ENVIRONMENTAL HEALTH
Water Supply
and Pollution Control
SERIES
Symposium on
STREAMFLOW REGULATION
FOR QUALITY CONTROL
U.S. DEPARTMENT OF HEALTH,
EDUCATION, AND WELFARE
Public Health Service
-------
SYMPOSIUM ON
STREAMFLOW REGULATION
FOR QUALITY CONTROL
Sponsored by the
Technical Services Branch
and
Basic and Applied Sciences Branch
of the
Division of Water Supply and Pollution Control
April 3-5, 1963
Chairman
John G. McLean
Chief, Water Resources Unit
Technical Services Branch, DWS&PC
Robert A. Taft Sanitary Engineering Center
Cincinnati, Ohio
U. S. DEPARTMENT OF HEALTH
EDUCATION, AND WELFARE
Public Health Service
June 1965
-------
The ENVIRONMENTAL HEALTH SERIES of reports was estab-
lished to report the results of scientific and engineering studies of
man's environment: The community, whether urban, suburban, or
rural, where he lives, works, and plays; the air, water, and earth he
uses and re-uses; and the wastes he produces and must dispose of in
a way that preserves these natural resources. This SERIES of reports
provides for professional users a central source of information on the
intramural research activities of Divisions and Centers within the
Public Health Service, and on their cooperative activities with state
and local agencies, research institutions, and industrial organizations.
The general subject area of each report is indicated by the two letters
that appear in the publication number; the indicators are
WP — Water Supply and Pollution Control
AP — Air Pollution
AH — Arctic Health
EE — Environmental Engineering
FP — Food Protection
OH — Occupational Health
RH — Radiological Health
Triplicate tear-out abstract cards are provided with reports in the
SERIES to facilitate information retrieval. Space is provided on the
cards for the user's accession number and additional key words.
Reports in the SERIES will be distributed to requesters, as sup-
plies permit. Requests should be directed to the Division identified
on the title page or to the Publications Office, Robert A. Taft Sanitary
Engineering Center, Cincinnati, Ohio 45226.
Public Health Service Publication No. 999-WP-30
ii
-------
PREFACE
The 1961 amendment to the Federal Water Pollution Control
Act authorizing the inclusion of storage for water quality control as
a feature in multi-purpose reservoir evaluation greatly increased
the need for techniques for evaluating the beneficial and adverse
effects of streamflow regulation. Particularly important are the effects
of storage on reservoir and downstream water quality, thermal
stratification, peaking power generation, and consumptive uses of
water.
This Symposium on Streamflow Regulation for Quality Control
was prompted by observations of low-quality water in the discharge
from several projects studied by the Technical Services Branch of
the Public Health Service, by similar observations of projects re-
ported by others, and by the interest this has generated in the de-
velopment of measures to correct these conditions. The Public Health
Service has initiated both intramural and extramural studies in this
field. The Basic and Applied Sciences Branch is undertaking studies
on the effects of streamflow regulation on water quality, which in-
clude in the initial phases quality changes in storage reservoirs and
the waste contributions of urban and rural runoff. The Research and
Training Grants Branch is financing research, training, demonstra-
tion, and fellowship grants, the results of which will contribute
toward solution of problems in the streamflow regulation field.
The consideration of quality changes is only part of the story,
however. Economic implications, which are equally important, are
not emphasized in this Symposium as the economics of quality con-
trol could be the subject of another symposium of equal importance.
The purpose of this Symposium was to bring together engineers,
scientists, and administrators from various levels and agencies of
government, from industry, and from universities to:
1. Facilitate exchange of information on the effects of streamflow
regulation on water quality,
2. Review available knowledge on the effects of streamflow
regulation on water quality,
3. Discuss studies presently in progress on quality control prob-
lems associated with streamflow regulation, and
4. Explore research needs in the field of streamflow regulation
for quality control.
To obtain a degree of informality and facilitate the exchange of
information, preprints of many of the papers were distributed to
registrants in advance of the Symposium. The technical papers were
presented, discussed by one or more reviewers, and then opened to
floor discussion. Discussions were recorded for inclusion in this
volume.
iii
-------
The papers and discussions in this volume were prepared for oral
delivery in the Symposium and necessarily have been edited to meet
publication standards of the Public Health Service. Every effort has
been made to present accurately the full data and the original views
anv.. meaning intended by the authors. The papers that follow may
therefore differ slightly from the preprints made available to Sym-
posium participants. Floor discussions have been edited to standards
for written presentation by technical reviewers. Publication of the
papers by the Public Health Service is not meant to imply endorse-
ment of the conclusions, or of any commercial products that may be
mentioned.
John E. McLean
Program Chairman
Iv
-------
SYMPOSIUM COMMITTEE
Program
J. E. McLean, Chairman, Chief, Water Resources Unit, TSBa
S. R. Weibel, Deputy Chief, Engineering Section, B&ASBb
R. Porges, Deputy Chief, Field Operations Section, TSB
F. H. Rainwater, Operations Officer, TSB (Washington, D.C.)
Arrangements
R. T. Bridges, Administrative Assistant, B&ASB
L. J. Potter, Public Health Engineer, TSB
Registration
M. H. Emshwilier, Secretary, B&ASB
R. E. Hyde, Administrative Assistant, TSB
Publicity
J. H. Durrell, Public Information Officer, SECC
S. R. Weibel, Deputy Chief, Engineering Section, B&ASB
J. M. Symons, Public Health Engineer, B&ASB
Hospitality
R. Porges, Deputy Chief, Field Operations Section, TSB
G. M. McDermott, Sanitary Engineer, B&ASB
Banquet and Luncheon
A. D. Sidio, Public Health Engineer, TSB
R. T. Bridges, Administrative Assistant, B&ASB
Publications
K. W. Cassel, Jr., Publications Officer, SEC
F. B. Layne, Publications Editor, SEC
T. W. Bendixen, Soil Scientist, B&ASB
J. E. McLean, Chief, Water Resources Unit, TSB
¦TSB—Technical Services Branch, Division of Water Supply and
Pollution Control, Robert A. Taft Sanitary Engineering Center,
Cincinnati.
b B&ASB — Basic and Applied Sciences Branch, Division of Water
Supply and Pollution Control, Robert A. Taft Sanitary Engi-
neering Center, Cincinnati.
c SEC — Office of the Director, Robert A. Taft Sanitary Engineering
Center, Cincinnati.
v
-------
CONTENTS
Welcome: Page
J. E. Flanagan, Jr 1
F. E. DeMartini - 2
L. W. Weinberger 4
Introduction of the Water Pollution Control Advisory Board:
J. M. Quigley 5
Session 1: WHY REGULATE STREAMFLOW FOR QUALITY
CONTROL
G. E. McCallum — Public Health Service Responsibilities for
Streamflow Regulation for Quality Control 11
E. W. Weber — Relation of Regulation for Quality Control to
Activities of the Corps of Engineers 15
D. R. Burnett — Relation of Regulation for Quality Control
to Activities of the Bureau of Reclamation 23
H. O. Ogrosky — The Influence of Conservation Practices and
Measures on Water Quality 29
F. S. Brown — Relation of Regulation for Quality Control to
the Activities of the Federal Power Commission 37
M. Stein — Flow Regulation for Water Quality Control and
Water Rights - 43
Session 2: EFFECTS OF IMPOUNDMENTS ON WATER QUAL-
ITY IN RESERVOIRS
F. W. Kittrell — Thermal Stratification in Reservoirs 57
D. R. F. Harleman — Discussion 67
Discussion from the Floor 74
C. H. J, Hull — Photosynthesis as a Factor in the Oxygen
Balance of Reservoirs 77
J. Verduin — Discussion 91
Discussion from the Floor 94
S. K. Love and K, V. Slack — Controls on Solution and Pre-
cipitation in Reservoirs 97
J. K. Neel — Discussion 120
Discussion from the Floor 124
Session 3: EFFECTS OF FLOW REGULATION ON WATER
QUALITY (Part 1)
R. A. Vanderhoof — Changes in Waste Assimilation Capacity
Resulting from Streamflow Regulation 129
A. J. Kaplovsky — Discussion 146
Discussion from the Floor 156
J. H. Svore — Mineral Quality Control Through Streamflow
Regulation 161
J. D. Hem — Discussion 175
Discussion from the Floor 177
M. A. Churchill — Control of Temperature Through Stream-
flow Regulation 179
G. E. Harbeck, Jr. — Discussion 193
J. J. Gannon — Discussion 195
Discussion from the Floor 196
vii
-------
Session 4: EFFECTS OF FLOW REGULATION ON WATER
QUALITY (Part 2)
K. M, Mackenthun — The Effects of Nutrients on Photosyn-
thetic Oxygen Production in Lakes and Reservoirs 205
C. N. Sawyer — Discussion 215
Discussion from the Floor 220
F. H. Rainwater — Hydrologic Facts Needed for Studies of
Flow Regulation for Stream Quality Control 221
C. J. Velz — Discussion 227
R. E. Lundquist — Discussion 231
Discussion from the Floor 240
M. B. Fiering — The Nature of the Storage-Yield Relationship 243
R. L. Woodward — Discussion 253
Z. Spiegel — Discussion 254
Discussion from the Floor 256
J. B. Coulter — Objectives and Criteria for Water Pollution
Control 261
R. L. Smith — Discussion 268
Discussion from the Floor 273
Session 5: MEASURES FOR IMPROVING THE QUALITY OF
RESERVOIR DISCHARGES
W. E. Knight — Improvement of the Quality of Reservoir
Discharges Through Control of Discharge Elevation 279
J. M. Hester — Discussion 290
Discussion from the Floor 295
T. F. Wisniewski — Improvement of the Quality of Reservoir
Discharges Through Turbine or Tailrace Aeration 299
W. S. Lee — Discussion 308
Discussion from the Floor 312
J. G. Bryan — Improvement in the Quality of Reservoir Dis-
charges Through Reservoir Mixing and Aeration 317
L. L. Falk — Discussion 334
Discussion from the Floor 336
Session 6: OPERATION AND ADMINISTRATION OF FLOW
REGULATION PROJECTS
S. Ragone and B. J. Peters — Water Quality Monitoring for
Water Quality Control 345
R. K. Horton — Discussion 365
Discussion from the Floor 369
W. King and E. C. Kinney — Cooperation in the Solution of
Water Quality Problems Associated with Flow Regula-
tion 373
W. W. Towne — Discussion , 401
Discussion from the Floor 404
K. S. Krause — Where We Stand on Streamflow Regulation
for Quality Control 409
Banquet Session:
M. K. Goddard — Public Expectations - 415
viii
-------
WELCOME
Joseph E. Flanagan, Jr.*
Acting Director, Robert A. Taft Sanitary Engineering Center
U. S. Public Health Service, Cincinnati, Ohio
One of the advantages of appearing on a program like this is
that the Program Chairman just says,"Come on down and greet the
group and make them welcome." I am always happy to do this.
I do not have to prepare a speech, and I am not limited in what I
may say. It's a very fine position to be in. I don't stand here to sub-
stitute for the Mayor, but I would like to use his usual phrase and
simply say to all of you, and particularly those of you who have not
been to Cincinnati before, "Welcome to the Queen City." It's a very
nice town.
I have been struck by what has happened in the field of symposia
since I came to Cincinnati eight and a half years ago. Then, they were
held in the auditorium at the main building of the Center. It could
seat 150 people and sometimes it wasn't even filled. The subjects were
usually quite circumspect. Generally a particular part of a branch
had a problem, and it was decided that it might be helpful to both
the Service and to some of the outside folks to bring a few technical
people together to talk about it, and, hopefully, to come up with a
solution. If we were particularly successful in solving a problem, it
would show up in one of our training courses, whether it was water
pollution, air pollution, radiological health, or something else.
That time has long since passed, and it is interesting that it
passed without any pushing on the part of administrative people at
the Center or, as far as I know, on the part of administrative people
in Washington. I think the group assembled for this Symposium on
Streamflow Regulation for Quality Control is a fine example of the
tremendous interest on the part of individuals in many different
government agencies, in educational institutions, and in private or-
ganizations, in the subjects presented. Nobody seems to come unless
he is particularly interested in the subject, and I think that the
Public Health Service gets a great deal from these meetings. I also
like to think the other agencies and individuals who participate get
a great deal out of them, too.
Another thing that is happening, in addition to the growth in
numbers of people who attend symposia like this is that the subjects
are getting more complicated. It would seem that symposia are
chosen in proportion to the difficulty of the particular problem. We
will talk about streamflow regulation for quality control for three
days. This no longer concerns only one of the branches in our water
program, it concerns all three of them. It also concerns others besides
the Public JHealth Service.
*Now Associate Director, Department of Environmental Health,
American Medical Association, Chicago.
Flanagan
1
-------
The next symposium scheduled to be held in Cincinnati is not
limited to one program of the Public Health Service. It is of interest
to me that it is jointly sponsored by a branch of the Water Supply
and Water Pollution Control Division and a branch of the Air Pollu-
tion Division. It wouldn't surprise me if some other divisions were
also involved in it, too.
I thought you would be interested in what is happening, and that
it is happening in spite of administration lines. The technical people
see these problems, get their heads together, and, regardless of their
particular program, get together and discuss it. As far as I am con-
cerned, this is all to the good. I would estimate that in another 10
or 15 years we are going to see symposia that will stretch across very
broad segments of the fields we now refer to as environmental health.
And when that day comes, I am sure the Water Division of the Public
Health Service will be one of the most important segments in this
broad activity.
The Water Pollution Control Advisory Board is also meeting in
Cincinnati this week and is attending this session of the Symposium.
Mr. James M. Quigley, Assistant Secretary of the Department of
Health, Education, and Welfare is with us and will describe the
Board and its activities.
Frank E. DeMartini
Chief, Technical Advisory and Investigations Section
Division of Water Supply and Pollution Control, USPHS
Robert A. Taft Sanitary Engineering Center, Cincinnati
Dr. Weinberger and I are shown on the program as part of the
welcoming group. Our comments, however, are primarily intended to
set a background for the meeting. I want to add to Mr. Flanagan's
welcome. Speaking for the Technical Services Branch, which is one
of the co-sponsors of the meeting, we are very pleased to see this
tremendous turnout and hope when these 3 days are over you will all
feel that it has been very worthwhile to have participated. We in
the Service believe we are going to gain a good deal from the papers
and discussions presented.
As to the purposes and objectives of the Symposium, they are
given in your program. Our objective was to bring together engineers,
scientists, and administrators from agencies of government, from in-
dustry, universities, private institutions, and organizations to discuss
flow regulation for quality control. We planned the program so that
time would be available after each presentation not only for a formal,
prepared discussion, but also for contributions and participation by
those who are in attendance. I think this is one of the chief values
that will come from this meeting. We urge your participation.
2
WELCOME
-------
Another bit of background that should be mentioned is the
Federal Water Pollution Control Act. Recent amendents provide
that in the survey or planning of any reservoir by a Federal agency,
consideration shall be given to the inclusion of storage for the pur-
pose of flow regulation for quality control, but that such storage is
not intended to be provided as a substitute for adequate treatment or
other means of control of wastes at their source. The need for and
the value of such storage is to be determined by the construction
agencies with the advice of the Secretary of the Department of
Health, Education, and Welfare, and his views are required to be
included in any report submitted to Congress proposing authorization
for construction of any reservoir that includes such storage.
Within the Public Health Service, the Division of Water Supply
and Pollution Control is responsible for evaluating these needs. The
Technical Services Branch of the Division directs the Regional Offices,
the Comprehensive Water Pollution Control Projects, and the Techni-
cal Advisory and Investigation Section, located at Cincinnati, in these
endeavors.
In making these evaluations, the various subjects to be discussed
in the next 3 days that may be pertinent to the particular reservoir
undertaking are considered, and the best engineering estimates possi-
ble are arrived at as to the effects of the proposed reservoir project
on water quality. Based upon the reservoir, the downstream uses,
the effects of wastes reaching the waters, the treatment that can be
provided, the operating schedule of the reservoir, and other pertinent
factors, the need for storage and its value for quality control are
determined.
These engineering estimates can be improved as we gain more
adequate knowledge in many areas. Gaps in our knowledge point
up the need for research and field studies. I have listed a few of the
areas that are worthy of further consideration:
1. The accounting for all wastes that reach our streams. We
need to know more about land drainage, urban run-off, chemical
poisons, pesticides, and weedicides that reach our waters.
2. The pollutional characteristics of new wastes. We must know
more about new types of wastes, many of which are the result of the
technological development taking place.
3. The treatment or control of persistent undesirable wastes that
are not easily oxidized.
4. The effects of recreational use of waters.
5. The control of siltation. Silt from highway construction,
timber harvesting, and land clearing for urban development is a
serious threat to streams and to reservoirs, and means of control are
needed.
6. The fate of radioactive materials in water.
DeMartini
3
-------
7. The effects of impoundments on the assimilation of organic
wastes, deoxygenation and reaeration, bacteriological changes,
patterns of biological life, and temperature changes. A great need for
more pre- and post-impoundment studies exists to develop a better
basis for quantitatively predicting water quality in reservoirs as it
affects the uses of water for various purposes including residual waste
disposal.
This mention of some of the problem areas indicates there is
still much we need to know about and do. Many interests can con-
tribute in arriving at answers and in fortifying our positions in these
areas. Among them are Federal, State, and interstate agencies, uni-
versities, private research institutions, and industry.
Dr. Weinberger will give you a brief resume of the work the
Public Health Service is doing in these areas, both directly and in-
directly by way of contracts and other arrangements. During the
3 days of the Symposium we will be hearing what many others are
doing, which will focus attention on many of these problem areas.
I believe we have very high caliber participants; in addition, the
enthusiastic response to the Program Committee's invitations to pre-
sent papers augers well for a very productive meeting. It is certain
that significant contributions will come from the sessions of the next
3 days, and that problems for future emphasis will evolve during the
discussions.
Dr. Leon W. Weinberger
Chief, Basic and Applied Sciences Branch
Division of Water Supply and. Pollution Control, USPHS
Robert A. Taft Sanitary Engineering Center, Cincinnati
It is usually not very difficult to welcome a group and normally
I would have felt quite at ease in carrying out this function. You
will note, however, that not one, not two, not even three, but four
people have already welcomed you. What am I supposed to do? My
conclusion is that the Program Committee had some doubts about the
meeting starting on time and I am to fill in, if necessary, with a
lengthy presentation until we have a quorum.
Since we are gathered I will keep my remarks short.
The function of the Basic and Applied Sciences Branch as the
research arm of the Division of Water Supply and Pollution Control
is outlined on the front cover page of the program. It indicates that
we undertake intramural and extramural studies and research on
the effects of streamflow regulation as part of our program.
I would like to expand upon the extramural part of the program.
Our program of research grants and contracts has evolved because
4
WELCOME
-------
of a realization that if we are to accomplish the research goals we
have in mind, it must be in cooperation with and through the efforts
of all other interested and competent people in this field. So our
extramural research grants and contracts in cooperation with other
Federal, State, and interstate agencies, universities, research institu-
tions, industries, and individuals, are an essential part of our research
program.
INTRODUCTION OF THE WATER POLLUTION
CONTROL ADVISORY BOARD
James M. Quigley
Assistant Secretary
Department of Health, Education, and Welfare, Washington, D. C.
I am here in the capacity of Chairman of the Federal Water
Pollution Control Advisory Board. I am sure you are all aware that
the Federal Water Pollution Control Act provided for the creation of
a nine-man advisory board appointed by the President of the United
States. The Board serves in an advisory capacity to review, oversee,
and comment on the operation of the Federal water pollution control
program and to make any suggestions it deems advisable that will
improve the operation of the program.
In addition to the nine Board members appointed by the Presi-
dent, the Secretary of the Department of Health, Education, and Wel-
fare has designated me to serve as its Chairman. The Board meets
about four times a year. We try to have no more than half our meet-
ings in Washington because we feel that by meeting in other places
we accomplish two purposes: First, we educate the Board as to the
particular pollution problems of the various areas of the country,
which as you know, and as I have learned, vary markedly from one
part of the country to the next; and second, we have found from
experience that our meetings tend to focus public attention on the
water pollution problems of an area. We feel this has served a very
constructive and useful purpose. Our presence and our activities help
to create an atmosphere in which governors can do more with their
legislative bodies; State water pollution control agencies experience
a little better cooperation with their appropriation and budgetary
officials; and public attention is focused on the need for community
action. An example of this was the bond issue in the City of St.
Louis where we held hearings some 5 or 6 weeks before the vote was
cast. We don't take full credit for the favorable results, but we like
to feel that the public attention that was focused on our presence and
the Federal government's program helped to make the voters of that
city and area much more aware of the problems involved and of their
local responsibility — responsibility they faced up to magnificently.
In accordance with this approach, we decided at our December
Quigley
5
-------
Board meeting to take advantage of the symposium you are having
here and to hold our April meeting in Cincinnati at the same time.
In addition to getting a certain exposure to many of the things you
are going to discuss, this is an opportunity to show the members of
the Advisory Committee the facilities Mr. Flanagan heads at the Taft
Sanitary Engineering Center, and generally to get a look at one of
the Public Health Service's more important operations in the water
pollution control program. So this is why we are here.
The members of the Board are a diversified group, and de-
liberately so. In making recommendations to the President, the De-
partment tries to put together a group that will be representative
of a variety of viewpoints in this very important field. They are di-
versified as to background, training and experience, geography, and
politics. The current members of the Board are:
Mr. John A. Biggs, Director, Department of Game
Olympia, Washington
Dr. Clair S. Boruff, Technical Director
Hiram Walker and Sons
Peoria, Illinois
Mr. Charles H. Callison, Assistant to the President
National Audubon Society
New York, New York
Mr. M. James Gleason,* Commissioner
Multnomah County
Portland, Oregon
Dr. Maurice K. Goddard, Secretary
Pennsylvania Department of Forests and Waters
Harrisburg, Pennsylvania
Mr. Lee Roy Matthias, Executive Vice President
Red River Valley Association
Shreveport, Louisiana
Mr. Tom McCann,* President
McCann Construction Company
Fort Worth, Texas
Mr. Ed E. Reid,* Executive Director
Alabama League of Municipalities
Montgomery, Alabama
Mr. William E. Warne,* Director
California Department of Water Resources
Sacramento, California
In addition to the Board members, Mr. William Towell, Director
of State Conservation, State of Missouri, a consultant to the Board,
is also present.
* Not able to attend because of other commitments.
6
INTRODUCTION OF ADVISORY BOARD
-------
As I have indicated, the purpose of our meeting in Cincinnati,
or anywhere, is twofold: (1) to educate the members of the Board,
and (2) to educate those who might come in contact with the mem-
bers of the Board. It is a two-way street. Many of you are here for
other purposes, but if you have any suggestions, if you have any ideas
as to how the Federal Water Pollution Control Advisory Board might
operate more effectively and more efficiently, or any suggestions as
to how the Federal Water Pollution Control Act might be changed or
extended or amended in a way that you think would make it a better
and more effective law to help clean up and purify the waters of this
country in the manner and in the time in which we have to do this
assignment, we would be very happy to hear from you.
Quigley
7
-------
Session 1
WHY REGULATE STREAMFLOW
FOR QUALITY CONTROL
Moderator: John E. McLean
U. S. Public Health Service
It is appropriate at the outset of this symposium to ask: Why
regulate streamflow for quality control? At the Federal level there
are at least two answers to this question. The first is in terms of
technical solutions to water quality problems; the second is in terms
of assigned responsibilities under applicable laws. In this session,
five agencies having primary responsibilities or activities significantly
affecting quality of water and its control by flow regulation will
present their parts of the answer under applicable laws. Technical
aspects are treated in the remaining sessions.
-------
PUBLIC HEALTH SERVICE RESPONSIBILITIES FOR
STREAMFLOW REGULATION FOR
QUALITY CONTROL
Dr. Gordon E. McCallum*
Chief, Division of Water Supply and Pollution Control
U. S. Public Health Service, Washington, D. C.
It is easy this morning to assess the importance of this Symposium
on Streamflow Regulation. In this group are representatives of the
five great Federal agencies that possess major responsibilities in water
management, plus administrators, engineers, and scientists from State
and local government, from business, and from our universities. I can
think of no phase of water resources other than streamflow regulation
that in this year of 1963 could bring together so diversified and
influential an audience.
There is no reason for surprise in this. For a good many years
we have recognized — and recognized rightly — that the two most
effective ways of protecting stream quality are by eliminating wastes
at their source or, failing that, by removing them from waste waters.
Streamflow regulation is a third method of quality control, necessarily
subservient to these other two, but a method nonetheless. It is time
we knew more about it — knew, in the rather stately words of your
program, of its "beneficial and adverse effects."
The particular event in history most responsible for this sym-
posium took place 2 years ago with the passage of the 1961 amend-
ments to the Federal Water Pollution Control Act. Among these
were the amendments concerning streamflow regulation, which now
appear in the Act as Section 2(b), and are set forth below:
Section 2.
(b)(1) In the survey or planning of any reservoir by the
Corps of Engineers, Bureau of Reclamation, or other Federal
agency, consideration shall be given to inclusion of storage for
regulation of streamflow for the purpose of water quality con-
trol, except that any such storage and water releases shall not
be provided as a substitute for adequate treatment or other
methods of controlling waste at the source.
(2) The need for and the value of storage for this purpose
shall be determined by these agencies, with the advice of the Sec-
retary, and his views on these matters shall be set forth in any
report or presentation to the Congress proposing authorization or
construction of any reservoir including such storage.
(3) The value of such storage shall be taken into account in
determining the economic value of the entire project of which
*Dr. McCallum's paper was presented by Mr. Robert R. Harris,
U. S, Public Health Service.
McCallum
11
-------
it is a part, asd costs shall be allocated to the purpose of water
quality control in a manner which wilt insure that all project
purposes share equitably in the benefits of multiple-purpose
construction.
(4) Costs of water quality control features incorporated in
any Federal reservoir or other impoundment under the provisions
of this Act shaJ] be deterrainad and the ^Tttiitiaries identified
and if the benefits are widespread cr national in scape, Ihe costs
of such features shall be nonreimbursable.
The net effect oi these amendments was to put the Federal water
pollution control program in partnership with the construction
agencies in determining the need of water storage for waste dilution.
It has had a further effect, however, both in our program in the
Public Health Service and in the scientific and research activities of
other govern men4, and private agencies. It has served to increase
scientific interest in the problem of waste dilution, has encouraged
many of our ablest students to turn their attention to this area of
water management, not only in their work in the laboratory but in
the field. This new interest, ir.cst obviously, is reflected ;n this room
today.
Neither in concept nor in practice is streamSow regulation new
since 2961. My first assignment vrithirv the Public Health Service v.-as
along the Ohio Hiver, srri 9ur 1&38-1S40 survey look augmented
streair.fl.ow into consideration, as did sortie other surveys of an earlier
date. But our approach, I am afraid, "P/as >amev?nat unsopfcisis&lec.
> i xrnf of taca? 5 prcblexs aid tc day's Kntivledge,
One reason toe this use the specialization ot our staff. Until
recently, water pollution control was a field occupied almost ex-
clusively by attorneys and engineers, the attorneys to prosecute the
transgressor and the engineer la huiM the remedi-al work. It takes
more than this to understand arid manage water quality today. We
need the sfci'Js of many scientific disciplines, including those of the
liie sciences and the social sciences, and we are beginning to assemble
these skills.
in an earlier day , we suffered from this lack of knowledge and
also from lack of legislative authority. The first Federal Water Pollu-
tion Control Act, as such, was riot passed until 1&4B, and a permanent
Federal law was not enacted until 1956. As I said earlier, stream/low
regulation for quality control did not appear in the Federal law
until 1961.
The legislative ¦development irora 1543 until 1961 occurred
against a background of itucx easing, FLa\fcpjia!s needsr and inrreasing
public discontent with our inability to meet these needs. The National
Conference on Water Pollution held in Washington, D.C., in I960,
expressed this r.e'j? p-jb^c feeling very rpre£L
" The goal of pollution abatement," the Conference recorded, "is
to protect asvd enhance the capacity oi the water resource to serve
12
PUBLIC HEALTH SERVICE'S RESPONSIBILITIES
-------
the widest possible range of human needs; this goal can be approached
only by accepting the positive policy of keeping waters as clean as
possible as opposed to the negative policy of attempting to use the
full capacity of water for waste assimilation."
The report of the Senate Select Committee on National Water
Resources, published in 1961, related this to water storage and
streamflow augmentation.
"Along with . . . waste collection and treatment facilities," the
Committee reported, . . very substantial reservoir storage will have
to be provided to supply water for dilution of the effluent from these
plants, if the job of cleaning up our rivers is to be done in the most
economical way."
The Committee estimated the cost of such reservoir storage to be
$12.1 billion by 1980, and $18 billion by the year 2000. This did jnot
include costs of storage needed for other specific purposes, such as
flood control, irrigation, and municipal and industrial water supplies.
Some of the storage provided for the purposes will, however, un-
doubtedly assist in the river regulation job the Committee envisioned
for pollution abatement.
The Public Health Service responsibilities in streamflow augmen-
tation for quality control tie in very closely with those stemming
from the earlier Water Supply Act of 1958, which provided for stor-
age in Federal reservoirs to meet anticipated future municipal or
industrial water supply needs. For this purpose and for the stream-
flow regulation requirement of 1961, we have by now prepared re-
ports for the Federal construction agencies on 125 projects and have
210 others in progress or scheduled.
Importantly interwoven with these forecast-evaluations is the
knowledge gained from our National Water Quality Network. This
system of 125 sampling stations on interstate watercourses — due
for expansion to 300 — is keeping check on the effects of pollution
in our streams. The Network is building up a very useful store of
basic data in some 15 physical, chemical, and biological parameters
in water — such as radioactivity, organic chemicals, coliform organ-
isms, temperature, and others.
These programs are among 15 or so authorized by the Federal
Water Pollution Control Act. Every one of them blend into and
complement our comprehensive river-basin planning studies in which
many of you are involved and cooperating. These are progressing in
various stages now in seven major basins, and four others are
scheduled for early starts. We hope to complete studies for all United
States basins by the year 1975.
In each of these river basin studies, needless to say, the "bene-
ficial and adverse" effects of streamflow regulation are being studied,
and in every case our decision-making process is calling for more in-
formation and better understanding. Your symposium subject this
afternoon, for example, is the "Effects of Impoundments on Water
McCallum
13
-------
Quality in Reservoirs." This is paradoxical, when one stops to think
about it, for it suggests that in storing water to protect quality, we
may also, to some extent, be impairing that quality. Paradoxical or
not, these effects must be determined.
There are many questions, but not yet enough answers. This is
why the Public Health Service has called this seminar and why, I take
it, you have been good enough to come to Cincinnati to attend it.
On behalf of the Division of Water Supply and Pollution Control, I
welcome you to this meeting and wish you every success in it.
14
PUBLIC HEALTH SERVICE'S RESPONSIBILITIES
-------
RELATION OF REGULATION FOR QUALITY
CONTROL TO ACTIVITIES OF THE
CORPS OF ENGINEERS
Eugene W. Weber*
Chief, Civil Works Planning Division
U. S. Army Corps of Engineers, Washington, D. C.
The enactment of Public Law 87-88 on 20 July 1961, less than
2 years ago, has made it possible to give full consideration in the plan-
ning and development of the Corps of Engineers' water resources
program to provisions for regulation of stream flow for water quality
control purposes on a par with all other project purposes. I will
outline briefly the import of this new authority on three categories
of projects in the Corps' program; i.e., completed projects, those
authorized but not yet completed or started, and those being studied
for possible future authorization.
COMPLETED PROJECTS
The Corps, in 1962, had in operation 214 reservoir projects con-
taining 169 million acre-feet of storage capacity, primarily for flood
control and power purposes, with provisions for recreational use at
most and with storage for water supply at some. Needs for flow
regulation for water quality control were considered in planning
many of these projects, but because of the absence of a positive policy
for Federal participation in providing for such regulation, specific
measures for quality control were included in relatively few projects,
such as Fall River, Toronto, and Kanopolis Reservoirs in Kansas;
Berlin and Mosquito Creek Reservoirs in Ohio; Youghiogheny River
Reservoir in Pennsylvania; and Tygart Reservoir in West Virginia.
The importance of low flow augmentation has been amply demon-
strated by these earlier projects. For example, the operation of Fall
River Reservoir, Kansas, in the very severe drought in the 1950's
provided vitally needed water supply and dilution of wastes involving
six downstream municipalities and also prevented complete shutdown
of three refineries and other industries. As another example, the
Berlin project, Ohio, was constructed in the early 1940's, at the re-
quest of the War Production Board, primarily for low flow regulation
in the Mahoning River Basin, to maintain maximum power and steel
production in this highly industrialized valley. In view of the con-
tinued critical situation with respect to steel production, the method
of operation specifically adopted as a war-time measure was con-
tinued until 1949, at which time a more balanced method of operation
between flood control and low flow regulation was adopted. Some of
our very early studies of low flow augmentation, which were carried
out in cooperation with the Public Health Service as far back as the
* Mr. Weber's paper was presented by Col. Robert J. Kasper, U. S.
Corps of Engineers.
Weber
15
-------
late 1930's, established the need for the inclusion of regulatory storage
in projects only recently undertaken, such as the Allegheny Reservoir
above Pittsburgh, During this same early period, the Ohio River
Pollution Survey was completed through oar joint efforts. This pro-
vided a very valuable background not only in the planning of Corps
projects but +o all other interests -concerned with pollution of the
OhLD Hirer.
Although water quality control was authorized as a specific pur-
pose of only a few of our 214 completed reservoirs, some planned
regulation for quality control is being made part of reservoir oper-
ating plans whenever it is consistent with the purposes for which the
projects were authorized. Furthermore, in nearly all ca&es, the effect
of regulation for authorized purposes is incidentally beneficial to
water quality. Also, as we learn more about the effects of reservoir
operations, steps can usually he taken to prevent adverse effects on
water quality.
On these completed projects, two general courses of action are
now possible. First, adjustments consistent with the authorized pur-
poses of the projects, particularly in operations, can continue \o be
made in the light of experience and changing requirements for water
quality. Second, completed projects can be resrudied, particularly in
ccnjxr>clioc w:tfc basin-wide stucies for additional projects to meet
future neecs and, when feasible, recommendations made to Congress
for modification of existing projects in the interests of quality control.
We do not have authority to modify completed projects to include
quality control without further action by Congress.
AUTHORIZED PROJECTS UNDER CONSTRUCTION
OR NOT STARTED
Including those most recently authorized last year in Public Law
81-874, the Corps has authority to construct over 200 additional proj-
ects containing' some 90 million acre-feet of storage. Of these, 81
reservoirs with 56 million acre-feet of capacity have been placed
under construction. On all of the projects not yet started and on
those where planning for certain phases of construction is still under
wayr the new PubJic Law 87-28 provides a basis fcr considering modi-
fications of previously authorized plans to include provisions for flow
regulation for water quality control if the following criteria are met:
!. The modification of the project must be economically justified.
2. The status of the planning &c.d construction of the project
must be such that it is practicable to accomplish the necessary changes
without undue delay or unreasonable increase in cost over that which
would have been incurred if water quality control had been originally
authorized as a project purpose.
3. The modification must not have a material adverse effect on
the purposes for which the project was authorized originally.
4. Satisfactory advise must be reoeived from the Secretary of
16
RELATION TO CORPS OF ENGINEERS
-------
the Department of Health, Education, and Welfare relative to the
need for and the value of the water quality control storage.
5. The Congress must be informed of our intention to modify
the project and of the views of the Secretary of the Department of
Health, Education, and Welfare by letters to the Public Works and
Appropriations Committees of the Senate and House.
Last year we informed Congress that we were modifying two
projects, John W. Flannagan Reservoir, Ohio River Basin, Virginia,
and DeGray Reservoir, Ouachita River Basin, Arkansas, to include
water quality control storage as a primary project purpose. We feel
certain that many other authorized projects can be modified to include
this important purpose as detailed planning studies proceed.
PROJECTS BEING STUDIED FOR FUTURE AUTHORIZATION
In the projections developed by Resources for the Future for the
1961 Report of the Senate Select Committee on National Water Re-
sources, it was estimated that by 1980 the United States would need
about 94 million acre-feet of additional storage for flood control and
about 315 million acre-feet for water quality control and other pur-
poses dependent upon low flow regulation. By the year 2000, it was
estimated that about 446 million acre-feet would be required for the
latter purpose. These formidable requirements for storage for water
quality, the 1980 amount being about double the capacity we have
now developed for all other purposes, were based on the assumption
that the desired quality of water in rivers would be achieved by the
cheapest combination of waste treatment facilities and storage to pro-
vide sustained flows for satisfactory dilution of effluent. On this basis
it was estimated that the degree of waste treatment at the source can
generally be expected to range upward from 70 percent treatment
in 1980 and 80 percent treatment in the year 2000, and up to 95 per-
cent or even higher in water-short regions or where costs of reservoir
storage are high.
Even with an early breakthrough in efforts to improve methods
for treatment of wastes at the source, it would be many years before
improved methods could be applied to replace present measures. Ac-
cordingly, the Corps has recently estimated that the programs for
which it has planning responsibility should be formulated with a
view to providing, by 1980, about 320 million acre-feet of reservoir
storage capacity. Under immediately prospective conditions, much
of this capacity is needed in the interests of water quality control. Even
if demands for storage capacity can be reduced by improved waste
treatment procedures, comparable requirements for flow regulation
will still be needed for other demands for use of our water resources.
This prospective program of reservoir storage, distributed as
shown in Table 1, would require a cumulative expenditure for reser-
voir construction totaling about $15 billion by 1980. This means in-
creasing the program from the current annual rate of about $0.5
billion to about $1.5 billion in 1980.
Weber
17
-------
TABLE 1. ADDITIONAL RESERVOIR CAPACITY NEEDED
BY 1980 AND ESTIMATED COST
Portion that
might be
Region a
Flood
control
Other
purposes6
Total
provided
under Corps'
program
Estimated
cost,0
Millions
of dollars
Millions of acre-feet
New England
0.8
4.0
4.8
4.8
365
Middle Atlantic
3.2
19.9
23.1
22.1
1,859
South Atlantic
3.7
26.3
30.0
28.0
1,010
Ohio Basin
8.3
13.9
22.2
21.2
1,550
Tennessee Basin
0
6.5
6.5
0
0
Great Lakes
0.4
41.4
41.8
41.8
1,969
Upper Mississippi
9.0
14.8
23.8
23.8
1,020
Lower Mississippi
0.8
18.7
19.5
19.5
370
Ark-White-Red
16.8
27.5
44.3
34.3
927
Gulf Southwest
12.6
38.1
50.7
43.7
1,660
Red of North
0.3
3.0
3.3
1.8
54
Missouri Basin
17.5
31.2
48.7
38.7
1,320
Colorado Basin
0.4
13.4
13.8
3.0
96
Great Basin
0.6
6.8
7.4
1.3
62
California Coastal
1.3
5.7
7.0
3.0
246
Central Valley
3.1
24.0
27.1
10.1
415
Columbia Basin
14.8
16.7
31.5
22.0
1,670
N. Pacific Coastal
0.8
3.3
4.1
3.6
230
Totals
94.4
315.2
409.6
322.7
14,823
° Substantially the regions shown on the map of United States Water
Resource Development, edition of 1958.
6 Based upon estimates by RFF of storage required for multiple-
purpose river regulation (the controlling factor is generally the flow
required to maintain satisfactory water quality).
cCost of capacity that might be provided under the Corps of Engi-
neers' program.
IMPLICATIONS FOK THE FUTURE
What does this prospect of requirements of reservoir storage ca-
pacity for water quality control mean for those concerned with water
resources planning and development? It means, of course, that water
quality considerations must be taken into account much earlier and
more completely than they have in past planning and development of
water resources. I will discuss three of the many significant implica-
tions for the future: projections of needs, evaluation, and research.
18
RELATION TO CORPS OF ENGINEERS
-------
PROJECTIONS OF NEEDS
A basic first step in planning for conservation and effective use
of water resources is the delineation of the probable future demands
for use of water. Since water development projects are and will con-
tinue to be an essential part of all human activity, projections of water
needs should take into account long-range as well as immediate
requirements. It is difficult to get planners, administrators, and legis-
lators to accept the implications of the doubling and tripling of popu-
lation, gross national product, and other indices of activity that we
face in coming decades. We must find ways to translate the current
projections of growth and change into estimates of requirements for
amounts and qualities of water needed at various times in the long-
range future. Insofar as water quality is concerned, this poses many
difficult problems, such as predicting what type and amount of con-
taminants will remain in treated effluents of the future and what
standards of quality should be maintained in specific reaches of
streams to permit use for water supplies, for recreation, for fish and
wildlife, etc.
In our current comprehensive basin-wide studies, such as on the
Ohio and the Susquehanna, the Corps is coordinating its arrange-
ments for economic base studies and projections of needs with the
Department of Health, Education, and Welfare in order that the
formulation of long-range basin-wide plans and specific immediate
projects may reflect the best possible consideration of water quality
requirements as well as other water needs. Joint undertaking of
economic base surveys serves to meet the individual and combined
needs of the participants and thus to reduce the overall cost of in-
vestigations necessary to devise effective water resource development
and management plans.
EVALUATION OF EFFECTS OF FLOW AUGMENTATION
A second area that will demand extraordinary efforts of water
planners is the evaluation of the effects of regulation of flows for the
purpose of water quality control. With the steadily mounting de-
mands for use of water and the increasing difficulty of finding prac-
ticable sites for reservoirs, it is becoming more and more important
that we be able to evaluate the effects of prospective water manage-
ment measures accurately. We need improvement in our evaluations
in absolute terms in order to be sure of the justification of our water
development proposals, which are becoming more costly as better
sites are preempted for various purposes. Even where decisions can
safely be made without full determination of absolute values, it is
necessary at least to have good definition of relative values or limiting
values to insure balanced project formulation in the light of com-
peting or conflicting demands for water use and alternative ways of
meeting various demands.
The techniques for evaluating flow regulation effects on flood
damage prevention, hydro-power generation, and irrigation in com-
parable terms have been reasonably well developed. Evaluation of
Weber
19
-------
flow regulation effects from a water quality standpoint presents many
more problems. When stream flows are regulated in the interests of
water quality, many types of effects may be pertinent. Some of these
are removal of sediment during impoundment and the resulting
changes in conditions downstream; dilution of acids, alkalis, salts,
and poisons; heat dissipation, which affects the amount of oxygen;
improved pH and its effects on the speed and character of biological
reactions; reduced costs in treatment of domestic water supplies; im-
proved fish and wildlife habitat and sport and commercial fishing;
improved recreational potential of waterways; improved quality of
domestic, industrial, and irrigation supplies; and reduced corrosion
damages to structures and floating craft. Although these end products
are deceptively simple in statement, we all recognize that literally
thousands of compounds may be introduced into rivers from our
municipal, industrial, and agricultural complexes and that these, in
combination with many possible patterns of stream use, produce a
wide variety of situations that should be considered in the evalu-
ation of flow regulation effects. Experts in practically every area of
human activity will have to contribute and work together to estab-
lish the parameters of the physical effects from water quality control
and provide a basis for evaluation of net benefits in monetary terms
for comparability with other project benefits. While some benefits
can be determined by existing procedures, many require research and
continued analysis. It is simply not enough to continue to use the
cost of alternative projects as a measure of water quality control
benefits, since, unlike power and municipal and industrial water
supply, we have no market-price-related basis for testing the eco-
nomic justification of such alternatives. We note that even though
evaluation and economic aspects are not being emphasized at this
Symposium it is contemplated that future meetings will do so. This
is essential if we are to make headway in incorporating water quality
control into water development programs such as those of the Corps
of Engineers.
RESEARCH
The emergence of water quality control as an increasingly im-
portant consideration in water resource planning and development
has brought about needs for research on many aspects of the problem.
For example, the differences in temperature and oxygen content of
reservoir releases compared with those of natural flows have effects
on fishery resources that we must learn more about so that we can
take measures to prevent adverse effects and produce improved con-
ditions if possible. Research is needed on qualitative and quantitative
effects on basic biological productivity, general limnology, fish popu-
lations, and countless other aspects of the use of streams for recrea-
tion and fish and wildlife. We need to know more about both old
and new substances being introduced into our rivers and their
effects on use of water in both old and new processes.
In the Corps of Engineers' program, we cannot hope to contribute
more than a fraction of the basic research that is obviously needed
20
RELATION TO CORPS OF ENGINEERS
-------
on matters related to water quality. We can investigate the possi-
bilities for changes in design of our dams to optimize the temperature
and oxygen content of water released in the downstream channels
or to provide multiple outlets to permit discharge of water from vari-
ous selected levels in the reservoir. We can. in appropriate cases,
monitor the dissolved oxygen content of water at various depths in
our reservoirs as we are now doing at the John H. Kerr Reservoir
Project on the Roanoke River, Virginia and North Carolina, As we
identify the most significant aspects from a water resource planning
and development standpoint, we are prepared to undertake and to
sponsor or support specific research activities- "by others as an essential
adjunct to our water resource development mission. We visualize as
particularly appropriate for our program those research activities that
will improve our ability to evaluate flow regulation effects and those
that will lead to more effective design of project features to ac-
complish beneficial control of water quality.
SUMMARY
In summary, water quality considerations have become a major
factor in the formulation of the water resource development programs,
which are a responsibility of the Corps of Engineers. In fact, present
iridicatioris are that flow regulation in the interests of maintaining
satisfactory water quality may be the dominant and controlling re-
quirement for reservoir storage in many areas in the next few decades
at least.
We have had a long-standing interest in this problem. For ex-
ample, our basic planning for the low flew regulation features of the
Allegheny Bawr Heservoir now under construction above Pittsburgh
was done in the 1938 to 1940 period with a major assist from the
Public Health Service. Over this same period the Ohio River Pollu-
tion Survey was completed through our joint efforts. This has pro-
vided a valuable background for the water management work —
quality and quantity wise — that has since been accomplished in the
Ohio Basin,
We are now striving to equip ourselves to understand more
about the significance of water quality in our economy, the effects of
changes of water quality, the effectiveness and value of measures to
control or improve it, and the interrelation of water quality to water
resource conservation and use in general.
One of the most significant effects of I his situation or the Corps
:: Engineers1 program is Iba; all planrurg must be carried out with
mulliple-piirpose needs and possibilities in mind. While we must
continue to solve and deal with many immediate and urgent problems
with whatever resources and data are immediately obtainable, we
must also relate these problems and their solutions as best we can to
all other existing and prospective needs for water. This has brought
about in recent months a concerted effort on our part to move towards
the earliest possible completion of integrated, comprehensive plans,
both in framework and in detail, for all of the Nation's liver basins,
Weber
21
-------
as recommended by the Senate Select Committee on National Water
Resources and as adopted as an objective by the President in his
policy pronouncements on water resources.
Perhaps the most significant development in water resource plan-
ning today is the growing realization that the water problem is so
complex that there is no way that all of the responsibility for dealing
with it can logically be assumed by any single agency or level
of government. We have adapted our procedures to facilitate working
with all agencies at Federal, State, and local levels, to achieve maxi-
mum possible coordination of water resource planning and develop-
ment. In so doing we are steadily gaining in our appreciation of
several types of interrelationships that are important to those con-
cerned with water resources.
First, through training of current employees and in new hiring,
we are equipping our own organization with personnel who encom-
pass the wide range of disciplines needed to work out solutions to water
problems. Our organization will include engineers who will try to
solve water quality problems, and also biologists, economists, law-
yers, and others needed to assimiliate the complex considerations in-
volved in modern water resource development plans.
Secondly, we are trying to work out appropriate and effective
arrangements for coordination and distribution of effort among all
agencies who can and should contribute to solution of water prob-
lems. Insofar as water quality considerations are concerned, we are
looking to the Department of Health, Education, and Welfare for
both day-to-day advice and continuing, concurrent consideration of
long-range studies.
Finally, we are convinced that exchanges of information and
ideas such as at this Symposium are becoming more and more neces-
sary for effective action on water problems. We know that our per-
sonnel who are participating will benefit and we plan to make the
proceedings widely available to our field offices. We hope that the
program of the Corps of Engineers can contribute significantly to
maintaining and improving water quality as well as to meeting other
needs for use of water, and we appreciate this opportunity to join
in efforts towards that end.
22
RELATION TO CORPS OF ENGINEERS
-------
RELATION OF KEGULATIOW FOR QUALITY
COXTttOL T
-------
States, th$ Water Supply Act of 1958 provided additional authority
for construction cf storage c^psi'irv fcr deferred use for this purpose.
The federal Water Pollution Control Act oi 1961 provides that
ia the survey and planning of any reservoir by the Ccrps t>f Er.gir.ters,
Bureau oi Reclamation, ar ether Federal agency consideration shaji
be give?} k> alar&ge !or regulation of streamflow for the
purpose of vvate.r quaJity contra]. It also provides that storage and
releases therefrom for water quality control sha.il not be provided
as a substitute for adequate treatment or other methods oi wnteolUng
wastes at their source.
There are mimercus specific project authorizations fox construc-
tion and speration of Reclamation projects, some of which include
works specifically for Abater quality control
The Bureau is making continuous studies of irrigation water re-
quirement, which Include consideration of water quality. We have
iearr.ed by some 6e adversely
affected for further use. In the planning of projects, it is therefore
essential that the quality of the potential water supply be carefully
investigated with respect to: (1) the nature and ennt-enir.afir't sf the
vri-y-.;.s_-is!_isis *2} js prc~«i>l2 iaecU ;f any, on the crops a.ntd
soils of the project under varying rates cf application s.r.d ecfid:ticiris
-------
eontroL structures on the oxygen content of the water to develop
methods of creating and maintaining a favorable oxygen content.
Many Reclamation projects supply municipal and industrial
water. The Bureau delivers raw water to the user, and although the
Burea j does not provide final treatment to make the water potable,
it does add chlorine, xvhen necessary, to prevent the growth of algae
in the conduits. The Cachuma Project ir, southern California is an
example of a project where chlorination is provided.
Sedimentation problems in connection with Reclamation projects
include loss of reservoir storage space as a result of sediment deposi-
tion, aggradation, and formation of deltas at the heads of reservoirs;
degradation of stream channels below reservoirs as the clear -water
picks up its natural sediment load; and deposit of silt behind diver-
sion dams, and in canals, siphons, and other water control structures.
These problems are under continuous study, as are those of stability
of both natural and artificial channels required to transport water.
Studies of sedimentation as a special subject are relatively new
in the field of engineering. Evaluation of many of the sedimentation
problems 011 early projects was based solely on judgment. As the
Bureau and other governmental agencies have gained experience in
operating projects and have accumulated data on sedimentation, the
practical operating knowledge acquired ha5 made it possible to im-
prove the techniques for evaluating the sediirver.tauon aspects in
project planning and development. Two outstanding accomplish-
ments of the Bureau in this continuing work include the desilting
works of the Imperial Dana on the Colorado and the recent program-
ing of the complicated Modified Einstein Procedure for computing the
total sediment transport of a stream.
Most Reclamation reservoirs are used for recreational purposes,
and the regulation of this use requires consideration of water quality.
The waters must meet quality standards to support such uses as
boating, swimming,, fishing, picknicking (drinking and cooking),
wildfowl nesting, and various other uses.
The Bureau is conducting research on the use of monomolecular
films to reduce evaporation losses ivoitv reservoir surfaces. A number
of other agencies are cooperating. These studies, especially on new
materials, include consideration of possible toxicity to fish, migratory
waterfowl, aquatic plants, and the various animals inhabiting reser-
voir areas. Any undesirable change in physical or chemical character-
istics, such as taste, odor, or color, is. not permissible in water for con-
sumption by humans or livestock; but studies to date have failed to
indicate any harmful effects.
It would be well, at this point, to mention several of the specific
problems now under study in certain areas as they relate to the
Bureau.
The Bureau is constructing the Spring Creek Debris Dam in Cali-
fornia to control polluted mine waters during periods of low flow in
Burnett
25
-------
the Sacramento River. These polluted waters will be released only
when sufficient dilution is available, to preserve the salmon runs in
the river.
The Bureau is constructing a project in the Malaga Bend section
of the Pecos River in New Mexico to intercept and divert flows of
salty spring water to an impervious offstream reservoir. There the
water will be evaporated. Interruption and diversion of the saline
water will improve the quality of the water in the main stream. A
similar situation exists at the potential site of the Dixie Project on
the Virgin River in Utah. Since no suitable storage site has been
found, the Bureau is tentatively planning to investigate treatment of
contaminating water by salt water conversion processes.
The investigation of the Delta Division of the Central Valley
Project in California includes practically the entire gamut of water
uses and controls — municipal and industrial water uses, irrigation,
drainage, navigation, stream pollution abatement, salinity repulsion,
fish and wildlife conservation, recreation, and even limited flood
control. The purpose of this investigation is to develop plans and
establish what facilities are needed to improve the quality of water
available in the delta area, to sustain and enhance existing develop-
ment, and to provide for continued economic growth. Water quality
control has not only been one of the controlling guidelines in the
planning of the Central Valley Project, but is a control on project
operations. This work is being carried out cooperatively with other
Federal agencies, State agencies, and local people. Studies involve
protection and enhancement of local fisheries and wildlife, as well
as the necessary studies to determine desirable levels of pollution
control as related to public health, recreation, and control of nuisance
conditions.
Reservoir operations at Folsom Dam have been modified and re-
medial construction measures taken involving the modification of the
river outlet to provide suitable water temperatures in the interests
of the American River fishery. The Central Valley Project, in pro-
viding supplemental water, also has assisted with improvement of the
water quality in many ground water pumping areas.
In our work in the field of water quality, our experiences work-
ing with the Fish and Wildlife Service, the Public Health Service,
and the National Park Service have been of paramount benefit. These
agencies have contacts and sources of information not readily avail-
able to the Bureau. They also have biological knowledge outside the
normal experience of our Reclamation personnel. The use of these
sources by our planners has allowed us to progress in water quality
control work at an accelerated pace.
The Bureau is conducting extensive water quality studies in the
Colorado River Basin, in cooperation with the Geological Survey. In
addition, the Imperial Irrigation District collects water quality data
on the A11-American Canal; and the Salt River Water Users Associa-
tion and the city of Phoenix are collecting water quality data on the
26
RELATION TO BUREAU OF RECLAMATION
-------
Verde River and the canals and wells of the Salt River Project. The
Bureau is collecting water quality data from wells of the Gila Project
and from Lake Mead.
Work is also proceeding on special studies, such as the one in the
Eden area of Wyoming, where samples are being taken every 10 days
at inflow and outflow points to characterize the effects of irrigation
on return flows. There is another study underway in the river reach
between Wanship Reservoir and below Echo Reservoir to character-
ize the effect of storage on quality of water. Although this area is out
of the Upper Colorado River Basin, data obtained here are expected
to be useful in the Upper Basin water quality studies. Samples are
being obtained at all sites where possible transmountain diversions
might be made out of the Upper Colorado River Basin.
In connection with the development of the Colorado River Stor-
age Projects of the Bureau, a cooperative study is underway with
the Geological Survey to evaluate the effect of storage and consump-
tive use of water in the Upper Colorado River Basin on the quality
of waters of the Colorado River.
Colorado River studies include determinations of discharge, total
salt concentration, and ion concentration relationships at key sites
within the Upper Basin in addition to a quality of water operation
study involving the four storage units and the eleven participating
projects. The operations will be extended downstream to determine
the quality of the releases from Glen Canyon Reservoir. As additional
projects or units are authorized, they will be incorporated into the
quality of water operations study.
