impetus needed for development of improved instrumen-
tation.
All Federal agencies having a responsibility in the water
quality area have an interest in water quality sensor de-
velopment. Many of these agencies have on-going programs in
sensor development and/or application. However, very little op-
portunity now exists for coordination of activities and sharing
of experiences. This workshop is a good beginning but unless
some form of formal coordination is established among the various
agencies, the momentum may be lost.
48
-------
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
by Nelson L. Milder
National Aeronautics and Space Administration
Washington, D.C. 20546
From its earliest days, the space program has been concerned
with the problems of water quality measurement and control. In
1971, the NAS-NRC Committee on Toxicology appointed a Panel on
Water Quality in Manned Spacecraft. Its charter included the
performance of studies and the submission of recommendations on
both potable and nonpotable water contamination problems.
Several studies were undertaken to investigate integrated
water and waste management systems for future spacecraft. One
such study, performed by the General Electric Company, led to the
development of a prototype system that served a multitude of
functions ranging from waste collection and disposal to water
reclamation and purification.
The potable water supply system developed for Skylab mis-
sions incorporated an inflight iodine bactericide monitoring and
control unit. This compact subsystem consisted of a water/
iodine concentration sampler, color comparator, reagent container
assembly i waste sample container and iodine injector.
There are numerous other situations within the NASA program
that have called for unique technological approaches designed to
deal with the special water quality control problem encountered
in the closed environment of a spacecraft.
But there is another side to the NASA interest. Embodied in
the words of the legislation that created NASA in 1958, is a man-
date that binds us to "the establishment of long range studies of
the potential benefits to be gained from the utilization of aero-
nautical and space activities for peaceful and scientific
purposes." In addition, we are required to apply these tech-
nologies "to the conduct of peaceful activities within and out-
side the atmosphere."
It is this aspect of the NASA program that is of interest
today. It is this aspect that has brought us here from NASA Ames
in California, from Langley in Virginia, Lewis in Cleveland,
Ohio, and from Headquarters.
The work we will be discussing with you during tue week
comes under the auspices of our Office of Space and Terrestrial
Applications .
As we are all aware, the environment for Federally supported
R&D has changed markedly over the past few years. Public
scrutiny has increased, and we are being asked at an ever
49
-------
increasing frequency to justify the commitment of public re-
sources to research that is often characterized by laymen as es-
oteric, and unrelated to the real needs of society. As with
other R&D organizations/ NASA has had to face more stringent re-
views of its R&D budget. We have accepted the challenge and,
since the launching of the last Apollo flight, have strived to
develop a strong program directed towards applying space and
space technology to the problems and needs of mankind. Also,
each year a certain amount of these resources is used on programs
aimed at utilizing NASA's competence as well as its technology in
dealing with needs that are not directly related to the aerospace
arena.
We lump these latter activities into a single classifi-
cation. We call them Technology Transfer projects. Last year,
we allocated approximately 15 to 20 million dollars to such
projects. More than 5 million of this amount was used in provid-
ing information about NASA's technology to prospective and active
users of this technology. The balance was used on projects to
implement, modify, and develop technology to meet specific user
needs.
I would suspect that many of you, or certainly the agencies
you represent, have been involved at one time or another with
this NASA program.
In Situ Water Quality Sensors
I have tried to present, in capsulized form, the
raison d'etre for NASA's involvement in the field of in situ
water quality measurement. It represents one aspect of a broad
program in which NASA, with the help of outside users, has de-
termined that the agency's technical competence can be used to
the advantage of the non-aerospace community.
I shall return to the subject of technology transfer later
in my talk. But right now, we need to deal with the topic at
hand. What are NASA projects in the area of in-situ water qual-
ity sensor development?
I should begin by describing some activities undertaken as
part of a long-standing joint NASA/EPA Interagency Agreement.
Since 1972 a cooperative program has existed between the EPA
Office of Research and Development and NASA. Under this Agree-
ment, we have been evaluating remote sensing techniques such as
four-color laser systems, multi-spectral scanners, multispectral
photography, dual differential radiometer, microwave radiometers,
thermal imagers, and infrafed photography for water quality mon-
itoring. As a necessary part of this program, numerous field
operations which required water sampling and in-situ water qual-
ity measurements have been carried out—primarily along the East
coast in coastal and estuarine waters. Usually in situ measure-
ment capabilities of other agencies (EPA, Corps of Engineers,
50
-------
NOAA, and various university groups) are depended upon in the
coordinated field test program requiring complementary remote and
ground truth measurements of water quality. In general, com-
mercially available sensors are used for the limited in situ data
obtained. Most of the measurements of interest still require
water samples to be taken back to the laboratory for anlaysis.
or example, on a recent water quality field experiment in the
James River near Hope well, Virginia, of the 25 parameters of
interest, only two were measured in situ. These were temperature
any secci depth. of the other 23 only 5 were considered measur-
able with commercial in situ sensors. Dif f iciencies of in situ
sensors include high maintenance, inadequate sensitivities, un-
or unaccepted standards, high operator skill requirements,
snort lifetime, and, sometimes, cost.
Responding to needs cited by EPA Region V, the EPA Monitor-
g^g Operations Division, Las Vegas, Nevada, requested NASA as-
istance in devising and implementing small portable water qual-
Y monitoring systems for use by all EPA regions. This request
occurred in 1976.
their needs, the EPA envisioned modular design concepts,
growth potential, utilizing NASA systems integration tech-
y along with potentially low-cost microelectronics for use
n on-board processing and data transmission components and tech-
iques. NASA is interested in this project because of its poten-
lal benefit in providing low-cost automated in situ ground truth
Rations for calibration and validation of remotely-sensed en-
ironmental data. Under the resulting interagency agreement,
ASA will demonstrate a one-person deployable water quality mon-
toring system. The package will accept and store data from a
arge variety of in situ sensors and can take water samples for
^ater laboratory analysis. The data will be stored in a non-
J-atile electronic memory system utilizing NASA technology. The
ystem can be deployed and retrieved from a small Bell 47
ellc°Pter (2 or 3 place) or from a small boat.
w Also, under the same agreement, a small, helicopter-borne
ater-quaiity monitoring probe is being developed by using a com-
J-nation of basic in situ water quality sensors and physical sam-
P-Le collector technology. The probe is a lightweight system
Cn can be carried and operated by one person as a passenger in
* small helicopter typically available by rental at commercial
irports. Real-time measurements are made by suspending the
ater quality monitoring package with a cable from the hovering
elicopter. Designed primarily for use in rapidly assessing
azardous material spills in inland and coastal zone water
°aies, the system can survey as many as 20 data stations that
re up to 1>5 kilometers apart and it can accomplish this in 1
°'-lr' The system provides several channels of sensor data and
«J-lows for the addition of future sensors. It will also collect
samples from selected sites with sample collection on command.
51
-------
An EPA Spill Response Team member can easily transport, deploy,
and operate the water quality monitoring package to determine the
distribution, movement, and concentration of the spilled material
in the water body. Test flights with the helicopter-borne
water-quality monitoring system have shown that it has great
promise for rapidly monitoring hazardous material spills, and
collecting other data.
Another project that has received support from
EPA—including this Laboratory—involves the application of an
electrochemical detection method to the monitoring of fecal
coliform concentrations in water.
For the past 3 years, research has been underway at the
Langley Research Center to develop an automated in situ coliform
monitoring system. The electrochemical sensor used to detect
coliforms was developed by Wilkins and Stoner. The technique was
based on measuring the time of hydrogen evolution as a function
of inoculum size. A linear relationship was established for this
parameter. A prototype system for monitoring levels of fecal
coliforms was evaluated in both fresh water and estuarine water.
These evaluations were performed in conjunction with EPA Region*
II. Thirty-two samples were processed by the in situ elec-
trochemical system during the evaluation. Correlation with
laboratory standard methods for determining fecal coliforms was
good for 14 samples, fair for 8 samples, poor for 6 samples, and
undetermined for 4 samples due to faulty electrodes and insuf-
ficient laboratory data. As a note of explanation for these
qualitative judgements, 'good' indicated that the laboratory data
were within the range of the predicted values, 'fair' results had.
differences by a factor of 10 to 100, and 'poor' results had dif-
ferences greater than 1000.
Very early in the development of the in situ monitoring
system, an interagency agreement was developed between EPA (Re-
gion II) and NASA. Four tasks were delineated. One of these
tasks detailed the conditions for a cooperative field evaluation
of the in situ system at Caven Point, New Jersey. This agreement
proved to be an excellent arrangement between the systems de-
velopment agency, NASA, and the user agency, EPA. It provided
NASA with the users requirements and at the same time afforded
EPA the opportunity to become familiar with the system.
In another project, we have been developing a system that
utliizes NASA's experience in data acquisition and processing
techniques as applied to continuous, near real-time water quality
monitoring and control. It has been directed towards dem-
onstrating the feasibility of the concept of continuous, on line
monitoring.
In 1971, NASA began a special study applying the technical
approaches and design philosophies of closed systems to the
52
-------
utility system of a 500-dwelling-unit garden apartment complex.
These studies led to an interagency agreement between NASA and
HUD (Department of Housing and Urban Development) relating to the
design and development of a MIUS (Modular Integrated Utility
System). Basic functions to be supplied by the MIUS included
integration of electrical power generation, water processing/
solid and liquid waste management, and environmental conditions,
and also the use of residual energy for utility functions. Only
production hardware items were to be incorporated in designs and
no building system changes were to be made. NASA's support of
this integration and test program resulted in a project to
develop an automated water quality monitoring and control system
to insure the safety and quality of treated wastewater so as to
permit its reuse (for non-potable purposes) in systems such as
the MIUS.
Our studies of water supply planning and wastewater rec-
lamation activities indicated a growing concern was motivating
responsible local, state, and national organizations to give
serious consideration to reuse applications. California, for ex-
ample, has begun numerous reclamation and reuse projects. The
most notable of these projects was the reuse of effluent from the
Lake Tahoe Tertiary Treatment Facility for recreational ap-
plication in the Indian Creek Reservoir. A study performed for
NASA further noted that many authorities in the water/wastewater
management field were convinced that the long-term solution to
the problem of insufficient water resources was in wastewater
reclamation and reuse. The City of Denver, Colorado, for
example, has been engaged in the planning for facilities that
will reclaim 100 mgd/day by 1986 for addition to the public water
supply. The City of Denver hopes to supply 2 percent of its
total water needs through the reuse by the year 2000. On the
basis of these studies, there appeared little doubt that
wastewater recycling and reuse in the domestic and civil sector
would become common practice in the United States and the world
in future years.
These study results, together with other NASA commitments,
resulted in a project to develop an automated water monitoring
system by applying space developed sensors and technology. The
basic concept should be applicable to all facets of water
monitoring—from natural water bodies to municipal drinking water
supplies to industrial and municipal wastewater.
The basic system has been installed by NASA in a converted
surplus radar van. Preliminary checkout was accomplished at the
City of Houston's southwest wastewater plant. The on-going field
test of the system is a joint demonstration test with Santa Clara
Valley Water District (SCVWD) at their experimental wastewater
facility. This facility was built with funding help from EPA and
53
-------
the State of California. It is located on the Palo Alto
wastewater treatment facility grounds. This field demonstration
test will evaluate the feasibility of the concept as a practical
means of monitoring water quality and thereby measure the effec-
tiveness of the wastewater treatment by this plant.
Although simple in principle, the actual system is complex
in fact. I must, therefore, defer its detailed description to
the Workshop sessions. Suffice to say, the system uses 18 com-
mercially available sensors to analyze 14 parameters. in ad-
dition, the system incorporates some NASA developed sensors to
monitor bacteria and trace organics. In addition, we are ex-
ploring some other unique approaches to monitoring biomass and
viral contamination in water. These are not wet chemistry ap-
proaches but rather are designed to rapidly provide an electronic
signal that measures concentration. This signal can then be
readily digitized for storage and processing.
Other instruments developed within the NASA program for
water quality determination are included in the Appendices to
this narrative. Finally, I will make mention of some fascinat-
ing research that we have been supporting, under a university
grant, that is aimed at developing miniature, self-contained,
highly reliable in situ water quality sensors. This work,
performed under the auspices of Professor C.C. Liu of the Uni-
versity of Pittsburgh, is based upon small scale-electrochemical
and immobilized enzyme concepts. To date, successful development
of a phenol sensor as well as miniature pO and pH sensors
has been accomplished. Dr. Liu's work is now being directed
towards the application of enzymatic reactions to the monitoring
of trace metal ions in water. This work originated as part of a
NASA program to develop an in situ water monitoring system for
use by EPA Region V for their oversight of the Great Lakes Basin.
This research has continued, although the proposed joint program*
was aborted.
Conclusion
This then briefly summarizes the NASA activities that we
will be discussing with you through the course of this week. I
suspect they may not be the only projects in water monitoring
being undertaken within the agency. One of the principal
activities for the coming.year will be to try to coordinate and
integrate NASA's programs in water quality assessment. That, in
fact, is one of the main reasons for our rather heavy attendance
at this Workshop.
I hope you recognized that laced through all our projects in
this area is the close interrelationship with some identified
users of the technology. I am again returning to the importance
of the transfer aspects of our program. Our objective here is to
develop applications of space technology and the capabilities of
54
-------
NASA's R&D organization to non-aerospace problems. In the coining
years we hope to strengthen our efforts in this area. But the
true measure of our success depends upon the cooperation and
support that our endeavors receive from you, the eventual
implementers and users of this advanced technology.
NASA has no chartered responsibilities in water resources
management but we have a deep and abiding desire to apply what we
have learned in our aerospace missions and what we have built—an
R&D institution designed to accomplish difficult technological
objectives of this important national concern.
55
-------
In addition to these summary remarks for verbal presentation
at the Workshop, the following Appendices are attached for inclu-
sion in the Workshop report to provide additional detail on NASA
water quality sensor development activities:
Appendix A - Automated Water Monitoring System
Appendix B - Electrochemical Method to Detect Coliform
Bacteria
Appendix C - Miniaturized Electrochemical and Immobilized
Enzyme Sensors for In-Situ Water Quality
Measurement
Appendix D - Other Activities within NASA
The Instrumental Detection of Viruses
A Neutron Gamma Ray Detection and Monitoring
System
A Thermal Plume Monitoring System
Acoustic Tracking of Woodhead Sea Bed Drifters
A Portable X-Ray Fluorescence Spectrometer
An Implantable Acoustic-Beacon Automatic Fish
Tracking System
56
-------
APPENDIX A. AUTOMATED WATER MONITORING SYSTEM
This project has been conceived to demonstrate the feasi-
bility of real-time or near real-time automated analysis of
overall water quality by applying space-developed sensors and
technology. The basic concept should be applicable to all facets
of water monitoring from reservoirs to municipal drinking water
supplies to industrial and municipal wastewater treated
effluents.
The basic laboratory has been installed by NASA in a
converted surplus radar van. Preliminary checkout of the basic
laborabory was accomplished at the City of Houston's Southwest
wastewater plant. The on-going field test of the laboratory is a
joint demonstration test with Santa Clara Valley Water District
(SCVWD) at their experimental wastewater facility build with
funding help from EPA and the State of California. It is located
on the Palo Alto wastewater treatment facility grounds. This
field demonstration test will evaluate the feasibility of the
concept as a practical means of monitoring water quality and thus
the effectiveness of the wastewater treatment by this plant.
As stated above, the SCVWD plant is an experimental plant
and thus equipped with several different waste treatment
processes including lime settling, filtering, activated carbon
treatment and ozonation. Furthermore, the influent to this plant
is the effluent from the City of Palo Alto wastewater treatment
facility (basic sludge plant), and the primary purpose of the
experimental plant is to evaluate the different wastewater
treatment methods to determine their effectiveness in producing
water that can be reused (recycled water).
Santa Clara Valley Water District (SCVWD) plans to use half
the plant output of four million gallons a day for injection into
the underground aquifers of the lower San Francisco Bay to see if
the intrusion of the saline water from the Bay into these
aquifers can be slowed, stopped, and eventually reversed. The
other half of the plant output will be sold for use as irrigation
water. First customers will be the City of Palo Alto and
Mountain View for irrigating a golf course and Baylands Park,
respectively. There is a good possibility that both NASA and the
Navy may also be using this water for irrigation and washing
aircraft in the near future. Thus, knowing the quality of the
wastewater in real time is important, and documentation becomes
imperative for future reference.
57
-------
Figure A-l shows the flow through the SCVWD experimental
plant schematically and also shows the locations in the
wastewater processing where continuous samples are drawn into the
WMS trailer for automated water quality analysis. Figure A-2
shows schematically how the water samples are conditioned and
distributed to the sensors on the WMS trailer.
The laboratory developed by NASA uses 18 commercially
available sensors to analyze 14 parameters to support the NASA-
developed biological and trace organic sensors. These commercial
sensors have been reworked for automated operation and measure
such parameters as: total oxygen demand, total organic carbon,
pH, turbidity and temperature. The bacterial sensors developed
by NASA are also automated and are capable of reducing the
analysis time required from days to hours; they include a sensor
for fecal coliform total bacteria, and total living bacteria. A
trace organic sensor using twin gas chromatograph columns has
been adopted from an instrument specifically developed for NASA's
Viking Project. This instrument will be used to detect suspected
carcinogenic contaminants in water. These sensors are expected
to provide SCVWD with data which will help them evaluate
operational efficiency of their plant. All activities of the
trailer, from sampling to sensor calibration and report
generation, are computer controlled.
Twin NOVA 1200 mini-computers are used for the above
activities. (Figure A-3).
On the following pages, the sensors are described followed
by summaries of operational problems that we have encountered to
date and finally some typical daily results are shown.
SENSORS
The sensors are grouped into the general categories, commer-
cial and NASA. The group of commercial sensors is described ""
first.
COMMERCIAL SENSORS
This group of sensors was obtained from government surplus
property or bought off-the-shelf from a commercial supplier. in
each case the sensors and their support hardware and electronics
were reworked to make them compatible for automated operation. AS
stated above some of the sensors and packaging were obtained from
surplus property and were not the optimum or latest state-of-the-
art sensors for the application. Consequently, as with the total
nitrogen sensor, problems have been encountered with reliability
and we have decided to replace them as they fail. y
58
-------
Total Organic Carbon (TOC)/Total Oxygen Demand (TOD)
The Astro Ecology Corporation analyzer simultaneously deter-
mines TOD and either total carbon or TOC on aqueous samples
containing solid particulates up to 2000 micrometers in diameter.
Dual measurement ranges of 0-10 ppm or 0-500 ppm of carbon and 0-
1000 ppm of TOD are available.
Fifty ml/minute of sample is continuously pumped into a
mixing chamber and gas-scrubber assembly where it is mixed with
hydrochloric acid to reduce the pH to 3. A portion of the
scrubbed, carbonate-free sample (4 ml/min.) is pumped into a
combustion chamber and combined with a metered air stream (79%
oxygen). The air provides the oxygen for combustion. The sample
remains in the chamber at 850°C long enough for full combustion
to occur. The products of combustion and excess air leave the
reactor and enter a water-cooled liquid/gas separation assembly
which removes condensable vapors. The remaining gases are routed
to an infrared analyzer where the amount of carbon dioxide is
measured and converted to units of TOC.
THE TOD analyzer determines oxygen demand for hydrogen,
nitrogen, sulfur, and carbon compounds found in the sample water.
It receives the noncondensable gases from the TOC analyzer
following the carbon dioxide analysis and passes them through a
solid electrolyte oxygen detector. The oxygen depletion, based
on the amount of air fed to the reactor, is measured and
translated into units of TOD.
Total Nitrogen (TN)*
The IBC/Berkeley Nitrogen Analyzer receives the non-
condensable combustion gases from the TOD analyzer and determines
the concentration of nitric oxide, by measurement of the poten-
tial between two electrodes. During the combustion at 850°C,
all nitrogen in the sample has been converted to nitric oxide,
thus a total nitrogen reading is provided by the instrument in
the range of 10 to 10,000 ppm nitrogen.
Hardness
An Orion Model 1132 Hardness Analyzer continuously monitors
the sample stream for hardness, a measure of calcium and
magnesium ions in water. The technique used is proprietary to
Orion Instrument Company. However, in general terms, it involves
the chelation of all devalent ions by a complexing agent,
*Note: This unit has been worn out and several models of
nitrogen analyzers are currently being evaluated as
a replacement.
59
-------
followed by the addition of a "substitution" ion which selec-
tively releases calcium ions. A "tag" ion is added at molar
concentrations 100 times greater than the maximum possible ionic
strength. A reference electrode is selective to this tag ion,
and the sensing electrode is selective to the substitution ion.
The electrodes are connected to the analyzer's electronics
system, which gives a direct hardness reading on a 4-cycle
logarithmic scale calibrated to read from 0.1 to 1000 mg/1.
Nitrate
A Delta Scientific Model 8138-153105-002XX1 Nitrate Analyzer
continuously measures nitrate/nitrite concentrations in the
sample stream by spectrophotometric analysis. Nitrates are
reduced to nitrites in a cadmium reducing column. The nitrates
are then reacted with sulfanilamide and N-l Napathy-
Ethylendiamine Hydrochloride (NEDA) in an acid solution to form
the azo dye. The color intensity developed is a measure of the
nitrate plus nitrite concentration in the sample. The concen-
tration of nitrites may be determined separately by bypassing the
cadmium column. Nitrate/nitrite concentrations above 0.4 ppm are
too dark for useful discrimination. Therefore, dilution of the
sample is required for most measurements.
£H
The Great Lakes Instruments (GLI) Model 70 Analyzer measures
pH for inprocess and pollution applications. The GLI probe uses
their Differential Electrode Technique to compare a pH electrode
to a standard electrode containing a chemical pH standard. The
probe is housed in a PVC tee (as described for the chloride ion
probe) to increase the flow velocity by the probe. An Orion
Model 401 specific ion meter provides a direct readout of pH
using an Orion 91-01 pH electrode and a Beckman 19033 Lazaran
process electrode. The pH electrode has a range of 2 to 11 pH
units. Both electrodes are mounted in a small tank that provides
continuous monitoring capability.
Total Residual Chlorine
An Orion Model 1125 Chlorine analyzer measures the residual
chlorine in continuous samples by the potentiometric method, it
operates on the principle that chlorine will liberate free iodine
from potassium iodide solutions when the pH is 8 or lower. The
sample is mixed with a reagent and pumped through a reaction
heater and constant temperature analysis chamber, in the
chamber, the mixture passes between a sodium electrode and a
redox electrode. The electrodes are connected to the analyzer's
electronics system which gives a direct chlorine concentration
reading on a 4-cycle logarithmic scale calibrated to read from
0.1 to 1000 mg/1.
60
-------
Sod i um
The Beckman 194204 Model 9415 Sodium Ion Analyzer determines
the concentration of sodium in a sample stream by measuring the
potential between a Beckman 633951 Sodium Ion Electrode and a
Beckman 19604 reference electrode. The electrode potential is
directly proportional to the logarithm of the active sodium
concentration. The response of the sodium electrode can be
affected by several other monovalent cations. Usually, hydrogen
is the only interfering ion encountered. To eliminate this
interference, all solutions are pH adjusted with ammonia to
suppress the hydrogen ion concentration. Temperature control is
provided by a heat exchanger located upstream of the electrode
flow chamber.
Temperature
Two Action Pac Resistance Thermal Detectors (RTD) are used
to measure sample temperatures. The RTD probe detects changes in
potential between two electrodes as the temperature changes and
converts this to a 0 to 5 volt signal. The probe is sensitive to
0.1°F and reads from 0 to 200°F.
Turbidity
A Sigrist Photometer Turbidimeter Model UP 52-TJ determines
the turbidity of a continuous sample stream by comparison with a
nephalometric standard. The Model UP 52-TJ has four measuring
attachments of different ranges. Two of the units use
falling-stream flow cells with ranges of 2 to 1000 and 2000 to
15,000 Jackson Turbidity Units (JTU); one uses a surface
scatter-flow cell with a range of 5 to 100 JTU, and one uses a
splash-flow cell with a range of 0.5 to 20 JTU. All of them use
a dual beam optical measure bridge. In the DAS, the units of
measurement are converted to mg/1
The Honeywell Model 551201-02-01 Turbidity sensor has an
optical head and a sampling tank for on-line analysis. A beam of
light is focused by a lens system down into a falling sample. A
photocell detects the light reflected by particles in suspen-
sion and supplies a proportional millivoltage to the signal
conditioner. The sensor has a nominal range of 0 to 25 Formazin
turbidity Units (FTU).
Ammon i a
A Delta Scientific Model 8119 Ammonia Analyzer continuously
measures ammonia in sample water by spectrophotometric analysis.
The intensity of the blue color developed by the reaction of
ammonia with phenol and hypochlorite in alkaline medium is
proportional to the concentration of ammonia in the sample. The
analyzer has a minimum sensitivity of 0 to 1 ppm ammonia, and no
upper limit.
61
-------
Chloride
A Great Lakes Instrument (GLI) Chloride Ion Probe measures
chloride by GLI's patented Differential Electrode Technique. The
GLI approach compares an Orion Model 94-17 solid-state chloride
electrode to a standard electrode containing a chemical chloride
ion standard. The probe is housed in a 1 1/2-inch PVC tee that
has been modified to reduce interior volume and provide a rapid
flow past the probe. Before reaching the probe, the sample is
mixed 50:1 with an ionic strength/pH adjuster. The probe has a
sensitivity range of 10° to 5 x 10-> moles per liter of chlo-
ride ion. It is relatively free from interferences except for
sulphur, bromide, iodide, and cyanide.
Conductivity
The sensor used to measure the ionic content of the water
sample is a Beckman Type R15 Solu Bridge Conductivity Indicator
with a temperature-compensated epoxy flow-through cell, type
CEL-VDJ4-KF. The cell constant is 4.0 permitting measurements in
the range of 0 to 2000 umhos/cm.
Dissolved Oxygen (DO)
A Delta Scientific Series 8310 Automatic Analyzer con-
tinuously measures dissolved oxygen in a sample stream with a
tank-mounted DO probe. The patented probe consists of gold and
silver electrodes mounted in a PVC body. A teflon membrane forms
an oxygen-permeable barrier between the water being tested and
the electrolyte in the probe. A voltage applied across the
electrode and as oxygen passes proportional to its concentration
DO readings in mg/1 and ppm in the ranges of 0 to 2, 0 to 10, or"
1 to 20 are displayed.
The Honeywell Model 551011-00-01 dissolved oxygen sensor is
a polarographic transducer consisting of a gold-silver-platinum
alloy cathode and a silver/silver chloride anode immersed in an
electrolyte cell. Its operation is similar to that of the Delta
Scientific instrument.
NASA DEVELOPED SENSORS
The following four sensors were developed for water quality
monitoring based on concepts developed by NASA in its space
program.
Biosensor, Luminol-Carbon Monoxide
This sensor was developed for continuous real-time detection
of total viable and non-viable biomass in water. The sensor
employs the chemiluminescent luminol-hydrogen peroxide reaction
with bacterial iron porphyrins and the carbon monoxide method to
62
-------
differentiate live from dead bacteria. Bacterial measurements in
water samples using this sensor are accomplished within an hour.
Essential hardware for this sensor consists of a photometer for
measuring chemiluminescence, a peristaltic pump for proportioning
sample and reagents and valves for sample selection and carbon
monoxide treatment.
The sensor interfaces with the WMS deionized water system,
100 psig air, 28 VDC power, 120 VAC power, wastewater drain
lines, CO gas line and the computer system. The sensor inter-
faces with the computer system for automatic control of the
sensor operation schedule, continuous measurements of photo-
metric cell output voltage, calculation of organism measured, and
storage and retrieval of sensor results.
Coliform Sensor
The coliform sensor detects presence of total and/or fecal
coliform organisms. It operates according to a "batch" process
schedule with up to eight incubator cells in operation. The
"batch" operation includes a four-phase operation: a) incubator
cell cleanup, b) nutrient fill, c) inoculation and d) total
and/or fecal coliform organism growth. The sensor interfaces
with the WMS facility for deionized water, 100 psig air, 28 VDC,
120 VAC power, wastewater drain lines. The computer system
provides automatic control of the sensor process schedule,
monitoring and measuring individual incubator cell performance.
In addition, it calculates total and/or fecal coliform organisms
present in inoculum for each incubator cell and provides storage
and printout of sensor results.
A second coliform sensor using the concept of electrical
impedance changes in the selective incubation nutrient solution
resulting from growth of fecal coliform bacteria has been
developed. It is in final checkout along with parallel efforts
to shorten the detection time via the use of optimum frequency.
Trace Organic-Gas Chromatograph
An existing gas chromatographic method has been adapted for
the rapid in situ analysis of chlorinated organic compounds at
the part-per-billion (1 ug per liter) level in water. This
method uses two gas chromatographs (GC) connected in series
through a system of valves and traps. The system operates
automatically and small samples of water (120 microliters) are
injected directly for analysis. The sample is injected through a
sample valve into a preparative GC for separation of organic
compounds from water. The organics are collected on a Tenax GC
trap which is heated to transfer the organics to the analytical
GC. In the analytical GC the mixture of trace organic compounds
is separated and detected by an electron capture detector. The
system is also equipped with a flame ionization detector for
monitoring aromatic compounds such as benzene and chlorobenzene.
63
-------
Viral Detector
This unit is in the development phase and various components
are being built for assembly into a prototype instrument. The
instrument is based on dyeing the virus and exciting the
fluorescence of the dye-viral complex with a Laser beam.
Initial work showed ethidium bromide, a high quantum yield
fluorescent dye combines with various types of enteric (polio and
influenza) and bacterial viruses. The fluorescent signals of
the dye-virus complex are enhanced 20 to 50 fold. The number of
dye molecules bound and the fluorescence emitted per virus was
found to be dependent on the number of base-paired regions of the
viral nucleic acid. Viruses possessing either a doubled stranded
DNA or RNA usually incorporated more dye molecules than single
stranded viruses. However, even small single stranded viruses
like polio or MS2 were found to incorporate enough dye molecules
(400 virions) to be experimentally detectable with current
photodetection technology provided the background noise is kept
to a minimum. The fluorescent dye was found to be freely
permeable through the protein coat of most viruses and where
permeability problems existed, heating the virus for 30 seconds
in buffered solution at 70°C was sufficient to overcome this
barrier.
An argon excited flow microfluorometer with a digital
readout component for rapid determination of viral titer in
solution is under construction at the Block Engineering firm in
Boston, Massachusetts. The instrument is designed to detect
individual viral particles in a flow system containing less than
10 viruses/ml in a 10-minute observation time providing the
background noise is eliminated or kept to a minimum. The
laboratory instrument in its present state (without removal of
background signals) has measured T2 virus at a concentration of
105/ml in a 10-minute counting time but its ultimate capability
still has not been determined. Interference by background noise
is still a problem but hopefully can be solved soon. The
instrument is scheduled for delivery, assembly and testing in
Ames in February 1978. Further experimentation will be required
to optimize the instrument before it is integrated to a viral
concentrator and eventually assembled as a unit for automated
real time flow systems.
RESULTS
All sensor data for the entire month are stored on disc and
at the end of the month, the data are transferred to magnetic
tape. At this time or any time during the month, the data can be
retrieved and displayed on the video screen or as tabular
printout (Table A-l) or in graphical form (Figure A-4).
64
-------
Currently the monthly data are reported in graphical form as
daily averages. They are generated from the hourly averages
stored on disc for each sensor and for each sample source. The
hourly averages are summed and scanned to evaluate the daily
average within one standard deviation (dashed line). The lower
grid is used to report the hourly peak (dot-dash line) and the
time of day the hourly peak occurred (numerals). These data are
"validated" data, i.e., bad data have been purged from the hourly
average prior to plotting. No other attempt has been made to
alter the recorded results.
Table A-l shows typical daily results for all the sensors
onboard the WMS trailer with the exception of temperature, total
oxygen demand, total nitrogen, sodium, chloride ion, dissolved
oxygen, hardness and nitrate. These sensors were not operated on
the 4 days used in this example. Only the TOD sensor, due to
overheating of its electronics, and the worn out total nitrogen
sensor were nonoperational. The results show that significant
reductions in the monitored parameters do occur in the treatment
process. For example, biomass reductions were on the order of 95
percent; turbidity reduction was about 28 percent and chloro-
organics about 87 percent. The data indicate only nominal
reductions in non-volatile organics (TOC) and ammonia during this
period because the ammonia stripping and carbon absorber treat-
ment processes in the plant were not activated. These results
indicate how the data from the automated water quality monitoring
system can be used to monitor the operational effectiveness of
water treatment processes in a wastewater plant.
Figure A-4 shows the graphical monthly data format for
dissolved oxygen. This is for the SCVWD reclamation facility "B"
effluent water sample station 6 (see Figure A-l). Similar
dissolved oxygen plots for any of the other five sample stations
are also available along with similar plots for all other
sensors.
The plots in the upper graph show the daily average reading
(solid line) bounded by plus or minus one standard deviation
(broken lines). The dispersal of the one standard deviation
lines gives an indication of the fluctuation in the sample
through the day. The lower graph shows the maximum hourly
average peak for each day and provides a measure of how extreme
the day to day maximum sample values were and, when compared to
the daily average value,, indicates the possible extreme variation
in the water quality for that day.
65
-------
SENSOR SHORTCOMINGS AND TYPICAL PROBLEM AREAS
Two modes of failure interrupt the continuous operation of
the system. The first most common failure is computer malfunc-
tion due to either overheating (corrected by installing a blower
in the chassis) or from running out of storage space. Installa-
tion of a larger computer would solve the latter problem. The
second type of failure involves the sensor electronics. These
failures occur primarily because the system is a prototype and
many continuing changes are made. Also, surplus equipment was
heavily used to minimize overall cost. These failures have not
been serious and certainly are not expected with an operational
system.
COMMERCIAL SENSORS
Probably the greatest problem has been that the commercial
sensors used, or their associated electronics and hardware, were
not engineered for continuous, automated, unattended operation
and, thus, the reliability of the overall sensor packages tends
to be poor for this application. The installation on the WMS
trailer has experienced difficulties such as drift in the zero
baseline of the TOD analyzer due to the sensor manufacturer
placing the electronics in a high temperature environment,
drifting of the chloride sensor and failure of the total nitrogen
sensor. Other problems that have been encountered are contami-
nation of the sensing elements and clogging in the feed lines to
the sensors, gasket and pump impeller failures. Seal failures
have been fairly frequent occurrences.
NASA DEVELOPED SENSORS
Biosensor, Luminol-Carbon Monoxide
This sensor is a first prototype and as such has operated
very well. There are certain limitations, however. in
particular, the system measures total living biomass and does not
distinguish between dividing and non-dividing cells. Thus, the
results do not correlate with standard plate counts. Also' in
order to avoid clogging in the sample tubes (0.4 mm diameter)
selective filtration is used which results in those clumps of'
bacteria, larger than the filter size, being removed from the
sample. This lowers the sample bacteria level. A survey of the
literature shows no acceptable methods are known for breaking UD
the clumps without affecting the actual bacterial colony. Some
problems with retained contamination, when too weak a cleansinq
reagent is used or solution is used, remain to be solved. These
are important but not necessarily debilitating problems.
66
-------
Coliform
This sensor for the present is not completely automated;
that is, the sample for the sensor is not automatically drawn
from the overall sample stream. Contamination problems resulting
from residual material in the sample feed line have been
experienced. This hinders the automatic sampling procedure.
Improvements in this area and in general cell sterilization
procedures are being investigated. Some procedures which will
help improve this sensor performance include (1) the variation of
the sample replicates via better sampling and sample preparation
techniques, (2) neutralization of anticipated sample interferring
agents found in different water samples and (3) increased sensi-
tivity and reduced detection time for the sensor.
Trace Organics
Gas Chromatograph. The calibration methods and gas chroma-
tographic separations require further refinement. Preparation
methods for calibration mixtures need to be standardized in order
to produce accurate and repeatable standards. Columns that will
separate all of the chlorinated species of interest are also
required. In addition, it would be desirable to expand the capa-
bility of the method to include analyses of important compounds
such as benzene, toluene, chlorobenzene and dichlorobenzene which
are not detected because of their poor response in the electron
capture detector. The flame ionization detector has not worked
due to high background noise from column bleed. A new column or
more sensitive detector should solve this problem.
Future work would include decreasing the size of the system,
decreasing the analysis time and expanding the capability of the
system to analyze for other compounds.
67
-------
SUMMARY
This summarizes the status for the WMS system including a
brief historical background, results, and most important, the
problems encountered in the project.
In spite of the difficulties, the experiences to date
strongly suggest that the eventual success of this concept for
automated water quality monitoring is certain. It is expected
that by this time next year the current concept verification test
phase will be complete. Definite recommendations for operational
system design will be possible, and should include recommen-
dations in the areas of increased sensor reliability, life, and
sensitivity and operational procedures. Also, our partners in
the test, SCVWD, will have enough data to evaluate the concept's
utility for managing the operation of a wastewater reclamation
facility, and make decisions on instrumenting their facility with
similar instruments.
68
-------
PRESSURE
REGULATOR
(SMM.L ORANGE)
TOTAL OXYGEN
DEMAND
TOTAL ORGANIC
*» UNFILTEREO CARBON
TURBIDITY
DISSOLVED
OXYGEN
PLANT i
EFFLUENT »
FITTING FOR
PRESSURIZED |
""""" PRESSURIZED
REGULATOR
(ORANGE)
EFFLUEKT
pH
CHLORIDE
AKMONIA
10y NITRATE
FILTERED CONDUCTIVITY
(TEMPERATURE
BIOSENSOR
RESIDUAL
CHLORINE
HARDNESS
DISSOLVED
lltlCTI TCDCf. OXYGEH
WNFILTERED TURBIDITY
50y
FILTERED
50v
FILTERED TEMPERATURE
XD-PRESSURE TRANSDUCER
R/V-RELIEF VALVE
Figure A-2. Sample conditioning and distribution
-------
TAP WATER
COMPRESSED AIR
INFLUENT
EFFLUENT
MUUI fOINT
IFfUIINT
ymm&-
i
•fftWiiw-x^Xvrf
I
•
., .i..
. %fltTEREt>
v ,Yf*t , f . i
I
»
I
TRAILER
TOC/TOD/TN
•TURBIDITY
L'D.O.
AMMONIA
CHLORIDE
NITRATE
CONDUCTIVITY
TEMPERATURE
•BIOSENSORS
kSODIUM
f-TEMPERATURE
•DISSOLVED
OXYGEN
L'TURMOITY
CHLORINE
HARDNESS
I*
.CTERIA
:ONCtNTRATOt
$8
DAILY
REPORT
HOURLY
REPORT
INSTANTANEOUS
DATA
I
Figure A-3, Water monitoring system
-------
i PRUUIRY EFFLUENT
a RECLmimiON rnc. n EFFLUENT
3 SECONDDRY EFFLUENT
i» CLniUFtER EFFLUUNT
SNOT fiCTIVE
@RECLfli:nTION KOC, a EFFLUENT
20
18
16
14
12
-i I*
i e
G
2
0
20
18
16
12
o
z: 8
6
u,
2 .
0 •
7. DIS OXYGEN RECM-l 3R£T E D OCT. 21 19/7'
_L
1
r
-
•
_Jjs
L
~
•--
.;
I
i
• -•-"
-
5
on
HOI
i i
1 - •!•"} o '
. i
]
i
—
i
<
'
•»
/
V
A.^.
i
!
|
i
3
^
—
c
-
--
1
."-'
1 1
[LY RVERROE
JRLY PEflK
ipmn:
tTTT'Ti
1
J
S B
i
i
i
^
i
!
._
\L- "'
J
• -
-
-k
i
!
i
5
»
(
~T
.V
-
_^_
:
-
H
J
i
"
- 1
i
:
—
i
1
1
i
-
'«
~\
/
-
-
1
.1
jfi
A/#
K"
}-•
-
i
(
L
J
^i--^
:\ /
V
7i
f
1
| !
]
i
i
~~i
i
irazr
"^"-)-T
=H? —
til i
»•' *
—
—
__L
—
1
i i
•
--11 STRMD OEV
o a •» £
3
/
-
i.
. :
l
1
!
J_.l
'I"
,
i
._„!
__i
t
£
i
1
~i
: " T 7 T " ! "
-r
1
-•
~H h—
_
i
*-..~j.-
i ' ' • —
,
i l !
|
1 ' ' 1
1
! i
-
1 1
-t !
i !
u_
' f
i
1
! jj
DRY OF I'.OMTH
Figure A-4. NASA/WMS - SCVWD Palo Alto water reclamation facility
72
-------
TABLE A-l . TYPICAL DAILY DATA
Reclamation Facility
Viable Bionass, cells/ml x 10^
Total Biomass. cells/ml x 106
Fecal Coliform
t-Dichloroethylene, ppb
Chloroform + Methyl Chloroform, ppb
Trichloroethylene, ppb
Bromodichloromethane, ppb
Tetrachloroethylene , ppb
Turbidity, MG/L SiO2(25°)
Turbidity, MTU (90°)
Total Organic Carbon, ppm
AmnDnia, ppm
Dissolved Oxygen, ppm
Influent Avg.
+ Std. Dev.
97.5 + 42.7
177.6 + 62.3
-
341.4
750.9
109.7
4.2
70.4
25.9 + 5.5
7.2 + 1.4
47 + 4
40 + 5
3.7 + 0.4
Conf ig.
*
1
2
1
2
3
4
4
4
4
4
1
2
1
2
1
2
1
1
Effluent Avg.
+ Std. Dev.
4.5 + 1.9
0.3
7.0 + 3.3
2.8
0
41.1
43.9
18.2
0.2
8.6
18.4 + 3.4
15.1
5.7 + 1.9
1.2 + 0.1
43 + 5
27
31 + 2
9.1 + 0.4
%
Reduction
95.4
99.7
96.7
98.4
-
88.0
94.0
91.3
95.2
87.8
29.0
41.7
20.8
83.3
8.5
42.6
22.5
—
(Continued)
-------
TABLE A-l. (Continued)
Reclamation Facility
Total Residual Chlorine, ppm
pH
Conductivity, MHO/CM
Influent Avg.
+ Std. Dev.
-
7.1 +0.1
1505 + 23
Config.
*
1
1
1
2
Effluent Avg.
+ Std. Dev.
2.8^0.6
7.3 +0.2
1455 + 40
1375
%
Reduction
-
-
3.3
* 1 - 10-27-77, Flocculation, ammonia stripping, recarbonation, ozonization, chlorination
2 - 10-27-77, same as above with filtration
3 - 10-25-77, same as 1
4 - 10-19-77, same as 1
-------
APPENDIX B. ELECTROCHEMICAL METHOD TO DETECT COLIPORM BACTERIA
The Federal Water Pollution Control Act of 1972 provides NASA
with the charter to apply technology to monitor water pollution.
Specifically, section 104 of this act charges the Environmental
Protection Agency (EPA) to provide a water quality surveillance
system for monitoring the quality of the navigable waters,
groundwater, the contiguous zone, and the oceans. In order to aid
in providing such a system, the EPA is charged to utilize the
resources of NASA, to the extent practicable.
For the past 3 years research has been underway at the
Langley Research Center to develop an automated in situ coliform
monitoring system. The electrochemical sensor used to detect
coliforms* was developed by Wilkins and Stoner. (1) The technique
was based on measuring the time of hydrogen evolution as a
function of inoculum size. A linear relationship was established
for this parameter. A prototype system for monitoring levels of
fecal coliforms was evaluated in both fresh water and established
fecal coliforms. These evaluations were performed in conjunction
with EPA Region II. Thirty-two samples were processed by the in
situ electrochemical system during the evaluation. Correlation
with laboratory standard methods for determining fecal coliforms
was good for 14 samples, fair for 8 samples, poor for 6 samples,
and undetermined for 4 samples due to faulty electrodes and
insufficient laboratory data. As a note of explanation for these
qualitative judgements, 'good' indicated that the laboratory data
were within the range of the predicted values, 'fair' results had
differences by a factor of 10 to 100, and 'poor1 results had
differences greater than 1000.
One important aspect is the relationship between in situ
predicted coliform counts and the values obtained with conven-
tional laboratory techniques using the Most Probable Number (MPN)
of Membrane Filtration (MF) method. There is a fundamental
difference between the two methods, viz., the electrochemical
technique is based on a graded or 'time to response' relationship
while conventional procedures are quantal or all-or-none observa-
tions. (2) In addition to these differences, the eventual use of
the data from the two methods also differs. For example, MPN or
MF results are used to determine the sanitary quality of the water
*Coliforms are the principal biological indicators of the sanitary
quality of water.
75
-------
for consumption, bathing beaches, etc. Coliform values beyond
those prescribed by law result in condemnation procedures, in its
current state of development, the electrochemical method cannot be
viewed in the same light as the MPN or MF procedures. Rather, its
explicit application in the in situ system was for monitoring the
degree of contamination at the test site. On that basis, the
in situ response times were arbitrarily divided into the followina
groups depending on the degree of contamination: detection time
endpoints between 3 and 6.5 hours would be indicative of heavy
contamination (400 to 6,000/100 ml), response times in the range
of 7 to 10 hours would be moderate (25 to 250/100 ml), 10.5 to
14 hours would indicate light contamination (1 to 15/100 ml), and
over 14 hours would indicate fairly low levels of fecal coliform
contamination. The results obtained during the field evaluation
would tend to support these arbitrary designations. For example
base station response times at Caven Point, New Jersey, were '
approximately 6.5 hours which indicated heavy pollution. This was
in agreement with the mean laboratory fecal coliform counts of
633/100 ml. On the other hand, base station response times at the
York River deployment were extended (12 to 17 hours), indicating
light contamination. This was also in agreement with mean
laboratory fecal coliform values of 27/100 ml.
The concept of remotely monitoring the levels of fecal
coliforms with the electrochemical method was verified although
utility of the in situ system would be limited to specific
ecological situations. It appeared that the in situ concept would
be most useful in defining areas heavily contaminated with fecal
coliforms. Placed in clean areas suspected of receiving large
amounts of pollution, the system could be utilized as an 'early
warning1 of impending contamination. Based on the limited data
and experience obtained during four deployments, the in situ
sampler has to be viewed as a monitoring system rather than a
system to obtain precise measurements to meet regulatory require-
ments. ~~
Based on extensive experience, it was felt that the primary
reason for the 'poor1 and to some extent the 'fair1 in situ
results was due to the combination redox electrodes. AS described
by Wilkins et al. (3) the electronic sensing of an endpoint con-
sisted of detecting two levels of response, 30 and 90 millivolts
Inherent in this requirement is a steady baseline. Any drifting *
will cause an early 'triggering1 of the electronics resulting in
false positive response. In addition, if the required millivolt
response levels are not reached, the electronics will cycle in an
attempt to locate the proper levels. Numerous tests have failed t
identify the source of these anomalies. The questionable relia- °
bility of the combination electrodes precludes their use in futur
in situ systems. In an attempt to circumvent this problem area S
investigations are underway to develop electrodes of a simple '
design with high reliability. Stoner has shown that electrodes of
76
-------
similar metals could be used to detect bacteria providing one of
the electrodes was shielded from the organisms. (4) This work was
funded by EPA and monitored by NASA. Wilkins has demonstrated the
efficacy of platinum electrodes in which both electrodes could be
in contact with the bacteria providing that a surface area ratio
of four to one was maintained. (5) Future applications of the
in situ concept will incorporate one of these electrode designs.
The following general conclusions were reached after review-
ing the data obtained during four deployments of the in situ
coiiform monitoring system. There were a number of features
associated with the current design of the in situ system that
restricted its general utility. For example, in view of the size
and weight of the system, deployment and retrieval required
experienced 'riggers' and the use of a heavy duty, mobile crane.
The 'batch' design limited the number of samples that could be
processed to 10. If the unit was located in water temperatures of
< 8.0° C, the power requirements on the batteries to maintain a
water bath temperature of 44.5° C restricted deployment time to
less than 10 days. As discussed previously, the questionable
performance of the combination electrodes distracted from the
required reliability of the system. In view of these problem
areas associated with a 'batch' system, the most promising future
technology appears to reside with a flow-through system. The major
advantage of a flow-through system would be the continuous
processing of samples without the need for frequent refurbishment.
Studies currently underway at the Langley Research Center indicate
that daily samples could be processed for up to 6 months before
the unit would be retrieved.
Very early in the development of the in situ monitoring
system, an interagency agreement was developed between EPA (Region
II) and NASA. Four tasks were delineated. One of these tasks
detailed the conditions for a cooperative field evaluation of the
in situ system at Caven Point, New Jersey. This agreement proved
to be an excellent arrangement between the systems development
agency, NASA, and the user agency, EPA. It provided NASA with the
users requirements and at the same time afforded EPA the oppor-
tunity to become familiar with the system and its limitations. In
addition, the interagency agreement also helped to prevent any
possible duplication of research and development effort.
77
-------
REFERENCES
1. Wilkins, J. R.; Stoner, G.E.; and Boykin, E.H.:
Microbial Detection Method Based on Sensing Molecular
Hydrogen. Appl. Microbiol. 27:949 (May 1974).
2. Wilkins, J.R.; and Boykin, E.H.: Electrochemical Method
for Early Detection and Monitoring of Coliforms. Amer.
Water Works Association Journal. 68:257 (May 1976).
3. Wilkins, J.R.; Young, R.B.; and Boykin, E.H.: Multichannel
Electrochemical Microbiol Detection Unit. Accepted for
Publication in the January 1978 issue of Applied and
Environmental Microbiology.
4. Stoner, G.E.: Unpublished Data. (December 1977).
5. Wilkins, J.R.: Use of Platinum Electrodes for the Electro-
chemical Detection of Bacteria. Submitted for Publica-
tion in the Journal of Applied and Environmental Micro-
biology.
78
-------
APPENDIX C. MINIATURIZED ELECTROCHEMICAL AND IMMOBILIZED
ENZYME SENSORS FOR IN SITU WATER QUALITY MEASUREMENT
During the past three years, research and development on
sensors and techniques for in situ water quality monitoring was
undertaken in the Chemical and Petroleum Engineering Department at
the University of Pittsburgh under the sponsorship of NASA.*
Efforts were devoted to the research on advanced electrochemical
sensors and immobilized enzyme sensors for possible application to
in situ water quality monitoring.
The sensors used in commercial water quality monitoring
instruments, by and large, are all electrochemical sensors. For
instance, dissolved oxygen, pH value and electrical conductivity
detectors all employ electrochemical sensors. The measurements
of chloride ions, ammonia and other chemical and biological
parameters in water quality are also performed by electrochemical
techniques. Hence, it is appropriate to devote research efforts
to the development of electrochemical sensors for in situ water
quality monitoring.
In situ measurements suggest the need for instruments that
permit continuous, unattended monitoring of the essential
environmental parameters quantitatively. To a lesser extent,
in situ measurements also imply a mobile and portable instrument
which is capable of detecting and identifying the pollutants in a
quantitative manner. Hence, in situ water quality sensors should
have good sensitivity, durability, reliability, minimum power
consumption and small sample volume. In our laboratories,
miniature pO2, and pH sensors are developed based on the
premises that miniature sensors may have the advantages of small
sample volume and low power consumption. Consequently, they may
be both portable and suitable for long-term in situ quality
monitoring.
The operation of a miniature pO2 sensor is based upon the
polarographic principle. .The gold cathode has a diameter of 0.127
mm (5 mil) and the anode is a sintered silver-silver chloride
electrode, 2 mm in diameter and 3 mm long. A selective membrane
is applied to the cathode to minimize ionic interference. The
polarizing potential is 0.7 volt, and the limiting current, of the
*Work performed.under NASA Grant NSG-3002. Principal Investi-
gator, Chung-Chiun Liu, Professor, Chemical and Petroleum
Engineering Department, University of Pittsburgh.
79
-------
order of nanoamperes, is in direct proportion to the oxygen
dissolved in the medium (water).
A miniature glass pH electrode was developed, fabricated and
evaluated. Hydrogen ion-sensitive glass (Corning 015) is used in
constructing the electrode. The inner electrolyte is a 0.1 N HC1
solution and the inner reference electrode is an Ag/AgCl elec-
trode, 0.127 mm in diameter. The average tip thickness of the
electrode is 0.484 mm compared to a wall thickness of 0.25 mm. An
integral circuit field-ef feet-transistor (FET) input operational
amplifier is incorporated into the pH electrode to widen the
effective range of measureable pH values.
These miniature p02r pH sensors have been used in field
tests and have been incorporated into a portable water quality
monitoring system. Limited field testing has demonstrated the
feasibility of using such sensors for in situ water quality
monitoring .
In addition to the research efforts on advanced electro-
chemical sensors, our other research efforts are focused upon the
development of immobilized enzyme sensors. Many enzymes are
excellent catalysts for oxidation or reduction reactions for
biological species. The high degree of sensitivity and selec-
tivity of the enzyme in these reactions is impressive, and can be
applied to in situ water quality monitoring. However, in order t
use the enzyme repeatedly over a reasonably long period of time °
the enzyme will have to be immobilized either by physical entran-
ment or chemical bonding techniques. This immobilized enzyme on
substrate can then be placed in the testing medium. The en2ymat?
reaction, if coupled with a redox reaction of electrochemicallv
active species, will result in a measurable electrochemical
potential. This zero-current potential is Nernstian in nature
and exhibits a linear relation with respect to the logarithmic'
values of the concentration (activity) of the detecting substrat
Based upon this concept, immobilized enzyme sensors are beinq
developed in our laboratories. Efforts have been devoted to th
studies of enzyme immobilization techniques and applications of
enzyme sensors to in situ water quality monitoring. Specif icallv
a phenol sensor has been developed, and the possibilities of us?'
enzyme sensors for trace metal monitoring are being investigated9
The phenol sensor incorporates the enzyme phenol oxidase
(tyrosinase) , immobilized within a polyacrylamide gel, into an
electrochemical system for phenol detection. Coupled with the
oxidation of ferrocyanide ions to ferricyanide ions and usinq
potentiometric techniques, the phenol sensor is capable of
producing aero-current potentials that are directly proportional
to the logarithm of phenol concentration over the range of 3 ft
10-7 M to 1 x 10"4 M. Although this sensor does not provide" *
80
-------
the greatest sensitivity available to detect phenol and related
compounds, it does provide accurate and reproducible measure-
ments which can be made on a continuous basis. Furthermore,
introduction of the sensor into the text medium is a simple
process. Once operating, the sensor response time is rapid, on
the order of 10 minutes or less.
Biochemically, phenol can be oxidized by oxygen in the
presence of the enzyme tyrosinase, to 1, 2 dihydroxybenzene and
then to orthobenzoquinone. When the reaction proceeds in the
presence of potassium ferrocyanide, the ferrocyanide ions are
oxidized to ferricyanide ions with the concurrent reduction of the
orthobenzoquinone to 1, 2, dihydrozybenzene. Thus, the oxidation
potential of ferrocyanide ions can be used as an indication of the
phenol concentration.
The performance of this phenol sensor has been evaluated in
terms of its calibration characteristic, reproducibility, per-
formance affected by oxygen concentration, performance in the
presence of phenol analogs, etc. The actual response of the
phenol sensor has also been evaluated by using industrial
effluents. An arrangement was made with Mobay Chemical Co.,
Pittsburgh, Pennsylvania, to test their effluent samples. Mobay
Chemical took seven effluent samples and divided each sample into
portions. One portion of the sample was analyzed by Mobay using a
chemical colormetric method. The other portion of the sample was
analyzed in our laboratory using the immobilized phenol sensor.
We had no prior information on the phenol concentration in each
sample and our analytical results were reported to Mobay for
comparison. The comparative results show:
TABLE C-l. PHENOL CONCENTRATION, PPM
Sample No. Our Method Mobay
1 2
2 --*
3 __*
4 0.3
5 1.0
6 80.0
7 <3
6
0.005
0.027
0.48
5.25
87.5
0.01
*Undetectable
In general, the results of the immobilized enzyme phenol
sensor are very close to those obtained by Mobay. Also, the
analysis procedure is simple. We are in the process of develop-
ing a prototype sensor for in situ automated water (industrial
effluent) quality monitoring.
81
-------
Another research area on immobilized enzyme sensors concerns
the applications of enzymatic reactions for trace metal analysis.
Many enzymes have been known to be highly sensitive to heavy metal
ions. Analytical methods for determining trace metal ions based
upon their inhibition in enzyme reactions have been demonstrated.
In general, the initial rate of an enzyme reaction will decrease"
with increasing inhibitor concentration. The range of this linear
region depends on the nature of inhibition; that is, whether it is
reversible or irreversible.
Based upon this principle, it is feasible to use the inhibi-
tion effects of heavy metal ions on selected enzyme reactions as a
means for monitoring the heavy metal ions. It is anticipated that
the presence of other heavy metal ions may cause interference in
the detection. This interference may be alleviated by applying
sample separation techniques or masking procedures.
The enzyme used in this study is immobilized in polyacryl-
amide gel over a platinum screen, in order to increase its
stability and reuse. This enzyme-gel platinum matrix (sensor) is
then used in conjunction with a redox reaction of electro-
chemically active species. The zero-current electrochemical
potential produced due to this coupled redox reaction is used as
the measuring output of the detection. The rationale of choosing
this mode of operation is based on the successful development of
phenol sensors described above. Other measuring techniques such as
ampermetric measurement, pH value change, pC>2 consumption, etc.
can also be considered as an alternative mode of operation. *'
At present, the inhibition effects by heavy metal ions on
enzymatic reactions is studied using the xanthine oxidase system
This enzyme catalyzes the oxidation of hypoxanthine and xanthine"
The inhibition effects of heavy metal ions on this enzyme reactio
are currently being investigated as the means to quantify the
heavy metal ions presented. This approach is novel and also
requires very simple instrumentation. Consequently, this techniau
may be applicable to the in situ water quality monitoring.
In summary, our research efforts on sensors for in situ wate
quality monitoring are devoted to advanced electrochemical sensors
and immobilized enzyme sensors. We hope the development of
miniature, low power requirement sensors would permit long-term
durable and highly sensitive in situ monitoring. Also, the '
development of highly-selective enzyme sensors would lead to
minimum interference, simple instrumentation and highly-sensitive
detection of essential chemical and biological water quality
parameters. We believe that these research efforts would aid in
the sensor development for in situ water quality monitoring and
yield substantial results in the near future.
82
-------
APPENDIX D. OTHER ACTIVITIES WITHIN NASA
The Instrumental Detection of Viruses is being investigated
for EPA at the Jet Propulsion Laboratory for use in monitoring
environmental conditions at sewage processing facilities. A
laboratory facility has been established to develop the appro-
priate instrumentation and feasibility has been demonstrated.
Further developments are continuing to increase the effectiveness
of the system. The technology is based upon tunable accoustical-
optical filters derived from the space program.
A Neutron Gamma-Ray Detection and Monitoring System is being
developed at Goddard Space Plight Center for use in analysis of
the superficial layers of the sea bottom, water quality and water
pollution. A prototype has been demonstrated. A system is being
designed for use in surface and submarine vessels for evaluation
in the ocean and rivers in the summer 1978. The technology is
based upon that used in remote analysis of the lunar surface
physical properties. The radiated material yields a signature
which can be analyzed by a computer to identify the chemical
properties. This project is a joint effort with DOE, EPA, NOAA
and USGS.
A Thermal Plume Monitoring System is being developed for NOAA
at the Langley Research Center for use in giving temperature depth
profiles in thermal plumes from power plants and ocean thermal
movements. The acoustic bathythermograph was demonstrated and
redesigned to overcome deficiencies. Another demonstration is
planned in the spring 1978. The technology is based upon some
variations with temperature. The inexpensive and miniature
sensors permit deployment from ships of opportunity to map ocean
thermal profiles.
Acoustic Tracking of Woodhead Sea Red Drifters was developed
for NOAA at the Langley Research Center in 1975 on a demonstration
cruise in the New York Bight. It was concluded that shipboard
sonar tracking of acoustic woodhead sea bed drifters could provide
useful Lagrangian information on bottom water movement caused by
tidal and other nonstorm effects. A NASA Technical Note NASA TN
D-8392 has been published.
A Portable X-Ray Fluorescence Spectrometer is being developed
at the Langley Research Center for use in water pollution analysis
(trace metals), oil analysis (metal contaminants), energy resource
83
-------
exploration (uranium core analysis) and fuel contaminant analysis
(sulfur in coal). The first prototype is to be demonstrated in
August 1979 for use by the Bureau of Mines to assist in deter-
mining the bearing strength of mine walls, for obtaining higher
yields from ores in mining operations and to minimize sample
selection and analysis from field exploration in wilderness areas.
The design is a modification of that employed in the Viking space-
craft for analysis of Martian soils.
An Implantable Acoustic-Beacon Automatic Fish Tracking System
has been developed for the Virginia Institute of Marine Science at
the Langley Research Center. The system was demonstrated in the
York River where pingers were implanted in small fish and were
successfully tracked up to 2.5 km. No changes in either fish
behavior or pinger performance were observed. The system is com-
mercially available and provides an effective approach to under-
water tracking of small fish within a fixed area of interest. A
NASA Technical Note (NASA TN D-8498) has been published. The
purpose of this effort was to investigate the effects on fish of
man-made perturbations in estuarine waters.
84
-------
NATIONAL BUREAU OF STANDARDS
by William H. Kirchoff, Ph. D.
Office of Air and Water Measurement
National Bureau of Standards
Washington, D.C. 20234
The mission of the National Bureau of Standards is to
provide standards of measurement and means for making measure-
ments consistent with these standards and to provide physical
constants and properties of materials. To this end, the major
products of the NBS Water Measurement Program are measurement
standards, measurement methods and evaluated and improved data
for relating effluents to ambient water quality. Thus, in terms
of water quality sensors, NBS is concerned with the reliability
and accuracy of specific types of sensors, with the possible
development of new sensors and with the study of properties which
may have some application to sensor design. NBS does not have
monitoring responsibilities and hence must turn to the monitoring
community for guidance on measurement method needs.
As already mentioned, the major products of NBS efforts are
measurement standards, measurement methods and evaluated data for
describing the distribution of substances in water. The measure-
ment standards include standards for the measurement of chemical
constituents in water and the measurement of water flow.
Standards for the measurement of the chemical constituents
in water can be check samples for interlaboratory comparisons or
NBS Certified Standard Reference Materials (SRM's). SRM's are
materials whose selected properties have been accurately measured
and certified. They can be used to evaluate measurement methods
and can serve as a basis for validation, through carefully docu-
mented comparison of working, reference or calibration standards.
To date the NBS Office of Standard Reference Materials has issued
and sold mixed gamma-emitting radionuclide solution standards;
two mercury-in-water standards (one at the ppm concentration
level and one at the ppb concentration level); a multielement
standard consisting of 16 elements in the concentration range of
1 to 100 ppb with sodium, potassium, magnesium, and calcium added
at the ppm level to simulate a natural fresh water sample; and a
sediment standard certified for the radioactivity. A sediment
standard certified for trace element composition as well as total
Kjeldahl nitrogen, phosphorous, total organic carbon and loss on
ignition will soon be offered. Check samples are generally de-
veloped for other Federal Agencies for the purpose of testing
laboratory performance. These standards are not subjected to the
same rigorous certification procedures as SRM's and, because they
are intended for use not long after they are prepared, they do
not receive the same stability testing as SRM's. Examples of
such check standards have included solutions with known
85
-------
radioactivity levels, synthetic rainwater samples, solutions of
polynuclear aromatic hydrocarbons in organic solvents, phenols in
organic solvents and water, sediment samples and biological
samples.
NBS scientists are also engaged in the development of
standards for water flow and the development of means for al-
lowing flow measurements to be compared with these standards.
The standards will consist of a closed circuit water tunnel in
which turbulence and wall effects can be predictably controlled
and an open channel facility with known flow. With these two
standards, common methods for flow measurement can be evaluated
and sources of error identified and quantified. The means for
ensuring comparability of field measurements to these standards
will consist of mathematical models and accompanying guidelines
With the guidelines, field engineers can measure prescribed
geometries and velocities as well as noting installation con-
ditions, and consulting engineers can analyze the observations
with the mathematical models to provide calibration of the
installation.
As already mentioned, another portion of the NBS mission is
the determination of physical and chemical properties of
materials and the development of methods for measuring these
properties more accurately. Research activities applicable to
water monitoring problems include the measurement of those
properties affecting the fate and distribution of chemicals in
the environment and the studies of properties which relate to an-
alytical chemical methods. Properties affecting the distribution
and fate of chemicals in the environment include solubility;
partition coefficients, adsorptivity and chemical stability'
(including photochemical stability). Properties which relate to
analytical determination of compounds in water include activity
coefficients (for electrochemical response), partition coef-
ficients (for separation efficiencies), membrane permeability
surface chemistry and spectra. It is in the provision of '
Standard Reference Materials for the evaluation of sensors and in
basic research into the properties of materials as they relate to
the measurement process that NBS can be expected to make its
greatest contribution to the development of in situ water
sensors.
86
-------
U.S. NAVY, OFFICE OF NAVAL RESEARCH
by Dr. James Bailey*
U.S. Navy, Office of Naval Research
Arlington, Virginia 22217
The Geography Program of the Office of Naval Research has
been conducting research in remote monitoring as it might apply
to aiding in the prediction of changes in conditions of coastal
environments.
The remote monitoring program is concerned with determining
the feasibility of using various sensors to acquire accurate and
timely measurements of significant diagnostic parameters.
Particularly, the program is dedicated to understanding the phys-
ics producing and affecting the energy signals received and re-
corded by the various sensors.
Our main concerns have been and are attempting to acquire
bathymetric data rapidly and accurately, in determining tem-
perature and salinity as a function of depth, and in general de-
termining the normal distribution of bioluminescent and fluores-
cent material in the world ocean. None of the research dealing
with these projects has been concerned with water quality per se,
but one can easily surmise that a spin-off from these projects
could be a capability to predict water quality from remotely
sensed data.
Our bathymetric studies have included the use of lasers,
cameras, and a multispectral scanner. We are actually dealing
with water turbidity when attempting to acquire bathymetric data.
Preliminary experiments have shown that there is a direct
correlation between the amplitude and shape of the elastic (no
change in wavelength) laser backscattered signal and water
turbidity. These experiments showed a linear relationship between
the magnitude of the laser backscatter and turbidity. Although
this relationship was found to exist at all wavelengths, the
greatest effect occurred for a laser excited wavelength of 440
nm. These preliminary measurements indicate the feasibility of
using the backscatter from an airborne laser transceiver as a
direct-reading alphameter (sediment load). The accuracy of this
technique for measuring alpha at 440 nm is estimated to be 5 to
10 percent.
A second method for determining water turbidity proved suc-
cessful; this method consisted of using the measured intensities
of subsurface reflections, determined for two water depths, to
yield the effective attenuation coefficient of the water.
This narrative not presented orally at the Workshop.
87
-------
A third method consisted of using a pulsed laser to excite
fluorescence in a dye cloud, which was composed of more than one
dye. The peak fluorescence signals from the multiple dyes are
detected. Ratioing techniques are then used to extract various
hydrographic parameters, such as water depth, temperature, salin-
ity and water turbidity, from the data.
Detailed analysis of the inelastic Raman backscattered
signal from the water should be able to provide one with a value
of the water turbidity in a manner similar to that using the
laser elastic backscatter signal. It has been shown that selec-
tive signatures can be obtained utilizing Raman and Fluorescent
Spectroscopy. This high degree of selectivity which is obtainable
using these technologies results from the fact that the
intensities and wavelength of emissions from a particular mater-
ial are directly coupled to the intensity and wavelength of the
source of illumination. This therefore enhances the probability
of being able to detect and identify specific material in the
water.
Remote sensing of fluorescent matter either floating on the
water or within the water column may yield information on ex-
isting currents and eddies at various water depths. Since tem-
perature and salinity affect the intensities and spectral
character of the fluorescent and Raman signals, it is possible
that remote detection of these signals can be used to measure
these quantities as well as turbidity.
The possible use of fluorescence as an indicator of sub-
surface disturbances is given by the following example: The
majority of the upper ocean is thermally stratified, sterile and
blue in color. When this stratification is overturned as by
natural currents/ human activity, and/or meteorological events
nutrients and cold water are transported into the sunlit layers
Our research program is concerned with all the in-situ and
remote sensing data obtained in the past on naturally occurring
matters such as gelbstoff, bioluminescence and phytoplankton.
The investigation also includes the assessment of research
programs which have been involved with the nature and dis-
tribution of other organics which under the right conditions
might fluoresce. In addition, this program is considering the
signatures and sensitivities of measuring equipment to inter-
ferences due to fish fluorescence and that due to accidental or
planned oil releases. The results of this program will be used
to assess the feasibility of using fluorescence, and
possibly Raman signatures, of material in the water as indicators
of various parameters, temperature, salinity, etc., in the
hydrosphere. The results of this study will also be useful in
guiding future research efforts in the measurement and predictio
of the fluorescent material in the waters of the world. n
88
-------
The Over-the-Horizon (OTH) drogue system consists of a low
cost, expendable sea sonde that can be used with OTH in either
the ground wave or sky modes to obtain real time coastal and
oceanographic data. Data that can now be obtained include wave
height and period, temperature and salinity as a function of
depth, current direction and speed as a function of depth,
turbidity, chemical oceanography, and meteorological
measurements.
The OTH/drogue system is already being utilized by industry
and various government agencies. Further development is needed
to make the system air-droppable and self-mooring, to add to the
system capabilities, and to bring it into operational read-
iness for the Navy. This OTH/drogue capability will be part of
the coastal reconnaissance system now being developed by the Navy
and Marine Corps.
This extremely brief glimpse into some of our research in
ONR Geography Programs serves to emphasize the importance of
knowing the environment being sensed, knowing the characteristics
of the sensor, and knowing when to use the sensor as well as how
to process the data for various levels of analysis.
89
-------
U.S. COAST GUARD
by Lt. Cmdr. Vincent B. DiPasqua
Environmental Protection Division
U.S. Coast Guard
Washington, B.C. 20590
Coast Guard pollution surveillance responsibilities can be
traced back to the mid 1960's. The Federal Water Pollution Con-
trol Act (FWPCA) as amended in 1966, 1972, and 1977 set the para-
meters for such a surveillance system. One section of the Act
calls for the preparation and publication of a National Con-
tingency Plan (NCP) for the removal of oil and hazardous sub-
stances. This legislation mandates that the NCP include:
". . .a system of surveillance and notice designated to
insure earliest possible notice of discharges of oil and
imminent threats of such discharges to the appropriate State
and Federal agencies." (FWPCA, Section 311(c)(2))
The NCP qualifies the type of surveillance system required of the
Coast Guard. Another section states, in part:
"... the Coast Guard maintains continuously manned facili-
ties that are capable of command, control, and surveillance
for discharges occurring on the waters of the United States
or the high seas." (MCP, Sec. 1510.22(1))
Research and development for aerial pollution surveillance
sensors was initiated in 1969. it was soon realized that while
similar principles could be employed, harbor surveillance called
for different types of sensor applications. Research and de-
velopment along this avenue began in 1971.
The Coast Guard is currently evaluating three types of water
quality sensors on the Rouge River in Detroit, Michigan. This
marine environment is known for its chronic pollution problems.
The sensors are placed in strategic locations along the river
which are historically known for oil spills and seepage. They
are linked to Coast Guard Group Detroit by a communications sys-
tem that employs both telemetry and a leased telephone line. The
equipment undergoing testing includes the Spectrogram Ultraviolet
Oil Detection Buoy (Spectrogram, Inc., North Haven, CT.), Ramble
Infrared Oil Sensor (Ramble, Inc. Irving, TX.), and the Buoy
Mountable Hydrocarbon Vapor Sensor (Midwest Research Institute
Kansas City, MO.). All three sensors have been previously em-'
ployed in New York (Bayonne) or New Haven Harbors where they
underwent initial testing.
Evaluative testing is proceeding satisfactorily and is
scheduled to run through September of this year. Provided that
90
-------
no major problems materialize, plans call for placing another
such sensor system in a saltwater port area in late 1978. The
port/harbor chosen will have to be of medium size, have areas of
chronic oil spillage, and have a Coast Guard Marine Safety office
or Captain of the Port Office which utilizes a 24-hour watch.
Two ideal possibilities that are under consideration are
Baltimore and Houston. A final decision will be made later this
year.
The environments under which these sensors have operated are
quite varied. In all three areas, New York, Connecticut and
Michigan/ the equipment operated in temperatures which fluctuated
anywhere between extreme heat (above 90 F) and extreme cold
(below 0° F). The Rambie and Spectrogram sensors are not
capable of detecting oil on ice. Oil must float past them for a
programmed period so the equipment can compare what it "sees" to
the background readings. The hydrocarbon vapor sensor can be
employed to fill this void. Humidity has negligible effects on
the sensors, however, in theory the range of Rambie's equipment
should be affected by fog or heavy haze. To date, this situation
has not been experienced. All three test areas are highly
industrialized. Heavy air and water pollution exist. The Rouge
River especially has been subjected to years of industrial abuse.
The auto industry, among others, has used the river's waters as a
sewer to deposit its industrial wastes. This condition, even
today, goes virtually unchecked due to the industry's political
and economic ties with the area. Detroit itself is honeycombed
with a complex outfall system. Many times it is impossible to
pinpoint a spill source after the pollutant reaches an outfall's
mouth.
To cope with a wide range of environments, the Coast Guard
requires sensors which are continuously monitoring, automatically
resettable, self sustainable, low cost, low power, and able to
detect, identify, and quantify oils and hazardous substances in
harbors and inland waters. Though it is within the capabilities
of these sensors to detect any visible sheen, this may be too
sensitive from an operational standpoint. A visible sheen (1-3
microns in thickness) constitutes a violation of federal water
pollution laws. In reality, this is unenforceable. In places
like the Rouge River, sensor alarms would be triggered
constantly. This is, in fact, what actually occurred immediately
after the system became operational. Sensitivities had to be
decreased. One way this conflict of interest can be approached
is to place less sensitive sensors in chronic spill areas and, as
the spill problems subside, slowly increase the sensitivities.
This will, undoubtedly, have to be resolved in the near future.
Sensitivity is not the only problem inherent in an in situ
sensor system. Another major problem experienced at the Rouge
River test site is an inconsistent communications link between
the sensors and display board (located at the Group Office). It
91
-------
was discovered that the source of the difficulties did not exist
in the telemetry system between sensors and the leased telephone
line as was originally thought, but in Michigan Bell Telephone
Company's equipment. The leased line was inoperative for 3
months during the summer of 1977. Throughout this period,
Spectrogram called Michigan Bell everyday without results. it
was only after the contractor threatened suit that the telephone
company repaired its equipment. Afterward, the system worked
well for approximately one month, then the line became
inoperative again and remained that way until the Spectrogram
buoys were removed 'from the river for the winter. Present plans
are to convert the telephone line to a more reliable teletvoe
1*1 *^
ink.
Another problem that has been experienced may be related to
the environment under which the sensors operate. The Rouge Rive
winds its way through downtown Detroit where there are numerous
tall buildings and industrial stacks. Transmission of sensor
data in this environment is more difficult requiring more power
to achieve clear transmissions. Larger power requirements lead to
more solar panels and bigger batteries resulting in increased
costs and weight. Most of this can be alleviated in a harbor
environment as communications can take better advantage of an
unrestricted line-of-site.
The Spectrogram buoys rely on solar energy to recharge their
batteries and for power during daylight hours. The solar cells
employed for this purpose have a 12-percent solar to electrical
conversion. Vandals, on a number of occasions, have thrown rocks
through the panels. Any small crack can effectively reduce panel
efficiency by 2-3 percent placing a heavier load on the
batteries. While not a system-related deficiency, it is one
which must be coped with in any urban environment.
Industrial receptiveness is yet another source of problems
Many industries who are using the river to discharge industrial"
wastes see the sensors as a threat to this comparatively
inexpensive practice. Faced with high treatment and/or disposal
costs to remove these contaminants, several companies have not
been receptive to Coast Guard requests to rent space on their
property for test equipment. Some companies have been completely
uncooperative, and on two occasions a Spectrogram buoy moored
adjacent to a major motor company's outfall was sabotaged.
Probably the most important problem associated with these
sensors is their limited field of view. They are spot sensors
and, as such, if oil does not pass within their scope they cannot
possibly detect it. To alleviate this, Rambie, Inc., is
incorporating a scanning feature into their models. In the case
of the hydrocarbon vapor sensor, deficiencies lie in the alarm
and logic circuits. Additional work is needed in this area to
make this sensor ready for operational use.
92
-------
As was mentioned before, sensor sensitivity is a problem.
Rouge River water contains a high percentage of oil. The sensors
are constructed in such a manner that, once in operation, they
will adjust to a background value over an 8-hour period and use
this as a standard to evaluate what it sees. When the background
reading is high, as on the Rouge, the sensor may actually be
seeing oil but not alarm.