Corollary studies will be conducted for the purpose of developing
data in support of the assumptions used in the operation study. These
will include: (1) determination of salt balance conditions on presently
irrigated areas in the basin; (2) study of equilibrium conditions of the
soils and irrigation water in presently irrigated areas; and (3) study
of leaching or salt accumulation in various irrigated areas of the basin.
Reports scheduled from 1962 through 1975 will present the
current status of chemical quality of water and predict future effects
of developments in the Upper Basin.
No subject is of greater interest or importance to the people of
the West than conservation of water. The Nation's water problem
cannot be one of total quantity because we withdraw only about one-
third of the total manageable supply and then actually consume less
than one-tenth of that. Therefore, the national problem stands forth
as one of management. Management of water involves getting the
desired amount of water of the required quality to the point of use
at the right time. Water use in most of the watersheds of the West
exceeds the total supply; the same waters are used for irrigation, in-
dustrial supply, recreation, power, navigation, and many of tke aflMr
multiple-purpose uses. Each use has its «flect on water quality, «nd
regulation of such waters must include OTNiaation of quality at wwli
Burnett
27
-------
point of demand, with proper consideration given to the later demands
to be placed upon the waste and return flow waters from that point
of demand. The Bureau of Reclamation, in cooperation with other
interested State and Federal agencies, as well as local entities, con-
tinues to give such consideration in the planning, construction, and
operation of its many projects. As we look to the future, we can fore-
see additional demands for increasing studies of water quality, such
as provisions for the disposal of radioactive wastes, increased radio-
activity in surface and ground waters from the effects of fallout, dis-
posal of waste byproducts of new processes and materials, and the
development of new management techniques resulting from more
sophisticated computer analyses of water quality studies.
Aii Bureau of Reclamation projects now being planned give
prime consideration to water quality control as a potential project
purpose under the provisions of the Federal Water Pollution Control
Act.
28
RELATION TO BUREAU OF RECLAMATION
-------
THE INFLUENCE OF CONSERVATION PRACTICES
AND MEASURES ON WATER QUALITY
H. O. Ogrosky
Hydrology Specialist, Soil Conservation Service
U. S. Department of Agriculture, Washington, D. C.
The Soil Conservation Service (SCS) is responsible for develop-
ing and carrying out a national program for the conservation of the
Nation's soil and water. Its primary objective is the establishment of
conservation treatment measures and patterns of land and water use
that are compatible with the capabilities of our land and water re-
sources.
To accomplish this objective, the Service employs scientists and
technologists from numerous disciplines. The knowledge and skills of
agronomists, biologists, engineers, geologists, hydrologists, economists,
soil scientists, and others are pooled to diagnose problems and de-
velop plans for the use and treatment of our Nation's farms, ranches,
watersheds, and river basins.
From its inception, the SCS recognized the necessity of local
participation and control in carrying out action programs. There-
fore, whether farms or watersheds are to be planned, the SCS pro-
vides assistance to individuals and local groups that want to develop
and apply conservation programs.
FARMS AND RANCHES
The types of measures applied to the individual farm or ranch
vary widely, depending on such factors as climate, topography, and
soils, and on the desires or interests of the farmers or ranchers. There
are over 100 items listed in the "National Catalog of Practices and
Measures used in Soil and Water Conservation." Most of these prac-
tices and measures will directly or indirectly influence the quality
or quantity of the water yield of an area. Although a practice may
be designed for the primary purpose of controlling erosion, it may
produce other benefits of significant value. For example, it is well
known that soil particles are detached by the energy of falling rain
drops. When vegetative cover protects the soil from the impact of
rainfall, erosion is reduced. But there are other effects of establishing
good vegetative cover, such as improving the condition of the soil.
An improved soil allows more water to infiltrate the soil surface. It
makes more water available for plant growth and contributes to the
underground supplies that feed our springs. It takes a little longer
for the water to reach a river by percolating through the soil, but
normally this is desirable.
Runoff from bare fields or bare watersheds occurs quickly and
may result in a flash flood of muddy water. Runoff from well-
vegetated fields and watersheds occurs more slowly and is not heavily
ladened with silt. Therefore, a simple measure such as the establish-
Ogrosky
29
-------
merit of vegetation for erosion control also tends to regulate stream-
flow. The installation of only a few acres of vegetation does not pro-
duce a significant effect in a watershed. But, if the vegetative cover
is improved on all of the farms within a watershed and if additiona
practices and measures such as strip-cropping, terracing, contour
tillage, etc., are installed, the effects of all these practices and meas-
ures acting together do have some influence not only on the rate o
runoff but also in reducing the sediment content and improving the
quality of the water produced by the watershed.
WATERSHEDS
In addition to the measures on individual farms and ranches,
the SCS assists local groups in developing project plans for upstream
watersheds. Each watershed is different and each plan is developed
to solve the local problems. A project may be designed to reduce
upstream floodwater damage or to provide for agricultural water
management. It may include measures to provide a water supply f°r
industrial, municipal, agricultural, or recreational use. Some degree
of flood prevention has been included in most of the watershed
projects planned to date. This usually involves the construction of
earth dams that temporarily store flood runoff and release it at a
relatively low uniform rate. These structures have a major influence
in regulating the flow below the dam. But in addition to land treat-
ment practices and measures, there are usually some other benefits
that result from the construction of floodwater retarding structures.
For example, even though these structures are not designed to trap
the sediment load, it is impossible to store the floodwaters even
temporarily without reducing the sediment load of the discharge
water to some extent. Since some of the sediment will be trapped by
the reservoir, additional storage is provided to permit the reservoir
to operate at full capacity for a period of 50 to 100 years. Often
structures are designed for storage of water for irrigation, municipal
or industrial water supply, or recreation, in addition to the temporary
storage provided for flood prevention. When permanent storage is
provided in a structure, it becomes even more effective in removing
some of the sediment being transported downstream.
Not all measures installed in watershed projects produce the
same kind of benefits. It may be necessary to reduce flood damages
by improving the existing channel. This could result in faster flowing
streams and greater flood peaks downstream. If small streams are
involved which outlet into lakes or major rivers, the flood effect may
not be significant; but channel improvements often must be limited in
scope or the effects of increased peaks must be offset by providing
structural storage in the watershed. This is pointed out to make clear
that all measures recommended by the SCS do not produce the same
kind of effect on streamflow.
Just as each farm is different and needs a conservation plan
prepared specifically for that farm, each watershed is different and
needs a watershed program prepared specifically for that watershed.
30 INFLUENCE OF CONSERVATION ON WATER QUALITY
-------
There are exceptions, therefore, to almost anything that is said about
the effect of watershed programs. It is safe, however, to say that
overall a complete watershed program will reduce the rate of runoff,
i.e., reduce flood peaks and the volume of sediment produced by the
watershed. Reducing the peak flow normally permits a greater use
of the total water yield of the watershed. Improving the quality of
water by reducing the sediment content, of course, produces a range
of benefits, depending on the use of the water supply.
RIVER BASINS
There is one other activity that should be considered in a dis-
cussion of the influence of conservation practices on water quality.
This is the planning of large watersheds or river basins. As our
population increases, it is necessary to expand the development of
our natural resources to meet the increased demands for products
of all kinds. Unfortunately, these resources are not distributed uni-
formly within major river basins nor does population expansion or
urban development take place uniformly everywhere. In basins where
our resources are limited or where there has been tremendous urban
growth, we already have serious competition for our land and water
resources. The SCS, therefore, participates with other Federal and
State agencies in planning major river basins to solve specific prob-
lems. The SCS has participated in studies of numerous river basins
to evaluate the effects of conservation practices and measures on flood
peaks, flood volumes, sediment loads, etc., but the recent Potomac
River Basin study is the first to emphasize the needs of upstream
storage for water quality control. It is interesting to note that pre-
liminary estimates of the cost of upstream storage were assigned as
follows: recreation, 2 percent; water supply, 8 percent; water quality
control, 40 percent; and flood prevention, 50 percent. In the Potomac
study, therefore, a very significant portion of the upstream storage
was recommended solely for the purpose of water quality control.
Let me summarize the activities of the SCS in this way. First, the
Service cooperates with soil conservation districts to assist individual
land owners and operators. Practices and measures are installed on
their land primarily to conserve soil and water on the farm. There-
fore, measures such as terraces are designed to produce a benefit to
the farm unit, although it is recognized that these practices and
measures may produce some very desirable effects downstream.
Second, the SCS cooperates with local organizations in planning
and applying conservation measures on a watershed basis. This work
encompasses all of the measures recommended for conservation pur-
poses on individaul farms, and in addition, includes measures re-
quired to solve the problems of the watershed organization. Such ad-
ditional measures might include a multiple-purpose structure to
provide irrigation water and flood prevention or it might consist of
a system of ditches to provide improved drainage. These measures
are designed to provide "on-farm" and "community" benefits, al-
though it is also recognized that many benefits will accrue to down-
stream interests.
Ogrosky L,t,r9ry 31
Pacific Northwes! Water Laftorator*
200 South 35;
-------
Third, the SCS cooperates with State and Federal agencies in
planning the development of the natural resources of major nv®r
basins. A comprehensive plan for a river basin would include the
practices and measures to be installed on individual farms ana
ranches and the system of structures, channels, or other works re-
quired in specific watersheds. It would also include the additional
on-farm or small watershed practices and measures that need to be
installed upstream to provide the level of protection or produce the
desired benefit in the downstream areas of major river basins.
PRESENT AND FUTURE BENEFITS
Since the effect of each farm or watershed program is different,
let me review some results and examine some expectations.
On two Texas watersheds that are farm-sized — that is, about
150 acres in size — it was found, by application of flow duration
curves to sediment yield data, that the untreated watershed had
nearly nine times more sediment moving past its streamflow station
than did the conservation treated watershed^)- T^is can be con'
sidered a typical comparison for watersheds of this size, although,
of course, the proportion of sediment yield varies somewhat across
the country.
On larger watersheds such as those above the Loch Raven and
Prettyboy Reservoirs, which are water supply sources for the City
of Baltimore, the sediment yield is also reduced when conservation
measures are applied in significant amounts. Installation dates and
sediment surveys in these two reservoirs show that for Loch Raven
Reservoir there has been a 70 percent reduction in reservoir sediment
deposits during recent years, while for Prettyboy Reservoir, which is
12 miles upstream from Loch Raven, there has been a reduction of
44 percent(2,3).
In Georgia, records from five of the larger waterworks show a
steady downward trend in turbidity of the water being pumped from
the rivers for treatment at the filter plants. For the Chattahoochee
River at Atlanta, for example, average turbidity has decreased from
over 300 ppm in the early 1930's to less than 100 in the early 1950's.
The decreases in turbidity are very plainly attributable to the in-
creases of conservation measures on these river watersheds(4).
Let me turn now from results on a watershed basis to what can
be done on a large river basin basis. Studies of watershed protection
projects indicate that results shown in Table 1(5) can be achieved
in major basins. In the New England area, for example, where present
sediment concentration is very low, a reduction of only 6 percent can
be expected by 1980. In the Lower Mississippi, however, where the
present sediment concentration is very high, a reduction of 30 percent
can be expected by 1980. How are these benefits going to be produced
nationally? Here are some ideas.
In river basin projects where the SCS has investigated and recom-
mended storage for water quality control to meet the requirements
32 INFLUENCE OF CONSERVATION ON WATER QUALITY
-------
TABLE 1. FACTORS INFLUENCING SEDIMENT CONCENTRATION IN MAJOR DRAINAGE BASINS
Predominant physical characteristics Sediment concentration
Basin
Climatic type
Land form
Relative
credibility
of SOlis
Major erosion
source of
sediment
Present
relative
rate
Estimated
reduction.
By By
19&0 2000
New England
Humid
Rolling
Low
Sheet
Very low
6
10.
Delaware-Hudson
Humid
Rolling
Low
Sheet
Low
6
10
Chesapeake Bay
Humid
Rolling to dissected
Low
SheeL
Medium
21
:i5
Southeast
Humid
Dissected to low rolling
High
Sheet, channel
High
36
60
Eastern
Great Lakes.
Humid
Rolling
Low
Sheet
Low
6
10
Western
Great Lakes
Humid
Rolling
Low
Sheet
Luw
ti
10
Ohio
Humid
Rolling
Medium
Sheet
Medium
12
20
Cumberland
Humid
Dissected
High
Sheet
Low
6
10
Tennessee
Humid
Dissected
High
Sheet
Low-
6
10
Upper Mississippi
Humid to subhumid
Low rolling to dissected
High to low
Sheet
High
21
35
Lower Mississippi
Humid
Dissected to low rolling
High
Sheet, channel
Very high
AO
50
Lower Missouri
Humid
Dissected to low rolling
High
Sheet
Very high
24
40
Lower Arkansas,
White, Red
Humid
Dissected to low rolling
Medium
Sheet
High
33
55
Upper Missouri
Subhumid to semj-
and
Low rolling lo mountainous
High to low
Sheet, channel
High to low
33
55
Upper Arkansas,
White, Red
Subhumid
Rolling
High
Sheet, channel
High
45
75
Western Gulf
Subhumid to semi-
arid
Rolling
High
Sheet
High
30
50
Rio Grande-Pecos
Semiarid to arid
Dissected
Medium
Channel
High
21
35
Colorado
Sermarid to arid
Dissected to mountainous
High to low-
Channel
High
18
30
Great Basin
And
Mountainous to plains
Medium
Channel, sheet
Medium
12
20
Pacific Northwest
Superhumid to arid
Mountainous
Low to high
Sheet, Channel
Low to high
9
15
Central Pacific
Superhumid to arid
Mountainous
Low to high
Channel, sheet
Medium
IB
30
Southern Pacific
Subhumid to arid
Mountainous
Law to high
Channel, sheet
Medium
18
30
-------
set forth by the Department of Health, Education, and Welfare, it is
expected that funds will be made available to the SCS to construct
the dams if the projects are approved by the Congress.
Under the watershed program the SCS can cooperate in the
planning of reservoirs to provide storage for water quality control.
At the present time, the SCS cannot provide funds for construction
of such reservoirs. Amendments are being considered that would
permit the Federal government to share with local agencies the cost
of providing such storage.
It must be kept in mind that the work of the SCS does not in-
clude spectacular engineering structures. SCS dams are limited under
PL-566 to 25,000 acre-feet of storage. But the volume of the work
is an indication of the influence these dams can have on water quality
control. As of January 1963, the SCS had received over 1800 appli-
cations for assistance in watershed planning; planning had been com-
pleted on over 800 watersheds; and programs had been authorized
on about 450 watersheds. Work is currently underway on slightly
under 300. Through fiscal year 1962, preliminary plans had been com-
pleted on somewhat over 2600 dams and construction had been
completed on about 650. In one state alone the applications pending
include the development of municipal water supplies for 53 towns
and cities.
The Department of Agriculture's inventory of conservation needs
indicates a need for watershed programs on 8300 watersheds in the
United States. Work has been completed on less than 1 percent of
this number. Construction has been started on about 3 percent, is
authorized on over 5 percent, and planned on over 8 percent. These
percentages may appear small when expressed in terms of the total
job; however, progress has been good, if we consider the fact that the
program was authorized in the fall of 1954, a certain amount of
tooling up was required to handle the work, and, of course, the re-
quests for assistance must be developed locally and cleared through
State and Federal channels before actual field work can be started.
Much more conservation work needs to be done, but enough has
been accomplished to prove the effectiveness of the various practices
and measures in the conservation of our soil and water resources.
This work is moving forward from approximately 3000 offices located
throughout the United States. The effects of these efforts have been
obvious on small areas such as farm units and small upstream water-
sheds. More and more evidence is being found to show that very
significant benefits are also occurring on large watersheds.
The cooperative effort of Federal, State, and local agencies in the
installation of conservation practices and measures on watersheds is
an effective means of achieving flow regulation in upstream areas
and silt pollution control. Water quality is being improved but pri-
marily as a byproduct of work to accomplish other objectives. If an
" >rt was made to determine upstream storage requirements for
lution abatement and to secure the necessary legislative amend-
INFLUENCE OF CONSERVATION ON WATER QUALITY
-------
ments to permit Federal-local sharing of construction costs, a much
greater contribution could be made by the SCS to streamflow regula-
tion and water quality improvement.
REFERENCES
1. Baird, R. W. Effect of conservation practices on sediment con-
centration and yield. Presented at Sedimentation Workshop,
sponsored by Agri. Res. Service and Soil Conserv. Serv., USDA,
Panguitch, Utah, Sept. 1962. 11pp.
2. Gottschalk, L. C. Report on the sedimentation surveys of Loch
Raven and Prettyboy Reservoirs, Baltimore, Maryland. Soil
Conserv. Serv. Spec. Rept. No. 5. USDA, Washington, D. C.,
1943. 22pp.
3. Koleman, J. N. The sedimentation of Loch Raven and Prettyboy
Reservoirs, Baltimore County, Maryland. Soil Conserv. Serv.
Tech. Publ. 145. USDA, Washington, D. C., Feb. 1965. 18pp.
4. Albert, F. A., and A. H. Spector. A new song on the muddy Chat-
tahoochee. In: Water — The Yearbook of Agriculture, 1955.
USDA, Washington, D. C., 1955. pp. 205-10.
5. Water resources activities in the United States. Estimated water
requirements for agricultural purposes and their effects on water
supplies. U. S. Senate. Select Committee on National Water
Resources. 86th Congress, 2d session. Committee Print No. 13.
1960. 24pp.
Ogrosky
35
-------
RELATION OF REGULATION FOR QUALITY
CONTROL TO THE ACTIVITIES OF THE
FEDERAL POWER COMMISSION
F. Stewart Brown
Chief, Bureau of Power
Federal Power Commission, Washington, D. C.
The Federal Power Commission is vitally concerned with the
comprehensive planning of river basins. Under statutes dating back
to 1920, the Commission has been given important responsibilities in
such planning activities. In fact, an early milestone of importance in
basin planning is the requirement of Section 10(a) of the Federal
Power Act. This section requires that to be licensed a water power
project must, in the judgment of the Commission, be best adapted to
a comprehensive basin plan.
I am pleased, therefore, to be able to discuss with you the rela-
tionship of the Commission's activities to the increasingly important
function of streamflow regulation for water quality control. The
views expressed, however, are my own and are not necessarily those
of the Commission.
Briefly, the Commission has two major activities relating to
water resources planning and development. The oldest of these is
its licensing function, under which the Commission issues licenses
to individuals, corporations, states, and municipalities authorizing
the construction, operation, and maintenance of water power projects
on government lands and on streams over which the Congress has
jurisdiction. It may also license these non-federal interests to utilize
the surplus water or water power from a government dam.
At the end of fiscal year 1962 there were licenses in effect for
268 major projects with some 16 million kilowatts of existing in-
stalled capacity and 25 million kilowatts of ultimate capacity. The
existing capacity under license represents 74 percent of all non-
federal hydroelectric capacity in the United States. The ultimate
installations under major licenses involve a claimed or estimated
cost of about $5.5 billion.
The Commission's second major activity relating to water and
related land resources planning is its comprehensive river basin
studies carried on generally in cooperation with the federal con-
structing agencies. The principal role of the Commission in such
cooperative investigations is to furnish advice on electric power
matters. This involves estimates of the amount and value of hydro-
electric power that can be developed in connection with proposed
plans, estimates of the future market for such power, and recom-
mendations of provisions for developing the power potential.
An important result of the Commission's river basin studies, made
both in connection with the licensing function and in cooperation
with the other federal agencies, is the nation-wide inventory of water
Brown
37
-------
power resources. Data on individual developed and undeveloped
hydroelectric power projects are compiled annually and published
periodically. As of January 1, 1963, preliminary estimates of the
total hydroelectric resources of the United States were 150 million
kilowatts of generating capacity capable, with average flow in the
streams and rivers, of producing 640 billion kilowatt-hours of electric
energy a year. About 38 million kilowatts of the total capacity was
in existing hydroelectric developments that generate an average of
about 200 billion kilowatt-hours annually. It may be noted that the
38 million kilowatts of developed hydroelectric capacity accounted
for about 18 percent of the total electric-generating capacity in the
United States.
Although the Commission makes studies of water resources
projects and keeps abreast of new developments in the fields of power
generation, transmission, and distribution, it does not carry on re-
search as such. This, of course, stems partly from the fact that the
Commission is not a constructing agency.
In our consideration of streamflow regulation for water quality
control, it should be noted that such regulation may have important
effects on hydroelectric power development. Storage included in a
reservoir for water quality control may provide head usable for
power development at the reservoir site. The streamflow regulation
may also increase the firm power output at the site and at hydro-
electric power plants downstream.
Although storage for water quality control may have important
effects on hydroelectric power development, the Federal Power
Commission is more often concerned with the effects the construction
of hydroelectric projects may have on water quality control. Storage-
reservoirs at hydroelectric projects are usually operated to supple-
ment natural flows during low flow periods and thus increase the
production of firm power. Such streamflow regulation may contribute
to water quality control by increasing natural minimum flows; how-
ever, whether such benefits are in fact provided depends upon how
the hydroelectric power plant is operated. A hydroelectric plant at
a storage reservoir usually is most valuable when operated as a peak-
ing plant. In a hydroelectric plant so operated, maximum water dis-
charges are made at times of peak power requirements. Optimum
operation of the plant in the interest of power would probably dictate
that no water be released during off-peak periods of power demands.
Obviously such operation might be detrimental to water quality
control. In some cases, in order to make the peaking power opera-
tions compatible with water quality control, a re-regulation reservoir
is constructed downstream from the power plant. The re-regulation
reservoir permits unlimited daily or weekly peaking operations of
the upstream power plant and makes possible uniform water releases
in the interest of water quality. In other cases minimum water re-
leases from the storage reservoir are prescribed. Sometimes, in such
cases, a small generating unit is installed to utilize and discharge the
required minimum flows during periods when the main generating
units are not operating.
38
RELATION TO FEDERAL POWER COMMISSION
-------
In issuing licenses for hydroelectric projects, the Commission,
acting under the authority of the Federal Power Act and other acts,
takes a number of actions to ensure proper consideration of the view-
points of various affected interests and of the needs for such purposes
as navigation, flood control, irrigation, fish and wildlife conservation,
recreation, public health, and water quality control.
The Commission notifies interested members of the Congress;
Governors; and Federal, State, and local agencies of applications for
preliminary permits and licenses for hydroelectric projects and gives
them an opportunity to submit comments. Notices of filings of appli-
cations are published in local newspapers and in the Federal Register.
In addition, before acting on an application, the Commission requests
comments from the Secretaries of the Army, Agriculture, Interior, and
Health, Education, and Welfare on the project as it affects their inter-
ests, and if other departments or agencies of the government are in-
volved, their views are likewise requested. Any party who believes
that his interests may be affected by a proposed project may submit
suggestions or protests to the Commission; and hearings in connec-
tion with an application may be ordered by the Commission, either
upon its own motion or upon the motion of any party to the pro-
ceeding.
Plans submitted as part of an application for license are reviewed
critically by the Commission's engineering staff to make sure that the
design will result in project works that are safe and are adequate to
develop properly the resources involved. Field supervision to ensure
compliance with license conditions, both during construction and
after a project is in operation, is assigned to one of the Commission's
five regional offices or to another authorized government agency,
specified in the license.
Several licenses include provisions for minimum flow releases or
other operating requirements in the interest of water quality control.
A few specific examples will illustrate such requirements.
The license for Project No. 2009, in Virginia and North Carolina,
issued to the Virginia Electric and Power Company, covers the Gaston
development now under construction and the downstream Roanoke
Rapids development now in operation. Both developments are on the
Roanoke River downstream from the federally constructed John H.
Kerr reservoir and power project. Immediately downstream from
the Roanoke Rapids development the river is polluted by discharges
of industrial wastes from the plant of the Halifax Paper Company
and sewage from the towns of Weldon and Roanoke Rapids.
Article 25 of the license for Project No. 2009 prescribes mini-
mum flow releases from the Roanoke Rapids pond in the interests
of the preservation of fishery resources and stream sanitation. It also
prescribes the minimum daily dissolved oxygen content to be
included in the flow releases. These provisions were determined
after a public hearing was held, particularly on the issue re-
specting the effect, if any, of the construction and operation of the
proposed Gaston project on the quality of water in the Roanoke
Brown
39
-------
River. The licensee has constructed submerged weirs above the power
intakes of the Gaston and Roanoke Rapids developments in order
to skim the higher quality water, particularly with respect to the
oxygen content, from the upper layers of the respective pools. It had
been found that the oxygen content of water released during certain
times of the year from the upstream John H. Kerr reservoir was less
than that required for sustaining fish life, and it was not anticipated
that the water quality would improve with normal passage of water
through the Gaston and Roanoke Rapids ponds and power outlets.
The license for Project No. 2232, issued to the Duke Power Com-
pany, covers 10 existing developments and one development, Cowans
Ford, under construction on the Catawba and Wateree Rivers in
North Carolina and South Carolina. This series of developments oc-
cupies a 215-mile stretch of river except for two undeveloped reaches
aggregating about 40 miles. Pursuant to an agreement between the
licensee and North Carolina State interests, the licensee has con-
structed a submerged weir at the Cowans Ford development in order
to skim the higher quality water from the upper layers of the
reservoir. It has been found that in a series of dams and reservoirs
without intervening stretches of free flowing stream the quality of
water in releases from each succeeding reservoir becomes lower,
particularly with regard to the oxygen content. The agreement also
provides for the licensee to make minimum water releases at the
six developments in North Carolina in the interests of stream sani-
tation, fish and wildlife conservation, and other beneficial purposes.
These minimum stream flows are prescribed in Article 31 of the
license. Minimum average daily releases from the five developments
in South Carolina are covered in Article 32. This article provides
that the licensee shall cooperate with the South Carolina Water
Pollution Control Authority to establish a mutually agreeable
schedule of minimum daily flow releases from the five developments
in South Carolina. During construction of the Cowans Ford dam,
the licensee received Commission approval to install an intake struc-
ture that could be used to take condenser cooling water from the
reservoir for a future steam-powered electric-generating plant with
ultimate installation of 2,000,000 kilowatts. The water would be re-
turned to the reservoir above the dam. The effect of this future use
on water quality is not known at present.
The Turlock and Modesto Irrigation Districts have filed an ap-
plication for license for the New Don Pedro Project No. 2299 on the
Tuolumne River in California. The purposes of the project are irriga-
tion, flood control, power, and exchange storage for the City of San
Francisco. The California Department of Fish and Game has recom-
mended inclusion of a provision in the license that the project shall
be constructed and operated in a manner that provides water tem-
peratures below 56°, as measured at the La Grange bridge, during
the period of October through March. This is the period for spawning
and incubation of chinook salmon in the Tuolumne River. At a hear-
ing held in October 1962, testimony was presented on temperatures
necessary for the preservation of fishlife and on the temperature of
40
RELATION TO FEDERAL POWER COMMISSION
-------
the water that would be discharged from the reservoir during the
critical and average water years. A decision on this matter has not
yet been reached.
When appropriate, other types of special conditions are included
in licenses in the interests of various water resources development
purposes or for the protection of the public. For example, several
licenses have required clearing of certain zones in the reservoirs in
the interest of public health. Licenses for a number of projects require
that the licensees take such measures for the effective prevention of
the breeding of malarial mosquitoes in the reservoirs as may be
required by the Commission, by the Commission and the affected
State health agencies, or by the United States Public Health Service.
In one case, the licensee is specifically required to place oil or other
materials, as recommended or approved by the U. S. Public Health
Service, in the shallow reaches of the reservoir to prevent the prop-
agation of mosquitoes.
Licenses have been issued for several projects that also include
storage for municipal water supply. Other licenses cover projects
that provide storage or other facilities for irrigation, in addition to
the power development. In one case a licensed project is operated
to re-regulate the flows from an upstream Federal reservoir in order
to ensure sufficient streamflow downstream for proper disposal of
municipal sewage plant effluent. In two cases the licenses specifically
permit the diversion of water from the reservoirs for condensing
purposes in steam-electric plants.
In addition to the relationship of hydroelectric power develop-
ment to water quality problems, there is also an important relation-
ship between thermal-electric plants and water quality. This, of
course, is the matter of thermal pollution of streams as the result of
using stream flows for condensing purposes. With the construction
of increasingly larger amounts of thermal-electric capacity, including
nuclear units, this problem will become increasingly important.
An indication of the significance of this problem is shown by a
recent Commission survey of the water requirements of the electric-
utility steam-electric plants in the United States. This survey, made
at the request of the Department of Commerce, showed that 29 trillion
gallons of water were used by such plants in 1959, or about 23 per-
cent of the total United States water withdrawals for all purposes that
year, according to information given in the January 1961 report of
the Senate Select Committee on National Water Resources.
Of the water withdrawn by electric-utility steam-electric plants
in 1959, 20.6 trillion gallons was fresh and 8.4 trillion gallons brackish,
primarily from the ocean. Of the fresh water, 15.4 trillion gallons was
withdrawn from rivers, 1.9 trillion gallons from the Great Lakes, and
3.3 trillion from other lakes and surface sources.
Since a portion of the 29 trillion gallons withdrawn was recir-
culated after passage through cooling towers or spray ponds, the
Brown
41
-------
total equivalent use of water amounted to 34 trillion gallons. This
is equal to 61.7 gallons per kilowatt-hour generated.
Cooling towers and spray ponds reduce the amount of water
required for steam-electric plants, but they increase the consump-
tive use of water. They are not used where water supplies are
adequate because they involve additional expense. In the 1959 survey,
plants with cooling towers or spray ponds accounted for 16 percent
of the power generation of all plants. Fresh water withdrawals at
these plants amounted to about 15 percent of what they would have
been without cooling towers and spray ponds. This gives a rough
indication of the potential for reducing fresh water withdrawals in
the future if it should become necessary to limit the water supply
to steam-electric plants.
Only about 0.3 percent of the 29 trillion gallons of water used
was actually consumed. The rest was discharged, generally to the
source from which the water was supplied. The temperature of the
returned water was raised some 12° to 13°F, but the water was other-
wise unaltered.
The importance of these factors is apparent when it is recognized
that the annual power generation in utility thermal-electric plants
is expected to increase from 568 billion kilowatt-hours in 1959 to
more than 2,000 billion in 1980. This estimate was made by the
Commission in connection with its National Power Survey now
under way.
42
RELATION TO FEDERAL POWER COMMISSION
-------
FLOW REGULATION FOR WATER QUALITY
CONTROL AND WATER RIGHTS
Murray Stein
Chief, Enforcement Branch
Division of Water Supply and Pollution Control
U. S. Public Health Service, Washington, D. C.
When I started preparing these remarks on the relation of water
rights to flow regulation for quality control, I had hoped that I could
continue the euphoria of opening session speeches. This paper would
not write that way, however. Water rights is a rather sticky subject
— a nasty one in many respects. The path we will be taking may
be very difficult and tortuous.
We should set the stage by referring to the basic law. Section
2(b), added by the 1961 amendments to the Federal Water Pollution
Control Act, provides that "storage for regulation of streamflow for
the purpose of water quality control" shall be considered in the
planning of Federal water resources projects. This is the key to my
topic — "for the purpose of water quality control."
No one appreciates more than I the difficulties incumbent upon
this task. The problem of releases from impoundments for water
quality control is the most difficult problem we have in the water
quality field. In the long run this may be the key to whether water
quality is controlled in this country or not. Flow regulation may very
well be the most important aspect of water quality control.
Let us assume for the purposes of this discussion that we have
secured adequate treatment of wastes at their source as required by
the law. Let us assume also that the hydrological problems are
solved, that the mathematical problems are solved, and that all the
other technical problems of releasing water at the right time to
achieve water quality control are solved. And here I recognize that
this is the subject of this Symposium and will no doubt take many
years to achieve — but let us suppose that these problems are solved,
that the water hai. been impounded, and that it now is being released
from the reservoir. After the water is released and flows down the
stream, it presumably becomes a part of the waters of the State, just
like any other waters.
Let us take an outrageous case first.
A farmer wants to irrigate downstream, so he pumps the water
from the river, puts it on his land, and the water does not get back
into the stream. Here, all the work, all the theory, all the operating
plans, all the Congressional objectives, all the interstate and inter-
agency agreements — all these things — go right down the drain
because the water is withdrawn. I think that this is the prototype
of the problem that we have to deal with and that we can have
many, many variations of the same thing.
Then there is the case of a water user who, though he does not
Stein
43
-------
withdraw water for a consumptive use, pollutes the water so that
stream quality is not improved by the flow regulation provided.
Another variant may be — and this is why water quality control
cannot be over-generalized or over-simplified — that of a public
water supply that is excessively mineralized. This was the case in
Dallas during the drought of the fifties. The water supply had such
high chlorides that it turned lawns brown and many people simply
could not drink the water. Water of low chloride content was brought
in and sold in milk cartons at $0.40 per gallon, which is more than
the price of gasoline. Would release of water to alleviate a situation
in a domestic water supply like that at Dallas be water quality
control?
These points are brought out to illustrate that case by case we
can have a complete spectrum of uses for quality control water and
that we certainly cannot provide answers to every question today
that may arise in the future. We can only dream of the answers.
The key point for this discussion, I think, is the first case cited,
that of someone such as the man engaging in irrigation farming. He
takes the water out of the stream, uses it to irrigate his crops, and
thereby thwarts the purpose of the Act to provide flows for water
quality control.
SYSTEMS OF WATER LAW IN THE UNITED STATES
In the field of water rights there has been a considerable amount
of controversy, as most of you know. We even have our own cliche —
that in the field of State and Federal water rights there is more heat
than light. Even in trying to give a dispassionate review of the situa-
tion — because of many people's legitimate bias — I may create more
heat than light. But even at this risk, I think we have to examine
the situation to determine what the facts are — what the background
is — before we can determine whether we can handle the water
rights problem in water quality control. I say this because in large
measure we are governed by history; as Justice Holmes said at one
time in a particular case, "In this case a page of history is worth
more than a volume of life." We have a tremendous history of con-
flict or of alleged conflict between Federal and State water rights
behind us, and we cannot divorce ourselves from it. Where we go in
this field may, however, very well be determined by legislation or
decisions outside the field of water quality.
This material is no doubt old hat to some, but to avoid the risk
of leaving unanswered questions for others, I will start by analyzing
the systems of water law in the United States. This is essential be-
cause, it seems to me, the critical conflict between State water
rights and Federal water quality control is going to emerge in the
western States.
In the 17 western States where water is not so plentiful, we have
only to compare the value of land with water rights with the value
44
WATER RIGHTS
-------
of land without water rights to learn the value of water and to
understand its critical importance. Because water was found only in
certain places and shortages were encountered almost from the time
of settling, in the western States the unique system of water law
known as the prior appropriation doctrine has developed.
Under the prior appropriation doctrine, priority of application of
water to a beneficial use gives one the right to use the water so ap-
propriated so long as the use is maintained. This system is often
characterized by the catch phrase, "First in time, first in right." To
illustrate, a junior appropriator located upstream cannot divert water
for his own use to the detriment of a senior appropriator downstream.
The doctrine of prior appropriation is written into several State con-
stitutions and defined by State laws. Generally, its administration
results in specific allocation of streamflow to water users under the
jurisdiction of a State administrator.
In the western States where the prior appropriation doctrine has
developed, the average annual rainfall is relatively low compared to
the humid East and the major appropriative use is agricultural irri-
gation, which is a highly consumptive use. It is estimated that two-
thirds of the water applied to crops does not return to the water
source. On the other hand, municipal and industrial use, which is
the large use in the East and is becoming increasingly more important
in the West, is generally nonconsumptive and more than 90 percent
of the water withdrawn is returned to the surface waters. In the 17
western States it is estimated that over 90 percent of the available
water is used for irrigation; in the 31 eastern States it is estimated
that less than 5 percent is used for irrigation.
The western States have embraced the prior appropriation doc-
trine, but laws vary considerably from State to State. Only seven —
Arizona, Colorado, Idaho, Nevada, New Mexico, Utah, and Wyoming
— have no elements of the riparian, or reasonable use, doctrine that
is prevalent in the East. Since the prior appropriation doctrine of
water rights deals with a commodity that is very precious in the
western States, it has been the subject of much litigation. In any
particular jurisdiction there are apt to be numerous judicial bench-
marks that delineate the rights, duties, and privileges of water users.
The riparian use doctrine, on the other hand, has been adopted
by the eastern and midwestern States. The riparian use doctrine was
embodied in the common law of England; it was practiced in the
early colonies and is presently applicable in modified form in all but
one or two of the eastern jurisdictions. The riparian doctrine is based
on the premise that running water is not susceptible to unqualified
ownership and that the right to use such water is incidental to prop-
erty ownership. Thus only riparian owners, that is, people who have
land abutting a watercourse or who have a watercourse running
through their land, are entitled to use of surface waters, or have
riparian rights. Under the riparian doctrine such persons have a
right to the use of the streamflow passing through their land in its
natural state, i.e., undiminished in quantity or quality.
Stein
45
-------
In the development of the riparian doctrine and in the resolution
of conflicting riparian rights, it soon became evident that the riparian's
right to receive the stream undiminished in quantity and, especially,
in quality is not unlimited. His right to the enjoyment of the stream
is subject to similar rights of all other riparian owners in their rea-
sonable enjoyment of the stream. Therefore, it is against the un-
reasonable use of this common benefit that legal action can be taken.
This modification of the riparian doctrine, enunciated in 1874 by
Justice Cooley of Michigan as the reasonable use doctrine, is the form
used by the midvestern and eastern States.
A corollary of the reasonable use doctrine is the right of the
riparian owner to be free of unreasonable interference in the use of
water. Reasonable use is the only measure of riparian rights and the
question of reasonableness is a question of facts. A characteristic of
the eastern system of water rights is the absence of frequent litiga-
tion. Water users have generally achieved an accommodation of their
needs through private arrangements and surprisingly few cases have
come before the courts. Although this has the advantage of providing
a flexible system of operation, without the guidelines provided by
numerous judicial decisions, it has left large areas of uncertainty.
Increasing and often conflicting demands of municipalities, industries,
and agriculture for water are making eastern water users acutely
aware of the lack of certainty of their water rights.
I would expect that in dealing with augmented flows provided
through flow regulation for water quality control we will continue
to have this uncertainly in the East. I would also expect that with its
tradition an accommodation will be reached in the East based on
negotiation rather than on legal action. In the 17 western States,
however, we are more likely to run into situations that will involve
legal actions. It boils down to the fact that there is more water in
the East and therefore more room for accommodation. In the West
water rights are essential to property rights and are therefore a
matter of life and death; every man is fighting for every acre-foot
of water he can get.
FEDERAL INTEREST IN WATER RIGHTS
Along with this understanding of where water rights conflicts
are likely to arise, we have to recognize that the Federal Government
is now in the business of water pollution control. The Federal interest
in water pollution control set forth in the Federal Water Pollution
Control Act, as amended, indicates that the clock is not going to be
turned back and that the Federal Government is in the water pollu-
tion control field to stay. The Federal interest arises for several
reasons. From the following list I think the cogency of the Federal
interest becomes apparent:
1. The relatively fixed fresh water supply. Give or take a wet or
dry cycle, the nation's fresh water supply is just about the same
now as it was when Columbus discovered the Americas.
46
WATER RIGHTS
-------
2. The uneven natural distribution of water geographically and
seasonally. There are 17 major river basins in the coterminous
United States. The 48 coterminous States certainly do not coin-
cide with the 17 major basins nor with the subbasins. Handling
water pollution problems on a river basin basis, therefore, natu-
rally brings in a national interest.
3. The rapid growth of population, industry, and agriculture —
often where water is not plentiful. The tremendous growth of
southern California where more than 10 million people depend
on an adjacent basin — the Colorado River — for their water
supply is an outstanding example.
4. The concentration of population and industry in urban communi-
ties and development of metropolitan complexes. The enormous
increases in water use by the rapidly growing population, ex-
panding industry) and intensified agriculture and the very large
increases in waste discharges resulting from population and in-
dustrial growth, new technologies, and changing land use prac-
tices are seriously jeopardizing the suitability of the nation's
water resources for needed use and reuse.
5. The failure to construct sewage and industrial waste treatment
facilities as fast as they are needed to provide adequate water
pollution control.
6. Lastly, the failure to develop means for dealing with new and
complex water pollution problems as they have emerged.
I think that the national legislature has recognized all of these
points in placing the Federal Government in the business of water
quality control.
COOPERATION IN QUALITY CONTROL
On many of these points, dealing with the States with regard to
water quality control has been relatively — and I emphasize that
word "relatively" — simple. Right now, for example, we have a
large water quality study going on in the Colorado River Basin.
California and Arizona are meeting amicably with five other States
and the Federal Government. California and Arizona are both inter-
ested in water quality control, despite their differences in the quantity
dispute. As a matter of fact, I can illustrate the difference in feeling.
During a meeting in Phoenix we needed a map of a particular portion
of the basin. A map was secured from the State files, but when it was
put up, a murmur arose throughout the room and would not subside.
On investigation we found that on the map there were numbers re-
lating to the distribution of water between Arizona and California.
Even though it was pointed out that the map was displayed only to
indicate where the river flowed, the people could not get their minds
off the numbers, and not until the map was removed could we pro-
ceed with the discussion.
Regardless of conflicting interests in quantity of water, I think
Stein
47
-------
that in the field of water quality we have reached an area of accom-
modation where all the States can work together, because whoever
gets the water in whatever quantity wants the water in the best
possible condition.
SOURCES OF FEDERAL-STATE WATER
RIGHTS CONTROVERSY
We might outline, now, as candidly as we can, the sources of
Federal-State controversy in water rights.
Many Federal statutes dating from the latter part of the nine-
teenth century led people to suppose that the United States had per-
manently transferred control of water rights to the States in the
patents that authorized the acquisition of water rights under State
law along with the acquisition of land in the West. In addition, in
legislation admitting several of the western States to the Union,
Congress accepted State constitutions with provisions that included
the recognition and confirmation of existing water rights. This also
led to the notion that the Federal Government was relinquishing
control of water rights to the States.
Then there are the acts under which the Federal Government
entered water resource development activity. So that there are no
hidden skeletons in the closet on this issue, I think it is useful to
review the acts.
The Federal Reclamation Act of 1902 included a provision that
Federal activities should not interfere with State laws relating to
the control, appropriation, use, or distribution of water used for
irrigation or any vested right acquired thereunder. This provision is
typical of subsequent reclamation laws.
The Federal Power Act of 1920 provided that the Act was not to
be construed as affecting or intending to affect, or in any way to
interfere with State laws relating to the control, appropriation, use,
or distribution of water used for irrigation, for municipal purposes, or
for other uses or in any other vested right acquired thereunder.
In the Flood Control Act of 1936, for the first time, it was de-
clared to be a national policy for the Federal Government to engage
in flood control activities because of the adverse effects of floods on
commerce between the States and on the national welfare. This
created quite a storm, and in the Flood Control Act of 1944, it was
declared to be the policy of Congress to recognize the interests and
rights of the States in determining the development of watersheds
within their borders and to limit the use of works authorized for navi-
gation to those purposes that do not conflict with any beneficial con-
sumptive use of water in the 17 western States. These statements of
policy have been repeated in the Omnibus Rivers and Harbors Bills
each year to give the notion of States' rights in this area.
Essentially the same declarations are included in the Watershed
Protection and Flood Prevention Act of 1954. Section 4 of the Act
48
WATER RIGHTS
-------
establishes as one of the prerequisites for Federal assistance in
watershed development the requirement that a local organization or
the land owners acquire such water rights pursuant to State law as
may be needed in the installation and operation of the works of im-
provement.
Finally, in the Federal Water Pollution Control Act, as many of
you know, it is . . declared to be the policy of Congress to recog-
nize, preserve, and protect the primary responsibilities and rights of
the States in preventing and controlling water pollution , . . Section
1(b) further provides that . . nothing in this Act shall be construed
as impairing or in any manner affecting any right or jurisdiction of
the States with respect to the waters (including boundary waters) of
such States."
This series of acts has been taken by many people to mean that
the Federal Government has relinquished its primary rights in water,
or water rights, to the States. On the other hand, if the United States
had relinquished its rights, why was it necessary to repeatedly
acknowledge State authority in these statutes? In other words, the
primacy of Federal rights in water was implicitly acknowledged in
each of these statutes. The fact is that Federal rights and responsi-
bilities in water could not be relinquished under the Constitution.
On the other hand, it should be noted that no change has been
made in the provisions of the Constitution dealing with the control
of water under the commerce clause, the general welfare clause, the
national defense clause, or the treaty-making clause. The Federal
sovereignty with respect to navigation has been generally affirmed.
As has often happened in American history, while the statutes
repeatedly have recognized State authority, Federal court decisions
have almost uniformly upheld Federal sovereignty against claimants
of water rights under State law. That is, the courts have generally
held for the Federal Government. So, while the Congress has con-
tinued to enact this series of laws recognizing States' rights, a series
of Federal court decisions have pretty uniformly upheld Federal
rights whenever they have conflicted (or apparently conflicted) with
States' rights.
I think it is appropriate to call attention to some of these cases.
The most famous cases —¦ United States vs. Eio Grande Irrigation
Company in 1899 (174 U.S. 690) and Winters vs. United States (207
U.S. 564) — generally have made it clear that the sovereignty of the
United States over waters coming under the commerce and treaty-
making clauses of the Constitution has not been broadly relinquished
by any statutory enactments.
Among other court decisions along this same line, the classic case
is the United States vs. Appalachian Power Company (311 U.S. 377)
decided by the Supreme Court in 1940. Although this case dealt with
structures, the scope of the language of the decision is instructive.
In it the Supreme Court stated:
Stein
49
-------
"This power of Congress to regulate commerce is so unfettered
that its judgment as to whether a structure is or is not a hin-
drance is conclusive. Its determination is legislative in character."
The Court went on to say:
"In our view it cannot be properly said that the constitutional
power of the United States over its waters is limited to control
for navigation. By navigation respondent means no more than
the operation of boats and improvement of the waterway itself.
In truth, the authority of the United States is the regulation of
commerce in its waters. Navigability, in the sense just stated, is
but a part of this whole. Flood protection, watershed develop-
ment, recovery of the course of improvements for utilization of
power are likewise parts of commerce control. That authority is
as broad as the needs of the Congress."
One of the cases that created — at least in my opinion — the
most heat and the least light was the famous case of the Federal
Power Commission vs. Oregon (349 U.S. 435), the case more famil-
iarly known as the Pelton Dam Case. In deciding this case in 1955,
the Supreme Court affirmed the supremacy of Federal jurisdiction
over unappropriated non-navigable water arising or flowing over
reserved lands of the United States, and in the western States the
Federal Government has quite a bit of reserved lands.
One case that is creating a lot of controversy but is still before
the courts is the United States vs. Falbrook Public Utility District
(165 Fed. Supp. 806). Although I do not believe it proper to talk
too much about this case, I do want to quote from the lower court
opinion to give the flavor of what is happening. The lower court
cited the decision of the Supreme Court in the case of the city of
Tacoma vs. Taxpayers of Tacoma (357 U.S. 320) and stated:
"It is no longer open to question that the Federal Government
under the Commerce Clause of the Constitution (Art. 1, Sec. 8,
CI. 3) has dominion, to the exclusion of the States, over navigable
waters of the United States."
The District Court then continued:
"It is our conclusion that as to the use of water for power proj-
ects, the Supreme Court has made it clear that Congress has
acted to endow the projects with such a public interest that
State action shall not be permitted to interfere with the func-
tioning of the projects."
From these citations you can see that the situation is a relatively
critical one. It is relatively easy to say, "We can solve Federal-State
water rights questions on a case-by-case basis," and this method
will continue unless we arrive at a conclusion acceptable to both the
States and the Federal Government. When water rights questions
are solved on a case-by-case basis before the courts, however, the
Federal Government has generally won.
50
WATER RIGHTS
-------
EFFECT OF EXERCISE OF FEDERAL WATER RIGHTS
We may be dealing with a theoretical issue. With all the con-
troversy over Federal and State water rights, perhaps we should ask,
"Who has been hurt?" Aside from the theoretical question of States
rights and State privileges versus the supremacy of the Federal Con-
stitution, who has really been hurt?
First I would like to quote Assistant Attorney General Ramsay
Clark, who heads the Lands Division, which handles water questions
in the Department of Justice. He said:
"For all the hue and cry arising from this controversy, not one
state, not one county, not one municipality, not one irrigation
district, not one corporation, not one individual has come forward
to plead and prove that the United States, exercising alleged
proprietary rights in the unapproved waters of public domain,
has destroyed any govermental power or property right. Why?
because it has not happened."*
Now a view from another spokesman, Senator Frank Moss of
Utah, who I think may fairly be said to represent the western States'
point of view in this issue. He said recently:
"I must confess that a rather detailed perusal of many articles on
the subject of Federal-State conflict in water rights has failed to
turn up many cases where actual damages can be shown. The
few cases I have uncovered where actual damages due to loss of
water seems to be demonstrated appear to be based on disputes
over water rights acquired by the Federal Government by pur-
chase or in other manner under State law."**
I think these positions are not very far apart.
Especially in dealing with water quality under the Federal Water
Pollution Control Act, we have a situation where there need be no
conflict between the States and the Federal Government. Although
the potential for conflict is certainly there and it would be expecting
the millenium to expect that none will develop, it will be the result
of baseless discord, both State and Federal, rather than principles,
if conflict does develop. All too often conflict has revolved around
the development of potential water rights that have never actually
been exercised.
REDUCTION OF CONFLICT
What can be done to prevent conflict over water quality control?
First the Federal agencies must reach an accord on how water quality
control will be handled, decide when water will be released for
*Speech, National Association of Counties, Grazing, Water, and Rev-
enue Conference, December 1662, Las Vegas, Nevada.
**Speech, Annual Midwinter Conference of the National District
Attorneys' Association, Los Angeles, California, March 11, 1963.
Stein
51
-------
quality control, who is to be responsible for releasing the water, and
on what criteria.
Agreement among the Federal agencies is certainly an essential
first step, but it is only a part of the solution. The Federal agencies
and the States must also be in agreement as to the purposes for which
water is released for water quality control. If the State governments
and all the agencies of the Federal Government can agree on methods
of operation — that waste control measures must be effectively en-
forced and that water released for quality control should remain in
the stream to do its intended job — we will have the problem more
than half licked.
I do not think that we should go beyond this. As in any water
pollution control case, once Federal and State Governments see eye
to eye, if anyone — individual, industry, or municipality — believes
that he is being deprived of property without due process, then by
all means he should have access to remedy through the courts. This
is quite analogous to the condemnation procedure used in the building
of Federal highways or the interstate system.
In the building of a road, taking of land or a building obviously
involves compensation and the question is, "How much?" In the
water field we are going to have two questions, "Is this man entitled
to compensation?" and "If so, how much?" If the Federal agencies
and the States can establish mutually acceptable guidelines and act
together, water rights questions can be handled normally just as a
private right or private compensation suit would be handled in a
State court.
Another question that is going to arise — and I think that this
has already come up on at least one river in the United States — is
that of the withdrawal of released water from a stream before it
has served its intended purpose. Who has the power to police these
withdrawals? As I have tried to point out, I believe that the ultimate
power and responsibility rests with the Federal Government. Water
quality control is also in the State interest, however, and to avoid
diversity of interest between the State and Federal Government, an
agreement should be reached on how much water should be left in a
streambed at any time for water quality control.
If on the other hand, a State joins with an adverse water ap-
propriator in the name of State rights when one of these private water
users takes water from a stream, thwarting the purposes for which a
project was built and for which the water is released, we may all be
in for a very, very difficult time.
In conclusion, let me come back to the Federal Water Pollution
Control Act. The statute provides that storage may be included in
Federal water resources projects "for regulation of streamflow for
the purpose of water quality control." Once the Public Health Service
has made a report on a project; once the needed storage has been
included in a project by the Corps of Engineers or the Bureau of
Reclamation; and once monies have been voted by the Congress,
52
WATER RIGHTS
-------
contracts have been let, the project put in operation, and releases
made for water quality control, I think that whoever tries to thwart
the purpose of the Act by exercising supposed rights to the water
made available is going to be reversed, if necessary, by the courts.
This is not to say, let me make clear, that in specific cases someone
may not have been deprived of a property right for which he is
entitled to compensation.
As a last word, let me say that we in the Public Health Service
recognize that we are facing a new problem and we are asking you
to join with us in thinking it through. I am sure that we cannot
leave it at, "You go your way and we'll go ours." We are extending
a hand to you in the other Federal agencies and you in the State
agencies and asking you to join with us in the recognition that we
are going to have some most difficult technical problems to solve
before we can properly incorporate water quality control in water
resource development projects. I hope we will not make it more
difficult by introducing unnecessary legal problems.
Stein
53
-------
Session 2
EFFECTS OF IMPOUNDMENTS ON
WATER QUALITY IN RESERVOIRS
Moderator: B. B. Berger
U. S. Public Health Service
This session is concerned with the quality changes associated
with the conversion of a free-flowing stream into an impoundment.
Specifically, we can ask these questions:
What is the phenomenon of stratification and how does it affect
water quality?
What is the phenomenon of eutrophication and how do we asso-
ciate it with stored water?
What are the particular chemical reactions that take place in
stored water that we associate with the usefulness of this water?
Each of these questions is treated in this session.
GPO 821-740-3
-------
THERMAL STRATIFICATION IN RESERVOIRS
F. W. Kittrell
Technical Services Branch, DWS and PC
Robert A. Taft Sanitary Engineering Center
Cincinnati
The principal purpose of this discussion is to provide a general
background for some of the more detailed presentations that are to
follow. This background is desirable for those who have not had
occasion to examine the phenomena of thermal stratification of water
in natural lakes or manmade reservoirs. It is hoped that a brief
review of some of the factors involved will be of value even to those
who have had reason to investigate the subject in the past, but may
not have retained all of the knowledge gained at that time.
The first man who ventured into a small pond on a calm, sunny
afternoon in prehistoric times probably noted one of the effects of
thermal stratification. Those who have enjoyed the pleasures of
swimming in the old mill pond have made the same observation.
When a swimmer stays on the surface, the water feels warm; but
when he lowers his feet to tread water, the deeper water feels cold.
The sun has warmed the surface layer of water and caused a reduc-
tion in density. The reduced density produces a resistance to mixing
of the warmer, lighter water with the colder, heavier water, and
the warm water floats on top as a thermal density layer or stratum.
This type of stratification usually is temporary and is soon destroyed,
either by surface cooling at night, or by mixing resulting from wind-
induced wave action.
Among the physical characteristics of water, its change in density
with change in temperature is one of the most important in the be-
havior of water and its constituents in lakes and reservoirs. Quanti-
tatively this change is small, being only 4.3 milligrams per cubic
centimeter, or less than 0.5 percent, between the maximum density
of 1.0 gram per centimeter at 4°C and that of 0.995673 gram
per cubic centimeter at 30°C, which is about the upper limit of
temperature for most surface waters.
Density changes of water with changes in temperature exhibit
two unique characteristics that exert important influences on the
characteristics of stratification in lakes and reservoirs. These char-
acteristics are illustrated in Figure 1. The figure does not present
actual densities, but rather differences in densities for 1°C for water
temperatures of 0° to 30°C.