Coast Guard research and development (R&D) in the field of
in situ sensors has centered upon the current Rouge River test
and improvements of that system. Spectrogram is looking into a
better data communications system while Rambie has developed a
small, low cost oil spotter and a scanning version of its
infrared sensor. While one type sensor may be used to detect all
types of oil, preliminary R&D indicates that this is not the case
for hazardous substances. Different chemical groups require
different technical applications for detection. R&D is
continuing in this field.
Coast Guard opinion dictates that while the basic principles
behind infrared and ultraviolet oil detection will not change,
future technology will center upon refining those sensor systems
the Coast Guard already has. This will most likely involve
creating sensors which have a larger effective field-of-view and
more efficient telemetry and data communications systems.
Quantification and positive identification of oil type might also
be within the realm of future technology. Deepwater ports and
other offshore facilities which employ underwater pipelines have
inherent oil detection problems. Pipeline surveillance is
currently conducted using flow meters (volume in, volume out).
Future technology in this area could be applied in building
better flow meters or developing a sensor which is capable of
doing the same job more efficiently. A major consideration for
the latter is the use of ultrasonic sensors. These would be
employed at strategic locations along a pipe to listen for fluid
leaks or metal stress.
The Coast Guard is not alone in its developmental efforts of
water quality sensors. Both the Environmental Protection Agency
and the Department of Energy have complementary programs in this
field. While the Coast Guard has sponsored the Rouge River
tests, these agencies have provided the bulk of the funding.
In situ sensors may be, at least, a partial solution to the
momentous problem of water quality surveillance of ports, harbors
and inland waters. In the year that they have been present on
the Rouge River, a marked decrease of oil in river water has been
noted. As was mentioned in the test, they do have a deterrent
effect on industry. This alone may, in part, help to justify
their existence. While still an operational concept of the
future, in situ sensors are fast becoming a viable surveillance
means of the present.
93
-------
U.S. GEOLOGICAL SURVEY
by James H. Ficken
U.S. Geological Survey
NSTL Station, MS 39529
The Geological Survey's Water Resources Division conducts
hydrologic programs to acquire, analyze, process, store, and dis-
seminate data and information on the quantity, quality, location
movement, and changes in the Nation's water resources. The formal
mission statement for the Water Resources Division is enclosed as
Appendix A.
The Survey, as the principal Federal water data agency and
as delegated by the Department of Interior, under OMB Circular
A-67, has established the Office of Water Data Coordination
(OWDC), which is charged with coordinating water data acquisition
activities of all Federal agencies. OWDC develops regional and
national plans for water data acquisition and has developed the
"National Handbook of Recommended Methods for Water Data Acquisi-
tion," which has been compiled with the cooperation of experts
from numerous Federal and non-Federal agencies. OWDC has con-
stituted two standing advisory committees, one representing the
private sector and one representing the Federal establishment, to
enhance water data coordination.
The Geological Survey routinely collects data on river dis-
charge, lake and reservoir content, surface and subsurface water
quality, and groundwater levels at almost 40,000 separate sites
Approximately 25 percent of these stations have continuously
operating unattended sensors. For example, over 9,000 stations
continuously monitor and record river or reservoir levels in
machine readable form, of which over 125 stations transmit their
data by satellite retransmission for experimental evaluation of
this technology. Several hundred stations monitor and record
surface water quality.
In recent years, the Survey established the National Stream
Quality Accounting Network (NASQUAN), which currently provides a
uniform national network of 445 stations, which at its full de-
sign level will include 525 stations in Alaska, Hawaii, Puerto
Rico, and the coterminous States. Information from this network
is used to assess current water-quality conditions and trends
The bulk of the data collected at NASQUAN and other water-quality
stations is derived from chemical, biological, and radiological
analyses of routinely collected water samples. The Survey's
field and project offices are served by two highly automated
water-quality laboratories in Denver, Colorado, and Atlanta,
Georgia, where analytical services and methods development are
provided.
94
-------
A key element in all Survey water resources programs is a
computerized data base known as the National Water Data Storage
and Retrieval System (WATSTORE). USGS offices daily and
routinely enter, edit, process, and retrieve water data from
WATSTORE files. Water-quality information for WATSTORE files is
entered into the EPA STORET system, of which Survey data com-
prises almost half the data base.
The Survey's water resources activities are responsive to a
combination of national and local needs—about half the support
is provided by the Federal/State cooperative program, where funds
are matched on a 50/50 basis by each party. Some 580 State and
local agencies participate in the program. The remainder of the
support is derived from Geological Survey budget line items and
other Federal agencies.
The Survey's interest in water-quality sensors stems from
the increasing need to collect continuous water-quality data
nationwide. Water-quality sensors presently in use measure water
temperature, pH, specific conductance at 25° C, and dissolved
oxygen concentration. There are also a number of "turbidity"
sensors in use; but, because of the difficulty in standardizing
and defining the measurement of turbidity, the Survey is de-
veloping instrumentation for measuring light transmittance and
scatter in water.
The Survey's network of hydrologic stations is national in
scope, and the environment in which sensors must operate ranges
from the tropical to the arctic. Some of the less obvious, but
real, problems of the environment include risks of vandalism in
urban and rural areas and the general difficulty of monitoring
water in sand channel or ephemeral streams. For example, a large
number of hydrologic stations are located on ephemeral streams
where conventional stream gaging, sampling, and monitoring tech-
niques are not effective. In such an environment, virtually all
the animal river discharge, sediment load, and constituent trans-
port occur during a few intense hydrologic events, which for
logistical reasons are especially difficult for personnel to mon-
itor. There are similar problems that are being encountered in a
new program to monitor stormwater runoff in urban areas. In this
program, measurements of discharge, precipitation, and water
quality are made during a number of short duration intense
events, where once again manual techniques are insufficient.
The Survey has developed and tested a variety of monitoring
systems that support its programs. They include:
• USGS Water Quality Monitor. A ten-
channel monitor designed by the USGS
to operate off line power. There
are about 85 monitors in operation
in the field, with 50 new units
under fabrication.
95
-------
• USGS In Situ Water-Quality Monitor.
A prototype four-parameter
battery powered system is under
test and is a scaled down version
of the ten-channel monitor.
• USGS Urban Hydrology Monitor.
An automatic system to monitor ur-
ban stormwater runoff. The system
records precipitation and runoff
and automatically collects samples,
which are stored in a refrigerated
chamber for later water-quality
analysis. Nine prototype systems
are in the field.
• Video Taping Hydrologic Events.
This is a feasibility study to
determine if video systems can
monitor remote ephemeral streams
to estimate discharge.
•Light Transmission-Scattering Meter.
A prototype system is currently
being tested. Further develop-
mental work is planned for a system
that can be used both in the field
and the laboratory.
• Satellite Telemetry. Two potentially
operational satellite telemetry systems
are under test to monitor and collect
data from hydrologic sensors.
The Survey participates in numerous interagency committees
to coordinate water data collection activities. These include
coordinating committees with NOAA, SCS, Fish and Wildlife, etc
as well as technology oriented groups, such as the Interagency"'
Working Group on Satellite Data Collection Systems, which
coordinates research in satellite telemetry technology. There i
a need for additional ongoing and formal liaison between groups
that are active in sensor development. The current workshop is a
good opportunity to define the need and opportunity for coordina-
tion in water-quality sensor development. ""
96
-------
APPENDIX A. U.S. GEOLOGICAL SURVEY, WATER RESOURCES DIVISION
BASIC MISSION AND PROGRAM
The mission of the Water Resources Division is to provide
the hydrologic information and understanding needed for the op-
timum utilization and management of the Nation's water resources
for the overall benefit of the people of the United States.
This is accomplished, in large part, through cooperation
with other Federal and non-Federal agencies by:
1. Collecting, on a systematic basis, data needed for the
continuing determination and evaluation of the quantity, quality,
and use of the Nation's water resources.
2. Conducting analytical and interpretive water-resource
appraisals describing the occurrence, availability, and the phys-
ical, chemical, and biological characteristics of surface and
groundwater.
3. Conducting supportive basic and problem-oriented re-
search in hydraulics, hydrology, and related fields of science to
improve the scientific basis for investigations and measurement
techniques and to understand hydrologic systems sufficiently well
to quantitatively predict their response to stress, either
natural or manmade.
4, Disseminating the water data and the results of these
investigations and research through reports, maps, computerized
information services, and other forms of public releases.
5. Coordinating the activities of Federal agencies in the
acquisition of water data for streams, lakes, reservoirs, es-
tuaries, and groundwaters.
6. Providing scientific and technical assistance in hydro-
logic fields to other Federal, State and local agencies, to
licensees of the Federal Power Commission, and to international
agencies on behalf of the Department of State.
97
-------
DEPARTMENT OF AGRICULTURE
by Frank R. Schiebe and O. W. Sanson
USDA Sedimentation Laboratory
U.S. Department of Agriculture
Oxford, MS 38655
This report summarizes the experience of several USDA
installations with regard to various types of water quality
sensors. The experience summarized is from the USDA Science and
Education Administration (SEA), formerly known as the
Agricultural Research Service. Inquiries to the U.S. Forest
Service and the Soil Conservation Service indicated a stronq
interest in such sensors but neither agency had initiated
monitoring activities. Both agencies are desirous of feedback
information from this workshop and both anticipate future
application of remote water quality sensors to their missions
The USDA is interested in (1) the quantity and quality of
water available for agricultural activities and (2) the qualit
of the effluent from various agricultural activities which is
passed downstream and affects those water resources. There is
particular emphasis on nonpoint sources of potential water DO!
lutants. ^
Within SEA there are four known locations where sensors ar
employed. The U.S. Salinity Laboratory at Riverside, California
has developed two different sensors for the purpose of measuri
the salinity of soil moisture. These sensors are implanted in VK
soil and have a sensitivity of + 1/4 to 1/2 millimhos/cm. Both
sensors are marketed by commercial firms and are now employed
around the world. They may be used wherever salinity
measurements of soil moisture or groundwater are desired.
The USDA Water Quality Management Laboratory at Durant
Oklahoma, is utilizing sensors to monitor temperature, con-'
ductivity, dissolved oxygen, and pH in 10 model farm ponds, 6
meters in diameter and 2 1/2 meters deep. The water is fresh
eutrophic, and relatively stagnant. The sensitivities desired'
are all at one tenth of a unit and the present equipment is
satisfactory from this standpoint at the time of calibration.
The temperature and conductivity sensors are working
satisfactorily at Durant. The pH sensors require weekly calibr
tion. The probes only last 6 to 9 months but the manufacturer ""
has been replacing them at no charge. Currently available DO
sensors will not maintain calibration more than a few days.
The USDA Southwest Rangeland Watershed Research Center at
Tucson, Arizona, has employed conductivity and pH sensors for
98
-------
short durations. Performance has been generally satisfactory in
an ephemeral environment.
An extremely sensitive sensor for measurement of specific
density has been employed in the field at this location to
continuously monitor suspended sediment in streams. This device
is based on changing the resonant frequency of a vibrating tube
when liquids of different densities flow through it. Performance
of this sensor has been generally good.
This model density sensor has been used in the laboratory at
the USDA Sedimentation Laboratory in Oxford, Mississippi, also
with satisfactory results. Sixteen such units have recently been
delivered to the Sedimentation Laboratory and will shortly be
employed in a new research watershed study. Successful
employment of these total solids resonance type sensors in the
field study at Oxford, Mississippi, will be of great importance
to our agency. The potential savings in labor compared to
conventional sampling and traditional laboratory analytical
techniques should be substantial. In this study research data
will be collected by a line of sight radio communications system
and acquired directly by the main laboratory computer.
The Sedimentation Laboratory also has eight automated water
quality data collection stations located in the Mississippi
Delta. These sensors are located in small streams and oxbow
lakes with moderate suspended sedimeixt loads. The drainage areas
are almost entirely covered by row crops. Sensors employed
include temperature, conductivity, DO and pH. Analog signals
from these sensors are digitized and logged on magnetic tape
cassettes. Automatic water sampling equipment is also employed
to allow determination of water quality variables which are not
determined by sensors.
Electromagnetic velocity sensors with no moving parts are
employed to assist in the determination of water discharge. The
velocity sensors have performed quite satisfactorily. One sensor
has been used continuously in a polluted stream for almost 18
months without failure. Performance of the temperature and
conductivity sensors has been satisfactory. Performance of the
DO and pH sensors has been almost identical to that of the
instruments at Durant, Oklahoma. We have been able to obtain
satisfactory DO data only by servicing and calibration in the
laboratory in the afternoon prior to field measurements the next
day.
In SEA the most serious deficiency has been with the DO
sensors. The problem seems to be one of poisoning of the membrane
and the electrolyte solution. One suggestion may be to increase
the volume of the electrolyte and modify the surface area of the
membrane. New technology in the area of DO measurement would be
very helpful. Field servicing and calibration is inconvenient,
and faster and easier field calibration procedures should be
sought.
99
-------
During the past 5 years research has been conducted at the
Sedimentation Laboratory to relate the spectral reflectance of
solar energy from a water surface to some water quality
variables, particularly inorganic suspended sediment. This work
was initiated to provide ground truth for interpreting Landsat
images and aerophotographs. A radiometer was used at water level
and incident and reflected solar energy was measured at several
wavelengths through the visible and near Infrared ranges. We
have found very good relationships between a corrected re-
flectance in the range from 700 to 800 nanometers and the total
suspended sediment load. A possibility exists for employing op-
tical sensors to indicate other water quality variables. It
should be emphasized, however, that optical sensors measure
optical properties of the water which should only be used to
interpret water quality variables after careful calibration.
The USDA Sedimentation Laboratory is presently cooperating
with the U.S. Corps of Engineers, Vicksburg District, on three
field research projects involving the water quality sensors
mentioned, and we are currently planning a fourth. We see inter-
agency cooperation as necessary to achieve the concentration of
effort and funds needed to provide a workable solution to the
sensor problem. Interagency cooperation would also tend to re-
duce duplication of effort.
The following persons were interviewed for this report and
their cooperation is acknowledged.
Mr. Paul Duffy - U.S. Forest Service, Oxford, MS
Mr. John Burt - Soil Conservation Service, Fort Worth, TX
Dr. Ronald Menzel - SEA, Durant, OK
Mr. Gary Miller - SEA, Durant, OK
Dr. Ken Renard - SEA, Tucson, AZ
Dr. Don Chery - SEA, Athens, GA
Dr. Jim Rhoades - SEA, Riverside, CA
Dr. Jerry Ritchie - SEA, NFS, Beltsville, MD
Dr. John Schreiber - SEA, Oxford, MS
100
-------
FEDERAL, STATE AND LOCAL
WATER QUALITY MONITORING NEEDS
As one of the primary water quality measurement needs deals
with environmental quality, and this need spreads across state
and local agencies as well as Federal programs, representatives
of two EPA Regional offices were invited to present a discussion
of these needs at the Workshop. As time would not allow such
presentations by each of 10 EPA Regional offices or several
representative states, it was felt representatives from two
Regional offices having particularly active interests in auto-
mated sensor applications would best provide the overview
desired. Through their implementation of environmental water
quality standards and regulations, EPA Regional offices have an
intensive working relationship with state and local environmental
and other agencies interested in water quality, and therefore
have the opportunity of being aware of the integrated, complex
needs of these various groups.
A narrative statement addressing this matter was prepared by
Mr. Clifford Risley of the EPA Region V office in Chicago and
Mr. Leonard Mangiaracina of the Region III office in
Philadelphia. That narrative is presented on the following
pages. Mr. Risley summarized these remarks in a 20-minute oral
presentation to the Workshop.
101
-------
FEDERAL, STATE AND LOCAL WATER QUALITY MONITORING NEEDS: AN EPA
REGIONAL VIEWPOINT
by Clifford Risley, Jr.
U.S. Environmental Protection Agency, Region V
Chicago, IL 60604
Leonard Mangiaracina
U.S. Environmental Protection Agency, Region III
Philadelphia, PA
Our message today can be stated briefly. The reduction of
the pollution of our environment is a tremendous challenge which
requires an effort at all levels of government and business
enterprises. It affects nearly everyone and, as a consequence,
is of interest to most of our concerned citizens.
pollution control has been the subject of much legislation
and a large expenditure of legal and technical effort. It has
resulted in the investment of billions of dollars by private
enterprise and by government. Ultimately, to make the whole
system effective, we will have to monitor the results of our
pollution control efforts. Maintenance of high water quality
standards, thorough and equitable enforcement measures, and
protection of public health mandate a complete surveillance
program.
We do not have the manpower to monitor pollution control in
an effective manner, nor can we afford the investment in manpower
necessary to do the monitoring effort using the traditional
techniques of collecting samples by hand and taking the samples
to the laboratory for analysis. We also cannot afford to wait
days for the analysis, data interpretation and reporting before
taking action. This is true whether we are talking about control
of the waste treatment process and subsequent discharge; or of
the local, state, and national agencies' ability to monitor and
take action to prevent pollution or to take prompt enforcement
action against violators.
The only way we can provide environmental monitoring in a
manner which is sufficiently thorough and rapid enough to give us
effective pollution control will be through the use of advanced
technology systems. We must utilize automated monitoring coupled
with a computerized evaluation and alerting system. This will
ultimately be greatly enhanced by integration with remote sensing
and rapid communication networks.
102
-------
Most of the technology needed to do this job effectively has
already been developed and such systems are in use in some loca-
tions. We have the automated monitoring packaging technology, we
have the computer technology, we have the analytical technology,
we have the remote sensing technology and we have the communica-
tion technology.
Our weakness is in sensor technology which at this time
limits the potential of the rest of the system.
I firmly believe that we will get a commitment to invest in
these advanced systems by Federal, State and local government,
and by industry, when they are convinced that the complete system
exists. Ultimately, they must make this investment because they
will not be able to ignore their monitoring responsibility and
they will not be able to afford the large expenditure necessary
to carry out the monitoring responsibility by less efficient
manual methods.
Now, I want to go back and elaborate on the points which I
have just outlined. I particularly want to emphasize the
monitoring needs and attendant sensor needs at local, state and
national levels.
The Pollution Control Challenge
In December of 1970, just seven years ago, the United States
officially declared war on pollution by establishing the Environ-
mental Protection Agency (EPA). Since then Congress has appro-
priated billions of dollars to wage this war. It has invested in
the building of municipal waste treatment facilities, in support-
ing state and local agencies, in legal action and in research and
development. But over the past seven years, EPA has learned that
an essential element for waging a successful campaign against
pollution is missing. That element is an adequate knowledge of
the enemy, its whereabouts, its source, its concentrations, its
characteristics, and its effects.
Maintenance of high water quality standards cannot be
assured unless there is a stringent operating program with
constant surveillance of river conditions and means to provide
quick and appropriate ameliorative action. Even the most skilled
and experienced engineer or scientist cannot determine the
quality of water just by observing it or from a few grab samples.
The common practice of the past and present is to send
people out to pick up samples for later analysis in a laboratory.
In order to conduct an effective monitoring program, EPA must
turn to sophisticated intelligence gathering techniques such as
remote sensing and automated monitoring. While remote sensing
and automated monitoring are not expected to totally replace the
103
-------
traditional manual sampling and laboratory pollution analysis,
they can provide a speed, perspective and mobility that can be
obtained in no other way.
All the latest scientific aids must be utilized to prevent
pollution. Effective and economically feasible measures and
methods to improve water quality can be achieved only by better
collection of pollution data, using automatic water monitoring
techniques across the waterway system involved. Processing data
in the form required by engineers and scientists, as the basis
for sound planning and action, is a necessary part of this
operation.
Trend Information
"Accurate and timely information on status and trends in the
environment is necessary to shape sound public policy and to
implement environmental quality programs efficiently." This
statement from the third annual report of the Council on Environ-
mental Quality underlines the fundamental need for the Environ-
mental Protection Agency (EPA) to acquire pertinent environmental
data.
Environmental quality data have been collected on a nation-
wide basis for a number of years but the ability to discern
trends on a national or even a regional scale is lacking.
Further aggravating this situation is the fact that the list of
mandated and suspected pollutants requiring surveillance is
growing.
Monitoring Expenditures
EPA currently spends approximately $33 million annually on
environmental monitoring. State and local pollution control
agencies spend approximately twice that amount. The private
sector is estimated to spend in the order of $50 million to $100
million annually on source monitoring. Even at these levels of
expenditure, monitoring coverage, both spatially and temporally/
is extremely sparse because of the relatively high costs of
monitoring. With the current state of the art, monitoring
expenditures will have to be increased severalfold in order to
provide essential data on standards violations, emerging
problems, and overall successes or failures of pollution abate-
ment efforts. Because of the magnitude of these expenditures,
effort toward improving monitoring efficiencies is warranted. In
that monitoring is an iterative process generally to be continued
indefinitely, any improvement in the efficiency of monitoring
systems will represent cumulative savings that are realized from
that point forward. Accordingly, even at the current level of
activity, a one time increase of only five percent in monitoring
efficiencies will result in an annual saving to the public sector
of approximately $7.5 million.
104
-------
Considerable effort has been expended by the National
Aeronautics and Space Administration (NASA), the Department of
Defense (DOD), and other governmental agencies, and a number of
private organizations in the development of remote sensing
systems for deployment in satellites, in aircraft, and on the
ground to observe and record certain conditions and characteris-
tics in the biosphere. Sufficient work has already been done
inside and outside EPA to demonstrate the feasibility of
utilizing certain of these remote sensing techniques, when
complemented by in situ measurement and associated data proces-
sing systems, to meet some environmental information needs of EPA
and state pollution control agencies. In addition, several of
these advanced techniques appear to have a good potential, with
some adaptation, for meeting a number of other environmental data
needs of EPA in more effective and efficient manner than the
approaches now employed.
In particular, automated monitoring station networks coupled
with data processing and situation centers would have direct
application to EPA and State regulatory agency needs.
The Jfleed for Monitoring
The basic items common to any water quality monitoring
program include: 1. program design, 2. flow measurement, 3.
in situ determinations, 4. sampling, 5. laboratory analysis,
6. quality control, 7. data management, interpretation and
reporting.
In considering the water quality monitoring efforts of the
federal, state, and provincial governments in both the deep-water
and near-shore monitoring and surveillance programs in the Great
Lakes, the agencies have concluded that insufficient resources
are committed to permit consistent and meaningful progress in
achieving water quality objectives.
Significant variations occur in the monitoring effort,
coverage, methods of sampling, analysis and reporting of data,
parameter selection, spatial coverage, and sample type. Among
the things not adequately assessed are the contributions from
erosion and rural and agricultural runoff. We have a great need
for obtaining a uniform and consistent data base for future
descriptions of waste loadings, changes in land usage, and
changes in water quality to facilitate comparisons and measures
of progress.
Other Regions in EPA and many State Agencies have drawn
similar conclusions regarding monitoring in the major rivers,
lakes and the ocean.
105
-------
When you examine the total effort in man years presently
being applied to the monitoring effort by all the agencies
concerned, it appears to be a sizeable effort, but when you
compare this effort with the total manpower requirements needed
to accomplish the goals of the legislation it becomes
over-whelming; beyond our most optimistic anticipated manpower
resources for many years to come.
The best way to extend this effort and to accomplish our
objectives will be approaching the problem from a new vantage
point. We need to develop new techniques which depend on
reliable sensors, automated analytical methods, automated
sampling and remote sensing.
Anticipating environmental problems depends on the ability
to identify and measure substances in the environment and to
observe their trends and effects through well designed monitoring
programs. Our present knowledge of the substances within the
environment is extremely limited and what we do know has been
gained through application of measurement techniques in limited
studies and from results accumulated over many years. These
techniques have identified hundreds of substances, many of which
are toxic or potentially toxic. However, the number of
unidentified substances continues to increase each year as more
synthetic substances enter the environment from more sources,
from the manufacture, use, and disposal of new commercial
products and from emerging energy technologies. Effective
identification and control of these pollutants will require more
monitoring than we have been able to provide. We need better
instrumentation, techniques, and quality assurance procedures for
accurately describing and evaluating the origin, vectors, and
effects on both human health and the environment.
State and Federal Monitoring Requirements
State and Federal Agency requirements for water quality
monitoring are similar. They are shared in many instances, but
the response time requirements are different. The state should
respond rapidly to municipal or industrial waste treatment plant
failures, bypasses of wastes, accidental spills from storage
areas, transportation accidents, or acts of nature. The state
needs to assess the hazard and protect the affected population,
whether they be in immediate contact with the incident or are
users of the water downstream. Federal agencies are called in
when interstate water bodies are affected or upon request by the
state agencies.
Historically, regulatory agencies have been alerted when a
sportsman reported a fish kill, a farmer had ill or dying live-
stock, or a municipality or industry detected a change in odor,
taste or appearance of the water supply. They respond by sending
out field crews to collect samples.
106
-------
This is obviously a reactive system which responds only
after damage has been extensive. It does not prevent major
pollution events nor can it detect them early enough to help
minimize the event. In fact, it relies on chance and may miss
the event until it is too late to prevent major damage to a
downstream water supply.
We cannot hope to prevent major pollution effects unless and
until we develop a real time automated monitoring system network
which can detect a spill, or discharge violation, evaluate the
data, and communicate the information to alert the concerned
regulatory agencies with minimum delay.
As we know automated, self-powered in situ water quality
sensors have been around for a long time. However, during the
last 5-6 years they have undergone extensive development and
improvement, especially in terms of the reliability of the
sensors, the data control package and readout device. Also,
automated field records (cassette tapes) and field data memory
units which dump data to satellites or telephone lines are now
well developed and used by NOAA, USGS, some EPA Regions and some
states.
The major problems with present systems are: the most
reliable sensors are T, Cond, DO, pH, but only a few specific ion
electrodes. Measurement of salinity by even the new probes is
rather weak; the best method is straight conductivity without
trying to convert, via electronics, to "salinity". Specific ion
electrode development is progressing slowly. The great need is
on specific chemical ion probes—our interest is high with regard
to ion probes.
Region III favors in situ water quality monitors and the
field "dipping" units like Hydrolab and Martek gear.
The big problem is that you need those fixed, in situ
devices (like Annapolis Field Office (Region III S&A Div) (AFO)
used to have and the states, basin commissions, and U.S.
Geological Survey now have) only if you have a full-blown
monitoring program.
As of now, there is no Region III dedication to a regional,
water quality monitoring program. If we had an estuarine, or
river, or region-wide stream program, then we would have need for
the fixed in situ monitors.
There are monitoring programs in Region III which use state-
of-the-art in situ stream monitors. Examples are: Ohio River
Sanitation Commission (ORSANCO), Susquehanna River Basin
Commission (SRBC), and Delaware River Basin Commission (DRBC)
which all have automatic in situ monitors throughout their
respective river basins. Interstate Commission of the Potamas
107
-------
(INCOPOT) has them, and some states (PA, MD) have their own
monitors in state stream/river programs.
The point is that the states and commissions with defined
programs already do use monitors in fresh water streams, but we
don't know of any monitors used in a major program within
estuarine Chesapeake Bay and estuarine rivers or creeks. There
are a few isolated personal studies such as Bob Corey's at
Smithsonian Institute (SI), one at Virginia Institute of Marine
Sciences (VIMS), and one at Chesapeake Bay Institute (John
Hopkins) (CBI) which use monitors.
The primary goal of a monitoring program is to provide the
information required to identify water quality issues and to
assess achievement of water quality objectives. The water
quality objectives in this general context include not only those
parameters with numerical limits but also the concept of non-
degradation of water quality. A secondary goal is to provide the
information needed to relate achievement or non-achievement of
the objectives to a particular cause. These goals require the
measurement of both water quality conditions in space and time
and material loadings to the receiving waters. The goals also
require or imply a framework in which to analyze the data. The
proposed monitoring program, therefore, includes all the elements
or components necessary to obtain these general goals.
Concerns for water quality are related to uses which require
a specific level of water quality and to uses which adversely
affect the water quality. Water must be protected for the most
sensitive use. The elements of environmental concern resolve
into a series of technical issues which include:
Enrichment
Organic contaminants
Metal contaminants
Radioactivity
Suspended materials
Microbiology
Dissolved materials
Trash, flotsam, jetsam
Thermal inputs
State and
toring for:
Federal regulatory agencies need full-time moni-
• Municipal waste treatment plant effluent permit
compliance
• Industrial waste treatment plant effluent permit
compliance
• In stream monitoring for water quality trends, to
determine loadings, and rural and urban runoff
characterization (sediment loadings, nutrient
and pesticide contributions)
• Provision of spill alert, storm and combined sewer
overflow inputs
• Lake monitoring in areas where potential for spills
exist
108
-------
• Estuary monitoring in areas where potential for
spills exist
These agencies need short term intensive monitoring capa-
bility (a few days to a few weeks):
• To provide data for enforcement actions and case
preparation
• For river, lake and estuary characterization, trends,
modeling, and planning corrective actions
They also need emergency response capability. They should
be able to move out to the field site rapidly via helicopter or
other rapid transportation so that they can define the affected
areas and determine the concentrations and seriousness of the
situation. They must then take actions to confine the incident
as much as possible, alert and protect the public, and take cor-
rective action. Rapid intensive monitoring is still necessary to
measure the effectiveness of these corrective actions.
Measurement needs--
Data read-out devices, memory packs, recorders, and power
supplies are well developed. We have needs for chemical sensors.
We are successful with t, Cond , DO, pH and Reference sensors. We
need reliable specific ion electrodes for field as well as for
lab use. We would be very interested to be informed on the de-
velopment of rugged, interference-free field probes for:
• Cl7F~, Br~
• CN7 SCN~
• CO2;alkalinity; acidity
• H2S,HS~ or S=; SOj
• N0~, NO"
• CIO"
• redox (fairly well developed)
• Metal species - still open for development and any
other inorganic complexes on the shelf or being de-
veloped
We know that many of these electrodes are already in use in
the laboratory for certain special analyses where the analyst
knows the general composition of the sample. The real need is
for specific ion electrodes which will work in a common "natural"
or "partially polluted" stream/river or estuary of unknown
109
-------
chemical compositions. We need specific ion electrodes which
will work equally well in "fresh", estuarine, and marine water—
i.e., across the range of ionic strength change due to increase
in dissolved halogens.
We would like to know of new methods of interference preven-
tion, electronic "block" or "nulls", strengths and weaknesses of
various new sensors. We would expect that NASA's chemical
instrumentation and miniaturization, medical science industries
advances, and the R&D of companies like Orion, L&N, Beckman could
make a workshop very worthwhile and valuable to someone knowl-
edgeable in the field of in situ chemical instrumentation.
If Regional management has a region or Bay-wide field
monitoring program we would need this information for maintenance
of permanent in situ monitoring stations. Presently, we could
use the information for our potential portable in situ monitors
(Martek & Hydrolab) and our laboratory specific ion probes.
The operating programs of Region III were contacted to
determine their interest and needs relating to Automated In Situ
Water Quality Sensors.
The Enforcement Division would like to see a water quality
sensor system developed that could be used for the detection and
early warning of oil spills in remote locations. Interest was
also expressed for the development of water quality sensors that
could be used to assist in achieving improved compliance
monitoring. This would involve the development of multi-purpose
sensors and placing the sensor units at strategic outfall
locations.
There were also suggestions that in situ monitoring capa-
bility be improved by developing a dissolved oxygen sensor that
has greater operational stability, reliability and respecta-
bility.
It was also felt that the possibility of developing a
technique that would provide the capability of determining sub-
surface conditions from surface or near surface reading using
remote sensing systems should be explored.
The Chesapeake Bay Program is currently undertaking the
development of a baywide monitoring strategy that would determine
the ambient water quality conditions in the Chesapeake Bay. It
would be considered extremely valuable and, in all probability,
cost effective to utilize multi-purpose in situ water quality
sensors placed in strategic locations throughout the Bay. These
should possess the ability to monitor and transmit water quality
conditions to a central receiver. This would provide the capa-
bility of both recording the quality of the Bay at any point in
time and to observe trends in water quality on a close to real
time basis.
110
-------
The need to develop practical in situ water quality sensors
is a necessity if extensive and effective water quality monitor-
ing of large water bodies is to be a reality.
Limitations of Manual Sampling
Monitoring has long been undertaken by field sampling and
laboratory analysis. The costs are high and the scale of infor-
mation is restricted. A significant limitation is the time
between water becoming suddenly polluted and this event being
recognized. Delay in tracing the source and taking corrective
action can cause considerable risk to public health and extensive
damage to water ecology.
Among the major disadvantages to manual sampling are:
• Manpower is the largest direct cost in a monitoring
program.
• Response time to transport men and equipment to the
field is extremely long especially where large
distances and combinations of air, land, and water
travel are involved.
• Time delay between sampling and subsequent laboratory
analysis is greater than response time, thus
aggravating the situation.
• Time delays can result in significant changes in the
quality of the sample.
• Time delay between sampling, analysis and reporting
is time lost before taking corrective action.
• Manual sampling is inconsistent, requires considerable
operator training and it is difficult to assure
quality control over difference in operator technique.
• Manual sample collection is inefficient. The number
of stations collected and frequency of collection is
limited by distance and number of men in the field.
• The frequent necessity to keep manpower in the field
around the clock requires several rotating shifts of
manpower. This is very expensive and inefficient.
• Field personnel cannot anticipate when and where
spills will occur. They can only respond well
after a spill has occurred.
• Field personnel cannot anticipate when infrequent
peak discharges from outfalls will occur; must keep
around the clock vigilance with the hope that they
will catch the peak.
• Significant variations occur in monitoring coverage,
methods, analysis, and reporting.
• It is nearly impossible to measure rainfall events,
urban runoff or agricultural runoff through manual
sampling methods.
• It is nearly impossible to provide any uniform and
consistent data sets through manual sampling methods.
Ill
-------
The decision whether to sample manually or use automatic
monitoring is far from straightforward, and involves many
considerations in addition to manpower and equipment costs.
The decision to use automatic sampling equipment does not
represent the universal answer to water and wastewater character-
ization. For initial characteri2ation studies, proper manual
sampling may represent the most economical methods of gathering
the desired data. It is also prudent from time to time to verify
the results of an automatic sampler with manual samples. Also,
manual grab samples are often taken during visits to sites where
automatic monitors are installed in order to obtain data on
certain parameters, e.g., metals, organics, DO, oil and grease,
coliform bacteria, etc., that cannot be meaningfully measured
from samples taken by automatic equipment. Additional sensors
may change the balance of future decisions in this regard.
The use of automatic samplers is indicated where frequent
sampling is required at a given site, where long-term composit-
ing is desired, where simultaneous sampling at many sites is
necessary, etc. Automatic sampling will often be the method of
choice for storm-generated discharge studies; for longer period
outfall monitoring, for treatment plant efficiency studies; where
24-hour composite samples are required, and so on.
The assessment of water quality is often based on infor-
mation and data obtained from surveys, investigations and routine
monitoring activities conducted by a large number of institutions
for a variety of purposes. The lack of a coordinated monitoring
program, including adequate data quality assurance, means that
the data are suspected and that long term trends in water quality
are difficult to develop. Automated monitoring should be of
considerable help in this effort.
Monitoring in Large Municipal Waste Treatment Plants
Automated Monitoring is no stranger to managers and
operators of modern waste treatment plants in large cities.
Likewise, sensor problems and limitations are well known in these
facilities.
Automated Monitoring systems have been used for portions of
the waste treatment plant control in some plants for many years,
but never extensively. Recent attempts to control most of the
treatment process have been attempted by a few plants without
complete success.
The primary reason for the paucity of instrumentation used
in existing wastewater treatment plants is the unsatisfactory
performance of many of the primary measuring elements and
112
-------
analytical sensors. Since most of the measuring elements
interface directly with raw sewage, mixed liquor, or thickened
sludge, these devices are subject to rapid fouling. Accordingly,
they need more frequent cleaning and calibration. Several
on-line measuring devices for assessing the organic concentration
(TOG, TOD, COD, and respirometry) of wastewater are commercially
available at this time. These online analyzers require copious
amounts of skilled maintenance, which is usually unavailable in
most wastewater treatment plants.
Despite the favorable economics of automatic process
control, most wastewater treatment plants use very little
automation. A recent survey indicated that automatic chemical
addition, residual chlorine control and digester temperature
control were used by about 1/3 of the plants.
Virtually all the large facilities utilize central infor-
mation so that all important events, alarms and treatment
information are displayed and recorded in a centralized location.
Most new plants used automatic data acquisition systems and
approximately 20% of the new facilities use data logging
computers. However, only 10% use dissolved oxygen control. Only
a few plants have tried to provide central control of a large
part of their operation.
Real time computerized supervisory control of large storm
and combined sewer networks is in use in several cities.
Minneapolis, St. Paul and Detroit are examples of such networks.
The vast numbers of variables and control points exceed human
computational and decision making capabilities.
Monitoring of the upstream and downstream water quality as a
guide to operational control of the municipal wastewater
treatment plant was accomplished by Minneapolis, St. Paul. They
utilized four water quality monitoring stations to ensure that
the city complied with state and Federal Water Quality Standards.
They were able to determine from the water quality and flow
measurements what degree of treatment was needed. In this way,
use of advanced waste treatment techniques was limited to those
periods of low flow and low water quality conditions when they
were required, rather than to try to operate them on a full time
basis or on guesswork.
Monitoring in Small Municipal Waste Treatment Plants
In small municipal wastewater treatment plants, many cur-
rent rules of operation are based largely on only experience and
common sense. Wastewater treatment is a complex, stochastic,
dynamic system. A detailed study of its performance requires the
use of appropriate techniques that take into account the highly
variable nature of relevant inputs such as wastewater flow,
strength, chemical composition, weather and outputs such as flow,
113
-------
composition and quality. Some factors have known diurnal and
seasonal trends. The practical difficulty in the control of the
wastewater treatment plants is that some important factors may
not be known; those that are known are not monitored adequately
and thus the plant can not be controlled.
There seems to be a tacit agreement between small treatment
plants and the states in which they operate that monitoring will
not be a topic for discussion. The state asks the plants to
monitor their effluent quality to show compliance with effluent
standards. Typically, a small plant does not have an operator
who can do the laboratory work and its budget is too small to add
the necessary personnel and laboratory equipment for extensive
testing. The few small plants that do report some effluent
quality data, do so infrequently and sometimes the data are
unreliable.
Ofttimes a compromise is drawn between a daily monitoring
program (which many states want) and no monitoring at all (which
small treatment plants want). The compromise is to sample a
little. How little is enough to serve both interested parties is
never resolved.
Monitoring is not control. It is only the prerequisite for
control. Effective monitoring is needed before one knows how
much control is required or how effective applied controls have
been. The design of an effective monitoring system requires
balancing the protection purchased against the cost.