The sharp break in the curve at 4°C illustrates one of the char-
acteristics. This shows that the minimum effect of temperature
changes on densities occurs at 4°C, and increases on either side of
this value. This coincides with the well-known fact that the maxi-
mum density occurs at 4°C, and is an important factor in the winter
stratification of water.
Kittrell
57
-------
D
UJ
>
UJ
OC
o *—• — u
o 5 10 15 20 25 30
WATER TEMPERATURE, °C
Figure 1 — Difference in water dentities for variout temperaturei.
The other peculiarity is the very large density differences for
1°C temperature changes at the higher temperatures, compared to
those at lower temperatures. Reference to the left vertical scale shows
that there is a density difference of only about 0.008 milligram per
cubic centimeter for 1°C change in temperature in the vicinity of 4°C,
whereas there is a difference of nearly 0.3 milligram per cubic
centimeter, or nearly 40 times as great, at 30°C. The relationships
throughout, the temperature range are shown by the right vertical
scale. The resistance to mixing, or conversely the amount of work
required to accomplish mixing, of water exhibits quantitatively
identical relationships to temperature; For example, a column of water
with a given height and cross section and a temperature of 22°C at
the bottom and 23 °C at the top would consume 30 times as much
energy in mixing as would a similar column of water 4°C at the
bottom and 5°C at the top. This fact explains why stable stratifica-
tion may persist late into the fall with temperature differences as
little as 2°C, whereas winter stratification may not occur in the ab-
sence of ice cover, and summer stratification does not start until the
water mass has warmed and substantial differences in temperature
from top to bottom have been established.
Knowledge of stratification in reservoirs has its origin in limno-
logical studies involving natural lakes. A manmade reservoir may
be considered as the upstream half of a natural lake, with a dam
58
THERMAL STRATIFICATION IN RESERVOIRS
-------
taking the place of the downstream 'nail. Obviously, many of the
characteristics of lakes are reproduced in reservoirs. There are two
important exceptions, however, which may lead to certain water
quality problems downstream from reservoirs.
One exception involves the facilities for control of discharge
from reservoirs, which may be operated to reduce downstream dis-
charges below natural miniaauro flows.
The other exception involves the depths from which water is
withdrawn from reservairs. Water generally passes through natural
lakes either by essentially complete displacement In the .absence af
stratification, ar by firwage through the surface stratum when strati-
fication exists. In bath cases, water leaving the lakes has been sub-
verted to normal aerobic processes of natural purification lot expended
periods of time and quality of the "water usually 13 good. On. the other
hand, water discharged below reservoirs frequently is withdrawn
from deep in the reservoir In title absence ot stratification this pre-
sents no problem. When stratification exists the water in the depths
of the reservoirs may have been subjected to anaerobic processes of
natural purification. These processes may produce undesirable "water
quality characteristics, such as lesw dissolved oxygen and excessive
hydrogen sulfide, color, iron, and manganese, in the water down-
stream from the dams.
Thermal sirsLocation may assume many patterns, dependent an
geographical location, clifnatological conditions, depth, surface area,
and olijer physical configurations of the lake or reservoir, This dis-
cussion is limited to description of only a lew i; U cab,' e. Jew c!b!te
cr a; iuk: a raw \weks. AMhcajn fkvw veiDciues are low, they ust-
ady sre appreciable end in a positive downstream direction Passage
of ivater through toe reservoir is by total displacetrwect, If pwer
production is in vol vac the [op of the penstock intake aasy extend
from near the bottom to -within 15 to 211 ieet of the water surface,
?h?:.Ta. s.traafvsaiic® irit^is type c-1 reservoir, v.hi:h occurs moat
frequently in warm weather and is somewhat similar to that described
in the old mill pond, is illustrated in Figure 2. The parallel hori-
zontal temperature lines indicate a gradual temperature gradient
decreasing from top to bottom through a moderate range of 25" to
2&°C.
The temvpet-Mure differences ar-fe caused by the manner in which
tJie sun warms bodies ct water. The upper layers of water absorb
the larger portions of the sun's energy with only minor portions
reaching t'ae jrwer depths. As the depth Increases arithmetically the
KittreU
58
-------
DAM
_____ 25°C
24°C
23°C — —
/ — INFLOW
. 22°C
PENSTOCK
Figure 2 — Main stream reiervoir — summer stratification,
energy absorption decreases geometrically, following the law of con-
stant proportionality. Although the rates of absorption differ greatly
among various bodies of water, it appears that a reasonable average
for clear natural water may be about 30 percent of the entering
energy in each 3-foot layer of water. Thus, roughly 70 percent of
the surface energy may remain at 3 feet, 5 percent at 15 feet, and 1
percent at 30 feet. The depth relationship to energy absorption is
reflected by the logarithmic spacing of the temperature lines in
Figure 2.
The warmest water, at the surface, has the lowest density and
therefore tends to remain at the surface, with the coolest and most
dense water on the bottom. This stratification does not occur if
stream velocities are sufficient to provide the energy necessary to
overcome the resistance to mixing inherent in the different water
densities. The velocities may be those of either the downstream flow
of the water or currents induced by wind. Many main stream reser-
voirs have neither the necessary flow velocities nor long, unbroken
surface reaches that would permit the wind to induce mixing veloci-
ties.
The effects of this type of stratification on water quality have
not been found to be nearly so important as are the effects of the
classical three-layer pattern of stratification, which occurs in many
storage reservoirs. The principal adverse effect of the temperature
gradient is the reduction in reaeration that results from the reduction
in vertical circulation of the water. Dissolved oxygen concentrations
as low as 5 milligrams per liter have been found in a number of
main stream reservoirs at locations far from significant sources of
pollution.
The classical three-layer stratification pattern that occurs in
many storage reservoirs is much more complex than the simple
thermal gradient pattern found in main stream reservoirs. The differ-
ences between the two types of reservoirs in physical configuration,
dam design, and reservoir operation contribute to the differences in
stratification patterns.
The typical storage reservoir usually is located on a tributary
60
THERMAL STRATIFICATION IN RESERVOIRS
-------
rather than on a main stream. The stream is impounded by a rela-
tively high dam of 100 to 125 feet or more. The impounded waters
extend far beyond the natural river channel and into many embay-
ments to form long unbroken reaches of surface on which wind may
act to induce currents, waves, and mixing of the waters. Fluctua-
tions in water level of as much as 70 to 80 feet or more throughout
the year are common. The storage capacity in relation to drainage
area is large, and detention is several months. Flow velocities in
much of the reservoir are negligible and may be influenced more by
wind-induced currents than by downstream movement of the water.
Passage of water through the reservoir is discontinuous, and quan-
tities of the water may remain in storage for nearly a year. Draw-
down of the reservoir dictates that the discharge intakes be located
deep in the reservoir, below the minimum level to which water will
be drawn.
Figure 3 shows the thermal pattern that occurs in late winter or
early spring. The reservoir water is cold, and at 6°C has a low re-
sistance to mixing. The water is easily mixed by wind action and has
a uniform temperature from top to bottom.
Temperature, °C
Figure 3 — Storage reservoir — spring temperature pattern.
As the season advances and atmospheric temperatures become
higher, both the inflowing tributary water and the surface water in
the reservoir warm up more and more rapidly, and become more and
more resistant to mixing. Finally, the resistance to mixing becomes
great enough to overbalance the ability of the wind to accomplish
circulation to the bottom of the reservoir, and stable stratification is
established. This may occur as early as April in the southern portion
of the country, and a month to 6 weeks later in the northern portion.
Kittrell
61
-------
The resulting summer stratification is illustrated in Figure 4.
This figure shows the classical three-layer stratification to which
limnologists have devoted so much study in natural lakes.
The upper stratum, or epilimnion, may be 30 to 50 feet deep.
The reservoir is mixed to this depth by wind-induced currents, and
has a uniform temperature, which in this instance is 26 °C.
The quality of water in the epilimnion usually is quite good.
Prolonged storage and reaeration permit the forces of natural puri-
fication to accomplish maximum benefit. Organic wastes are stabil-
ized, and dissolved oxygen usually is near saturation. Bacterial con-
tamination is reduced to a minimum. Silt is practically eliminated
and color is decreased. Wide variations in water quality are reduced
and relatively constant quality is achieved. The only adverse effect
may be excessive growth, or blooms, of algae, with attendant taste
and odor production.
Below the epilimnion is the thermocline, which may be 10 to 20
feet deep. As its name implies, the thermocline is a stratum in which
temperature decreases rapidly as depth increases. The thermocline
has been defined for convenience, though on a strictly arbitrary basis,
as a stratum of water in which the temperature decreases 1°C or
more in a depth of 1 meter, or 0.55°F in 1 foot. The decrease in
temperature is accompanied by an increase in density, with a cor-
responding increase in resistance to mixing. In Figure 4 the tempera-
ture of the thermocline decreases from 26°C at the top of the stratum
to 10 °C at the bottom.
A well-established thermocline is quite stable and has a strong
^ INFLOW
—>-»—1—i—I i L
o 10 20 30
Temperature, °c
Length
Figure 4 — Storage reiervoir — iummer stratification.
62
THERMAL STRATIFICATION IN RESERVOIRS
-------
resistance to mixing. It serves, in a manner similar to a flexible
diaphragm, to separate the top and bottom strata of water. Figure 5
illustrates the stability of the thermocline. Here the wind has been
blowing strongly from left to right across the reservoir. The water
of the epilimnion has been driven toward and piled up on the right
side of the lake, forcing the thermocline on that side deep into the
reservoir. Correspondingly, the left side of thermocline has risen
toward and actually intercepted the surface of the reservoir. As a
result, a portion of the cold water of the thermocline is exposed at
the left bank of the reservoir. Despite this displacement, when the
wind stops blowing, the thermocline returns to its former horizontal
position without breaking up. This return is accomplished by a series
of oscillations during which the opposite sides of the thermocline
alternately rise above and fall below the normal horizontal level
until finally the normal position is resumed. The series of oscillations
is known as a seiche.
Wind Direction
Figure S — Storage reservoir — summer stratification wind tilted thermocline.
The bottom stratum of the reservoir, below the thermocline, is
the hypolimnion. Here the water is excluded from the atmosphere
by the overlying thermocline and epilimnion. Very little light and
heat from the sun penetrate to these depths, and very little increase
in temperature occurs after stratification is established. Most of the
increase takes place in the upper portion of the hypolimnion, which
produces a slight temperature gradient from top to bottom. In Figure
4 the temperature of the hypolimnion is 10°C at the top and 8°C at
the bottom.
Kittrell
63
-------
Summer stratification may persist until late October or early
November in the southern portion of the country, and may terminate
a month or so earlier in the northern portion. During the 4 to 6
months of summer stratification, the hypolimnion is unable to replen-
ish its dissolved oxygen, and decomposition of organic matter present
in the water, in bottom deposits, and in dead plankton, which settle
from the top strata continuously, depletes the available oxygen. In
the hypolimnions of many reservoirs, even though man-made pollu-
tion may not be involved, total depletion of dissolved oxygen occurs.
During the period of anaerobic, or septic, decomposition that follows,
hydrogen sulfide may be produced, color may increase, and the iron
and manganese contents of the water may reach objectionable levels.
Water withdrawn from such a hypolimnion and discharged to the
stream below the dam contains the objectionable constituents and is
deficient in dissolved oxygen.
Some storage reservoirs do not exhibit this classical type of
stratification. Those reservoirs that do not store substantial volumes
of water at winter temperatures or that discharge such water before
warm weather occurs do not develop thermoclines. Some reservoirs
that are relatively shallow and have broad expanses of surface areas
exposed to strong winds may not develop thermoclines. These types of
reservoirs may, however, exhibit temperature gradients like those
in main stream impoundments.
Total oxygen depletion does not occur in the hypolimnions of
some reservoirs. There is some evidence that water poor in the
nutrients that support aquatic life may not develop sufficient organic
material to exhaust the dissolved oxygen through decomposition.
As atmospheric temperatures drop in the fall, inflowing tributary
streams and surface water in the reservoir become cooler. This cooler
water is mixed throughout the epilimnion and the upper portion of
the thermocline by convection and wind action. Cooling continues
until the temperatures and densities of the epilimnion and thermo-
cline approach those of the hypolimnion. Finally, the resistance to
mixing is reduced to the extent that wind action can mix the water
to the bottom of the reservoir, and the temperature becomes uniform
from top to bottom as illustrated in Figure 6. This occurrence is
known as the fall overturn.
In the southern portion of the country, where surface water
temperatures rarely drop below 4°C, no stratification occurs in winter,
and temperatures remain uniform throughout impoundments. In
northern areas, where temperatures drop as low as 0°C, and especially
where ice cover forms, winter stratification occurs.
The temperature pattern of winter stratification is the reverse
of that in summer, as shown in Figure 7. The colder, but in this case
lighter, water at 0°C is at the surface, below the ice cover. The
warmer water, with its maximum density at 4°C, is on the bottom.
The temperature increases sharply from 0° to 4°C in the surface
stratum, and resembles a thermocline, though it is not commonly
designated as such. The temperature of most of the water is 4°C.
64
THERMAL STRATIFICATION IN RESERVOIRS
-------
Temperature, °C
Figure 6 — Storage reservoir — fall temperature pattern.
Figure 7 — Storage reservoir — winter stratification.
Kittrell
65
-------
Total depletion of dissolved oxygen during winter stratification
may occur near the bottom of the reservoir, as a result of decomposi-
tion of bottom deposits. Total depletion throughout the entire depth
is rare and apparently occurs only in shallow lakes under somewhat
abnormal conditions. At the start of stratification, the dissolved oxy-
gen concentration is greater in winter than in summer. At winter
temperatures, rates of organic decomposition are low, and reduced
plankton productivity yields less dead plankton than in summer to
deplete the oxygen.
The quality of water withdrawn from the lower levels of reser-
voirs in winter generally should be better than that in summer.
As atmospheric temperatures increase in spring, reservoirs that
have been stratified during winter pass through the spring overturn
and return to the uniform temperature distribution illustrated in
Figure 3.
A type of stratification that has been observed in reservoirs but
has no known counterpart in natural lakes deserves mention. This
has been found only in main stream reservoirs that receive cool water
from upstream storage impoundments. It is produced only when the
water flows through a reservoir at appreciable velocities and requires
large portions of the cross sectional areas of the reservoir for passage.
The resulting stratification is illustrated in Figure 8. The incoming
water at 20°C is about 6°C colder than normal surface water during
midsummer. The flowing water occupies the entire cross sectional
areas of the upper end of the reservoir. From about the middle of
the reservoir on downstream, only portions of the larger cross sec-
tional areas are required. The portions not required by the flowing
water, above the top of the penstock intake, are occupied by a wedge-
shaped body of warmer water, consisting of a thermocline and an
epilimnion. The thermocline is not horizontal, but is approximately
parallel to the bottom slope of the reservoir and is held in this position
by the cooler water flowing beneath it. The stability of the thermo-
cline is revealed by the fact that its upstream end, which intersects
the water surface, moves upstream when streamflow is reduced and
downstream when streamflow is increased, but is not destroyed unless
streamflow becomes extremely high.
~am
66
THERMAL STRATIFICATION IN RESERVOIRS
-------
This type of stratification is not common at present, but will
become more prevalent as river systems are developed more com-
pletely, with storage reservoirs pouring cold water into downstream
reservoirs.
The principal effect of such stratification on water quality results
from the elimination of reaeration of the water flowing beneath the
thermocline. If the flowing water has a significant oxygen demand
and if it requires several days to pass through the lower end of the
reservoir, excessive depletion of dissolved oxygen may result.
DISCUSSION
Dr. Donald R. F. Harleman*
California Institute of Technology, Pasadena
Mr. Kittrell has described clearly the phenomena of thermal
stratification in reservoirs. The purpose of this discussion is to show
quantitatively how thermal stratification in a reservoir affects the
quality of the water released.
We are concerned with a complex fluid-mechanics problem in-
volving the interrelationship between the vertical density distribu-
tion resulting from the thermal structure and the vertical velocity
distribution induced in such a fluid by the inflow to the reservoir at
the upstream end and by the withdrawal of water by a turbine intake
at the downstream end.
To simplify the problem somewhat, let us consider a large res-
ervoir in which the flow pattern is primarily controlled by the dis-
charge at the turbine intakes on the face of the dam. This is an
especially good example if water is stored in the reservoir during
spring runoff, and during the summer season of thermal stratification
the rate of water release is appreciably greater than the rate of
inflow. We can further visualize a rectangular box in one end of
which the wall represents the dam and a horizontal slot represents
the turbine intake, or reservoir outlet.
If the fluid in the box is of constant temperature and density,
vertically, the streamline pattern a short distance upstream of the
outlet is a series of equally spaced parallel lines as shown in Figure 1.
Thus the percentage of the total flow from all depths is the same re-
gardless of the elevation of the outlet slot. Hence the vertical position
of the outlet has no effect on the quality of water released, it being
a simple average of the vertical distribution of any constituent such
as iron content or dissolved oxygen within the reservoir.
As a second case, we will consider a similar box in which the
fluid has a vertical temperature distribution such that the fluid
density increases linearly with depth. This is actually a more com-
*Now with Massachusetts Institute of Technology, Cambridge.
Harleman
67
-------
Dam
Water Surface
Outlet
1
y
Figure I — Stream lines in a reservoir of uniform density
plicated problem from the fluid-mechanics viewpoint than the three-
layer system described by Kittrell. The three-layer system of epi-
limnion, thermocline, and hypolimnion (with turbine intakes deep
in the hypolimnion) is rather simple, fluid-mechanically, since al]
the water is drawn from the hypolimnion. There are many reser-
voirs, especially the deep ones that store during the spring and then
release during the summer, in which the distribution of temperature
is perhaps better approximated by a linear gradient rather than by
the sharp thermocline previously discussed. Theoretically and ex-
perimentally we can show that, under conditions of a density gradient,
internal velocity concentrations will occur; that is, there will be a
strong peak in the vertical distribution of horizontaJ velocity in the
vicinity of the elevation of the intake and, therefore, the quality of the
water discharged will be strongly influenced by the concentration of
constituents at the elevation of the intake. The velocity concentration
is due to the fact that the vertical density gradient tends to stiffen
the fluid against vertical motion and relax it to horizontal motion.
Several interesting and analogous situations occur that are
familiar. One is in the atmosphere where a vertical temperature
gradient is always present. If horizontal pressure gradients exist, as
they almost always do because of high and low pressure areas, the
phenomenon known as the jet stream develops. Furthermore, this
phenomenon also exists in the ocean where there are always temper-
ature gradients. The main currents of the ocean — the gulf stream,
and so on — are manifestations of these velocity concentrations.
Theoretically, there are two methods of analyzing the problem
of velocity concentrations in a stratified fluid with a linear distribu-
tion of density. Although the basic assumptions are quite different
both methods point to the existence of selective withdrawal.
The first method of analysis (1) neglects viscosity and heat con-
duction and considers only the dynamics of an incompressible fluid
under gravity whose density increases linearly with depth. The solu-
68
DISCUSSION
-------
tion for the flow pattern in a reservoir, or in the hypothetical box,
due to the withdrawal of water as from a turbine intake can be found
in terms of a single dimensionless parameter known as the densi-
metric Froude number
Where: Q == turbine discharge in cubic feet per second at the time of
temperature measurement.
W = width of the reservoir in feet at place where the tem-
perature measurement was made and at the elevation
of the centerline of the turbine intakes.
z0 = vertical distance in feet between the water surface and
the centerline of the turbine intakes.
pz0 = density of water at the center line of the turbine intake
in slugs per cubic foot.
g = acceleration of gravity (32.14 feet per second per second).
/} = slope of the linear density curve for the underflowing
layer of water, b, computed in slugs per cubic foot per
vertical foot.
This parameter incorporates the characteristic dimensions of the sys-
tem such as depth of the intake, the flow rate, and the vertical
density gradient. For values of this parameter of the order of magni-
tude of unity the streamline pattern is essentially the same as for a
homogeneous fluid. At a densimetric Froude number of 1/tt, or
roughly one-third, however, the solution becomes mathematically
imaginary and the physical interpretation is that the process of in-
ternal velocity concentrations or selective withdrawal must begin at
this point.
Within the range of reservoir dimensions, flow rates, and temper-
ature gradients of interest in actual practice, the value of the densi-
metric Froude number is of the order of 0.01. This means that under
conditions of even the smallest vertical temperature gradient, a
selective withdrawal can be expected to occur. This theoretical ap-
proach is therefore of little value in quantitative prediction of the
flow mechanics, but it has proven the existence of the effect we are
seeking. More importantly, it indicates that in the range of interest
the inertial forces are very small compared to the gravitational effects
of the density gradient since the densimetric Froude number expresses
physically the ratio of the two forces.
This has led to the development of a second theoretical approach,
which we are now analyzing and experimenting with at the California
Institute of Technology.
The second theory(2) considers both viscosity and heat con-
duction in an incompressible fluid having a vertical density gradient.
Harleman 69
-------
To solve the equations of motion, the inertial forces, which were
shown to be small, are neglected The analysis indicates a strong
velocity concentration in the reservoir at the elevation of the outlet.
In addition, it indicates a reverse flow over a significant portion of
the velocity distribution in a vertical section. The shape of the velocity
distribution is a function of a single dimensionless parameter incor-
porating the gravity constant, the density (or temperature) gradient,
the coefficient of heat conduction, and the viscosity of the fluid.
The objective of this research program, which is supported by the
Division of Water Supply and Pollution Control of the Public Health
Service, is to obtain a method for determining the characteristics of
the water released through an outlet from an impounding reservoir.
If we know the vertical distribution of any constituent of interest,
such as DO ior example, we want to be able to predict the concentra-
tion in the water released. Furthermore, we want to predict the
change in the concentration of the constituent as a function of the
elevation'of the outlet.
Figures 2 to 5, which deal with a study in which I cooperated
with the Walla Walla District of the Corps of Engineers and the
Tennessee Valley Authority (3), illustrate some of the points I have
made.
Figure 2 shows a typical series of vertical temperature profiles
taken 700 feet upstream from Fontana Dam. The profiles indicate
that the stratification during the summer months is not of the classi-
cal three-layer pattern, but rather show a gradual increase of tem-
perature with depth.
35 40 45 SO 55 60 65 70 75
TEMP ERA T"U RE, »F
Figure 2 — Typical temperature profilesr Fanlarta Retervcir. Doto for 1945.
80
70
DISCUSSION
-------
Figure 3 shows typical temperature and density curves for
Fontana Reservoir. The data available were observations of temper-
ature at regular intervals about 700 feet above the dam. From this
the vertical distribution of density was calculated. This showed that
over an appreciable region the distribution of density was essentially
linear.
Figure 4 is a schematic of an internal density current. It was
assumed that the bottom of the moving layer was established at the
break in the density curve. The thickness was estimated as the verti-
cal depth, b, necessary to give the observed discharge temperature,
based on the assumption that velocity was uniform over the depth
of the density flow.
The correlation of the vertical thickness of the moving layer with
the densimetric Froude number was only moderately successful as
shown by the scatter on Figure 5; it indicated, however, that further
research and perhaps laboratory experiments would be fruitful.
Figure 6 shows the experimental apparatus used in laboratory
experiments at the California Institute of Technology.
1700
COOLING WATER TEMP = 61.0°F
DISCHARGE - 2585 cfs
b \ t. Turbine Intake
1.932 1.934
1.936 1.938 1
DENSITY, slugs/ft*
60 70
TEMPERATURE, °F
1.940
1.942 1.944
40
50
80
90
100
Figure 3 — Typical temperature and density profiles in Fontana Reservoir.
Data for October 3, 1945.
Harleman
71
-------
b thickness of moving layer, ft
/j density, slugs/ft3
B slope of density gradient,
slugs/ft4
Figure 4 — Schematic diagram of internal current due to vertical temperature gradient.
At the end of the box is the horizontal slot, which simulates the
reservoir outlet or turbine intake. The vertical density gradient was
obtained by the rather tedious procedure of putting 1-inch layers of
water of different densities in the box until the desired depth of about
12 inches was reached. Molecular diffusion would then iron out the
stair steps. It is possible to get a very nice linear gradient, but it
takes about 1*A days to prepare for each test.
As soon as flow is started a velocity concentration is induced at
the level of the slot. Figure 7 is a plot of the velocity distribution
some distance upstream of the slot. Note the strong concentration of
velocity in the downstream direction at the level of the outlet. Obser-
vations show that the thickness of the moving layer tends to in-
crease in the upstream direction. This growth in thickness of the
moving layer probably is responsible for observations of the type
Mr. Kittrell described as a tilted thermocline.
Obviously, even if the thickness of the moving layer should ulti-
mately equal the reservoir depth (as it might in a very long reser-
voir) there will still be a strong selective withdrawal of water along
the elevation of the outlet and water quality will be primarily de-
pendent upon the water quality at that depth.
The most significant conclusion of these studies is the need for
providing flexibility and variability of outlet elevations in impound-
72
DISCUSSION
-------
0.005
0.010
a / pz.
0.015
1 WZ,2 J 8jS
Figure 5 — Verficcrl thtckneis of "internal current vs. denjimelrit fraude number.
Sink
Q
/
16 cm
250 cm
0.9 cm
1
12.6 cm
p 0.6 crri J
1
PLAN
12.6 cm
7"
ELEVATION
t-0:
13 cm
45 cm
i
9 cm
Figure A — Schematic diagram of experimental apparatus.
Harleman
13
-------
ments in which the quality of water discharged is an important
consideration. The increased cost of such outlet structures should be
more than compensated by the benefits of improved water quality
in most cases.
Figure 7 — Velocity concentration upstream of an outlet (sink).
REFERENCES
1. Yih, C. S. On the flow of a stratified fluid. Proc. 3rd Nat. Cong.
Appl. Mechanics. 1958.
2. Koh, R. C. Y. Viscous stratified flow towards a line sink. Keck
Lab. Rept. No. KH-R-6. California Inst. Technol,, Pasadena,
Calif. Jan. 1964.
3. Internal density currents created by withdrawal from a stratified
reservoir. Cooperative study by Eng. Lab.t Div. Water Control
Planning, TVA, Norris, Tenn., and Planning and Reports Branch,
Eng. Div., U. S. Army Eng. Dist., Walla Walla Corps Engrs.,
Walla Walla, Wash. Feb. 1962. 20pp.
DISCUSSION FROM THE FLOOR
Mr. Sidio, U. S. Public Health Service: Mr. Kittrell, what work
is actually necessary in determining the existence of thermal stratifi-
cation in reservoirs?
74
DISCUSSION
-------
Mr. Kittrell: Such a determination can cover a wide range of
degrees of work. Probably the simplest attack would be to determine
the temperature at the surface of the reservoir and, if intakes are
located deep in the reservoir, the temperature of the water discharged
through the penstock. It is desirable, of course, to determine the dis-
solved oxygen at the surface and in the water flowing through the
penstock at the same time. If temperatures and dissolved oxygen
values are much lower in the penstock water than at the reservoir
surface, the chances are very good that the classical thermal stratifi-
cation does exist.
Of course, one can go much beyond this and actually determine
the temperature gradient from top to bottom with a thermocouple-
type instrument. Such an instrument — which costs only about a
hundred dollars, incidentally — can have a cable of almost any length
necessary to reach the bottom of the reservoir and is very simple to
drop over the side of a boat or along the upstream face of the dam.
Temperature would be measured at various intervals of depth —
10 feet is a fairly standard depth interval over which temperatures
are taken.
The same thing can be done with dissolved oxygen, which is one
of the principal things of interest. There is a probe-type instrument,
which I believe costs as much as 400 or 500 dollars, that can be used
to determine dissolved oxygen. When I was with the Tennessee
Valley Authority, we became interested in determining which reser-
voirs discharged water low in dissolved oxygen, and we developed a
simple method for determining dissolved oxygen, which at one time
was in Standard Methods. I do not believe it is included now. I be-
lieve the name was the amidol method. We prepared capsules of two
reagents; I believe one was citric acid or some citrate salt that acted
as a buffer and the other, amidol. This produced a yellow color in
the presence of dissolved oxygen. We prepared permanent color
standards from various mixtures of chromate and dichromate. We
sent these to the operators of all the dams with instructions as to how
to use them and asked them to run dissolved oxygen tests every week
on water discharged through the penstock. Although this was a rather
gross measurement, it did indicate whether there was significant
oxygen depletion in the water discharged from the penstocks. We
were very well satisfied with the results, although later on, I believe,
Churchill went to titration right at the reservoir sites. Any agency
responsible for a number of reservoirs might well consider deter-
mining which reservoirs are discharging waters deficient in oxygen.
Mr. O'Connor, Manhattan College: Dr. Harleman, in the labora-
tory experiments you described, were there any turbulence fluctua-
tions in that stream that approached the intake?
Dr. Harleman: No, the scale of the experiment was not large
enough to get turbulence fluctuations.
Mr. O'Connor: Is that generally true in the water that ap-
proaches the intake in reservoirs, too?
From the Floor
75
-------
Dr. Harleman: One of my Figures shows the vertical distribu-
tion of velocity up to the center line of the intake. The theoretical
curve is shown by a solid line, and the experiments we have
done for low Reynolds number flows are indicated by points. In
the Fontana data, the magnitude of these internal currents is still
of the order of a tenth of a foot per second, even at the maximum
value. Therefore, I would not expect turbulent mixing to be a very
serious consideration. The thicknesses of the currents in Fontana,
we think, are of the order of 70 feet in a depth of about 400 feet.
Growth of the internal current thickness may well explain the
tilted thermocline Mr, Kittreli showed in his Figure 8, because if you
follow a line of constant temperature back from the dam during this
type of flow it wou2d exhibit an upward slope.
76
DISCUSSION
-------
PHOTOSYNTHESIS AS A FACTOR IN THE
OXYGEN BALANCE OF RESERVOIRS
C. H. J. Hull
Research Associate, School of Engineering
Baltimore, Maryland
INTRODUCTION
Since September 1957, there has been under way at The Johns
Hopkins University an investigation of low-flow augmentation for
stream-pollution abatement. This study has involved the legal, eco-
nomic, and technical aspects of the use of water stored in wet seasons
for the control of stream water quality during dry seasons.
The evaluation of the need for and benefits of flow regulation
generally requires consideration of the oxygen balance as related to
stream-flow rates and waste-assimilative capacity.
A factor long recognized as generally significant in controlling
the oxygen concentration of natural water bodies is the photosynthesis
of aquatic plants, including microscopic algae that make up the
phytoplankton. In spite of this general recognition, however, this
factor has been frequently neglected in oxygen-balance studies. The
following statement is quoted from a translation of a Russian-
language paper by Vinberg(i):
"Untii recently little attention was paid to photosynthetic re-
aeration during the investigation of the capacity of polluted
waters for self-purification, Soviet research workers(2,3, 4) have
already admitted the urgent necessity of taking photosynthetic
re-aeration into account, but methods for investigating this ques-
tion are only beginning to be developed and have not yet been
introduced in practice in sanitary-hygienic and sanitary-hydro-
biological researches. Quantitative methods for the investigation
of photosynthetic re-aeration are needed firstly, for understand-
ing the frequently-underestimated significance of green organ-
isms as agents in self-purification; and secondly, they are
essential for the investigation of the oxygen balance of polluted
waters."
This paper describes techniques that have been used in studies
of photosynthetic oxygenation in streams, estuaries, and impound-
ments, and discusses the general significance of this factor in control-
ling the oxygen balance of impoundments and regulated streams.
BALTIMORE HARBOR STUDY, 1«49
In 1947, The Johns Hopkins University began a 3-year study
of water quality in Baltimore Harbor, an arm of Chesapeake Bay.
Observations of diurnal fluctuations of dissolved-oxygen concentra-
tions in the near-surface water levels indicated a strong photo-
synthetic effect, and therefore, attempts were made to measure
Hull
11
-------
oxygen production by phytoplankton. This phase of the investigation
took place in 1949. Although Garland (5) summarized the results,
the study has only recently been reported in detail(6).
The technique used in the study — the only one reported in the
English-language literature up to that time — was that employed by
Gaarder and Gran(7) in their 1916 study of Oslo Fjord, the now-
familiar light- and dark-bottle oxygen technique.
The objective of the light-dark test is to expose paired samples
of water to conditions that are as nearly natural as possible, except
that light is excluded from one of the samples. In this "dark" bottle,
photosynthesis cannot take place, whereas in the "light" bottle, de-
pending upon the depth of exposure and upon the presence of auto-
trophic organisms and nutrients, photosynthetic production of organic
matter and the release of oxygen can occur. Respiration consumes
oxygen in both bottles of course. One of the assumptions of the test
is that respiration is equal in the two bottles. This assumption be-
comes less valid as the duration of the exposure period increases,
because of diverging populations in the two bottles. This is one
reason for the frequent precaution of limiting the test period to the
shortest time that will produce a significant difference between the
oxygen concentrations of the light and dark bottles.
Figure 1 shows one method of suspending the samples at the
various depths tested.
Figure 1 — Float for measuring the effect of photosynthesis by the light- and
dark-bottle technique.
78 PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
-------
After exposure, the difference in oxygen concentrations between
the two bottles is a measure of oxygen production at the depth of
exposure. Figure 2 shows the variation of gross oxygen production
with depth below the water surface for a typical 1949 test in Balti-
more Harbor. The upward concavity of the curve near the surface
reflects the inhibiting effect on photosynthesis of supra-optimal light,
as observed by Marshall and Orr(8), who found maximum production
well below the surface, with lesser production at shallower as well
as deeper layers. This condition was not observed in Baltimore Har-
bor, probably because of the relatively high turbidity that limited
supra-optimal light intensities to a very shallow layer near the sur-
face, too shallow to be detected with the size of bottles and the vertical
spacing used.
Surface
x
H
0-
Ld
Q
10
15
20
1
J
1
/
t
1
/
~
O i
~
° 1
.a*
Compensation jc 1
X
Depth S P
Final Light DO "
i! /
E!/ 8
-
> /
1/ £
V c
mt
Lower Limit of £
Photosynthesis 1
1
1
1
1 I J
.. \ L
E
3 1
J h-
CL
Ul
a
4 6
DO Concentration, mg/l
I , L.
10
-L
2 4
PRODUCTION, mg/l
Figure 2 — Recults of light- and dark-bottle test for photosynthetic oxygen production,
Baltimore Harbor, Station B, August 19 to 23, 1949. (Ref. 6).
The solid vertical line in Figure 2 represents the initial dissolved-
oxygen concentration in both light and dark bottles at all depths
tested. The broken vertical line shows the final oxygen values in the
dark bottles at all depths. These two lines are not necessarily vertical;
in waters that are stratified within the depths to be tested, samples
collected and exposed at different depths would be expected to have
different initial oxygen concentrations and different respiration rates.
Since Baltimore Harbor was found to be generally well mixed down
to the natural bottom(5), a simplifying procedure was used in which
a sample taken from a single depth was exposed at all depths tested.
The horizontal distance between the "final light DO" curve and
Hull
79
-------
the "initial DO" curve (or line) is a measure of the net production
at any depth. The distance between the initial DO line and the "final
dark DO" line represents the respiration at any depth. The distance
between the final light and final dark curves at any depth shows the
gross production at that depth. The intersection of the final light and
initial DO curves is the "compensation point," the point at which
photosynthesis just balances respiration. The depth of this point is
called the compensation depth. The intersection (or point of
tangency) of the final light and final dark DO curves gives the lower
limit of the productive or euphotic zone. The vertical distance from
the surface to this latter intersection is the depth of the euphotic
zone. This depth is determined by the transparency of the water to
the wave lengths of solar radiation that support photosynthesis.
For a test in nonstratified water, if the origin of the horizontal
scale is shifted to the right to coincide with the final dark DO line,
the horizontal scale then shows the gross oxygen production directly.
The area enclosed by the "final light DO" curve, the "surface"
line, and the "final dark DO" curve is a measure of the total oxygen
production for the column of water, or
e
Pa = JP.dz (1)
O
in which Pa = gross photosynthetic oxygen production for the
column of water, grams per square meter (g-m-2)
o = depth at the surface (zero)
e = depth of euphotic zone, meters
Pz = gross photosynthetic oxygen production at depth
z, mg/1
Pa can be determined by planimetering the area under the curve
of production versus depth. The area can also be determined with
acceptable accuracy by approximate integration methods such as the
trapezoidal rule or Simpson's rule.
During the summer of 1949, 12 light-and-dark tests in Baltimore
Harbor showed an average gross oxygen production of 2.7 grams per
square meter per day (24,0 pounds per acre per day). For the entire
harbor area, this amounted to 605,000 pounds per day. When com-
pared with the quantity of atmospheric oxygen absorbed by the
harbor, 187,000 pounds per day as estimated by Garland(5), the
contribution of photosynthetic oxygen is seen to be of major
significance.
In 1961, a similar, but more intensive, study of photosynthetic
oxygenation was carried out in the Delaware River estuary. Before
starting this investigation, we reviewed the various newer techniques
for analysis of photosynthesis and other factors of the oxygen balance,
80
PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
-------
which had been introduced into the literature since the 1949 Balti-
more Harbor study(9,10,11,12). These methods do not depend on
bottle measurements and therefore eliminate the known and suspected
errors related to bottle effects. (For a discussion of the bottle effects,
see reference 13,) A modification of Odum's method based on slack-
water sampling(14) was also considered. As explained elsewhere
(25), these various techniques were not attempted because they either
did not account for some of the oxygen-controlling factors believed to
be important in the Delaware estuary, or required an extensive
sampling program far beyond the resources of personnel and equip-
ment available. With full recognition of the potential sources of
error in bottle methods, we decided to use one of the available bottle
methods because their limitations appeared less invalidating than
those of any feasible open-water technique.
In 1952, Steemann Nielsen(16) introduced the light- and dark-
bottle carbon-14 method of measuring primary production. Because
of its greater precision, the carbon-14 method has been widely ac-
cepted as an improvement over the oxygen method. However, an ap-
parently unsettled controversy over interpretation of the results of
the carbon technique led us to reject this method also (15). This left
us with the light- and dark-bottle oxygen method of Gaarder and
Gran (7) as used in Baltimore Harbor in 1949, but with improved
equipment and procedures.
In the 1961 investigation of the Delaware River estuary, 100 light
and dark tests (oxygen method) were made from March through
December 1961. A series of 24-hour tests in August and September,
when oxygen levels in the estuary were at their annual minimums,
indicated that phytoplankton were responsible for a gross oxygen
production of well over 1 million pounds per day (17).
In this study, a concept was developed that should have some
application in studies of reservoir oxygen balance. This concept em-
ploys a new yardstick of self-purification that we have called the
compensated-zone depth, shortened for convenience to the "cozode."
(The cozode should not be confused with the compensation depth.)
The cozode is defined as the depth of the water column, measured
from the surface, for which the total respiration of the planktonic
community is balanced by the gross photosynthesis of the phyto-
plankton in the euphotic zone. In mathematical terms, the cozode is
defined as that depth at which
e C
j PBdz= j R,dz (2)
o o
in which C = the cozode, meters
R, = respiration at depth z, mg/1
and the other terms are as previously defined.
In a stratified reservoir, respiration would be expected to vary
Hull
81
-------
with depth. In such a case, the simplest way to determine the cozode
is by a graphical solution, in which the function
z
R,)Z= | Rzdz
o
is determined for a series of values of z by planimetering the corres-
ponding areas under the curve showing respiration as a function of
depth. These areas are then plotted as a function of z as shown in
Figure 3. Then the value of C is taken from the graph at the inter-
section of the curve with the horizontal line representing the value
of P..
Figure 3 ¦— Illustration of graphical determination of the cozode.
If one is dealing with a vertically non-stratified system, i.e., if
respiration does not vary with depth, then the cozode can be deter-
mined simply by dividing the gross production per unit of surface
area by the respiration per unit of volume or
e
| Pzdz (g*m~2)
C==_L O)
Rz (g-m~s or mg/1)
Inasmuch as the cozode varies with the duration of time con-
sidered, the period should be specified. For many purposes, the 24-
hour day is the most useful period. The 24-hour cozode can be deter-
82
PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
-------
mined by use of light and dark tests exposed for 24 hours. Greater
accuracy will usually result from a series of shorter tests whose
periods aggregate 24 hours.
In the study of the Delaware River, which is generally taken as
a non-stratified estuary, equation 3 was used to determine the cozode.
Figure 4 shows the frequency distribution of values of the cozode for
two stations in June. Also shown are the mean depths of the river
cross-sections at these stations. Comparison of the data for the two
stations shows an advantage for the station at Eddystone over that
at Kaighn Point. Not only is the cozode greater at Eddystone, but also
the mean depth is less than at Kaighn Point. In fact, the cozode
exceeded the mean depth of the river at Eddystone in two of the six
1-day tests. This indicates that about one-third of the time in June,
the total oxygen demand of the river water in this area is more than
balanced by the photosynthetically liberated oxygen. During such
times, the excess photosynthetic oxygen, together with oxygen ab-
sorbed from the atmosphere, is available to balance the demand of
bottom deposits and other oxygen-removing factors.
15.0
<
tfi
Z
LJ
a.
5
O
o
Figure 4
KAIGHN POINT , 11.3 m
_L
_L
J I L
I
_L
_L
5 10 20 30 40 50 50 70 80 90 95
PERCENT OF OBSERVATIONS LESS THAN INDICATED VALUE
Frequency dittribution of compensated-zone deplhi (1-day tests only), Delaware
River, June 1961.
Hull
83
-------
Seasonal Variation of Photosynthesis
A search of the literature has revealed very little information on
the seasonal variation of photosynthesis. In the 1949 study of Balti-
more Harbor(6), the tests were not designed to investigate this
question. In the 1961 study of the Delaware River estuary, however,
tests were made in all seasons. The data appear to show a strong
correlation between temperature and photosynthetic oxygen produc-
tion, as indicated in Figure 5. This figure does not show the true
relationship between temperature and photosynthesis, but only an
empirical relationship between the two for a single station in the
Delaware River. The consistency of the data is noteworthy, however,
in view of the many factors that control photosynthesis. The apparent
correlation indicates that many of these factors are strongly correlated
seasonally with temperature. In discussing the relative importance
of light and temperature as factors influencing algal periodicity,
Prescott(JS) observed that increases or decreases in plankton some-
times attributed to temperature changes are actually due to light
variations. On the other hand, the Delaware River observations
showed maximum photosynthetic rates in August and September,
when temperatures were maximum but duration and intensity of
sunlight were well below their maximum values at the time of
summer solstice.
Whatever the explanation, the data for the Delaware River, pre-
sented in detail elsewhere (17), indicate that photosynthesis varies
markedly with the time of year, ranging from negligible values in
March to maximum values in late summer. The 1961 variation at
Kaighn Point is described approximately by the equation
Ps = 0.04 (1.218)T (4)
in which PB = daily oxygen production in the surface samples,
mg/l/day
and T = water temperature, °C
This indicates that the rate of photosynthesis increases about 22 per-
cent per degree rise in temperature in the Delaware River. It is em-
phasized that this is not the effect of temperature alone, but represents
the combined effects of all factors controlling photosynthesis, many
of which are apparently themselves strongly correlated with the
annual temperature cycle.
Reservoir Studies
In 1957, a study was made to determine the effect of a submerged
weir in the Roanoke Rapids Reservoir on water quality discharged
through the penstocks at the dam (IS). The submerged weir served
to retain the deep waters of low oxygen concentration while allowing
the shallow layers of water, with their relatively high oxygen con-
centrations, to be selectively withdrawn from the reservoir and passed
downstream. The oxygen was needed downstream for the assimila-
tion of wastes from several towns and industries,
84
PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
-------
I I I I I I I L
4 8 12 16 20 24 2B
TEMPERATURE, °C
Figure 5 — Grow oxygen production at lurfoee (depth — % foot) ot a function of water
temperature, Delaware River ot Kaighn Point.
The successful hydraulic operation of the submerged weir raised
a question concerning the ability of the system to replenish the
oxygen so removed. For this reason, a study of the reoxygenating
capacity of the reservoir was made. Although the writer was closely
associated with all phases of the investigation of Roanoke Rapids
Reservoir, the primary responsibility for this phase was in the hands
of Dr. Donald W. Pritchard and Dr. James H. Carpenter, both of
the Chesapeake Bay Institute of The Johns Hopkins University. Their
analysis appears in an appendix in the unpublished report of the
overall study compiled by Fish et al. (19).
Hull
85
-------
Based on observations of diurnal changes in the DO concentra-
tions of the surface layer of water, and using a modification of Odum's
method of analysis (9), Pritchard and Carpenter estimated the mean
rates of photosynthetic oxygen production, atmospheric oxygen ex-
change, and respiration for the reservoir. Briefly, their procedure was
as described below.
The oxygen balance of a cubic-meter parcel of water in the
surface layer 1 meter in depth was considered. Based on the absence
of a vertical gradient in oxygen concentrations at the 1-meter depth
and of any consistent horizontal gradients in the top 1-meter layer,
exchange of oxygen between the parcel considered and adjacent
waters was assumed to be negligible. The oxygen balance of the
parcel was described by the equation
(CH —C) +P_r (5)
in which C = concentration of dissolved oxygen
at time, t, g-m~a
Ko = atmospheric reaeration coefficient, hr"1
Ce = saturation concentration of dissolved
oxygen, g*nrs
P = rate of photosynthetic oxygen production,
g ¦ m'3 hr._1
and R = rate of respiration, g-m"1 hr ]
For the condition of daily maximum dissolved-oxygen concentra-
tion in the surface layer, usually occurring about 3 p.m., the rate of
change in concentration, dC/dt, was set equal to zero, which simplified
equation 5 to
o = k2 (cb — c)a-|_pa~_r (6)
Based on Odum's hypothesis that photosynthetic production
varies directly with light intensity (9), it was assumed that maximum
production occurs at noon and that the daylight diurnal variation is
in the form of a sine curve. {Disagreement among various authori-
ties on the type of variation was noted, but the use of other types of
diurnal variation did not markedly affect the final results,) From
Odum's hypothesis and based on the characteristics of sine curves,
the average rate of photosynthesis between 9 a.m. and 3 p.m. was
taken as
= 0.92 Pm„ (7)
in which PM A = mean value of photosynthetic rate of production
between the time of morning (M) and afternoon
(A) observations
and Pa—*0.70 (8)
8#
PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
-------
The respiration rate, E, was assumed to be a constant. Based on
a finding by Ryther(20) that respiratory rates equalled 10 percent
of the optimal photosynthetic rate for healthy laboratory cultures, it
was assumed that
R = 0.1 Pnllx (9)
A mean value of equation 5 for the period between morning
observations (about 9 a.m.) and afternoon observations (about
3 p.m.) was written as
(dC/dt)MiA = Ko (Ch — C)M A + P^ — R (10)
in which (dC/dt)M A = difference in concentration between
afternoon and morning observations
divided by the time interval
(CK—C)MA = average deficiency between morning and
afternoon observations
and PM A = mean value of the photosynthetic
rate of production between the time
of morning and afternoon observations
Simultaneous solution of equations 6, 7, 8, 9, and 10 gives
0.60 (dC/dt)M A
K-'= 0.60 (CH—C)M,A — 0.82 (CN —C)A (U)
~K2(CS —C)A
and Pmnx= '-± (12)
By use of equations 11 and 12 and the observed oxygen con-
centrations at about 9 a.m. (M) and 3 p.m. (A), K2 and Pmax were
found to be 0.157 per hour and 0.248 gram per cubic meter per hour,
respectively. Because the parcel of water considered was 1 meter in
depth, the surface exchange coefficient, Ks, was numerically equal to
K2, from the relationship
Kh = -X- Ko = iHji (0.157 hr-') (13)
or Ka = 0.157 m*hr_1
From the value of Pmax the gross daily photosynthetic oxygen
production for the upper meter, Pday, was computed to be 2.14 grams
per cubic meter per day. The depth of the euphotic zone was taken
to be 4.3 meters. Several methods of integrating the production over
the entire depth of the euphotic zone gave values of the total produc-
tion in a column of water under one square meter of surface, Ptotan
ranging from 4.19 to 5.98 grams per square meter per day.
The respiration rate was computed to be
R = 0.1 Pmajc = 0.025 g • m-3 hr-1
= 0.6 mg/l/day
Hull
87
-------
-which agreed closely with an observed BOD value of 0.5 milligram
per liter per day. The total respiration for the euphotic zone was
therefore computed to be
Rt«tni = 4.3 x 0.025 x 24 = 2.6 g*m~2. day-1
The net production of oxygen was then found to be
= = 4.2 — 2.6 = 1.6 g-rrr2. day 1
The net exchange of oxygen was found to be negative, represent-
ing a loss of oxygen from the system of 1.0 gram per square meter
per day.
The concept of the cozode used in the 1961 study of the Delaware
River has been applied to the above results obtained by Pritchard
and Carpenter. The cozode is
^ 4.2 grams m~2 day-1
7 meters
0.6 gram m-3 day-1
This indicates that photosynthesis in the euphotic zone of Roanoke
Rapids Reservoir is equivalent to the oxygen demand ol the water
down to a depth of 7 meters (23 feet). At a pool elevation of 133
feet, Roanoke Rapids has a volume of about 82,000 acre-feet and a
surface area of about 4,800 acres. Thus, the mean depth is approxi-
mately 17 feet. Since the cozode exceeds the mean depth, the oxygen
balance of the entire reservoir water volume must show an overall
oxygen surplus, which is consistent with the finding of Pritchard and
Carpenter that there was a net flux of oxygen from the reservoir
surface into the atmosphere during the study period.
A need for study to find ways of capturing the oxygen lost to
the atmosphere is apparent. Perhaps mechanical mixing of the reser-
voir water to prevent supersaturation at the surface is one method
of retaining and using for waste decomposition the oxygen now lost
to the atmosphere. If the average surface oxygen concentration can
be reduced by this means to values significantly below saturation,
the atmosphere will become an additional oxygen source instead of
an oxygen sink.
Oxygen Demand of Photosynthetically Fixed Carbon
I am not unaware that most of the literature on the relationship
between plants and the oxygen balance is concerned primarily with
the detrimental effects of algae and higher plants. The oxygen de-
mands of algae following "population explosions," or blooms, can be
a serious problem, a problem that has focused attention upon this
aspect of aquatic algae. These detrimental effects of blooms should
not be allowed, however, to obscure the fact that aquatic plants, in-
cluding the phytoplankton, can be credited in many cases with a net
contribution to the critical-period oxygen balance of water bodies;
this has been indicated in Roanoke Rapids Reservoir. It seems highly
probable that much of the reoxygenating capacity, which has in the
PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
OPO Bfcl-740-4
-------
past been credited to atmospheric reaeration on the basis of the
oxygen-sag equation of Streeter and Phelps(21) and other formulas,
should really be credited to photosynthesis.
I have been frequently reminded that for every gram of oxygen
released by photosynthesis a gram of biochemical oxygen demand is
produced by the synthesis of organic matter, and therefore there can
be no net gain of oxygen. This is a gross oversimplification — it must
be qualified by consideration of the relative rates, periods, and loca-
tions of the oxygen liberation and subsequent oxygen demand. To
support this, one need only point to the vast standing crop of fixed
carbon on this earth in the form of petroleum, coal, plants, and
animals, including several billions of human beings. All of this carbon
was fixed by photosynthesis, but some of it has waited millions of
years to be oxidized, before it will eventually take back the oxygen
liberated at its birth.
Organic matter fixed photosynthetically in a reservoir may settle
to the bottom where it may be oxidized anaerobically, or may remain
until the fall turnover, when the critical period with respect to oxygen
conditions is passed. Organic matter fixed in Roanoke Rapids Reser-
voir may show up still unoxidized in Albemarle Sound, where oxygen
demand is not a problem.
In any event, oxygen-balance studies in any water body, whether
a river, an estuary, or a reservoir, must take into account the effects
of photosynthesis during the period and throughout the area of
critical oxygen conditions.
ACKNOWLEDGMENTS
The preparation of this paper and the investigations on which it
is based were supported in part by research grants WP-153(C4),
(C5), and (C6) from the Division of Research Grants, National Insti-
tutes of Health, Public Health Service, U. S. Department of Health,
Education, and Welfare. The Roanoke Rapids Reservoir study was
supported by the Virginia Electric and Power Company in coopera-
tion with the North Carolina State Stream Sanitation Committee.
The Delaware River estuary investigation was supported in part by
the Water Department, City of Philadelphia. The writer is indebted to
Dr. Bernard C. Patten, Associate Marine Scientist at the Virginia
Institute of Marine Science, Gloucester Point, Virginia, for suggestions
used in defining the Cozode.
REFERENCES
1. Vinberg, G. G. Zhachenie fotosinteza Jlya obogashcheniya vody
kislorodom pri samoochishchenii zagryaznennykh vod. (Signifi-
cance of photosynthesis for oxygen enrichment of water during
self-purification of polluted waters.) Trudy Vsesoyuz. Gidrobiol.
Obshchestva. 6:46-69. 1955.
2. Drachev, S. M. Protsessy samoochishcheniya v silno sagryaz-
nennykh rekakh s malym taskhodom. (Processes of self-puri-
Hull
89
-------
fication in very polluted rivers with a small flow.) Vodosnabzh.
i santekhnika. 15(7): 1940.
3. Sibiryakov, M. A. Kizucheniyu kislorodnogo rezhima vodoemov.
Sb. "Sanitarnaya kharskteristika, etc." (Towards the analysis of
the oxygen regime of reservoirs. Coll. "Sanitary characteristics
of reservoirs.") Trudy AMN S.S.S.R. 10:14-43. 1951.
4. Nesmeyanov, S. A. Donnye otlozheniya i kilorodnyi rezhim
vodoemov. (Benthic deposits and the oxygen regime of reser-
voirs.) Izd-vo Akad. Med. Nauk SSSR, Moscow, 1950. 157 pp.
5. Garland, C. F. A study of water quality in Baltimore Harbor.
Publ. No. 96, Chesapeake Biol. Lab., Md. State Dept. Research
Educ. Sept. 1952. 132 pp.
6. Hull, C. H. J. Oxygenation of Baltimore Harbor by planktonic
algae. JWPCF. 35:587-606. May 1963.
7. Gaarder, T., and H. H. Gran. Investigations of the production of
plankton in the Oslo Fjord. Conseil Permanent International
pour l'Exploration de la Mer, Rapport et Process-Verbaux des
Reunions. 42:1-48. 1927.
8. Marshall, S. M., and A. P. Orr. The photosynthesis of diatom
cultures in the sea. J. Marine Biol. Assoc. United Kingdom.
15:321. 1928.
9. Odum, H. T. Primary production in flowing waters. Limnol. and
Oceanog. 1(2): 102-17. Apr, 1956.
10. Ryther, J. H. The measurement of primary production. Limnol.
and Oceanog. 1:72-84. 1956.
11. Ryther, J. H. Photosynthesis in the ocean as a function of light
intensity. Limnol. and Oceanog. 1:61-70. 1956.
12. Verduin, J. Primary production in lakes. Limnol. and Oceanog.
1:85-91. 1956.
13. Goldman, C. R. Discussion of "Photosynthetic oxygenation of a
polluted estuary" by C. H. J. Hull. In: Advances in Water Pollu-
tion Research, vol. 3. E. A. Pearson, ed. Proc. Intern. Conf. on
Water Pollution Research, London, England, Sept. 3-7, 1962.
Pergamon Press, London, England, 1964. pp. 386-90.
14. Kaplovsky, A. J. Estuarine pollution investigation employing
"same-slack" technique. Sewage and Ind. Wastes. 29(9): 1042-
53. 1957.
15. Hull, C. H. J. Discussion of "Photosynthetic oxygenation of a
polluted estuary" by C. H. J. Hull. In: Advances in Water Pollu-
tion Research, vol. 3. E. A. Pearson, ed. Proc. Intern. Conf. on Water
Pollution Research, London, England, Sept. 3-7, 1962. Pergamon
Press, London, England, 1964. pp. 391-95.
16. Steemann Nielsen, E. The use of radioactive carbon (Cu) for
measuring organic production in the sea. Conseil, Conseil per-
manent intern, exploration mer. 18:117-40. 1952.
90
PHOTOSYNTHESIS AS A FACTOR IN OXYGEN BALANCE
-------
17. Hull, C. H. J. Photosynthetic oxygenation of a polluted estuary.
In: Advances in Water Pollution Research, vol. 3. E. A. Pearson,
ed. Proc. Intern. Conf. on Water Pollution Research, London,
England, Sept. 3-7, 1962. Pergamon Press, London, England,
1964. pp. 347-74. Abstract in: London Conference abstracts.
JWPCF. 34:275-76. Mar. 1962.
18. Prescott, G. W. Some relationships of phytoplankton in lim-
nology and aquatic biology. In: Problems of Lake Biology. Am.
Assoc. Advance. Sci. Publ. No. 10. 1939. pp. 65-78.
19. Fish, F. F., C. H. J. Hull, B. J. Peters, and W. E. Knight. A study
of the effects of a submerged weir in the Roanoke Rapids Reser-
voir upon downstream water quality. Special Report No. 1,
Roanoke River Studies. Compiled by Special Report Committee
for Committee for Roanoke River Studies, Feb. 6, 1958. 63 pp.,
23 tables, 63 figures, Appendixes A-F.
20. Ryther, J. H. 1954. The ratio of photosynthesis to respiration in
marine plankton algae and its effect upon the measurement of
productivity. Deep-Sea Research. 2:134-39. 1954.
21. Streeter, H. W., and E. B. Phelps. A study of the pollution and
natural purification of the Ohio River. III. Factors concerned in
the phenomena of oxidation and reaeration. Public Health Bull.
No. 146. Feb. 1925. Reprinted in 1958. 75 pp.
DISCUSSION
Dr. Jacob Verduin
Bowling Green State University
Bowling Green, Ohio
Hull's researches have a close parallel with much of the work
that has interested me during the past 15 years. My motive was
basically one of establishing aquatic ecologic relationships rather than
making applications to sanitary engineering problems. Polluted waters
have interested me only as one end of the aquatic habitat spectrum,
but many of the data gathered can be drawn on for comparison with
those of Hull. Our experience with clear and dark bottles has shown
that photosynthetic yields under such conditions are lower than those
that occur under completely natural conditions and that respiration
rates in the dark bottles are higher than those in nature. Conse-
quently, Hull's estimates of the oxygenation capacity of a phyto-
plankton community are conservative ones, and his estimates of the
depth of the euphotic zone are conservative when they are based on
enhanced respiration rates inside black bottles. Our studies under
completely natural conditions have revealed photosynthetic yields
amounting to 6 grams of carbon fixed per square meter of lake
surface per day during the summer months. This work was repeated
during several summers and represents the mean of many determina-
Verduin
91
-------
tions. Converted to oxygen production, it amounts to 16 grams of
oxygen produced per square meter per day, which is similar to the
highest values reported by Hull.
Another aspect of our studies has a bearing on this problem. We
have observed that most ponds and lakes usually exhibit a fairly close
balance between diurnal photosynthetic oxygen production and noc-
turnal respiration by the total community. If a sudden supply of
organic matter is added, as for example by autumn leaf fall in a
small pond, an excess of respiration over photosynthesis will develop
with a consequent excess of C02 supply, as is most conveniently de-
tected by reduced pH values. This enhanced CO., supply does, how-
ever, stimulate the autotrophic community to higher photosynthetic
yields and a new balance is soon achieved in which the photosynthetic
O., production by day matches the nocturnal O., consumption of the
total community, including those components that feed on the added
organic matter. Conversely, periods of photosynthetic excess may be
observed. In Western Lake Erie, these are most evident in April
when the spring phytoplankton pulse usually drives the pH values
up to about 9.0. The C02 deficiency (and others as well, no doubt)
resulting from this imbalance tends to reduce the photosynthetic
capacity, and the enriched supply of organic matter tends to stimulate
reducer components in the total community, resulting in a return
to the balance described.
Hull's observation that there are autotrophs capable of photo-
synthesis at low light intensities also has been confirmed by our in-
vestigations. Some of the highest phytoplankton densities and oxygen
concentrations we ever observed were found in January under a sheet
of ice in Urschel's Quarry where a Chlamydomonas community flour-
ished in light supplies that averaged less than 2 percent of full sun-
light. Such high production in 0°C water casts some doubt on the
widely held opinion that photosynthesis is inhibited by low tempera-
tures. Because efficient phytoplankton crops develop wherever light
and nutrient conditions are favorable, it seems one is justified in as-
suming that, within reasonable limits, C02 from any source in an
aquatic environment will be converted to 02 by the photosynthetic
process. By reasonable limits, I mean that it is obviously possible to
inject loads of organic matter into a pond that will overload the
capacity of the autotrophic community to perform this conversion.
But the eminent success of the sewage stabilization pond device
(which confounded the expectations of many sanitary engineers)
attests to the effectiveness of the process.
My personal reaction to the increase of impoundments and reser-
voirs in our civilization is most favorable. I would like to see the
trend accelerated considerably. I am convinced that the best thing
we can do with the millions of tons of water that cascade down our
streams today is to create impoundments that will hold these waters
quietly for a while, exposing them to the purifying action of the
phytoplankton community, thus improving the water quality while
providing numerous other well-known and beneficial byproducts.
At Bowling Green, we are considering establishment of a shallow
92
DISCUSSION
-------
pond to hold the effluent of the city sewage treatment plant for a few
weeks before sending it down the Portage River. It is hoped that this
pond can he excavated by judicious borrow-pit management during
the construction of the bypass for highway 75. Such a pond will
further improve the quality of the effluent (which is already well up
to public health department standards as a result of recent remodel-
ling of our treatment plant) and will serve as a safety factor in
emergencies. Many other cities would be well advised to consider
similar projects, especially where they can be designed as a byproduct
of other construction. The resulting ponds should provide fishing for
the young as well as improve the quality of water on the downstream
side of the city.
TABLE 1. QUANTITIES DERIVED FROM INVESTIGATION OF
PHOTOSYNTHESIS IN NATURAL PHYTOPLANKTON
COMMUNITIES
Average phytoplankton density
Photosynthetic yield
C02 absorbed at optimal light
C02 gross absorbed, summer
Net C02 absorbed (in excess of
diurnal respiration by total
community), summer
Estimated C02 exchange with
atmosphere, (02 exchange is likely
to be of the same order of
magnitude)
Rate coefficients
CO;, transfer across air-water
interface in ponds
02 transfer across air-water
interface in rivers
1-6 microliters per liter
1-8 micromoles per microliter
of phytoplankton per hour®
500 millimoles per m2 per day
250 millimoles per m2 per day
13 millimoles per m2 per day
1.0-3 cm per secb
range is 10~2 to 1Q~5 cm per
sec depending on degree of
turbulence
4 x 10-3 cm per secb
a Much lower values may be encountered in dense phytoplankton
communities.
D Q D Q
11 This is —in the Fickian equation: = — A C where
Li AT L AT
is transport per square centimeter per second and A C is the concen-
tration difference between the water sample and the interface, which
is assumed to be in equilibrium with the atmosphere. Hence, a
liquid boundary layer of unkown thickness L is presumed to be the
major transport barrier.
Verduin
93
-------
Rivers, as they exist in most of the United States, are too turbid
to provide a favorable environment for photosynthetic processes.
The average rate of oxygen transport into an O, deficient river (O.,
near zero) is about 30 grams per square meter of surface per day*
This computation is based on an average transfer coefficient across
the air-water boundary layers of 4 x 10";! centimeters per second
(_P.= 4 x 10-3 in the Fickian diffusion equation). This value was
L
computed from data for several rivers in Tennessee, gathered by M. A.
Churchill. Similar values have been obtained experimentally with
Lake Erie water, mechanically stirred in a finger bowl. The O , con-
tent of rivers is, however, seldom near zero. Consequently, this
maximal estimate must be reduced in proportion to the departure
from atmospheric equilibrium. In O,, deficient rivers, however, O,,
absorption from the air is frequently the dominant source of oxygen.
Table 1 gives the quantities derived from our years of investiga-
tions of the aquatic photosynthesis. These may aid in making first
approximations of the 02-producing capacity of natural phytoplank-
ton communities.
DISCUSSION FROM THE FLOOR
Mr. O'Connor, Manhattan College: Dr. Hull, in your paper you
indicated there may have been some vertical gradients in DO in the
Roanoke Rapids Reservoir case. Could you give me an order of
magnitude of what they were?
Dr. Hull: Pritchard and Carpenter's analysis of oxygen changes
in Roanoke Rapids Reservoir was based on the top 1-meter layer of
water. At the depth of 1 meter we could find no detectable oxygen
gradient; we were well aware, however, that one could occur at times.
Taking this into account by assuming a conservatively high oxygen
gradient of 0.1 milligram per liter per meter of depth at the 1-meter
depth, and going through the calculations again, Pritchard and
Carpenter found that this added factor of a vertical flux across the
bottom of the 1-meter layer of water tended to increase the photo-
synthetic production and decrease the atmospheric exchange co-
efficient. One-tenth of a milligram per liter per meter of depth was
greater than we ever observed at the 1-meter depth.
Mr. O'Connor: You indicated that the BOD exertion was 0,5
milligram per liter per day. At what depth was that BOD sample
taken and at what time of day?
Dr. Hull: The respiration value of 0.5 milligram per liter per
day was obtained by observing the DO change in a dark bottle
suspended near the surface for a 24-hour period. The sample was
obtained near the water surface after dark.
Mr. O'Connor: Dr. Verduin, from my own readings in the lim-
nological area, I suppose the assumption is usually made that the
94
DISCUSSION
-------
respiration rate through the course of the daylight period is constant,
as was assumed in the paper by Dr. Hull. Are there any data on
diurnal variation in respiration rate, and if not, would you express
an opinion as to how that rate might vary throughout the course of
the daylight hours?
Dr. Verduin: I cannot give you the actual reference for the
paper in which I have reported this, but I have done some work on
the respiration rate in the afternoon as compared to midnight and
morning. I did it by collecting samples; by exposing them for photo-
synthetic measurements; then putting these samples that had been
exposed to light, in the dark; measuring their respiration after a
5-hour interval; and measuring again 8 hours later. I found that the
respiration rate was nearly twice as high during the 5-hour period
immediately after exposure to light as it was during the second
period that would correspond, let us say, to from 10 o'clock at night
to 6 the next morning.
I think this is a fairly good estimate of the kind of respiration
increase that occurs under illuminated conditions. This has been
reported by others working with cultures. In other words, when you
expose phytoplankton to light, they build up a substrate that supports
a higher rate of respiration, roughly about twice as much.
Mr. Churchill, Tennessee Valley Authority: I have a question
along the same lines. Some of our English contemporaries maintain
that respiration rates of phytoplankton are higher at high oxygen
concentration than at low concentrations. Do either of you gentlemen,
or anyone else in the room, have any data on this?
Dr. Hull: I have not had any personal experience with this prob-
lem, but information in the literature is a little contradictory. Zobel
has reported that no relationship exists between respiration of bac-
teria and oxygen concentration between the limits of approximately
0.3 and 36 milligrams per liter. Only in one or two of our experiments
on the Delaware River did oxygen concentrations approach either
end of this range. We did get oxygen concentrations as high as 32
milligrams per liter at the end of a few, long-term, 5-day experiments.
Other literature (I cannot cite the references) indicates there is
a relationship between the photoinhibition effect of supraoptimal
light intensity and oxygen concentration.
Dr. Verduin: I do not know what kind of experiments you are
referring to, Milo, but in my experiments the oxygen concentration
was higher during the first 5 hours than it was later, so that a correla-
tion might exist between the oxygen concentration and respiration
rate. I did not, however, have such phytoplankton concentrations that
the oxygen was seriously depleted, so I interpreted this as being due
to changes in substrate rather than change in oxygen concentration.
Dr. Lewis, U. S. Public Health Service: One other point might be
brought up here. We have been considering optimal conditions. If,
however, there is a period in which good sunlight is available over
From the Floor
95
-------
several days so that quite a large growth of phytoplankton organisms
results, followed by a period of cloudy weather, the respiration re-
quirement of the organisms will be greater. Changes in temperature
and other conditions may cause considerable variation in respiration
requirements.
How often a good increase in oxygen by these organisms could
be expected during the day would be of interest to the engineers who
are considering the use of these organisms in supplying oxygen.
Dr. Hull: We examined this factor in our work on both the Dela-
ware River and the Roanoke Rapids Reservoir. Variations do occur,
of course, and on cloudy days you get less production, A very wide-
spread misconception, however, is that on cloudy days no photo-
synthesis occurs. My Figure 1 showing the variation of photosynthesis
with depth showed measurable production in Baltimore Harbor
down to depths of some 14 feet. Even on the brightest, sunny day it
is pretty cloudy at a depth of 14 feet so all photosynthesis on cloudy
days cannot be ruled out. In fact, on the Delaware River we found
concentrations of 2 milligrams per liter or more in the surface samples
even when it rained all day and we did not see the sun at all.
Dr. Heukelekian showed as early as 1931 that these organisms
did not need direct light at all. They can liberate oxygen in very
diffused light. The literature is full of information indicating that
very low levels of surface light, less than 1 percent, support photo-
synthesis. So, although variability exists, as was pointed out, this
factor must still be considered in the oxygen balance. It is a significant
factor, especially at shallow depths.
Mr. Churchill: I would like to add a little fuel to what Jack just
said. We have been working on a very deep and very sluggish river
in our area, and we anticipated a net oxygen use by phytoplankton.
But how wrong we were! In three separate 24-hour studies — one
in July, one in August, and one in September — we actually found a
net addition of oxygen to this very deep, very sluggish stream by the
phytoplankton. So, for my money, phytoplankton are pretty good
oxygen producers.
96
DISCUSSION
-------
CONTROLS ON SOLUTION AND PRECIPITATION
IN RESERVOIRS*
S. Kenneth Love and Keith V. Slack
U. S. Geological Survey
Washington, D. C.
INTRODUCTION
Reservoirs have many variable characteristics. Some are related
to natural processes whereas others are related to management prac-
tices. It is the purpose of this paper to discuss both kinds of charac-
teristics insofar as they have a bearing on the solution and precipita-
tion of inorganic and organic substances in reservoirs.
The composition of water that flows into a reservoir is controlled
by the interaction of climate, geology, topography, biota, and time,
as shown by Gorham(l). When this water comes into contact with
solids different from those with which it has been in contact or when
unlike waters are mixed, chemical reactions may occur (2, p. 218)
The environmental events that bring about encounters between water
and new solid phases or between unlike water masses in reservoirs
will be examined first, followed by a more detailed discussion of the
chemical reactions that result.
THE RESERVOIR ENVIRONMENT
Reservoirs are designed for a variety of purposes, and both the
type of construction and the management greatly influence water
quality. Two principal kinds of impoundments were recognized by
Kittrell(3): (1) storage reservoirs, characterized by high dams on
streams with relatively steep gradients, large surface areas with
many extensive embayments, low velocity as a consequence of slow
passage of water through the impoundment, variable surface levels,
deep location of intake penstocks, and typical thermal stratification,
which commonly leads to a deficiency of dissolved oxygen in bottom
waters; and (2) main stem reservoirs, characterized by low dams,
small surface areas with much of the water confined to the original
channel, more rapid flow and exchange of water, relatively constant
surface levels, intake penstocks near the surface, small temperature
gradients, and dissolved oxygen intermediate between that of un-
regulated rivers and storage reservoirs. The summer distribution of
temperature in the two classes of reservoirs is shown in Figures 1
and 2.
The distribution of dissolved oxygen is of utmost importance in
determining the composition of discharge water and is related to the
cycles of nearly all of the important elements. Hutchinson(4, p. 701)
suggested that the presence of an oxidized surface layer on sediments
throughout the year in some lakes and its disappearance seasonally
•Publication authorized by the Director, U. S. Geological Survey.
Love and Slack
97
-------
^ DAM
WATER SURFACE
IN PLOW
Figure 1 — Representative profile showing summer stratification in a typical storage reser-
voir. (Ref. 3).
98
CONTROLS ON SOLUTION AND PRECIPITATION
-------
in the deeper parts of the basin in others may be the most important
phenomenon in the chemical classification of lakes. The occurrence
or absence of this so-called oxidized microzone and the related con-
centration of iron is intimately involved in most of the chemical
transformations with which we are concerned. This subject will be
discussed in more detail in a later section.
CLIMATIC CONTROLS
The climatic variables that influence composition of reservoir
water are precipitation, temperature, wind, evaporation, and solar
radiation. Solar radiation is included because of its role in photo-
synthesis. The part played by climate in weathering, erosion, plant
growth, and other processes outside of the impoundment will not be
considered. The contributions of these sources to the chemical con-
tent of natural waters has been reviewed recently by Hutchinson{4)
and Gorham(5). It is enough to note that the composition of inflow
water reflects the conditions of its origin and that climate is one of
the primary controls on the availability of substances for transporta-
tion by water.
The composition and temperature of inflow waters are particu-
larly significant. If inflow water differs in composition or temperature
from that of the water mass in the reservoir, chemical reactions may
occur or density flows may result. In general, the dissolved solids
concentration is lower during periods of high flow; however, concen-
trations of suspended solids usually are greater during high runoff.
Temperature
Both the concentration and proportion of ions in solution may
be affected by such physical processes as freezing, dilution, and
evaporation, all of which are manifestations of climate. The forma-
tion of an ice cover may be a regular event in northern areas. Freez-
ing is initiated by sudden cooling of the reservoir surface during a
calm, cold night, and a stratification of low stability in the top few
meters results(4, p. 453). This inverse stratification is preserved
under the ice. Size of a reservoir has an effect upon the probability
of freezing. A deep basin loses heat more slowly than a shallow one,
and a smaller surface is more likely to remain undisturbed during
potential freezing conditions.
In his studies of English lakes, Mortimer(6) found that freezing
removed 94 percent of the dissolved salts, all humus coloring matter,
and a large part of the dissolved gas content. This led to increased
concentrations in the water immediately under the ice. In Alaska,
the increase in ice thickness on lakes was accompanied by an almost
linear increase in mineral content of the unfrozen water just below(7).
Highly mineralized waters of closed basins in Saskatchewan exhibited
alternate freezing out and re-solution of sodium sulfate crystals dur-
ing cold and warm periods(8, p. 149). It is unlikely that the freezing
out of dissolved substances is a major control upon the composition
of reservoir waters; however, stagnation resulting from an ice cover
can produce significant changes in composition as was shown by
Love and Slack
99
-------
Mortimer(6). Kittrell(3) stated that winter stagnation in northern
reservoirs does not result in total oxygen depletion in bottom waters,
owing to the high initial dissolved oxygen concentration, the low
rate of oxygen demand, and the shorter period of stratification.
Thermal stratification has a major influence on the water chem-
istry of reservoirs. The winter temperature pattern in a typical
temperate zone lake is characteristically isothermal. There is com-
plete vertical circulation as a consequence of the uniform density.
In spring, solar heating warms the waters of the surface and tribu-
taries until a density difference results sufficient to prevent mixing
of the warm upper water with the colder, denser water below. As
the surface becomes progressively warmer the density gradient
steepens and the depth to which wind can mix the upper waters is
diminished. The result is a separation of the reservoir into two iso-
lated masses of water — a circulating surface volume floating upon
a relatively stagnant volume below. This stratified condition typically
persists until autumn, when cooling of the surface results in the
diminution and eventual breakdown of thermal stability. At this
time the reservoir "overturns" under the influence of wind, resulting
in circulation to the bottom.
Wind keeps the water in circulation unless a persistent ice cover
forms. In that event, an inverse type of stable thermal stratification
persists until it is destroyed by warming and wind action. If winter
stagnation occurs, it is followed by a spring overturn and a second
period of circulation.
Although this generalized picture undoubtedly holds for many
reservoirs, especially those having intakes at shallow depth, Church-
ill's^, 10) reports on TVA impoundments clearly show wide de-
partures from the ideal model when the intakes are deep. These de-
partures are due primarily to the location of the intake penstocks
through which water is discharged from the reservoir. The type of
control exerted in this way may be considered as a special type of
morphometric influence on water quality. The draft through the
penstocks, of course, is an aspect of management control, since there
is evidence that except for this factor a more typical thermal strati-
fication would prevail.
The following generalized statements seem to apply to the sea-
sonal thermal patterns of reservoirs as indicated by the references
previously cited:
1. During cold months the water circulates freely (in absence
of ice cover) and tends to be isothermal. Discharge is from
the entire cross section of the pool regardless of penstock
depth. The flow pattern is diffuse.
2. Shortly after stratification occurs in early summer the flow
from a deep intake is from the hypolimnion. Water at or
above the level of the intakes is discharged first according to
the density. Warm surface water descends to replace the
withdrawn cold water. In reservoirs with a shallow intake,
flow is confined to the mixed surface layers.
100
CONTROLS ON SOLUTION AND PRECIPITATION
-------
3. By late summer or early fall most of the cold winter water
has been removed by discharge from deep intakes. Stratified
withdrawal of warmer water then occurs.
4. In late fall inflow from tributaries is cooler (denser) than
reservoir waters and tends to form density flows until such
time as complete circulation develops.
Large variations in salinity, temperature, or suspended solids of
inflowing waters may produce density flows. Anderson and Pritch-
ard(ll) identified four characteristic types of inflow water that ac-
counted for the seasonal distribution of salinity and circulation pat-
terns in Lake Mead (Figures 3 and 4). The first was a homogeneous
winter water with a temperature of about 50 °F and salinity near 700
ppm. Type two occurred during the spring high-flow period. The
temperature was about 67°F and the salinity 275 ppm. The summer
water had a temperature of about 76°F and a salinity of 800 ppm,
reflecting the reduced inflow. The fourth type occurred during the
fall when inflow was further reduced. Temperatures at this time
were near 66°F and salinity about 900 ppm.
Figure 3 — Sectional salinity distribution and circulation patterns, Lake Mead.
Love and Slack
101
-------
Figure 4 — Seasonal variations in temperature gradients. Lake Mead, 1948.
The influence of density currents on water quality has been
described by Wiebe(I2), Churchillf 13, 9, 10), and Love(I4). They
described three types of density flows: overflows, interflows, and
underflows. The significant point for our purpose is that water masses
are brought into contact with solids or with waters of different com-
position, the results of which may produce chemical reactions.
One of the most significant aspects of density currents is their
effect on dissolved oxygen. Wiebe( 12) reported observations in TVA
impoundments in which he related oxygen minima near the thermo-
cline to density interflows rich in decomposable organic matter. The
opposite effect was reported by Bryson and Suomi(15). They re-
ported that oxygen increased in the hypolimnion of Lake Mendota
as a result of a density flow following a heavy rain. Hutchinson (4,
p. 297) believes there is evidence for the general occurrence of density
currents running down the slope of the bottom of a lake, as dissolved
material diffuses from mud into the water. These currents probably
are best developed during summer stagnation when reduction proc-
esses liberate ionic material.
Wind
Wind influences ionic concentration by affecting evaporation(16)
and by inducing water movements. Wind also delivers dust or dry
fallout to the reservoir, some of which is soluble. Most of the wind-
transported matter probably is of local origin and does not differ
greatly from that supplied by runoff. In a similar way, material may
be removed from the exposed margins of the basin by deflation dur-
ing periods of low water, resulting either from evaporation or from
regulated drawdown. Deflation has been shown to be important in
natural lakes in semi-arid regions (17), but its importance in reser-
voirs in other climatic areas has not been reported.
102
CONTROLS ON SOLUTION AND PRECIPITATION
-------
Two types of currents are generally present in natural lakes:
water movement from inlet to outlet and dominant wind-driven
circulation(4, p. 363). There is reason to believe that the hydro-
graphic slope current system dominates in many reservoirs, especially
at certain periods of the year. They are, for example, responsible
for density flows. Nevertheless, wind acting on an unfrozen water
surface produces turbulent movement of water, resulting in transfer
of material and heat. Winds are one of the main causal agents for
seiches. The most important type of deep-water movement in strati-
fied lakes is believed to result from internal seiches.
Evaporation
Evaporation combines the effects of wind and temperature to
produce one of the major water losses in most reservoirs. For exam-
ple, it has been estimated that approximately 12.3 million acre-feet
of water is lost annually by evaporation from the principal reservoirs
in the 17 western states(18). Most of the evaporation occurs during
the warm months and in drainage basins with large reservoirs. It
was concluded from this study that total water loss by evaporation is
more closely related to the status of development in a river basin
than to anything else.
Evaporation losses from Lake Mead are about 5 percent of the
average inflow, or about 700,000 to 900,000 acre-feet per year. The
effect of evaporation on water composition was investigated by
Howard(J9, p. 123) who found increasing concentration of soluble
salts remaining in solution. Small increases in magnesium, potassium,
and nitrate were observed in the outflow from the impoundment.
These increases in ionic concentration were partly offset by decreases
resulting from precipitation of dissolved materials. Howard esti-
mated that during 14 years of storage Lake Mead water precipitated
more than 1 million tons of silica and in excess of 9 million tons of
calcium carbonate. It is doubtful whether evaporation or inorganic
precipitation are the only controls in this case.
INORGANIC CHEMICAL REACTIONS
We have seen how climatic variables function in some of the
compositional changes that occur in reservoir waters. We will exam-
ine now the chemical changes that result from the physical influences
on distribution of water and materials.
Solution
When a liquid comes in contact with a solid, solution can occur
only if atoms, ions, or molecules of the solid are strongly enough
attracted by the liquid so that the strong forces that hold the particles
in the solid are overcome. Most inorganic compounds go into solution
as positive and negative ions rather than molecules. All inorganic
ions are soluble unless combined with some other ion that prevents
its going into solution. For example, silver nitrate and sodium chloride
are both soluble, but silver chloride is only slightly soluble.
Love and Slack
103
-------
The quantity of a material that can go into solution in pure water
at a given temperature is expressed by its solubility-product con-
stant. This is a quantitative statement of the limit of solubility of
any slightly soluble substance that forms ions. When the product
of the concentrations of the ions in the solution exceeds the value of
the solubility-product constant, either precipitation occurs or a
supersaturated solution is formed. Supersaturated solutions tend to
be unstable, and precipitation is the usual result of excess concen-
tration of the ions.
The solubility of a gas is related to the ratio of concentration of
gas in solution to that in the gas phase in contact with the solution.
Such gases as oxygen, nitrogen, and carbon dioxide are usually present
at or near atmospheric saturation levels during the period of the year
when the water is in complete circulation. Under reducing conditions
that prevail in the hypolimnion during stagnation, other gases are
present, sometimes in sufficient concentration to form bubbles that
rise to 'the surface. Methane is one of the products of anaerobic de-
composition, but carbon dioxide, nitrogen, ammonia, hydrogen, and
hydrogen sulfide may also be present.
Solution of pre-existing material within the land area of an
impoundment is a familiar example of a solubility process. Trees,
brush and other obvious sources of organic matter are commonly
burned or removed during construction. For example, TV A im-
poundments are cleared of trees and brush in the zone between full
pool level and the normal level of maximum drawdown(9). The trees
and brush remaining below the summer thermocline tend to deplete
oxygen and produce hydrogen sulfide, although it is reported that
little additional influence on water quality remains after a full annual
cycle of operation. Some of the problems that can arise during site
preparation are described by Benedetti and Roller (20) for the upper
Green River, Washington. Hydrogen sulfide was leached by a small
stream from hydrothermally altered rock of an extinct hot spring,
slopes had to be stabilized with rip-rap, and a swamp deposit consist-
ing of 600,000 cubic yards of organic material was encountered. A
test covering of bank-run sand and gravel was tried as a means of
sealing off this deposit, but methane gas was evolved from around
its periphery. Investigation of this problem was being continued be-
cause of its possible effect on water stored.
Lake Mead provides a well-documented example of solution from
an inundated reservoir area. Releases from this lake initially showed
higher dissolved solids content than the average for the inflow(29).
From records of dissolved solids inflow and outflow, it was estimated
in 1948 that a total of 20 million tons of soluble salts had been dis-
solved from the bed, but that the concentration had only increased
bv half this amount because of precipitation of calcium carbonate
and silica from the stored water. The leached material consisted
largely of gypsum (calcium sulfate) and some halite (sodium chlo-
ride) As a result of this solution, sulfate concentration in the out-
flows increased more than 50 percent over that in the inflowing water
104
CONTROLS ON SOLUTION AND PRECIPITATION
-------
in several years. Calcium increased in the outflow despite the large
amounts precipitated as calcium carbonate within the basin. Increased
concentrations of sodium and chloride are attributed partly to solu-
tion of halite. Since 1951 the differences in concentration have been
relatively small, salinity being sometimes higher in the inflow and
sometimes higher in the outftow(24).
Many substances undergo an annual cycle of solution and precipi-
tation correlated with the cycle of stratification. An example of this
type of solubility cycle is provided by calcium carbonate, which
precipitates from surface waters under certain conditions and redis-
solves at the bottom in the presence of carbon dioxide produced
during decomposition.
Precipitation
Solids may be dispersed in water in one of three different forms,
depending on the particle size. These are solutions, colloids, and
suspensions. Particles in solution are single molecules or ions only
a few angstrom units (A) in diameter. Colloidal particles are of the
order of 10 to 2000 A in diameter and may consist of many molecules.
They are of such small mass that they exhibit Brownian movement
and do not settle under the influence of gravity. Colloids differ from
suspensions only in size.
As mentioned earlier, a substance is precipitated when its solu-
bility product is exceeded. All precipitations from natural waters
are made from solutions that contain ions other than those of the
precipitate, and some of these ions are frequently dragged down with
the precipitate, a process known as coprecipitation. The amount of
coprecipitation varies with the nature of the precipitate and the con-
ditions of its formation. A precipitate first forms as particles of col-
loidal size before they become large enough to settle out. Coprecipi-
tation usually occurs in the colloidal stage.
Under certain conditions of formation of ferric hydroxide pre-
cipitates, such as when solutions containing ferrous iron are raised
in pH or redox potential, other ions may be removed from solution
by coprecipitation (21). Hem and Skougstad studied the coprecipita-
tion effects of ferrous, ferric, and cupric ions in laboratory solutions.
They noted that the effects of ferric hydroxide precipitation on the
content of other heavy metal ions in natural waters are approxi-
mately indicated by the behavior of copper. The results of these
experiments suggest that the conditions most favorable for coprecipi-
tation include a neutral or alkaline pH so that the normal positive
charge on ferric hydroxide particles is neutralized. If precipitation
occurs at low pH some anions may be adsorbed. Colloidal suspen-
sions of ferric hydroxide in natural water probably do not carry any
large amounts of other cations by adsorption on the hydroxide
particles because of the small concentration of ferric hydroxide that
can occur in stable suspensions; however, the behavior of some of the
trace constituents in natural water may be affected by adsorption on
colloidal ferric hydroxide.
Love and Slack
105
-------
It has been suggested that the scavenging effect of ferric hydrox-
ide precipitates may explain the presence of certain rare metals in
seawater in amounts much less than saturation(22, 23, 21, p. 96). This
effect has apparently not been studied in inland waters. The work of
Aarnio(24) on coprecipitation of humus and ferric hydroxide has been
discussed by Hutchinson(4, p. 711). Since much smaller amounts of
Al, 0.( were needed to form a precipitate with humus material, it was
suggested that in nature aluminum may be rapidly depleted, leaving
the iron in solution.
Colloids occur in aqueous solutions as suspensoids and emulsoids.
Suspensoids have low viscosity and little affinity for water. Because
of their physical similarity to solutions they are known as sols, or in
water, hydrosols. Dispersions of metals and salts in water are exam-
ples. Emulsoids have high viscosity and a high affinity for water.
They are known as gels, or in water, hydrogels. Phenomena associ-
ated with surfaces are intimately involved with particles of colloidal
size because of the very great surface area per unit mass.
The properties of surfaces are related to exposed atoms or ions
lying on the surface. Within a crystal each ion is surrounded by op-
positely charged ions except for those at the surface. These surface
ions or atoms can attract and retain other ions or atoms by adsorption.
Adsorption occurs at all surfaces, large or small, but the quantity per
unit area is so small that it usually becomes significant only when the
adsorbate is finely divided, as in the colloidal state.
Many substances function as adsorbents in nature. These include
finely divided crystalloids, especially clay particles of the montmoril-
lonite group, organic and inorganic colloids, gels of ferric hydroxide
and silicic acid, humus colloids, polymorphic inorganic and organic
complexes, and surfaces and integuments of living and dead organ-
isms^, p. 187). The role of ferric hydroxide in controlling phosphate
has been described by Einsele(26). Hutchinson(4, p. 886) summarizes
the effects of organic colloids on some physical properties of lakes.
Some of the ecological relationships of sorption reactions were
studied by Carritt and Goodgal(27). They proposed a two-part mech-
anism, rapid adsorption followed by a slower diffusion-controlled
process, to explain the observed uptake of phosphorous ions from
solution. This process suggests that during desorption both the amount
of phosphorus and the rate of its removal depends on the length of
time that the complex has had to form. With long contact time the
desorption is slow, and this stability of the phosphorus-solid complex
is important because of the transport of suspended substances through
regions of different pH, temperature, and salinity. It was further
shown that many dissolved substances compete in sorption reactions.
For example, phosphorus reduced the uptake of sulfate.
The initiation of crystallization from a supersaturated solution
is not completely understood, but it is known that attainment of equili-
brium may be very slow for some natural reactions. Sooner or later
crystal nuclei appear; the greater the relative supersaturation of the
solution, the more nuclei occur. Precipitation may proceed by co-
106
CONTROLS ON SOLUTION AND PRECIPITATION
-------
agulation of colloidal particles and by deposition of ions from solution
to build up the crystal lattice of the nuclei. Both usually occur, and
the domination of one process over the other depends on the condi-
tions. In the laboratory it is found that keeping the degree of super-
saturation low favors growth of original nuclei by deposition of ions
from solution. Conversely, rapid mixing of reagents produces in-
creased supersaturation and favors the formation of more nuclei. The
importance of these relationships to natural systems in reservoirs is
apparently unknown.
The formation of marl may be used as an example of natural
chemical precipitation. Marl is deposited from a solution in which
the solubility product of calcium carbonate is exceeded. The revers-
ible reaction involved may be represented by
Ca(HCO:))2 = C02 + H20 -f CaCO, .
Removal of carbon dioxide by any of several means drives this re-
action to the right. Welch(28) lists the following mechanisms that
can result in the reduction of CO., concentration: activities of living
organisms, agitation of water, a rise in temperature, evaporation, and
incorporation of carbon dioxide in bubbles rising from bottom deposits.
In the sea large quantities of marl are thought to be deposited
as a result of bacterial action, but no such relationship has been
demonstrated in fresh waters. The process considered to be of prime
importance in lakes is the utilization of carbon dioxide by plants for
photosynthesis in accordance with the following reaction:
6 HsO + 6 CO, = CaHiaOfl + 6 02 .
Calcium carbonate that precipitates in the photosynthetic region
settles into the decomposition zone where it may combine with carbon
dioxide to increase the bicarbonate content. Ruttner(25) showed that
in low calcium waters the calcium carbonate precipitated is insuffi-
cient to fix the available C02 and there is no accumulation of marl.
In many lakes there is a net removal of precipitated carbonates by
the subsequent deposition of inorganic sediment that forms an
effective seal.
Ruttner(25) reported that 100 kilograms of fresh Elodea cana-
densis precipitated 2 kilograms of calcium carbonate in a day with
10 hours of sunlight. Howard(19, p. 123) estimated that 9 million tons
of calcium carbonate were precipitated in Lake Mead. He calculated
that the amount precipitated from the water accounts for only 0.45
percent of the 10.3 percent estimated to be present in the 2 billion
tons (dry weight) of sediment accumulated in Lake Mead during the
same 14-year period. Evidently large quantities of calcium carbonate
are brought into the lake with the suspended load of the Colorado
River.
Complex Formation
The term "complex" is applied to many natural substances of
poorly defined composition, especially if the chemical species is partly
Love and Slack
107
-------
organic. The chemical definition of a complex is more specific and
implies a definite structure consisting of a central atom or ion sur-
rounded by atoms, ions, or molecules in a stable configuration. Com-
plexes in solution may originate by direct solution of complex min-
erals from rocks or by reaction of dissolved substances with simple
minerals to form complex ions.
Chelates are a special type of complexes between inorganic and
organic structures. A chelate compound or ion is a ring structure
containing a central atom, usually of a metal, tied by coordinate and
polar valence bonds to an organic segment. Chelate ions or com-
pounds generally behave very differently from the uncomplexed
metallic ions. The term "chelation" is sometimes loosely applied as
a general term for formation of any complex of a metal and an
organic ligand. Organic complexes were reviewed by Bjerrum,
Scharzenbach, and Sillen( 29).
Complexes directly affect the composition of water by inducing
or preventing precipitation. When complexing takes place, both redox
potential and pH may be altered and the rate of oxidation or reduc-
tion may be slower than when complexing is absent (3D).
Hem(30) noted that iron combines readily with many other ions
to form complexes that generally have an ionic charge. In most cases
the charge is positive, although neutral and negatively charged com-
plexes exist. Among the iron complex formation reactions listed by
Stumm and Lee(3I) those of possible significance in reservoirs are
the sulfato and phosphato complexes ([Fe(SO,)]^ , (Fe(HPO,)]-t ,
[Fe(HPO,)J-). These workers suggested that complex formation
may be partially responsible for the high concentrations of humic
acid and lignin derivatives often associated with high concentrations
of soluble iron. Colloidal ferric hydroxide may be stabilized as a
sol by organic compounds such as the yellow coloring matter studied
by Shapiro(32). In his discussion of silica, Hutchinson(4, p. 799)
wrote that he believes there is evidence for the existence of colloidal
silica and possibly of complex aluminosilicate ions such as Al(OH)
(HSiO.,)+ or Al(HSiOit) + "l-. Ingols and Wilroy(33) noted that the
presence of natural tannins seems to induce precipitated manganese
to dissolve, and it was suggested that organic complexing of trivalent
manganese may keep this element in solution. Mortimer(6) sug-
gested that cations adsorbed on sediments may be released into
anaerobic bottom waters during summer stagnation by the break-
down of an iron-silica-humus adsorption complex. At the onset of
autumn circulation this complex reforms and the cations again are
adsorbed. Gorham(5) indicated, however, that he believes the marked
fall in alkalinity of the bottom water at overturn, cited by Mortimer
as evidence of base adsorption, may be due mainly to dilution of a
very small volume of concentrated hypolimnion water with a large
volume of more dilute epilimnion water.
Oxidation-Reduction Reactions
Oxidation is an increase in positive valence resulting from loss
10S
CONTROLS ON SOLUTION AND PRECIPITATION
-------
o£ an electron by an atom, ion, or molecule. Reduction is the opposite
process, a gain of electrons or loss of positive valence. In the reaction
4 Pe' ; -f 2 S 4 Fc -U S2
sulfide ion is oxidized to sulivjr, and ferric ior. is. reduced to ferrous-
ion.
Under ideal conditions all oxidation-reduction reactions are re-
versible and none ever goes entirely to completion. For example, the
reaction
Fe~U4- — Fe^—I- -L- electron
may go in either direction, depending on the presence of other sub-
stances that furnish or accept electrons. In general, ions with the
greatest tendency to acquire electrons will do so at the expense of
other ions present.
Oxidation-reduction has two aspects of importance: intensity, as
expressed by the electrical potential with reference to an arbitrarily
chosen standard, and the buffering, or poise of the system, i.e., the
ability to carry on a given amount of oxidation or reduction without a
significant change of potential^, p. 195).
The theoretical aspects of redox potentials in natural waters have
been summarized by Hutchinson(4) following Cooper's{35) treat-
ment. In general, the potentials are insensitive to changes iti the
partial pressure of oxygen down to very low concentration. Reducing'
th*» oxygen concentration from 100 percent saturation to 1.0 percent
lowers the potential only about 0,03 volt; however, a rise of one unit
in pH decreases the potential 0.058 volt, a value that can be used to
adjust potentials at different pH values to some arbitrary pH when
an oxygen potential is involved.
The hypolimnions of some lakes exhibit a uniform redox poten-
tial, even with low oxygen concentrations. It is assumed that the
mud surface remains oxidized under such conditions. More com-
monly the potential falls rapidly near the bottom and frequently the
hypolimnion has potentials considerably lower than those in the
epilimnion. The near-bottom decrease results from reducing sub-
stances derived from the mud. the most important of which is ferrous
iron(4, p. 725), The most complete series of quantitative data show-
ing chemical changes in the hypolimnion during stagnation are those
of Mortimer(5) for English lakes. This work and that of Einsele(26)
demonstrated that the redox potential is inseparably linked with the
concentration of iron and that an understanding of these relationships
is essential to an understanding of most other chemical cycles.
Iri deep water, reducing conditions probably always exist below
the sediment surface at a potential of about 0.0 volt. When the sedi-
ment is in contact with oxygenated water, an oxidized microzone
develops at the mud surface. This micro zone has the brown color of
hydrated ferric oxide, and when fully formed, exhibits a potential
of about 500 millivolts. Because of a break in redox potential-depth
curve below the sediment surface, Mortimer concluded that the oxi-
Love and Slack
19S
-------
dized microzone was maintained by molecular diffusion of oxygen
into the mud to a depth determined by the reducing power of the
sediment. Thus the potential break represented the thickness of the
winter oxidized layer. More recently Hayes anci Co-workers(36) and
Gorham(l) presented evidence suggesting that the depth of the po-
tential break represents the depth to which turbulence had stirred
the sediment during circulation. *n a°y case, as the oxygen concen-
tration decreases at the beginning of stratification, the oxidized
microzone becomes thinner and finally disappears when the potential
in the surface layer of mud reaches above 200 millivolts. When pres-
ent, the oxidized layer at the sediment-water interface serves as a
barrier to diffusion from the mud to the water because of the greater
adsorptive power of the oxidized phase. Also, if ferrous ions and
phosphate ions diffuse into an oxygen-rich region, a precipitation of
ferric phosphate occurs. Mortimer(fi) showed that reduction was ac-
companied by the appearance in the water of ferrous, ammonium,
manganous, silica, phosphate, and sulfide ions and soluble organic
compounds.
Experimental evidence in support of this mechanism has recently
been provided by artificially mixing a lake(37).
The natural restraints on redox potentials have been the subject
of much study. It appears well established that living organisms
exert a major control on oxidation-reduction systems in natural en-
vironments. Although the evidence cannot be presented here, the
important paper by Baas Becking, Kaplan, and Moore(38) illustrates
the point. It was concluded that the redox potential — pH character-
istics of water are determined chiefly by photosynthesis, respiration,
and oxido-reductive changes in the iron and sulfur systems.
MORPHOMETRY CONTROLS
With regard to natural lakes, Hutchinson (4, p. 164) observed
that the form of a lake basin and of the lake that occupies it depends
partly on the forces that produced the basin and partly on the events
that occur m the lake and lts drainage after it has formed. The same
statement can be made of impounded water bodies, if the special
considered °eS °f ^ °rmation and subsequent management are
The effect of reservoir size and shape on solution phenomena has
been mentioned in connection with climatic controls. From the stand-
pom o waer quality, tl}e influence of morphometry on stratification
ls. ^ mer(J9) ^cognized three consequences of
stratification (j) sufficient light for photosynthesis exists only in the
- and m moVements in hypolimnion are much re-
duced, and (3) dead organic matter rains down from the productive
zone at a rate depending on the productivity This gives rise to the
rSe Thus' th* «¦ "iS
auaiitit *n shallow lakes because of the smaller volume and
smaller quantity Df oxygen available at the beginning of stratifica-
110
CONTROLS ON SOLUTION AND PRECIPITATION
-------
tion. By similar reasoning, it is obvious that reduction products tend
to reach high concentrations in the hypolimnions of shallow basins.
Morphometry sometimes acts directly to control chemical trans-
formations. Hutchinson(40) explained the increase in bicarbonate
in bottom waters of Linsley Pond by showing that more of the ion
per unit volume could go into solution from precipitated carbonate
in the sediment where the bottom slope was gentle than where it was
steep. A more gradual slope provided more mud surface per unit
volume. Shapiro(4i) attributed a mid-water oxygen minimum in Lake
Washington to a similar bed effect in which a submerged shelf with
sediments of high oxygen demand was in contact with the water.
Reservoirs act as traps for sediment carried in by influent waters,
with the result that the effective storage capacity and depth are re-
duced. The rate at which this process proceeds is extremely variable.
Although the effect of reservoir sedimentation on water quality is
probably minor compared to the serious loss of water storage space,
other consequences of sediment accumulation have been reported.
Gartska et al. (42) observed that most of the turbidity currents reach-
ing Hoover Dam in Lake Mead occurred during the first 7 years of
the impoundment when the original channel of the Colorado River
was still well defined. Later, sediment deposited in the channel in-
creased the area of the interface, resulting in reduced velocities and
increased desilting of the inflowing water. The density currents
spread into thinner layers and became increasingly less capable of
reaching the dam. A similar observation was reported for Elephant
Butte Reservoir (43). In discussing the chemical dissimilarity between
small lakes and adjacent ground water, Hutchinson(4, p. 232) stated
that lake water is separated from ground water by a seal of clay and
fine silt deposited as an early lake sediment. This is not true for all
lakes, however, and probably not for all reservoirs. This relationship
needs further study.
In their studies of radioactive phosphorus in lakes, Hays and
Phillips(44) found that the morphometric properties of a lake were
one of the important controls on the final level of total exchangeable
phosphorus. The volume determines the dilution, and the depth influ-
ences both thermal stratification and the relative size of the photo-
synthetic zone. They further concluded that rooted aquatic plants
are probably unimportant in phosphorus uptake in large, deep lakes
where the size of the littoral region is negligible compared to the total
volume. In smaller and shallower lakes with minimal profundal zone,
growth and bacterial decay of rooted aquatic plants probably dom-
inates the phosphorus exchange.
BIOLOGICAL CONTROLS
The emphasis thus far has been on inorganic controls; however,
the seasonal cycle of abundance of major and minor elements in
water can only be interpreted with a knowledge of the ecological
processes involved. Some of these effects have already been touched
upon in previous sections.
Love and Slack
111
-------
Biological requirements for nitrogen and phosphorus have been
stressed in water quality investigations because of their frequent
occurrence in quantities too small to support plant growth. Other
elements also are essential to organisms. Of these, potassium, calcium,
sulfur, and magnesium are required in relatively large amounts.
Hewitt(45) lists the following 10 trace elements, which are known
to be essential to plants: iron, manganese, copper, zinc, boron, sodium,
molybdenum, chlorine, vanadium, and cobalt. The abundance of these
elements in water is controlled by the rates of supply and utilization
by plants and animals, and by inorganic reactions.
Some organisms require still other elements. Thus, Hutchinson
(4, p. 799) stated that the development of diatom blooms constitutes
the most important mechanism by which silica is removed from lake
waters. In reviewing the work of Vinogradov and Eoichenko (46),
Hutchinson concluded that the dissolution of aluminosilicate clay-
minerals by benthic diatoms greatly accelerates inorganic liberation
of silica present in clay minerals of sediments.
Recent reports have emphasized the role of bacteria in controlling
the redox potential of water and sediment and in the distribution of
individual elements. Myers(47) and Ingols and Wilroy(33) found that
bacteria are essential to the solution of manganese, although the
mechanisms involved are not completely understood. Myers reported
that manganese carbonate precipitates when carbon dioxide is re-
moved from bicarbonate by bacterial action. Other bacteria are
thought to utilize organic substrates that act as protective colloids,
causing manganese to precipitate as the hydroxide. Under reducing
conditions, this substance redissolves and may form high concentra-
tions in bottom waters. Ingols and Wilroy(33) suggested that a bac-
terial byproduct rather than manganous ions may reduce manganic
oxide by direct utilization of oxygen. The high concentrations of
manganese reported in lakes with bottom temperatures of 23° to 28°C
further suggest the importance of biological processes.
In a study of the transport of iron by clay minerals, Carroll(48)
found that free iron oxide films could be removed from clay particles
by lowering the redox potential through the metabolic processes of
bacteria. Theil's(49) finding that bacteria can affect the solubility of
alumina and silica was also substantiated by Carroll.
Products of biologic activity are important in the formation of
complexes and in sorption reactions. Beauchamp(50) wrote that he
believes diatom frustules have the capacity to adsorb sulfates and to
prevent their reduction to sulfides. The accumulation of a partially
decomposed organic layer on the bottom of a marsh was thought by
Kadlec(57) to increase greatly the capacity of the sediment to adsorb
ions. The role of dissolved organic substances in chelation of trace
elements in natural waters was discussed by Saunders (52), During
a study of the responses of a marl lake to additions of chelated iron
and fertilizer, Schelske et al.(53) observed that a phytoplankton
bloom removed iron from the water, leaving concentrations below
the pretreatment level. Apparently the chelate molecule made avail-
112
CONTROLS ON SOLUTION AND PRECIPITATION
-------
able to the algae iron that had not been available previously. It was
further noted by Schelske(54) that the same chelating agent, tri-
sodium salt of N-hydroxyethylethylenediamine triacetic acid, was
effective in bringing iron sediments into solution under aerobic
conditions.
Although the subject of marl formation is still poorly understood
in several respects, there is general agreement that photosynthesis
by plants and shell formation by molluscs are two of the most im-
portant processes by which calcium carbonate (and to a lesser extent,
magnesium carbonate) is removed from solution.
Nitrogen is intimately involved in organic cycles, since it is an
essential constituent of protoplasm. Dugdale and Dugdale(55) pre-
sented evidence for nitrogen fixation by Anabaena, a blue-green alga,
in Sanctuary Lake, Pennsylvania. Kadlec(5/) studied conditions in
soils exposed during drawdown. He found a decline in bases, probably
as a result of leaching from the soil, but nitrogen increased through-
out the system as a result of microbiological nitrification in the
organic soil.
The dynamics of phosphorus in water-sediment systems has been
the subject of continuing investigation since Einsele's(26) early work.
As shown in an earlier section, inorganic phosphorus is inseparably
involved with redox potential and the iron cycle. Nevertheless,
Rigler(56) concluded that the turnover of phosphate under natural
conditions appeared to be caused primarily by bacteria. Hays and
Phillips (44) attributed to bacteria the decisive role in exchange of
phosphorus between solid and aqueous phases, reversing the con-
clusions of Hays(57). A normal cycle was postulated as consisting
of (1) uptake and synthesis of organic phosphorus in the water;
(2) fallout to the sediment surfaces; (3) bacterial breakdown to
inorganic phosphorus; and (4) restoration of phosphorus to the water.
That the matter may not be so easily settled is shown by a recent
study by Livingstone and Boykin(58) in which it was found that
the most important factors in phosphorus release from mud are the
exchange capacity of mud and the ionic activity of water.
In discussing biologically active elements, Lauff(59) wrote that
cycling of mineral nutrients is most efficient when the zone of active
plant growth is close to the zone of most active decomposition of
organic matter. Shallow depths are particularly favorable for this
type of regeneration cycle.
MANAGEMENT CONTROLS
In this category are included those factors manipulated by man
in order to regulate conditions within the reservoir or downstream
from it. These operations are not necessarily attempts to regulate
the quality of impounded water, in fact they usually are not, but they
are consequences of management of the reservoir for one or several
purposes for which it was designed.
During the water loss investigations of Lake Mead, consideration
Love and Slack
113
-------
was given to the possibility of reducing evaporation loss by with-
drawing warm water from the surface. Gartska and co-workers(42)
concluded, on the basis of field and model studies, that once a steady
state is attained withdrawals of water from a homogeneous reservoir
consist of contributions from the entire reservoir. Upper-level with-
drawals of water from a reservoir with stable stratification would
contain a greater proportion from the layers nearer the surface than
would be expected under idealized hydrodynamic conditions. Sim-
ilarly, when withdrawals are made from the lower portions of the
reservoir, they would include a greater proportion from the layers
nearer the bottom than would be expected on theoretical grounds.
Furthermore, withdrawal of warm surface water might have other
undesirable consequences downstream, even if it were feasible from
the engineering standpoint. Churchill's(9,10) oxygen data for TVA
reservoirs did, however, indicate that withdrawal of surface water
is indeed feasible.
After finding that most deep-water reservoirs show increased
manganese concentration near the bottom, Myers(47) concluded that
the depth of withdrawal and the nature of circulation determined
whether this substance would create a problem. In general, man-
ganese can be avoided if water is withdrawn from the epilimnion.
Ingols and Wilroy(33) suggested that small reservoirs should be
kept in continual circulation, possibly by the method described by
Riddick( 60), as the best means of controlling the increased hardness
and manganese concentration in the hypolimnion during stratification.
Ford(6/) described the benefits of circulation induced in a small res-
ervoir by air injection. Stratification was eliminated with marked
improvement in the quality of discharge water, notably because of
the elimination of hydrogen sulfide.
Reduction of the volume of the epilimnion by surface withdrawal
has been proposed for an entirely different purpose by Murphy (62).
It was suggested that biological productivity might be enhanced by
mixing deeper water, richer in nutrients, into the photosynthetic
zone during the productive season. If this is true, then surface with-
drawal may be detrimental in cases where it contributes to phyto-
plankton blooms.
Finally, attention may be drawn to two other aspects of reservoir
management; these involve solution and precipitation. First, im-
proved watershed practices that reduce the sediment load can be
expected to benefit the reservoir, not only by extending the effective
life, but also by substantially diminishing the amount of adsorbed
substances carried into the basin. Second, factors affecting the ex-
posed marginal areas of impoundments during drawdown may pro-
duce undesirable secondary effects when reflooded. Ingols(63) ob-
served excessive organic loading following submergence of exposed
lake bottoms on which plants had grown. Kadlec(52) studied the
effects of drawdown on the productivity of a wildfowl marsh. Fre-
quent and severe drying of the soil was the most effective inhibitors
of release of plant nutrients after reflooding. The demonstration of
the influence of nitrogen fixation by alder trees on lake productiv-
114
CONTROLS ON SOLUTION AND PRECIPITATION
-------
ity(64) and the possible leaching of manganese from oak trees of a
forested watershed(47) suggest that vegetation control may be ef-
fective in the management of small impoundments.
CURRENT AND NEEDED RESEARCH
Much of what has been presented thus far is based upon studies
of lakes, ponds, or laboratory tanks. With the increasing number and
importance of impoundments in the development of water resources,
more detailed knowledge of the reservoirs themselves is required if
they are to be managed for optimum results in terms of water quality.
The need is especially great for increased understanding of funda-
mental processes that may be applied in the early design stage for
control or at least forecast of the ultimate water quality to be ob-
tained.
A number of research projects in the Geological Survey are
applicable to this type of study of the dynamic controls on water
quality in reservoirs. The following areas of current research illus-
trate the scope of these investigations:
1. Occurrence and distribution of radioelements in water.
2. Exchange reactions of radioactive substances.
3. Mineralogy and exchange capacity of fluvial sediments.
4. Solute composition and minor element distribution in lacus-
trine closed basins.