Calculated results from the few available cost studies show
that a sampling interval of two or three days should provide
protection, yet not overburden a small operation financially.
Increasing the penalty for polluting would drive one to more
frequent sampling.
If daily sampling is legally required of a small plant, the
responsible legislative body is assuming (perhaps without basis)
that the cost of monitoring is money well invested. It is, in
effect, assuming that the external and public costs of possible
occasional pollution exceed the cost of monitoring and control.
For small plants receiving industrial wastes which may discharge
toxins, and smaller plants in environmental sensitive situations,
one might expect the real penalty of being out of control to be
great enough to justify not only daily monitoring, but perhaps,
even more frequent monitoring (hourly or continuously).
Wastewater flows and concentrations vary constantly. They
are influenced by diurnal, weekly and yearly seasonal forces.
Sanitary engineers have made few efforts to define these
variations quantitatively, fewer efforts to use dynamic models
for process design or control, and even fewer attempts to use
time series techniques.
114
-------
One possible reason why time series methods have not been
used more often may be because the data base has not existed. Few
plants have day to day composite sample analyses. It is rare,
indeed, to find frequent grab sampling programs, say, on an
hourly or bi-hourly basis.
Monitoring in Industrial Waste Treatment Plants
Many industries have been using monitoring systems for
process control. They have developed the sensors necessary to
control their process and have been able to do a good job because
they can control their flow rates and their inputs of raw
materials; and they can conduct either a steady state or
controlled batch process. However, when it comes to waste
treatment control they have suffered the same problems as the
municipal wastewater treatment plant.
The waste flows are intermittent, uncontrolled, variable in
quantity, strength and composition and contain a mixture of
water, sewage, organic and inorganic process chemicals, oils,
toxics and non-toxic substances. Historically, most industries
were unconcerned about their waste discharges and only recently
have been faced with the necessity to know what is in their
waste, how to treat it and how to control their discharge. They,
too are faced with lack of good sensors, good automated
monitoring systems and know how.
Some material plants have a distinct advantage over the
municipal plant in that they do have expertise on hand familiar
with automated control systems and sensor operations and mainte-
nance. These plants will have the capability to extend their
existing control systems to include the wastewater monitoring and
control functions. However, there is no profit motive in this
and they will not do so until the regulatory agency demonstrates
that it can detect their effluent permit violations and threaten
prompt action. At that time industry will step up its monitoring
effort in order to protect itself from regulatory actions, and to
provide legal proof that it is providing adequate treatment.
Smaller industrial plants face the same problem as smaller
municipal plants. They do not have automated monitoring systems
for process and control purposes. Even if they have laboratory
facilities and personnel, they may not have the equipment or
personnel with sufficient time, interest or pollution control
knowledge to solve their problems.
Most industrial plants would be very interested in an
automated monitoring package that was complete enough and
reliable enough to provide their effluent monitoring require-
ments. Here again, the availability of such systems is limited
by the availability of adequate sensors.
115
-------
Status of Advanced Technology
Automatic water monitoring by a network of stations is
clearly the best approach to providing data on the scale required
for improved control planning.
The basic chemical and physical parameters to be measured
automatically are fundamentally common to all waterways.
However, technological difficulties have, until now/ restricted
the development of satisfactory automatic water monitoring
stations capable of operating efficiently within a network.
Scientists and engineers have been forced to accept compromises
in their basic requirements for data, and higher costs of
operation.
Laboratories have had to continue tedious gathering of
routine information, rather than concentrating on specialized
investigations and analysis of pollutants requiring techniques as
yet too complex for incorporation into automatic monitoring
on-site.
Automatic water monitoring stations for continuous measure-
ment are currently in use and have demonstrated capabilities of
running unattended for periods of four to five weeks. These
stations are located in housing on the shore of water bodies and
require power lines and communication lines. Such stations use
pumps to deliver the water to sensors and sampling equipment
located inside the housing where it is protected from the
elements and from vandalism and may be heated in winter.
Such systems currently measure pH, Redox (ORP), Cl,
conductivity, dissolved oxygen, temperature, turbidity, level,
flow, wind speed, wind direction, ambient temperature and solar
radiation. These stations can also provide alarms if pre-
determined levels are exceeded or if any sensor, pump or other
component indicates a malfunction. Provision can also be made to
automatically collect a series of samples subsequent to a
predetermined level alarm or from telemetered instruction.
Automatic cleaning and automatic calibration of sensors can
also be provided in such a system.
Experience with such systems indicates that the costs are
still too high, they tend to collect too much irrelevant data and
the stations are vulnerable to vandalism and contamination.
However, these problems will be overcome with greater use of such
systems.
A small, helicopter-borne water-quality monitoring package
is being developed by the NASA/EPA using a combination of basic
in situ water quality sensors and physical sample collector
technology. The package is a lightweight system which can be
116
-------
carried and operated by one person as a passenger in a small
helicopter typically available by rental at commercial airports.
Real-time measurements are made by suspending the water quality
monitoring package with a cable from the hovering helicopter.
Designed primarily for use in rapidly assessing hazardous
material spills in inland and coastal zone water bodies, the
system can survey as many as 20 data stations up to 1.5 kilo-
meters apart in one hour. The system provides several channels
of sensor data and allows for the addition of future sensors.
The system will also collect samples from selected sites with
sample collection on command. An EPA Spill Response Team member
can easily transport, deploy, and operate the water quality
monitoring package to determine the distribution, movement, and
concentration of the spilled material in the water body.
Another Water Quality Monitoring System is being developed
by NASA for use by the Environmental Protection Agency (EPA) in
lakes, rivers, estuarine, or coastal waters for the purpose of
monitoring water quality. This monitoring system will utilize
state-of-the-art technology in the areas of sensors, electronics,
data storage, and packaging; and shall be capable of deployment
and recovery from a small surface craft or from an amphibious or
low hovering helicopter. The system shall measure and record the
concentrations of major constituents and parameters of water in a
selected areas, unattended, for a predetermined period of time.
The system consists of a subsurface data buoy, a portable surface
base station, a special anchoring mechanism, and a sixteen-cell
.water sampling device.
The subsurface data buoy is deployed manually and will
remain anchored in the area of deployment, at a predetermined
height above bottom, until retrieved. Data are collected and
stored internally in a nonvolatile electronic memory. Sensors
shall be selected prior to deployment and data will be taken on a
fully programmable basis. The buoy shall have an acoustic
transceiver to transmit stored data to the surface on command and
to receive control and command instructions from the surface
station.
A minimun of six channels of data will be recorded on a
programmable time share basis. Specific data are determined by
choice of sensor for each channel prior to deployment.
Sensors
It is readily apparent that the primitive state of present
sensor technology is a primary constraint on the application of
advanced technology to the management of water resources. Many
of the needed parameters cannot now be sensed automatically. High
priority must be placed on sensor development because a lack of
adequate sensors probably will impede other developments for some
time to come. We have not committed sufficient resources to
sensor development.
117
-------
The use of aircraft—both high-flying, and low level—
has provided remotely sensed water data that have relevance and
reliability. These include measurements of the water content of
ice and snow, cropcanopy emittance and other parameters related
to subsurface water, the dynamics of surface water flux, and many
of the water quality and environmental constituents. Present
sampling, monitoring, and laboratory techniques are available
and, in most cases, standardized to obtain these measurements.
In order to utilize space technology, accurate automated stations
which can operate service free over long periods of time need to
be developed. Their data output can be relayed by satellite or
aircraft without the need for human intervention and thus
coverage can be provided for many locations, some of which may be
remote and difficult to service.
There are two key areas in which substantial breakthroughs
are needed to develop ground sensors compatible with the capa-
bilities of satellite-borne data collection systems. First,
present equipment is not rugged enough to survive the severities
of weather, vandalism, and environmental contamination for
periods of six months to one year. The improvement required is
largely an engineering problem and should be attainable.
Secondly, present sensors either cannot duplicate the laboratory
analysis required for detection of many parameters or they
require in-place equipment which impedes navigation or other uses
of inland water.
In closing, I want to make a plan for use of imagination in
sensor development. It is essential that sensor technology be
sensitive to user needs, but it is also important that new
capabilities not be forced uselessly into old patterns. Since
developers of sensors and water resources experts usually are not
expert in each other's field, a continuing dialogue between the
two is required. The result can be beneficial to both and may
produce new approaches in the management of our nation's water
resources.
118
-------
CANADIAN PROGRAMS
In addition to a review of automated water quality sensor
applications, research and needs in the United States, Workshop
participants heard a description of Canadian activities in this
area by Mr. Kenneth Birch of the Canada Centre for Inland Waters.
Prior to the Workshop, communication with various U.S. scientists
and administrators produced references to active water quality
sensor research and leadership in several Canadian agencies. When
Canadian representatives were contacted, they were not only re-
ceptive and willing to contribute to the Workshop by sending an
individual to attend as a special guest, but were highly sup-
portive of such a gathering. The program description
narrative, summarized by Mr. Birch in his oral presentation, is
given on the following pages.
119
-------
CANADIAN PROGRAMS
by Kenneth N. Birch
Canada Centre for Inland Waters
Burlington, Ontario, Canada 17R4A6
It is estimated that Canada has about 15 percent of the
world's known freshwater resources. Management of these water
resources is shared by the federal government and ten provincial
governments. The provinces have jurisdiction over property and
civil rights and can legislate on their own local water resource
matters. The federal government has jurisdiction in specific
areas, including national parks, northern territories and situa-
tions where fisheries and navigation matters are concerned or
where international and interprovincial agreements are made. The
activity of monitoring the quality of these various bodies of
water is conducted by both levels of government, primarily in the
traditional manner of bottle sampling, followed by analysis in a
laboratory. (1) Something in the order of 5 percent of this
total monitoring effort makes use of in situ sensors or automated
field analyzers.
There are 22 automated river monitoring stations currently
operating continuously, year round, in Canada. All of these
stations are relatively permanent installations involving
significant facilities on each site. The locations are shown on
Figure 1, while the details about the site, the equipment and the
operating agency are outlined in Table 1. The parameters:
temperature, pH, specific conductance and dissolved oxygen are
continuously measured from a sample pumped from a single intake,
which is generally located off the bottom in mid-channel. The
sensors are installed in commercial robot monitor systems housed
in trailers or local buildings. Heating systems and buried
waters with high suspended load, floating debris and ice cakes
can cause problems in the spring. Most rivers have vigorous
algal growth in the summer which necessitates more frequent
system cleaning. The following paragraphs give a brief outline
of the equipment and mission of the various stations. (2)
The Province of Alberta operates a network of six Schneider
Model 25 Robot Monitors for general surveillance and spills moni-
toring. Sites are typically located up and downstream of major
centres and industrial plants. Data not recorded locally on
stripcharts are also telemetered to Edmonton to a central
computer facility where they are merged with air quality and
stream flow information.
120
-------
The Water Quality Branch of the Department of Fisheries and
the Environment has operated a Schneider Robot Monitor on the Red
River at Emmerson, Manitoba since 1968. Two new Phillips PW9835
monitors have been recently installed at the U.S.-Canada border
crossing of the Souris River. These stations are part of the
SOD, nutrient and dissolved solids monitoring programs in
connection with the International Joint Commission and the U.S.
Garrison Diversion. The Emmerson station is located in a local
water treatment plant and is serviced by the plant operator. It
was recently instrumental in the detection and effective documen-
tation of a short duration, but serious, oxygen depletion problem
on the Red River. (3) The Souris River stations are equipped
with the Phillips ultrasonic electrode cleaners, the automatic
high/low level standardization manifolds, and automatic, refrig-
erated samplers. These stations are scheduled to have GOES
satellite telemetry to Saskatoon operational by summer "78.
The Saint John River System is monitored by a network of six
stations operated by the Federal Government in co-operation with
the Province of New Brunswick. (4) The network is a key part of
the monitoring of the transboundary pollution in tributaries to
the Saint John River. The monitoring equipment, made by
Automated Environmental Systems, has been extensively reworked by
its operators and now gives reliable service. The stations are
tied together by a leased telephone line and are polled
automatically by a PDP8/L based data acquisition system located
in Moncton.
The Province of Ontario operates three Schneider Model 25
Robot Monitors and two NERA-Hydrolab units in support of special
harbour studies and more recently, studies of pollution from land
runoff (PLUARG) in the Grand River. These sites are less
permanent than the previously mentioned stations and are serviced
on 5- to 30-day schedules. In addition, they also operate two
Plessey MM4, self-contained, submersible monitor packages. Data
are used with numerical models, together with stream gauge and
recording current meter data. (5)
The operators of these river monitoring stations have
learned to overcome most of their monitor apparatus problems. In
general the temperature, pH, conductivity and dissolved oxygen
sensor performance is satisfactory. Six- to ten-percent data
loss with 10- to 12-week service periods is reported on the more
established system. There is not yet enough experience with the
ultrasonic cleaners in the Phillips system to comment
meaningfully on the effectiveness, although satisfactory 1-month
unattended service over a year was obtained by the National
Research Council experimental station on the Ottawa River.
121
-------
The universal problem seems to be turbidity sensors. Hach,
AES and Schneider models have all been tried at several sites and
all were found to drift too quickly to be of present value.
Turbidity in itself is not considered to be important enough to
increase the service frequency at the stations for this
parameter. However, if a sensor could be made which had good low
range sensitivity, dynamic range from 1 to 10,000 ppm P, and
reliable cleaning mechanisms (handling algal growth, suspended
material and oil films), then it would be used in many of the
Canadian stations. There are some who believe that useful
statistical information, related to factors like nutrient
loading, can be found from continuous turbidity measurements.
Another problem area seems to be drifting reference elec-
trodes as used for pH and chloride ion measurements. Some of the
solid state combination electrodes have been found to work best
in the Saint John's network stations.
Apart from the river monitors, Canadians are also making
good use of the new types of portable water quality monitors like
the Hydrolab Surveyor - NERA system and the Montedoro Whitney
Mark II system. These systems have a surface readout and
recording unit with sensor sonde which can be lowered to
100 meters with sensors for temperature, depth, pH, conductance
and dissolved oxygen.
Regional biologists and limnologists are finding these in-
struments very useful for surveying the hundreds of small lakes
in their regions which they must characterize. Measurements in
profile can be quickly taken and plotted on site, such that
adequate surveys of small lakes can be completed in hours. Fur-
thermore, these quick surveys permit selection of water sampling
sites where the water will be representative of specific zones
within the lakes. The net result is a reduction in the number of
samples to be carried back to the analytical lab, and an
enhancement of coverage within a shorter time than could be
realized without these instruments. Generally, users of this
equipment experienced a variety of annoying problems initially
until they developed their own procedures of transporting,
operating and calibrating and after the manufacturers had
corrected assorted faults. Defective power packs and leaking
cables and connectors were the frequently reported faults.
The Canada Centre for Inland Waters uses water quality
sensors in a variety of special systems put together in support
of particular research projects. One system (6) automatically
measures dissolved oxygen and temperature profiles in three limno
corrals by gradually lowering a YSI oxygen sensor through the
centre of each corral in turn. The system includes a motor
driven jiggling mechanism to effect the necessary flow past the
122
-------
sensor and allows an atmospheric calibration check. A separate
system (7), used for estimating primary production in these limno
corrals, measures pC02 in the air and the water by an IR
absorption instrument equipped with an automated sample standard
sequencer and packaged for unattended service on a raft. A
commercial turbidity meter (8) employing a transmitted/scatter
light ratiometer technique was packaged for an in situ profiling
application where it was required to trace large sections of
water with high suspended load. Adams (9, 10) used two Datel
LPS-16 data logger systems to monitor a special array of sensors
(22 thermistors, 4 Hydrolab conductivity, 10 wavelength selective
irradiance) mounted on an underwater tower in support of studies
on effects of oil spills in Arctic lakes.
When asked to identify needs for new sensors the user com-
munity in Canada responded with requests for nutrient monitors
which are sensitive and reliable over several weeks. This re-
quirement stems from a need to estimate trends of loadings into
rivers and lakes within a timeframe of only 1 or 2 years. There
is a high degree of temporal variability in nutrient concentra-
tions and discharge such that conventional grab sampling tech-
niques are questionable for this purpose. Nutrient loadings are
also becoming reference items in international treaties on
transboundary water movements. It is not clear just what
nutrient parameters should be measured for this purpose, but most
users would suggest total and orthophosphorus, total nitrogen,
nitrate and ammonia. These must be measured over the range 1 to
0.005 ppm as N or P with accuracy on the order of 20 percent of
reading at the low end range being acceptable. There have been
attempts to use a Technicon CSM-6 to continuously monitor
orthophosphorus, ammonia and nitrate at a test site on the
Red River in Emmerson. It was found that the equipment needed at
least daily standardization and frequent cleaning of the plumbing
and colorimeter. The filter systems provided with the equipment
would repeatedly clog under the conditions at the site.
Nevertheless, with one junior technician assigned to attend to
this monitor, a year long record was obtained. The experience
demonstrated the high variability in these parameters and the
lack of reliable equipment to monitor them.
There are other parameters for which sensors would be re-
quired. Total organic carbon or total carbon in the range 5 to
20 + mg/ was requested from some users, but no one had eval-
uated the UV types now oh the market, or had plans to do so.
Sensing devices for hydrocarbons and toxic substances were also
requested to aid in detecting spill and special runoff situa-
tions.
Much of the research and development work in water quality
monitoring is being done in support of laboratory based, analyt-
ical technology. There is no agency in Canada with a specific
mandate to develop or evaluate sensors for water quality
123
-------
monitoring in general. Nevertheless, there is some activity,
satisfying some particular research interests, in the
universities, National Research Council and the major water
resource centres in Canada.
The Analytical Methods Research Section at CCIW has been
active in the development of new types of selective ion elec-
trodes and in applying them in semi-automated systems. Sekerka
and Lechner (11) have developed an inexpensive solid state
chloride ion-selective electrode based on HgS/Hg2Cl2 membrane
showing overall better performance characteristics than those
based on AgCl membranes. It has a response over the concentra-
tion range 1 x 10"1 M to 5 x 10"? M Cl , an electrical imped-
ance in the 100 ohm range, and a response time in the five second
range (at 1 x 10~3 M C). A prototype of this electrode is
being evaluated in a robot river station in the St. John Network
where it is reported to work well in low chloride ranges. The
HgS matrix has been applied (12) to electrodes for other ions in-
cluding bromide ( to 1 x 10~7 M) , thiocyanate (to 5 x 10~7 M),
cyanide (to 5 x 10~6 M) , and iodide (to 1 x 10~7 M) . Analy-
tical behaviour in respect to selectivity, sensitivity and
response is superior to electrodes based on
More recently a new zero-current chronopotentiometric
technique (13) with ion-selective electrodes has been developed.
In principle, it utilizes electrodes and gas-permeable membranes
for measurement under non-equilibrium conditions, where the speed
of the potential change of the sensing electrode is related to
the concentration of the ion of interest in the sample. (This
technique is capable of determining a variety of species,
assuming that a system of chemical reactions leading primarily to
the formation of a gaseous species, which passes through a
hydrophobia semipermeable membrane, and reacts with ions of the
internal solution of the sensor, produces a change in potential
of the ion-selective electrode. ) A working laboratory prototype
has been built and used to measure CN~. In operation a sealed
sample is treated with a hexamine buffer to pH 5.5 and the sensor
electrode is mechanically removed and replaced over the membrane
so as to trap about 1 M/ of fresh inner electrolyte. As the
gas formed passes through the membrane it reacts with a known
initial concentration of AgNC-3 at pH 11.5 and produces a chang-
ing potential at the silver ion-selective electrode. With a
15-second measuring time, AE readings of 10 to 130 mV are
obtained for 10 to 100 ppb CN~respectively. The technique has
been applied to 803- and is expected to be applied to NH4+,
HCC-3, N02~f and P- with detection limits in the order of
107 technique, there is hope of being able to make a new family
of water quality sensors exploiting its features of solid state
electrodes, high selectivity and sensitivity, simple sample
pretreatment and potential for neat in situ probe-like packaging.
124
-------
The need for a relatively inexpensive dissolved oxygen
monitor capable of unattended, in situ service for several months
has been another area where specific sensor research is being
sponsored. Membrane fouling and aging limits the use of
available membrane diffusion-limited voltametric cells as sensors
for dissolved oxygen over long time periods. Fowler and Oldham
(14) have developed a cell that employs a pulsed mode of
operation and special signal processing called semi-integration.
It works on the principle that oxygen will diffuse through a
semipermeable membrane to establish partial pressure equilibrium
with a small pocket of internal electrolyte, such that when a
cathode in that electrolyte suddenly falls to a pot ntial where
oxygen can be reduced, then the pulse of current which flows can
be semi-integrated to produce a single, m, directly proportional
to the dissolved oxygen partial pressure, aQ?, This can be
shown as:
m
= (nAFKv/D). ao2
where n is number of electrons, A is the area of the cathode, F
is the Faraday, K is the reaction rate constant, and D is the
coefficient of diffusion in the electrolyte. The current is
independent of membrane permeability terms and relatively insen-
sitive to temperature and convection factors. For continuous
monitoring of dissolved oxygen, the cell is pulsed for 5 seconds
every 15 minutes. Also, because the cell rests open circuit, it
is very amenable to applications where arrays of DO sensors are
needed. Unfortunately, we have been unable to produce a
practical field-usable sensor based on this design, apparently
because there is a significant background signal which is
difficult to control by the fabrication ideas attempted so far.
In lakes research and surveillance programs there has been
an outstanding need for an in situ oxygen sensor which could
quickly and accurately measure the oxygen profile in lakes.
Present day dissolved oxygen sensors respond too slowly to
temperature and oxygen changes for them to be operationally
useful in profiling instrument systems. CCIW has a study in
progress to design and test a prototype oxygen sensing device
capable of at least 0.5 second response to both temperature and
oxygen. Essentially this work is following the line of medical
researchers who have already demonstrated in vitro millisecond
response times in blood by miniature membrane-covered probes.
In 1973 the Instrument R&D Unit at CCIW undertook a project
to develop a prototype of a new type of water quality monitoring
system based on a "Robot Experimenter" concept first proposed by
Birch. (15) The research need addressed was that of getting
economical in situ and time-series data for a variety of water
quality parameters of importance in lake and river studies.
125
-------
Since suitable sensors for most of these parameters are not
available, it was decided to approach the problem by constructing
an underwater robot laboratory and programming it to emulate the
analytical procedures by which water quality parameters are
usually defined. To implement this approach, a family of
submersible apparatus modules (e.g. pumps, valves and electrode
packages) was created. The underwater robot sensing head is made
up of a number (maximum of 18) of apparatus canisters and two,
20-cm-diameter pressure cases containing electronics held
together and serviced by a main frame structure. This sensing
head is placed on the lake bottom and is connected to shore by a
cable which carries power, data and control-signals. An
Interdata 74 minicomputer (8K x 16-bit memory) is sited onshore.
Resident software, specifically developed for this system, allows
experiment routines to be written in a concise, interactive,
interpretive language called REM code. It is arranged so that a
central computer or a user at a data terminal can telephone the
shore based minicomputer to get a readout of data and status
conditions or to modify the experiment routines. The flexibility
afforded by the modular apparatus and software control makes this
system very useful for developing and field testing new ideas for
sensor devices and sampling techniques'. A recent configuration
(V2) of this system operated last year and measured hourly
observations of temperature, pH, chloride ion, conductivity,
dissolved oxygen, turbidity and total alkalinity from three
intakes fixed by a float line at 0.2, 2 and 7 meters off the
bottom at a site 2 km offshore in Lake Ontario. This system
configuration is being evaluated for time series monitoring
applications related to lake surveillance programs. A sample
intake profiling (SIP) subsystem has been built for the
underwater station so that samples can be taken for analysis from
any point or depth interval and so that readings of temperature,
dissolved oxygen and turbidity can be taken in profile.
The continued demand for inexpensive, reliable, unattended,
nutrient monitoring capability and the apparent lack of excite-
ment with available continuous (auto-analyzer styled) monitors
have prompted the development of a phosphorus monitor based on a
micro-batch analyzer concept. The monitor will use a modified
form of the familiar ammonium molybdate colorimetric method em-
bodied in specially designed apparatus. It will work under com-
puter control to perform the sample preparation, digestion or
hydrolysis, reagent addition, colour development and measurement
steps in a sequential, timed order. In concept, it will analyze
for several forms of phosphorus, one after the other, with self-
standardizing methods and be capable of unattended operation for
weeks or about 3,000 data points. The first prototype is being
built as a module for the Robot Experimenter system utilizing the
computer control, data handling and sample taking facilities that
the system provides. The apparatus and techniques developed
should be extendable, with the use of modern micro-computers, to
other parameters and system configurations.
126
-------
Dauphinee of the National Research Council has developed a
prototype zooplankton counter using a conductance cell designed
for towing in seawater. (16) The system uses a D.C. current,
tubular conductance cell (3 to 5 mm diameter and 1 to 2 cm long)
with evenly spaced ring electrodes. Plankton, being near
insulators relative to seawater, produce a characteristic signal
which can be distinguished from other spurious signals and which
contains information on the size and length of the plankton.
In my opinion the most promising technological improvements
are to come from good system analysis and design. This will
suggest which water quality factors should be sensed and what
equipment configurations and sensor devices would be most
appropriate to do that job. I sense that managers are expressing
their need for monitoring equipment in terms of their present
surveillance programs and are limited by the concepts that went
into the legislation that authorizes such programs. For example,
the so-called "parameters" are often defined by complex
analytical accuracy with that achieved by these methods. I
believe there are many instrumental techniques which are
sensitive to water quality changes and which could be developed
as valuable tools in automated monitoring stations, but their
value cannot be fully appreciated as long as they must compare to
conventional laboratory practice.
System design at the monitor station level should be
questioned. Since only a few millilitres of water are really
necessary for any water quality measurement and since only one to
six samples per hour are generally adequate to describe
variations, then is it really necessary to pump huge volumes of
water continuously, suffering the cost of installing, main-
taining and operating the engineering works that provide it? This
need for major site engineering works can be avoided by
miniaturizing the plumbing apparatus and packaging sensors so
that the entire monitor can be planted in the stream directly.
Furthermore, a stable thermal environment without overheating and
freezing conditions is afforded by going underwater.
The Robot Experimenter system, mentioned previously, has
demonstrated the feasibility of such a submersible station in
concept and in many practical ways as well. Of course there are
some new engineering problems to be faced in designing such
self-contained underwater stations, but there are many advantages
to be exploited as well.
Another point of system design which should be reviewed
against other alternatives is the need for continuous monitoring
and its corollary, parallel direct readout of each parameter. One
attractive alternative is frequent batch sampling with some
sequential and parallel analysis. In addition to the energy
saving in the pumping operation alone, there are advantages to
127
-------
be gained from the fact that fouling and reagent consumption are
generally proportional to sample pumping time. Many presently
available and accepted analytical techniques become more practi-
cal in batch operation as we hope to demonstrate with the phos-
phorus monitor project at CCIW. The common practice of having
completely independent sensor-electronics for each parameter re-
sults in duplication of hardware with no reliability benefit. By
reducing the hardware to the smallest complement and by employing
common signal conditioning and computer compensation techniques,
then it should be possible to reduce the cost per monitor.
128
-------
REFERENCES
Gale, R.M. , and A. Demayo, "Water Quality Data Collection
Programs, Environment Canada," Proceedings of the Inter-
national Seminar and Exposition on Water Resources and in
strumentation, 1974, V4, p 8-19.
Birch, K.N. et al . , "Automated Water Quality Monitoring in
Canada - Present Practice and Trends for the Future," Proc.
Water Research Centre Conference, Reading, England, 1975.
Gummer, Win. D. "The Red River and Low Dissolved Oxygen for
the Period March 31 to April 6, 1975." A report submitted
to the International Red River Pollution Control Board,
1975.
Cullen, D.H., "Automatic Water Quality Monitoring within
Saint John River Basin." ASTM Special Publication, Water
Quality Parameters, STP-573, 1973.
Palmer, M.D., and D.J. Poulton 1973. "Water Chemistry Data
from the St. Clair River at Corunna, Ontario, as Determined
by Continuous Water Quality Monitor." Proc. 16th Conf .
Lakes Res. p. 309-320.
6. CCIW Engineering Services. "D.O. Profiling System," Handbook
Notes, ESS-1035, 1974.
7. Weiler, R.R. "Carbon Dioxide Exchange and Productivity in
Lake Erie and Lake Ontario." Presented at 23 SIL Con-
ference, Winnipeg, 1974.
8. Desrosiers, R.M. "A Wide Range Submersible Ratio Turbidity
Sensor for Use in Canada's Inland Waters," CCIW unpublished
technical note, 1974.
9. Adams, W.A. , B.F. Scott and N.B. Snow. "Environmental
Impact of Experimental Oil Spills in the Canadian Arctic."
ASTM Special Publication, Water Quality Parameters
STP-573, 1973.
10. Adams, W.A. "Continuous Water Quality Monitoring Associated
with Experimental Oil Spills." Technical Bulletin, Inland
Waters Directorate, Department of the Environment.
129
-------
11. Sekerka, I., and J.F. Lechner, "Chloride Ion-Selective Elec-
trode Based on HgS/Hg2Cl, "Journal of Electroanalytical
Chemistry, 57(1974), p. 317-323.
12. Sekerka, I., and J.F. Lechner, "Preparation and Evaluation
of Halide Ion-Selective Electrodes Based on HgS Matrices."
Journal of Electroanalytical Chemistry, in press.
13. Sekerka, I., and J.F. Lechner, "A New Zero-Current Chro-
nopotentiometric Technique with Ion-Selective Electrodes,"
Analytica Chemica Acta, 93 (1977) p. 129-137.
14. Fowler, J.K., and K.B. Oldham. "Voltammetrie Membrane Cell
Used in the Equilibrium Mode for Dissolved Oxygen Assay."
Chemistry and Physics of Aqueous Gas Solutions, Special
Publication of the Electrochemical Society, 1975.
15. Birch, K.N. "REX - A Computer Controlled Robot for Water
Quality Monitoring." ASTM Special Publication,
Water Quality Parameters STP-573, 1973.
16. Dauphinee, T.M. "Zooplankton Measurements Using a Con-
ductance Cell," Oceans 77, 39B-1 MTS-IEEE.
130
-------
TABLE 1: MAJOR RIVER MONITORING SITES IN CANADA
MAP
KEY
1
2
3
4
5
6
7
8
9
10,11,
12
13
14
15
16
17
18
i *._ _ i- j".
WATER BODY
North Saskat-
chewan
Bow River
Bow River
Oldman River
Red Deer River
Wapiti River
Qu1 Appelle
River
Red River
Red River
Toronto Harbor
Ottawa River
St. John River
Aroostook River
Presquile River
Meduxnekeag
River
Kennebecassis
River
SITE
Vinca
Bowness
Calgary
Drumheller
Grande Praire
Edenwald
St. Agathe
Emmerson
(3 stations)
Ottawa
Grand Falls
Dam
Tinker Dam
Can/US Border
Belleville
Apohague
MONITOR
AGENCY MANUFACTURER
Alberta
Alberta
Alberta
Alberta
Alberta
Procter and
Gamble
Saskatchewan
Federal
Water Survey
Federal Water
Quality Branch
Ontario
Ntl Research
Council
Federal Water
Quality Branch
Federal Water
Quality Branch
Federal Water
Quality Branch
Federal Water
Quality Branch
Federal Water
Quality Branch
Schneider
Schneider
Schneider
Schneider
Schneider
Schneider
Schneider
In house
developed
Schneider
Schneider
Phillips
A.E.S.
A • Ei • O •
A » Ei • O •
A.E.S.
A.E.S.
131
-------
TABLE 1. (Continued)
MAP
KEY
19
20
21
22
23
WATER BODY SITE
St. John River Beechwood Dam
St. John River Mact aquae Dam
Sour is River Canada/U.S.
Border
Souris River Canada/U.S.
Border
Bow River Medicine Hat
MONITOR
AGENCY MANUFACTURER
Federal Water
Quality Branch
Federal Water
Quality Branch
Federal Water
Quality Branch
Alberta
Alberta
A.E.S.
A.E.S.*
Phillips
Phillips
Schneider
* presently decommissioned
132
-------
U)
u>
STJOHN RIVER
BASN NETWORK
Figure 1. Major river monitoring sites in Canada
-------
UNIVERSITY RESEARCH PROGRAMS
Representatives of six universities active in water quality
sensor research and development were invited to present discus-
sions of their research activities and participate in Workshop
work panel discussions. Each was requested to prepare a written
narrative for inclusion in this Workshop report, as well as give
a 20-minute oral presentation on his topic. The narratives are
given on the following pages of this report. The participat-
ing university representatives were:
Dr. Roger Bates University of Florida
Dr. Khalil Mancy University of Michigan
Dr. Walter Blaedel University of Wisconsin
Dr. F. H. Middleton University of Rhode Island
Dr. Richard Newton Texas A & M University
Dr. Charles Whitehurst Louisiana State University
134
-------
RESEARCH ON ELECTROCHEMICAL METHODS FOR ON-SITE DETERMINATION OF
TRACE METAL IONS IN NATURAL WATER SYSTEMS: Anodic Stripping
Voltametry with Collection
by W. J. Blaedel, Ph.D.
University of Wisconsin
Madison, Wisconsin 53706
Introduction
On-Site Trace Analysis by Electrochemical Methods--
Electrochemical methods of analysis have great capabilities
for on-site monitoring. Almost all such methods involve the ap-
plication and measurement of potentials up to a volt or two,
together with the observation of currents that are of the order
of microamps. Equipment and techniques for such measurements are
basically very compact and simple, and their use need not be con-
fined to the controlled environment of the large, central an-
alytical laboratory.
Equipment and instrumentation that are commercially avail-
able and in general use have usually been designed for high
stability, high sensitivity, high accuracy, long life, and
versatility, and these qualities tend to locate the instruments
in our analytical laboratories. Some important requirements for
on-site use in the field are simplicity, portability, and econ-
omy, and these are not generally compatible with the require-
ments for laboratory instrumentation. To adapt existing or
conventional equipment and methods to on-site use therefore often
requires a compromise, in which some desirable characteristics
are relinquished or achieved by other means, in favor of
characteristics that are desirable for on-site use. The on-site
application of anodic stripping voltametry (ASV) may be achieved
by such compromise.
In principle, electrochemical methods of analysis are highly
sensitive. The potential at an electrode in a chemical system is
generated at the electrode surface—it is an interfacial
phenomenon—and to affect that potential, only a monolayer of
electroactive material may be needed for a potentiometric
measurement. An amperometric or coulometric measurement can usu-
ally be made easily with the consumption of 10""^ coulombs (one
nanoamp flowing for one second), which corresponds to lO""^
equivalents transformed, and which approaches the femtomole level
of detection. In any case, the parts-per-billion (ppb) levels
encountered in environmental analysis appear to be well within
the reach of electrochemical methods. For some elements, the
sensitivity of electrochemical methods exceeds the sensitivities
of the more popular atomic absorption methods. Even the
sensitivity of activation analysis for some elements may be ex-
ceeded by electrochemical methods.
135
-------
In describing the appropriateness of electrochemical
methods for the measurement of ppb concentration levels, it is
important to avoid the impression that these levels are easy to
measure. On the contrary, the measurement of ppb concentrations
has been likened to trying to measure the golf ball concentration
of a golf ball in Lake Mead. While this comparison is not ac-
curate, it does point up a very important consideration in low
concentration measurements: getting the material to the sensor
surface. While sensor quality and sensitivity are necessary,
they are not sufficient. For example, installation of the best
and most sensitive golf ball detector in the world at the north
end of Lake Mead would be of little avail if the golf ball is
located near the south end of Lake Mead. Or, a bucket-dipping
type of sampling operation would result in a lot of samples with
zero golf balls per bucket. (In fact, if a lucky bucket sample
were obtained with the golf ball in it, that result would prob-
ably be rejected on statistical grounds for being out of line
with other replicate samples!)
The message in these remarks is that, in trace analysis, the
operations of sampling, enrichment, and separations are
intimately bound up with the sensing step, and may require more
effort than the sensing step. This is particularly true for
electrochemical measurements, which are not highly selective. A
main goal of on-site methods is to reduce the sampling effort,
and to eliminate the effort, delays, and errors associated with
the storage and processing there before the sensor measurement is
made.
Anodic Stripping Voltametry (ASV)—
In ASV, the trace metal ions in a water sample are deposited
on an electrode held cathodically for a period of time under con-
trolled conditions. After a sufficient amount has been deposited
the electrode potential is scanned in the anodic direction. The'
trace metals are oxidized, beginning with the most active, giving
current peaks whose location on the voltage scale characterizes
the metal ion, and whose peak height is proportional to the con-
centration of the metal ion in the original sample solution. ASV
is applicable to 10-20 elements at the ppb level—ones which can
be electrodeposited. The elements of particular interest are Cu,
Pb, Cd and Zn.
Like any sensitive method, ASV suffers some shortcomings. A
principal disadvantage is encountered in the stripping process,
as the electrode is being made anodic. This is accompanied by a
large current, called the charging current, which gives a high
and variable background, and which tends to obscure the smaller
stripping currents. The electrode material is mercury, usually
in the form of a hanging drop. Mercury is used because of its
high overvoltage for hydrogen liberation, which interferes with
the deposition of active metals like zinc at other electrode
materials.
136
-------
Anodic Stripping Voltametry with Collection (ASVWC) (1,2)
Summary Description of ASVWC—
ASVWC is a modification of ASV designed to eliminate charg-
ing current. Equipment is schematized in Figure 1, which re-
presents a tubular channel through which the sample solution is
passed. Two tubular electrodes separated by a small gap are
located on the inside of the channel. For the deposition of the
trace metals, the upstream electrode (lower electrode in Figure
1) is made cathodic, and the sample is passed through for a con-
trolled length of time (1-10 minutes). The downstream electrode
is kept anodic during the sample deposition, to prevent the ac-
cumulation of plated materials. For the stripping step, a cath-
odic potential is applied to the downstream electrode, and the
upstream electrode potential is increased linearly with time. As
the metals are stripped from the upstream electrode, they are
carried by the flowing solution to the downstream electrode,
where they are deposited, giving a peak in the plating current.
With properly controlled conditions, the peak height of the plat-
ing current is proportional to the concentration of the trace
metal in the original sample.
Equipment—
The electrodes are housed in an assembly of two Plexiglas
blocks (Figure 2), held together with bolts (not shown). The
lower block contains a cavity with two 1-mm thick glassy carbon
plates separated by a Teflon sheet spacer (0.010" thick) epoxied
into the bottom. The solution channel (2 mm dia.) is formed by
drilling the assembly to give the tubular electrode configuration
depicted in Figure 1. The upper Plexiglas block contains a
silver-silver chloride reference electrode wound around a post in
the cavity. Bridging between the tubular and reference elec-
trodes is achieved through a glass frit washer or stack of ion
exchange membrane washers, sandwiched between the end of the post
in the upper block and the floor of the cavity in the lower
block.
Samples are introduced from a covered beaker, and air is re-
moved by sparging with a stream of nitrogen. Flow of solutions
through the cell is measured with a rotameter on the effluent
end, with flowrate control achieved by regulation of the head of
solution. The reference electrode compartment is flushed with
0.1 M KC1 (0.3 ml/min), to prevent accumulation of contamination
by slow leakage of sample solution through the bridge.
Startup includes electrodeposition of a mercury film onto
the glassy carbon electrodes, which lasts for many de-
terminations. Shutdown includes stripping of the mercury film
and scrubbing the tubular electrodes with a pipe cleaner and de-
tergent solution.
137
-------
Characteristics of the System--
In systems of small dimensions, hydrodynamic regimes are
highly stable and reproducible, and many are mathematically
describable. For ASVWC with tubular geometry, the dependence of
peak collection current (ipc) on the operating and system
parameters is theoretically derivable:
ipc = 2.023 N IQII2 vD2/3Vfl/3 X2/3 C T
RT
N is the collection efficiency at the downstream electrode, v is
the anodic voltage scanning rate at the upstream electrode, D and
C are the diffusion coefficient and bulk concentration of the
electroactive analyte, Vf is the volume flow rate of solu-
tion, X is the tubular electrode length, and T is the deposition
time. Some of the dependences have been verified experimentally.
For example, Figure 3 is a log-log plot showing the cube-root de-
pendence of the peak collection current on flow rate.
Applications—
Figures 4 and 5 represent ASVWC analyses on synthetic
solutions prepared from triply distilled water and reagent grade
chemicals. Figure 4 shows the dependence of peak stripping and
collection currents on stripping potential for ppb levels of Cu
Pb and Cd. The advantage of measuring collection currents instead
of the conventional stripping currents is shown clearly. For ex-
ample, the collection current peak for copper is easily percepti-
ble and measurable, but the corresponding stripping current peak
is not even perceptible. Figure 5 shows the collection current
peaks given by 0.02 ppb of Cd and 0.10 ppb of Pb.
Table 1 shows results for the analysis of copper in the city
of Madison tapwater by several methods. Study of the table
indicates that ASVWC results do not represent the total copper
content as determined by atomic absorption. This is ascribed as
soluble complexes and colloidal particulates (3). Further ex-
amination of Table 1 also indicates that the ASVWC results do not
represent the copper ion activity as determined with the copper
ion-selective electrode, or by an ion exchange procedure that is
presently under development (4). Apparently, many of the com-
plexed or bound forms of copper that do not contribute to the
cupric ion activity are electrodepositable and do contribute to
the ASVWC copper content. These findings are not at variance
with those of other workers who have investigated ASV.
138
-------
TABLE 1. COPPER IN TAP WATER
Method
ASVWC
pH 7.1
pH 5.0
Copper ISE
pH 7.1
pH 5.7
Ion exchange
pH 7.1
pH 5.7
Atomic absorption
Copper Ion Concentration yumol
160
132
2 (activity)
80 (activity)
2 (activity)
58 (activity)
240 (total)
Conclusions
Apparatus and techniques are described for ASVWC that appear
to have the capability for on-site determination of Zn, Cd, Pb,
and Cu in natural waters. On-site capability has been achieved
by the following modifications of conventional instrumentation
and equipment: (1) use of a mercury-coated tubular glassy carbon
electrode instead of conventional hanging drop mercury electrodes
or rotated ring-disk electrodes; (2) replacement of versatile
commercial polarographic instrumentation with simple battery-
operated solid-state circuitry for the application of potentials
and the measurement of currents; (3) use of a small battery oper-
ated recorder for the recording of currents instead of con-
ventional high-precision line-operated laboratory instruments.
Adaptation of laboratory-type equipment for ASVWC has been
realized by relatively simple modifications and compromises that
are within the expertise of the analytical chemist. However,
automation of ASVWC requires duplication of the manipulations of
the operator who performs the analysis. Such manipulation
involves not only operation of the equipment and instrumentation,
including startup and shutdown, but also selection of the sample
and its transport through the various stages of the analysis.
The automation is largely a research problem in mechanical en-
gineering. If unattended operation over a long period of time is
required, the automation effort will be difficult, time-
consuming, and expensive. These disadvantages should be weighed
139
-------
against the disadvantages of on-site use by an operator, or
against the disadvantages of only partial automation.
Acknowledgement
This work was supported in part by the Environmental Protec-
tion Agency (Grant No. 804179-02-1) and by the National Science
Foundation (Grant No. CHE-7615128).
References
1. G. W. Schieffer, and W. J. Blaedel, Anal. Chem., 49, 49
(1977). —
2. G. W. Schieffer, and W. J. Blaedel, Anal. Chem., 50, 99
(1978).
3. W. Davison, and M. Whitfield, J. Electroanal. Chem., 77 59
(1977). —
4. R. A. Niemann, "Donnan Equilibrium Applied to Enrichment
and Speciation by Ion Exchange for the Determination of
Submicromolar Cationic Activity". PhD thesis, University
Wisconsin-Madison, 1977.
List of Illustrations
Figure 1. Electrode configuration for ASV.
Figure 2. Twin tubular electrode cell for ASVWC.
(A) O-ring; (B) Cation exchange membranes; (C) Lead
to upstream stripping electrode; (D) Cast epoxy;
(E) Sample solution inlet; (F) Teflon spacer;
(G) Lead to downstream collection electrode;
(H) 0.1 M KC1 inlet; (J) Lead to reference electrodes
(SSCE's); (K) Sample solution outlet; (L) 0.1
M KC1 outlet.
Figure 3. Dependence of peak stripping current flow rate.
0.1 MM lead in 0.1 M HC1. One minute deposition time
1 V/mTn.
Figure 4. Stripping and collection peaks for 6 nM Cd, 7 nM Pb
and 6 nM Cu in 0.04 M acetate buffer. ~ '
5-min deposition time. 1 V/min. Collection electrode
voltage, -0.80 v.
Figure 5. Collection peaks for 0.017 ppb Cd and 0.10 ppb Pb
in 0.01 M acetate buffer.
12-min deposition time. 1 V/min. Collection
electrode voltage, -0.80 V.
140
-------
irt
Figure 1. Electrode configuration Figure 2. Electrode cell
200
0.5
12 5 10
FLOW RATE, ml/MIN
Figure 3. Stripping current dependence
COLLECTION
Cd(ll)
-0.8 -0.4 0.0
POTENTIAL OF STRIPPING EUECTRODE. VOLTS
Figure 4. Stripping and collection
peaks
-o.e -0.4
POTENTIAL OF
STRIPPING ELECTRODE, VOITS
Figure 5. Pb collection peaks
141
-------
A LIDAR POLARIMETER FOR WATER QUALITY MONITORING
by R. W. Newton, Ph.D.
Texas A&M University
College Station, TX 77843
Introduction
For several years the Remote Sensing Center has been
involved in programs for developing methods of measuring water
quality parameters. Several devices have been investigated
including infrared photography and multispectral scanner systems
Controlled measurements have demonstrated that these techniques
have a potential for quantitative measurements of certain water
quality factors such as total suspended solids and presence of
algae (1), (2). Although worthwhile, these techniques cannot be
easily incorporated into in situ measurement devices. in ad-
dition, the resulting parameter analysis is generally not real-
time .
A new and unique approach to measuring water quality param-
eters has resulted from the development of an oil spill de-
tection system for the U.S. Coast Guard. The system is a dual
polarization laser "radar" termed a lidar polarimeter. It was
designed to be located at a fixed position in a harbor area, and
scan the harbor as a continuous and automatic oil spill detection
system. While the system was not designed as an in situ measurinq
device, the measurement technique that it utilizes could prove to
be useful for in situ water quality measurements. The sensor de-
sign also demonstrates present capabilities for automated sensor
systems.
The lidar polarimeter oil spill detection system developed
by the Remote Sensing Center had to be capable of detecting oil
on the surface of water independently of the "background" water
turbidity. As a result/ it had to be capable of distinguishing
between various turbidity levels. It will be shown that the
measurement technique employed in this sensor has the capability
of continuous, real-time identification and measurement of oil on
water and selected water quality parameters including con-
centration of suspended solids.
Suspended solid particulates are supportive of pollution
transport in water due to the adsorptive mechanism. As a result
the concentration of suspended solids can be indicative of pol- '
lution potential resulting from precipitation run-off from non-
point sources such as timber cut areas, oil shale excavation
areas, rangeland, agricultural areas, as well as discharge from
industries and municipalities. This is demonstrated by studies
(3) that have linked the transport of electrically charged
chemicals to suspended clay particles. In addition, sediment and
corresponding loads of suspended material in an aquatic system
142
-------
reflect hydrochemical and ecological conditions and their change
over long periods of time (4). For all of these reasons sediment
and other suspended solids are important parameters in water
quality measurements.
Measurement Concept
The measurement technique is to illuminate a water surface
with linearly polarized coherent light and measure the light
scattered back (backscattered) toward the receiver in both the
polarization of transmission (like polarized) and in the
polarization orthogonal to the transmit polarization (cross
polarized). The interpretation of the backscattered energy is a
direct result of theoretical work on electromagnetic scatter by
Rouse (5). The development by Rouse indicated that for linearly
polarized radiation incident on a surface, the cross polarization
back scatter is due to scatter within a volume below the surface.
Since the volume scatter that produces the cross polarized back-
scatter is a multiple scattering process, it is dependent on the
concentration of scatterers within the volume.
Recent active optical studies have described the dependence
of electromagnetic backscatter upon the subsurface volumetric
scattering properties of an aquatic target (6). This work es-
tablished that the cross polarized backscatter component was al-
most exclusively dependent upon the number density (turbidity)
and size of suspended particles within the medium. These
findings led to the development of a laser system transmitting
vertically polarized coherent light at 632 nm (red) and receiving
both the vertical and horizontal polarized backscatter which
provided an accurate measure of suspended solids in natural
waterways.
Since most oils act as a lossy dielectric at visible
wavelengths, their presence on turbid water can be readily de-
tected as an equal attenuation of both polarization components of
the laser backscattered signal. Consequently, reliable detection
of oil spills is possible with a single wavelength dual-polarized
laser sensor. In addition, the attenuation characteristics of
oil vary with wavelength and oil types, therefore the potential
exists to actually identify the oil type by employing a multiple
wavelength laser system. Further, once the type is known, the
thickness of the oil can be inferred from the remote measurement.
As a result of the analysis summarized above, the Remote
Sensing Center undertook the development of a prototype dual
wavelength system (6) and the single wavelength sensor described
herein. The present environmentally hardened sensor employs a
442 nm (blue) laser as the transmitter. The received signal is
processed digitally and analyzed in real-time in a small special
purpose computer to provide a direct readout of turbidity, pres-
ence of oil, the mean and variance of the data and other system
flags.
143
-------
Measurement Result.8
In order to demonstrate and quantify the utility of a lidar
polarimeter to measure water quality parameters and to detect
oil, a series of experiments was devised. Controlled laboratory
measurements were performed first. These were successful in
verifying theoretical calculations and demonstrating the effects
of surface roughness, field of view, scatter concentration and
adsorbing material. Laboratory measurements were also performed
to demonstrate that the presence of oil on turbid water does not
affect the depolarization ratio, and that oil type and thickness
could be inferred from a dual wavelength system (6).
Measurements were then made under actual field conditions.
Measurements made of the Brazos River and the Houston ship chan-
nel verified the correlation between depolarization ratio and
water quality parameters under actual water conditions (Figure
1-3). These measurements support the use of a lidar polarimeter
for remote measurements of turbidity, suspended solids, and
transmittance.
The System
The optical head of the 442 nm (blue) lidar polarimeter
built for the Coast Guard is composed of three major components;
the transmitter, receiver and scanning system. The transmitter
is 12 mw Helium Cadmium vertically polarized laser (Liconix Model
401M). The laser is internally modulated with a bandwidth of
modulation greater than 100 kHz at 100% depth of modulation. The
output of the laser is expanded with a 40x beam collimator to re-
duce the beam divergence at long ranges and to maintain a safe
power density of 2.5 mW/cm^. The laser transmitter is housed
in a 6" diameter PVC tube which is supplied with an externally
controlled atmosphere for proper laser cooling.
The backscattered energy from the target is collected with
300-mm f/5.5 telephoto lens. The entire receiver section with
associated electronics is housed in a 4" diameter PVC tube
located directly above the transmitter. The telephoto lens is
mated to an extension tube which houses the spatial and spectral
filters along with the recollimating lens. After collimation,
the energy is directed through a calcite beam-splitting polariz-
ing tube. The separated vertically and horizontally polarized
components are detected utilizing RCA 8645 photomultiplier tubes.
The entire transceiver is mounted on a scanning system which
is under the control of a special purpose processor. The scan
system is capable of providing 350 arc scan at a rate of
144
-------
1° per second. The entire system, including the scan system, is
constructed of PBC and Lexan capable of withstanding a very harsh
environment.
The analog signal processor is composed of four subas-
semblies, which as a unit are housed aft of the optical system in
the receiver column. The system is essentially a dual channel,
frequency locked synchronous demodulator, the output of which,
after output power normalization, is converted to serial digital
data for transmission to the data processor.
The data processor is fabricated around a basic Intel 8080A
system with 4K of EPROM and IK of scratch pad. Within the
processor the mean and variance of the data are calculated and a
data integration period is automatically selected based on a
selected signal-to-noise ratio and system sensitivity. An oper-
ating region is established for the selected site and the data
are subjected to various filters and tests loops to determine the
variance of media turbidity and the presence of oil. Azimuth
scan of the instrument is a necessary functional subprogram in
the software. Scan speed, direction and current position are
monitored and controlled by the CPU. Besides operational
functions, the processor maintains system housekeeping for self-
test, thermal conditions and status flags.
Outputs from the data processor include an RS-232C standard
communications frame which contains .all system flags and data
along with a buffered alarm signal which is compatible with ex-
isting Coast Guard early warning systems.
The operator's console houses the data processor, power sup-
plies for the laser and photomultiplier tubes, the system cooling
unit and the operator interface for the selection of functional
parameters. The case itself is an environmentally sealed unit
which, when closed, maintains a constant atmosphere for the sys-
tem.
Conclusion
The development of an optical technique of measuring the
concentration of suspended solids was described. The unique
features of the technique are such that it has the potential of
making a quantitative measure of the concentration of suspended
solids without many of the difficulties associated with other op-
tical techniques, as well as detecting and qualifying oil on
water. The specific sensor discussed herein was designed as a
remote sensor and will measure water quality parameters only in
the surface layers of water. However, the measurement concept
has tremendous potential and could be incorporated in an in situ
sensor. Such a sensor could be used to profile water quality
parameters in real-time by lowering it through the water medium.
145
-------
References
1. Blanchard, B. J., and R. W. Learner, "Spectral Reflectance of
Water Containing Suspended Sediment," Remote Sensing and Water
Resources Management, American Water Resources Association, June
1973, pp. 339-347.
2. Clapp, J. L., "Application of Remote Sensing to Water Re-
sources Problems," Fourth Annual Earth Resources Program Review,
Vol. II, Houston, Texas, January 1972, pp. 46-1 - 46-40.
3. Richardson, E. M., and E. Epstein, "Retention of Three
Insecticides on Different Size Soil Particles Suspended in
Water," Soil Science Society of American Proceedings, Vol. 35,
1971, pp. 884-887.
4. Hahn, H. H., and R. Klute, "Potential Effects of Suspended,
Sedimented and Eroded Particulate Material in the Aqueous En-
vironment," International Conference on Environmental Sensing and
Assessment, Vol. I, Las Vegas, Nevada, September 14-19, 1975.
5. Rouse, J. W., "The Effect of Subsurface on the De-
polarization Ratio of Rough Surface Backscatter," Radio Science.
Vol. 7, No. 1, October 1972. '
6. Sheives, T. C., "A Study of a Dual Polarization Laser Back-
scatter System for Remote Identification and Measurement of Water
Pollution," Technical Report RSC-53, Remote Sensing Center, Texas
A&M University, College Station, Texas, May 1974.
146
-------
.5
n) .4
OS
c
o
-------
.5 .
*>.
00
o
•H
4-1
CO
OS
o
•H
t-i
0)
S-.2
.1
4- •
100
200
300 400
Turbidity (FTU)
500
600
700
Figure 2.
Graph of lidar depolarization ratios as a function of turbidity observed
along Brazos River from a boat
-------
.6-
.5..
.4
o
•rl
4-1
.3
•H
k-i
CO
H
O
a,
a>
o
•2,.
.1
20
-I
40 60
Transmittance (%)
100
Figure 3. Graph of lidar depolarization ratios as a function of
transmittance observed along Brazos River from a boat
149
-------
NEW WATER QUALITY SENSOR TECHNOLOGY EMPLOYING AUTOMATED WET
CHEMICALS
by Khalil H. Mancy, Ph.D.
University of Michigan
Ann Arbor, MI 48109
Among the various water resources research activities at the
University of Michigan, there exists an active research program
devoted to the development and application of water quality
sensor systems. During the last 10 years, this research program
has been primarily concerned with the development of new sensor
systems and the application of new measurement techniques. This
included water quality monitoring and surveillance in lakes (es-
pecially the Great Lakes), rivers, and municipal water supplies
(1).
This presentation provides an overview of recent research
findings, future needs and possible new applications of auto-
mated in situ water quality sensors for field application.
In Situ And Qn-Line Measurement
Water quality surveillance and monitoring programs rely on
both field and laboratory measurements. Field measurements are
primarily based on either non-contact remote sensing or contact
in situ and on-line measurement techniques. Schematic il-
lustrations of in situ and on-line measurement techniques are
shown in Figure 1. In this arbitrary classification, the dis-
tinction is being made between in situ measurement where the
sensor is directly placed in the environment to be measured, and
all other field systems offer the unique advantage of the
elimination of the sampling step and one needs not to be con-
cerned with the problem of collecting representative samples.
Responses from in situ sensors are proportional to certain
physical or chemical characteristics of the aqueous phase, based
on established relationships. Typical examples include thermo-
electric, photoelectric and electrochemical transducers. Examples
of in situ electrochemical sensors are listed in Table 1.
In spite of the distinct advantages offered by in situ
measurement, it is frequently more feasible to utilize on-line
techniques, which include collecting water samples and bringing
them in contacet with the sensor system. This can be done with
or without sample pretreatment. The main limitation with in situ
measurement is that the sensor response is generally influenced
by environmental factors such as temperature, hydrostatic pres-
sure, hydrodynamic characteristics, light intensity, chemical
interferences, or nonspecific interferences, e.g., accumulation
of oil or grease, silt, or biological growth on the surface of
the sensor. The significance of these environmental factors
150
-------
TABLE 1. ELECTROCHEMICAL SENSORS FOR WATER QUALITY MONITORING
Sensor Type Equation
n
Conductome trie L = Kc 2 Ci A^ z^ (!)
i
RT
m
,
Potentiometric E = k + zF In Z ^
(2)
-Glass electrode pH = -log ajj-f (3)
(e.g., pH)
-Membrane electrodes
(e.g., ion-selective
electrodes)
- cationic pM+ = -log a^j+ (4)
- anionic pA~ = -log a^- (5)
where:
L = specific conductance
Kc = cell constant
Ci = ionic concentration
\i = ionic equivalent conductance
zi = ionic valency
Em - measured electrode potential
F = Faraday constant
KJ = selectivity coefficient
i(3 = diffusion current
A = electrode surface area
Pm = membrane permeability coefficient
b = membrane thickness
depends on the type of the sensor system and the particular ap-
plication. Automatic compensation for some of these effects,
e.g., changes in temperature or hydrostatic pressure, is some-
times possible within certain ranges.
In view of these limitations, it is frequently necessary to
conduct on-line monitoring. In certain applications, measurement
can be done without sample pretreatment. In other cases, sample
preconditioning is necessary (A) to minimize the effect of
interferences or (b) for the appropriate use of the sensor sys-
tem. This may include temperature control, filtration, dis-
solution, dilution, digestion, or reagent addition. Furthermore,
151
-------
sample preconditioning may be required for the appropriate
functioning of the sensor system regardless of the effect of en-
vironmental factors. For example, water samples should be
adjusted to approximately pH 5.0 prior to the application of the
lanthanum fluoride selective ion electrode. This is required to
minimize inteferences by hydroxyl ions at high pH or the forma-
tion of hydrogen fluoride at low pH values.
In situ and on-line water quality monitoring is frequently
conducted by automated systems in mobile laboratories (2), on
board ships (3), or by means of remote robot systems situated on
shore, or by buoys (4). These are used for water quality mon-
itoring of rivers, lakes, estuaries and marine environments (5).
Furthermore, they are frequently applied for monitoring of
industrial and municipal effluents and the control of water and
wastewater treatment processes (6).
Examples of in situ and on-line water quality sensor systems
are given in Table 2. It should be noted that all in situ
sensors can be used for on-line measurements, with or without
sample pretreatment.
Qn-Line Measurement With Sample Pretreatment
In view of the previous discussion it is evident that rel-
atively few water quality parameters are presently amenable to
in situ measurement or even on-line measurement without sample
pretreatment. Such measurements are in general limited to phys-
ical parameters and to those chemical constituents for which
electrochemical sensors are available. Even in these cases, how-
ever, sample pretreatment may be required. For example, since
selective ion electrodes measure the activity of a single species
(e.g., only Sand not HS or H2S) it is common practice to
account for both the pH and the ionic strength when making
specific ion electrode measurements so that the results are
directly interpretable in terms of the total analytical con-
centration. Adjustment of pH is required not only to convert all
analyte to the requisite form but also to control the degree of
electrode response to H+ or OH since these ions can, at higher
levels, be sensed respectively by cation-and anion-specific ion
electrodes.
On-line measurements with sample pretreatment provide ex-
tensive possibilities for automated analysis. Examples of such
measurements systems are given in Figure 2.
The most common on-line measurement with sample pretreatment
is the automated colorimetric analyzer and, in particular, the
Technicon AutoAnalyzer system. It should, however, be evident to
the reader that the techniques that will be discussed are ap-
plicable to many sensors in addition to selective ion electrode
and colorinietric analysis. Automated systems have been used in
152
-------
atomic absorption spectrophotometry to present samples to the
instrument or to add reagents (such as lanthanum for suppression
of phosphate interference), or to perform liquid-liquid extrac-
tions prior to measurement (13). Examples of the types of para-
meters which can be measured by the Technicon AutoAnalyzer are
given in Table 2. These are primarily based on automated col-
orimetric procedures (10).
TABLE 2. TYPICAL APPLICATIONS OF SENSOR SYSTEMS FOR IN-SITU
AND ON-LINE AUTOMATED WATER QUALITY MEASUREMENT
(PARTIAL LISTING)
In Situ;
a) Separation and/or concentration:
- Controlled potential electrodeposition
- Gas permeable membranes
b) Sensor Systems:
- Temperature transducers (Thermistors,
Thermocouples, etc)
- Pressure (hydrostatic)
- Current speed and direction
Light intensity
Electrical conductance
- pH-glass electrodes
Redox potential-inert metal electrodes
- Selective ion electrodes (7)
Solid state (Free ions)
Liquid ion exchange (Free ions)
Enzyme electrodes (organic substrates)
Gas permeable electrodes (e.g. CO2* NH3)
- Voltametrie membrane electrodes (8)
(e. g. 02, HOC1, 03)
On-Line Without Sample Pretreatment:
a) Turbidity
Light scatter
Light absorption
b) UV absorption (Certain dissolved organic compounds,
e.g., phenols). Advanced techniques include com-
pensation for turbidity effects (1).
c) Fish toxicity measurement. Advanced techniques
include measurements of both survival and viability (9)
d) All of the above listed in situ sensors can be used
for on-line measurements.
153
-------
TABLE 2. (Continued)
^ On-Line Measurement With Sample Pretreatment;
a) AutoAnalyzers, e.g., Technicon AutoAnalyzer (10).
Parameter
Ammonia
Nitrate+Nitrite
Ortho-Phosphate
Total Inorganic
Phosphate
Silicate
Total Soluble Iron
Hexavalent Chromium
Copper
Sucrose
COD
Total Phosphorus
Detergents
Range(mg/1)
0-10
0-2
0-1
0-10
0-10
0-10
0-5
0-10
0-100
0-500
0-50
Parameter
Cyanide
Phenol
Fluoride
Chloride
Hardness
Sulfate
Alkalinity
(pH 3 min)
Alkalinity
(pH 8 min)
Total Kjeldahl
Nitrogen
Range(mg/1)
0-3
0-5
0-2
0-10
0-300
0-300
0-500
0-100
0-100
b) Specialized automated systems;
Metal analysis by atomic absorption spectrophotometry
Trace metal analysis by differential pulse
polarography or differential pulse anodic stripping
voltametry (2)
Free and total ions using commercially available
selective ion electrodes (e.g. Ca++, Mg4"1",
Cd++, Cu++, Pb++, Hg++, S—, F-, C1-,
Br~, i~, N03~, etc.) (11)
Dissolved gases or neutral species using voltammetric
membrane electrodes (Residual Chlorine-free and
total, S02, Ozone) (12)
Total Oxygen Demand (TOD)
Total Organic Carbon (TOC)
Calcium Carbonate Deposition Test (CCDT) (2)
Automated Titrators
A large number of continuous titrators are commercially
available. These can be advantageously utilized to both reduce
laboratory costs and to improve analysis precision. These de-
vices utilize motor driven, constant delivery speed syringes for
titrant addition. Although end points can be sensed, and the
titrant addition rate can alter the instrument's derivative cir-
cuitry, the analyst should note that this is usually not the
154
-------
limiting factor; the one to which the most concern should be
given is the response time of the electrode sensor system. Slow
response can lead to over-shooting the end point and thus to an
analysis bias„
Automated analysis systems for the monitoring of low con-
centrations of materials require frequent recalibration. This
poses serious limitations to their use in field applications
where only weekly maintenance visits to the monitoring station
are made. Some presently available equipment utilizes solenoid-
activated valves to switch the sample intake from the stream
being monitored. Based on the sensor output, the instrument can
then be automatically recalibrated from their concentration to
that of the sample may not be valid. Dilute solutions often lose
strength by adsorption of the material onto the container walls?
this can perhaps be prevented by use of inert standard storage
vessels such as quartz or Teflon,, <
Field Applications
Several examples of automated sensor systems are given in
Table 2. From this limit it is obvious that there exists a con-
siderable number of possibilities for automated water quality
monitoring systems. Depending on the particular use, a given
system may not be suitable for field applications. The main
limiting factors are usually:
0 Ease of operation and maintenance.
0 Portability.
° Ruggedness and ability to withstand field condi-
tions with minimum attendance.
0 Frequency of calibration and servicing.
° Cost—both capital and running cost.
In a recent study, a number of automated systems were used
for monitoring the quality of water supplies in a number of
municipalities in the USA (2). This included water quality mon-
itoring at the source, the treatment plant and throughout the
distribution system. The purpose of this investigation was to
assess water quality deterioration in the distribution system and
assess the feasibility of a prototype mobile monitor (a specially
equipped van).
A listing of the sensor system used in the prototype monitor
is given in Table 3. A summary of the performance characterist-
ics of these sensor systems is shown in Table 4. Examples of
automated wet chemical measurements using electrochemical sensors
include the use of the lanthanum fluoride selective ion electrode
as shown in Figure 3(2), and differential anodic stripping
voltametry as shown in Figure 4(2).
155
-------
TABLE 3. SENSORS -IN THE PROTOTYPE MONITOR
Parameter
Temperature
Conductivity
PH
Chloride
Dissolved
Oxygen
Free Residual
Chlorine
Total Residual
Chlorine
Turbidity
Corrosion Rate
Free Fluoride
Total Fluoride
Alkalinity
Hardness
Nitrate
Cadmium
Lead
Copper
Calcium
Carbonate
Deposition Test
Sensor Type
Thermistor
A-C Conductivity Cell
Glass Electrode
Solid State Ion Selective
Electrode
Voltammetric Electrode
Galvanic Cell
Galvanic Cell
Nephelometer
Polarization Admittance
Technique
Solid State Ion Selective
Electrode
Solid State Ion Selective
Electrode
Potentionmetric Combination
pH Electrode
Liquid Junction Ion
Selective Electrode
Liquid Junction Ion
Selective Electrode
Differential Anodic Stripping
Voltammetry (DASV)
DASV
DASV
Potentiostatic Rotating
Ring Disc Electrode
Unmodified
Commercially
Available
System
/
/
/
/
/
/
/
/
/
Sample
Pre-
/
/
/
/
V
/
/
/
/
Requires
Special
Timing
/
/
/
J
/
01
-------
TABLE 4. PERFORMANCE CHARACTERISTICS OF MOBILE LABORATORY SENSOR SYSTEMS
en
L Pj-jniftf'
Kjrjnsjs
Nitfite
Chloride
Temperature
I
Conductivity
PH
Dissolved
Oxyjen
Fr*e Residual
Chlarine
Total R-1-sidual
Cnionne
Turbidity
Detection
L'mit
1 0 rr.;/!
eq'jiv.
CaCOj
1 ni}/l
3.0 mj/l
N.A.
N.A.
N.A.
0.1 mg/l
0.01 mg/l
0.01 mg/l
0.01JTU
T
Ran^els)
One Dccacie.
e.g.
10—100 mg/l
equiv. CaC03
Two Decides.
= •9-
1-500 mj/l
0-240 mj/l
0-1200
o-z«o
micro r.'HOS/
pH2-12
0-12
0-2-5
mg/l
0-1 .0
0-2.0
m^/l
o-o.i
0 — 1 .0
0-3.0
0-30
JTU
Selectivity Coefficients
or
|nKr!crcncas
Zn-*-3.5 N;a2»=l.o
Fe2* = 3.5 B32' = 0.04
Cu2+=3.1 5r2 +=OiS
r:;2*=i.35 iNa*=o.is
Ca2*=1.0 K* = 0.10
HCO^O.C2
CQ2-O.COG
Cl- =0.006
OH'=C.O!3
None
NJ*
None
None
A:r bubbles inwrnplc
s',rc;m must be re-
moved by a bubblB
trjp
Prec'u!oi>
•*-5 frig/*
OI20-!50mg/l
equiv. CaC03
N.D.'
±2 mg/l
* 0 2°F
—5 micro MHO/cm
i0.02 units
± O.I mg/l
Ifl.Ol mg/l
-10.02 JTU
Temperature
Oepencicnci
Consifiercd
in calculation
of hardness
from divalent
cation activity
measurements
Thermistor
Compensated
Thermistor
Compensated
N.A.
Thermistor
Compensated
Thermistor
Corr.pen:atcd
Tliermistor
Compensated
Thermistor
Compensated
N.A.
Applicable
pH Rjnge
5.5 to 12
2-12
2-12
N.A.
5-12
2-12
2-12
Sample pH
adjusted to
pH 4.5 by
reagent
addition
N.A.
loiliC
Str^n.jth
E.'fcct
Considered
in calculation
of hardness
from divalent
cation activity
measurements
N.S.2
N.S.
NA.3
This parjmcter
Measures
Chanyes in ionic
strt'nijth
INS.
N5.
Ionic strength
adjusted to
constant value
by reagent
addition
HA.
Long Term Stability,
nocaliliraticn a.id
Mainl.'.iivinc" R'TJ jireir.?n'.i
T.'iese Ii;;'ii'l junction ion st-
Icctive Ci'.'Ctrod"*, liavc been
shown to h.ive stable poten-
tial; which permit continuous
uso for pcrodJ up to seven
d.iys v/ithont recaSiU'jtion;
the clvc'.rcj^i' tip. './.-. cc d;i!y
attention ii n-ciu^'C'! to .n'.ufe
thai bu'JoU; hwvc no! !c.r.-n<.d
on the tips. C on'plc'.t rejuven-
ation ol the probes must be
performed evf-ry 30dayi.
7 days: No known main-
finance
'
7 days: No kiiown snain-
ta-.iwnce
7 days: Internal flow chan-
nel ;nd rlectrodc suHaccs
must be cleaned monthly
7 days; No known piain-
tainjnco
7 dnys; Electrode must be
rejuvenMed every 30 days
OJily c!'.e-ckino ol calibra-
tion-. |jy tompi'isnn with
jrr.|:ercmi-lr:C litriliu.'t >s
must be rt:!.:ud every 7 days
7 days; Periodic clcamn? of
photocell windows maintains
icns.tivity. LiQht loulmg of
tlic windows doe» not «ifcct
accuracy within the 7 dsy
cjlibvation period
-------
TABLE 4. (Continued)
Parameter
Frca Fluoride
Total FluorUs
Alkalinity
Cadmium
Lesd
Copper
! C~OT
Corrojlon
Detection
Limit
0.1 mg/l
0.1 mg/l
! Omg/t
0.001 mg/l
O.COImg/1
0.001ms/!
N.A.
NA.
Rangets)
0.1— 2.0 mg/1
60mg/l spin
e.g.
20-50 ma/I
equiv. Ca£o3
0-0. 1 mg/l
0—0.1 mg/l
0-0.1mg/l
NA.
10 calibrated
ranges from
C-O.I MPY to
0-5000 MPY
(Mils per yrj
Selectivity Coefficients
or
Interferences
OH-=0.7@[F-J =
10-G tolO-'M
OH'=0.07€>IF'] =
IO"4 to 10'3M
OH-=O.OSO|F-] =
10-3 to!0'2M
Residual chlorine inter.
feres, but is eliminated
by addition of Ihiosul-
fate to the sample stream
Trace organic; can in-
terfere by giving false
peaks. Accurst; iden-
tification of peak po-
tential indicates the
urcsencf! of interferences
Unknown
Nona
Precision
±0.02mg/l
±3mg£
equiv. C»CC>3
N.D.
N.A.
*• 5% of
range
Temperature
Dependence
+0.17% increase
in apparent sig-
nal for IOC in-
crease in temp.
(Theoretical)
Unknown
N.D.
Unknown
Temperature
variation re-
sults in chjngc
in corrosion
rate
Applicable
pH Ran;c
Sample pK
".dj listed to
pH 5.0-
5.4
NA.
Sample pH
adjusted to
pH 3
Variation in
pH results
in v.inclion
in fJtc of
lilm depo-
sition
Vatiation in
pH rcs-jlis in
variation in
rate of cor-
rosion
Ionic
Strength
Effect
Ionic Strength
adjusted to
constant value
by reagent
addition
Ic-nic Strenglh
idjui'.cd to
constant value
by rc;ocnt
addition
Ionic strength
adjusted to
constant value
by rcj^cnt
addition
KJ\.
NA.
Long Term Stjljllily.
Scc.Tlibrat :cn 2nd
M.'intain.inc!" r\rrj'ji''?mc''ts
Daily introductljn of stan-
dards riaintcins ;:C'jr;cy :r.d
dfrnonstr.ilcs i;'!r^n'.y cf the
flow system. E!rc;'C£»S ar»
knov/n to be 5'ci:1'' ov^- very
Ion; pcrio-Js o' '.ir:c. Prcj'.-fit
rcscr>o:r: rrvji! be /*Jillcd
evi-'y l^-' (Jays
Rcjrjer.t rcscrviors mjst b;
refilled every lour dsy;. Rc-
C3l;br.i'bion mull b^ p-j.'fcrm^J
for each new stt of reagents
Standards r.re ana!yJ':"1 ao'.o-
matic.TtJy every *;.<< ho;;rs.
OASY electrode; must be
rcplated doily
£j::h ana!/;;; ••. di'.C'-^lc.
hl'Clrc^.1 .:v_-,; f" fj:-»cn«;cd
L^l'.vr^c rich run hy il:>,>nin5
'.vith ^i'ult rt'l;ic DCiti
No call'jrjtlon or MJin'.iir.ancc
rcquir c-'T»tr.t'. . Internal electron-
ic s'.-Tr.daicI dci-.:onvt.'3'.os In-
strumcnt integrity.
Ul
00
N-D." not determined; I.e., a specific relationship between the referenced characteristic and the panmetcr of interest wa: not determined for the mobile laboratory sensor system.
2N.S. » not jljijflcant;!.*., variations in ionic strength cf t^p v/cters in the US. would not produce more than ten percent variation in the apparent :ctivity of tht pirjrnc'.er of
i. * not spplicjbl:; l.«, tilt referenced characteristic is not rented to tho parameter of Interert.
-------
The operation of automated monitoring systems is best con-
trolled by means of a minicomputer. This is particularly useful
in cases where several systems are running simultaneously. An ex-
ample of such an application was reported by the above mentioned
study (2). All sensor systems in the mobile laboratory are con-
trolled by an on-board minicomputer system which includes a Texas
Instruments digital computer, Model 960-A; a Texas Instruments
Silent 700/300 teleprinter; a combination high speed paper tape
reader perforator; and a Computer Products wide range analog-to-
digital (A/D) converter, Model RTP7480). The computer contains
16,384 words of semiconductor memory, expandable internally to
32K, and, with the addition of an external chassis, to 64K. An
internal timer A16 input/output digital switching board wired for
interrupt, and an internal communications register unit (CRU) ex-
pansion chassis, required for 5 to 20 peripheral connections, are
options used with the computer. Seven CRU connections are used
in the mobile laboratory.
Within the data sampling cycle, appropriate digital outputs
are sent to each measurement system (a total of 18). Independent
programs, operating from the internal timer in the computer, send
digital output signals to each of the measurement systems.
In the data sampling cycle, the internal interval timer con-
trols the timing for reading the analog signals from each of the
different sensor instrumentation packages. The data acquisition
program in the computer determines the gain to be used in the
analog-to-digital converter (ADC) in reading each analog input
channel. Within the computer, all program operations are under
the control of the Texas Instruments supervisor program, PAM
(Process Automation Monitor).
Recommendations
The recommendations given below are not restricted to auto-
mated wet chemical applications.
1. It is recommended that a unified terminology
be adopted for sensor performance
characteristics. This can be separated into
(a) primary performance characteristics
including sensitivity, selectivity,
detection limits, response time, and long
time stability, and (b) secondary
performance characteristics including
environmental effects such as temperature,
hydrostatic pressure, hydrodynamics, light
intensity, ionic strength, etc.
159
-------
2. Future research and development should emphasize
both the improvement of existing sensor systems
and their adaptation for field applications, and
the development of new sensor systems.