5. Hydrosolic metals in natural water.
6. Organic substances in water.
7. Geochemical controls of water quality.
8. Analytical methods — water chemistry.
Specific Areas of Needed Research
1. Effects of reservoir aging on adsorption capacity of bottom
sediments and on other physical, chemical, and biological processes.
2. Controls on bacterial activity in aerobic and anaerobic water
and sediments with reference to the regeneration of chemical sub-
stances.
3. Role of organic matter in cycling of elements.
4. Improved sampling and analytical techniques for inorganic
ions and complexes. (This is essential to a satisfactory understanding
of chemical dynamics in reservoirs.)
5. Techniques for predicting the chemical and thermal conditions
to be expected in new construction.
6. Improved management methods for optimum control of water
quality.
7. Better understanding of the role of producer and decomposer
organisms in controlling water composition.
Love and Slack
115
-------
8. Distribution and importance of organic and inorganic trace
substances.
9. Fate and role of cultural additives, specifically the influence
of these materials on natural processes. (This presupposes a knowl-
edge of the uncontaminated environment.)
10. Nature of deep-water currents in thermally stratified basins
and their effect on water composition.
11. Relative influence of bottom and suspended sediments (in-
organic and organic) on adsorption processes.
12. Relationships between reservoir water and adjacent ground
water under uniform and fluctuating levels.
REFERENCES
1. Gorham, E. Observations on the formation and breakdown of
the oxidized microzone at the mud surface in lakes. Limnol. and
Oceanog. 3(3): 291-98. 1958.
2. Hem, J. D. Study and interpretation of the chemical character-
istics of natural water. USGS Water-Supply Paper 1473. 1959.
269 pp.
3. Kittrell, F. W. Effects of impoundments on dissolved-oxygen
resources. JWPCF. 31:1065-78. 1959.
4. Hutchinson, G. E. A treatise on limnology. Vol. 1. Geography,
physics and chemistry. John Wiley & Sons, New York, N.Y.
1957. 1015 pp.
5. Gorham, E. Factors influencing supply of major ions to inland
waters, with special reference to the atmosphere. Bull. Geol.
Soc. Am. 72:795-840. 1961.
6. Mortimer, C. H. The exchange of dissolved substances between
mud and water in lakes. J. Ecol. 29:280-329. 1941. 30:147-201.
1942.
7. Boyd, W. L. Limnology of selected arctic lakes in relation to
water-supply problems. Ecology. 40:49-54. 1959.
8 Rawson, D. S., and J. E. Moore. The saline lakes of Saskatchewan.
Can. J. Research, D. 22:141-201. 1944.
9. Churchill, M. A. Effects of storage impoundments on water
quality. J. Sanit. Eng. Div., ASCE. 83:SA1. Feb. 1957.
10. Churchill, M. A. Effect of impoundment on oxygen resources.
In: Oxygen relationships in streams. Proc. of seminar, Cincin-
nati, Ohio, Oct. 30-Nov. 1, 1957. Tech. Rep. W58-2. SEC. Mar.
1958. pp. 107-24.
11. Anderson, E. R., and D. W. Pritchard. Physical limnology of Lake
Mead. Lake Mead sedimentation survey. Navy Electronics Lab.
Rept. 258. 1951. 153 pp. and figs.
116
CONTROLS ON SOLUTION AND PRECIPITATION
-------
12. Wiebe, A. H. The effect of density currents upon tne vertical
distribution of temperature and dissolved oxygen in Norris Res-
ervoir. J. Tenn. Acad. Sci. 15:301-08. 1940.
13. Churchill, M. A. Effect of density currents upon raw water
quality. JAWWA. 39(4): 357-60. 1947.
14. Love, S. K. Relationship of impoundment to water quality.
JAWWA. 53 (5): 559-68. 1961.
15. Bryson, R. A., and V. E. Suomi. Midsummer renewal of oxygen
within the hypolimnion. J. Marine Research. 10:263-69. 1951.
16. Harbeck, G. E., Jr. A practical field technique for measuring
reservoir evaporation utilizing mass-transfer theory. USGS Pro-
fess. Paper 272-E. 1962. pp. 101-05.
17. Langbein, W. B. Salinity and hydrology of closed lakes. USGS
Profess. Paper 412. 1961. 20 pp.
18. Meyers, J. S. Evaporation from the 17 Western States. USGS
Profess. Paper 272-D. 1962. pp. 71-100.
19. Howard, C. S. Chemistry of the water. In: Comprehensive Sur-
vey of Sedimentation in Lake Mead. USGS Profess. Paper 295.
1960. pp. 115-24.
20. Benedetti, A. J., and J. A. Roller. Preparation of an impound-
ment area in Washington. JAWWA. 54(2): 197-202. 1962.
21. Hem, J. D., and M. W, Skougstad. Coprecipitation effects in solu-
tions containing ferrous, ferric, and cupric ions. USGS Water -
Supply Paper 1459-E, 1960. pp. 95-110.
22. Krauskopf, K. B. Factors controlling the concentrations of thir-
teen rare metals in sea water. Geochim. et Cosmochim. Acta.
9:1-32. 1956.
23. Goldberg, E. D., and G. O. S. Arrhenius. Chemistry of Pacific
pelagic sediments. Geochim. et Cosmochim. Acta. 13:153-212.
1956.
24. Aarnio, B. Ober die Ausfallung des Eisenoxyds und der Tonerde
in finnlandischen Sand-und Grusboden. Geoteckn. Medd., 16:
1-76. 1915.
25. Ruttner, F. (Trans, by D. G. Frey and F. E. J. Fry). Funda-
mentals of limnology. Univ. Toronto Press, Toronto, Canada.
1953. 242 pp.
26. Einsele, W. Ober chemische und kolloidchemische Vorgange in
Eisenphosphat-systemen unter limnochemischen und limno-
geologischenGesichtspunkten. Arch. Hydrobiol. 33:361-87. 1938.
27. Carritt, D. E., and S. Goodgal. Sorption reactions and some eco-
logical implications. Deep-Sea Research. 1:224-43. 1954.
28. Welch, P. S. Limnology, 2d ed. McGraw-Hill, New York, N.Y.,
1952. 538 pp.
Love and Slack
117
-------
29. Bjerrum, J., G. Schwarzenbach, and L. G. Sillen (compilers).
Stability constants of metal ion complexes. Part 1, Organic
ligands. London Chem. Soc. Spec. Publ, No. 6. 1957. 105 pp.
30. Hem, J. D. Complexes of ferrous iron with tannic acid. USGS
Water-Supply Paper 1459-D. 1960. pp. 75-94.
31. Stumm, W., and G. F. Lee. The chemistry of aqueous iron.
Schweiz. Z. Hydrologie. 22: (1960) Fasc. 1:295-319.
32. Shapiro, J. Yellow acid-cation complexes in lake water. Science.
127:702. 1958.
33. Ingols, R. S„ and R. D. Wilroy. Observations on manganese in
Georgia waters. JAWWA. 54(2): 203-07. 1962.
34. Allee, W. C., A. E. Emerson, O. Park, T. Park, and K. P. Schmidt.
Principles of animal ecology. W. B. Saunders, Philadelphia, Pa.,
1949. 837 pp.
35. Cooper, L. H. N. Oxidation-reduction potentials in sea water.
J. Marine Biol. Assoc. United Kingdom. 22:167-76. 1937.
36. Hayes, F. R., B. L. Reid, and M. L. Cameron. Lake water and sed-
iment. II. Oxidation-reduction relations at the mud-water inter-
face. Limnol. and Oceanog. 3(3):308-17. 1958.
37. Freshwater Biological Assoc. Thirtieth Annual Report for year
ended March 31, 1962. Report of the Director, Operation "Swiz-
zlestick."
38. Baas Becking, L. G. M., I. R. Kaplan, and D. Moore. Limits of
the natural environment in terms of pH and oxidation-reduction
potentials. J. Geol. 68(3): 243-84. 1960.
39. Mortimer, C. H. Underwater "soils." A review of lake sediments.
J. Soil Sci. 1:63-73. 1949.
40. Hutchinson, G. E. Limnological studies in Connecticut. IV. The
mechanism of intermediary metabolism in stratified lakes. Ecol.
Monographs. 11:21-60. 1941.
41. Shapiro, J. The cause of a metalimnetic minimum of dissolved
oxygen. Limnol. and Oceanog. 5(2):216-27. 1960.
42. Garstka, W. U., H. B. Phillips, I. E. Allen, and D. J. Herbet. With-
drawing water from Lake Mead. In: Water-loss Investigations,
Lake Mead Studies. USGS Profess. Paper 298. 1958. pp. 63-75.
43." Lane, E. W., and E. J. Carlson. Some hydraulic engineering as-
pects of density currents. U. S. Bur. Reel. Hydr. Lab Rept.
HYD-373. 1954.
44. Hayes, F. R., and J. E. Phillips. Lake water and sediment. IV.
Radio-phosphorus equilibrium with mud, plants, and bacteria
under oxidized and reduced conditions. Limnol. and Oceanog,
3(4): 459-75. 1958.
118
CONTROLS ON SOLUTION AND PRECIPITATION
-------
45. Hewitt, E. J. Some aspects of micro-nutrient element metabolism
in plants. Nature. 180:1020-22. 1957.
46. Vinogradov, A. P., and E. A. Boichenko. Decomposition of kaolin
by diatoms. C. R. Acad. Sci. URSS. 37:135-38. 1942.
47. Myers, H. C. Manganese deposits in western reservoirs and dis-
tribution systems. JAWWA. 53(5):579-88. 1961.
48. Carroll, D. Role of clay minerals in the transportation of iron.
Geochim. et Cosmochim. Acta. 14:1-27. 1958.
49. Theil, G. A. The enrichment of bauxite deposits through the
activity of micro-organisms. Econ. Geol. 22:480-93. 1927.
50. Beauchamp, R. S. A. Utilizing the natural resources of Lake
Victoria for the benefit of both fisheries and agriculture. In: 7th
Tech. Meeting I.U.C.N., Athens, Greece, Sept. 1958. Vol. IV, Soil
and Water Conserv. 1960. pp. 357-62.
51. Kadlec, J. A. Effects of a drawdown on a waterfowl impound-
ment. Ecology. 43 (2): 267-81. 1962.
52. Saunders, G. W. Interrelations of dissolved organic matter and
phytoplankton. Botan. Rev. 23(6): 389-410. 1957.
53. Schelske, C. L., F. F. Hooper, and E. J. Haertl. Responses of a
marl lake to chelated iron and fertilizer. Ecology. 43(4): 646-53.
1962.
54. Schelske, C. L. The availability of iron as a factor limiting pri-
mary productivity in a marl lake. Ph.D. thesis, Univ. Michigan.
1960.
55. Dugdale, V. A., and R. C. Dugdale. Nitrogen metabolism in lakes.
II. Role of nitrogen fixation in Sanctuary Lake, Pennsylvania.
Limnol. and Oceanog. 7(2): 170-77. 1962.
56. Rigler, F. H. A tracer study of the phosphorus cycle in lake
water. Ecology. 37(3):550-62. 1956.
57. Hayes, F. R. The effect of bacteria on the exchange of radio-
phosphorus at the mud-water interface. Verh. Int. Ver. Limnol.
12:111-16. 1955.
58. Livingstone, D. A., and J. C. Boykin. Vertical distribution of
phosphorus in Linsley Pond mud. Limnol. and Oceanog. 7(1):
57-62. 1962.
59. Lauff, G. H. The role of limnological factors in the availability
of algal nutrients. In: Algae and Metropolitan Wastes. Trans,
of seminar, Cincinnati, Ohio, Apr. 27-29,1960. Tech. Rept. W61-3.
SEC. 1961. pp. 96-99.
60. Riddick, T. M. Forced circulation of reservoir waters. Water &
Sewage Works. 104(6): 231-37. 1957.
Love and Slack
119
-------
61. Ford, M. E., Jr. Artificial control of reservoir limnology and its
effects on water quality. Paper presented at the Calif. Section
Fall Conference, Santa Monica, Calif., Oct. 25, 1962. 8 pp. mimeo.
62. Murphy, G. I. Effect of mixing depth and turbidity on the pro-
ductivity of fresh-water impoundments. Trans. Am. Fisheries
Soc. 91(1): 69-76. 1962.
63 Ingols, R. S. Effect of impoundment on downstream water qual-
ity, Catawba River, S. C. JAWWA. 51(1): 42-46. 1959.
64. Goldman, C. R. The contribution of alder trees (Almts tennifolia)
to the primary productivity of Castle Lake, California. Ecology.
42(2):282-88. 1961.
DISCUSSION
Joe K. Neel*
U. S. Public Health Service
Kansas City, Missouri
The preceding presentation is based upon a review of literature,
and a critique of the usual sort seems inappropriate, inasmuch as a
large number of people other than the two authors are responsible
for the data given. Mr. Love and Dr. Slack are to be congratulated
for their exhaustive search for and selection of material, as well as
their summation of pertinent information. It was impossible for them
to enter into all related areas in the time allotted, and their omission
of a number of phenomena is not assumed to reflect their unfamili-
arity with them.
A number of people here are aware of a 5-year study I directed
to ascertain the effects of main stem reservoirs on water quality in
the Central Missouri River. Progress reports had a limited distribu-
tion and the final report(l) was received from the printer only a
few days before this symposium.
As indicated, the Love and Slack paper is difficult to review. I
would assume that discussion of the various findings reported would
require several times the space used for listing them. Since Love and
Slack were largely concerned with the role of individual factors
and actions, a brief discussion of some more general relationships that
determine water quality in impoundments will complement their
presentation. Water quality in streams below reservoirs is often
greatly influenced by reservoir operation, which may at times negate
or unduly emphasize certain events occurring in the impoundments.
I have studied various phenomena affecting quality of surface
inland waters for more than 20 years. During this period I have
become very impressed with the effects of living organisms operating
*Now with The Potamological Institute, Louisville, Kentucky.
120
DISCUSSION
GPO 821-740-3
-------
within limitations imposed by physical and chemical conditions,
which they themselves often modify. Only in its course through the
atmosphere to the ground may water be assumed to be insignificantly
affected by life. When it strikes the earth and moves across or into it,
materials owing their existence to living processes in the soil are
taken into solution or suspension and carried across or just beneath
the earth's surface to water courses, or deeper into the ground to be
discharged later. Carbon dioxide, which makes water much more
corrosive of limestone, dolomite, apatite, etc., chiefly owes its exist-
ence to the biota of the soil, as do numerous other substances. In-
organic suspended materials and soluble minerals are also acquired
from contact with the earth.
When runoff reaches a stream, pond, lake, or impoundment,
materials it contains are subject to modifying influences of various
factors operative therein, and it may also pick up additional solutes
and suspended particles, Surface runoff usually or frequently trans-
ports loads of suspended sediment into streams from which it enters
reservoirs. Most particles may be expected to settle out when cur-
rents are slowed or stilled in headquarters. Deltas of silt and sand
are normally built up in these areas, but they may be eroded and
displaced down reservoir if the water level declines sufficiently to
permit their channeling and undercutting by the inflowing stream.
Surface runoff discharges often differ in density from reservoir sur-
face waters and, upon entering impoundments, flow above or under
the latter, dependent upon their relative density, as density currents.
Much of the silt contained in runoff segments of this sort may be
carried some distance into reservoirs where it precipitates slowly.
The character of suspended materials and processes encountered in
reservoirs may markedly affect rate of sedimentation. Particles bear-
ing the same charge (+ or —) repel each other and tend to remain
in suspension if small and light, but they may be precipitated rapidly
if neutralized by ions bearing opposite charges. Many finely divided
clay particles carry negative charges that are often neutralized by
hydrogen ions arising from respiration and/or decomposition of
aquatic organisms. This reaction leads to aggregation and sedimenta-
tion of clay particles, which may often take organisms down with
them.
Surface runoff differs in density from reservoir surface waters
during several periods of the year, and it is common for such dis-
charges to flow under or across upper waters as density currents.
Such movements are designated as over-, inter-, and underflows,
depending upon the level at which they move down reservoir. Cur-
rents that move down at some depth often displace upper layers and
induce an upstream flow along the reservoir surface. The upstream
surface movement is stilled at the point where the inflow sinks into
the reservoir, and this "convergence" line is often marked by a
narrow zone of floating debris that has been carried both into and up
the reservoir. Water quality may fluctuate markedly at this point
with a change in wind direction from upstream to downstream or
vice versa. Development of a convergence line is not a necessary
Neel
121
-------
prerequisite for changes in water quality to accompany shifts in wind
direction in headwater areas,
Turbid density currents often carry suspended sediment consid-
erable distances into reservoirs and may play a large role in accumu-
lation of "bottomset" siltation. They may lose turbidity very slowly
and if perched in upper levels may restrict penetration of light.
Behavior of turbid inflows is by no means static, and few patterns
may be relied upon uniess there is considerable familiarity with the
reservoirs in question.
It has long been assumed that seasonal inflows are generally
homogenized by mixing in reservoirs at some time during the year
and that overall water quality reflects the proportional influences of
various seasonal discharges. Water contributed during seasons of
high runoff is usually less mineralized than that entering during low
flow periods, and could be expected to dilute the more highly miner-
alized segments. In some quarters reservoir reduction of hardness and
alkalinity concentration has been assumed due to such dilution
Records(l) on the Central Missouri River indicate that seasonal flows
tend to maintain their individuality in the huge main stem reservoirs
mixing only along their forward and rear margins. Prior to im-
poundment, rivet discharges, with lowest mineral content occurred
in July and August. As impoundment was augmented, appearance
of these waters was delayed, and, finally, they left the reservoir area
in midwinter. The most highly mineralized flows, normal to the mid-
winter season of low river discharges, were relea&ed during the sum-
mer period that formerly contributed flows with lowest mineral
content. A reversal in the native seasonal mineral variation pattern
was effected. These changes in water quality were not as abrupt av
before impoundment.
Studies ot the Central Missouri River furnished data on pre- and
post-impoundment phenomena in Ft. Randall, Garrison, and Gavins
Point Reservoirs, and their effects upon quality of discharges. None
of these impoundments developed stratification or density layers dur-
ing the period of our investigation. Water level rise in each instance
inundated previously open soils and leached various substances
from them. Suspended sediment carried in the muddy river was
dropped to the bottom in headwaters and phytoplankton growth soon
began, and was evidently stimulated by materials leached from the
soils. In Ft. rtandali Reservoir, photosynthetic precipitation of hard-
ness and. ^ soon exceeded or overshadowed quantities picked
up from the tosiii and the net effect on an annual basis was reduc-
tion of frj,ca* . nai? fCa and Mg) in the water passing through the res-
ervoir- " Reservoir reduced the hardness and alkalinity of its
inflows duri g its first year 0f impoundment but leaching of its floor
increased n ness and alkalinity during the second year. Operation
during tihe ona year resulted in a net increase in surface elevation
Leaching e less noticeable in both impoundments with passage
of tiine. £ gressively smaller areas were inundated per foot of
elevation 1 ®ase, and sedimentation of organic and inorganic par-
ticles soon gan to form a barrier between the water and original
122
discussion
-------
soil substances. The sediment layer was disrupted and leaching pro-
moted by fluctuations in surface elevation. It is expected that photo-
synthetic precipitation will eventually overshadow bed solution in
all units of this reservoir system.
Reservoirs are lakes, a rather unique type to be sure, but a type
that can be formed by natural actions when an earthquake or land-
slide blocks a stream. Construction of a reservoir therefore imposes
lentic or lake conditions in a previously and exclusively lotic or
stream environment. It is normal for a standing body of water to
work toward its own extinction by filling its basin with various mate-
rials. This tendency is quite contrary to the normal lotic pattern of
enlarging, deepening, or changing channels and flood plains. Settling
or precipitation of materials in a reservoir is the major action behind
its influences on water quality as aging proceeds, but these processes
can be disrupted by fluctuation in water depth.
The subject of reservoir classification has not received any con-
centrated nationwide attention. Distinctions that may be noted in
the Tennessee River system, for example, should not be expected to
characterize other areas where water demands, operation, climate, etc.
are different. The nature of a reservoir is essentially determined by
its surface area, depth, capacity, rate of flow through, topography of
its surroundings, soils in its contributory drainage system, climate,
mode of operation, orientation to prevailing winds, age of impound-
ment, etc.; location on a tributary stream or main stem is at most
secondary in the great majority of cases. In many instances, major
storage is provided by main stem impoundments.
The authors (Love and Slack) devote considerable attention to
climate, chemical reactions, stratification, etc. that have marked in-
fluences on water quality. The classical picture of stratification and
the effects they ascribe to it, supported by the findings of various
investigators, assume that the dividing line between the upper and
middle water layers (epilimnion and thermocline) is below the
level of effective light penetration. This is not always true, and since
quality of water may be and often is dominated by photosynthesis,
thermally separate layers of water may exhibit identical or very
similar chemical characteristics if light penetrates both. Two distinct
types of photosynthesis occur in surface waters, and it is normal for
the less common type, that carried on by certain bacteria and liber-
ating sulfur instead of oxygen, to occur beneath the "photosynthetic
zone" of green plants. The bacterial type utilizes longer light
rays and seemingly is restricted to levels where sulfides exist in
excess of demands or are continually being formed for use. Hence,
organisms carrying on this activity may exist in the upper oxygenless
zone or at the boundary between aerobic and anaerobic conditions.
Sulfur that they liberate sinks toward the bottom and builds up
noticeable layers under certain conditions. Some may be oxidized
to form gypsum if it enters aerobic water. Activities of photosynthetic
bacteria may therefore afford hardness increases under certain condi-
tions when CaCOs is being precipitated by algal or flowering plant
photosynthesis.
Nooi
12S
-------
Major effects of the commoner (O., liberating) sort of photosyn-
thesis on water quality are removal from solution of various elements
necessary to plant growth and metabolism, supply of oxygen, and
precipitation of CaCOs (MgCO.,) from Ca(HCO.t)2 (Mg(HCO;!)2) as
the latter compound is decomposed to provide CO,,. Algae can remove
great quantities of phosphorus and nitrogen when these elements are
present in abundance, and in most natural waters they are usually
effective in restricting phosphorus to the organic or "unavailable"
state. Great quantities of alkalinity and hardness were removed from
water by photosynthetic precipitation in Ft. Randall Reservoir on the
Missouri River during 1953 through 1957.
Photosynthetic organisms may restrict depths to which they may
occur by proliferating sufficiently to limit penetration of light to
upper layers. Photosynthetic bacteria may be forced to migrate
nearer the surface or cease photosynthesis when light is restricted to
upper levels. Phytoplankers produce hydrogen ions by respiration
or decay, and these ions often neutralize negatively charged silt
particles and contribute to clarification when the latter aggregate in
sufficient quantity to precipitate and carry organisms down with
them. A number of normally photosynthetic organisms are able to
exist as saprophytes or with other modes of energy acquisition when
they are deprived of light. Under these conditions they may have
influences foreign to those associated with their normal modes of life.
REFERENCES
1. Neel, J. K., H. P. Nicholson, and A. Hirsch. Main stem reservoir
effects on water quality in the central Missouri River. USDHEW,
PHS, Kansas City, Mo. Mar. 1963, 112 pp.
DISCUSSION FROM THE FLOOR:
Mr. McLean, U. S. Public Health Service; The lack of questions
indicates that we have not been as much concerned with chemical
quality as we may be in the future. In work on the Kansas River,
we have been concerned with chemical quality, and I think we will
be increasingly concerned with this problem in the future. I expect
that the next time a similar symposium is held there will be more
questions, because these problems will have intruded themselves
upon us.
Dr. Ingols, Georgia Institute of Technology: Dr. Neel, what were
the orders of magnitude of the high and low hardness values? In
other words, was the range 1 or 2 milligrams per liter out of a hun-
dred, or two or threefold in magnitude?
Dr. Neel: Hardness values ranged from about 300 to about 150
milligrams per liter.
Dr. Ingols: This is a twofold variation in magnitude at a very
high level. We have observed a fivefold change in ionic concentration
in the hypolimnion in some of the Piedmont zone lakes, but of course
4
DISCUSSION
-------
when this is discharged from a lake or reservoir, the dilution of the
hypolimnion in the total flow is so great that it does not show as a
very big change downstream. Your orders of magnitude are tre-
mendous for a stream downstream from a reservoir where there is
mixing.
Dr. Neel: We did not detect any type of stratification in these
reservoirs during the periods of our study.
Dr. Ingols: To what do you attribute the transition? Are ground
water discharges causing this order of magnitude of change in mineral
quality?
Dr. Neel: The low flows are made up mainly of ground water, yes.
Dr. Ingols: Is it a matter of change in the time of occurrence
of low flow versus high flow?
Dr. Neel: Yes. The high concentrations originally occurred in
the winter; they occurred in the summer after the impoundments
were completed, which was a reversal.
Mr. Whitehouse, Central Electricity Research Labs, England: Mr.
McLean said that apparently you were not as yet concerned very
much with the chemical control of water. In England we have had
to concern ourselves with this. An interesting and useful paper both
from the biological and chemical point of view was published in 1959
in the proceedings of the Society for Water Treatment and Examina-
tion by Dennis Hammerton, who was the biologist with the Bristol
Water Works in England.
From the Floor
125
-------
Session 3
EFFECTS OF FLOW REGULATION
ON WATER QUALITY (Part 1)
Moderator: L. W. Gebhard
U. S. Public Health Service
-------
CHANGES IN WASTE ASSIMILATION CAPACITY
RESULTING FROM STREAMFLOW REGULATION
R. A. Vanderhoof
Director, Ohio River Basin Project
U. S. Public Health Service, Cincinnati, Ohio
INTRODUCTION
My first impression of this subject was that it should be limited
to a discussion of the effects of increased minimum flows downstream
from a dam on dissolved oxygen concentrations resulting from the
assimilation of organic wastes by biochemical oxidation processes. It
became apparent, however, that a much broader consideration of the
changes in waste assimilation capacity resulting from streamflow
regulation was necessary. Changes in temperature, changes in pH,
and changes in coliform numbers observed in field studies will be
described. Since flow regulation requires that water be stored,
changes in quality of impounded waters will also be discussed.
It is assumed that the reader is familiar with the Streeter-Phelps
oxygen-sag formula and that its basic assumptions are generally
understood (1,2), i.e., that the oxidation of decomposable matter is
proportional to the amount present, that it is independent of the dis-
solved oxygen level, that reaeration is proportional to the dissolved
oxygen deficit, that it is dependent on surface turbulence and internal
mixing as well as absorption, and that both deoxygenation and re-
aeration are temperature dependent.
It is further assumed that the reader is generally familiar with
the provisions of the Federal Water Pollution Control Act (as
amended in 1961 by Public Law 87-88) relative to streamflow regula-
tion. Section 2b of the amended law provides that "in the survey or
planning of any reservoir by the Corps of Engineers, Bureau of
Reclamation, or other Federal agency, consideration shall be given
to inclusion of storage for regulation of streamflow for the purpose
of water quality control, except that any such storage and water
releases shall not be provided as a substitute for adequate treatment
or other methods of controlling waste at the source." Note that both
storage and water release are mentioned in the Act. Therefore, both
waste assimilation in impoundments and in the stream below are
discussed here.
WATER QUALITY CHANGES IN A RESERVOIR
Let us first consider water quality changes that might be ex-
pected when a flowing stream is converted to a reservoir. When a
large river is impounded, the tributary streams that are flooded be-
come the "arms" of the reservoir. My first major discussion concerns
such "arms."
Vanderhoof
129
-------
Let us look at a hypothetical situation: a waste load discharged
to a small stream that runs into a larger river. After impoundment
of the larger stream, a part of the small stream and its valley would
be an arm of the impoundment. What changes in water quality, and
hence in assimilation capacity, should be anticipated? The cross-
sectional area in the arm of the impoundment would be larger than
it was in the flowing stream. Consequently, the time of passage be-
tween any two points along the trace of the stream would be longer.
As a result, the shape of the oxygen-sag curve would be changed
significantly; i.e., you would expect that the long curve of degrada-
tion and recovery extending many miles in the free flowing stream
would be compressed to perhaps about 1 mile, though the time ele-
ment may not be greatly changed. Because of the loss of the mechan-
ical aeration and mixing effects found in the free flowing stream, the
critical oxygen deficit in the reservoir arm might be expected to be
greater than the deficit observed in the free flowing stream. The
minimum dissolved oxygen concentration must be modified and re-
vised upward, however, for three reasons:
1. The increased absorption of oxygen from the atmosphere
resulting from greater water surface area.
2. The greater clarity of water, which encourages increased
growth of algae if necessary nutrients are available, and the
resulting addition of dissolved oxygen resulting from photo-
synthesis.
3. The availability of dilution water within the reservoir arm.
With the exception of the effects of increased activity of algae,
the major considerations that are expected to change water quality
resulting from impoundment are included in the oxygen-sag formula.
With a knowledge of the cross-section of the flooded valley, the flow
rates of the incoming stream, and the quality of incoming waters,
the quality of water in the arm, after reservoir stabilization has taken
place, can be approximated.
If the assimilation capacity is defined as the organic waste load
that can be accepted without degrading the water below a minimum
acceptable quality, then in the preceding case there would likely be
a loss of assimilative capacity. But is this loss the most important
consideration? While a small area may be degraded below the quality
objective, the water quality in a much larger downstream area may
be greatly improved. In this small degraded section, the flow through
time is greatly increased. Instead of rapidly passing on down the
stream, the water is retained in the reservoir arm. While the oxygen
resources are heavily taxed and are replaced slowly, the increased
detention time permits the satisfaction of BOD and the die-off of
coliform organisms within a confined area.
A further consideration is the use of the land and water in the
vicinity of the area of greatest quality degradation. If there is a loss
of legitimate water use in the area of greatest degradation in the
impoundment, the question arises as to who should be responsible
130
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
for the mitigation of the damage; the one(s) discharging the waste
or the one who built the impounding reservoir? No attempt is made
here to answer this question.
Now suppose we leave the hypothetical situation and consider
actual observations that were made in a free flowing stream in 1956
before impoundment and in the arm of a reservoir in 1960 after the
stream was impounded (3). Both sets of observations were made in
the summer season. The stream was relatively small, not more than
5 yards wide over most of its length. The arm of the impoundment
is not more than 5 yards wide at the upper end and about 300 yards
wide at the end of the first mile.
The major source of waste is an industrial plant located about
2 miles above the head of the reservoir arm. The organic waste
load was about 25 percent larger at the time of the post-impoundment
survey than at the time of the pre-impoundment survey. The number
of coliform organisms discharged at the time of the post-impound-
ment survey was only half what it was during the pre-impoundment
survey, however. In neither case were the industrial wastes treated.
The 5-day, 20°C-BOD (BOD5) observed in 1956 showed a steady
reduction from about 80 milligrams per liter to an estimated 30
milligrams per liter, or a reduction of about 60 percent, over a dis-
tance of 5,000 feet. In 1960, after impoundment, the BOD5 dropped
from 51 to 1 milligram per liter, or 98 percent, in approximately the
same distance (Figure 1). While oxygen depletion, or an oxygen
block, might be expected in the arm of such a reservoir, considering
the BODr, load, no area was observed where the oxygen resources
were exhausted. Where the free flowing stream merged into the
reservoir, the rate of oxygen use roughly approximated the rate of
reaeration, as indicated by the flat section in Figure 2. This section
was about 4,000 feet long. For comparative purposes, all oxygen
measurements were taken at a depth of 1 foot. With the exception of
the flat section at the head of the reservoir, the rate of recovery in
the impoundment was observed to be approximately equal to that
of the free flowing stream. Vertical temperature and dissolved oxygen
profiles (Figure 3 and 4) indicate that the dissolved oxygen concen-
tration decreased rapidly through the thermocline and then more
slowly until values below 1 milligram per liter were observed below
75 feet.
The coliform die-off occurred as expected. Figure 5 shows a
reduction of 50 percent in the free flowing stream and 99 percent in
the equivalent portion of the impoundment.
Thus this heavily polluted stream whose only use was to convey
wastes to a larger body of receiving water was converted into a body
of water usable for every recreational purpose, without benefit of
manmade treatment devices. Except for a distance of about 4,000
feet where oxygen consumption equalled reaeration in the surface
layers, the oxygen concentration increased in the reservoir. At the
head of the reservoir, the oxygen content was approximately 5
Van der hoof
131
-------
1000
f I II I I I II I I I I f U I I I I I I
100
10
I II | M I II I I I I I I IT^
o.i ^ l i
o
I 1 I
E
4)
o
e
o»
D
e
o
IMPOUNDMENT , RUNNING STREAM
< |
i
10,000
I I I I M I I I
20,000
o
£
4>
a
•a
w
I
LEGEND
Pre-lmpoundment Survey s
August 1956 Q == 2 cfs °
o
SMAX «j
Post-Impoundment Survey 5
July i960 Q = 4 cfs
AVG
¦ ¦ I I I I I I I 1 M i ¦ I | |
40,000
30,000
DISTANCE FROM MAIN CHANNEL, ft
Figure 1 — Companion of BODs before and after impoundment.
milligrams per liter. If the industrial waste is treated, the quality
picture should be even brighter.
In another arm of the reservoir, prior to impoundment, the free
flowing stream received untreated municipal and industrial wastes;
after impoundment, the wastes were adequately treated. The BODb,
coliform content, and dissolved oxygen content were observed under
both conditions over a distance of 11,000 feet(4).
The BODg declined in the free flowing stream from 292 to 51
milligrams per liter, or 82 percent (Figure 6). In the impoundment
a reduction from 54 to less than 1 milligram per liter was observed
over the same distance. This last reduction was observed for a waste
132
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
12
IMPOUNDMENT FLOWING STREAM
LEGEND
Pre-lmpoundment Survey
4)
8?
ID
¦C
U
(ft
10,000
20,000
30,000
40,000
DISTANCE FROM MAIN CHANNEL, ft
Figure 2 — Companion of diuolved oxygen content before and after impoundment.
uf
O
£
0£
D
(0
QC
LlI
H
<
5
5
o
u
ul
m
0.
LU
D
Sample points |
M ¦ 4 ¦ ^ I ^
5,000 10,000 15,000 20,000
DISTANCE FROM MAIN CHANNEL, ft
Figure 3 — Temperature profile, °C. July 14, 1960.
25,000
treatment plant effluent, not for a raw waste load. These comparisons
were made in terms of concentration rather than load since the true
load in the impoundment, if the organics that were undoubtedly
leached from the bottom lands after they were inundated are con-
sidered, was not known. This certainly points to the need for more
understanding of this complex problem.
Vanderhoof
133
-------
Figure 4 — Dissolved oxygen profile, mg/l. July 14, 1960.
Although the coliform content of the flowing stream was ex-
tremely erratic, as shown in Figure 7, the reduction amounted to
about 99 percent. The coliform content of the waters discharged
from the study area was still very large, however — over 3 million
per 100 milliliter. On the other hand, the impoundment was ex-
tremely effective in reducing the coliform content. In a distance of
10,000 feet the coliform content was reduced from 790,000 per 100
milliliter, a condition of heavy contamination, to 40 per 100 milliliter,
a condition acceptable for swimming.
The average dissolved oxygen content in the free flowing stream
below the source of pollution was less than 1 milligram per liter over
its entire length (except where tributary inflow provided a temporary
improvement), as shown in Figure 8, After impoundment (and
treatment of wastes), the waters entered with a dissolved oxygen
content of 4.5 milligrams per liter, or 50 percent of saturation, and
left the study area with an oxygen content of 7.8 milligrams per liter,
or nearly 100 percent of saturation. No oxygen blocks were observed
in the upper end of the reservoir. Temperature and dissolved oxygen
profiles (Figures 9 and 10, respectively) show that the dissolved
oxygen declined steadily from about 10 feet below the water surface,
that the decline was more pronounced along the sloping bottom, and
that a zone of degradation with one small pocket below 1 milligram
per liter was observed. In this discussion, I am most interested in
the effects on minimum water quality of changing a free flowing
stream to an impoundment. It is recognized that in certain seasons
of the year reservoir stratification may occur below depths of 15 to
134
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
30 feet, which can result in reduced dissolved oxygen concentrations
at the lower depths.
While the dissolved oxygen sag question can be used to approxi-
mate the changes in water quality that occur in the upper end of an
J l I I [ U l I | I I I ! | I I I I | I I I I | I I I I | I I I I jjU ! I | i I l l
IMPOUNDMENT
U
ra
E
W
0>
u
c
c
o
o
t
FLOWING STREAM
i"
i ¦ iJ
I I I I 1 I I I I I
10.000
20.000
o
t
a
K
¦D
10
0)
X
±1
>
-------
1000 C7
rr
100
&o
E
Q
O
ffi
o
ft
>-
<
D
10
y
0.11
I I I I I I ' I I I I I r
IMPOUNDMENT RUNNING STREAM
< 1 =>
J*
| LEGEND
I Pre-lmpoundment Survey
| August 1956
MAX
Post-Impoundment Survey
AVG July 1960
~
& w
St O E
U u
tr
¦D ^ m
0)
X
c 2
o E
a w
<9 —
Se effects of impoundments are always beneficial. On the other hand,
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
10"
I 1 1 1 I I
<5
I 1 I [ I I I I I I I
POUNDMENT1 . RUNNING STREAM
n
10«
10'
_ 10°
E
8
«
a
z
a.
5
5
£E
o
u.
Ij
O
o
io»
104
?EAM
1Q3
10»
10
Max
Avg
B
LEGEND
Pre-Impoundment Survey
August 1956
Post-Impoundment Survey
July 1960
+J*-r
c o E
® «<*
E82i
^_T3 C£
"S ac 5
a# c o E
x£u(/>
ij
i i i i i
Se°
S!I
co i— a.
A.
5,000 10,000 15,000
DISTANCE FROM MAIN CHANNEL, ft
Figure 7 — Comparison of conform concentration] before and after impoundment.
one should be most cautious in drawing the conclusion that impound-
ment of a small polluted stream will always result in nuisance con-
ditions and the depletion of oxygen resources, even in a small area.
Confinement of the polluted waters to a relatively small area can
make the remaining portion of the water course available for higher
water uses.
Vanderhoof
137
-------
5,000 10,000 15,000
DISTANCE FROM MAIN CHANNEL, ft
20,000
Figure 8 Comporiton of diiiolved oxygen content before and after impoundment.
Figure 9 — Temperature profile, °C. July 11, 1960.
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
DISTANCE FROM MAIN CHANNEL, ft
Figure 10, Dissolved oxygen profile, mg/l. July 11, 1960.
WATER QUALITY CONTROL DOWNSTREAM FROM A
RESERVOIR
Now, let us consider the changes in organic waste assimilation
capacity resulting from streamflow regulation downstream from a
dam. Since temperature is discussed in another paper, only general
remarks on its effects will be made. To compute the change in assimi-
lative capacity, the characteristics of the stream and the quality of
water prior to flow regulation by impoundment must be known.
The determination of dissolved oxygen content of waters released
from a reservoir requires careful study; previous papers dealt with
the development of thermal stratification and consequent oxygen
changes that may be expected. This difficulty should now be con-
sidered prior to construction of any dam and, if depths are over 40
to 50 feet, if the period of storage is significant, or if organic wastes
and/or nutrients are present in substantial amounts in the incoming
water, appropriate measures should be taken to assure the discharge
of waters of adequate quality. For a dam with no hydropower instal-
lation, it is usually recommended that facilities be provided to allow
discharge of waters from the upper, middle, and lower levels. Then,
according to downstream uses and needs and the quality of water
available, water of appropriate quality can be released. Multiple
level outlets provide a flexible system that can accommodate changing
water quality and altered downstream uses. While today's uses may
be known, it is not possible to be completely sure of tomorrow's
uses. The general subject of improving the quality of reservoir dis-
charges is discussed in the fifth session. Of most importance is the
recognition that quality changes do occur in reservoirs, that there
Vanderhoof
139
-------
are methods for improving the dissolved oxygen content of waters
discharged from reservoirs, and that the quality of the water dis-
charged must be considered. The BOD of waters released from a
reservoir should, of course, also be determined. It is not unusual for
waters to have a long detention time in the reservoir and it can be
expected that the BOD will be low. BODr> values of less than 1 milli-
gram per liter have been observed in many of our field studies.
The temperature of waters released from reservoirs is gener-
ally lower than in the flowing stream. Water temperatures can be
expected to approach the mean air temperature, the change being
proportional to the difference between the air and water temperature,
the rate of flow, the distance downstream, and other factors(4).
The reaeration constant is a characteristic of the stream hy-
draulics and should be the same after as prior to impoundment. The
reaeration constant varies greatly, however, with different flows(5, 6),
It is assumed for purposes of comparison that the waste loads
are not changed and that the deoxygenation rates of the waste loads
entering the stream below the reservoir are the same before and after
impoundment.
Observations should also be made to compare the pH of waters
from the reservoir with that of the free flowing stream. If there was
an acid problem in the stream before impoundment, we can anticipate
some improvement in the water released from the reservoir through
dilution and neutralization, since the most severe acid problems are
generally associated with short periods of runoff following periods of
less than average flow. Although the stoichiometry of acid mine
drainage waters is not well understood, the low flows and high flows
are mixed in the reservoir, and dilution and neutralization of the slugs
of low pH water should assure that the discharge from a reservoir
would be more nearly neutral than the slug discharges occurring
occasionally under natural flow conditions. It should be determined
whether the changed pH has any effect on the deoxygenation rate.
Acidity and alkalinity should be considered in the evaluation of pH
changes that may result from impoundment.
Flow Augmentation
The major variables in the oxygen sag equation have been con-
sidered, except the effect of quantity of flow released from the reser-
voir. How much water should be released from a reservoir? What
pattern of flow will the reservoir operation produce? The answers
to these questions depend upon the purposes for which the project
was authorized and the water uses downstream. Flow augmentation,
or regulation of streamflow, can be provided to serve a number of
purposes including public and industrial water supply, navigation,
power production, propagation of fish and aquatic life and wildlife,
recreation, and full or supplemental irrigation as well as water qual-
ity control. All such uses are considered in the multipurpose concept.
140
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
Two of these purposes — municipal and industrial water supply
and water quality control — are the primary interest of the Public
Health Service. The following discussion is confined generally to
storage for regulation of streamflow for purposes of water quality
control.
Flow augmentation implies the increase of an existing flow.
The stream's total flow cannot be increased except by unusual means
such as importation of water, but its ability to continuously as-
similate wastes without excessive quality deterioration can be im-
proved by storage of flood flows and subsequent controlled release
so that flows can be increased when natural flows are low and water
quality conditions are poor. Although flow augmentation is the topic,
we are really interested in increased waste assimilation capacity pro-
vided by increased flow and this, of course, means maintenance of
some selected quality objective.
Municipal and industrial waste treatment plants usually are
designed to maintain a desired water quality objective in the receiv-
ing stream based upon the anticipated waste load and the conditions
that most severely affect stream quality. For organic wastes for
which the oxygen-sag equation was developed, these generally occur
in the late summer and fall when flows are low and water tempera-
tures are high. Generally speaking, it has not been considered prac-
tical to design waste treatment facilities to maintain the desired
quality under the most severe or absolute minimum flow conditions.
The design flow has been determined after consideration of the
effect of lower flows on other water uses.
Treatment plants have thus generally been designed to allow
maximum use of the stream for waste assimilation, consistent with
other uses. The river below such a properly designed treatment plant
should maintain the water quality objective essentially all of the time
and be satisfactory for all legitimate uses.
The design flow often used is the annual 7-day mean low flow
that has a frequency of occurrence of once in 10 years. The 10-year
recurrence interval and the 7-day duration are used, for example,
by the North Carolina State Stream Sanitation Commission. The
design flow occurs within the critical period when, if dissolved oxygen
is considered as the parameter of quality, the most adverse combina-
tion of low flow and high temperature can be expected. Thus, if
minimum flow occurred in the winter when more oxygen is available
because of the higher dissolved oxygen saturation level and the lower
rate of waste decomposition, then minimum flows would be con-
sidered for the summer-fall period only, rather than for the entire
year. If color is the controlling consideration, however, minimum
flows at any time of the year would probably result in critical con-
ditions. In the eastern states, the area with which I am most familiar,
the design flow assures protection of aquatic life and the maintenance
of a minimum acceptable quality of water over 99 percent of the
time since the design flow is found to be equaled or exceeded in
excess of 99 percent of the time in this area. If this be true, quality
Vanderhoof
141
-------
lower than the objective occurs less than 1 percent of the time so
long as waste treatment satisfactorily removes excess waste constitu-
ents. Beyond that point flow regulation is needed, and the duration
and frequency of adverse conditions rapidly increase in its absence.
Peaking Power Discharges
If the main purpose of a reservoir is production of peaking hydro-
power, however, the downstream flow characteristics are seriously
altered, even for those projects that plan to release the design flow
at all times. In the eastern United States, flows at or below the design
flow occur less than 1 percent of the time under natural conditions.
It is estimated that flows at or below the design flow could occur as
much as 75 percent of the time, however, based on 8 hours of dis-
charge on weekdays under the altered conditions of peaking power
production; flows would of course become more uniform as a result
of channel' storage as the distance from the dam increased. The
changes in flow that result from peaking power production are illus-
trated by conditions downstream from Alatoona Dam, near Carters-
ville, Georgia(7, 8). Comparison of the duration curve of daily flow
before and after power production was begun shows that the flow
formerly occurring only 1 percent of the time now occurs 30 percent
of the time. Thus the daily average flow exceeds the design flow
only 70 percent rather than 99 percent of the time. Since daily
average flows were used, the actual occurrence of low flow would
actually be a much higher proportion of the time.
Hydropower plants that do not provide for at least the release
of the design flow used for sewage treatment plant design create even
more difficult downstream problems with respect to waste assimila-
tion. Such hydropower plants can drastically alter the hourly, daily,
and even weekly flow patterns when operated to generate peaking
power. In some instances, stream flows may be completely shut off
below the dam, especially during weekends. Such practices may not
appreciably affect or may even improve the average monthly flow,
but can completely alter the minimum continuous flow available to
receive treated waste loads downstream while maintaining acceptable
quality conditions. Peaking hydropower plant operations can thus
become a more limiting factor affecting the waste assimilation capa-
city of the river than the natural flows.
What is the effect of retarding the minimum flow? When the flow
is reduced as a result of storage during off-peak hours, pollution
continuously entering the river from municipalities and industries
accumulates in the channel downstream from the dam; when water
is later released for power production, the accumulated pollution is
swept downstream. The initial hydraulic wave is dampened as the
water flows downstream, but the wastes are carried downstream as
a slug of polluted water and are diluted only gradually. Under such
conditions, wide fluctuations in water quality may persist for a con-
siderable distance downstream, generally further than the fluctua-
tions in discharge, and may seriously affect downstream uses.
142
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
There are other and equally important effects of limited dilution
in addition to reduced dissolved oxygen. For example, toxic concen-
trations of wastes may adversely affect the aquatic life of the stream
for considerable distances. Concentrations of taste- and odor-produc-
ing materials may cause periodic conditions of unpalatable water in
downstream municipal water supplies. Coliforms and pathogens may
be carried to downstream water plants relatively undiluted and with
little die-off, resulting in increased hazard to the health of the com-
munity. Studies indicate that a modern, well operated water plant
can handle the above described situation(9). Equipment can break
down, however, and operation is not always at top efficiency!
The wastes discharged concurrently with the higher flows may
be more than adequately diluted, but the excess of available oxygen
is generally of little or no help in diluting the succeeding slugs of
polluted water. The larger volume of water discharged during the
production of peaking power will eventually mix with the slug of
waste, but often not until after the maximum effect of the wastes
has been felt. Average monthly, weekly or even daily discharges do
not adequately disclose the ability of a stream to assimilate wastes.
There is an additional difficulty observed with peaking power
hydroplants. The release of water accompanying peaking power pro-
duction makes the water course downstream hazardous for recrea-
tional uses and in a sense, limits the use of the river to power pro-
duction. The dedication of a river to a single water use is no longer
considered reasonable. All streams should be available for a variety
of uses, and should be physically safe for instream uses, as well as
of an acceptable quality.
Are we saying that peaking hydropower and waste assimilation
are mutually exclusive uses? That we can't have both? The answer is
definitely, "No!" These uses can exist together to mutual advantage.
There are means of improving the dissolved oxygen content of
reservoir discharges. A re-regulating structure can be built a short
distance downstream to release waters at a more uniform rate, thus
increasing the minimum flows and reducing the peaking flows. As
holidays frequently involve 3-day weekends, sufficient storage should
be provided in the re-regulating dam, or releases should be planned,
to maintain a continuous minimum flow over the 3-day period. Re-
regulation would probably eliminate the need to discharge water at
the power dam during off-peak power periods. The downstream flow
could be provided from storage by the re-regulating dam.
An advantage to recreation would be that the rapid rise of water
resulting from peaking flows would no longer be a threat to safety of
swimmers, boaters, or wading fishermen.
Since power production often involves increased summerflows,
an overall increase in waste assimilation capacity could result from
power production and re-regulation in that the design flow would
always be exceeded, compensating somewhat for the previously dis-
Vanderhoof
143
-------
cussed extension of the duration of occurrence of the design flow.
Some loss of reaeration could be expected, however, in the re-regu-
lation reservoir.
COMPATIBILITY WITH OTHER PROJECT PURPOSES
The single-purpose flood control reservoir generally has little
influence on the waste assimilation capacity of a stream since it has
a relatively small conservation pool. An exception is the provision
of quality control storage during the part of the year when floods
are less likely to occur. In the Ohio Valley a part of the flood control
pool may safely be used to store water for purposes other than flood
control from about April 1 to December 1. This part-time storage,
or "seasonal storage," may be used for flow regulation for quality
control. Such storage is extremely useful in the Ohio Valley as critical
periods (defined as the period when high temperature and low natural
flow result in the least assimilation capacity) occur within that period.
This storage can be added to a project with relatively little addi-
tional expenditure and can be very useful at the time of need. Some
modification of the normal outlet works may be necessary to provide
needed flexibility for water quality management.
Dams built to provide storage of water for navigation purposes
are certainly also useful in increasing the waste assimilation capacity
by increasing minimum flows during normal low flow periods. Addi-
tion of multiple outlets in such structures would make the project
equally applicable for regulation of stream flow to control water
quality. Whether there is a downstream need for water quality control
is another subject.
The navigation lock and dam structures may reduce the reaera-
tion coefficient of the canalized stream. Whether the reduction in the
overall reaeration coefficient is significant depends on the individual
situation. Differences in surface area, depth, and velocity of water are
all factors to be considered. A factor that should not be overlooked
is the increased flow that will be available in the stream if flow is
maintained for navigation purposes.
Reservoirs built for municipal and industrial water supply un-
doubtedly interfere with the waste assimilation capacity of the stream
during periods of low flow. Water stored for these uses is returned to
the stream as the effluent from a waste treatment plant and is not
available to assimilate wastes. In at least one situation in Ohio, a
city pays damages to downstream riparian owners where nuisance
conditions result.
In some of the 17 western states, use of water for quality control
purposes is not recognized as a legitimate use. If storage for regula-
tion of stream flow for the purpose of quality control is included in
such a project, it is not now apparent how such water can be reserved
for water quality control purposes when prior legitimate water rights
are not satisfied.
144
CHANGES IN WASTE ASSIMILATION CAPACITY
-------
CONCLUSION
Multipurpose reservoirs that include provision for storage for
regulation of streamflow for the purpose of water quality control
should of course be designed to increase the waste assimilation capa-
city by increasing the flows above the design flow as well as releasing
such flows with proper quality. The need for such flows is estimated
according to results of an economic base study. The nation's water
resources developments must be timely to ensure that lack of water
will not constitute a limit on our economic growth. On the other hand,
construction should not get too far ahead of forecasted needs for
water, for this would result in a part of the investment being un-
productive and therefore not efficient, It must be apparent that water
quality control storage must efficiently meet the anticipated need in
a timely fashion.
REFERENCES
1. Phelps, E. B. Stream sanitation. John Wiley and Sons, Inc., New
York, N.Y. 1944.
2. Streeter, H. W., and E. B. Phelps. A study of the pollution and
natural purification of the Ohio River. III. Factors concerned in
the phenomena of oxidation and reaeration. PH Bull, No. 146.
Feb. 1925. 75 pp. Reprinted.-
3. DWSPC, Region IV, Atlanta, Ga. and Robert A. Taft Sanitary
Engineering Center, Cincinnati, Ohio. Lake Lanier study, Lime-
stone and Balus Creeks. Dec. 1960. 65 pp. and 24 fig. Unavail-
able for distribution.
4. LeBosquet, M. Cooling water benefits from increased river flows.
J. New Engl. Water Works Assoc. 6:111-16. 1946.
5. O'Conner, D. J. and W. E. Dobbins. The mechanism of reaera-
tion in polluted streams. J. Sanit. Eng. Div., ASCE. 82: (SA6):
Paper 1115. 1956. 30 pp.
6. Churchill, M. A., H. L. Elmore, and R. A. Buckingham. The
prediction of stream reaeration rates. J. Sanit. Eng. Div., ASCE.
88; (SA4): 1-46. 1962.
7. Ingols, R. S. Effects of impoundment on downstream water qual-
ity, Catawba River. JAWWA. 51:42-46. 1959.
8. Ingols, R. S. Pollutional effects of hydraulic power generation.
Sewage and Ind. Wastes. 29:292-97. 1957.
9. Walton, G. Effectiveness of water treatment processes as meas-
ured by coliform reduction. PHS Publ. No. 898. 1961. 68 pp.
Vanderhoof
145
-------
DISCUSSION
Dr. A. Joel Kaplovsky*
Director, Delaware Water Pollution Commission
Dover, Delaware
In his paper, "Changes in Waste Assimilation Capacity Resulting
from Streamflow Regulation," Mr. R. A. Vanderhoof presents a hypo-
thetical case in which he discusses the effect of an impoundment
upon the stabilization of a waste originally discharged into a one-
directional-flow stream. The author follows with an actual example
by presenting a segment of his extensive studies with regard to the
change in water quality before and after an impoundment construc-
tion. The opportunity to evaluate an area of size, before and after
impoundment, is certainly the envy of most investigators. The author
is to be complimented on his approach and more specifically upon
his apparent capability of presenting a potentially complex problem
in a manner that has practical significance.
When Mr. Vanderhoof discusses changes upstream resulting from
streamflow regulation as induced by the construction of a reservoir,
he stresses the effect of time and its indirect influence upon resultant
water quality. This time aspect, when altered by changes in the
runoff pattern and specifically with regard to recovery capacity, is
probably the most important and least stressed factor in this type of
quality investigation. It was hoped that the author would have
pursued this time influence consideration at greater length.
In discussing potential change downstream, at various con-
trolled flows, the author stresses the resultant effect of sudden large
releases and/or "diversion" represented by releases of less than
critical design flow. Here again, however, it would have been desirable
for the author to pursue the time aspect and its indirect influence
upon resultant water quality.
The author's comments lead one to conclude generally that if
controlled releases are made from stored water in a manner proper to
prevent sudden fluctuation in dissolved oxygen, temperature, andl
flow, only beneficial effects with regard to waste assimilative capacity
would result downstream with increasing flow. At one time this con-
cept was accepted completely; however, in recent years extensive
estuarine studies have forced us to conclude that "dilution is not the
solution to pollution." This finding also has led us to look more
acutely into data from one-directional-flow systems.