3. Research and Development of water quality sensor
systems should include the characterization of
soluble and insoluble matter, both in the aqu-
eous phase and sediments. Most trace metals in
aquatic environments associate with the particu-
late matter and sediments, including aquatic
biota. The soluble fraction of trace metals may
be insignificant in certain cases.
4. It is proposed that a better means of utilization
of easily measured water quality parameters be
examined. This could be done by first.es-
tablishing correlation coefficients between one
or more easily measured parameters and those
parameters which cannot be easily measured. It
is important to note that these correlation coef-
fients will vary from one application to another
and could also be time-dependent.
5. Similar to item 4, an easily measured parameter
could be used to trigger an automated sampling
device and an alarm system. The sample will be
then collected for detailed laboratory
analyses.
6. It is proposed that the application of in situ
separation and concentration techniques be
investigated. This could be done for sampling
of trace metals or organic matter using elec-
trodeposition or especially designed
membranes. Furthermore, in situ separation and
concentration could be used in conjunction
with voltammetric and selective ion electrodes
to provide more effective in situ sensor sys-
tems.
7. It is recommended that the application of
advanced polarography for trace organic
analysis be encouraged. Unfortunately,
little emphasis is given to this ap-
plication in recent literature.
160
-------
References
1. Mancy, K.H., and Allen, H.E. "Automated Water
Quality Monitoring and Analysis." Chapter in
Water Quality Analysis Manual, Regional Office >
For Europe, the World Health Organization
(1977).
2. McClelland, N.E., and Mancy, K.H. "Water Quality
Monitoring In Distribution Systems,"
EPA-600/2-77/074 (1977).
3. Kramer, J.R., Allen, H.E., Baulen, G.W. and Burnes,
N.M., "Lake Erie Time Study (LETS)," Canada Centre
for Inland Waters Paper No. 4., Burlington, Ontario,
1970.
4. "Proc. of Nat'l Symposium on Data and Instrumentation
for Water Quality Management." Univ. of Wis., 1970.
5. Riley and Skirrow. "Chemical Oceanography." Academic
Press, New York, NY (1965).
6. Briggs, R. "Proceedings of International. Workshop of
Instrumentation, Control and Automation for Wastewater
Treatment Systems," IWRP, London (1973).
7. Durst, R.A. (Editor). "Ion Selective Electrodes." National
Bureau of Standards Special Publ. No. 314. U.S. Govt.
Printing Office, Washington, D.C. (1965).
8. Mancy, K.H., and Jaffe, T. "Analysis of Dissolved Oxygen
in Natural and Waste Waters." Public Health Service
Publication No. 999-WP-37 (1960).
9. Grains, J. Jr., Lanza, C.R., Sparks, R.E. and Waller, W.T.,
"Developing Biological Information Systems for Water
Quality Management." Water Resources Bulletin, American
Water Resources Association, 9^, 81 (1973).
10. Technicon Instruments Corporation, Tarrytown, N.Y.
11. Orion Research Incorporated, Newsletter/Specific Ion
Electrode Technology, Cambridge, MA. 2_, 5 (1970).
161
-------
12. Mancy, K.H. "In Situ Measurement of Ozone by a Voltametric
Membrane Electrode System." Unpublished report, The
Environmental Chemistry Laboratory, SPH-I, The University
of Michigan, Ann Arbor, Michigan (1977).
13. Goulden, P.D., Brooksbank, P. and Ryan, J.F. Am. Lab. 5 (Q)
10 (1973). '
162
-------
A. IrwSItu Measurement
Aquatic - _ -
Environment
.
•^f^^^
sensor
Systen
^
Signal 1
Condi tioning^
Readout j
M
o\
U)
D. On-Hnc Measurement without Sample Pretreatrent
Aquatic
Envi ronment
t
Intake
(°
°)
Sensor j
iystem i
iignai
Conditioning
Readout i
Continuous
Sample
Discrete
Samples
C. On-line Measurement with Sample Pretreatment
Aquatic •
Environment
s^-^-v*
*^-< • -^^.
Intake
Sample
'rctreatnent
Reaction
Developnent
Sensor
Systen
Signal
Conditioning
Readout
continuous
Sample
Discrete
Samples
Figure 1. Automated measurement concepts
-------
Sample stream
or
Discrete samples
SAMPLE
PREPARATION
Dialysis
Di2estion
Dilution
Dissolution
Distillation
Extraction
Filtration
REACTION
DEVELOPMENT
Heating
Mixing
Reagent addition
INSTRUMENTAL
TRANSDUCERS
Colorimeter
Fluorimeter
Flame spectrophotometer
Atomic absorption spectrophotometer
Electroanalytical systems
READOUT
Meter
Recorder
Punch tape
COMPUTER
FEEDBACK
ACTION
Figure 2. Basic operations of continuous flow analysis
-------
tn
Automatic Sompier
13/hr. II
F-FAB
Air
Sample
Air
T-FAB
Proportioning
Pump
Dual Pen Recorder
1
I
(ml/m)
'
-2O-
- ? O-
f\ A*)
Ofl-
~]
> 1
> 1
V
1
X 1
•v
3
t
t_
^/
5
..-•-• i
Mixing Coil 1
^TyYlW^ v
LMAAJ '
/^yyy^y^yyyt>yyyO
-------
DIFFERENTIAL
ANODIC STRIPPING
VOLTAMMETRY
STANDARD
I ,
i
I"
I
I
•
h-
STRIP-CHART
RECORDER
i
ELECTRONICS
MODULE
SAMPLE /
STANDARD
RELAY
FLOW /
NO FLOW
RELAY
I
SENSORS'
FLOW
Cl'L
WASTE
TAP
WATER
SAMPLE
r
SAMPLE/
STANDARD
VALVE
SAMPLE
FLOW/
•YPASS
VALVE
N2 FLOW
BYPASS
VALVE
n
MIXING
COIL
-AIR
PROPORTIONING
PUMP
WASTE
ACID
N, SUPPLY
figure 4. Schematic illustration of trace metals monitor in the mobile laboratory.
-------
A STATE VIEW OF THE NEED FOR RESEARCH IN IN SITU AUTOMATED WATER
QUALITY SENSORS
by Charles A. Whitehurst
Louisiana State University
Baton Rouge, LA 70806
In looking for thoughts on which I could base my comments
about the needs in sensor research and development, I came across
a Request for Proposal from the USGS. It was issued on November
29, 1977, and it asked for the development of a prototype system
to measure water stages and water surface velocity in natural
streams on a continuous or time series basis using microwave
techniques. Responses to this request were to be received in
Washington no later than January 4, 1978, which makes it a very
recent effort in sensor development. The objectives of the
effort as stated in the RFP were to define, manufacture, test,
and deliver a prototype system which would meet the following ob-
jectives:
•Operate unattended for a minimum of 30 days.
•Be capable of continuous or intermittent measurement
of water stage and water surface velocity at
programmable time intervals.
•Have no moving parts and no parts in contact with
the water.
•Operate under a wide range of environmental conditions
such as:
1. Expected temperature range -30 to 130° F.
2. Expected humidity range 0 to 95 percent.
• Be capable of measuring water stage with a resolution
of +0.04 feet over a range in stage of 20 feet.
However, it would be most desirable to achieve a
resolution of +0.01 feet over a range of 50 feet.
In this respect, be capable of mounting permanently
at least 20 feet above minimum water surface.
• Be capable of measuring water surface velocity with an
accuracy of +4 percent for velocities ranging from 1
to 25 feet per second.
• Be as compact as practical consistent with a sound
design.
• Be battery operated with minimal battery drain
consistent with reliable and unattended operation.
167
-------
I think the request emphasizes the need for automated in
situ sensors of a very basic nature. Although this RFP may not
be completely relevant to the objectives of this workshop (we are
water quality oriented), it does put emphasis on the continuing
need for this type of research and development. The design and
evaluation of in situ sensors for environmental monitoring must
be a priority area for agencies involved in the protection of our
natural resoruces. My comments are, in a large part, upon
experiences in south Louisiana, where stresses on the coastal
environment have been noted for several decades, and based upon
interaction with colleagues in the "scientific" community at LSU.
It is appropriate that I stress the importance of the Louisiana
environment and the activities which prompted a few of us at LSU
to initiate reports in environmental monitoring and sensor
development.
The coastal zone of Louisiana extends approximately to the
5-foot contour line at its northern boundary, west to the Sabine
River and eastward to the Chandeleur Islands. The area is es-
timated to be about 10.5 million acres or 16,400 mi2.
Marshlands comprise the bulk of the wetlands, some 4.2 mil-
lion acres or 6,563 mi2. These marsh acres support 10 mil-
lion user-days of recreation per year. They also hold some
10,220 camps, which serve as away-from-home bases for a range of
marsh activities, including fishing, hunting, and boating.
Further categorization of the marsh types shows that saline
marshes account for 932,000 acres or 1,456 mi2, freshwater
marshes stretch for 1.3 million acres or 2,031 mi2, while mixed
marshes, being the most extensive, comprise roughly 2 million
acres or 3,040 mi2. Swampland accounts for about 1.6 million
acres or 2,500 mi2.
Thirty-one percent of the coastal region, or 5,279 mi2, is
surface water. Predominant water bodies include Timbalier,
Vermillion, Caminada and Baratarta bays; and lakes Borgne,
Salvador and Pontchartrain. Eight river basins together with a
large number of manmade water features give coastal Louisiana a
total land-water interface of 30,190 miles, second only to
Alaska. The shoreline-gulf for the state is only 363 miles, the
smallest of the land-water interfaces. Over 11,800 miles of
navigable channels exist in the coastal zone. About 8,850 of
these miles are located south of the gulf intercoastal waterway.
Tides in Louisiana are diurnal ranging from one half foot to
a maximum of 2 feet. The average is slightly greater than one
foot.
A great part of the environment described above has been
created by government and private interests, namely the Corps of
Engineers and the major oil companies. Their dredging activities
168
-------
account for many miles of the navigable waters. These activities
continue and, in fact, are increasing due to the energy shortage.
Further, there are relatively new developments in the coastal
zone.
The Louisiana Offshore Oil Port, Inc. (LOOP), a consortium
of major U.S. oil companies, has planned the construction of an
oil terminal 18 miles off the coast of Louisiana to handle deep-
draft supertankers. The port, its connecting pipelines and re-
lated land-based facilities will provide a means for supplying
imported crude oil to refineries in southeastern Louisiana and in
areas of the midwest, and will have a capacity to initially!
handle up to 1,400,000 barrels of crude oil daily. The oil port
project, as proposed by LOOP, will consist of: 1) an offshore
terminal for unloading deep-draft crude oil tankers, 2) large
diameter buried pipelines from the offshore terminal to a booster
station at the shore, then to an onshore storage facility, 3) the
onshore storage facility itself, and 4) a pipeline from the on-
shore storage facility to the terminal of capline, one of the
world's largest pipelines, located on the Mississippi River at
St. James, Louisiana. Port construction is to begin in early
1978 with crude oil handling to commence by late 1979.
Another activity is the storage of oil in salt domes. There
is already storage in three areas with proposed domes in
Napoleonville and other areas suggested for use if pending
investigations are successful. Once oil is stored in the sub-
surface reservoirs, there will be the ever-present possibility of
leakage.
The Louisiana state legislature acknowledged the en-
vironmental problems that could be caused by a superport and its
associated onshore facilities when it established the offshore
terminal authority as the primary licensing and regulatory agency
for the proposed Louisiana superport. The legislation creating
the authority (LA. RS 34:3101-16) requires that protection of the
environmental monitoring program is mandated, the purpose of
which is to ascertain the existing ambient environmental con-
ditions and then detect and evaluate any of the potentially del-
eterious effects the superport may have on the coastal en-
vironment during port construction and subsequent operation.
This evaluation will form the basis for determining where, when
and how necessary remedial measures are to be taken.
The Environmental Monitoring Program required by the en-
abling legislation includes assessing:
The environmental stresses caused by oil
spills and other polluting instances;
169
-------
the air and water pollution levels in
the Louisiana coastal environment;
the offshore water column and offshore benthos;
shoreline, estuarine marsh, and freshwater swamp
vegetation and fauna;
the number, movement, and spawning and breeding
cycles for wildlife in the Louisiana coastal
environment;
saltwater intrusion;
water circulation and drainage; and
shoreline erosion and erosion along pipeline
canals.
The list of environmental information needed to support the
authority's monitoring program is both explicit and extensive and
suggests the means used for acquiring the basic information must
be:
accurate
long-term/reliable
cost effective
based on "real time" data
To meet the need for a comprehensive Environmental Mon-
itoring Program which will satisfy the above criteria, a well
planned, coordinated effort is required. The Louisiana De-
partment of Wildlife and Fisheries has been analyzing water qual-
ity in the coastal zone for many years. The typical measurement
technique is to go to the site by boat or helicopter, dip sensors
into the water and record readings or take water samples back to
the laboratory for analysis. While this is the only way that
some data can be obtained, the cost is high, sampling is limited
by weather and the data are not available on a real time basis.
There are no "new" monitoring techniques proposed by the F & WL
personnel.
A continuous in situ data acquisition system could provide a
significant improvement in the present and planned monitoring
program. Data from a real time system, through cooperation with
the Department of Wildlife and Fisheries, would be analyzed to
determine the relationships between changes in physical para-
meters such as temperature, salinity, dissolved oxygen, etc.,
and changes in plant and animal life. The in situ monitoring
station would be used to sample those physical parameters found
170
-------
to be the most critical as often as needed (for example, on an
hourly basis) and transmit the data to a receiving station.
Analysis of these data would be used to determine where and when
more detailed analysis of water quality was required. The in-
situ monitoring platform could also provide data on wind and cur-
rents which would be necessary to model regional dynamics. These
models would be useful in predicting the movement and dispersion
of pollutants, the flushing time of estuaries, etc.
The requirements that data acquisition in the environmental
monitoring program must be provided in "real time" and be "cost
effective" imply that measurements be made from unmanned in situ
monitoring platforms. The areas of Louisiana, and, in fact, most
coastal zones, are entirely too vast to continue operating in a
bucket brigade mode. Platforms must be designed for operation in
remote areas with power supplied from batteries and/or solar pan-
els. Although the list which follows is not all-inclusive, it
does represent those environmental parameters which would be de-
sirable or useful in the characterization of an estuarine area.
The parameters are grouped in order to facilitate this discus-
sion. Those in Group I comprise the basic parameters found in
many water quality analyzers such as those manufactured by
Hydrolab, Martek, Beckman, etc. Sensors for measurement of tem-
perature, salinity (calculated from conductivity), depth, and pH
are relatively reliable for short periods of time. This is not
the case for continuous measurements, especially for dissolved
oxygen (DO). As pointed out earlier in this meeting, a major
problem with this type sensor is fouling of the membrane. This
reduces the permeability of the membrane which results in an ap-
parent decrease in DO.
TABLE 1. DESIRABLE MEASUREMENT PARAMETERS IN ESTUARINE AREAS
GROUP I GROUP II GROUP III
Water Temperature Current Velocity Chemical Oxygen Demand
Salinity Wind Velocity Suspended Particle
Concentration
Dissolved Oxygen Evaporation Chlorophyll a_
pH Rainfall Hydrocarbon Content
Depth Acoustical Signature
Trace Metals
Chemical Elements
171
-------
Experience by personnel with the Corps of Engineers in New
Orleans indicates that sensor degradation takes place rapidly in
areas where the water is not fast moving and where the water con-
tains oil and/or fine silt. The estuarine areas of Louisiana
have precisely these characteristics. Therefore, development
work is necessary to provide better sensing techniques or to
adapt presently available sensors to make them suitable for con-
tinuous use in an in situ monitoring platform application.
In the summer of 1977 we initiated an experimental effort
designed to measure the error in recordings from in situ water
quality sensors. The sensors were located in a range of en-
vironmental conditions from fresh water to sea water.
Two sets of data were recorded simultaneously at all experi-
mental locations. The experimental data set consisted of read-
ings taken from in situ sensors, and included the measurement of
dissolved oxygen, conductivity, pH, temperature, and depth. The
control data set was obtained by measuring the same para-
meters with auxiliary instruments. A set of differences was
produced over a 2-week period which represented the error caused
by environmental degradation of the in situ sensor. After the
2-week period the unit was removed from the water column and
prepared for deployment in the next experimental site.
Preparations for deployment included inspection of the assembled.
instrument, disassembly and inspection, cleaning, reassembly, and
calibration. Several methods of statistical analysis are being
used in analyzing the data sets. For instance, simple regression
analysis is being used to determine the rates of error inducement
for a location in a time period.
We can make several recommendations for research efforts
based on biofouling alone since this does appear to be a prime
factor affecting the longevity of sensor stability. Several ap-
proaches to this problem would include:
• Modifications of the sensor to restrict light.
• The development and utilization of anti-fouling
agents which inhibit biological growth, and
• The development of new probes which will be
resistant to fouling of all kinds. This would
call for basic research in materials science.
Physical factors which were observed to affect hydrolab con-
trol in water columns were tides, waves, and bottom sediments.
Parameters in group II describe some of the driving forces
which cause changes to take place in the other parameters. They
are also useful in developing models to predict the movement and
172
-------
dispersion of pollutants as well as various living organisms.
(The sensor to be developed by the USGS RFP could very well be
placed in this category.) Correlation of these parameters with
others can significantly improve the understanding of the complex
forces at work in the estuarine area. For example, knowing
rainfall, wind velocity and evaporation rate would be useful in
explaining and predicting changes in salinity and dissolved oxy-
gen. Relatively little development work on sensors for this
group is anticipated; however, signal conditioning and interface
circuits must be designed and tested.
The parameters in group III are those for which major de-
velopment work is needed. At present there are no field
instruments for in situ monitoring of chemical oxygen demand
(COD). This parameter is generally determined by taking water
samples and returning to the laboratory for analysis. Suspended
particles are usually detected by optical techniques employing a
light source and lenses. Fouling of the optical components is a
major problem. One chlorophyll sensor using a filtered
fluorometer was incorporated in a water quality indicator system
developed by Magnavox for the NOAA data buoy office. The system
was developed in 1975 and has been used only on a limited basis.
There are sufficient data to evaluate the applicability of this
sensor to the estuarine environment. It is anticipated that
fouling of the optical system will again be a major problem.
There is no in situ sensor for detecting hydrocarbon content in
the water or for detecting a thin layer of oil on the surface.
Large oil slicks can be detected by remote sensing; however,
small amounts of oil are difficult to detect, especially when
emulsified in a surf zone and then carried into the estuarine
area.
There are other parameters which will require monitoring
that are very difficult to measure and will require intensive
work in the R and D sector before continuous in situ measurements
can be made. These include BOD, heavy metals, pesticides, toxic
chemicals, petroleum products in general, and others.
These deficiencies emphasize a need for basic research and
the development of new approaches to monitoring systems.
If we take the definition of water quality to be the
capability of supporting both plant and animal life, then it can
be used as a basis for suggesting areas of potential sensor re-
search.
Support of plant and animal life implies two important
ideas: (1) the required elements are present in sufficient
quantity; and (2) these elements have been combined into the
proper molecular structure for use in normal metabolic processes.
The latter implies that some molecular structures of the same ele-
ments may indeed produce abnormal physiological processes if used
as metabolites by a living system. It is felt that less than
173
-------
sufficient attention has been focused on the second idea,
molecular combination of elements integral to our water sources.
In particular it has been suggested by Professor F. Warner,
Louisiana State University, that basic research be undertaken to
develop water sensors capable of providing some structural
information about biological trace metal (lead, zinc, selenium,
copper, iron, cadmium, platinum, etc.) complexes inherent In
water sources. New knowledge gained from present biophysical re-
search (investigation of iron and platinum in malignancy) clearly
shows that there are three basic reasons for abnormal biological
development in the presence of trace metals: (1) the presence of
the element itself; (2) if present, the molecular structure of
the metal-containing complex; and1 (3) the quantity of metal com-
plex available for biological reaction. For example, we have
found, from a study of platinum binding to serum proteins, that
several complexes could be formed depending on the reacting metal
complex and the quantity of the complex present. The protein
complexes formed ranged from primary binding of platinum to
transferring to forming polymers of albumin. In a recent issue
of the journal, "Clinical Chemistry," the problem of reactions
between platinum chemotherapeutic complexes and serum proteins
was exposed and lack of knowledge concerning the bio-structure of
these secondary complexes was discussed. It is obvious that
there needs to be additional work done to identify not only ele-
ments present but, just as important, the chemical complexes
formed from these trace metals.
As an initial research subject for development of sensors
to abstract some chemical structural data, on line, Professor
Warner suggests that consideration be given to nuclear magnetic
resonance spectroscopy. What is involved here is developing a
small flow-through spectrometer to look at chemical shifts in
carbon, nitrogen, or hydrogen, such as those used currently in
measuring in vivo blood flow. The transmitted data could then be
compared with known spectra generated by metal-containing com-
plexes. Since magnetic resonance is a phenomenon dependent on
molecular magnetic fields, further information may be obtained by
photo excitation or paramagnetic electrons during NMR ex-
amination.
There are numerous physiochemical processes that may,
indeed, provide the basis of water quality determination in the
future, but the time is now to accomplish the basic research. Of
utmost importance in this research will be the investigation of
innovative techniques for taking advantage of physiochemical
processes.
Professor Jim Robinson, Louisiana State University, has
suggested that we develop a package for monitoring turbidity on a
continuous basis by using a deuterium light source which is quite
174
-------
stable and a detector. Monitoring would be accomplished by
measuring the intensity of radiation reaching the detector.
A second area of interest to Professor Robinson would be in
monitoring of oils and greases by using a source of infrared
radiation and monitoring a reasonably narrow wavelength region
which is typical of oils and greases. This would not be difficult
to do and is well within our capability to demonstrate.
A third area of great interest would be in the monitoring of
heavy metals. Most of the available sensors in this area are
based on electrochemistry and they break down in practice because
they determine only the ionic concentration rather than the total
metal concentration. We feel that some development work should
be carried out in this area as a priority item.
These are certainly not the only areas open for good re-
search and development projects. There is a continuing need to
upgrade power sources/ microprocessing techniques, and com-
munication linkages.
175
-------
ELECTROCHEMICAL TECHNIQUES IN WATER QUALITY EVALUATION
by Roger G. Bates, Ph.D.
University of Florida
Gainesville, FL 32611
Modern electrochemical analytical techniques applicable to
the assessment of water quality may be divided into six classes*
conductivity, potentiometry, amperometry, coulometry, chro-
nopotentiometry, and voltammetry. These procedures are based on
either an evaluation of the thermodynamic state of the water sys-
tem or on a measurement of transport properties. The quantity
ultimately measured is either resistance, emf (voltage), time, or
current. Some of t'hese techniques are simple and adapted with
relative ease to both field use and automation, while others de-
mand elaborate equipment and are better suited to the laboratory
(1). Y
Conductivity is a classic technique for the routine and
rapid examination of overall water quality. The measurement is
easily performed in the field and is a superior technique for the
determination of salinity. Accurate conductivity measurements may
be used as a check on the gross accuracy of a water analysis.
Nevertheless, conductivity is nonspecific, reflecting the sum
total of ionic impurities in the sample and ignoring the presence
of nonionic solutes such as soluble organic material.
As applied to water analysis, potentiometry has already
proved highly useful and further applications of this technique
appear highly promising. Measurements of pH utilize the emf de-
veloped between a glass electrode and a reference electrode to
assess the level of hydrogen ion activity or concentration and
the alkalinity of the sample. The newer ion-selective elec-
trodes, which are now available for some 20 different ionic
species, enable the method to be extended to the routine evalua-
tion of other important parameters such as the concentrations of
sodium, chloride, ammonium, fluoride, lead, copper, sulfide, and
cyanide ions. Water hardness can be measured by this method, and
simple modifications permit the level of dissolved carbon dioxide
to be established.
The measurement of emf with highly sensitive digital
voltmeters is simple and readily made in the field. Temperature
regulation need not be elaborate, but specially designed flow
cells may be needed, plus some provision to avoid contamination
with atmospheric carbon dioxide. The emf responds in logarithmic
fashion, and thus the method becomes more sensitive as the con-
centration decreases.
In spite of their versatility, potentiometric methods are
not without their drawbacks. The emf responds to the activity of
the ionic species rather than to its concentration. Furthermore
176
-------
serious errors may be involved if liquid-junction potentials at
the reference electrode are not balanced out of use of calibra-
tion standards which match closely the ionic strength of the
water sample. This objective is difficult to achieve at the
present time, and further study of this problem is needed.
Finally, no electrode is perfectly selective, and care must be
taken to avoid interferences; for example, for the accurate
estimation of the concentration of sodium ion, the concentration
of this ion must be at least a thousandfold greater than that of
the hydrogen ion in the solution.
In amperometry, a steady-state current is measured and
related to the concentration of an electroactive substance. A
polarizable electrode (dropping mercury electrode or rotating
platinum electrode) is used, together with an external applied
voltage so chosen that the limiting current is proportional to
the concentration of the species to be measured. A simple cir-
cuit utilizing a microammeter to measure the current may be suit-
able for the highest accuracy. The need for an external source
of potential can be avoided by choosing electrodes such that a
constant potential is developed within the cell.
The amperometric determination of dissolved oxygen in the
field is highly satisfactory, and commercial instrumentation is
available for this purpose. Modifications more suitable for
laboratory use permit chemical oxygen demand and total oxygen
demand to be evaluated. An instrument applying amperometry to
the estimation of chlorine residuals in the field has been
described (2). Its sensitivity appears to be a few parts per
billion. Iodine released from a KI reagent by the dissolved
chlorine is determined amperometrically. Amperometric procedures
can also be modified to permit the separate estimation of
chloramines.
Coulometry and chronopotentiometry are highly precise tech-
niques, adaptable both for major constituent analysis and for
trace analysis. Nevertheless, their automated application in the
field appears limited.
Voltammetric methods of analysis depend on the current-
voltage relationships at a polarizable electrode. The many
variations include conventional polarography, square-wave and
pulse polarography, cyclic voltammetry, and anodic stripping
techniques. The latter procedure is a highly sensitive trace an-
alytical method, especially for the estimation of metal ions.
Its application in monitoring water quality is discussed in de-
tail by another participant in this workshop. Voltammetric
methods are useful for determining aluminum, arsenic, cadmium,
copper, lead, and mercury, as well as certain anions, notably the
halides, cyanide, sulfate, and sulfite. Sometimes separations
are required, however, which render these techniques time-
consuming and poorly adapted for automated field use.
177
-------
A complete monitoring system for following the technological
process of surface water treatment of Danube raw water has re-
cently been described (3) and is under further development. The
data acquisition system uses 15 sensors with analog outputs and 7
instruments with digital outputs. The physical and chemical
properties monitored are flow rate, temperature, turbidity, pH,
pCl, conductivity, chemical oxygen demand, alkalinity, hardness,
and free chlorine.
178
-------
References
(1) E. J. Maienthal and J. K. Taylor, Chap. 32 in "Water and
Water Pollution Handbook", ed. by L. L. Ciaccio, Vol. 4,
Marcel Dekker, New York, 1973.
(2) G. Marinenko, R. J. Huggett, and D. G. Friend,
Journal of Fisheries Research, Canada, 33, 822-826 (1976).
(3) T. Cserfalvi, T. Meisel, B. Tarnay, K. Seybold, P. Galina,
and E. Pungor, Z. Anal. Chem., 282, 351-355 (1976).
179
-------
WATER QUALITY SENSOR TECHNOLOGY USING ACOUSTIC TECHNIQUES
by F.H. Middleton
University of Rhode Island
Kingston, R.I. 02882
Introduction
This talk will emphasize an acoustic method of sensing the
thickness of an oil layer floating on the top of either fresh or
salt water. The system is capable of functioning in quite rough
sea conditions because of the effective employment of a micro-
processor. Some mention will be made of other situations in
water quality sensing wherein accoustic techniques offer some
unique advantages in comparison to any other device.
Oil Thickness Sensor Description
An early model of the acoustic oil thickness sensor was de-
veloped in 1974 without the incorporation of a microprocessor.
The first system served to establish the validity of the tech-
nique even though it required taping large amounts of acoustic
data for subsequent processing on a lab computer. The most
notable disadvantage to this first system was the turnaround time
required between taking the test data and reading out the layer
thickness as a function of time.
Figure 1 is a photograph [Note: Photographs are not re-
produced in this report] showing a typical oil management test
wherein the Sonar Oil Thickness Sensor (SOTS) was employed for
the first time off the California coast. Figure 2 is a pho-
tograph of an oil recovery device in a test performed by the U.S.
Coast Guard. Measurement of the actual oil layer thickness both*
inside and outside the recovery device is the most important
measurement of all.
Figures 3, 4, and 5 outline the system evolution, output
capabilities and data retrieval capabilities. Figure 6 shows a
block diagram of the buoy SOTS system. The SOTS electronics is
little more than a refined acoustic Fathometer used to sense the
water depth from a boat by acoustic means. The SOTS transducers
are small commercial acoustic devices which are resonant near the
frequency of 1-MHz. One transducer radiates a short acoustic
pulse and the other receives the total acoustic echo from any
acoustical reflectors located above the transducer. These echoes
are highly amplified and the microprocessor seeks out the
specific echoes of concern from both the underside and the top
side of the oil layer. It is the travel time difference between
these two echoes which corresponds to the oil layer thickness.
180
-------
SOTS Output
There are three crucial points that should be mentioned
because of their importance to the ultimate performance of the
SOTS. The first is the convenient physical fact that because of
the acoustical geometry, the first echo (from the underside of
the oil surface) is the weaker of the two acoustical echos. This
means that even a very high intensity from the top of the oil
surface cannot "mask" or override the weaker echo.
The second major point is that even though a rough sea sur-
face can have a relatively large tilt relative to the horizontal,
the acoustic system can still function. This is because a steep
wave slope is both preceded and followed by a wave crest and a
trough,, both of which have essentially zero slope. Finally, the
microprocessor memory makes it simple to ride smoothly over these
echo "drop outs" without losing important thickness information.
The microprocessor makes it possible to focus on any one of
a large variety of output oil thickness parameters. Figure 15
illustrates this point and a typical parameter of interest to the
system user might be the average oil layer thickness at the SOTS
transducer location for the immediately preceding 5-minute peri-
od. The microprocessor makes it possible to change to any other
parameter such as the highest value of layer thickness during a
particular time period of a test. One might be more interested
in the oil thickness RMS value or its variance during a
particular time period.
Controlling the SOTS buoy system underutilizes the Motorola
6800 microprocessor so that other functions may be handled by its
reserve capacity. One of these is the pre-conditioning of data
and even the storage of data over considerable time periods.
This stored data might be rapidly extracted from a disk or tape
file (i.e., by satellite) or recovered when the buoy is retrieved
from its field station.
Figure 7 shows the ultimate oil layer thickness sensing res-
olution limit as it depends upon the selected acoustic frequen-
cy. The ultimate best use for the SOTS is in measuring the
thickness of significant oil layers rather than in detecting a
very thin oil sheen on the surface of the sea or within an es-
tuary. Optical techniques would appear to always have the
advantage over acoustics in this thin film case. However, at
about perhaps 1/2 mm and on up to many cm, the acoustic technique
would probably be the best possible approach that is available.
Conclusions
Figure 8 shows in an elementary way some other regions where
acoustic sensing can be of unique value in water quality sensing
applications.
181
-------
Perhaps the most important two of these would be in the de-
tection of suspended particulate (sediment) matter and in the de-
tection of flow profiles. The "scatter cross-section" for op-
tical energy in the atmosphere depends upon the data. The bucket
dipping, weighing, drying technique applied to measuring particle
size distribution is indeed crude, time consuming, and costly.
Conventional current meter arrays are very expensive, even if
they would have a chance to survive in the violent debris-laden
run of a river in the spring.
182
-------
EVOLUTION
GENERATION
1 - AFTER THE FACT PROCESSING AWAY FROM
COLLECTION SITE
- STATISTICS COMPUTED BY HAND MEASUREMENT
2 - REAL TIME PROCESSING
- STATISTICS COMPUTED BY HAND MEASUREMENT
3 - REAL TIME PROCESSING ON SITE
- STATISTICS COMPUTED REAL TIME WITHOUT ADDED
ERRORS DUE TO MEASUREMENT BY HAND
Figure 3. SOTS system evolutioi
183
-------
OUTPUT CAPABILITIES
REAL TIME MONITOR
TELETYPE - FULL MESSAGE CAPABILITY
7 SEGMENT LEDS - LIMITED MESSAGE CAPABILITY
ANALOG - METER - NO MESSAGE CAPABILITY
STORAGE MODE
DIGITAL - MAGNETIC TAPE
PUNCH TAPE
DISK
ANALOG - STRIP CHART
MAGNETIC TAPE
Figure 4. SOTS output capabilities.
184
-------
DATA RETRIEVAL
BASIC SYSTEM
OIL THICKNESS VS. TIME
BAD DATA DETECTION
EXTENDED SOFTWARE SYSTEM
MAXIMUM THICKNESS
MINIMUM THICKNESS
MEAN THICKNESS
THICKNESS VARIANCE
EXTENDED HARDWARE SYSTEM
NAVIGATION/LOCATION
OIL REFLECTION COEFFICIENT
Figure 5. SOTS data retrieval capabilities.
185
-------
DATA RETRIEVAL
BASIC SYSTEM
OIL THICKNESS VS. TIME
BAD DATA DETECTION
EXTENDED SOFTWARE SYSTEM
MAXIMUM THICKNESS
MINIMUM THICKNESS
MEAN THICKNESS
THICKNESS VARIANCE
EXTENDED HARDWARE SYSTEM
NAVIGATION/LOCATION
OIL REFLECTION COEFFICIENT
Figure 5. SOTS data retrieval capabilities.
185
-------
00
a\
MPU
,_ I] (^OUTPUT
Figure 6. SOTS block diagram
-------
S 1,0
OIL THICKNESS
RESOLUTION
FREQUENCY (fflz)
oil thickness resolution
187
-------
00
00
TURBIDITY
MEASUREMENT
FLOATING DEBRIS
DETECTION
Figure 8. SOTS alternate applications
-------
WORKING PANEL REPORTS
This section contains the reports prepared by six Working
Panels at the Workshop. The panels consisted of a Measurement
(Sensor) Needs panel which addressed Federal agency water quality
measurement needs that might be met by in situ sensor technology,
four Sensor Technology panels which reviewed specific sensor areas
in terms of applicability to automated in situ deployment, and a
Needs and Technology Integration Panel which correlated and
summarized the outputs of the needs and technology panels. Panel
titles and their chairmen, in order of presentation on the
following pages are:
Measurement (Sensor) Needs Clifford Risley, EPA
Electrochemical Sensor Technology Richard A. Durst, NBS
Electrophysical Sensor Technology James H. Picken, USGS
Optical Sensor Technology Vincent G. Di Pasqua, USCG
Automated Wet Chemical Sensor Harlan L. McKim, COE
Technology
Needs and Technology Integration Barbara S. Pijanowski, NOAA
and
John D. Koutsandreas, EPA
The first five panels met concurrently during the last one and a
half days of the Workshop; the last panel deferred deliberations
to the day following the Workshop since it needed reports from the
other panels to complete its work. The last panel consisted of
chairmen from the other five as well as other interested
participants.
In order to approach discussions in a preplanned, systematic
and somewhat uniform manner, the chairmen participated in prework-
shop meetings to design panel discussion topics and formats.
Various ideas on sensor category breakdowns were explored by the
workshop planners and consultants before selecting the four
technology categories used in the Workshop. Recommended individ-
uals were recruited as panel chairmen and draft discussion topics
and outlines were reviewed and drafted. All chairmen met several
weeks before the workshop to exchange ideas and to develop uniform
approaches to panel discussions and reporting. This preworkshop
preparation proved to be valuable in minimizing confusion. (The
panel chairmen are to be commended for their extra effort and
dedication in preparation for the Workshop.)
189
-------
Once involved in the deliberations at the Workshop, panel
chairmen found it necessary to modify slightly the "standard
format" for written reports in order to accomodate their specific
breakdown of sensor categories. Panel reports are presented on
the following pages. In compiling these reports, limited
editorial license has been exercised in an attempt to provide
further similarity and uniformity in the writings contributed by
numerous authors and chairmen. Complete uniformity was not
achieved or sought in order to preserve thoughts and philosophy as
reported.
The Panel membership is listed in the beginning of each panel
report. All Workshop attendees (see the roster at the end of this
report) were requested to specify their interests for panel
participation. As much as possible, within the limits of
balancing panel membership, assignments were made according to
these desires.
MEASUREMENT (SENSOR) NEEDS: Working Panel Report
Members
Clifford Risley, EPA, Chairman
01in D. Bockes, USDA
Richard W. Paulson, USGS
Charles R. Eastwood, NASA
Earl E. Eiker, COE
Victor W. Lambou, EPA
Enrico Mercanti, NASA
Nelson L. Milder, NASA
Charles A. Whitehurst, Louisiana State University
Mary S. Hunt, University of California
Paul C. Etter, MAR Inc.
Leslie G. McMillion, EPA
Barbara Pijanowski, Chairman of the Integration Panel, spent
considerable time with this panel.
Summary
This workshop has brought together representatives of a number of
agencies who have considerable experience and requirements for
monitoring water quality. All are in agreement that the need for
monitoring is increasing exponentially and that Federal agencies
are frustrated by their inability to expand their monitoring
efforts sufficiently to keep up with the demand. No agency has
sufficient manpower to provide the required monitoring effort
using the traditional approach of field sampling and laboratory
analysis. There is complete agreement that we must develop
automated monitoring systems that can measure the parameters of
need, with accuracy, simplicity and reliability.
190
-------
There is also concurrence that we have an already demonstrated
ability to build reliable in situ monitoring systems which have
operated dependably for weeks at a time with reliable sensors,
data control packages, memory units and recording on tape.
Also, automated field record and field data memory units which
dump data to satellites or telephone lines are well developed and
used by several agencies.
The major problem with present systems is that so few reliable
sensors exist. It is readily apparent that the availability of
only a few reliable sensors for field applications is the primary
constraint on the application of automated monitoring to the
management of water quality and water resources. Many of the
needed parameters cannot now be sensed automatically. High
priority must be placed on sensor development.
The panel recognized that for some of the parameters of interest,
we do not have good analytical procedures in the laboratory. We
feel that the effort to develop good field sensors will have a
payoff benefit to the laboratory as well.
An important part of the automated monitoring package should be a
sample collection component which can be triggered to collect
samples when the sensors detect pollution levels in excess of
prescribed limits.
The panel also found that there may be excellent opportunities for
reducing the costs of obtaining and analyzing groundwater samples
by use of automated in situ devices. These devices, once
installed, would reduce the time and cost that is entailed by
field personnel going to the well sites, setting up pumping
equipment over each well, obtaining the samples and transporting
them to the laboratory for analysis.
The automated devices could be selected for determination of key
parameters, installed and occasionally serviced in conjunction
with water-level measuring devices which are generally installed
in the present operational scheme.
State and Federal Agencies need automated monitoring:
• Spill alerts, ocean, lake and river
• Lake monitoring, where spill potential exists
• Estuary monitoring, where spill potential exists
• Ocean monitoring, shipboard and buoy
• In-stream monitoring for water quality trends
191
-------
• Ocean monitoring, shipboard and buoy
• In-stream monitoring for water quality trends
• Wells and ground water supplies
• Short term intensive studies for enforcement actions
• Within the soil mantle
• Data for water quality model development
• Data for planning corrective actions
• Data to support best available technology (BAT)
requirements for 129 Toxic Compounds
• Municipal waste treatment plant effluent permit compli-
ance
• Industrial waste treatment plant effluent permit com-
pliance
• Rural and urban runoff characterization
• Storm and combined sewer overflow measurement
• Data to determine long term toxic buildup
• Data to develop land use relationships
• Nonpoint source water pollution monitoring.
State and Federal Agencies also need emergency response
capability. They need to move out to a field site via helicopter
or other rapid transport to define affected areas and determine
the concentrations of pollutants and assess the seriousness of the
incident. Action must be taken to confine the pollution, alert and
protect the public and take corrective action. Rapid intensive
monitoring is necessary to make these decisions and to measure the
effectiveness of corrective actions after they are taken.
Monitoring for the above needs requires that automated devices and
their sensors be able to perform at the water surface, sub-
surface at many depths (ocean, lakes, rivers and groundwater)
referenced to water surface and to bottom surface, on the bottom
and at various depths in the soil mantle^ The devices must also
work in a temperature range from 0 to 40 C, and must be
resistant to fouling by bacteria, algae, sediment, and chemical
corrosion. They must also withstand physical disturbance from
currents and wave action.
192
-------
The Measurement Needs Panel decided that it should concentrate its
efforts on the development of a list of parameters for which the
greatest present need and anticipated near-term needs exist. The
parameter list which the panel developed was grouped into three
categories and is attached to this panel report (Table 1).
Group 1 Parameters of need to all agencies (state, federal
and private).
Group 2 Water quality indicators needed by many agencies
but carrying less concensus among agencies as to
importance ranking.
Group 3 Water quality indicators of specialized importance
to some agencies but not of high importance or
concern to all agencies.
We see a great need to further develop these needs by having each
agency user group address the parameter list to determine its own
applications and requirements. We propose that a committee be
appointed to make this survey and that the return information be
made available for use by a subsequent interagency water quality
sensor workshop.
General Requirements for All Sensors and Systems
Reliability and Quality Assurance — Due to the increase in costs
of travel and manpower, and the need for monitoring data, our goal
would be to have sensors with a 60-day station life-time and up to
6-month time interval between maintenance. The type of sensor
will largely determine the possibility of realizing this goal.
Other desired goals:
• Longevity (3 to 5 years or longer).
• Mean time between failure or service cycle (resupply of
reagents, filters, etc.) (3 months to 1 year).
• Cost of sensor less than $1,000 each in production of
greater than 1,000 units/year.
Complexity — The sensors should be kept as simple as possible;
should be modular, plug in, easily replaceable.
Operation and Maintenance — Sensors should operate unattended for
a 60-day maintenance schedule. Maintenance should be performed by
one person in the field with little equipment necessary.
193
-------
Modular components must be small enough to be replaced in the
field.
Accuracy — This is dependent upon the parameter under considera-
tion, environment, considered applications, frequency of measure-
ments.
Conclusion — In conclusion the Panel wants to reemphasize its
previous statement that we see a great need to further develop
these needs by addressing the parameter list to each agency user
group to determine their applications and requirements. We
propose that a committee be formed to make this survey and that
the return information be made available for use by a subsequent
interagency water quality sensor workshop.
194
-------
TABLE 1. HIGH PRIORITY MEASUREMENT NEEDS
Group 1 - Needs of all agencies (state, federal, and
private)
Temperature
Depth
Water velocity and direction
Dissolved oxygen
Conductivity (fresh water)
pH
Optical properties
Density (sea water, ground water)
Suspended particles (concentration, size,
distribution)
Group 2 - Needs of many agencies
Total dissolved solids (fresh water)
Oil presence (visible sheen 1-3 micron)
Oil quantity
Phytoplankton biomass (chlorophyll a)
Total phosphorus
Dissolved phosphorus
Available nitrogen
Organic carbon
Dissolved organic carbon
Chlorine - wastewater
Group 3 - Specialized needs (of major importance to some
agencies, but not to all agencies.)
Chloride - fresh water
Coliform
ATP
BOD
Viruses
Sulfate, Sulfides
Zooplankton
Metals*
Pesticides/insecticides (herbicides)*
Hydrocarbons
Aromatics*
Industrial solvents*
Phenols*
Benzene toluene group*
Miscellaneous compounds*
Radioactivity (gross oc, gross j3, gross y)
Ozone (fresh water)
7see~~~TabTe2
195
-------
TABLE 2. GENERIC GROUPING OF 65 SETTLEMENT
AGREEMENT TOXIC SUBSTANCES*
Pesticides/Insecticides
Aldrin/Dieldrin
Chlordane and metabolites
DDT and metabolites
Endosulfan and metabolites
Endrin and metabolites
Heptachlor and metabolites
Isophorone
Toxaphene
2,3,1r8-Tetrachlorodibenzo-para-dioxin
Metals
Antimony and compounds
Arsenic and compounds
Beryllium and compounds
Cadmium and compounds
Chromium and compounds
Copper and compounds
Lead and compounds
Mercury and compounds
Nickel and compounds
Selenium and compounds
Silver and compounds
Thallium and compounds
Zinc and compounds
Iron
Manganese
Calcium
Magnesium
Potassium
Silicon
Aromatics
Polynuclear aromatics
Polychlorinated biphenyls (PCB)
Napthalene
Chloronated Napthalene
Phthalate esters
Industrial Solvents
Carbon tetrachloride
Chlorinated benzenes
Chlorinated ethanes
(Continued)
196
-------
TABLE 2. (Continued)
Chloroakyle ethers
Chloroform
Dichlorobenzenes
Dichlorobenzidine
Dichloroethylene
Dichloropropane and dichloropropene
Haloethers
Halomethanes
Hexachlorobutad ine
Hexachlorocyclohexane (all isomers)
Hexachlorocyclopentadiene
Tetra chloroethylene
Trichloroethylene
Phenol Family**
Chlorinated phenols
2-Chloro phenol
2r 4 Dichlorophenol
2, 4 Dimethylphenol
Nitrophenols
Pentachlorophenol
Phenol
Benzene Toluene Group
Benzene
Benzidene
Dinitro-toluene
Ethyl benzene
Nitro benzene
Toluene
Miscellaneous
Acrolein
Acenaphthene
Acrylonitrile
Asbestos
Cyanides
Diphenylhydrazine
Fluoranthene
Nitrosamines
Vinyl chloride
Fluoride
**Not part of Settlement Agreement list. Added by Working
Panel.
*From Consent Decree between Natural Resources Defense Council
and U.S. Environmental Protection Agency, June 7, 1976.
197
-------
OPTICAL SENSOR TECHNOLOGY: Working Panel Report
Members
Vincent G. DiPasqua, USCG, Chairman
Richard W. Newton, Texas ASM University
William H. Kirchoff, NBS
Eduardo D. Michelena, NOAA
Robert F. Middelburg, USGS
Jack A. Salzman, NASA
Frank A. Schiebe, USDA
Summary
This panel considered a number of optical techniques including
passive and active spectroradiometry, transmissometry, and
polarimetry. Some of these techniques do not require direct
contact with the water and, as a result, require less mainte-
nance than water contact sensors. All are dependent upon
electromagnetic radiation from the water bodies. They can be
made to scan large areas, and provide synoptic measurements.
The active systems and some passive are capable of performing
measurements night and day. Water parameters which can be
detected include temperature, water velocity, optical properties
suspended particles, salinity, oil presence and quantity, sedi- '
mentation, turbidity, and chlorophyll a_. The quantification of
suspended sediments and oil presence has been demonstrated by
lidar polarimetry. Spectrographic techniques could be developed
to identify the more complex pollutants such as pesticides and
hazardous materials. Differential radiometry can be used to
measure chlorophyll concentration and turbidity.
Sensor Technology
Definition of Technology—
There are a number of optical in situ and near in situ
water quality sensors now available. Many of these devices
use the same measurement technique. As a result, it was decided
to group the measurement concepts into several broad classes and
subgroup specific sensors within the appropriate measurement
concept class. The general classes of measurements considered
were:
CATEGORY I: Passive Spectroradiometric
a. on site non-contact
b. in situ
CATEGORY II: Active Spectroradiometric
a. on site non-contact
198
-------
CATEGORY III: Transmissoffieters
a. artificial broadband light source
(transmission, 90% scatter)
b. artificial narrowband coherent light source
(forward scatter, 90% scatter)
c. natural light source
(index of refraction)
CATEGORY IV: Lidar Polarimeters
a. on site non-contact
b. in situ
Advantages—
• many optical sensing techniques do not require contact
with the water; measurements can be made from a distance
• capable of scanning large areas to provide synoptic
measurement
• maintenance requirements are less for those sensors not
in water contact
• pretreatment of samples is not required
• provides the capability of multi-parameter measurements
• active systems are available 24 hours per day
Disadvantages—
• fouling of the lens
• susceptible to other radiation source
• passive systems have daytime limitations
Detection Capabilities—
Group I
a. temperature
b. water velocity
c. optical properties
d. suspended particles
e. salinity
Group II
a. oil pressence
b. oil quantity
c. phytoplankton biomass (chlorophyll a)
Group III
a. potential to detect many items listed via spectrographi<
techniques
Near-Term (5-year) R & D Recommendations—
Emphasis should be placed on anti-fouling engineering and
miniaturization and size reduction of existing sensor systems.
These sensors which detect suspended particles should be
advanced to operational field use as soon as possible.
199
-------
Long-Term R & D Recommendations—
All sensors require improved reliability and. precision.
Many sensors have multi-scanning detection capability that needs
to be developed over a long time period. This is particularly
true of microwave spectroradiometry.
Remarks and Conclusions—
Optical properties, water turbidity, oil detection and
salinity are now detectable, and to a certain extent quanti-
fiable measurements can be made by existing sensors. We still
are unable to detect most of the priority measurements listed in
Group II and III. The performance of existing in situ optical
sensors today is limited. Of the Group II parameters, develop-
ment of water velocity by laser velocimeters could be considered,
Spectrographic techniques should be developed to identify
Group III parameters. The spectrographic wavelength variability
allows for multiple parameter detection using optical sensors.
Individual Sensor Applications/Characteristics—
Results of panel discussions on various individual sensors
appear in outline form on the following pages, addressing them
by measurement concept, as listed in the four categories
described above.
200
-------
Spectroradiometers (passive)
(General)
1. Applications
Reflectance at various wavelengths.
Inorganic sediment particles suspended in the air.
2. State of the Art
a. Status
b. Capabilities/Limitations
Error limits not defined at this time; small size,
weighs about 20-25 Ibs; region (large area) oriented;
used for fresh water but can be used for salt water,
also easily adapted for use in water; limited by
sunlight.
c. Expertise/Source
Dr. Frank Schiebe, U. S. Department of Agriculture.
Dr. James Bailey, U. S. Navy, Office of Naval Research,
d. Cost
$2,500 off-the-shelf.
3. Short-Term (5-year) R & D Possibilities
Once wavelength parameters are established, cost should
decrease, many substances could be identified, i.e., algae,
chlorophyll a.
201
-------
Infrared Reflectance (Active/Passive Radiometer)
1. Applications
Detects oil on water.
2. State of the Art
a. Status
Has been developed/ presently attempting to manu-
facture for routine use.
b. Capabilities/Limitations.
Mounted above the water. Can be roonostatic
(transmitter and receiver side by side) or biostatic
(transmitter and receiver mounted opposite each other
across a body of water). In the biostatic receiver
mode it scans (by a motor) the water, allowing
examination of a broad expanse of water body. The
method is 90% accurate, has proven very reliable.
The two units are about 1 foot square, water and
explosive proof, free from weather, require no
servicing unless the pole they are mounted on changes
the line of vision. Can scan 500-plus feet, two feet
in diameter.
c. Expertise/Sources
Donald R. Jones, EPA, Washington, D.C.,
Lt. William Plage, U. S. Coast Guard, Washington, D .C.
Wright & Wright, Inc., Massachusetts. *'
3. Short-Term (5-year) R & D Possibilities
R & D complete at this time.
Plan to use system, 5-10 units, in the ports of Baltimore
and Houston in 1979 and 1980.
4. Remarks
If this system is successful in the ports listed above
then purchase by major oil companies is anticipated to
give early alarm to prevent expensive spill cleanup
costs due to late knowledge of spill.
202
-------
Spectroradiometer (Active)
(General)
1. Applications
a. Measurement of radiation reflected from a water surface
at various optical wavelengths; energy source is
provided internally. Measurements to be made remotely
(reflectance) or underwater (back scattering).
b. Inorganic and organic particles suspended in water,
infrared from the optical measurements.
2. State of the Art
a. Status
(1) Oil detection - USCG work.
(2) Other R & D unknown.
b. Capabilities/Limitations
Accuracy - good potential.
Reliability - good potential.
Size/Weight - could be developed in small package.
Water Type - fresh/salt.
Power Requirement - Could be considerable if CW is
used. Strobe systems could reduce power consumption
considerably.
Serviceability - Should be no problem.
Environmental Conditions - In air or underwater (fresh
or saline).
In air (-10° to 50°) .
Not affected by rain, wind, etc.
c. Expertise/Sources
No known research in direct subject. Manufacturing
by ISCO, Texas Instruments, Rambie.
Related research in passive spectroradicmetry at USDA
sedimentation laboratory (F. K. Schiebe and
J. C. Ritchie).
d. Cost
Operational - None known in routine operation
Capital - Could be quite low. Estimated less than $5000
for ultimate off-the-shelf instrument. Automation
would add to cost.
3. Short-Term (5-year) Development Possibilities
a. No known R & D for spectroradiometry.
R & D at Texas Instruments for specific bands (IR) .
Product available from Rambie.
b. Feasibility study needed to assess potential.
203
-------
Fluorescene Spectrometer (Active)
1. Applications
Chlorophyll £/ oils, dissolved organics.
2. State of the Art
a. Status
Chlorophyll a - Routine application for continuous
monitoring of strips in marine environments.
Developmental system on NOAA buoy system for field
testing.
Development needed for fresh water application.
Oil - Field-deployed buoy system for detection and
identification.
Total Organic Carbon (TOC) - Research.
Algae species identification - Research.
b. Capabilities/Limitations. Material identification
based on excitation wavelength and/or fluoresced
wavelength intensity measurement.
Chlorophyll a_ (marine environment) , 0-20 mg/m^ + 10%.
Oil detection of presence on surface, coarse identi-"
fication.
TOC: Not established.
Fouling of optical windows and components reduces
reliability (cleaning cycle is a function of water
condition).
Chlorophyll a and oil detection system - Jj cubic feet
and battery powered.
c. Expertise/Sources. Chlorophyll a - Turner Designs,
Mountain View, CA (modified for buoy use by Magnavox
Fort Wayne, Indiana).
Oil - Spectrogram Corp., North Haven, Conn.
TOC - Michael Bristow, EPA, Las Vegas.
Algae Species Identification - Jack Hall, NASA, Langley
d. Cos t.
Chlorophyll a - $10K.
Oil - $10K.
3. Short-Term (5-year) R & D Possibilities.
R & D ongoing for chlorophyll
-------
Active Light Transmission
(General)
1. Applications
Light Attenuation Measurements; 90° scatter, forward
scatter, differentiate between organic and inorganic
particulate matter.
2. State of the Art
a. Status
Light attenuation and scatter - operational, some
development.
Forward scatter - developmental.
b. Measurement of parameters of interest is dependent on
the wavelength selection and angle of measurement.
Accuracy, optical characteristics quite accurate but
relating to parameters of interest is only fair;
reliability is good; size and weight - light, compact,
easily portable; Water type - all; serviceability -
requires frequent service; environmental conditions -
dependent on sensor or design.
c. Expertise/Sources - Light attenuation by manufacturer.
Forward Scatter - Jack Salzman, NASA-Lewis.
d. Cost - Light attenuation and 90° scatter - $5K.
Forward scatter unknown.
3. Short-Term (5-year) Development Possibilities
Resolve fouling problems.
Increase unattended operation capability.
Make more automated.
Redefine and standardize calibration and measurement units.
205
-------
Transmissometer
(Active)
1. Applications
Transmission of light through the water to determine
trophic levels.
2. State of the Art
a. Status
Operational commercially available underwater use,
operates continuously. Presently used on ocean buoys
b. Capabilities/Limitations
Within 1% accuracy when first deployed. Life expec-
tancy one month underwater due to fouling problem.
Coating of sensor to retard organism fouling may be
possible 20 Ibs. (21 x 3" in diameter).
c. Expertise/Sources
Manufacturers - Hydro Products, MARTEK, KAHL,
Inter-Ocean; Dr. Ed Michelena, NOAA, NSTL,
Bay St. Louis, Missouri.
d. Cost
$3K off-the-shelf plus adaptation for NOAA use.
3. Short Term (5-year) R & D Possibilities
R & D ongoing is to develop methods for reducing fouling.
Another R & D need is for lower power consumption.
206
-------
Refractometer (Active)
1. Applications
Refractive index for salinity determinations
2. State of the Art
a. Status
Operational - commercially available hand-held model.
Developmental - automatic recording.
b. Capabilities/Limitations
Accuracy +.1% salinity for hand-held model.
+.2% for automatic recording model.
If pollutant concentration is high, the accuracy will
be degraded. Easy to clean.
Limited to source of light.
c. Expertise
Mr. Edward Brainard, ENDECO, Marion, Massachusetts.
d. Cost
Hand-held model (which is the operational unit) -
$1100. Recording type unit - $4500. Present record-
ing scheme uses photographs of the viewing window.
3. Short-Term (5-year) R & D Possibilities
Conversion to the use of a monochromatic light source.
Will increase accuracy and allow 24-hour operation.
4. Long-Term R & D
Develop sensor with an electronic digital output and also
with an analog voltage output.
207
-------
"Turbidity" Meter
1. Applications
Measures light transmission and scattering through water
to determine turbidity.
2. State of the Art
a. Status
Developed, being evaluated.
b. Capabilities/Limitations
Accuracy hard to define, probably better than 1%.
Fresh water use but can be adapted for salt water.
Window needs constant cleaning. Needs 110 V AC
power; % cubic foot size. Needs weekly service
and weather protection.
c. Expertise/Sources
James Ficken, U. S. Geological Survey, NSTL, Mississinn-5
d. Cost ppl
$20K for development: $2-5K for production.
3. Short-Term (5-year) R & D Possibilities
Used as routine monitor short- and long-term. Production
model could be completely automatic.
208
-------
Microwave Instruments (General)
(Microwave Spectroscopy and Microwave Radar)
1. Applications
a. Microwave spectrum for specific identification of
compounds.
b. Reflections from changes in dielectric constant in
a bulk medium.
2. State of the Art
a. Status
Present Application
Research into molecular structure and compound
identification.
Measurement of water content of materials (soil,
grain, snow, concrete, etc.) research stage only.
b. Capabilities/Limitations:
Highly specific but requires sample preparation
identical to GC mass spectrometry, i.e., isolate
compound in gas phase at low pressure.
Has not been applied to water monitoring.
Points to consider: accuracy, reliability, size/
weight, water type, serviceability, environmental
conditions.
c. Expertise Sources. At this state these instruments
require highly skilled operators. There are no longer
commercial sources of microwave spectrometers. Instru-
ments are laboratory-assembled from purchased components.
Sources are:
Microwave Spectroscopy - William Kirchoff, NBS
Gaithersburg, MD.
Microwave Radar - Hal Boyne, NGS, Boulder, CO.
d. Cost:
Microwave spectrometer cost is approximately $100K.
Microwave radar instrument could be assembled for
less than $10K. As there are no sensors in routine
operational use, operational costs are not available,
however, an instrument for measuring atmospheric
formaldehyde based on microwave absorption has been
used.
3. Short-Term (5-year) Development Possibilities
Development of microwave Spectroscopy as a routine analyt-
ical instrument is currently not being pursued. Packaging
of data handling and development of a GC-microwave spectrom-
eter interface is needed. A major drawback is the inability
to identify a new substance if its spectrum is not already
known. A major strength is specificity and capability of
identifying components of gas phase mixtures. Quantifi-
cation using isotopically-labeled spikes is also simple
209
-------
since spectrums of different isotopes are distinct.
Microwave radar has not yet been tested on water systems
but has been tested on the materials listed above.
Feasibility research would be useful.
210
-------
Lidar Polarimeter (Active)
1. Applications
Oil on water/ concentrations of suspended solids.
2. State of the Art
a. Status
Developmental
b, Capabilities/Limitations
Detect oil on water and detect concentrations of
solids in the water column; very accurate, depends
on particle size distribution; not affected by sea
state; scans an area; measurement can be adapted
to in-water use with research; 24-hour operation;
weather protected; serviceability depends on laser
efficiency; weight of 50 pounds; 24 inches long,
4-inch diameter cylinder.
c. Expertise/Sources
Dr. Richard Newton, Texas A&M University.
d. Cost
$9OK to date. Production cost indeterminate.
3. Short-Term (5-year) R & D possibilities
Usefulness can be demonstrated in short-term with addi-
tional research.
211
-------
Coherent Lidar Polarimeters (Active)
1. Applications
Concentrations of suspended solids, size distribution of
suspended solids, presence of oil on a water surface,
thickness of oil on the water surface, gross classifi-
cation of oil type.
Measurement technique is to transmit narrowband polarized
coherent light (laser) and measure the light backscattered
from a water target in both polarization states. The
measurement technique has been used as a non-contact measure-
ment technique, but is applicable to a true in situ sensor.
2. State of the Art
a. Status
The measurement concept has been demonstrated both
analytically and experimentally. An automated non-
contact single wavelength sensor has been built to
demonstrate detection of oil on water and the capa-
bility to measure concentrations of suspended solids.
A dual wavelength laboratory system has been built
and used to demonstrate the capability of oil thick-
ness determination and gross oil classification. The
status of the sensor concept would be considered
developmental.
b. Capabilities/Limitations
The ratio of the "cross" polarized backscatter to the
"like" polarized backscatter (termed depolarization
ratio) is related to the concentration of suspended
solids in the water volume. This relationship is
independent of surface characteristics of the water
when the instrument is operated in a non-contact mode.
The depolarization ratio is not only dependent on
concentration of suspended solids, but also on the
size distribution of these solids. As a result, by
using a multiple wavelength light source, measurement
of particle size distribution is conceivable. In
addition, in the in situ mode the instrument could be
used to profile concentrations of suspended solids
within a water volume by lowering the sensor through
the water or by pumping water from different levels
to the sensor.
In the non-contact mode (sensor above the water) r
the presence of surface contaminants such as oil
can be detected by the magnitude of the like or
cross polarized backscatter measurements at a single
wavelength. Two or more wavelengths add the capa-
bility of oil thickness determination and oil
212
-------
classification.
Serviceability and reliability depend primarily
upon the lasers utilized. The remainder of elec-
tronics required should be highly reliable.
c. Expertise/Sources
Dr. Richard Newton, Texas A & M University
Dr. John Rouse, University of Missouri
d. Cost
Production cost indeterminate at this time.
Short- and Long-Term Developmental Possibilities
a. Research and development is currently not being pursued
by any Federal agency either in-house or by outside
funding.
b. Development needs include:
(1) Investigation of effect of organic materials.
(2) Investigation to determine ability to profile
suspended solid concentrations.
(3) Investigation of ability to determine particle
size distribution.
(4) Investigation into ability of Raman scattering for
foreign substance identification.
213
-------
ELECTROCHEMICAL SENSOR TECHNOLOGY: Working Panel Report
Members;
Richard A. Durst, NBS, Chairman
Roger G. Bates, University of Florida
Walter J. Blaedel, University of Wisconsin
Chung-Chiun Liu, University of Pittsburgh
J. Anthony Llewellyn, University of South Florida
Gary Ward, 1JOAA
Judd R. Wilkins, NASA
Summary
The work of this panel was hampered by two deficiencies: time
and user expertise. Without the first-hand knowledge of sensor
users on the panel, it was difficult in many cases to thoroughly
address the types of problems encountered in real-world moni-
toring situations. Most of the panel participants were labora-
tory scientists having little experience with the problems of
field measurements. Future panels of this type should have the
representation of users with diverse monitoring experience.
Because of the lack of time, all parameters which are amenable
to electrochemical sensor monitoring were not discussed and
should be addressed at some future workshop. The parameters
which were not discussed included alkalinity, oxygen demand,
calcium, boron, carbon dioxide and carbonates, total organic
mercury, nickel, potassium, sodium, nitrate and nitrite, urea,
amino acids, redox, and ATP.
The Electrochemical Sensor Technology Working Panel divided
its discussions into two primary areas, amperometric and
potentiometric sensors. The amperometric sensors are a very
valuable and versatile technique for evaluating water parameters,
They are useful with analytical techniques as detectors for
liquid chroroatography, and in titration procedures, and have a
great potential for organic analysis. The operational mode
determines the response time. This technology is capable of
detecting electroactive substances such as: organic compounds,
metals, halides, synthetic organic compounds, oxygen and
chlorine, and oxygen demand. The potentiometric sensors are
small and inexpensive, and can be easily adapted to automated
in situ monitoring applications. They will normally require
sample pretreatment, but yield good accuracy with frequent
calibration. Capabilities exist for detecting selected cations
and anions, including fluorides, chloride, cyanide, sulfide,
calcium, coppers, cadmium, and potassium, coliform bacteria,
and redox.
214
-------
A summary of discussions of sensors in the amperometric
category follows:
Sensor Type--
Amperometric (Summary)
Measurement Applications—
Electroactive substances:
Metals
Organic Compounds
Halides
Synthetic Organic Compounds
Oxygen and Chlorine
Oxygen Demand
Advantages—
High sensitivity; economical; wide applicability - any
electroactive substance (direct or indirect); simplicity -
potentially portable for on-site applications (can be minia-
turized and amenable to battery operation); usable in all types
of water; amenable to a variety of operational modes including
amperometric titrations, differential pulse polarography,
anodic stripping voltammetry, cyclic voltammetry, and chroma-
tographic detectors. When operated in the steady-state mode,
the response is rapid.
Disadvantages—
Poor selectivity and interferences when used without
sample preparation. Electrodes will require periodic cleaning.
Near-Term R & D Recommendations—
Development of amperometric methods for providing specific
redox information; investigation of oxidation-reduction poten-
tial has a descriptive parameter for characterizing the redox
strength of water systems. Development of automated systems
for on-site monitoring.
Remarks and Conclusions—
Amperometric sensors are very valuable and versatile
techniques for the evaluation of many water parameters. High
capabilities for automated on-site measurements. Useful in
conjunction with other techniques, for example, as detectors
for liquid chromatography and in titration procedures. Great
potential for organic analysis. Operational mode determines
the response time.
Summary remarks regarding potentiometric sensors included
the following:
Sensor Type—
Potentiometric (Summary)
215
-------
Measurement Applications—
Selected cations and anions, including fluoride, chloride,
cyanide, sulfide, calcium, copper, cadmium, lead, potassium, and
sodium, and water hardness. Gases, including carbon dioxide,
ammonia, oxygen, sulfur dioxide, nitrogen oxides, and hydrogen
sulfide. Enzymes and substrates; coliform bacteria; pH; redox
(unsatisfactory).
Advantages—
Inexpensive; simple mode of operation; sensitivity suitable
for many monitoring applications; easily automated; good selec-
tivity with proper sample preparation; small size; adequate
response time for many applications; can be modified in various
ways for gas, biochemical and microbiological measurements;
portable; useable in all water types with sample preparation;
useful for both direct and indirect measurements; rugged; good
accuracy with frequent calibration and control of operating
conditions.
Disadvantages--
Selectivity normally requires sample pretreatment; long-
term instability; susceptible to fouling; temperature sensi-
tivity; liquid junction problems with reference electrodes.
Near-Term R & D Recommendations—
Devise methods for the elimination of liquid junction
problems. Improve methods for prevention or removal of fouling.
Development of better automated pretreatment methods. Selected*
enzyme electrode development. Development of microbiological
sensors; development of automated on-site systems for ion-
selective electrodes. Miniaturization.
Long-Term R & D Recommendations--
Major program in enzyme electrode research. Research and
development into new membrane technologies and ion-sensitive
field-effect transistors.
Remarks and Conclusions—
Should be easily adaptable to automated on-site monitoring
applications for continuous real-time analysis. Susceptible to
a variety of interferences and normally will require sample
pretreatment. Versatile for both direct and indirect measure-
ments. Membrane fouling and long-term instability are problems
which must be overcome. Response time needs improvement.
Present potentiometric redox methods are unsuitable and new
technology should be developed.
in
The Panel's review of various individual sensors is reported
the following outlines, which cover these parameters:
glass electrode
PH
216
-------
ion selective electrode
sulfide ion
chlorjde .ion
fluoride ior
heavy metals (copper, cadmium, lead and silver)
gas-sens ing_gTlec_trpdes
ammonia
hydrogen sulfide
carbon dioxide
sulfur dioxide
nitrogen oxides
redox hydrogen electrode
coliform bacteria
amperometric
heavy metals (copper, cadmium, lead, zinc)
amperometric^ galvanic, ^hronocouloinetric
dissolved oxygen
electrochemical
conductivity
217
-------
Potentiometrie (glass-electrode)
1. Applications
pll
2. State of the Art
a. Status
Operational
b. Capabilities/Limitations
The accuracy and precision is sufficient for most
applications when the electrode is functioning
properly but in practice may not be achievable. The
size and weight are not problems. There are no
interferences in natural waters. The sensor requires
a calibration frequency commensurate with the
accuracy needed. The electrode is sensitive to
electrical noise and is fragile.
c. Expertise/Sources
Roger G. Bates, University of Florida;
Becknian Instruments; and Corning.
d. Cost
$50-$400
3. Short-Term (5-year) Development Possibilities
R & D ongoing or programmed - No information.
R & D needs and gaps - Prevention of fouling,
calibration methods and high-pressure studies.
4. Long-Term Development Possibilities
R & D ongoing or programmed - Mo information.
R & D needs and gaps - The development of a pK
sensor based on a new technology but of low
priority compared to other sensor needs.
218
-------
Fotentiometric (ion-selective electrode)
1. Applications
Sulfide don
2. State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Susceptible to interferences and equilibria between
chemical forms. Usually requires pretreatment; pK
control is important. Poisoning of membrane by
mercury. Good stability.
c. Expertise
James Ross, Orion Research
d. Cost
$250-$500
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - No information
b. R & D needs and gaps - Development for deep ocean
measurements required.
219
-------
Potentiometric (ion-selective electrode)
1. Applications
Fluoride ion
2. State of the Art
a. Status
Operational
The sensitivity is sufficient for most applications.
It is useable in all types of natural waters , but
may require pre treatment in certain cases. It has
good stability, has a precision better than 5 percent
and 95 percent response in less than two minutes.
c. Expertise/Sources
James Ross, Orion Research; Martin Frant, Foxboro
Company.
d. Cost
$250-$600
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - Development of pre treat-
ment procedures for special applications.
220
-------
Potentiometric (ion-selective electrode)
1. Applications
Chloride ion
2. State of the Art
a. Status
Operational
b. Capabilities/Limitations
Rugged; useable in all types of water (requires
pretreatment in effluent monitoring); fast response?
good stability and sufficient sensitivity.
c. Expertise/Sources
Orion Research and Hydrolab
d. Cost
$250-$600
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - The development of
sample pretreatment methods for effluent monitoring.
221
-------
Potentiometric (ion-selective electrodes)
1. Applications
Heavy metals (copper, cadmium, lead, and silver)
2. State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Sensitivity is borderline for natural ambient levels;
precision is approximately 5-10 percent. Useable
in all water types but pretreatment necessary by
ionic strength adjustment and masking of interferences,
Electrodes require frequent calibration and recondi-
tioning.
c. Expertise/Sources
James Ross, Orion Research
d. Cost
$250-$600.
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - Research and development
ongoing.
b. R & D needs and gaps - Presently not as sensitive as
voltammetric devices but may be improved by new
membrane materials and the development of pretreat-
ment techniques.
222
-------
Potentiometric (gas-sensing electrodes)
1. Applications
Ammonia, hydrogen sulfide, carbon dioxide, sulfur dioxide
and nitrogen oxides.
2. State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Sufficient sensitivity; slow response time. Reference
solution requires periodic replacement; pretreatment
of sample usually needed; susceptible to membrane
fouling.
c. Expertise
James Ross, Orion Research
d. Cost
$300-$500
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - Systems development for
monitoring, including pretreatment of the sample.
b. R & D needs and gaps - Development of electrodes for
other gases than those indicated above.
223
-------
Potentiometric (redox hydrogen electrode)
1. Applications
Coliform bacteria
2. State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Freedom from interferences; requires monthly calibra-
tion; does not function in high salt solutions;
response time is a function of the number of cells
and rate of division.
c. Expertise/Sources
Judd Wilkins, NASA; C. C. Liu, University of Pittsburgh.
d. Cost
$10-$200.
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - Development of electrode
configurations and flow systems.
b. R & D needs and gaps - Fundamental studies of electro-
chemical mechanisms and biochemical reactions.
4. Long-Term Development Possiblities
a. R & D ongoing or programmed - Investigate possibilities
of using similar systems to monitor other micro-
organisms.
224
-------
Amperometric
1. Applications
Heavy metals (copper, cadmium, lead, and zinc).
2. State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Sensitivity is sufficient for most monitoring applic-
ations. Not automated; sample preparation required;
good capability for on-site automation; stability
not high; application to all water types with pre-
treatment; accuracy and precision not high but could
be improved.
c. Expertise/Sources
Walter Blaedel, University of Wisconsin; Judd Flato,
PAR; Tom Hoover, EPA, Athens; Ray Clem, Lawrence
Berkeley Laboratory; Al Zirino, Naval Undersea
Research Center; W. Davison and M. Whitfield,
reference: Journal Electroanalytical Chemistry,
75., 763 (1977).
d. Cost
Inexpensive (potentially disposable) .
3. Short-Term Development Possiblities
a. R & D ongoing or programmed - Seek new applications
for available systems and development of new electrode
designs.
b. R & D needs and gaps - Engineering research and
development of systems packages for on-site monitoring
225
-------
Electrochemical
(amperometric, galvanic and chronocoulometric)
1. Applications
Dissolved oxygen
V State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Membrane instability and fouling; requires sample
around sensor to be continuously renewed, except
for pulsed operations; sensitivity is sufficient;
accuracy and precision depend on calibration pro-
cedure. Interferences - chlorine and hydrogen
sulfide and possibly volatile organic sulfides.
Required calibration frequency unsuitable for long-
term monitoring. Response time is too slow for
profiling. Useable in all water types (except for
specific interferences).
c. Expertise/Sources
K. H. Mancy, University of Michigan; J. A. Llewellyn,
University of South Florida; G. Ward, NOAA; C. C. Liu,
University of Pittsburgh; Yellow Springs Instrument
Company; Beckman Instruments; Orbisphere; Martek.
d. Cost
$100-$200 ($1000 for high-pressure sensor)
3. Short Term Development Possibilities
a. R & D ongoing or programmed - Considerable R & D in
progress.
b. R & D needs and gaps - Studies into membrane properties*
improvement of stability.
226
-------
Electrochemical
1. Applications
Conductivity
2. State of the Art
a. Status
Operational
b. Capabilities/Limitations
Sufficient accuracy and sensitivity; calibration
frequency about two - three days in salt water?
useable in all types of water; response decreases
with decreasing temperature; may require filtration
of sample; nonspecific response.
c. Expertise/Sources
James Sprenke, NCAA
d. Cost
Less than $500
3. Short-Term Development Possibilities
a. R & D ongoing or programmed - Improvement of precision.
b. R & D needs and gaps - Miniaturization and improvement
in stability.
4. Long-Term Development Possibilities
a. R & D ongoing or programmed - Research into the
improvement of the correlation between conductivity
and salinity.
227
-------
ELECTROPHYSICAL SENSOR TECHNOLOGY: Working Panel Report
Members
James H. Ficken, USGS, Chairman
Mary S. Hunt, University of California
Frank H. Middleton, University of Rhode Island
Walter J. Blaedel, University of Wisconsin
Stuart Garner, Hydrolab Corporation
John McFall, NASA
Bob Farland, NOAA
Thomas B. Harris, U. S. Navy
J. Anthony Llewellyn, University of South Florida
Summary
Some of the most powerful tools discussed by this panel include
acoustical and radioactive methods. The acoustical methods can
provide information on the thickness of oil and hazardous
materials (e.g., densities less than water), and sedimentation
concentration. Acoustical methods (passive) can provide infor-
mation on activities of marine life, such as snapper shrimp
and schools of croakers. Radioactive techniques include the use
of neutron activation and energy dispersive x-rays for measure-
ment of heavy metals. Electromechanical methods included the
use of rotors and floats for measuring water velocity and tides.
It was emphasized that macroscopic measurements are often over-
looked, but are essential to understand the large picture such
as total stream flow and oil coverage. For example, rather
than measuring flow velocity in a stream or area, acoustical
devices could be used to measure the average velocity over a
much longer path distance.
It was noted that the advent of the microcomputer devices
incorporated in the sensor package will considerably reduce
the need for expertise in the operation of these systems.
This coupled with new technology such as solid state devices
(thermistor circuits) for detecting thin films of oil and
other pollutants shows great promise. Additionally, packaging
and the auxiliary devices required to keep sensors in operation
should not be neglected. It must also be remembered that most
sensor systems will operate much more satisfactorily if they
have anti-fouling and cleaning devices attached to them.
Sensor Technology
Five types of electrophysical sensors were discussed in detail
by the Working Panel. These were:
• acoustical
• electrical
228
-------
• electromechanical
• differential pressure
• radioactivity
jk summary of these discussions is provided in the following
outlines.