We have endeavored to look hard at the influence of time and
temperature in association with increasing regulated flow. With in-
organic constituents, increasing flow results in lower concentrations
by dilution alone. With decomposable organic material subjected to
variations in time, temperature, and flow, however, we have found
~Now Director of Research and Control, Metropolitan Sanitary Dis-
trict of Greater Chicago.
146
DISCUSSION
-------
that reduced quality in estuarine systems can result with increased
flow under specific conditions. We have also observed that when a
drainage basin is considered in its entirety, it is possible to have the
overall beneficial effects exceed the undesirable aspects. Much de-
pends, however, upon which portion of a drainage basin is affected
by the highest concentration of pollution. The distance, timewise,
from the center of population and industry to a specific best water-use
area becomes critical and controlling.
In recent years we have been forced to conclude that assimilative
capacity may not have the same meaning to all investigators. The
author defines assimilative capacity as that "load which can be ac-
cepted without degrading the water below a minimum acceptable
quality." His effort to establish assimilative capacity before and
after impoundment for his findings, however, raises some doubt as
to whether our understanding of assimilative capacity is the same.
Changing the environment by means of an impoundment and thereby
inducing a change of water quality followed by a comparison of
conditions before and after does not automatically ensure that this
comparison constitutes a measure of the change in assimilative capa-
city. An understanding of the cause and effect relationship is required.
A measurement of change in concentration alone, before and after,
does, however, record the change between two points by changing the
flow pattern and time aspect, but it does not measure the full recovery
capacity of the system under varying time, temperature, and loading
conditions. The former approach may reflect the work done by the
system or net change in concentration, whereas the latter provides a
progressive measure of capacity of the system to assimilate wastes
without ill effects for future consideration. The tendency is to create
or induce a degree of dilution and conclude that the comparative
findings represent the extent of assimilative capacity. For example,
unless we provide a measure of ultimate oxygen denumd (only one
of many potential parameters) on a pound basis and fully consider
the multiple discharges of variable age, we are over-simplifying and
erroneously judging a complex problem. We must have a full under-
standing of the total input and output covering the period under
consideration. Perhaps, if we tentatively divorced from our thinking
the standard of quality desired for a specific use at a given point,
which incidentally can be achieved in some areas by dilution alone
with unlimited resources, and confined our evaluations to a true
measure of capacity to assimilate, we would more readily achieve our
objective. Assimilative capacity can be more concisely stated as a
measure of the net change in water quality within a given area under
a specific set of conditions when subjected to a series of increasing
and/or decreasing loadings added as continuous discharges.
In any applied-research stud^ of assimilative capacity, one is
unable to separate the evaluation of this capacity from the location
simply because in any implementation «f the findings the best water
usages of a specific area will eventually determine the adopted, criteria,
or desired water quality. Quality at any location is affected notably
by the proximity to the source of pollution and/or by the time of flow.
Kaplovsky
147
-------
As pointed out by the author, when we impound water we change
the flow pattern or the transit time in the "arms," i.e., the affected
upstream tributaries. Similarly, in a free-flowing system below an
impoundment, an increase or decrease in released water changes the
time of transit between two given points. The question arises, "What
will be the net progressive change in quality at various temperatures
and rates of flow between two given points?" Does the loss of time
for stabilization result in a condition sufficiently degrading to offset
the benefits attributed to dilution with increasing flow? For example,
if we had a waste discharge at point A, and a water intake down-
stream at point B, a loss of degradation time between A and B might
result in an undesirable water quality condition at B in spite of the
increased flow or dilution. This could be particularly significant with
bacterial discharges, where a loss of time corresponds to a lessening
in total number of organisms destroyed between the two points under
consideration. In an estuarine system, where the time of transit is
markedly affected by the river configuration as we move downstream,
the net water quality in the lower reach could become considerably
depressed with an increased rate of flow or regulated release, in spite
of the beneficial aspects of dilution evident in the uppermost reaches.
The generally accepted concept that increased flow improves
water quality is open to serious question when considering pollutional
aspects, particularly in estuarine waters. Studies of the Delaware
River prototype during the past 8 years have shown that the effect
of increased flow is not the same in all portions of the tidal area.
Initially, it was assumed that the shift producing decreased quality
in the lower reach of the estuary during periods of increased flow
was of a temporary nature. This reduced quality became more
marked with increased flows, and at much higher than normal flows
there was a tendency of reversal in which the quality began to im-
prove in the lower reach of the estuary. The question remained
whether this conclusion was sound and whether a series of hydraulic
model experiments could be performed to resolve this apparently
unpopular observation.
In an attempt to simplify or generalize what takes place within
an estuarine segment when a waste discharge enters the upper end,
the following analogy is made. If the flow or transit time to pass
through a segment is constant, a specific amount of stabilization of
organic matter is assumed to take place, depending on the deoxygena-
tion coefficient "k," at a given temperature. If "k," or the tempera-
ture should increase, the amount of stabilization would increase or
vice versa. If the flow increases, the time to traverse the segment
under consideration decreases, with a corresponding decrease in time
for stabilization. The latter would result in not only less stabilization
of organic matter within the segment under consideration, but would
project an increased loading of less stabilized organic matter into the
next downstream segment. On the credit side, however, increased
flow would result in decreased buildup concentrations. The question
is whether the loss of stabilization time within a given segment is
sufficiently compensated by the lower concentration buildup level
induced by this increase in flow.
148
DISCUSSION
-------
A series of tests with the Delaware River Model at Vicksburg,
Mississippi, was carried out with equal dye" loadings applied to the
same location at progressively increasing flow stages. It was con-
sidered reasonable that the effect of flow on organic stabilization
could be shown if the rate of deoxygenation and the decreasing
buildup with increased flow could be compared under similar condi-
tions. Initially, four consecutive segments of equal length were com-
pared. The model data were able to provide, for each of these seg-
ments, the percent change in buildup concentration with increasing
flow. In addition, the time of transit within each segment was related
to flow. Finally, the stabilization, expressed in percent of the original
concentration, was plotted against time under various deoxygenation
coefficients. The effect of flow on stabilization could then be illus-
trated. The average concentration within a segment during a specific
flow period was used. This procedure allows for the normal con-
centration gradient corresponding to increasing volume. In combining
the two effects (changes in stabilization and concentration), we
derived what we termed relative amounts of unstabilized organic
matter. The amount of stabilization is initially substracted from
100. This results in the percent of unstabilized organic material.
These values are multiplied by the corresponding concentrations. The
net increase or decrease of organic matter resulting from increased
flow conditions at various deoxygenation coefficients was plotted for a
series of consecutive segments. The Delaware River Model data pre-
sented in Figure 1 showed that at a constant "ki" value of 0.04, in-
creased flow is always beneficial in the uppermost upstream segment
under consideration and similarly, although not as marked, in the ad-
jacent downstream segment. In the more remote downstream segments,
0£
LU
E
5
Z
<
Z
D
IL.
0
H
Z
3
1
UJ
>
5
0.06
o 0-05
c
Q>
O
C
8 0.04
*
0.03
0.02
!
—-—
SEGMENT
50-80
80-110
1
110-140
1140-170
STEADY-STATE FLOW AT TRENTON, N.J., cfs
Figure 1 — Relative amounts of unstabilized decomposable organic matter remaining be-
tween progressive river segments under different flow conditions and at a deoxygenation
coefficient of .04.
Kaplovsky
149
-------
however, a decrease in stabilization was progressively evident up to a
streamflow of 5,500 cfs and thereafter additional flow increases were
beneficial; in the downstream stations flows as much as 8,000 cfs,
66 percent of the annual average, did not result in net stabilization
greater than at 2,000 cfs. This general pattern varies, however, with
the deoxygenation rate and/or temperature.
A typical example of the change resulting from different rates
of deoxygenation and/or changes in temperature is shown in Figure
2 for a specific 30,000-foot station interval. In general, at deoxygena-
tion rates of .072 and greater, increased flow results in progressively
Figure 2 — Relative amounts of unttabiliced decom potable organic matter remaining under
different flow condition* and at various deoxygenation rates between channel segments
110-140.
poorer conditions. At a lower deoxygenation rate of .05 (20°C) and
with decreasing temperatures comparable to the late fall, winter, or
early spring, conditions could result in essentially less change or even
improved conditions with increasing flow, all at the same location.
This phenomenon is even more evident as we move downstream.
The analogous effect of the transit time in an estuary leads to a
lummary observation. Temperature and transit time are critical and
1*0
DISCUSSION
-------
controlling. Accordingly, one can understand why a river system
subjected to a continuously uniform organic loading may show, at a
single location, a wide variation of quality with season. Further, if
the design of treatment facilities is to be guided by dry-weather flow
conditions, the resultant water quality and the controlling factors
occurring during this period directly influence the final design. Water
quality levels during other seasons normally become more of
academic interest. Understandably, the location of the pollution
source and the proximity of the recipient water user may create a
situation whereby the adopted dry-weather period might not be
the most critical time. Obviously, each system and each problem
area must receive an independent evaluation.
The following discussion of water quality changes in Brandywine
Creek by Mr. Vasuki* presents data compiled from a one-directional
flow stream and depicts the effect of season and flow upon certain
water quality parameters.
DISCUSSION: N. C. VASUKI
The Brandywine Creek has been an important source of water
supply in the lower Delaware Valley since the early colonial settle-
ments. The City of Wilmington (population approximately 98,000 —
1960 Census), Delaware, is situated at the lower reach of the Creek
and draws approximately 35 mgd for public water supply. In recent
months it has become increasingly important to establish existing water
quality and potential change with flow. The proposed construction
of an upper-reach reservoir and, therefore, controlled release of water
precipitated an immediate need for the evaluation of projected water
quality with varying controlled flow regimen.
The Wilmington Water Department has on record the various
chemical and bacteriological analyses dating back several decades.
For this evaluation, the period January 1, 1943, through December
31, 1961, was chosen. This represents a period of 228 consecutive
months. Analyses were made on samples collected at 8:00 a.m. each
day. The same two people analyzed the samples during the entire
period of the study. The results of the analyses for 6,947 days were
punched on IBM cards. Each card contained the following informa-
tion: date, corrected average river flow, turbidity, alkalinity, chlo-
rides and most probable number (MPN) of coliform bacteria. An
IBM 101 Statistical Machine was used for statistical analyses. These
analyses indicated the frequencies of occurrence of various concentra-
tions of the variables and the mean values of each variable for each
month, season, and year. The annual mean concentrations of turbidity,
alkalinity, and chlorides are given in Table 1. The annual loads are
shown in Table 2 and in Figure 3.
* Hydrologist, Water Pollution Commission, Dover, Delaware.
Kaplovsky
151
-------
TABLE 1. ANNUAL MEAN CONCENTRATIONS OF TURBIDITY,
ALKALINITY, AND CHLORIDE. BRANDYWINE CREEK AT WIL-
MINGTON, DELAWARE
Year
River flow,
cfs
Turbidity,
mg/1
Alkalinity,
mg/1
Chloride,
mg/1
1961
540
15.8
39.7
13.1
1960
585
21.9
38.0
15.5
1959
333
20.1
42.4
14.3
1958
722
26.8
35.4
9.5
1957
306
14.2
42.5
12.0
1S56
430
20.0
38.7
10.7
1955
339
17.0
41.7
8.6
1S54
247
18.0
40.4
10.1
1953
585
29.8
33.8
8.5
1952
677
27.0
32.5
9.4
1951
469
28.6
34.8
11.0
1950
402
23.9
35.8
11.0
1949
484
20.1
36.1
11.0
1948
544
31.7
31.9
11.1
1947
357
24.9
37.8
13.0
1946
453
33.9
36.4
10.4
1945
440
41.9
35.4
9.7
1944
325
37.3
35.6
14.7
1943
437
31,9
40.3
13.3
Routine coliform measurements, by season, are presented in
Figures 4 (a, b, and c). Grouping by season was intentional to
minimize the effect of variable temperature and thereby improve the
evaluation of the effect of the variable of flow. Most interesting is
the apparent increase in coliform content with increasing flow.
Sources of coliform reaching the Brandywine Creek are well up-
stream, and the reduced bacterial decay between source and down-
stream' intake associated with increasing flow is evidenced by the
progressively higher coliform levels.
On the basis of these preliminary examinations, further analysis
of the data is planned utilizing an IBM 1620 Computer. Additional
information on change in population, industrial output, land use by
agriculture, etc., is being collected. It is hoped that a sound deter-
mination of the change in water quality as a result of other man-made
changes will be possible.
Several interesting observations may be made from the data as
analyzed:
152
DISCUSSION
OPO 021 —740—6
-------
1. Chloride values declined commencing with 1953, followed by a
uniform increase since 1956. The latter trend has continued and
now chloride loads exceed values of former years.
2. Turbidity levels show two distinct patterns, namely, a continuous
decline over the entire period of record and a marked reduction
commencing with 1954. The latter coincides with new waste
treatment facilities at a paper mill 4 miles upstream. Imple-
mentation of contour farming during this period may have con-
tributed to the steady decline.
3. Alkalinity showed no marked deviation over the period of
evaluation.
4. A segregation of coliform data by season showed an increase in
coliform content with increasing flow. This persistent pattern
is attributed to a loss of time, associated with increasing flow,
for bacterial decay between upstream sources and the down-
stream water intake.
TABLE 2. ANNUAL LOAD OF TURBIDITY, ALKALINITY, AND
CHLORIDE (in Lbs. x 10«). BRANDYWINE CREEK AT WIL-
MINGTON, DELAWARE
Year
Turbidity
Alkalinity
Chloride
1961
16.770
42.139
13.917
1960
25.321
43.935
13.335
1959
13.156
27.753
9.377
1958
38.138
50.376
13.480
1957
8.541
25.563
7.192
1956
16.997
32.889
9.079
1955
13.034
31.972
6,562
1954
9,250
19.668
4.935
1953
34.360
38.973
9.798
1952
36.028
43.367
12.555
1951
26.293
31.993
10.156
1950
18.729
28.055
8.658
1949
19.175
34.438
10.481
1948
34.083
34.298
11.934
1947
17.521
26.598
9.147
1946
30.268
32.500
9.286
1945
36.337
30.874
8.412
1944
23.959
25.886
9.442
1943
27.476
33.247
11.456
Kaplovsky
153
-------
Noissiusia
Ml
TURBIDITY
lbs * 10"
1943 -
ALKALINITY
lbs x 1Q»
S o $ ggf
~n—I rr
CHLORIDES
lbs x ID8
TT
1944
1945 -
1946 -
1947 "
1948 -
I 1950
1955
1956
1957
1958
1959
1960
1961
I LI
-------
95
t*
o
<
09
X"
•<
50,000
50,000
400 600 800 1,000 1,200 1,400 1,600
RIVER FLOW, cfs
a. March, April, May.
50,000
10,000 -
100 200 300 400 500 600 700
RIVER FLOW, cfs
b. June, July. August.
100 120 140 160 180 200 220
RIVER FLOW, cfs
c. September, October, November.
Figure * — Col if or m bacteria (MPN per mt) versus flow. Bra nd yw't ne Creek at Wilmington, Delaware.
(n
(n
-------
DISCUSSION FROM THE FLOOR
Mr. Grounds, U. S. Army Engineers: Over the years, the Corps
of Engineers has improved the navigation channel on the Mississippi
River. They have done this by the construction of locks arid dams arid
by the dredging of the channels. This results, of course, in a series of
impoundments up and down the Mississippi River.
Do you think that the incidental benefits resulting from the
assimilation of waste in these impoundments could be applied in the
economic justification of future channel improvement projects?
Mr. Vanderhoof, Ohio River Basin Project: There is a great big
plus for quality control in the navigation system, insofar as upstream
storage dams are built to maintain a quantity of flow. Generally this
is released at the time of the year when it is also needed for quality
control. This is a plus.
At this time we are not sure just how much the lock and dam
activity affects quality and how much benefit or loss may be attribu-
table to the navigation locks and dams. In fact, the Ohio River
Division has us working on the effects of changing the Cincinnati
pool from a series of smaller locks and dams to the larger Markland
Lock and Dam project. We are only halfway through that, since
Markland Dam was just put into operation this winter. Post-
impoundment observations are scheduled for this summer (1963).
Mr. Grounds: In other words, we are not at a point yet where
we can actually say there is any definite benefit from main stem
navigation dams for assimilation of waste?
Mr. Vanderhoof: At this point, no, not from the locks and dams
themselves. But the upstream reservoirs that control quantities, if
they happen to release water at the time we need it, do benefit water
quality.
Dr. Kaplovsky: You were probably acutely aware in my dis-
cussion, that I represent a downstream state below a heavily popu-
lated, industrialized area in a tidal estuary, and also that we have to
take a slightly harder, more practical view of development upstream
because we are not only the sole recipients of everybody's beneficial
development, but we are also the sole recipients of everyone's errors.
I am not completely familiar with the area that you are referring to,
but if you are interested in the tidal portion — and I assume this is
why you have directed the question to me as well as Mr. Vanderhoof
— you would find that as you deepen and widen the navigation
channel, the pollution problems tend to change; e.g., saline water
moves further upstream,
If you build a reservoir and release water at low-flow periods,
the time of transit in the area downstream is reduced and the transfer
of pollution from one point to the next point downstream is decreased
time-wise. In other words, your buildup of pollution will be higher
in most cases in your downstream area unless you provide sufficient
156
DISCUSSION
-------
additional augmentation of flow to keep that build-up level at a
reasonable point.
A lot also depends upon what use you will make of the water
behind your impoundment. If it is to be used for water supply that
you are going to divert out of the area, then your use of this storage
will not provide additional increased flow.
Mr. Grounds: If you are contemplating an increase in the channel
depth, you are going to provide a greater surface area for oxygena-
tion in the impoundment. Is there any way to estimate this before-
hand?
Dr. Kaplovsky: I think we are talking about two different things.
While damming a river to increase depth results in a greater surface
area, the Corps of Engineers in our estuary area deepens the channel
by dredging, so that there is no increase in surface area.
It would probably be possible to estimate the change in aeration,
but I believe it would require extensive study. I should say that
observations at short intervals both vertically and horizontally would
be required. I, personally, would be hesitant to make an estimate
of how much improvement you would get in DO until you get an
indication of the depths involved.
Mr. McLean, TJ. S. Public Health Service: Mr. Vanderhoof and
I have worked on this problem in the Ohio River. The items involved
have already been outlined — the changing of time of flow and the
changing of the cross section. The changing of the cross section gives
more surface area for aeration, but reducing the velocity and in-
creasing the depth tend to reduce the amount of oxygen that goes
into the system per unit of surface area.
There will likely be increases in photosynthesis.
Our conclusion in the Ohio pool was that we need to observe
the changes in at least a few situations before we reach an answer to
this question. Our hesitancy in arriving at a definite answer to this
question is one of the reasons this symposium was called.
Dr. Krenkel, Vanderbilt University: The idea of reaeration is
increased because the presence of an impoundment increases surface
area is not always true. The entry of oxygen into water depends upon
the area over velocity (A/V) factor which is proportional to the
depth. If you have a long, deep channel the increase in depth may
actually cause a tremendous reduction of oxygen absorption or waste
assimilative capacity.
Furthermore, if we consider that the reaeration coefficient is
proportional to some form of turbulence intensity — I am sure every-
one will agree that it is — and if we take the mean stream velocity
as a measure of turbulence intensity, then the turbulence intensity
is reduced behind these impoundments, and the corresponding re-
aeration must also be reduced. Also, if we take the depth as a measure
of mixing length, then the mixing length is increased. I think it has
From the Floor
157
-------
been fairly well shown that the reaeration coefficient varies inversely
as the square of the depth. In other words, for a small increase in
depth, a tremendous reduction in reaeration capacity may result.
One other thing that has been encountered in several reservoirs
in the Southeast is thermal stratification — not the classical type we
talked about yesterday, but that which results from thermal dis-
charges from power plants. The cooling water may essentially in-
hibit mixing in the downstream or impounding reservoir. This can
cause a tremendous reduction in waste assimilative capacity.
A lot of factors in impoundments have not been considered that
I think should be.
Mr. Whitehouse, Central Electricity Research Labs, England:
Mr. Vanderhoof, considering the graphs in your paper, wouldn't you
regard the change in oxygen sag more apparent than real on the
following grounds: You said that when you put an impoundment on
a stream you leach out or oxidize organic materials in the soils that
are inundated and are, in effect, trapped in the impoundment.
I would suggest that this might be true in the upper reaches —
or the water in the upper layers — of the impoundment, but if you
take into account the carbon: nitrogen ratio of the bottom deposits a
different conclusion might be reached. If the carbon:nitrogen ratio
is greater than about 10, organic material accumulates on the bottom;
therefore, only in the surface water will you have this change in the
oxygen sag curve. This is a feature of normal stratification and will
increase the depletion of oxygen in the lower waters of a reservoir.
Do you think a real change occurs in the oxygen conditions
throughout the whole depth of the water, or are you deluding your-
self by just looking at surface waters for these results?
Mr. Vanderhoof: I think what you have said is quite true. In
many cases the improvement in DO is more apparent than real, but
it is observed. In the case of untreated waste you would of course
expect the deposition of settleable solids. I cannot disagree with any
of your comments. I think you again emphasize why we are here
today; we are facing a most complex problem.
Dr. Ingols, Georgia Institute of Technology: Having made some
observations downstream from a power impoundment receiving a
very heavy waste load, I would like simply to underline Dr. Kap-
lovsky's comments concerning the lower assimilative capacity of
somewhat higher flows. The nighttime minimum flow below this
reservoir frequently receives a greater pollution load than the peak-
ing flows because of a very much higher reaeration capacity. If only
one turbine is put on line the flow is about double the minimum, but
the increased dilution below the dam is not adequate to compensate
for the loss in reaeration so a very much more severe dissolved oxygen
problem occurs. Sometimes the scour of the organisms from the
bottom of the river at maximum flow creates a much more severe
oxygen deficiency than occurs at lower flows.
58
DISCUSSION
-------
Mr. Coutant, Lehigh University: Mr. Yanderhoof, you made a
couple of rather dangerous generalizations in your presentation. One
is that reservoir releases generally tend to be low in BOD as a result
of storage. I wonder if you have any comment on the fact that hypo-
limnetic waters quite often have a high BOD aside from any additional
wastes that are added to them?
Mr. Vanderhoof: Yes. I agree that I have made many assump-
tions here, but I have based these assumptions on the fact that high-
BOD hypolimnetic waters can be avoided by use of multiple-level
outlets. With proper selection of waters from the various levels of the
impoundment, you can avoid the very high BOD of which you speak.
Mr. Coutant: Also, with regard to pH, what are your comments
regarding an impounded bog valley in which seepage tends to be of
very low pH as a result of organic acids? This would be quite
different from a stream having high mineral acidity.
Mr. Vanderhoof: I hope I did not say that pH's were increased.
I meant to say that if there was a pH problem, it would tend to be
corrected as a result of mixing as the high flows tend to neutralize the
low flows in the reservoir.
With respect to the pH of water from bogs, I have never had
occasion to observe such a situation. Perhaps you could enlighten me.
Mr. Coutant: My very limited knowledge comes from the Pocono
Mountain area in eastern Pennsylvania where rather small reservoirs
have been built, mainly for recreational purposes, which have very
often been put across extinct lake areas. A good peat bog deposit
forms the bottom. When flows coming into the reservoir are low,
the seepage water coming up through the peat deposits on the bottom
and sides of the reservoir constitute a large percentage of the total
flow. This is rich in humic acids, and despite the fact that the normal
tributary waters are more on the alkaline side, it is a fairly common
occurrence to find low pH values, especially if the water is drawn
primarily from the hypolimnion.
Mr. Vanderhoof: Could not such a condition be corrected by
multiple-level outlets?
Mr. Coutant: Yes, definitely.
Mr. Whitehouse: Not in all situations, I would suggest, because
humic acids in impounded waters completely control the chemistry
of the waters irrespective of the oxygenation rates. Though photo-
synthesis may provide considerable oxygen, this does not prevent de-
oxygenation below a depth of 3 or 4 feet of water. As the organic
acids are being brought in all the time by seepage water, in the case
of humic acid impoundments, multiple-level outlets do not solve the
problem of drawing out water extremely low in pH.
From the Floor
159
-------
MINERAL QUALITY CONTROL THROUGH
STREAMFLOW REGULATION
Jerome H, Svore
Regional Program Director
Division of Water Supply and Pollution Control
U. S. Public Health Service, Dallas, Texas
Water is found everywhere on the surface of the earth, but in no
iwo places does it occur in the same quantity, quality, frequency, and
duration. Man's effort, historically, has been to improve this resource
to meet his needs or desires. Among those factors for which control has
long been sought in areas of limited water resources is mineralization,
either in the form of dissolved solids or suspended materials. The
amounts of materials carried into the ocean are tremendous. In the
United States alone, it is estimated that 270,000,000 tons of dissolved
solids and 513,000,000 tons of suspended matter are delivered by our
rivers to the tidal waters each year. The size of these figures makes
one wonder:
1. What areas contribute these materials?
2. Are they associated with the high rainfall areas?
3. Can these stream loads be controlled to an acceptable degree?
4. Can control be accomplished by managing stream runoff?
Consideration is given to these questions (not necessarily in the
order stated) in this paper.
WATER QUANTITY DISTRIBUTION
This country, logically, can be divided into three principal runoff
areas according to topography and moisture conditions:
1. The area east of the 95th Meridian, which has an annual
average runoff of 10 inches or more, may be classified as
moist. Within this area, all the larger streams have con-
tinuous-flow and well-developed stream networks.
2. The Great Plains area running north and south through the
central portion of the United States characteristically has a
minimum number of streams with continuous flows because
of low precipitation, low relief, and high evaporation-trans-
piration. There are periods of the year when runoff may be
very high, such as during spring thaws in the northern por-
tion or during severe rainstorms in the southern half of the
region.
3. The western part of the country is comprised alternately of
mountains and dry lowlands or plateaus. (The North Pacific
Coast Region is an exception to this pattern.) The runoff
in this area comes substantially from the mountains, which
provide the water used on the flat lands.
Svore
102
-------
For all the runoff regions, the flow is seasonal. In the arid and
semi-arid parts of the country, 75 percent of the year's runoff usuallv
occurs within a very few weeks. These widely varying runoff char
acteristics produce uneven streamflow patterns. In humid regions"
between 50 and 70 percent of the total annual runoff occurs durine
a short interval following rainstorms or the melting of snow or ice
Short periods of fluctuation occur in all streams, but in many of the
smaller ones peak flows are several hundred times the average flow
over the year. Salvaging quantities above the average flow would
m many cases, increase the stream's ability to supply a potential de'
mand. Generally, the larger and climatically more diverse the water
shed of the stream, the less severe the fluctuation in flow is likely to
be. The influence of extensive tree cover, as well as cultivation or
other land uses in farming areas, is clearly perceptible in runoff.
Infiltration rates are notably greater for grass-covered lands than
for crop lands. Close-growing crops, such as grasses, are associated
with higher infiltration rates than row crops. Fluctuations in runoff
are reduced where infiltration rates are increased. Thus, not only is
runoff distributed unevenly geographically, but also seasonally and
daily.
MINERALIZATION
Attempts to reduce variations in quantities of water traveling the
stream, unfortunately, produce other undesirable effects. Water, be-
cause of its superior capacity as a solvent, tends to increase in mineral-
ization with longer contact times. Other factors also influence the
suspended and dissolved solids loads carried by the streams.
In general, the mineral sediment content of surface water in the
United States is greatest in the arid or semi-arid sections. The sedi-
ment content also tends to be relatively large where water flows over
loosely consolidated sedimentary rocks, as in the Missouri Basin, or
over unconsolidated marine sediments, as on the Atlantic and Gulf
Coastal Plains. It is significant to note that mineral sediment in
stream flow or natural and artificial storage may come from as far
as 100 or more miles away.
Plate 3 in the U. S. Geological Survey Atlas HA-61 shows mean
annual weighted sediment discharge concentrations of river systems
in the United States. This value represents the amount of suspended
solids passing a given point in a definite time divided by the volume
of water discharged during the period. Differentiation should be made
here between suspended sediment and total sediment discharge. Sus-
pended sediment is made up of those particles that can be lifted by
the water from a place of residence and transported as a suspension to
a new site. Total sediment includes suspended particles plus the
material sliding along the stream bed.
Sediment concentrations shown on the plate represent the major
flowing portion of the stream. Streams of the vast majority of the
land east of the 95th Meridian and of the Northwest United States have
the lowest annual sediment concentration. The largest suspended
162
mineral quality control
-------
sediment concentrations occur in the band between the 95th and
107th Meridians and in the southwestern portion of the United States.
The concentration of suspended solids in streams draining 50 percent
of this land area averages less than 600 milligrams per liter. Water
flowing from 10 percent of this land area has over 8000 milligrams
per liter.
Plate 1 in the U. S. Geological Survey Atlas HA-61 shows the
general modal concentration of dissolved solids in streams of the
United States. Values shown represent larger streams, usually at the
lower-flow conditions. Smaller tributaries vary widely in water qual-
ity and quantity, making it difficult to present a value that can be
considered representative of the waters. A comparison of this plate
and the one showing sediment load discloses many similarities. This
is to be expected, since water assumes characteristics of the materials
it contacts.
With some exceptions, water in the United States east of the
100th Meridian, particularly from the Mississippi River eastward, has
a low dissolved solids content. The exceptions are the saline under-
ground waters, such as those underlying Kentucky and Michigan, and
certain industrially caused saline conditions such as on the Holston
River downstream from Saltville, Virginia. On the other hand, high
dissolved mineral content tends to characterize streams west of the
100th Meridian, particularly after they leave their mountain sources.
Exceptions to this pattern occur in the mountain sections themselves,
on the Pacific Coast north of the San Francisco Bay area, and west
of the Cascades and Sierra Nevada Mountains. As distance from the
oceans increases, mineralization becomes heavier. The water, in
general has its highest concentration of dissolved matter in the areas
of lesser rainfall. Fifty percent of the country's waters contain less
than 250 milligrams per liter total dissolved solids; 90 percent con-
tains less than 900 milligrams per liter.
CHEMICAL QUALITY
The mineral content of the waters of this country varies greatly.
The principal differences center on the amounts of sodium, chlorine,
calcium, carbon, magnesium, and sulphur contained. Other constitu-
ents that show minor but important differences are iron, manganese,
potassium, boron, silica, arsenic, zinc, selenium, lithium, and fluorine.
Surface and shallow ground waters in the arid and semi-arid sections
of tb-i United States, in general, contain more sodium, chlorine,
mag;:n:'.um, and sulphur than waters in the more humid sections of
the country. Waters of high bicarbonate and carbonate content are
found in both the humid and arid sections of the country. High cal-
cium concentrations are found in ground and surface waters over a
large section of the midwest and over the interior southeast, as well
as in the arid and semi-arid west. The occurrence of iron in water
at harmful concentrations is geographically sporadic, but has occurred
in all sections of the country. The same may be said for each of the
remaining secondary elements, such as boron, zinc, or fluorine. Plate
Svore
163
-------
2 m the U. S. Geological Survey Atlas HA-61 identifies surface water*
according to the more commonly identified gross water contaminants
The various divisions indicate regions where the waters are of th^
calcium-magnesium, carbonate-bicarbonate type; calcium-magne
sium, sulfate-chloride type; sodium-potassium, carbonate-bicarbonate
type; or sodium-potassium, sulfate-chloride type. Eighty-seven per
cent of the waters of the country are of the calcium-magnesium type"
WATER QUALITY CRITERIA FOR MINERAL CONSTITUENTS
Even when mineralization alone is considered, the criteria for
judging a raw water supply for its many uses, regardless of the means
of attaining it, are varied and complex. Sources of supply found
satisfactory for one purpose are not necessarily (in fact, usually are
not) adequate for other uses.
Domestic water supplies certainly are most important to us
individually and collectively. The criteria generally used for judging
drinking water in the United States are the U. S. Public Health
Service Drinking Water Standards, last revised in 1962. Technically
these standards apply only to water supply systems providing drink '
ing water used by interstate carriers and others subject to the Federal
Quarantine Regulations. The characteristics identified in Table 1 do
not directly measure the safety of the water, but they are related to
its acceptability to the consumer. At 5 units of turbidity, the water
becomes unacceptable to a number of people. Chlorides, sulfates and
dissolved solids are important because of their taste and laxative
properties. Iron is objectionable to the consumer because of the taste
it imparts to drinking water and the brownish color it imparts to
laundered goods. Excessive amounts of these constituents can result
in rejection of the supply by the consumer or in his providing indi-
vidual treatment. Desirable mineral quality ranges for several in-
dustrial uses are also shown,
SATISFYING QUALITY REQUIREMENTS
Because of the sensitivity of a number of uses and the infrequent
natural occurrence of a completely satisfactory supply of water
techniques for adjusting the quality-quantity demand requirement
are extremely important. In the past, three common ways have been
used to reduce the quality problem: (1) treatment of the supply
^21 development of uses tolerant to high dissolved solids content'
and (3) restriction of activities causing quality deterioration. All of
these methods are important because, in one way or another, thev
affect availability and regulation of aqueous flows. In this paper
however, control of quality by managing available water will be
given primary consideration.
The use of impoundments to dilute and secure a water of uniform
quality has been suggested as a means of making available the opti-
mum surface water resources of a drainage area. Although this
method may be useful insofar as quantity is concerned, there is seri-
164
MINERAL QUALITY CONTROL
-------
TABLE 1. DESIRABLE CONCENTRATION RANGES FOR COM-
MON CONSTITUENTS FOR VARIOUS WATER USES
*
£
Ui -T5
H 3
10
w
0)
C
TS ^h
« SP
X £
• n
c
•H
13 rr1
< a
T) 5P
~ s-
—I w
3 ss
° £ o
h T3 v:
CO
0)
T3
» H
2
o £
Cfl
0)
13 ^
^ W)
tfi E
o "oB
A 6
Domestic
water
supply
5 or
less
250 or
less
Hardness
plus 35
mg/1
or less
500 or
less
250 or
less
250 or
less
0.3 or
less
General
food
processing
1 - 10
10-250
30 - 250
850 or
less
0.2 or
less
Pulp and
paper pro-
duction
0-50
0-400
0-200
0-400
0-100
0-300
0-1.5
Textile
manufac-
ture
0.3 - 25
0-50
100 or
less
100 or
less
0.1 - 1.0
Irrigation
water (for
most plants
under most
conditions)
700 or
less
175 or
less
950 or
less
Cooling
water
50 or
less
50 or
less
0.5 or
less
* Silica scale.
ous doubt that dilution is economically justified if mineralized waters
containing high concentration of chlorides and sulfates contribute
significant loads and volumes over most of the year. An example of
the mixing technique can be shown from information available on one
of the larger manmade lakes in the southwest.
Lake Texoma was authorized by Congress in 1938 and con-
structed by the Corps of Engineers for flood control, hydroelectric
power, municipal and industrial water supply, and other allied
purposes. The surface area is 91,200 acres at the top of the power
pool and 144,100 acres at full flood control pool. Water depth at the
dam can reach a maximum of 135 feet. The major objectionable
contaminant in the lake waters is chloride. Two large rivers, the
Washita River entering from the north and the Red River entering
from the west, are the principal streams contributing to the reservoir.
The Washita River is over 600 miles long and drains nearly 8,000
square miles (primarily agricultural and ranch lands) of Central
Svore
165
-------
and Northwest Oklahoma. Some relatively small petroleum opera-
tions are located in the lower reaches of the stream. Flow and quality
information has been taken near the mouth of this stream for a
number of years. Seven years of data on chlorides is summarized in
Table 2. It is clearly evident that the concentration of chlorides may
be regarded as low — at least for the southwestern section of the
country.
TABLE 2. CHLORIDE AND FLOW DATA — LAKE TEXOMA
Washita River Red River Red River
above above below
Lake Texoma" Lake Texomab Lake Texomac
Water Flow,d
year cfs
Chloride,4'
Flow,d
cfs
Chloride,*'
Flow,4
cfs
Chloride,R
mg/1
ton/
day
mg/1
ton/
day
mg/1
ton/
day
1953
522
41
58
651
698
1230
1850
304
1520
1954
1260
36
122
3090
394
3290
4370
299
3540
1955
878
33
79
2630
462
3280
2760
309
2310
1956
445
62
75
2180
532
3140
3550
343
3290
1957
3490
26
245
7790
280
5890
11,100
258
7700
1961f
1135
46
142
3044
644
5296
4299
399
4630
1962'
1835
49
180
2605
464
3258
4340
403
4726
aAt Durwood, Oklahoma; the Washita River is 626 miles long and
drains 7945 square miles.
bAt Gainesville, Texas; the Red River above Denison Dam is 495 miles
long and drains 30,782 square miles (probably 5936 square miles is
noncontributing).
cAt Denison Dam.
^Yearly average.
eWeighted average.
'Values from U. S. Public Health Service Network.
The Red River enters Lake Texoma from the west. It is a sluggish
stream draining almost 31,000 square miles of northern Texas and
southern Oklahoma. The river and its several tributaries collect run-
off from a number of large oil-producing areas and 10 large natural
salt sources. Many of the tributaries in the upper Basin, including the
Red River itself, flow intermittently, allowing considerable storage in
the alluvial plains. As a result, the water entering the lake from this
stream is high in chlorides. None of the 7 years of record, 1953
through 1957, 1961, and 1962, lists a monthly average chloride con-
centration below the Public Health Service's recommended Drinking
Water Standard of 250 milligrams per liter.
Within the lake, mixing occurs between the two major streams,
other small tributaries, and runoff from the immediate area. Yearly
166
MINERAL QUALITY CONTROL
-------
weighted average concentrations tended to increase within the lake
for the period cited. In fact, with the exception of 1 year (1957), the
yearly averages substantially exceeded the levels recommended by
the Public Health Service for drinking water supplies. A review of
the monthly average concentrations reveals variation from 1 to 50
percent. If inflow remained fairly constant, concentration in the out-
flow tended to remain near the same level. Floods occurring in either
the Washita or the Red River Watershed diluted the chloride concen-
trations in the lake.
Some investigators, having noted the changes in mineral con-
stituents flowing into and out of large bodies of water, have expressed
the opinion that reservoirs in some manner act as storage sites for the
unaccountable amounts of minerals. Stratification of waters in the
relatively shallow lakes of this region has been found, but not deposi-
tion of the more soluble minerals on the lake bed. Studies conducted
by the Public Health Service during the Arkansas-Red River Basins
Water Quality Conservation Project revealed that dissolved chlorides
will eventually leave the lake. Attempts to find water of usable
quality at constant lake depths have not been successful. Apparently,
water entering the lake represents a highly unstable system, which
varies with time, stream volume, temperature, concentration, and
other chemical and physical factors. The combining of the water
from the two streams in Lake Texoma results in degradation of a good
quality water.
The period 1953 through 1957, selected for illustration of the
quality data for Lake Texoma, represents a period of severe drought
in this portion of the United States. The need for maintaining
desirable water quality in the lakes and rivers was most urgent.
Unfortunately, waters being released were already so highly mineral-
ized that their use for improving the quality of other waters was
limited. It is interesting to note that the city of Dallas, which was
suffering from an extreme shortage of water at the time, obtained
water from the Red River above Lake Texoma during 1957. In a
period of 3 months, 45,000 tons of salt flowed into Dallas homes from
the Red River, causing extensive damage to property and plants. The
city of Dallas since that time has made provisions for an adequate
supply of water.
An interesting quality control operation has been conducted for
a number of years on a tributary to the Red River above Lake
Texoma. It consists of a two-lake system constructed on the Wichita
River to supply water to irrigable lands in the vicinity of Wichita
Falls, Texas. The uppermost lake, identified as Lake Kemp, impounds
560,000 acre-feet of water when full. This lake catches the drainage
of three branches of the main stream and, more importantly, the
chloride contamination of three large natural salt-emission areas. An
average of 650 tons of chloride per day was found to be entering the
lake during 2 years of continuous study by the Public Health Service.
This structure at the present time is used for flood control, storage of
brine-contaminated water, and as a make-up source for approxi-
Svore
167
-------
mately 10,000 acre-feet per year needed to satisfy irrigation water
demand.
Lake Diversion, the lower lake, is a smaller impoundment, hold-
ing approximately 40,000 acre-feet of water. It is from this lake that
irrigation waters for the area are obtained. Insofar as possible, water
stored and used from this supply comes from the watershed below
Lake Kemp and is good water. Unfortunately, water demands, which
average about 30,000 acre-feet per year, exceed by about 10,000 acre-
feet the quantity that Lake Diversion can supply. Drawing the water
from Lake Kemp to make up this deficiency results in a decrease in
the quality of the irrigation water.
Quality characteristics from such a supply vary greatly, as would
be expected. During years of heavy rainfall, total dissolved solids
decrease. As rainfall lessens, chlorides and solids increase until the
water becomes unfit for use. Generally, water is considered unfit for
use when the total dissolved solids climb to about 2500 milligrams
per liter or when the chloride concentration becomes approximately
one-half of this value. Even now, after 2 years of above-average rain-
fall, the water for irrigation supply contains 1800 to 2000 milligrams
per liter of dissolved solids and 900 or more milligrams per liter of
chlorides. Only when Lake Kemp has a better quality water than
Lake Diversion is the latter lowered and filled with Lake Kemp water
prior to the irrigation season. Most of the time, however, Lake Kemp
holds the spring runoff without overflowing and thereby allows Lake
Diversion to fill with better quality water from the drainage area
below Lake Kemp. Nevertheless, this practice is not entirely satis-
factory and some land is often out of production because of low
tolerance to minerals.
The Public Health Service, Region IX, recently prepared a re-
port on the proposed Kellogg Reservoir in California. One purpose
of the reservoir would be improvement of water quality in the Contra
Costa Canal. At present, the canal receives its water from Rock
Slough, an off-channel slough of Old River, which over the years has
deteriorated in quality. During certain times of the year, however,
the water in Old River contains less than 100 milligrams per liter
chlorides and 400 milligrams per liter total solids. As planned, water
would be pumped from the Delta-Mendota Canal, which receives
water from the river, to the Herdlyn Reservoir and from there to the
Kellogg Reservoir. Water will be taken from the Delta-Mendota
Canal only during periods when water quality is good. The proposed
Kellogg Reservoir, with a maximum capacity of 135,000 acre-feet,
would provide good water to the Contra Costa Canal during other
periods of need. It is recognized that this comes more under the
category of selective withdrawal than under flow regulation, but it
does provide for the availability of better quality water.
A somewhat similar method of quality control was practiced by
a city in the north-central part of the country. This, too, was selective
withdrawal, as the city diverted river water during the spring runoff
to a large abandoned gravel pit. By this method, the underground
168
MINERAL QUALITY CONTROL
-------
gravel basin from which the city obtained its water was annually
recharged with excellent water. Since the construction of a large
reservoir above the city, both the quality and quantity of the flow
have been stabilized.
The water is still of good quality, but the city no longer has the
spring snow melt available as its source of supply. A major reservoir
on the main stem of a river did not, in this case, improve the quality
of the water available to a municipality; it merely averaged it out.
For the past few years the Public Health Service, with assistance from
Corps of Engineers, State Health Department, U. S. Geological
Survey, and others, has been conducting a study (Arkansas-Red
River Basins Water Quality Conservation Project) of the mineral
concentrations and loads in the Arkansas-Red River Basins. Specific
attention has been focused on chlorides, sulfates, and total dissolved
solids. In these basins, most chlorides appear to originate from 15
large natural-occurring salt (NaCl) sources and from brines from
numerous past and present oil-producing operations (See Figure 1).
The distribution of chloride load in the Arkansas River is pre-
sented in Figure 2 in average tons per day for the 2-year period of
the study. Contributions from the natural salt sources varied from
205 to 2850 tons of chlorides per day. Brine loads from oil-producing
operations were highly variable, depending on size of field, years of
operation, past and present water disposal practices, and other factors.
It is obvious from the many measurements that desirable water
quality can be obtained only if a considerable number of sources are
regulated or controlled to prevent the release of chlorides into the
receiving stream.
To obtain an idea of the control measures necessary to maintain
quality in the Arkansas River at the Keystone dam site near Tulsa,
Oklahoma, data obtained during the last 2 years of intensive study
were used to estimate the effect of controlling natural and manmade
chloride sources with flow conditions as they existed during the
drought period of 1951 through 1957. It was assumed that uniform
chloride concentrations existed at the site and that the majority of
the loads not originating from the natural salt sources were man-
made. Table 3 indicates that if a high degree of control of both
natural and manmade loads can be maintained this undesirable
contaminant can be reduced to a concentration well below that recom-
mended in the Public Health Service Drinking Water Standards and,
at times, to acceptable levels for other critical uses.
The feasibility of control is presently being studied by the Corps
of Engineers, Tulsa District Office. Basically, the measures considered
consist of one, or some combination of the following:
1. Diversion of fresh water around salt source areas.
2. Construction of lined channels through salt source areas.
3. Isolation of salt sources by levees, dikes, and dams.
Svore
169
-------
se
H
SB
>
£>
d
>
r
H
k!
O
O
5!
si
O
r
LEGEND:
j Monitoring sites with PHS
* identification numbers
Natural salt source areas
General oil and gas producing area
Figure 1 — Public Health Service Arkansas
Red River mineral surveillance network.
-------
Cfl
<
9
¦«
9
Little Arkansas River
224 — Valley Center 229 — Winfield
ARKANSAS
403 Van Buren
A NATURAL SALT SOURCES
* ONLY 1 YR. OF RECORD
2 Figure 2 — Typical material balance worksheet for the Arkansas River showing distribution of chloride load (in tons per day) — July 1960 to September 1962.
-------
4. Construction of lined channels through recharge areas.
5. Construction of dams and levees to impose a hydrostatic head
on salt sources.
6. Interception of brines and pumpage to subsurface strata.
7. Interception of brines and pumpage to off-channel storage.
8. Interception of brines and pumpage through a pipeline to the
Gulf of Mexico.
9. Provision of brine storage reservoirs on the channel and
diversion of fresh water around these reservoirs.
TABLE 3. EXPECTED CONCENTRATION OF CHLORIDES IN
ARKANSAS RIVER AT KEYSTONE DAM SITE NEAR TULSA
OKLAHOMA. UNDER VARIOUS CONTROL CONDITIONS," mg/1
Excedence
frequency/' Condition
%
1
2
3
4
5
6
1.0
1900
276
220
220
100
50
10
215
165
140
65
35
20
189
135
120
50
25
30
166
115
100
45
40
147
100
90
40
20
50
450
119
90
80
35
60
112
80
70
30
15
70
98
75
60
25
80
72
65
55
20
10
90
250
56
50
45
15
a Condition 1: No control of natural or manmade sources.
Condition 2: 100% control of natural sources; no control of man-
made sources.
Condition 3: 90% control of Areas, I, II, and III; 85% control of
manmade sources.
Condition 4; 90% control of Areas I, II, and III; 75% control of Area
IV; 85% control of manmade sources.
Condition 5: 100% control of Areas I, II, III, IV, and XII; 85% con-
trol of manmade sources.
Condition 6: 100% control of natural and manmade sources.
b Excedence frequency is expressed as percent of time the monthly
average chloride concentration will equal or exceed quoted value
under the specified control system.
172
MINERAL QUALITY CONTROL
-------
The usefulness of a stream is affected by the volume and quality
of its minimum flow. Any changes that increase flow at low stream
stages without increasing total dissolved solids excessively improve
the potential value of that streamflow and may reduce the need for
impoundment. The optimum combination of land use and water yield
practices should be the deciding element. Water-conserving farming
practices are directed not only toward the infiltration of water that
otherwise immediately would runoff, but also toward the reduction
of on site evaporation and unproductive transpiration. Such practices
include mulching, weed control, and shelter-belt planting. To the
extent that the additional soil moisture thus is directed toward pro-
ductive use, a net water gain and flow regulation are both achieved.
There are times, of course, when situations demand a supply of
water without regard to its mineralization. Such requirements have
been receiving increased attention during the last few years and are
of interest here because they center around having a constant water
supply available for use. Once available, a number of methods have
been used to make the quality of the water acceptable to the cus-
tomers. The basic methods or processes that are in use can be classed
as distillation, osmosis, solvent extraction, freezing, adsorption, electro-
dialysis, thermal diffusion, and ion exchange. With each of these
methods there are two major difficulties: (1) producing an accept-
able quality and quantity of water at a cost comparable to that for
water from alternative sources, and (2) disposing of mineral residues.
In this limited consideration of controlling mineralization of
streams, some attention needs to be paid to abilities of streams over
various portions of this country to maintain regulated flows as com-
pared to the estimated uses of these streams now and in the future.
Since it is impractical to consider streams individually here, the 22
regions defined in the Senate Select Committee on National Water
Resources Print No. 32, will be used as a basis for indicating a real
variability of the supply and demand for water from a physical
point of view.
Table 4 summarizes the estimated water needs for consumptive
plus waste dilution uses for the years 1954, 1980, and 2000. Three
broad categories of use were considered in arriving at the values
under each of the yearly headings. They were (1) demands necessi-
tating withdrawal such as agriculture, mining, manufacturing, and
municipal use, (2) in-stream flows required by hydroelectric power,
sport fish habitats, and waste dilution, and (3) consumptive losses.
Of all the demands considered, that for waste dilution was greatest
in most cases. These demands were based on organics, however,
not minerals.
Supplies of water, as listed in the last column, appear adequate
to meet the needs, except in those areas pertinent to this discussion.
A review of the regions with the previously defined highly mineral-
ized waters shows the coincidence of high total dissolved solids with
shortage of supply.
Svore
173
-------
TABLE 4. WATER DEMAND VERSUS MAXIMUM SUSTAINED
LOW FLOW FOR YEARS 1954, 1980, AND 2000
Maximum
Region
Water demand,"
bgd
sustained
1954
1980
2000
low flow, bgd
1
New England
48.9
14.2
19.4
64.7
2
Delaware and Hudson
25.1
13.8
19.8
31.1
3
Chesapeake Bay
32.1
24.0
26.9
50.8
4
Southeast
27.4
75.4
99.4
207.4
5
Eastern Great
Lakes
49.8
15.8
28.9
38.6
6
Western Great
Lakes
129.1
36.2
59.1
40.0
7
Ohio
10.2
19.4
34.3
108.5
8
Cumberland
3.1
7.7
12.7
15.5
9
Tennessee
12.6
24.4
29.7
42.6
10
Upper Mississippi
58.9
30.4
32.8
58.9
11
Lower Mississippi
6.5
22.2
36.1
47,5
12
Upper Missouri
33.0
33.6
51.1
26.9
13
Lower Missouri
1.9
3.6
7.1
14.5
14
Upper Arkansas-
Red-White
17.7
7.9
11.7
10.5
15
Lower Arkansas -
Red-White
16.1
16.9
27.3
70.3
16
Western Gulf
33.5
40.9
43.5
40.4
17
Upper Rio
Grande Pecos
10.9
7.3
10.3
1.0
18
Colorado
21.7
19.3
21.6
10.4
19
Great Basin
15.4
12.1
13.5
9.3
20
Pacific Northwest
17.5
42.8
44.3
136.3
21
Central Pacific
33.9
44.5
54.5
55.3
22
South Pacific
23.8
10.3
15.7
0.3
a Consumptive losses plus waste dilution flows.
The importance of quality improvement or quality adjustment
techniques in American water development is accentuated by the
size of most American drainage basins. Although water quality
problems have always been met locally, any upstream deterioration
of water quality is of concern everywhere within a basin. In the
extreme situations found throughout much of the West, cumulative
quality deterioration may eventually cause even the better waters to
become unusable if their quality is not protected. Quality improve-
ment by methods associated with streamflow regulation cannot be
ascertained without regard for the ultimate use, but it certainly does
not appear to be more than a stop-gap measure at best in securing
174
mineral quality control
-------
mineral reduction. Because of the nature of our available water
supply and its use and re-use, control of mineralization of the streams
as close to the source as possible would seem to offer the best single
solution.
Large quantities of controllable high-quality water in the upper
reaches of our rivers can improve conditions throughout the entire
length of a stream. Reservoirs below sources of mineral pollution can
level off the quality but mineral concentration may still be too high
to permit normal use of the waters.
BIBLIOGRAPHY
McKee, J. E., and H. W. Wolf. Water quality criteria, 2d ed.
Publ. No. 3A. State Water Quality Control Board, Sacramento,
Calif., 1963. 548pp.
Public Health Service drinking water standards. PHS Publ. 956.
Rev. 1962. 61pp.
National Water Quality Network annual compilation of data.
Public Health Service, Washington, D. C.
Water resources activities in the United States. Water supply and
demand. U. S. Senate. Select Committee on National Water Re-
sources. Committee Print No. 32. 1960, 131pp.
Reed, P. W. Augmenting stream flow as a pollution control
measure. Public Works. 93:92-96. Oct. 1962.
Ackerman, E. A., and O. G. Lof. Technology in American water
development. Johns Hopkins Press, Baltimore, Md., 1959. 710pp.
Desalination research and the water problem. Publ. No. 941.
Nat. Acad. Sci. Nat. Research Council, 1962.
Kneese, A. V. Water pollution, economic aspects and research
needs. Resources of the Future, Inc., 1962, 107pp.
Arkansas-Red River water quality conservation project. Public
Health Service, Region VII, Dallas, Texas, 1963.
Rainwater, F, H. Stream composition of the conterminous United
States. Hydrologic Investigations Atlas HA-61. U. S. Geological Sur-
vey, Washington, D. C., 1962.
DISCUSSION
John D. Hem
U. S. Geological Survey, Menlo Park, California
The geochemical role of rivers involves transporting the soluble
products of rock weathering to the ocean. The quality of the water of
a stream is dependent in general on two characteristics of its drainage
Hem
175
-------
basin — the available supply of soluble material and the amount of
water available for dilution. Generally, the second is more subject to
variation than the first. Hence, at low flow the concentration of
solutes in the water may become objectionable.
The factors governing sediment loads are different, in that the
concentration that may be carried is a function of the kinetic energy
of the stream and is at a maximum at high stages of flow. By de-
creasing the stream movement nearly to zero, a reservoir impound-
ment is very effective in sediment deposition, although at a cost in
storage capacity that may be uneconomic.
Because the dilution effect of released impounded water is the
average of the flow entering the reservoir, it cannot be as effective
as runoff from rainfall. As Mr. Svore has pointed out, in unfavorable
circumstances the impounded water may be too poor in quality to
be of any benefit to downstream water users.
Nevertheless, improvement in water quality by storage is readily
observable in many areas. The disappearance of suspended sediment
load has made most of the muddy Missouri River I remember from
younger days into a relatively clear stream. An important instance
of quality benefits is to be found in the Colorado River, as a result of
the construction and operation of Hoover Dam. The Geological Survey
has obtained water quality data on this stream for many years, and
effects can be evaluated closely from these data.
Lake Mead, with a storage capacity of some 28 million acre-feet,
can store about 2 years of normal river flow. The reservoir has nor-
mally been more than half full. The composition of the water entering
the reservoir varies continuously. Analyses by the Geological Survey
of samples taken daily at Grand Canyon, Ariz., during a period of 30
years show that the median dissolved solids concentration, the level
exceeded 50 percent of the time, was 960 milligrams per liter, and
that the dissolved solids concentration exceeded 1,000 milligrams per
liter more than 40 percent of the time. The median hardness of the
water during this period was 460 milligrams per liter. These are the
values characteristic of normal flow. During flood periods, especially
during the spring snow-melt season, dissolved solids values are lower,
but very seldom has a concentration less than 250 milligrams per liter
been observed. Values between 250 and 500 milligrams per liter
occurred about 20 percent of the time. At these times, the water
contained mostly calcium and bicarbonate ions.
The large amount of stored water in Lake Mead tends to iron out
the fluctuations in composition of the inflow, even though mixing is
incomplete. Considered on a time basis, the improvement in quality
that results is striking. Water released from the reservoir since 1937,
when the amount of stored water reached a moderate level, has never
exceeded 1,000 milligrams per liter in dissolved solids, and has been
between 750 milligrams per liter and 500 milligrams per liter about
80 percent of the time. The range of fluctuation of hardness also has
decreased, and the water has a hardness between 300 and 400 milli-
176
DISCUSSION
-------
grams per liter 87 percent of the time. The improvement in water
quality available to the southern California area and to others who
use this water for a public supply is highly significant.
It should be noted that reservoir storage has some quality effects
besides averaging out fluctuations. In Lake Mead some solutes are
added by solution of exposed gypsum and rock salt, and concentra-
tions of some ions are increased as a result of evaporation of water.
These effects are partly offset by precipitation of calcium carbonate
and silica in the reservoir. The net effect of these influences in Lake
Mead has been to give a small increase in dissolved solids in the water
as a result of storage. Effects to be expected in other reservoirs will
depend on local conditions.
DISCUSSION FROM THE FLOOR
Mr. Swiggart, U. S. Army Engineers: I would first like to make
a statement, as a citizen of Dallas, on the Dallas water supply. Yester-
day, Murray Stein mentioned the bad Dallas situation. Jerry just
briefly mentioned what happened there during the droughts of the
fifties. At the time of the droughts, Dallas had only about 100,000
acre-feet of water supply storage. We were in the process, through
cooperation with the Corps of Engineers, of increasing that storage to
500,000 acre-feet. The war, of course, delayed some of these plans.
Unfortunately, the drought came along before the storage was com-
pleted and we therefore had a bad problem. But the city of Dallas
now has over a million acre-feet of water supply storage and we are
in the process of adding about 400,000 acre-feet more. So, we now
have a good water supply in Dallas.
Mr. Svore, in your example regarding the mixing of water in
Lake Texoma, where we do not have good water primarily because
about 70 percent of the inflow is of poor quality and only 30 percent
can be considered good, isn't it true that the flow past the dam site
now is of considerably better quality than it was prior to the im-
poundment?
Mr. Svore: Yes, Forrest, that is certainly true. As far as Lake
Texoma is concerned, the resultant quality of water at the dam un-
questionably is much better than it was in the past.
I might add, too, that because of the quality problem Dallas
experienced 6 or 7 years ago, they are doing their planning a long
way into the future. If the Corps of Engineers plan is authorized, the
next hundred years are pretty well taken care of.
Dr. Gloyna, University of Texas: Mr. Hem, you showed that the
reservoir reduces the large fluctuations in the dissolved solids content
of the stream, but don't you agree that evaporation results in solids
build-up, so that the average base level of the dissolved solids will
very likely continue to rise in the reservoir over a long period of time?
From the Floor
177
-------
Mr. Hem: Yes, there is no doubt that the effect of evaporation,
along with the effect of solution of soluble rock from within Lake
Mead, has had this effect. The Geological Survey report on the opera-
tion of Lake Mead, Professional Paper 295, pointed out that the net
effect of the reservoir has been to give a somewhat higher dissolved
solids load below the dam than above it.
Dr. Gloyna: This is important for the smaller reservoirs, particu-
larly where water is recycled or reused within a reservoir system. I
think this is a very important point concerning the dissolved solids,
which should be stressed at this time.
Mr. Hem: Yes, I agree.
Mr. Churchill, Tennessee Valley Authority: I want to comment
on that point because I think we need to understand something that
apparently has not been brought out. This is the matter of the time-
weighted average concentration of total dissolved solids compared to
the flow-weighted average concentration. Although the mean con-
centration in terms of volume will be higher, the time-averaged con-
centrations will be lower. There is a big difference, and it is important
to understand it.
178
DISCUSSION
-------
CONTROL OF TEMPERATURE THROUGH
STREAMFLOW REGULATION
Milo A. Churchill
Chief, Stream Sanitation Staff, Division of Health and Safety
Tennessee Valley Authority, Chattanooga
SYNOPSIS
If water is released through low-level outlets in the dam, signifi-
cant effects on downstream water temperature result during the
warmer months. If water is released through high-level outlets, the
reservoir may have little effect on water temperatures. This paper
presents data observed in, and downstream from, certain impound-
ments operated by the Tennessee Valley Authority. The data have
been collected over a period of approximately 26 years. The paper
not only includes information to illustrate the magnitude of thermal
effects produced,# but also data designed to explain why and how
these effects are produced.
RESERVOIR OPERATION
In the TVA system, reservoirs are operated primarily for pur-
poses of flood control, navigation, and the production of electricity.
Two basic types of reservoirs are in use. Large, deep storage reser-
voirs are located on the principal tributaries of the Tennessee River.
In these pools, water levels usually vary over a wide vertical range
during the course of a year, with a vertical range of 100 feet or more
being not uncommon. On the main Tennessee River, navigable depths
must be maintained at all times throughout the full length of each
of the nine contiguous pools. This means drawdown must be limited,
and as a result the vertical range of pool-level fluctuations is rela-
tively narrow. Other reservoirs also operated within a narrow range
of pool levels are those relatively small pools located on principal
tributaries below storage impoundments. These pools are operated
primarily for power production.
Storage reservoirs are operated on an annual cycle. They are
drawn down during late summer, fall, and early winter to reach a
predetermined minimum flood-control level by January 1. The pools
are allowed to fill slowly after that date. As the probability of heavy,
widespread, flood-producing rains diminishes with advancing spring,
these reservoirs are permitted to fill gradually to their maximum
levels by early summer.
Since storage reservoirs are operated primarily to control floods,
these projects reduce the magnitude of high streamflows and raise
the general level of low flows. Although the general level of low flows
is raised, the usual practice is to reduce or eliminate releases over
weekends, and during other periods of off-peak power loads.
To fulfill its basic function, a flood-control reservoir must be
drawn down to and held at a relatively low level during the season
Churchill
179
-------
when floods can be expected. So that power can be generated with
water released during low-pool periods, low-level power intakes are
provided in the dams. As explained later, these low-level intakes are
the key to the effects exerted by a storage impoundment on water
temperature.
A map of the eastern half of the Tennessee Valley is shown as
Figure 1. Reservoir volumes, areas, depths, etc., for the projects
discussed in this paper are listed in Table 1.
TABLE 1. PERTINENT DATA ON SELECTED IMPOUNDMENTS
Full pool
Date of Volume, Area,
Depth at
Length, dam, ft.
Reservoir
River
closure
acre-ft
acres
miles
(approx
So. Holston
So. Fk. Holston
11-20-50
744,000
8,750
24
250
Watauga
Watauga
12-1-48
678,800
7,200
17
330
Boone
So. Fk. Holston
12-16-52
196,700
4,520
17
120
Ft. Patrick
893
Henry
So. Fk. Holston
10-27-53
27,100
10
80
Cherokee
Holston
12-5-41
1,565,400
31,100
59
150
Douglas
French Broad
2-19-43
1,514,100
31,600
43
150
Norris
Clinch
3-4-36
2,567,000
40,200
77
210
Parksville
Ocoee
12-15-11
91,300
1,900
7
120
Ft. Loudoun
Tennessee
8-2-43
386,500
15,500
55
95
Watts Bar
Tennessee
1-1-42
1,132,000
43,100
72.4
80
EFFECTS of storage reservoirs on
WATER TEMPERATURES DOWNSTREAM
Effects in open river channels downstream
A striking illustration of the effect impoundments can exert on
water temperatures downstream is shown in Figure 2. Water temper-
atures in the Clinch River at Clinton, Tennessee, are shown for 1935
and 1941. Norris Dam was built on the Clinch River approximately
20 river miles above Clinton, and was closed in 1936. Uncontrolled,
or natural, water temperatures and streamflows are shown for 1935,
The lower section of Figure 2 shows that significant releases from
Norris reservoir during 1941 were not made until late in May and
were continued through June and into early July. During most of
July and August, Norris releases were relatively insignificant again.
The water temperature at Clinton dropped approximately 30°F when
Norris releases began in May, and rose approximately an equal
amount when releases were cut off in early July. In late August,
temperatures again dropped approximately 30°F when releases began
anew. By the time the releases starting in late August were finally
cut back in late November, the temperature of local inflows had been
lowered seasonally to approximately 10°F below that of the released
180
CONTROL OF TEMPERATURE
-------
-------
water. (The small crosses on the temperature plot for 1941 will be
discussed later.)
This illustration of the marked influence on downstream water
_+,,rp, averted bv an upstream storage project may not be
completely understood by those unfamiliar with thermal stratifica-
tion in reservoirs. Temperature observations made throughout Chero-
r reservoir in 1945 (Figure 3) may help gain an understanding of
thfs important physical phenomenon - important because thermal
ll.
°
Ui
ca
100
D
h-
90
<
a:
80
Lxi
CL
70
S
LU
60
h-
50
cc
UJ
40
1-
<
30
20
18
Vi
16
O
14
o
12
o
o
10
LU
e>
CL
<
I
O
W)
u
cr
D
h-
<
X
UJ
CL
5
u
h
tc
UJ
I-
<
b RIVER WATER TEMP. AT CLINTON
JAN
FEB j MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
YEAR 1935
YEAR 1941
UJ
a
x
<
I
o
U)
5
100
90
80
70
60
50
40
30
20
18
16
14
12
10
8
JAN FEB MAR APR MAY JUN JUL AUG
SEP OCT NOV DEC
RIVER WATER TEMP. AT CLINTON
w. COMPUTED TEMP. OF
*IS WATER TEMP. EL. 866"
RELEASE FROM NORRIS
figure 2 — Temperature* and ditchargei in th* Clinch River below Norrii Rwervoir.
182
CONTROL OF TEMPERATURE
-------
stratification controls the hydraulic path followed by water flowing
into and through a reservoir. An understanding of reservoir hydrau-
lics is essential to an understanding of the effects a reservoir may
exert on water temperatures.
On March 1, 1945, there was little thermal stratification in Chero-
kee reservoir, as the isotherms show. Flow through the power intakes
during such isothermal periods is drawn from the entire cross section
of the pool, as the flow-net theory would indicate.
As spring advances, the warmer inflows from the tributaries and
the increased direct heating of the lake waters by the sun produce
thermal stratification in the pool. By June 6, the warm upper stratum
(the epilimnion) was separated by the thermocline (which shows
rapid changes in temperature with depth) from the colder waters of
the hypolimnion below. Wind and diurnal atmospheric temperature
changes mix and aerate the waters of the warm epilimnion. The
thermocline marks the lower limit of the mixing effect.
During thermally stratified periods, flow through the power in-
takes is drawn from the hypolimnion exclusively, since the warmer
waters of the epilimnion simply float on the cooler, more dense waters
below, just as would oil on water. Since draft from the pool is
through the deep intakes, the warmer waters from above settle down-
ward to occupy the space vacated by the colder discharged water.
Water velocities in the hypolimnion, induced by the draft, are, of
course, very low except near the intake. Water is discharged accord-
ing to its temperature (density), the coldest (most dense) water at or
above the intake level going out first. In this way, water many miles
upstream from the dam may be discharged ahead of water at the dam
stored only 50 feet or so above the intake elevation but having a tem-
perature several degrees warmer than that at intake elevation. As
shown for June 6, the cold water trapped below the intake level cannot
be discharged (unless the sluices are opened) since its density keeps it
at the lowest elevation possible. Naturally, some mixing of this water
with the overlying moving water takes place at the interface, but
even so the water in this low pocket is changed very slowly.
By September 8, the waters of the hypolimnion, as the hypolim-
nion existed on June 6, had been discharged from the pool. On Sep-
tember 8, the thermocline was at the intake level. Very little cold
or cool water from the previous winter remained in the pool.
Figure 3 also shows the effect in Cherokee reservoir on October
18, 1945, of then-cooler inflows from upstream. In the fall of the
year, the tributary inflows are, of course, cooled more quickly by
cooler fall air than the great mass of reservoir water. Consequently,
these cooler inflows move into and through the pool along the bottom
as density underflows (1). Such underflows move, by gravity, rather
slowly along the old river channel, most of them having velocities of
a few tenths of a foot per second. With a velocity of 0,25 foot per
second, approximately 12 days would be required to pass through
a pool 50 miles long. For a given volume of inflow, the velocity of
the underflow is increased as the density difference of the two waters
is increased and as the slope of the riverbed is increased.
Churchill
183
-------
«h>>M irti •>!(
oo
O
o
w
o
r
o
"9
M
S
« 2
sS
s >
' 3
f §
!> m
w m uli
Pu«t Ml*
Figure 3 — Thermal stratification in Cherokee Reservoir, March to November 1945
(isotherms shown in degrees centigrade).
-------
Density underflows continue in the fall until the overlying pooled
waters are cooled by vertical circulation to such an extent that there
is little if any difference between the temperature of the inflowing
waters and the pooled waters. For the latitudes of the Tennessee
Valley, the pools are essentially isothermal from the middle of De-
chamber until about the end of March or the middle of April. Dunring
this period, the isothermal water offers little resistance to wind-
induced mixing, and consequently, mixing continues all winter. There
is no period of winter stagnation in reservoirs at these latitudes.
Figure 4 illustrates several interesting facets of reservoir effects
on downstream water temperatures. A continuous record of water
temperatures below Cherokee Dam on the Holston River for 1945
is shown, together with a plot of water temperatures on the Holston
River 47 miles downstream. Note that very little difference existed
between these two records in the summer months except when Chero-
kee releases were cut off. In other words, very little warming of the
released water took place in the 47 miles of open river flow. (Little
data are available on the time of water travel through this reach,
but it is known that a flow of 7,900 cfs (cubic feet per second) re-
quires 24,5 hours to flow from Cherokee to a point 50 miles down-
stream (2).)
Figure 4 — Temperatures and discharges of the Holston Itiver.
Note that the temperature of the released water, just below the
dam, gradually increased as the summer progressed. If Cherokee
reservoir were larger and deeper or releases during the spring and
summer had been at significantly lower rates the temperature of the
released water would have increased more slowly as summer pro-
gressed.
Churchill
185
-------
Water temperatures for 1934 and 1935, as observed for Clinch
River at Clinton, Tennessee, are included in Figure 4 to indicate at
least approximately what the temperature of the Holston River would
have been during the summer of 1945 had Cherokee reservoir not
existed. There was no reservoir on the Clinch River during 1934
and 1935.
During periods of thermal stratification, water is drawn from
storage impoundments according to its density if the impoundment is
provided with low-level power intakes. To demonstrate this, the
temperatures of releases from Cherokee were computed for the sum-
mer months from a combination of (1) the temperature "profile" of
the reservoir for June 6, 1945, (2) a record of water volumes re-
leased day by day after June 6, and (3) an elevation-volume curve
for the reservoir.
In this computation, the starting date was June 6. The tempera-
ture of water in the reservoir at the centerline of the intakes on this
date was taken at 15°C (59°F) from the reservoir survey data (see
Figure 3). Horizontal lines were drawn by eye through the plotted
isotherms shown for June 6 in Figure 3 to determine a specific eleva-
tion to be used in computing the volumes of water in the pool with
temperature between 15"C and 16°C, between 16°C and 17°C, etc.,
on up in temperature and elevation to the upper edge of the thermo-
cline which in this case showed 22°C. (Such computations probably
should stop at the center of the thermocline.) The date on which
water at 16°C should have been discharged was determined by adding
up daily discharges after June 6 until a volume was obtained equal
to that volume in the pool on June 6 between the 15°C and 16°C
lines previously drawn. This date was found to be June 15. Corre-
spondingly, a cross was plotted for June 15 at 16°C (60.8°F) in
Figure 4. By continuing such computations, the date on which water
at each degree of temperature between 16°C and 22°C should the-
oretically have been discharged was determined and plotted. The
results, as indicated by the crosses in Figure 4, confirm the hypothesis
that water is withdrawn from such pools during the summer months
from the thin, horizontal stratum at or very near the intake level,
and only after this water is discharged can warmer (less dense)
water from the stratum next above settle downward and be dis-
charged in turn.
A similar computation was made for estimating the temperature
of Norris releases during 1941, starting with the thermal distribution
of water in the reservoir as of June 6, 1941. Computed temperatures
for the released water are shown as small crosses on the plotted data
in Figure 2. Here also, the above hypothesis of draft from a stratified
pool is confirmed.
It follows that the temperature of releases from a deep reservoir
can be predicted for several months in advance during the spring and
summer if estimates are available on reservoir releases.
Effects in reservoirs located downstream from storage impoundments
Just as density underflows are created during the fall months
in the most upstream storage impoundment by cool unregulated river
186
CONTROL OF TEMPERATURE
-------
inflows, underflows are created during the spring and summer months
in downstream reservoirs receiving cool releases from upstream stor-
age impoundments. One example of this phenomenon is shown in
Figure 5.
Fort Patrick Henry reservoir, a small pool on the South Fork
Holston River is located downstream from two large storage projects
(South Holston and Watauga) and immediately below Boone reser-
voir. On Thursday, April 29, 1954, the rate of release of cool water
from Boone reservoir to the upper end of Fort Patrick Henry reservoir
was 3,400 cfs. This inflow passed through Fort Patrick Henry reser-
voir as a density underflow. The inflow rate was cut to zero at 11
p.m. on April 30, and held until 8 a.m. of May 3. Note the isotherms
for Sunday, May 2, show the last of the water in the cool inflow that
was cut off at 11 p.m. on April 30, had by that date (May 2) slid
down into the lower levels of Fort Patrick Henry reservoir, and
warmer water from the epilimnion had moved upstream to fill the
space recently vacated by this sinking, sliding mass of cool inflow.
Underflows also existed throughout this small pool on July 16 and on
September 27, 1954 (Figure 5). In other words, density underflows
normally exist in this reservoir during all of the warmer months.
The data on dissolved oxygen also shown in Figure 5 confirm the
flow patterns indicated by the isotherms.
Another example of a density underflow produced by thermal
effects is shown in Figure 6. Here the underflow is passing through
Fort Loudoun reservoir along the bottom as a result of the combined
cool releases from the lower levels of both Cherokee and Douglas
reservoirs during the summer months.
Note how parallel to the river bottom (disregarding the irregu-
larities of the riverbed) are the sloping isotherms of the thermocline.
The underflow mixes but little with the warmer water above and
maintains essentially the same velocity (about 0.25 feet per second
in this particular case) from the duck-under point at about river
mile 630 on down to the dam at mile 602.3. A method for determin-
ing whether a density underflow will be created and, if so, for estimat-
ing the velocity of the density current has been presented earlier(3).
The cool water discharged at Fort Loudoun Dam flows through
Watts Bar reservoir as a density interflow (Figure 7). In other words,
this water does not pass through Watts Bar along the bottom, nor
through the surface stratum, but in between. The reason for this is
that still-cooler (and thus denser) inflow from the Clinch River arm
of Watts Bar forces the water from Fort Loudoun up off the bottom.
The cooler Clinch inflow passes through Watts Bar as a true density
underflow.
Figure 7 shows a profile of the lower end of the Clinch River
embayment of Watts Bar pool. The position of the left end of the
Clinch profile in the figure denotes the river mile (567.7) on the
Tennessee at which the Clinch River enters the Tennessee. Also
shown is a profile of the Emory embayment, which is tributary to the
Clinch and joins the Clinch at Clinch River mile 4.4.
Churchill
187
-------
TOP or MTU
TEMP, °C
'54
AY
00, mg/l
7/16/54
FRIDAY
»o» »f wro iau
•f#C
•fcUMV CWIT
DO, mg/|
9/27/54
MONDAY
Figure 5 — Thermal stratifications and dissolved oxygen concentration*, Fort Patrick Henry
Reservoir — 1954.
188
CONTROL OF TEMPERATURE
-------
TIMNCSSK v _ C ruQJWtiE - *I»C*
Figure 4 — Water temperoturej in Few* Loudoun Reservoir, July 20-21, 1944 (isofharm shown
in "fj.
WATTS
BAR DAM
S29 9 _ 540
740
720
RIVER MILE
sso
-------
drains back out into the Clinch River and on downstream along the
bottom.
Density underflows are more than just interesting hydraulic
phenomena. They can produce both beneficial and detrimental effects
on man's many uses of reservoirs and reservoir waters. In the case
of Fort Loudoun reservoir (Figure 6), the effluent from the Knoxville
sewage treatment plant is absorbed into the cool water and carried
through the major portion of the reservoir along the bottom, thus
reducing the detrimental effects of the sewage plant effluent on
recreational uses of the downstream portion of the pool.
Probably the most beneficial use of density underflows in the
TVA region, from an economic standpoint at least, is the use of these
cool waters for steam condensing purposes in large steam-electric
power plants. One such plant, the Kingston Steam Plant on the lower
Clinch River, is so located that it can draw cool water from the Clinch
via the lower Emory embayment. Details of this interesting situa-
tion and of how a "skimmer" wall was designed and built to exclude
the warmer surface strata have been published in two earlier papers
(4, 5). On the Cumberland River near Nashville, Tennessee, the TVA
Gallatin Steam Plant takes advantage of cool underflows passing
through Old Hickory reservoir.
The underflow in the Tennessee River below the mouth of the
Clinch River absorbs the treated municipal and industrial wastes
from the city of Harriman, Tennessee. These colored wastes are dis-
tributed across the streambed by means of a multiport pipe diffuser.
They are absorbed into the underflow and do not rise to the surface.
If a density underflow did not exist here, some surface discoloration
could be expected that might be objectionable to recreational and
other interests.
EFFECTS OF RESERVOIRS WITH HIGH-LEVEL OUTLETS
ON TEMPERATURE OF RELEASED WATER
As stated earlier in this paper, reservoirs provided with high-
level outlets in the dam have much less effect on water temperature
than do those with low-level outlets. Perhaps the best available
illustration of this situation is Parksville reservoir on the Ocoee River
in east Tennessee. Water temperatures observed a short distance
above the dam in this reservoir for each of the years 1943 to 1947
are shown in Figure 8. Temperatures of the penstock discharge are
also shown for 1955.
Since the power intakes are located high on the face of the dam,
the outflow is drawn from near the surface of the pool. This means
that the warmest water available in the pool is drawn out. The result
is that temperatures very similar to those that would have been ob-
served under natural river conditions are found here. The warm
inflow simply passes through the surface layers of the pool, over the
cooler water below, and is discharged at the dam with essentially
the same temperature it had on entering.
190
CONTROL OF TEMPERATURE
-------
JM Fit MM AN Mt JW JIH MW XF OCT KC
Figure 8 — Thermal Gratification in Parktville Reservoir at dam showing effect of high-level
power intakes (isotherms in degree* C.).
Churchill
191
-------
If cool inflows from upstream, released from the lower levels of
a storage impoundment, enter a pool with high-level outlets, the
pattern of flow through this pool will be essentially the same as in
Parksville but the outflows will be somewhat cooler than local inflows
during the summer months. Inflows during the early spring from the
storage pool upstream will have essentially the same temperature as
the pooled water, and under such conditions will mix with the pooled
water As the supply of cool water from upstream gradually dimin-
ishes and slowly warms up, the flow path gradually will change,
however from one that occupies the entire cross section to one that
occupies'only the upper strata of the pool. The extent of warming
these overflowing waters receive during the summer months will
depend upon the length and width of the pool and the rate of draft
from the pool. Naturally, in short, small pools with relatively high
rates of draft, little warming will be produced. Conversely, long,
large pools with relatively low rates of draft will retain the inflow
much longer and it will be warmed more not only by the sun and
atmosphere, but by warm local inflows that enter the surface stratum.
CONCLUSION
Water control structures may exert a profound influence during
the warmer months of the year on water temperatures of both im-
pounded and released waters. By understanding the forces that
control these influences, the desirable effects can be used advan-
tageously and the less desirable effects can be controlled or perhaps
avoided.
references
1 Fry A S., M. A. Churchill, and R. A. Elder, Significant effects
of density'currents in TVA's integrated reservoir and river sys-
tem. Proc. Minn. Intern. Hydraulics Conv., IAHR and ASCE,
Minneapolis, Minn. Sept. 1953.
2. Churchill, M. A. Discussion of "Translatory Waves." Trans.
ASCE. 110:1229. 1945.
3. Churchill, M. A. Effects of impoundments on oxygen resources.
In: Oxygen Relationships in Streams. Proc. of seminar, Cincin-
nati, Ohio, Oct. 30-Nov. 1, 1957. Tech. Rept. W58-2. SEC. Mar.
1958. pp. 107-24.
4. Elder R. A., and G. B. Dougherty. Thermal density underflow
diversion, Kingston Steam Plant. J. Hydraulics Div., ASCE.
84: (HY2) : Paper 1583. Apr. 1958.
5. Harleman, D. R. F„ R. S. Gooch and A. T. Ippen. Submerged
sluice control of stratified flow. J. Hydraulics Div., ASCE, 84.
(HY2)-.Paper 1584. Apr. 1958.
192
CONTROL OF TEMPERATURE
-------
DISCUSSION
G. Earl Harbeck, Jr.
Hydraulic Engineer, U. S. Geological Survey, Tucson, Arizona
The effects of varying temperatures of reservoir outflows are
generally considered to be important with respect to biological, chem-
ical, and physical quality. Most aspects of these effects have been or
will be discussed by several well-qualified speakers. Certain features
of temperature control of outflow, as mentioned by Mr. Churchill,
and some results thereof appear worthy of additional comment.
Of particular interest in Mr. Churchill's paper are the results of
his computations of temperature of releases from Cherokee and
Norris Reservoirs. Using reservoir temperature data and information
on depths and volumes of withdrawals, he was able to verify his
hypothesis that summer withdrawals come from the thin, horizontal
stratum at or very near the intake level. If the volume of outflow
were to be increased greatly, however, the validity of the hypothesis
would have to be demonstrated.
Some work has already been done on the physical theory of with-
drawals from stratified reservoirs, particularly by D. R. F. Harleman
and R. A. Elder, but as far as the writer is aware, little has yet ap-
peared in the literature. The U. S. Geological Survey has begun pre-
liminary work on the problem using analog techniques. Most engi-
neers will concede that if water is withdrawn from a million-acre-
foot reservoir by means of a Vfe-inch diameter pipe through the dam,
the outflow temperature will be that of the water in the reservoir at
the depth of the withdrawal. If the volume of outflow is increased,
however, to say 100,000 cfs, the validity of the assumption is ques-
tionable. What is needed is a technique for withdrawal temperatures
that will take into account withdrawal depth, variations in reservoir
density, volume of outflow, and physical characteristics of the dam
and reservoir.
It appears that dam designers in the future will give careful
consideration to the desirability of providing for reservoir with-
drawals at any selected depth. It appears also that the operator of
the dam will continue to be subject to criticism for his choice of with-
drawal level, for it will be difficult if not impossible to satisfy all
downstream interests at the same time. For example, a sudden in-
crease in outflow temperature, which might result from changing
from a deep to a shallow withdrawal, may evoke anguished wails
from avid trout fishermen; the reverse may bring on outraged howls
from downstream rice growers who insist that the colder water will
delay the harvest.
The temperature of reservoir releases can have a pronounced effect
on the hydraulic characteristics of the stream, particularly in winter.
Although the effect on the stage-discharge relation resulting from
temperature-induced changes in viscosity (1) may be statistically
significant, it is small. The possible effect of discharging warm water,
Harbeck
193
-------
however, (as from a steam powerplant using river water for cooling)
should not be ignored. If the site is in the North where streams are
normally ice-bound throughout the winter, men engaged in the wide-
spread stream gaging activities of the U. S. Geological Survey will
applaud because locations where open-water conditions prevail in
winter are few. The discharge of warm water may not be a blessing
to all downstream riparian residents, for although the stream may be
completely ice free for some distance below the point where warm
water enters, shore ice will begin to form, and slush ice may cause
ice jams downstream where complete ice cover again prevails. With-
out the warm water inflow, the stream might remain under complete
ice cover all winter throughout its length, with no riparian damage
save that traditionally associated with the spring breakup. On the
other hand, the addition of warm water in winter does not guarantee
any such unwelcome consequences; an orderly transition from open
water to complete ice cover may occur.
A seldom-considered effect of cold and warm water withdrawals
on the physical characteristics of reservoirs is that of evaporation
losses. Generally, this is of little consequence in the humid East, but
it can be shown, theoretically at least, that some saving in reservoir
evaporation would result if it were possible to make withdrawals by
skimming only the warmer surface water. For Lake Mead it was
shown that the annual saving in evaporation might be 8 percent (2),
This was a maximum figure, however, and it was considered that
skimming the lake was not practicable from an engineering stand-
point. Moreover, it was realized that the evaporation loss from the
next reservoir downstream would be increased. Incidentally, the
computed annual saving of 8 percent at Lake Mead was equivalent
to about 70,000 acre-feet, which is more than the capacity of most
other reservoirs in the United States; so the idea is attractive if
perhaps impracticable.
An example of extreme temperature conditions is the shallow,
meromictic lake, described by Anderson (3). Located in north cen-
tral Washington, the lake has no outlet, and is saline of course. It
has an average salt gradient of approximately 100 grams per liter at
the top to 400 grams per liter at the bottom. The maximum tempera-
ture was more than 50°C in July 1955 at a depth of about 2 meters.
Perhaps more startling was the fact that in December 1954, the lake
was completely frozen over, yet the maximum temperature at the
bottom was 33°C. Admittedly these are conditions that are not to be
expected in reservoirs, but they serve to illustrate the effects that
storage can have on water quality.
The control of temperature through streamflow regulation may
well have its broadest application in reservoirs used for steam power-
plant cooling. The effects of adding heat from a powerplant on the
energy budget for a reservoir can be computed with adequate accu-
racy (4) when air-water temperature differences are small. When
large temperature gradients exist, however, the commonly used
Bowen ratio concept appears inadequate for computing the amount of
194
DISCUSSION
-------
energy conducted from the water surface to the atmosphere. The
basic equation for this type of heat flux is
H = —c p K„ (a T/d z + r )
P
in which H = heat flux
c = specific heat of air at constant pressure
r
p = density of air
Kh = heat less coefficient
aT/az = temperature gradient
r = adiabatic lapse rate
Considerable work has been done to develop instrumentation for the
direct measurement of heat flux, and in the writer's opinion, further
research in techniques for measurement of eddy flux of heat, momen-
tum, and water vapor is most desirable. The theory is basically sim-
ple, but additional information concerning the eddy diffusivities of
those three items is needed.
REFERENCES
1. Eisenlohr, W. S. Effect of water temperature on flow of a natural
stream. Trans. Am. Geophys. Union. 29(2): 240-42. 1948.
2. Harbeck, G. E., M. A. Kohler, and G. E. Koberg. Water-loss in-
vestigations — Lake Mead studies. USGS Profess. Paper 298.
1958.
3. Anderson, G. C. Some limnological features of a shallow saline
meromictic lake. Limnol. and Oceanog. 3:259-70. July 1958.
4. Harbeck, G. E., G. E. Koberg, and G. H. Hughes. The effect of the
addition of heat from a powerplant on the thermal structure and
evaporation of Lake Colorado City, Texas. USGS Profess. Paper
272-B. 1959.
DISCUSSION
John J. Gannon
Associate Professor, School of Public Health
University of Michigan, Ann Arbor
Where cool water is discharged from impoundments during the
warm-weather months, the temperature change will be most notice-
able in the open river channel immediately downstream from the
impoundment. Mr. Churchill presented several illustrations of TVA
reservoir systems that regularly discharge cool water to downstream
river stretches because of low-level power intakes at dams that draw
water from lower reservoir levels. This temperature change is not
permanent, however; it is a temporary condition with the river
water temperature gradually increasing and approaching what might
Gannon
195
-------
be called a natural water equilibrium temperature at some noint
downstream This equilibrium temperature will prevail under natural
conditions where heat losses due to evaporation, convection, and radf
ation will balance the heat gain from solar radiation. Computational
methods employing energy budget relationships have been preseS
by Velz and Gannon (I for determining this equilibrium water tem
perature, using the meteorological elements of air temperature w£d
velocity, vapor pressure, and solar radiation.
The discharge of cool water and the resulting increase in
temperature until it approaches the natural equilibrium temperature
.s analogous to the condition where waste heat is discharged to »
river, resulting m elevated river water temperatures with *,,?
decrease toward the natural equilibrium temperature Here^S
computational methods are available for defining either thTvf^ ?g • '
or heat dissipation and the resulting watef tu e profit
Ploying energy budget relationships and using the previous!v- m™"
tioned meteorological elements, together with the water snST
If river channel characteristics arc available
distance, it then becomes Dossihlp tn if- UildLe area to
temperature to specific downstream locations. r6SU mg mer water
The methods employed by Velz and
weather bureau data from nearby weather hnwa VI- USe routine
most applicable to average conditio"and«*
possibly a month. Generally, they should nnt an„r ! * ek 0r
shorter than a week. Because many of the mir-m +¦ Periods
to perform, a computer program has bo™ wS ™ , ",1 tedious
employing the MAD computer language for both dot! ^ 7090
equilibrium water temperature anSlh^^,? wa "oS
the ehan,!t rcsuuine
change, bu/rather ,. tempo r™^^^
eventually approaching natural equilibrium conditions If %Z$Z
river-water-temperature profile information is needed for ?, 8
poses as evaluation of waste assimilation capacity etc it is
to calculate this profile with reasonable accuracy " possibl*
REFERENCES
'' SSms. wre/: Jn%Tnl ZT^e "eat loss itt p°nds an,i
DISCUSSION FROM THE FLOOR
Mr. Churchill, Tennessee Valley Authority: This paper is an ex-
tension of Mr. Kittrell's fine paper yesterday, and of course the dis-
cussion of it by Dr. Harlem an. I attempted to show actual instances
196
DISCUSSION
-------
where reservoirs have affected — controlled, if you like — water
temperatures downstream. We need to understand the magnitude of
these changes in actual instances, and also to go into the mechanics
a little bit more, into the reservoir hydraulics, if you please, because
if you fail to understand the reservoir hydraulics picture, you fail to
understand many of the impacts reservoirs have on water quality.
The temperature is important because it controls the hydraulics and,
therefore, it controls the effects produced; hence the emphasis on
temperature in this symposium.
Mr. Coutant, Lehigh University: Mr. Churchill, in Figure 2b of
your paper, the temperature values for water released from the dam
in December, January, and February are apparently above the normal
river temperatures. Are any data available to support the belief that
this difference in temperature can be called insignificant, while the
differences in temperature noted for May, June, and July can be
considered significant, with regard to the ecological effects on the
downstream areas?
Mr. Churchill: The dotted line is the temperature at elevation
866 in Norris Reservoir, the center line of the intake in the pool. The
water in the river downstream is somewhat colder because of local
inflow and is also more responsive to the seasonally colder air tem-
peratures than the great mass of water in the reservoir. The trend of
the temperature of elevation 866 is downward all through that period
and hits a low point at about the middle of March.
Mr. Coutant: Perhaps my choice of your Figure 2b was unwise.
The point I would like to make is that higher winter temperature
is a very typical phenomenon of reservoir discharges. Especially in
the higher latitudes, the water released from reservoirs having deep
outlets tends to be considerably higher in temperature than the
stream temperatures normally occurring in the area, which in the
winter months may be zero degrees centigrade. I was wondering if
you had reasons for not considering this temperature difference as
significant as the summer temperature difference.
Mr. Churchill: Of course, the difference is not as large. I suppose
that is the first reason I did not consider it. Also, when the water is
that cold in the TVA area, a few degrees difference one way or an-
other doesn't make too much difference, but if you are talking about
30 degrees difference in summer time, that does make a difference.
Something I did not bring out, but probably should, is that this
cold water released from Norris is used at the Kingston steam plant,
one of our own large steam electric power plants downstream near
the mouth of the Emory River. Dr. Harleman, by the way, had some-
thing to do with the design of the skimmer wall that was built there
to hold back the warm water in the surface layers and allow the
cold water flowing down from Norris to be drawn underneath. The
steam plant thus takes advantage of the cold water released from
Norris some 80 miles upstream. Its use for cooling is a very important
economic aspect.
From the Floor
197
-------
Mr. Coutant: Well, I hope we agree that the microbiology of the
stream is an important aspect of its quality — the plankton, periphy-
ton organisms, plus the bacteria. In a number of cases, microbiological
populations have been shown to be fairly responsive to temperature
changes, and I wonder if the change in the winter temperature of the
stream, such that the minimum temperatures do not get down to the
freezing point, has changed the microbiology to a significant extent.
Mr. Churchill: The answer to that is easy: I don't know. I have
no data on that.
Mr. Dougal, Iowa State University: Mr. Churchill, could you
describe your program of temperature measurement? What size sec-
tion do you use? What types of boats? How do you make the meas-
urements, and do you take them every year at all of your reservoirs?
Mr. Churchill: In the early days we did observe temperatures in
every reservoir every year, but we found, as you might suspect, that
the temperature pictures repeated themselves so closely year after
year that we no longer do this.
We have portable equipment and use thermisters, which are very
easy to use, so it is simple to get a vertical traverse. It is a job that
can be done very easily with a boat; we can survey a whole reservoir
50 miles long in less than a day. Data are easily and quickly col-
lected.
Mr. Dougal: What is the last year for which you have taken
measurements? I notice some of the charts extend through about
'55 to '57.
Mr. Churchill: We observed temperatures on Watts Bar and
Chickamauga Reservoirs just this last year, but not on any of the
reservoirs shown on the figures. We are currently working on these,
but do not attempt to keep records on all of the reservoirs each year.
We do collect weekly samples of water from the outflow of each of
our reservoirs, and the local collector observes water temperature at
the time of collection. In that way we have a complete record of
outflow temperature every week from each of our dams.
Mr. O'Connor, Manhattan College: Would you indicate, please,
the inflow hydrograph to Cherokee Reservoir for the period June to
October 1941, that coincides with the outflow hydrograph on your
Figure 2b?
Mr. Churchill: That would depend upon the operation of reser-
voirs upstream, now. Before the upstream reservoirs were con-
structed, it would have been something like the hydrograph shown
on Figure 2a for the Clinch River at Clinton below the Norris Reser-
voir site. Under natural conditions, summer time temperatures are
higher and discharges are somewhat lower, but with the regulation
going on upstream, this picture would be changed. I cannot give
information, off-the-cuff, on the effect of current operation; I would
be glad to send it to you.
198
DISCUSSION
-------
Mr. O'Connor: I have noticed from your previous work that the
minimum depth of reservoir that you dealt with was about 80 or 90
feet and that in all cases you had significant stratification.
Mr. Churchill: That is right.
Mr. O'Connor: At what depth of reservoir would you expect min-
imal stratification — at the 30-, 40-foot range?
Mr. Churchill: I do not know. Most of our pools are deeper than
that, so any answer I would give would just be a guess. Your guess
would be better than mine, I expect. Somebody else here may be
able to answer your question.
Mr. Towne, U. S. Public Health Service: I have been listening to
this discussion with considerable interest. I think we in the Columbia
Basin are extremely interested in temperature and have done some
work on the subject of predicting the temperature effect of reservoir
releases for considerable distances downstream. Mr. Robert Jaske has
been involved in the General Electric Hanford operation and will, I
hope, give us a little bit of background on this. I think they have a
considerable amount of information that would be very valuable.
Mr. Jaske, Hanford Atomic Products Operation: Our contract
operation with the Atomic Energy Commission at Hanford involves
the use of a tremendous amount of cooling water for the reactor
operation at that site. We are interested in the optimum use of this
water resource for our own cooling purposes, and also for the sec-
ondary purpose of minimizing the effects of our heat releases to the
communities downstream.
We noted in the records, the development of a thermocline at
Lake Roosevelt through the years much as Milo Churchill has shown;
in fact, the data we collected could almost be superimposed on the
Cherokee thermocline information in his figures. Consequently, we
considered the possibility of using this cold water from Lake Roose-
velt to our advantage and to the advantage of the downstream com-
munities.
We started releasing cold water from the depths of Lake Roose-
velt in 1958, originally on a daily cycle to counteract the diurnal
cycle and later on a continuous basis to take care of the average
effect on the river temperature. We have been successful in the past
5 years in reducing the average temperature of the Columbia River as
much as 1.5 to 3 degrees centigrade through this process.
We are currently studying the heat budget associated with this
release. Before we consider this, I should stress the size of this par-
ticular river. On the Columbia River, anything less than 30,000 cfs is
considered drastically low flow. This is a large stream. The stream-
flow past the Hanford plant during the peak period is generally about
400,000 cfs; in some years, it has been as high as 800,000 cfs.
One lesson we have learned from our work there is the danger
From the Floor
199
-------
of excessive generalizations from results in other areas. I note in
this conference and in other discussions with people a tendency to
take the experience of one particular region and to immediately apply
it to every situation that comes to mind. For instance, the turbidity
in the Columbia peaks at the period of high flow, which is just the
opposite of what one would expect. Other changes associated with
the annual and diurnal cycles in our area vary widely from what you
would expect in some of the smaller rivers.
In our work on the heat budget, originally we followed the
example of the people who worked on Lake Mead and then on Lake
Hefner, and we attempted to correlate the factors in the heat budget
with conditions that obtained downstream. We were not as suc-
cessful as Milo Churchill in determining the fate of the cold water
that was released from Lake Roosevelt, for two reasons: First, the
volume of bank storage is very high in our area because a lot of
porous gravel exists that is very permeable, resulting in the release
of large quantities of water; second, we found to our dismay that we
need a tremendous amount of micrometeorological data to correlate
the heat transfer and radiation effects to the various temperature
waters.
We do not have the equipment to conduct the extensive program
needed to do a complete investigation. We have satisfied ourselves,
however, that the temperature of the Columbia River can be regulated
successfully for public purposes by the proper operation of the reser-
voirs. This proper operation, however, as noted by Murray Stein
yesterday, does not always coincide with the most efficient use for
power generation or for other purposes. Our own use of the river
has been inadvertently compromised to some extent by the manipu-
lation of reservoirs operated by the various agencies upstream.
One conclusion we have drawn from our entire work — and we
drew this very reluctantly and in the face of some opinion to the
contrary — is that the erection of dams and reservoirs on the main
stem rivers in the Columbia Basin, notably the Columbia and the
Snake Rivers, has served to increase the temperature of these rivers.
This temperature increase we estimate in the range of 1 to 2 degrees
centigrade for each major structure that has been erected. The fish-
eries biologists in the Pacific Northwest have been studying the inci-
dent of columnaris disease in salmon, and they note an exponential
effect on the incidence of columnaris with increases in temperature
above 70 degrees. For this reason, everyone out our way is concerned
with temperatures of the river. We are hopeful more agencies will
become interested in this particular facet of water quality as it relates
to the erection of dams and storage reservoirs.
Mr. Kittrell, U. S. Public Health Service: I want to ask Milo
about the two reservoirs where they have been practicing repumping
during off-peak power periods. These were Hiwassee and Apalachia,
were they not? Have you had an opportunity to observe any possible
effects of this repumping on temperature or dissolved oxygen in these
reservoirs?
9AA
DISCUSSION
-------
Mr. Churchill: I believe the use of that equipment to pump water
back upstream has been so intermittent and so infrequent that to
attempt to find the effect would be straining at gnats. We have not
tried to find it.
Dr. King, U. S. Fish and Wildlife Service: I am interested in Mr.
Churunni's explanation of temperature conditions on the Cherokee,
as I am sure all biologists are. There is one point I think would be
well to mention to show the close relationship between temperature
variations and the fishery. On his temperature chart for Cherokee
Reservoir, the cold water is exhausted some time during the summer,
certainly during August or by the first of September. The water is
going down in elevation very rapidly during that period, too. About
9 months of the year the stream below Cherokee has temperatures
in the mid-fifties and mid-sixties for 50 or 60 miles — it's a beautiful
stream, ideal for boating and at that time of year good for trout.
You can imagine, though, what happens when the cold water is ex-
hausted and the temperature rises into the seventies during the late
summer and early autumn. This is the problem the fishery managers
face. After attempting to stock the stream with trout for several
years with very little success, they investigated and found the tem-
perature pattern mentioned here.
In this case, 9 months of the year a high recreational potential
exists, but for 3 months the temperature conditions are such that
the fishery disappears. The water is too cold to support warm-water
fish most of the year since it does not permit spawning, but during
late summer and fall it is unsuitable for trout. So it is impossible,
with our present knowledge, to develop a useful fishery in this water.
Another instance of interest is Lake Mead below Hoover Dam.
There are now 11 or 12 miles of excellent trout fishery below Hoover
Dam that weren't there originally. There is also a good trout fishery
below the thermocline in Lake Mohave. The Fish and Wildlife Service
has just completed a new hatchery at a cost of over half a million
dollars to supply trout to stock the water there. Here is a case where
the temperature phenomena in reservoirs are contributing very
materially to an excellent fishery, and we all expect it to continue.
From the Floor
201
-------
Session 4
EFFECTS OF FLOW REGULATION
ON WATER QUALITY (Part 2)
Moderator: E. C. Hubbard
North Carolina Department of Water Resources
-------
THE EFFECTS OF NUTRIENTS ON
PHOTOSYNTHETIC OXYGEN PRODUCTION IN
LAKES AND RESERVOIRS
Kenneth M. Mackenthun
Technical Services Branch, Division of Water Supply and Pollution
Control
V. S. Public Health Service, Cincinnati, Ohio
A reservoir or lake is the settling basin of a drainage area. The
potential productivity of a lake is determined to a great extent by the
fertility of the land that drains into the lake and by the contributions
of civilization. Biological activity within the lake influences such
chemical parameters as dissolved oxygen, pH, carbon dioxide, iron,
manganese, phosphorus, nitrate-nitrogen, etc.; it is governed by
temperature and stimulated by nutrients (e.g., phosphorus and nitro-
gen). A lake's basin gives dimension to biological activity and may,
because of unique physical characteristics, concentrate the nutrients
it receives as well as the developing biomass.
BASIC NUTRIENT SUPPLIERS
Basic sources of nutrients to lakes and reservoirs are tributary
streams and waste discharges, precipitation from the atmosphere, and
the interchange of bottom sediments.
Sewage and sewage effluents enrich tributary streams. Rudolfs(l)
studied the content of sewages from 12 separate sources and con-
cluded that the annual per capita contribution of phosphorus ranged
from 0.6 to 1.5 pounds. Studies of Wisconsin oxidation ponds indi-
cate annual per capita contributions of 4.1 pounds of inorganic nitro-
gen and 1.1 pounds of soluble phosphorus(2). The secondary treat-
ment effluent of the combined sewage and industrial wastes of Madi-
son, Wisconsin, has an annual per capita contribution of 8.5 pounds
of inorganic nitrogen and 2.8 to 3.7 pounds of soluble phosphorus.
By diverting its treated sewage effluent around Lakes Waubesa and
Kegonsa, this city reduced the inflow of nutrients into those waters
by 3,143 pounds per day of inorganic nitrogen and 1,342 pounds per
day of soluble phosphorus(3).
Lakes and reservoirs located on heavily used duck flyways re-
ceive "flying in" nutrients from a transient duck population. Sander-
son (4) found the annual raw waste contribution of a domestic duck
to be 2.1 pounds of total nitrogen and 0.9 pound of total phosphorus.
Paloumpis and Starrett(5) applied a factor of 0.5 to these data and
determined that the annual nutrient contribution to Lake Chautauqua,
Illinois, from the wild duck population was 12.8 pounds of total
nitrogen and 5.6 pounds of total phosphorus per acre.
Land runoff may often be the principal contributor of nutrients
to the tributary stream. The annual loss per acre from a planting of
Mackenthun
205
-------
corn on a 20 percen s^°Pe of Miami silt loam was found to be 38
pounds of nitrogen an 1.8 pounds 0f phosphorus(6). On an 8 percent
slope, the annual loss P®r acre was reduced to 18 pounds of nitrogen
and 0.5 pound of phosphorus, in a study of the lower Madison lakes
Sawyer et al.(7) and Lackey and Sawyer(S) found the annual con-
tribution of inorganic nitrogen per square mile of drainage area
tributary to Lake Monona to be 2,800 pounds, to Lake Waubesa, 3,130
pounds, and to Lake Jvegonsa, 4,100 pounds.
Lake Koshkonong, a 10,000-acre lake in southcentral Wisconsin,
receives the Rock Riv®r and two minor tributaries. Based on calcu-
lated normal flows, La e Koshkonong receives 91 pounds of inorganic
nitrogen and 40 poun s of soluble phosphorus per acre per year.
During high flows, the nutrient influx may reach as much as 8 pounds
of inorganic nitrogen and 1 pound of soluble phosphorus per acre
per day. Tributaries to the Rock River account for 64 percent of the
drainage area to the outlet of Lake Koshkonong. Twelve monthly
samples collected at the outlet in 1943-44 showed a mean total phos-
phorus concentration of 0.21 milligram per liter and an inorganic
nitrogen concentration °f 0.37 milligram per liter(7). These samplings
were repeated in 1959-60. The mean total phosphorus concentration
was doubled (0.42 mg/1); and the mean inorganic nitrogen concen-
tration was more than tripled (1.36 mg/1). Total annual precipitation
was 15 inches greater for the basin during 1959-60.
Inorganic nitrogen compounds are present in small amounts in
rainwater where they are predominantly nitric acid and ammonia.
These compounds come from the atmosphere and are the products
of electrical discharges, terrestrial decomposition, and volcanic erup-
tions. If the concentrations quoted by Hutchinson(9) are used and a
30-inch annual precipitation is assumed, the contribution of ammonia
and nitrate nitrogen in the temperate region would be 5.5 pounds
per acre. In an 18-month investigation at Hamilton, Ontario, Mathe-
son(10) determined the annual atmospheric nitrogen fall to be 5.8
pounds per acre. Sixty-one percent of the total nitrogen fell on 25
percent of the days when precipitation occurred, and the balance was
attributed solely to the sedimentation of dust.
As nitrogen enters the reservoir it is incorporated in the biomass
in the form of protein. Upon death or excretion, nitrogen is liberated
for re-use. During this process some is lost (1) in the effluents, (2) by
diffusion of volatile nitrogen compounds from surface water, (3) by
denitrification in the lake, and (4) in the formation of permanent
sediments.
Likewise, phosphorus is taken up in the web of life, is liberated
for re-use upon death of the organism(ll), may settle into the hypo-
limnion with the sedimentation of seston or in fecal pellets, and some
may be released at the mud-water interface(22).
Sawyer et al.(7) found the nitrogen content of bottom muds in
the Madison lakes to be 7,000 to 9,000 micrograms per gram (dry
weight) and phosphorus, likewise, 1,070 to 1,200 micrograms per gram
EFFECTS OF NUTRIENTS ON PHOTOSYNTHETIC
206 OXYGEN PRODUCTION
-------
(dry weight). Some of this is recirculated from the bottom ooze into
the upper lake water by the movement of organisms, eddy diffusion,
and the thorough mixing of the lake waters at the time of lake
overturns.
NUTRIENT UTILIZATION
The principal factors affecting aquatic growths include tempera-
ture; sunlight; size, shape, and slope of lake basin; water quality,
including water clarity and dissolved oxygen; and nutrients. The
total supply of an available nutrient depends on the total volume of
water, as well as on the concentrations of the element in the water.
Gerloff and Skoog(13) determined in the laboratory that 5 units of
nitrogen plus 0.08 unit of phosphorus —• a ratio of 60:1 — will pro-
duce 100 units of algae. The N-P ratio, as it occurs in algae and sub-
merged plants, is more nearly 10:1, and in sewage 8:1(14). Allen(15)
found the maximum algal crop that can be grown on the nutrients
present in domestic sewage to be 1 to 3 grams per liter (dry weight);
to obtain any appreciable increase, it was necessary to supplement
the sewage with nitrogen as well as carbon.
Sawyer(16) studied the southeastern Wisconsin lakes and con-
cluded that 0.30 milligram of inorganic nitrogen per liter of water and
0.015 milligram of soluble phosphorus per liter at the start of the
active growing season could produce nuisance, blooms. Nitrogen ap-
pears to be the more critical factor limiting algal production in nat-
ural waters(13), since phosphorus is stored in plankton as excess
and may approach 10 times the actual need.
A continued high rate of nutrient supply does not appear to be
necessary for continued algal production. After an initial stimulus,
the recycling of nutrients within the lake basin is sufficient to promote
algal blooms for at least a number of years. The initial stimulus is
most often supplied by dissolved phosphorus.
Studies of the standing crop of submerged aquatic plants in oligo-
trophy Green Lake and eutrophic Lake Mendota, Wisconsin(17,18)
indicated a wet weight of 14,000 pounds per acre and a dry weight of
1,800 pounds per acre. Harper and Daniel(lfl) found that submerged
weeds were 12 percent dry matter and contained an average of 1.8
percent total nitrogen and 0.18 percent phosphorus. Birge and Juday
(20) found the largest crop of spring plankton was approximately
360 pounds per acre on a dry weight basis (10 percent dry matter)
and the largest crop of autumn plankton, 324 pounds per acre. The
summer and winter minimums were 124 and 98 pounds per acre
respectively. Blue-green algae are approximately 6.8 percent nitrogen
and 0.69 percent phosphorus on a dry weight basis(21). These latter
values are for growths occurring in a fertile lake that does not rou-
tinely produce nuisance blooms.
Based on the above computations, a standing crop of submerged
aquatic plants in a fertile lake could contain, and liberate on decom-
position, 32 pounds of nitrogen per acre and approximately one-tenth
Mackenthun
207
-------
as much phosphorus. Likewise, under similar circumstances an algal
population could theoretically tie-up about 15 pounds of nitrogen
and 1.5 pounds of phosphorus per acre, or approximately one-half
of the nutrient components of a normal submerged weed population.
An algal population often covers the entire lake; submerged plants
are normally limited to that area where the depth is 20 feet or less.
PRODUCTION IN ABUNDANCE
The literature records many lakes capable of excessive algal pro-
duction and as Hasler(22) states, "it is clear that any increase in
the rate of eutrophy, even if this involves only the acceleration of a
natural and inevitable process is, from a human point of view, thor-
oughly undesirable."
Anderson(23) discusses recent eutrophication of Lake Washing-
ton near Seattle. In 1950, the standing crop of phytoplankton was
0.6 parts per million (by volume); in 1955, 1.6; and in 1956, 4.2.
The phosphate accumulation in the hypolimnion was 23 ppb in 1950,
89 ppb in 1957, and 74 ppb in 1958. These factors were correlated
with changes in the flora, especially the initial observation in 1955 of
Oscillatoria rubescens, "a notorious indicator of pollution in many
lakes." Hasler(22) described eutrophication in the ZUrichsee and
Hallwilersee, Switzerland, in Lake Windermere, England, and in sev-
eral other lakes. Phinney and Peek(24) discussed Klamath Lake,
Oregon; and Benoit and Curry(25), Lake Zoar, Connecticut; Deevey
and Bishop (26) found Linsley Pond to be the most biologically pro-
ductive of 30 lakes studied in Connecticut. Here, again, evidence was
found of rapid eutrophy in comparatively recent times.