229
-------
Acoustical
1. Definition of Technology
Involves measurements of sound transport and character-
istics which are peculiar to parameters of interest.
2. Applications
Parameters such as depth, profiling, velocity, direction,
density, oil concentration are important applications.
3. Advantages
These methods cover a wide range of applicability.
4. Disadvantages
Probably the main disadvantage at this time is the cost
of these systems, but indications are that significant
progress is being made in this area of cost reduction.
5. Near-Term R & D Recommendations
The need for increased application to sediment concen-
tration by acoustical scatter, using such things as
cavity resonance, frequency slev/ing.
6. Long-Term R & D Recommendations
Doppler shift velocity profiling in swift water environ-
ments and the use of acoustics in measuring water
velocities of low values (0-10 cm/sec).
7. Remarks and Conclusions
Many sophisticated acoustical devices developed by
military and space research can be selectively applied
to water quality sensing. Even considering the complexity
and cost, the advent of microprocessors improves the
capability for acoustical measurements.
230
-------
Electrical
1. Definition of Technology
Involves the measurement of resistance/ resistivity,
and voltage as the result of physical change. This can be
associated with two basic applications—the measurement of
temperature and the measurement of specific conductance.
In the measurement of temperature various elements, in-
cluding RTD's and/or thermistors and thermocouples are
used. Conductance systems utilize inductive and resistive
techniques.
2. Applications
Temperature and specific electrical conductance, which is
related to salinity and dissolved solids.
3. Advantages
Temperature elements are developed and commercially
available for most applications. The conductivity measure-
ments are fairly well defined for salt water and have
recently been applied to fresh water.
4. Disadvantages
None identified for temperature. Conductance sensor
configurations are not as small as desired for some
requirements.
5. Near-Term R & D Recommendations
Ho specific recommendations for temperature sensors other
than those needed for special applications are necessary.
For conductivity, more improvement is needed in reducing
the affects of fouling and in reducing the size of
inductive cells.
6. Long-Term R & D Recommendations
None identified
7. Remarks & Conclusions
The methods for measuring temperature seem to be as far
advanced as necessary for most monitoring applications.
Much work has been done in the specific conductance
measurements and we have considerable expertise represented
throughout the country. The relationship of conductance
to salinity in salt water is well defined. It is not
well defined in fresh water. The measurement can be used
as a secondary input to correct other measurements such
as correcting the concentration of a dissolved oxygen
probe.
231
-------
Electromechanical
1. Definition of Technology
Depth measurements include physical sounding, floats,
gages, and staffs. Flow measurements include drag
devices, tracers, rotors, and vanes. Density measure-
ments include vibrating tubes.
2. Applications
Depth and flow.
3. Advantages
Most of these parameters are measured by devices that are
fairly inexpensive, easily operated and in most cases
durable. However, in long submergence in water, quite a
few are subject to fouling.
4. Disadvantages
Disadvantages associated with mechanical devices are
that they usually require some type of fixed installation
and that generation of an electrical signal usually requires
other transducers or devices.
5. Near-Term R & D Recommendations
Could be replaced by more advanced methods, e.g. acous-
tics.
6. Long-Term R & D Recommendations
Some of the mechanical methods are still quite good if more
work is done to repackage and provide protection from
fouling.
7. Remarks & Conclusions
These methods are used quite frequently. Thousands of
stations are located throughout the country and will con-
tinue to be used in the future. Some of the simple
mechanical methods cannot be discounted.
232
-------
Differential Pressure
1. Definition of Technology
New generation pressure transducers of the quartz crystal
type exhibit changes in resonant frequency with variation
in pressure. Direct digital outputs are possible.
2. Applications
Density and determination of size and concentration of
suspended sediment in a settling water column.
3. Advantages
Relatively non-fouling methods for measuring density, water
stage and wave height and relating these to total suspended
solids. Can replace expensive float structures. Many
devices available.
4. Disadvantages
Relative high cost for accurate stable sensor.
5. Near-Term R & D Recommendations
Efforts should be made to reduce the cost of accurate,
stable sensors.
6. Long-Term R & D Recommendations
The use of differential pressure methods in determinating
sediment concentration and size distribution.
7. Remarks & Conclusions
The use of differential pressure can be applied to the
measurement of density which could be related to salinity
and suspended solids. The method can also be applied to
the measurement of velocity and flow.
233
-------
Radioactivity
1. Definition of Technology
Includes ionization detectors, scintillation counters,
and semiconductor devices.
2. Applications
Measurement of radioactive materials through the identi-
fication of radionuclide and type of radiation. There
may be some application for density and suspended sediment
measurement.
3. Advantages
High precision and accuracy for some applications.
Fairly low cost, reliable.
4. Disadvantages
For some applications, may require preconcentration.
5. Near-Term R & D Recommendations
This working panel felt the need for more inputs from
experts in the field. Much of the development has
probably already been completed. The application of
this technology to suspended sediment measurement should
be explored.
6. Long-Term R & D Recommendations
None
7. Remarks & Conclusions
These methods show great promise and are being used in
field measurements of the above parameters.
234
-------
AUTOMATED WET CHEMICAL SENSOR TECHNOLOGY: Working Panel Report
Members
BarIan L. McKim, COE, Chairman
Ken N. Birch, Canada Centre for Inland Waters
Ted Major, Magnavox
Andrew J. Green, COE
Ray W. Lovelady, NASA
Anthony F. Mentink, EPA
Khalil H. Mancy, University of Michigan
K. Nishioka, NASA
Summary
A number of sensor technologies that could be used individually
or in combination with automated pretreatment systesm were
considered. These include ultraviolet and colorimetric detec-
tion techniques, membrane electrodes and specific ion electrodes.
These would present signal processing problems for complete
automation, but most are resolvable with microprocessors,
A certain amount of pretreatment will always be necessary.
Though a high capital cost can be expected initially, a very
large number of water quality parameters lend themselves to
detection and quantification by this technology. Automatic
cleaning would be necessary, often, depending upon types of
waters. It is expected that automatic measurements can be
extended from the present capability of a few hours to a few
hundred hours by application of available technology. Measure-
ments of total dissolved phosphorus, nitrogen compounds, and
dissolved total organic carbon should be pursued.
Automated wet chemistry is the automation of on-line sample
pretreatment for detection of parameters by sensor systems.
It involves elimination of substances that would interfere
with detection and requires minimum attention and maintenance
so that it is applicable for field use (Figure 1). The sensor
and package should operate unattended in, on, or immediately
adjacent to the water body being monitored, and at the field
location where the measurement is desired, as opposed to a
laboratory facility, either fixed or mobile.
235
-------
I (bv
Source
(bulk)
.-*_.
Intake
System
r
Sample
P re treatment
T
Detection
Analyzer
Output
1. Biological
2. Chemical
3. Electrical
4. Mechanical
5. Thermal
6. Radiation
1. Electrochemical
2. Magnetic
3. Optical
4. Wave length
Figure 1. Automated Wet Chemistry Technology
236
-------
A summary of sensor technology follows:
Applications (Water Type)--
a. reservoir g. portable water supplies
b. rivers h. rainfall
c. lakes i. oceans
d. estuaries j. reclaimed waters
e. groundwater k. dredged material waters
f. wastewater
Advantages—
• Ability to measure:
a. Expands the capability of a specific sensor type
and minimizes effects of interferences and
environmental factors.
b. Allows concurrent multiple parameter analysis.
c. Provides ability to measure large number of
samples.
• Precision:
a. Permits reduction of operator-induced error.
b. Amenable to automated calibration and operation
under computer control in or near real time.
• Frequency:
a. Provides a high production rate.
• Cost:
a. Cost effective.
b. Provides flexibility of application.
c. Requires minimum maintenance.
Disadvantages—
• Capital cost can be high.
• Operator skill requirements are high.
• System packages can be highly complex.
Near-Term R & D Recommendations—
• Development of miniaturized modular packaging for in
situ and on site field installation.
• Assess performance characteristics and cost effectiveness
• Design modules amenable to microprocessor control.
• Develop combination of wet chemistry modules tailored to
meet needs associated with specific classes of water.
• Develop new wet chemistry methods for various water
quality parameter in different classes of water.
• Expand/ demonstrate, and improvement of available
methodologies.
Long-Term R & D Recommendation—
• Continued emphass on all the items identified as short
range R & D goals.
• Increased emphasis on:
a. Miniaturization
b. Remote operation capability
237
-------
c. Product reliability
Continued search for new technology within related
field of research.
Remarks and Recommendations—
Automated wet chemical procedures lend themselves to the
detection and quantifying of a large number of water quality
parameters. The main challenge is the adaptation of these meas-
urement techniques to field applications. Without sample pre-
treatment, a large number of sensor systems cannot be applied
to water quality surveillance and monitoring. Another important
aspect of unattended field installation of wet chemistry method-
ology for water quality analysis is automatic cleaning. In
system design and packaging, procedures for determining the
type of cleaning (chemical, electrical, mechanical) required for
the various water classes need to address such variables as
frequency, time, chemical constituents and specified conditions
as functions of the plumbing and detection components.
The Working Panel believes that the terminology for iden-
tifying water quality parameters of concern to the various
agencies should conform to conventional, analytical chemistry.
Such parameters as available nitrogen and dissolved phosphorus,
if used, should be defined. Our suggestion is to use ortho and
polyphosphate and total nitrogen, TN soluble nitrogen (free NH4,
N03, NO2) and unsoluble nitrogen (particulate) to define what
is meant by phosphorus and nitrogen.
It is recommended that interdisciplinary research and
investigations with the biological and medical disciplines be
initiated with the objective of developing new sensing tech-
niques .
Some of the sensor types, particularly the ultraviolet,
colorimetric and electrochemical types could be developed for
completely submersible in situ systems for deployment near
shore or on buoys. Much of the basic technology for designing
and controlling underwater electromechanical apparatus has been
developed in oceanography and processes control.
Individual Sensor Applications/Characteristics—
Automated Wet Chemistry Technology Work Panel discussions
of specific sensors were broken dov/n to address four types:
• colorimetric
• electrochemical
• on-line ultraviolet absorption
• energy dispersive x-ray
Sensor Technology
The Panel comments on individual sensors are summarized in
the following outlines.
238
-------
Colorometric
1. Applications
Total N, NO3/ NH4, Kjeldahl N, Ortho P, Cl, COD,
Total P and others.
2. State of the Art
a. Status
Developmental
b. Capabilities/Limitations
Accuracy and reliability-variable, temper-mental;
Operational longevity-2-hr life at present, 1 week
is needed for unattended life;
Suspended solids will need to be controlled in water;
Precision-not accurate;
Water type applications (see water type classification
in summary, item 2)
high probability of measurement - types a, c, e,
9/ h, j.
medium probability of measurement - types d, i.
low probability of measurement - types b, f, k.
Serviceability-requires highly qualified individual;
Environmental conditions-enclosed (no freezing or
sunlight).
c. Expertise/Sources
Federal, commercial, university, and clinical.
d. Cost
For sensors in routine operation:
Developmental - $250K
Capital - $15K
Operations and maintenance - $30K (annually)
3. Short-Term (5-year) Development Possibilities
a. R & D ongoing or programmed - not presently known.
b. R & D needs and gaps - automation, packaging for
field application.
4. Long-Term Development Possibilities
a. R & D ongoing or programmed - not needed for engineering
needs.
b. R & D needs and gaps - not needed for engineering needs.
239
-------
Electrochemi ca1
1. Applications
NH^, N03-N, residual chlorine, ozone; trace metals,
alkalies, alkali earth metals, florides, cyanide and other
electrochemically detectable species where some pretreat-
ment is needed.
2. State of the Art
a. Status
Research, Developmental, Operational
Sensors are available for the measurement of parameters
shown in each of the measurement categories.
b. Capabilities/Limitations:
Accuracy-precision varies with the sensor;
Size/weight-portable;
Water type applications (all in summary list, item 2)
Serviceability-varies with the type of sensor on a
hourly, daily, weekly basis;
Environmental conditions-can withstand many types of
environmental stresses, does not require large
quantities of water.
c. Expertise/Sources
Federal, university, commercial, and clinical
d. Cost
For sensors in routine operation: $500 - $2000 per
package.
3. Short-Term (5-year) Development Possibilities
a. R & D ongoing or programmed - some are demonstrated
under field conditions.
b. R & D needs and gaps - for process control in treat-
ment plant in situ conditions.
4. Long-Term Development Possibilities
a. R & D ongoing or programmed - Technology needs to be
accomplished for unattended, reliable, low cost
packaging for field and in situ use.
b. R & D needs and gaps - None identified.
240
-------
On-line Ultraviolet Absorption
1. Applications
Unsaturated resonating organic compounds, phenolic groups,
pesticides, etc.
2. State of the Art
a. Status
Developmental - Limited to effluents where normal
consituents are known.
b. Capabilities/Limitations
Reliability-good with further development;
Size/weight-small
Water type applications-clear water;
Serviceability - unknown (expected to be good);
Environmental conditions - no housing can be put in
the field.
c. Expertise/Sources
Federal, commercial, university and clinical.
d. Cost
For sensors in routine operation:
(1) $5K per package
(2) Operation and maintenance - unknown
3. Short-Term (5-year) Development Possibilities
a. R & D ongoing or programmed - Investigation of field
applications to difficult types of water. Non-speci-
ficity - can only detect organics which absorb in
that specific light wavelength.
b. R & D needs and gaps - None identified
4. Long-Term Development Possibilities
a. R & D ongoing or programmed - Investigate field
applications for difference types of waters.
b. R & D needs and gaps - in situ measurement.
241
-------
Energy Dispersive X-ray
1. Applications
All heavy metals with the exception of the metal used for
the X-ray window.
2. State cf the Art
a. Status
Developmental routine: Detection ability - ppb range
at low end and at upper end. Filter paper saturation
problems exist because of matrix interference and
very high metalic ion concentrations.
b. Capabilities/Limitations
Reliability-unknown but expected to be within design
expectations over a target period of 30-day unattended
operation;
Size/weight-13 cu ft, 420 Ibs.
Water type applications-all except dredged sediment;
Serviceability-unknown;
Environmental conditions - no freezing temperatures,
etc.
c. Expertise/Sources
Industry, Federal, university, and clinical
A. F. Mentink, EPA, EMSL-Cincinnati
d. Cost
Development and demonstration - $500K. For sensors in
routine operation - $5OK for a production package.
Operations and maintenance - $20K annually/station
3. Short-Term (5-year) Development Possibilities
a. F & D ongoing or programmed - Development of addi-
tional filter papers and the associated resins.
b. R & D needs and gaps - Greater portability.
4. Long-Term Development Possibilities
a. R & D ongoing or programmed - Testing to be completed
under 3-a.
b. R & D needs and gaps - Completely automated remote
installation.
242
-------
NEEDS AND TECHNOLOGY INTEGRATION: Working Panel Report
Members
Barbara Pijanowski, NOAA - Chairman
John D. Koutsandreas, EPA - Co-Chairman
Donald t. Wruble, EPA
Richard A. Durst, NBS
Karlan L. McKim, COE
James H. Ficken, USGS
Vincent G. DiPasqua, USCG
Cliff Risley, EPA
Ted Major, Magnavox
Enrico Mercanti, NASA
Nelson L. Milder, NASA
Charles R. Eastwood, NASA
Summary
It was the unanimous consensus of the agency representatives
attending the workshop that there is indeed a great need for
development of automated in situ measurement systems^ to
satisfy requirements for aquisition of water quality information.
It was noted that because such sensors do not exist, our ability
to collect the nationwide field information necessary for
adequate resource management and protection is seriously
limited. Also noted was the fact that because these sensors are
not available, development of more advanced measurement systems
such as satellite and aircraft remote sensing is also limited as
adequate means for obtaining supportive ground truth information
do not exist.
^•Automated in situ measurement systems for the purpose of
this workshop were defined to be sensors or sensor systems
that operate in contact or near contact with the water they are
measuring; they must operate unattended in, on, or immediately
adjacent to the water body being monitored, and at the field
location where the measurement is desired versus a laboratory
facility. A sensor system may consist of several elements which
are used to pre-condition the sample as in the case of automated
wet chemistry, with the overall system producing a signal that
is proportional to the parameter being measured.
243
-------
It was recognized that natural waters, composed of ground-
water, streams/ rivers, lakes, estuaries and the oceans, are
keys to survival because of both the benefits they provide
and the disaster they may carry. Water as drinking water and
for agricultural purposes is vital to existence. It plays a
key role in the removal of wastes, the production of energy
and the establishment of climate. It is also of significant
importance for its ability to distribute pollutants and
hazardous materials throughout the environment, contaminating
drinking water, food supplies and recreational facilities. It
is important therefore, that we both understand and manage our
water resources effectively, not only from the standpoint of
conservation, but also from the perspective of publdc health
and safety.
The aquisition of adequate reliable water data is the first
step in any water management program, whether the objectives
are conservation oriented or scientific in nature, i.e., for
the allocation of water resources or for the understanding of
the mechanisms and pathways of pollutants.
Data must be collected frequently, sometimes continuously, in
remote locations, under adverse environments and often over
very large areas. Practice at the present time is to send
people into the field to collect samples for analysis or to
perform the frequent maintenance required for the few automatic
instruments that are available. Such practice provides neither
the real time information that is required in many cases, nor
the adequate quantity and quality of data required for proper
management.
It was acknowledged by workshop participants that present
capabilities for data transmission, storage, retrieval and
manipulation far exceed capabilities to make accurate, reliable
real-time measurement of necessary water quality variables.
It was concluded that automatic sensor systems would have wide
application for water quality monitoring by many agencies in
fresh water streams and lakes, in groundwater and in the marine
waters of the ocean and coastal zones. For this reason, it
was strongly recommended that agencies cooperate as much as
possible to develop these automated in situ systems for their
mutual benefit through joint interagency research and develop-
ment efforts.
Although agencies require water quality measurements for differ-
ent purposes and in different environments, it was noted that
there are many areas where requirements overlap. Areas of
mutual interest regarding high priority measurements were
identified and grouped into three categories according to degree
244
-------
of commonality. These are detailed in the Measurement Needs
Working Panel report. Group 1 parameters were found to be of
high priority and great interest to most agencies with water
measurement programs. Group 2 parameters were of high priority
interest to many agencies, and Group 3 parameters were of high
priority but specialized interest to a few agencies. For example,
the monitoring of toxic substances is extremely important, how-
ever, EPA and the U. S. Coast Guard are the agencies with
primary concern for such measurements. On the other hand,
suspended particulate measurements are of high priority concern
to all agencies since such information is necessary for the
quantification of sediment transport (U. S. Department of
Agriculture, U. S. Geological Survey, U. S. Army Corps of
Engineers) as well as the study of pollution transport mechanisms
(EPA, NOAA, Department of Energy) and ground truth support for
remotely sensed phenomena (NASA).
All agencies represented at the workshop, with the exception
of NASA, had active water measurement programs and could
therefore contribute to the discussion of measurement require-
ments. NASA was recognized, however, to have a substantial
interest in development of water measurement technology since
NASA is anxious to promote application to national problems
through the considerable technical expertise it has acquired
through the space program. Furthermore, NASA has a budget
to facilitate such technology transfer.
Findings and Recommendations
1. Although the Measurement Needs Working Panel was able
to identify high priority measurements required by most agencies,
it was felt that definition of requirements specific enough
for design and development of measurement systems related to
various application areas could not be generalized in terms
of overall agency requirements by the relatively small number
or panel members present. A preliminary summary of agency
needs is included in Measurement Needs Panel report. The panel
prepared a series of questions to be answered regarding specific
requirements.
It is Recommended that:
(a) "Requirements Survey" be formally conducted within
each agency to reflect overall agency needs and that
results of this national inventory be published by a
workshop follow-up committee.
2. It was recognized that this workshop was the first
meeting of an interagency group for discussion of common needs
and requirements for automated water quality measurement tech-
nology. General agreement existed on the value of continuing
such dialogue in order to establish national priorities,
245
-------
accelerate technological development and coordinate agency
efforts to minimize duplication. Although little duplication
could be identified except in a few specialized areas of
hydrology, the need for better interagency coordination was
recognized. It was felt that improved communication would lead
to better coordination and more efficient application of the
always limited budgetary resources.
It is Recommended that:
(a) more sensor development be carried out through inter-
agency cooperative efforts utilizing the considerable
expertise that already exists in the Federal system.
(b) a similar workshop be held approximately six months
after the report of this workshop is issued and that
coverage be expanded to include areas that were
inadequately covered (biology and radioactivity) at
this workshop.
(c) a steering committee be formed to follow through on
the action items of this workshop and to plan future
coordination.
(d) efforts be made to interest one or more of the pro-
fessional societies (e.g. IEEE, ISA, etc.) in in-
cluding sessions devoted to automated in situ sensor
development for water quality measurement in their
future conference plans.
3. It was noted that the research and development capabil-
ities of private industry are considerable and if a profitable
market could be identified, private industry would be willing t<
use its own resources to develop many of the devices needed for
automated in situ measurement.
It is Recommended that:
(a) an interagency briefing on the results of this work-
shop be delivered to interested industry representa-
tives after the workshop report has been completed.
4. The technology exists at the present time to develop
many of the automated in situ sensors with real time data
collection capabilities required for Federal agency programs.
With minimum effort, short term development could yield auto-
mated sensors to meet some of the interagency requirements
that have been identified. Additionally, interim stages of
development will frequently result in a number of automated
devices for immediate use in the laboratory. Such intermediate
development stages will provide useful laboratory devices that
can reduce manpower requirements as well as the human variabil-
ity inherent in manual sampling and analytical methods. Since
little funding has been directed toward such development to
data, inadequate programming to focus ..on this area has resulted.
246
-------
It is Recommended that:
(a) agencies assume lead roles in the development of
indicated sensors as outlined in Table 1 on page 10,
pursuing additional funding and support in-house and
through cooperative arrangements with other interested
agencies to support research and development efforts.
(b) the technological expertise of NASA be applied to
address the need for imporved sensors through the
technology transfer program, as it is apparent that
NASA could make substantial contributions in this
area.
5. One of the areas that has been specifically identified
as having a high potential for short term development success is
automated wet chemistry. For many of the "Group 2," parameters
cited by the Measurement Needs Panel, such techniques are either
already in use or have demonstrated feasibility for near term
implementation. It is recognized that the simplest of these
systems are presently limited to periods of about one week for
unattended operation. It was felt, however, that in a fairly
short time, minimum efforts at improvement and imaginative
packaging could result in being able to extend time for field
use as well as extend this technology to more complex operations.
It is Recommended that:
(a) a lead agency role be accepted by EPA to promote and
carry out additional efforts toward development of
small automated wet chemical systems for the measure-
ment of total dissolved phosphorus, nitrogen compounds,
and dissolved and total organic carbon.
6. Additional technologies that have been specifically
identified as having high potential for successful development
in the next few years are acoustic echo sounding and optical
techniques as applied to suspended particulate measurements, and
atomic absorption spectrophotometry; neutron activation, and
energy dispersive X-ray spectroscopy for metal detection.
Further evaluation is required however, before assessment of the
most effective technique for in situ sensing can be made.
It is Recommended that:
(a) on-going efforts involving the optical measurement of
suspended particulates be accelerated and expanded to
include existing expertise, particularly that within
NASA,
(b) NOAA serves as the lead agency to further evaluate
existing metal ion measurement systems and accelerate
development of in situ sensors,
(c) additional funding to support these areas be secured
through elevating their priorities within
247
-------
agency programs and through interagency agreements
for sharing research and development costs.
7. All sensor technology panels, regardless of the tech-
nology specialty being considered, stressed the need for improved
sensor cleaning and anti-fouling techniques. The performance
of any sensor that must operate unattended in a natural water
environment is greatly degraded by the presence of sand, sediment
and biological organisms.
It is Recommended that:
(a) additional efforts be directed toward development of
cleaning techniques and prevention of fouling for
unattended automatic sensors in natural water environ-
ments.
8. Although the workshop focused on development of systems
applicable to point type measurements, it was recognized that
there is also a need for mesoscale measurement schemes to obtain
synoptic data over large areas and to acquire large scale infor-
mation such as total flow in streams and oil spill coverage.
It is Recommended that:
(a) consideration be given to design or systems to obtain
large scale measurements.
248
-------
WATER QUALITY SENSOR WORKSHOP
February 14-16, 1978
Las Vegas , Nevada
ATTENDEES
Bates, Dr. Roger G.
Professor of Chemistry
Department of Chemistry
Gainesville, FL 32611
(904) 392-0561
Birch, Kenneth N.
Head, Instrument R&D Unit
Department of Fisheries
and Environment
P.O. Box 5050
Burlington, Ontario
CANADA 17R4A6
(416) 637-0128
Blaedel, Dr. Walter J.
Professor of Chemistry
Department of Chemistry
Madison, WI 52706
(608) 262-3033
Bockes, Olin D.
Remote Sensing Specialist
Soil Conservation Service
P.O. Box 2890
Washington, DC 20013
(202) 447-6267
FTS 447-6267
Cameron, Ben
17302 Daimler
Irving, CA 92713
(714) 540-4435
University of Florida
Canada Centre for Inland
Waters
University of Wisconsin
U.S. Department of Agriculture
Martek Instruments, Inc.
249
-------
Di Pasqua, Lt. Cmdr. Vincent G.
Surveillance & Monitoring
Branch Chief
Environmental Protection Divison
Nassiff Bldg.
Washington, DC 20590
(202) 755-7938
FTS 755-7938
U.S. Coast Guard
Dowd, Dr. Richard M.
Science Policy Advisor
Washington, DC 20460
(202) 755-0263
FTS 755-0263
Durst, Dr. Richard A.
Research Chemist
Chemistry Bldg., Rm. 1-221
Washington, DC 20234
(301) 921-2881
FTS 921-2881
Eastwood, Charles R.
Manager, Environmental
Program
Washington, DC 20546
(202) 755-3140
FTS 755-3140
Eiker, Earl E.
Hydraulic Engineer
Office of the Chief
of Engineers
HQDA (DAEN-CWE-HY)
Washington, DC 20314
(202) 693-7330
FTS 693-7330
fitter, Paul C.
Associate Engineer
1335 Rockville Pike
Rockville, MD 20852
(301) 424-1310
Farland, Bob
8508 Woodside Ctr.
Lanham, MD 20801
(301) 443-8444
U.S. Environmental Protection
Agency
National Bureau of Standards
National Aeronautics and Space
Administration Headquarters
U.S. Army Corps of Engineers
MAR, Incorporated
National Oceanic and
Atmospheric Administration
250
-------
Ficken, James H. U.S. Geological Survey
Hydrologist
GCHC-NSTL-Bldg. 2101
NSTL Station, MS 39529
(601) 688-2161
FTS 494-2161
Garner, Stuart Hydrolab Corporation
Development Engineer
P.O. Box 9406
Austin, TX 78766
(512) 837-2050
Green, Andrew J. U.S. Army Corps of Engineers
Environmental Engineering
Division
Waterways Experiment Station
P.O. Box 631
Vicksburg, MS 39180
(601) 636-3111, ext. 2408
Harris, Thomas B, Naval Surface Weapons Center
Development Coordinator
Code 201
Silver Spring, MD 20910
(301) 394-2465
Hunt, Dr. Mary S. Lawrence Berkeley Laboratory
Chemist
University of California
Berkeley, CA 94720
(415) 843-2740, ext. 6297
FTS 451-6297
James, C. E. "Gene" U.S. Environmental Protection
Sp. Asst. Office Monitoring Agency
401 M Street, S.W.
Washinton, DC 20460
(202) 426-4452
FTS 426-4452
Kirchoff, Dr. William H. National Bureau of Standards
Acting Chief
Office of Air & Water Measurement
Washington, DC 20234
(301) 921-3775
FTS 921-3775
251
-------
Koutsandreas, John D.
Sr. Advisor Advanced
Monitoring
EPA ORD/OMTS
401 M Street, S.W.
Washington, DC 20460
FTS 426-4477
Lambou, Victor W.
Chief, Water & Land
Quality Branch
P.O. Box 15027
Las Vegas, NV 89114
(702) 736-2969 ext. 391
FTS 595-2969 ext. 391
Liu, Dr. Chung-Chiun
Professor of Chemical
Engineering
1249 Benedum Hall
Pittsburgh, PA 15261
(412) 624-5285
Llewellyn, Dr. J Anthony
Professor of Engineering
College of Engineering
Tampa, PL 33622
(813) 974-2581, ext. 246
Love lady, Ray W.
Research Engineer
Hampton, VA 23665
(804) 827-3581
FTS 928-3581
Lowry, H. Michael
Physical Scientist
P.O. Box 15027
Las Veags, NV 89114
(702) 736-2969, ext. 391
FTS 595-2969, ext. 391
Major, Ted
Manager, Environmental
Monitoring Systems
1313 Production Rd
Fort Wayne, In 46808
(219) 482-4411, ext. 5349
U.S. Environmental Protection
Agency
U.S. Environmental Protection
Agency
University of Pittsburgh
University of South Florida
NASA-Langley Research Center
U.S. Environmental Protection
Agency
Magnavox
252
-------
Mancy, Dr. Khalil H.
Professor of Environmental
Chemistry
School of Public Health
2530 SPH-I
Ann Arbor, MI 48109
(313) 763-4296
McFall, John
Aerospace Engineer - Regional
Applications Transfer Officer
MS325
Hampton, VA 23665
(804) 827-2486
FTS 928-2486
University of Michigan
NASA-Langley Research Center
McKim, Dr. Harlan L.
Program Manager
Wastewater Management Program
Cold Regions Research &
Engineering Laboratory
P.O. Box 282
Hanover, NH 03755
(603) 643-3700, ext. 344
FTS 862-9600
Mentink, Anthony F.
Chief, Instrumentation
EMSL-Cincinnati
26 W. St. Clair Avenue
Cincinnati, OH 45268
FTS 684-7324
Mercanti, Enrico
Project Manager
12415 Shelter Lane
Bowie, MD
FTS 982-2697
Michelena, Dr. Eduardo D.
Physical Scientist
Data Buoy Office
NSTL Station, MS
(601) 688-2806, ext. 2806
FTS 494-2806
Middelburg, Robert F.
Hydrolegist
Quality of Water Branch
National Center, MS 412
Reston, VA 22092
(703) 860-6834
FTS 928-6834
U.S. Army Corps of Engineers
U.S. Environmental Protection
Agency
National Aeronatics and
Space Administration
National Oceanic and
Atmosperic Administration
U.S. Geological Survey
253
-------
Middleton, Dr. F. H.
Professor of Ocean
Engineering
Kingston, RI 02882
(401) 792-6244
Milder, Nelson L.
Manager, Civil Systems
Applications
Code ET-4
600 Maryland Avenue, S.W.
Washington/ D.C.
Murray, Tom
Biologist
401 M Street, S.W.
Washington, DC 20460
(202) 426-7790
University of Rhode Island
National Aeronautics and
Space Administration
U.S. Environmental Protection
Agency
3n, Dr. Richard W.
Director, Microwave
Systems Laboratory
Remote Sensing Centei
326 Teague Bldg.
College Station,
(713) 845-5422
Texas A&M University
TX 77843
Nishioka, K.
Engineering Operations
Management
Moffett Field, CA 94035
(415) 965-5897
FTS 448-5897
Paulson, Richard W.
Instrumentation Coordinator
Water Resources Division
MS 460 National Center
Reston, VA 22090
(703) 860-6014
FTS 928-6014
NASA-Ames Research Center
U.S. Geological Survey
Pijanowski, Barbara
Ocean Engineer
Office of Ocean
Rockville, MD
(301) 443-8444
FTS 443-8444
Engineering
20852
National Oceanic and
Atmospheric
Administration
254
-------
Risley, Clifford
Research & Development
Representative
U.S. EPA, Region V
230 S. Dearborn
Chicago, IL 60604
(312) 353-2200
FTS 353-2200
Salzman, Jack A.
Head
Water Quality Project Unit
Lewis Research Center
21000 Brookpark Road
Cleveland, OH 44135
(216) 433-4000, ext. 6181
FTS 294-6181
Schiebe, Frank R.
Research Hydraulic Engineer
USDA Sedimentation Laboratory
P.O. Box 1157
Oxford, MS 38655
(601) 234-4121
Sprince, R
NASA Hdqtr
Code EC
Washington, DC 20546
(202) 755-3591
FTS 755-3591
Strauss, William A.
Consultant
Mason, NH 03048
(603) 818-2500/1669
Ward, Gary
Chemist
Department of Commerce
NOAA/OMT - C651
Rockville, MD 20852
(202) 426-9090
U.S. Environmental Protection
Agency
National Aeronautics and
Space Administration
U.S. Department of
Agriculture
National Aeronautics and
Space Administration
PR Mallory Company
National Oceanic and
Atmospheric Administration
255
-------
Whitehurst, Charles A.
Associate Dean
Engineering for Graduate Students
LA State University
College of Engineering
6436 LaSalee Avenue
Baton Rouge, LA 70806
(504) 388-5309
FTS 688-5309
Wilkins, Judd R.
Microbiologist
Langley Research Center
Hampton, VA 23665
(804) 827-3458
FTS 928-3458
Wruble, Donald T.
Director
Monitoring Operations Division
EMSL-Las Vegas
P.O. Box 15027
Las Vegas, NV 89114
(702) 736-2969, ext. 342
FTS 595-2969, ext. 342
Zabarsky, Oscar P.
17302 Daimler
Irvine, CA 92713
(714) 540-4435
Louisiana State University
National Aeronautics and
Space Administration
U.S. Environmental
Protection Agency
Martek Instruments, Inc.
256
-------
WORKSHOP AGENDA
WATER QUALITY SENSOR WORKSHOP
(With particular consideration to in situ small
waterborne system deployment)
Cochairmen - John D. Koutsandreas
Donald T. Wruble
TUESDAY, February 14, 1978
0800 - 0900 Registration & Get-Acquainted Coffee
INTRODUCTORY SESSION
0900 - 0905
0905 - 0910
0910 - 0930
0930 - 0935
Call to Order
Welcome
Keynote
Introductory Remarks
0935 - 0945
0945 - 0955
AGENCY PROGRAM PRESENTATIONS
0955 - 1005
1005
1030
1055
1120
1030
1055
1120
1145
1145 - 115
115
140
205
230
250
315
340
405
140
205
230
250
315
340
405
430
Overview
EPA
NOAA
Department of Energy
Corps of Engineers
Lunch
NASA
National Bureau of Standards
Department of Defense (Navy)
Break
U.S. Coast Guard
U.S. Geological Survey
U.S Dept. of Agriculture
Participant Comments/
Observations
Mr. Donald T. Wruble
Mr. George B. Morgan
Dr. Richard M. Dowd
Dr. Richard W. Paulson
Workshop Schedule, Logistics,
Plans, Procedures Mr,
Break
Donald T. Wruble
Mr. John D. Koutsandreas
Mr. Tom Murray
Ms. Barbara Pijanowski
Dr. Mary S. Hunt
Mr. Earl Eiker
Mr. Nelson Milder
Dr. William H. Kirchoff
Mr. Thomas Harris
Lt. Cmdr. Vincent DiPasqua
Dr. Richard W. Paulson
Dr. Frank Scheibe
Open Discussion
257
-------
WEDNESDAY, February 15, 1978
SPECIAL GUESTS
0830 - 0855
0855 - 0930
Sensor Needs in EPA Regional, State
& Local Operational Monitoring
Programs Mr. Clifford Risley
Water Quality Sensor R&D in Canada-
Canada Center for Inland
Waters Dr. Kenneth Birch
UNIVERSITY PRESENTATIONS (New Technology)
0930 - 0950
0950 - 1010
1010 - 1035
1035 - 1100
1100 - 1125
1125 - 1150
1150 - 1215
University of Wisconsin
Break
University of Texas ASM
University of Michigan
Louisiana State University
University of Florida
University of Rhode Island
LUNCH
Dr. W. T. Blaedel
Dr. Richard Newton
Dr. K. H. Mancy
Dr. Charles Whitehurst
Dr. Roger G. Bates
Dr. F. H, Middleton
PANEL SESSIONS
130 - 145 Panel Assignments & Instructions
145 - 500 Panels - 1st Session
THURSDAY, February 16, 1978
0830 - 1200 Panels - 2nd Session
LUNCH
Report Preparation
BREAK
Panel Report Presentations
Report Discussion/Comments
Concluding Remarks
130 - 250
250 - 300
300 - 400
400 - 425
425 - 430
Mr. John Koutsandreas
FRIDAY, February 17, 1978 - (Panel Chairmen only)
Workshop Report Finalization
258
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO,
EPA-600/9-78-03A
2.
3. RECIPIENT'S ACCESSION-NO.
TITLE AND SUBTITLE
AUTOMATED IN SITU WATER QUALITY SENSOR WORKSHOP
February 14-16, 1978
6. PERFORMING ORGANIZATION CODE
REPORT DATE
October 1978
AUTHOR(S)
Wruble, Donald T., Barbara Pijanowski and
John D. Koutsandreas, Compilers
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Monitoring and Support Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Las Vegas, NV 89114
10. PROGRAM ELEMENT NO.
1HD620
11. CONTRACT/GRANT NO.
2. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency-Las Vegas, NV
Office of Research and Development
Environmental Monitoring and Support Laboratory
Las Vegas, NV 89114
13. TYPE OF REPORT AND PERIOD COVERED
Workshop Report
14. SPONSORING AGENCY CODE
EPA/600/07
5. SUPPLEMENTARY NOTES
Document is a compilation of prepared presentations and discussions at a U.S.
Federal agency workshop held February 14-16, 1978, in Las Vegas, Nevada.
6. ABSTRACT
A Federal agency workshop to discuss a common interagency need for development
of automated in situ water quality sensors was held in February 1978. The meeting
was organized to focus interagency attention on the lack of adequate automated in
situ devices for meeting national water quality measurement needs, and to explore
possible solutions to the problem by identifying technologies that might be applied
and initiating interagency cooperation to consolidate required research and
development efforts.
Agency programs and academic research programs are described. Working panels
addressed sensor needs and technological areas that might be applied to sensor
development, including electrochemical, automated wet chemistry, optical, and
electrophysical sensors. Recommendations for greater development emphasis,
greater coordination within lead-agency responsibilities, and technology emphasis
are presented.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Water pollution
Water quality
Sensors
Workshop, Water
In situ monitoring
In situ sensors
08H
09C
13B
17C,D,E,H
18. DISTRIBUTION STATEMENT
RELEASE TO THE PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
266
22. PRICE
A-12
EPA Form 2220-1 (9-73)
*U.S. raVERNriENT PRINTING OFFICE: 1978 - 684-043/1902-f, 2037. 9-1
------- |