Thirteen of the southeastern Wisconsin lakes studied by Sawyer
et al.(7) were re-examined by this writer in 1959-60. Twelve to
sixteen more inches of precipitation occurred during the latter study.
With the exception of Lake Koshkonong (discussed earlier) and Lake
Delevan, the chief difference noted was a significantly higher organic
nitrogen concentration in six of the remaining eleven lakes. Total
phosphorus, soluble phosphorus, organic nitrogen and inorganic nitro-
gen determinations of the lake effluents were made by the same
chemist during both studies.
Lake Delevan, like Lake Koshkonong, is a flow-through lake with
an area of 1,038 acres and a maximum depth of 52 feet. The resident
population of its shores exceeds 1000 dwellings, and it receives addi-
tional treated sewage effluent from Elkhorn, Wisconsin (1960 pop.,
3,586). When the 1959-60 monthly effluent-nutrient concentrations
were compared with those found in the 1943-44 study, a 2-fold in-
crease was noted in total phosphorus and organic nitrogen and a 2.7-
fold increase in inorganic nitrogen. During the summer of 1948, this
writer noted an abundance of higher aquatic plants such as pond-
weeds, coontail, wild celery, and water milfoil. The secchi disc
visibility was 9 feet. Wisconsin Conservation Department records for
August 16, 195l}
again showed rooted aquatics abundant with a
number of species present, and a secchi disc reading of 7.5 feet.
EFFECTS OF NUTRIENTS ON PHOTOSYNTHETIC
208 OXYGEN PRODUCTION
-------
Another record for August 11, I960, indicated the Tooted vegetation
scarce, the west shore devoid of plants, a dense algal growth, and a
secchi disc visibility of 2.8 feet. In recent years, an accelerated algal
control program, in which copper sulphate is used, has been estab-
lished.
The parameters of eutrophication are many; most important to
the layman on the scene are those that can be readily noted through
visual inspection. The secchi disc is a more polished measure of
visual inspection. The shape of vertical dissolved oxygen curves, the
build up of nutrients in the hypolimnion, significant changes in the
algal population and in the fishery are all factors closely correlated
with enriched conditions in a lake basin.
PHOTOSYNTHETIC OXYGEN PRODUCTION
As early as 1931, Rudolfs and Buekelekian(27) noted the effects
of sunlight and green organisms on the reaeration of streams and
found that the dissolved oxygen in water containing large quantities
of algae could be decreased from supersaturation to 17 percent satu-
ration by placing the water in darkness, and could also be increased
to 282 percent saturation by subjecting it to diffuse light.
In recent years, the measurement of primary production ha.s
stimulated interest among investigators(28). Several methods of de-
termination are applicable in a situation in -which the inflows of
energy and material balance the outflows. These include the oxygen
method (light-dark bottle, diurnal oxygen curve, and oxygen deficit
in the tvypolimnion); the carbon dioxide method; determinations with
radioactive materials; and the chlorophyll method(29).
According to Dice(30), "the ultimate limit of productivity of a
given ecosystem is governed by the total effective solar energy falling
annually on the area, by the efficiency with which the plants in the
ecosystem are able to transform this energy into organic compounds,
and by those physical factors of the environment which affect the rate
of photosynthesis."
Oswald and Gotaas( 31) state that green algae utilizing energy
from the sun produce carbohydrates from carbon dioxide and water,
and then assimilate these carbohydrates together with the liberated
ammonia and other essentials to produce additional algal cells. Quan-
titatively the growth of 1 pound of algae is usually accompanied
by the production of a minimum of 1.6 pounds ol molecular oxygen.
In stabilization ponds it has been found that a photosynthetic efficiency
of 1 percent is equivalent to the production of about 25 pounds of
organic matter or the liberation of about 40 pounds of oxygen per
acre per day. Photosynthetic efficiencies in pilot-plant sewage stabil-
ization ponds have ranged from 2 to 9 percent under varying condi-
tions of depth, detention time, recirculation time, and mixing.
The literature records many ways of expressing oxygen produc-
tion, e*g., micromoles of Oa per microliter of organisms per hour,
millimoles per square meter per day, milligrams of carbon per square
Mackenthun
-------
meter per day, milligrams of 02 per square centimeter per month,
grams per square meter per day, pounds per acre per day, etc. This
writer has taken the liberty of converting these data to pounds of
oxygen per acre per unit of time and to parts per million, to permit
rapid comparison.
Light-dark bottle data on sewage stabilization ponds in the
Dakotas indicated gross oxygen production of 231 pounds per acre
per day, respiration of 169 pounds per acre per day, or net oxygen
production of 62 pounds per acre per day(32). Gross production was
highest during mid-morning at Lemmon, South Dakota, (18.7 lb
per acre per hr) and lowest during early evening (0.3 lb per acre
per hr). Highest net oxygen production was 10.7 pounds per acre
per hour. There is generally no measurable oxygen production during
winter under ice in stabilization ponds.
In measuring in situ aeration of Wisconsin's sewage stabilization
ponds, McNabb(33) found net oxygen production proceeding at a
rapid rate in the morning, and oxygen consumption by biota exceed-
ing production throughout most of the afternoon in spite of light in-
tensities favorable for photosynthesis. Highest oxygen production was
21.4 pounds per acre per hour with a phytoplankton population of
178 parts per million (by volume) and a nutrient inflow of ap-
proximately 4.0 pounds per acre per day total nitrogen and 1.3 pounds
per acre per day total phosphorus.
Verduin(34) summarized the literature on primary production in
lakes and concluded that the net photosynthetic rate of autotrophic
organisms under optimum light was 35 x 10~6 pounds of Oa per milli-
liter of organisms per hour. Lakes with an epilimnion layer of the
order of 1 meter are likely to have standing crops of autotrophic
plants on the order of 50 parts per million (by volume), and lakes
with an epilimnion depth of 10 meters are likely to have standing
crops of about 5 parts per million (by volume).
Computations of photosynthetic oxygen production for several
lakes yielded values lying mostly between 42 to 57 pounds per acre
per day. A year-round study under completely natural conditions in
western Lake Erie showed winter yields of about 11 pounds per acre
per day and summer maxima of about 85 pounds per acre per day.
The annual oxygen curve closely followed the solar radiation curve.
Net oxygen photosynthesis in two Alaskan lakes ranged from 3.4 to
4.0 pounds per acre per day(35). The net oxygen production rate for
East Okoboji Lake in Iowa, a producer of high plankton population?
was found to be 79 pounds per acre per day, with production largel;
confined to the first 2 meters(55). Average gross photosynthesis foi
Grand Traverse Bay (Michigan) has been recorded at 10.8 pounds
per acre per day and for Douglas Lake, 3.7 pounds per acre per
day(37).
An increase in the hypolimnetic oxygen deficit has been taken as
evidence of increased lake productivity. In Lake Washington, near
Seattle, the hypolimnetic oxygen deficit was determined to be 105
pounds per acre per month in 1933, 178 pounds in 1950, and 279
EFFECTS OF NUTRIENTS ON PHOTOSYNTHETIC
210 OXYGEN PRODUCTION
-------
pounds in 1955(38). The standing crop of phytoplankton in the top 20
meters of water in 1950 was 0.6 parts per million (by volume) and
in 1955, 1.6 parts per million (by volume) (23).
As pointed out by Whipple et al.(39), supersaturation in the
upper waters is not cumulative to a great extent because of the
circulation that is maintained by wind action and convection currents,
both of which promote contact of the water and the air with a con-
sequent loss of oxygen. Highest saturation is frequently found in
the upper region of the thermocline in oligotrophic lakes. Wind action
seldom disturbs the water of this zone, convection currents are absent,
and diffusion is a slow process. Chlorophyll-bearing organisms find
an abundant supply of carbon dioxide and sufficient light in this area
to promote photosynthesis, resulting in saturation values that may
exceed 300 percent.
There is a need to relate photosynthetic oxygen production to
(1) practical problems associated with lakes and streams, (2) some
quantitative expression of phytoplankton, (3) nutrient balance, and
(4) seasonal fluctuations and climatic phenomena.
THE PRICE OF EUTBOPHY
The disadvantages of algae as a source of oxygen have been well
summarized by Bartsch(40). Algae respond to complex, changing,
unpredictable environmental factors including solar radiation, opacity
of the medium, rate of bacterial activity, rise and fall of nutrients,
climatic phenomena, and ecological succession. When algal cells die
and sink to the hypolimnion, oxygen is used in decomposition. The
stimulation of algal production by nutrients can lead to the formation
of a mass of organic matter greater than that of the original waste
course(42). In an enriched environment, algae respond so well to
incoming nutrients that the oxygen required for the respiration of
the resultant algal mass alone surpasses the BOD of the incoming
food material. Lake Winnebago, Wisconsin, (area, 213 mi2) produces
heavy algal populations. In July, when the lower Fox River carried
a heavy algal load from Lake Winnebago, the ultimate BOD in the
river above the sources of industrial and municipal wastes ranged to
660,000 pounds of oxygen demand each day(42).
Enrichment often results in domination of the algal mass by a
relatively small group of blue-green algae, which become well estab-
lished. Indications are that many species of fresh water algae are
capable of producing physiologically active metabolites that may
function as toxins, growth inhibitors, or growth stimulators to them-
selves or to associated algae(43). Most of the adverse effects resulting
from an algal mass occur when one species of algae dominates the
population.
Fish kills have resulted from a supersaturation of oxygen (44),
as well as a depletion of oxygen (45). An example of the latter oc-
curred in October 1946 when tremendous quantities of Aphanizomen
ftos aquae entered the Yahara River from Lake Kegonsa near Mad-
Mackenthun
211
-------
ison, Wisconsin, decomposed in passing downstream, and caused
oxygen depletion resulting in the death of tons of fish.
Provost(46) indicated that overproduction of chironomids (midge
larvae or bloodworms) in lakes is caused by excessively nutritious
waters. Midgeflies have become a nuisance in several areas where
conditions are especially suitable for the concentration of a swarmine
mass of adults following an emergence (e.g., Clear Lake, California-
Lake Winnebago, Wisconsin; and several lakes in Florida). Larval
development is no doubt fostered by the deposition of the dead bodies
of a rich plankton population on the bottom sediments.
Both weed and algal nuisances develop in enriched water. Fish-
ing may be impaired; and bathing, boating and water skiing often
become indefinitely postponed sports in waters that otherwise offer
maximum multiple recreational use. Industrial or municipal water
treatment is hampered or made inefficient by extensive aquatic
growths; property values are lowered; and resort trade is often can-
celled as a result of these nuisances.
Rapid decomposition of dense algal scums with associated organ-
isms and debris gives rise to odors and hydrogen sulfide gas, which
creates strong citizen disapproval and often stains the white paint on
residences adjacent to the shore.
Controls are needed; these are feasible(47), but frequently are
costly, time consuming, and offer only temporary relief. Long-term
remedial measures can be focused on reducing the nutrient concen-
tration in the troublesome area or on altering some aspect of the
physical topography that concentrates or fosters the development of
nuisance algae or aquatic weeds. These considerations often involve
costly measures to correct existing conditions, and future planning
to assure wise use of the natural aquatic resources of an area.
REFERENCES
1 Rudolfs, W. Phosphates in sewage and sludge treatment. I. Quan-
tities of phosphates. Sewage Works J. 19:43-47. 1947.
2 Mackenthun, K. M., and C. D. McNabb. Stabilization pond studies
* in Wisconsin. JWPCF. 33:1234-51. 1961.
o Mackenthun, K. M., L. A. Leuschow, and C. D. McNabb. A study
of the effects of diverting the effluent from sewage treatment
uDon the receiving stream. Trans. Wisconsin Acad. Sci. 49:51-
72.
4 Sanderson, W. W. Studies of the character and treatment of
wastes from duck farms. In: Proc. 8th Ind. Waste Conf., La-
fayette, Ind., May 4-6, 1953. Eng. Ext. Ser. No, 83. Eng. Bull.,
Purdue Univ. 38(l):170-76. Jan. 1954.
5 Paloumpis, A. A., and W. C. Starrett. An ecological study of
benthic organisms in three Illinois River flood plain lakes. Am.
Midland Naturalist. 64:406-35. 1960.
effects of nutrients on photosynthetic
212 OXYGEN PRODUCTION
-------
6. Eck, P., M. L. Jackson, O. E, Hayes, and C. E, Bay. Runoff analy-
sis as a measure of erosion losses and potential discharge of
minerals and organic matter into lakes and streams. Summary
report, Lakes investigations, Univ. Wis., Madison, Wis., 1957.
Mimeo.
7. Sawyer, C. N., J. B. Lackey, and R. T. Lenz. An investigation of
the odor nuisance occurring in the Madison Lakes, particularly
Monona, Waubesa and Kegonsa from July 1942-July 1944. Report
of Governor's Committee, Madison, Wis., 2 vols. 1945.
8. Lackey, J. B., and C. N. Sawyer. Plankton productivity of certain
southeastern Wisconsin lakes as related to fertilization. Sewage
Works J. 17:573-85. 1945.
9. Hutchinson, G. E. A treatise on limnology. John Wiley and Sons,
Inc., New York, N.Y., 1957. 1015 pp.
10. Matheson, D. H. Inorganic nitrogen in precipitation and atmos-
pheric sediments. Can. J. Technol. 29:406-12. 1951.
11. Cooper, L. H. N. The rate of liberation of phosphates in sea water
by break-down of plankton organisms. J. Marine Biol. Assoc.
United Kingdom. 20:197-220. 1941.
12. Hooper, F. F., and A. M. Elliott. Release of inorganic phosphorus
from extracts of lake mud by Protozoa. Trans. Am. Microscop.
Soc. 72:276-81. 1953.
13. Gerloff, G., and F. Skoog. Nitrogen as a limiting factor for the
growth of Microcystis aeruginosa in southern Wisconsin lakes.
Ecology. 38:556-61. 1957.
14. Sawyer, C. N. Some new aspects of phosphates in relation to lake
fertilization. Sewage and Ind. Wastes. 24:768-76. 1952.
15. Allen, M. B. General features of algae growth in sewage oxida-
tion ponds. California State Water Pollution Control Bd. Publ.
13. 1956. pp. 11-34.
16. Sawyer, C. N. Fertilization of lakes by agricultural and urban
drainage. J. New Engl. Water Works Assoc. 61:109-27. 1947.
17. Rickett, H. W. A quantitative study of the larger aquatic plants
of Lake Mendota. Trans. Wisconsin Acad. Sci. 20:501-27. 1922.
18. Rickett, H. W. A quantitative study of the larger aquatic plants
of Green Lake, Wisconsin. Trans. Wisconsin Acad. Sci. 21:381-
414. 1924.
19. Harper, H. J., and H. R. Daniel. Chemical composition of certain
aquatic plants. Botan. Gaz. 96:186-89. 1939.
20. Birge, E. A., and C. Juday. The inland lakes of Wisconsin. The
plankton. I. Its quantity and chemical composition. Wis. Geol.
Nat. Hist. Sur. Bull. 64, Science Series 1, No. 13. 1922. 222 pp.
Mackenthun
213
-------
21. Gerloff, G., and F. Skoog. Cell content of nitrogen and phos-
phorus as a measure of their availability for growth of Micro-
cystis aeruginosa, Ecology. 35:348-53. 1954.
22 Hasler A. D. Eutrophication of lakes by domestic drainage.
Ecology. 28:383-95. 1947.
23. Anderson, G. C. Recent changes in the trophic nature of Lake
Washington — A review. In: Algae and Metropolitan Wastes.
Trans of Seminar, Cincinnati, Ohio, Apr. 27-29, 1960. Tech. Rept.
W61-3. SEC. 1961. pp. 27-33.
24. Phinney, H. K., and C. A. Peek. Klamath Lake, an instance of
natural enrichment. In: Algae and Metropolitan Wastes. Trans,
of Seminar, Cincinnati, Ohio, Apr. 27-29, 1960. Tech. Rept.
W61-3. SEC. 1961. pp. 22-27.
25. Benoit, R- and J. C. Curry. Algae blooms in Lake Zoar, Con-
necticut. In: Algae and Metropolitan Wastes. Trans, of Seminar,
Cincinnati, Ohio, Apr. 27-29, 1960. Tech. Rept. W61-3. SEC.
1961. PP- 18-22.
26. Deevey, E. S., and J. S. Bishop. Limnology. Sec. II. A fishery
survey of important Connecticut lakes. State Bd. Fish & Game,
Lake and Pond Survey Unit Bull. No. 63. State of Conn. Publ.
Doc. 47:69-121. 1942.
27. Rudolfs, W-, and H. Huekelekian. Effect of sunlight and green
organisms on re-aeration of streams. Ind. Eng. Chem. 23:75-78.
1931.
28. Ryther, J- H. The measurement of primary production. Limnol.
andOceanog. 1:72-84. 1956.
29. Odum, E. P Fundamentals of ecology. W. B. Saunders Co.,
Philadelphia, Pa., 1959. 546 pp.
30. Dice, L. R- Natural communities. Univ. Mich. Press, Ann Arbor,
Mich., 1952. 547 pp.
31. Oswald, W. J., and H. B. Gotaas. Discussion — photosynthesis in
the algae. Ind. Eng. Chem. 48:1457-58. 1956.
32. Bartsch, A. F. Algae in relation to oxidation processes in natural
waters. Pymatuning Spec. Publ. No. 2, Ecology of Algae. Pyma-
tuning Lab. of Field Biology, Univ. Pittsburgh. Apr. 1960. pp.
56-71.
33. McNabb, C. D. A study of the phytoplankton and photosynthesis
in sewage oxidation ponds in Wisconsin. Ph.D. thesis, Univ. Wis-
consin. I960.
34. Verduin, J- Primary production in lakes. Limnol. and Oceanog.
1:85-91. 1956.
35. Goldman, C. R. Primary productivity and limiting factors in
three lakes of the Alaska Peninsula. Ecol. Monographs. 30:207-
30. 1960.
EFFECTS of nutrients on PHOTOSYNTHETIC
214 OXYGEN PRODUCTION
-------
36. Weber, C. I. Some measurements of primary production in East
and West Okoboji Lakes, Dickinson County, Iowa. Proc. Iowa
Acad. Sci. 65:166-73. 1958.
37. Saunders, G. W., F. B. Trama, and R. W. Bachmann. Evaluation
of a modified C14 technique for shipboard estimation of photosyn-
thesis in large lakes. Great Lakes Research Division Publ. No. 8.
Michigan Univ., Ann Arbor, Mich. 1962. 61 pp.
38. Edmondson, W. T., and G. C. Anderson. Artificial eutrophication
of Lake Washington. Limnol. and Oceanog. 1:47-53. 1956.
39. Whipple, G. C., G. M. Fair, and M. C. Whipple. The microscopy
of drinking water. John Wiley & Sons, Inc., New York, N. Y.,
1948. 586 pp.
40. Bartsch, A. F. Algae as a source of oxygen in waste treatment.
Sewage and Ind. Wastes. 33:239-49. 1961.
41. Renn, C. E. Allowable loading of Potomac River in vicinity of
Washington, D. C. A Report on Water Pollution in the Washing-
ton Metropolitan Area. Sec. Ill — Appendixes Feb. 1954. AB-1
— AB-17.
42. Scott, R. H., B. F. Lueck, T. F. Wisniewski, and A. J. Wiley.
Evaluation of stream loading and purification capacity. Commit-
tee on Water Pollution, Madison, Wis. Bull. No. 101. 1956. Mimeo.
43. Hartman, R. T. Algae and metabolites of natural waters. Pyma-
tuning Spec. Publ. No. 2, Ecology of Algae, Pymatuning Lab. of
Field Biology, Univ. Pittsburgh. Apr. 1960. pp. 38-55.
44. Woodbury, L. A. A sudden mortality of fishes accompanying a
supersaturation of oxygen in Lake Waubesa, Wisconsin. Trans.
Am. Fisheries Soc. 71:112-17. 1941.
45. Mackenthun, K. M., E. F. Herman, and A. F. Bartsch. A heavy
mortality of fishes resulting from the decomposition of algae in
the Yahara River, Wisconsin. Trans. Am. Fisheries Soc. 75:175-
80. 1948.
46. Provost, M. W. Chironomids and lake nutrients in Florida. Sew-
age and Ind. Wastes. 30:1417-19. 1958.
47. Mackenthun, K. M. The chemical control of aquatic nuisances.
Wisconsin Committee on Water Pollution, Madison, Wis., 1958.
64 pp.
DISCUSSION
Dr. Clair N. Sawyer
Senior Associate, Director of Research
Met calf and Eddy Engineers, Boston, Massachusetts
After the comprehensive discussion Mr. Mackenthun just pre-
sented, there is not too much left to say except perhaps to elaborate
Sawyer
215
-------
around the fringes, emphasize some things, and note one or two places
where I might disagree.
First, we are concerned, in the long run, with what is commonly
called ecological control in the management of problems related to
the superabundance of the algal forms we encounter in our aquatic
areas. Let us look once again at the sources from which the nutrients
come.
The air is one source we have generally ignored in the past. We
can do very little about controlling contributions from the air. We
cannot control the lightning, for example, or what comes down from
the upper atmosphere, although sometimes we would like to. The
people in Minnesota heard about the problem of water fowl con-
tamination of lakes in Illinois, so they have been doing something
about it by spilling soybean oil into their rivers with a tremendous
kill of ducks in that area. I do not mean anything against the people
in Minnesota, but I just thought you might be interested in what
cooperation you sometimes get between States.
As to the land, the character of the soil is very important, of
course, and the nutrients that reach our receiving waters depend
both on the water that percolates through the soil and the surface
runoff. We cannot do much about the nutrients that percolate
through the soil because farmers are going to grow crops, fertilize
the land, and so forth. An area in which we find tremendous amounts
of nutrients in the receiving waters is in the development of peat
anas. Peat bogs have very poor retentive capacity for nutrient mate-
rials and furthermore, serve as tremendous nitrifying beds. I have
seen small creeks coming from such beds that contained as much
total nitrogen as an ordinary sewage treatment plant effluent. Not
much can be done about this source of nutrients as the farmer is
going to work these peat lands.
I have always been much concerned about the way farmers
spread manure on the land. When they spread manure on frozen
7a?S lt rai^S- What haPPens to the soluble fertilizing sub-
stances. They wash into the receiving stream; if the manure is spread
™°T fl?nd and iX rains> the ammonia and phosphates are
leached into the soil and held there by ion exchange. What can be
enc°"rage county agents to educate the farmers to im-
prove their practices in handling animal manures. We are all familiar
with contour farming so I will not dwell on that.
wat s°urce of nutrients in this modern age is our waste
waters, particularly those of domestic origin. Mr. Mackenthun has
fJIfnifrT m*ormation on per capita contributions of phosphorus
fi" 4 7*° Warn you about back into the literature
flnd data on Phosphorus, because the picture today is en-
dXSLnJ f]ren* *r?"V what ^ was even 5 years ago and is certainly
as ~ years ag0" We used to think of domestic sewage
as having a nitrogen to phosphorus ratio of 8 to 1. As a result of the
use of synthetic detergents that are loaded ~
216
DISCUSSION
opo eai-74o—a
-------
phates (they are often missed because we do not analyze for them),
the nitrogen to phosphorus ratio in domestic sewage today is closer
to 3 to 1. A total phosphorus concentration of 8 to 10 milligrams per
liter is very common in domestic sewage today whereas Pr*or to W°r
War II it was on the order of 3 milligrams per liter. Tremendous
quantities of this fertilizing element have been brought into the pic-
ture as a result of the development and widespread use of household
detergents. This is an important consideration.
Many of our industrial wastes are very potent sources of nutri-
ents — phosphorus, in particular. Great amounts of synthetic deter-
gents are used in place of old-fashioned soap. Textile wastes are
ordinarily rich in phosphorus; fortunately, they are deficient m nitro-
gen. Animal packing wastes are extremely rich in both nitrogen and
Phosphorus. Both of these are major sources of nutrient materials.
Mr. Mackenthun talked about the secondary effect of the cycling
of nutrients between the water and the bottom deposits, Once you
get the wheel turning, so to speak, algal blooms develop, the algae
die, settle, and enrich the bottom, bacteria decompose the algae thus
releasing the nutrients, and then there is a feedback from the bottom.
This feedback from the bottom, or the hypolimnion, is extremely
important; more important than is often realized. When a lake strati-
fies, stratification usually occurs at a depth somewhere around 20
feet, and whether the hypolimnion goes anaerobic or stays aerobic
makes quite a difference in the feedback of nutrients.
Let us consider a lake that goes anaerobic. Tremendous amounts
of ammonia nitrogen, phosphorus, and other nutrients come from the
bottom that would normally be held there fairly tightly if the lake
stayed aerobic from top to bottom. This feedback, then, is a source of
nutrients throughout the summer months. The surface waters can
be fairly well stripped of nutrient materials by early season blooms,
but as wind on the lake surface sets up currents in the epilimnion, a
reverse current is established in the hypolimnion. Transmission of
energy through the thermocline sets up this lower circulation. Chem-
ical data on hypolimnetic waters demonstrate that dispersion of
nutrients from bottom muds is not dependent upon ordinary diffusion,
but is due to circulation of the hypolimnion. Actually, analysis of the
Waters of the hypolimnion reveals that the nutrient content of the
hypolimnion is more or less constant throughout its entire depth
during the summer stagnation period.*
Mr. Churchill mentioned the lowering of the thermocline within
stratified reservoirs as a result of the drawoff of cold waters from
the hypolimnion. The thermocline will go down to some extent any-
how as the summer progresses, because of the mixing referred to
above. Thus the depth of water in the epilimnion becomes greater
and greater. As this takes place, of course, water keeps moving from
the hypolimnion into the epilimnion, and this serves as a continuing
supply of nutrient materials for the epilimnion.
*See Journal New England Water Works Association, 61, 122 (1947).
Sawyer
217
-------
In regard to nutrient utilization, analyses of algae and rooted
aquatics indicate that they take about 10 parts of nitrogen out of
the water for every part of phosphorus. Although the work of Gerloff
and Skoog that Mr. Mackenthun referred to indicates algal growth
with a ratio of 60 to 1, this is very definitely a starvation diet; algae
will grow, but the blooms will not be normal. This ratio of 10 to I
is quite important. In the old days when domestic sewage had a ratio
of nitrogen to phosphorus of 8 to 1, algae would grow and there was
not much of an excess of phosphorus left in the water. Today, with
a ratio of 3 parts of nitrogen to approximately 1 of phosphorus in
wastes, and a ratio of approximately 10 to 1 in synthesized plant
material, a great abundance of residual phosphorus is in the water.
In the Madison Lakes situation referred to, we often found the
surface waters essentially stripped of nitrogen in the middle of the
growing season, but 0.3 to 0.5 milligram per liter of phosphorus re-
mained in the surface water afteT the nitrogen had been essentially
all removed. I suspect if the municipal wastes were still reaching
the lakes in the same way today they would show much higher levels
of phosphorus.
How about lakes that are heavily fertilized with phosphorus?
What does phosphorus mean to us. The important thing is that
nitrogen in surface waters is not necessary to produce extensive algal
blooms. The blue-green algae often come into the picture in late
summer about the time the nitrogen is pretty well stripped from the
surface waters. If there is sufficient phosphorus (and other necessary
nutrients), extreme blooms of blue-green algae may occur. Therefore,
we must consider both nitrogen and phosphorus in the nutrient
problem.
I have searched and searched for a good definition of eutrophica-
tion. Mr. Mackenthun indicates it is based upon several parameters.
This makes me feel better because here is a biologist who says more
than one criterion must be used. The criteria I use are:
1. How far does light penetrate into the water? As produc-
tivity increases, algal blooms increase, so obviously the
light will not penetrate so far. Shallow light penetration
is therefore a major indicator.
2. In lakes that stratify, does the hypolimnion go anaerobic?
If it goes anaerobic, I am sure it is a eutrophic lake. This
is probably as important as any criteria and is the one
I would use.
These are, in general, the major criteria.
The disappearance of rooted aquatics that Mr. Mackenthun re-
ferred to is a natural occurrence. We know we can control these
growths ii we can block out the light, but we generally do not build
shelters over our reservoirs, The algae may do this, however, as far
as the Tooted aquatics are concerned. When a dense bloom of phyto-
218
DISCUSSION
-------
plankton develops, they create a natural shade for the rooted aquatics,
which, of course, do not survive.
Mr. Mackenthun discussed quite extensively the problems re-
lated to eutrophication in lakes; he also mentioned the problem that
develops in the receiving stream when these accumulations of phyto-
plankton get loose and float down the receiving stream. If these algae
would just keep their heads under water so they would not be con-
centrated by gentle breezes in coves or in the outlet river, they could
be quite beneficial in many ways, But, they do accumulate in coves
and they do go out the outlet and there they, of course, decompose
and cause the algal nuisances to which Mackenthun referred.
There is one other aspect of this problem of the effluents from
reservoirs or lakes, which is related to the nutrients themselves. In
the Madison Lakes chain, the waters leaving Lake Waubesa were
extremely nutritious. Rooted aquatics came into the picture in the
relatively shallow river below there to such an extent that they
restricted the flow of water down the river and raised the water level
in the lake. The cottagers, of course, rebelled at that.
Another case is Spring Creek in Pennsylvania, a rich fishing
stream. This is below a sewage treatment plant rather than an im-
poundment, but it illustrates the potential problem. The nutrients
getting into this shallow stream have created a very extensive rooted-
aquatic growth in the stream. These growths have two effects. In
the daytime, of course, there is a diurnal effect that makes the DO go
sky high. The water with its high oxygen concentration does not
stay there, however; it moves on down the river, and when night
comes fresh water that does not have the photosynthetically pro-
duced oxygen moves into the area where the rooted aquatics are, and
the DO in the water flowing through the area is depressed. As you
all know, it takes a little while for streams to recover, so that water
with a low DO now goes downstream as a slug and produces some
very serious fish kills.
There are many aspects of this problem to be considered. Before
closing, I want to raise this question: How do we harness eutrophica-
tion to do us some good? My experience has been that we create
these impoundments, everybody wants to use them as recreational
areas and they do not want wastes discharged into them. Mr. Mack-
enthun has indicated we can expect to have 45 to 57 pounds of oxygen
per acre per day that ought to be harnessed. We are living in a very
affluent — I sometimes say, effluent — society and it is time we begin
to think in terms of reserving some of these areas for industry, par-
ticularly the highly fertilized areas such as we are talking about here.
We know industrial areas are going to develop and that they will
probably become a nuisance to us; so why not plan to put them to
work? We could encourage industries rather than cottagers with
their speed boats to locate around the reservoirs that have high
nutrient levels and thus assimilate considerable nutrient while reduc-
ing the potential adverse effects in other areas.
Sawyer
219
-------
DISCUSSION FROM THE FLOOR
Mr. LeBosquet, U. S. Public Health Service: I wonder, Dr. Saw-
yer, if you could elaborate a little on corrective measures. I know that
keeping nutrients out is a problem, but, when this has been accom-
plished, is it successful? In reclaiming some of these eutrophied
lakes, how successful have you been in keeping the nutrients out, and
how fast have the lakes come back?
Dr. Sawyer, Metcalj and Eddy: The classical example is Lake
Waubesa at Madison, Wisconsin. I think Mr. Mackenthum is in a
better position to discuss this than I am because he had it under sur-
veillance after the diversion was put into operation.
Mr. Mackenthun, U. S. Public Health Service: As I recall, the
Madison effluent was first diverted around the lower lakes December
1, 1958. Tests on nutrient content of the effluents of the lower lakes
showed no great reduction during the first and second summers fol-
lowing diversion. The first winter following diversion there was a
slight reduction in nitrogen, but no reduction in phosphorus. The
second winter, however, there appeared to be a significant reduction
in both nitrogen and phosphorus in the effluents as determined by
weekly samples. Algal determinations were not made on these waters
following diversion. Approximations were made, however, from the
suspended solids and organic nitrogen and, as I recall, the algal popu-
lation has been as high since diversion as before. There is some
indication that the concentrations of nutrients are now going down.
Mr. Hester, U. S. Fish and Wildlife Service: Dr. Sawyer, you
referred to the feedback, or recycling, of nutrients from the lower
layer of the reservoir into the upper layer. I wonder if you would
comment on whether bottom-level discharge or higher-level discharge
would be more effective in increasing the concentration of these
nutrients in a reservoir?
Dr. Sawyer: I would think that under summer conditions it
would be better to discharge the surface waters rather than the lower
waters because the lower waters ordinarily are quite a bit richer in
nutrient materials. That has been my experience, anyhow.
220
DISCUSSION
-------
HYDROLOGIC FACTS NEEDED FOR STUDIES OF
FLOW REGULATION FOR STREAM
QUALITY CONTROL
Frank H. Rainwater
Physical Science Administrator, Technical1 Services Branch
Division of Water Supply and Pollution Control
U. S. Public Health Service, Washington, D. C.
This symposium has as one of its primary objectives a delineation
of technical problems encountered in complying with Section 2b of
Public Law 660, as amended, the segment of the Federal Water
Pollution Control Act of 1961 having to do with streamflow regulation
for purposes of water quality control. Discussion between agencies in-
volved with water resource development will lead to mutual under-
standing and, we trust, to better advance planning for water quality
management.
Some of the basic questions facing us today are:
How does the science and practice of hydrology enter these
quality control studies? What are the specific questions that must be
answered? Where are the principal deficiencies? Once we establish
these basic points, we can then proceed to discuss the scope and limita-
tions of flow regulation studies and ways for strengthening reports.
USES OF FLOW REGULATION FOR QUALITY CONTROL
Several relations between hydrologic and water quality param-
eters may be exploited by planned manipulation of streamflows.
Under the general heading of storage and regulation, the within-
reservoir relations must be considered as well as the downstream
relations.
Regulation of streamflow below a reservoir might be used to
control concentration by changing volume, to change time of passage
for decay-type wastes, to improve reaeration and self-purification by
increasing turbulence, or to prevent the alternating deposition and
fiush-out of taste- and odor-producing substances. These uses do not
exhaust the possibilities, but represent the principal ones.
The volume-concentration relation is expressed as:
Pollution load
Concentration
Volume of flow
This simple dilution mechanism is applicable to control of almost
all types of pollutants; however, it may not be the most effective or
most efficient means in all cases.
Increasing the time of passage between a source of pollution and
a downstream point of use may be more efficient than dilution for con-
trol of bacteria, radioactive wastes, and degradable organic wastes.
Rainwater 221
-------
The generalized behavior of die-away pollutants is demonstrated bv
the equation: *
Ct = C0e~Kt
where Ct is concentration at time t,
C0 is concentration at time o,
K is decay rate constant,
t is time.
Degradation of such pollutants through a stream reach might h«
enhanced by maintaining low fl0ws and velocities. This time of
passage relation is very important, too, within reservoirs.
^ „ When deoxygenating wastes (B0D) are involved, the oxygen
deficit in the stream depends in part on the rate of reaeration (K i
as described by Streeter and PhelpS(j).
~ - K,L - K„D
where D is dissolved oxygen deficit,
t is time,
L is residual BOD,
Kj is the deoxygenation constant,
Ko is the reaeration constant.
Turbulence and reaeration are related. To the degree that turbulence
is a function of stream discharge and velocity, flow regulation provide,
a way to increase reaeration. &
Fluvial sediment transport and the hydraulic variables affectin*
it can be important to taste and odor problems(2). Some taste- and
odor-producing substances are readily adsorbed on sediment Stream
velocity may be manipulated in such a way as to preclude slow
deposition and later batch flush-out of these sediments. Of course
sediment itself can be troublesome in a stream used for water supply.'
Within reservoirs, storage tends to smooth out the peaks in con
centration of inflow. Reservoirs also slow the time of travel and oro
vide opportunity for decomposition of degradable soluble wastes
Sedimentation may provlde relatively safe storage for some settleable
materials while they decay For example, uranium mill wastes de-
posited in Lake Mead probably will never be a threat below Hoover
Dam. The surface exposure to air and sunlight afforded by reservoirs
may also be conducive to water quality improvement.
the analytical problem
The hydrologic analysis for flow regulation studies is a four-sten
procedure that: ^
1. Describes the present annual streamflow and quality char-
acteristics in terms of typical companion hydrographs having
selected average recurrence intervals.
222
HYDROLOGIC FACTS NEEDED
-------
2. Projects the future streamflow and quality characteristics
that would exist under projected water use and management
in year(s) "X" without flow regulation.
3. Computes the regulated streamflow regimen needed in
year(s) "X" to meet the quality objectives.
4. Schedules the flow regulation needs, including probable
monthly distributions, over the life of the proposed structure.
A study of the needs of flow regulation is a forecast. It involves
reservoirs to be built, water to be used, wastes to be discharged, and
stream depletions to be experienced. The hydrologist has no choice
but to accept this ground rule — projections into the future. History
is essential to, but only a part of, projection. Statistical probability
can be a very useful forecasting tool if the available data are amen-
able to statistical analysis. The method, therefore, dictates the kind
of raw data needed.
A second ground rule is laid down in the law under which these
studies are made. The responsibility assigned by the Congress is to
control stream quality, not to maintain a streamflow. The two are
entirely different entities. The law reads, in part, "— consideration
shall be given to the inclusion of storage for regulation of streamflow
for the purpose of water quality control,—The term "regulation"
was chosen advisedly. Pollution burdens imposed by nature and man
on a stream are not constant. Hence, maintenance of a constant or
minimum streamflow is not the most efficient answer to quality
control.
Maximum regulation demands do not necessarily coincide in time
with either minimum streamflows or maximum concentrations of
pollutants. For example, the process of counteracting a BOD load
imposed on a stream involves the oxygen concentration of receiving
water, the reaction rate (deoxygenation), and reaeration factors. Each
of these is temperature sensitive. Hence, the streamflow requirement
is dependent, partially, on temperature. In mineral quality control
the tons of mineral to be diluted may be much less at minimum flow,
or maximum concentrations, than at some higher flow.
THE PRINCIPAL DEFICIENCIES
As is evident from the nature of the problem, its solution is ap-
proached through a chain of dependent sequential analytical treat-
ments of facts and figures. The old cliche that "a chain is no stronger
than its weakest link" is particularly appropriate here. Relative
weaknesses exist both in the raw hydrologic data and in knowledge
of how best to analyze and interpret these data. Only the four most
significant deficiencies are considered in this paper.
1. Description of low-flow and quality characteristics of streams.
2. Obsolescence of data.
3. Hydrochemical dynamics of reservoirs.
4. Statistical treatment of multivariable systems.
Rainwater
223
-------
These deficiencies are in no way Der„na, ^ ,
They are common to many other wat^r stream quality control,
cept of flow regulation for stream qualit^™*^6 ,u"ctions- The con-
a new set of problem areas, but ratW qLi . does not generate
solving the "standard" ones. s lmPetus to the need for
Description of low-flow and quality characteristics of stream.
Probably the most significant defiHpn™ ; 4.u
relating water quality and streamfiow Stream" _.^ealm of inter -
discharge are inseparable in these studies For ?, 3nd streain
event in the past and future time serieso 'II ' purP°ses, each
behavior has two dimensions — quantity t a comprislnfi stream
quality. The one objective of flow reffulatirm°fr»?°W concurrent
satisfactory concentration of Water-born* m + f? contr°l is a
involves manipulation of tonnages which 3/ h? f mec^anism
current flow and concentration. Knowledge of streamflo'°' C°n"
istics is of little use unless it can be a«nrfat0,q streamfiow character-
stream quality. The same is true of independent *t W*th attendant
« or independent stream quality facts
With respect to streamfiow alone ,•«# .
low-flow characteristics is weak in both a/°n availal,le on the
ity. What are the problems in this area' <;°verage and valid-
flow characteristics of streams is relatival ' 6- concern with low-
country where water has been pSSP'ln'SI ? •m
-------
Regulation, diversion, and increasing consumptive use are continually
changing the water quantity regimen of streams. The water quality
regimen responds to changes in flow regimen and also to man's use
of water to transport wastes. Data obsolescence is particularly pro-
nounced in consideration of low flows.
This problem of obsolescence has several consequences in flow
regulation studies. The hydrologist may be seriously misled by pub-
lished low-flow frequency data compiled, unadjusted, from the his-
toric record. Usually, such data require considerable reworking be-
fore they can be used with confidence. Then too, the analyst may be
forced to base his predictions of streamflow and quality on a rela-
tively short time series. Techniques are needed to improve the pre-
cision of estimates so based. Although not designed specifically
for this purpose, the mathematical synthesis of streamflow sequences
proposed by Thomas and Fiering(3) offers promise for both stream-
flow and stream quality investigations. Certainly, additional research
and ingenuity in this field are needed.
Stream depletion is another very important aspect of obsoles-
cence. To fulfill its responsibility in making recommendations to the
Federal construction agencies, the Public Health Service must base
its analysis of regulation needs on future depleted or modified stream-
flows. The flow to be augmented 20 or 50 years hence will not be
the same as the flow measured now, but rather a stream flow depleted
by the then existing water use and management. Hence, we need
accurate projections of depletions and other modifications of stream
flow. For some types of depletion projections, the Public Health
Service turns to other agencies who are the responsible authorities.
In terms of national averages, consumptive loss by municipal
use is about 10 percent of withdrawals; by industrial use (including
cooling), about 2 percent; and by irrigation use, about 60 percent.
In the current distribution of water by these major users 30 to 40
percent of the water withdrawn in the United States is lost from the
supply. No stretch of the imagination is required to see that future
depletions will make increasing inroads on the static available supply.
National averages may be misleading, however, because both water
supply and water users differ so widely from place to place.
Furthermore, streamflow may be lessened by factors other than
municipal, industrial, and irrigation uses. How about stock ponds,
phreatophytes, reservoir evaporation, and changes in land use? Some
of these factors are particularly influential in low-flow and drought
periods.
Concerted attention of several agencies to the over-all problem
of projecting stream depletions will be required to strengthen this
weak link in the chain of flow regulation studies.
Hydrochemical dynamics of reservoirs
If impounded water is to be used in the future to improve down-
stream quality, some method must be developed to predict more
Rainwater
225
-------
accurately the quality of releases from reservoirs yet unbuilt. Both
the importance of, and the deficiency in, this type of knowledge is
attested by the number of papers in this symposium that relate to
this facet. Consequently, I will not belabor the point other than to
say that several groups are attacking this problem either from a
theoretical or empirical angle.
The Division of Water Supply and Pollution Control is organiz-
ing a research project that hopefully will develop a scheme of pre-
diction. This scheme would be based on measurements of existing
environmental conditions and analytical tests that would show the
potential for chemical and biological changes. Pollution from waste
discharges and land drainage will be included in the study. Labora-
tory work has started, and sites for field studies are being solicited.
Statistics of multivariable systems
Even the simplest problem of flow regulation for quality control
involves a hydrochemical system that contains several variables of
quantity of water and quality of water. For example, the draw-down
on a reservoir required at a given time in the future will depend on
the quantity and quality of uncontrolled runoff, the quantity and
quality of wastes, the quality of reservoir releases, and the stream-
flow depletions. Each of these variables has a probability distribution
of occurrence. Suppose that one sets as a design objective the main-
tenance of acceptable stream quality 9 years out of 10. The problem
is to interrelate the plausible combinations of the variables to compute
volume of reservoir releases necessary to maintain quality at the
design frequency.
The analysis would be straightforward if all of the variables were
perfectly correlated. Similarly, if all of the variables were completely
independent, a system analysis based on combinations of random
values for each variable would be realistic. The fact is, of course, that
neither extreme exists in any given drainage basin.
This problem provides a fruitful field for research and develop-
ment in methodology. Particularly, it calls for experience and prac-
tical judgment in deciding those simplifying assumptions that can be
tolerated while maintaining a precision of estimates commensurate
with other aspects of flow regulation studies.
SUMMARY
The kinds of hydrologic data and techniques required in deter-
mining flow regulation needs are dictated by the unique nature of
the analytical problem. Consequently, these requirements are rather
specific and exacting.
Projections of flow regulation needed for water quality control
can be improved through better raw data, increased knowledge of
environmental processes, and improved methodology. In review, the
major deficient areas are:
1. Description of low-flow and quality characteristics of streams.
226
HYDROLOGIC FACTS NEEDED
-------
2. Obsolescence of data.
3. Hydrochemical dynamics of reservoirs.
4. Statistical treatment of multivariable systems.
They cover a rather broad spectrum of topics. Likewise, a rather
broad spectrum of effort — administrative, operational, and research
— is required to overcome them.
REFERENCES
1. Streeter, H. W., and E. B. Phelps. A study of the pollution and
natural purification of the Ohio River. III. Factors concerned
in the phenomena of oxidation and reaeration. Public Health
Bull. 146. Feb. 1925. Reprinted in 1958. 75 pp.
2. Ryckman, D. W., et al. Methods of characterizing Missouri River
organic materials of taste and odor interest. JAWWA. 53:1392-
1402. Nov. 1961.
3. Thomas, H. A. Jr, and M. B. Fiering. Mathematical synthesis of
streamflow sequences for the analysis of river basins by simula-
tion. In: Design of Water-Resource Systems. A. Maass et al.
Harvard Univ. Press, Cambridge, Mass., 1962, Ch. 12 pp. 459-93.
DISCUSSION
C. J. Velz
Chairman, Department of Environmental Health
University of Michigan, Ann Arbor
Mr. Rainwater has laid the basis for a very wide discussion. In
the brief time allocated I can but touch the high spots.
The synthesis of what we are after is the quantitative definition
of the waste assimilation potential of each particular stream so that
it can be intelligently utilized without waste or abuse. This synthesis
I call the art and science of stream analysis.
Actually, the scientific tools are at hand to predict with reason-
able reliability in advance of construction what can be expected in
terms of stream quality under a range of streamfiows, waste loads,
and man-made river developments. With notable exceptions, what
is generally lacking is the art of application — the rational process
of thorough analysis preceding design. This has long been standard
practice in many fields of engineering, yet in the field of pollution
control (again with notable exceptions), designs, stream classifica-
tions, quality standards, and control procedures have been arbitrarily
decided upon without benefit of rational stream analysis.
One need not be diverted by new names given to this old process
such as "systems analysis," nor should one delay for the last scientific
Velz
227
-------
re&necient, or qutiable aboct deEaar-cjas it J lit Wsat is jsast pressf-
ingty e-^i-sc. is ts= stppTitaliiHi ct the :at;XTsl analysis process now.
A fir.® anriaeoU 3»cl laois af Erjuysas av-aEafcle Ixia^. T"=rt j-
natjBty, is tV, United States there is. a er-.i=ice:aVe body ct VjW.ji
srui dkraatogojiesl data available. Trafr Uien= ds5-:fercas aad
m:• fgyifl 3?
utilisation can be msde.
Since the warm-wather low runoff season is generally critical,
drought gtreariflov probability framework is of special importance
as it provides the reference base frost vfhich the need far and the
benefit of Haw regulation are measured- We have found that the
graphical adaa'-atiort at GumbeV-s. theory af extreme values is par-
uni'.ariy uscr.il in naming oJOGghr tx-«v eha-acterstici. Quite widely
throughout the United States we find the distributions of drought
{Jaws are logarithmically extremal. Furthermore, indices ca« be de-
rived from comprehensive basin or region studies that are useful ih
estimating drougit severities at locations alea^ the- ararse oi a river
cr3»er thaa at tine ga&ng staticas.
V&riarieis is raorcihly mean cliinaicilagicai raea^jr^rierJ^
as a:r tempera tare, w:r.& velocity, vapcr pressure, a ad solas- radiaticjfi,
quite generally describe normal cistriautiariE Srora which s-ettscmal
patterns of most probable *rA expected confidence ranges can be
readily constructed.
A segment of hydralogic data that too frequently is neglected
fcath in ecfLectioi: ami analysis is detailed characterisation a{ the river
ch^nneL ITis is fcast JulEted bf checeal cross-section sotwdiyigs at
Marrait of appresiiaaw!y Sli> test along Ibe c-wjrsa ct the river by-
means of echo sound eqstipiBeftt. These provide a heais far obtaining
reach by reach the occupied channel volume, surface area, effective
depth, and accurate tires of passage, the importance of which "was so
eloquently argued this marking. These are factors essential far refined
computation of water quality profiles along ifrfe river casirse.
Incidentally, in computing water quality profile -the accvsr.-
titwial oxygen sag equation as usually applied ignores thesfi detailed
ebaiwel factors and in ow experience is inadequate tc cSsfic^j the DO
•premie in all but the most elementary situations. Tbfc ^oc^dute we
favor stems from Phelps" earlier work, m which deosyge&^tion and
m
msfccwoN
-------
reaeration are computed separately taking into account the measured
channel factors reach by reach. This procedure, we believe, is essen-
tial if one is to evaluate adequately radical changes in channel in-
duced by man-made river developments such as navigation works,
dams, and flow augmentation and regulation. The computations are
extensive but are greatly relieved by electronic computers. At this
point, although the use of statistical methods is advocated, a word of
caution. Statistical methods and modern computers are only tools —
not substitutes — for thinking and experienced professional judg-
ment. The machines do only what they are instructed. If you put in
nonsense, out comes nonsense. There is presently a tendency to over-
work data, particularly in multivariate analyses, with the danger of
developing spurious correlations. Statistical analyses based solely on
observed end results are no substitute for understanding of the funda-
mental mechanisms at play and identification of the controlling factors
involved.
A few comments on flow regulation and the impact of multi-
purpose and hydropower developments: I have long been an advocate
of low-flow augmentation. In studies of river basins throughout the
United States, one is struck by the severity of drought flows relative
to the average annual flow available and how depressing it is to gear
community and industrial development to these minimal water
resources.
In many instances a mo lest storage or multipurpose development
can lift a river basin substantially above the restricting framework
of natural unregulated drought flow. On the Savannah River, multi-
purpose developments with reregulation below the hydropower plant
insures downstream flow completely above the normal and the severe
drought levels.
In too many other instances, however, the potential benefits of
multipurpose developments are not realized. For example, on the
Chattahoochee River above Atlanta, Georgia, the multipurpose devel-
opment provides storage that could easily assure flow three to four
times the natural drought runoff. As initially constructed, however,
no reregulator was provided to iron out diurnal pulses and weekend
shutdowns and, as a result, off-peaking flow at Atlanta is no better
than the natural unregulated drought runoff. I could cite a whole
series of such cases.
One might ask, does hydropower plant practice have the right to
inflict this waste of water resources on downstream users? Or should
it be obligatory to provide reregulation below all hydropower devel-
opments? Granted, peaking practice is inherently essential in hydro-
power operation — but if we are to achieve the benefits of flow aug-
mention, reregulation is also essential.
Returning to the stimulating argument of this morning, I would
like to support Mr. Vanderhoof's position. I hope with all the special
difficulties that have been cited you are not left with the impression
that low flow augmentation is bad. Much depends upon where and
how the augmented flow is added.
Velz
229
-------
In detailed studies on the Miami River where the lower river
from Dayton to the mouth is a continuous urban-industrial complex
we find that adding all the augmented flow from the headwaters is
not the best solution. It corrects a serious oxygen depletion and a
high temperature rise from steam power in the upper reach below
Dayton, but it moves the residual waste loads downstream faster to
create a secondary, though less critical, zone in the lower reach This
is similar to the situation Dr. Kaplovsky objects to on the Delaware
Splitting the augmentation, however, and introducing an increment
also in the lower reach produces a greater overall stream improve-
ment with less storage.
This only emphasizes the need for rigorous stream analysis. It
also emphasizes the dangers of generalization and the need to con-
sider each river basin individually.
Notwithstanding difficulties and problems, however, based upon
my years of experience on rivers, I am convinced that low flow aug-
mentation offers great opportunity for stream improvement.
In conclusion I wish to make a strong plea for rational adminis
tration of the 1961 Amendments to the Water Pollution Control Act
Section 2(b)(1) provides that in federal river development "consid-
eration shall be given to inclusion of storage for regulation of stream"
flow for the purpose of water quality control, except that any such
storage and water releases shall not be provided as a substitute for
adequate treatment or other methods of controlling waste at the
source." This was universally greeted with enthusiasm. There is
growing concern, however, over the differences in opinion as to how
this section is to be interpreted.
One rational view is expressed in the concept that reasonable
treatment should be provided with additional benefits to be derived
by augmenting low stream flow, utilizing the waste assimilation
capacity of the stream. On the other hand, irrationally, there are
those who feel that the highest degree of waste treatment possible
should be enforced before any benefit to waste assimilation capacity
can be considered from augmented stream flow. This is equivalent
to saying that all community and industrial wastes must receive com-
plete treatment and only then would consideration be given to low
flow augmentation. To my thinking this will delay for years the
benefits clearly intended in the Act and make difficult or impossible
inclusion in current multipurpose river development of features that
substantially increase waste assimilation capacity and other water-use
benefits.
Also, this viewpoint would result in heavy excess expenditures
for waste treatment plant construction and operating costs for de-
grees of treatment beyond what is "adequate" for current and reason-
ably foreseeable needs. With the pressures on the national economy
and demands on the tax dollar, can we afford this kind of waste? I
sincerely hope we shall proceed rationally and prudently. If we do
not, Mr. Khrushchev may make good his threat to bury us.
230
DISCUSSION
-------
DISCUSSION
Roy E. Lundquist
Hydrologist in Charge, U. S, Weather Bureau River Forecast Center
Cincinnati, Ohio
As a representative of the U. S, Weather Bureau in the Depart-
ment of Commerce, it is an honor to participate with you in this
inter-agency symposium on streamflow regulation for quality control.
Some of you may not knew that the Weather Bureau conducts a very
up-to-date hydrclogic service program that includes streamflow fare-
casting.
Mr. Rainwater, in his excellent paper, has delineated many of
the technical problems that may be encountered, and has pointed
out the kinds of factual hydrologie information needed for studies
associated with streamflow regulation for quality control. But more
than this, he has made some very profound statements that indicate
a need to live more intimately with streamflow and quality control
data. There is time for me to discuss only briefly a few highlights of
his paper.
The role of streamflow regulation for quality control differs
widely in season and location throughout the country because of many
factors such as multiple water uses involved, increases and concen-
trations of population and industry in the watersheds, relative loca-
tion upstream or downstream along the river systems, and ever-
changing flow characteristics due to climatic variations and planned
retention in reservoirs. It is no wonder that the obsolescence of flow
data and quality information, as well as other weaknesses in factual
materials, are considered by the author to be some of the principal
deficiencies in regulation studies. Effective information on sampling
or monitoring for quality would require almost continuous observa-
tion and daily analysis at representative points in the same manner
that flow data are measured and reported automatically for immediate
checking or processing into useful information for further study,
planning, and operation. To us, this is called living with the river
daily.
Mr. Rainwater's analysis for flow regulation by a four-step pro-
cedure is also very interesting because of his conclusion: "A study
of the needs of flow regulation is a forecast . . . projections into the
future. History is essential to, but only a part of, projection." This
creates an enthusiastic response in the mind and heart of an old river
forecaster who has been projecting for many years! Our forecasting
procedures are based on published detailed streamflow records and
stage-discharge rating curves furnished by the U. S, Geological Sur-
vey and the U S. Corps of Engineers. To bring the procedures up to
date and make them capable of handling ever-changing situations,
we must incorporate current data and even anticipate forecasting
problems and changing methods of reporting data. We must also
meet changing conditions in quality control programs, such as Mr.
Rainwater anticipates.
Lundquist
231
-------
The ability to regulate streamfl0w for quaUty CQntrol Qr
of control varies with the amount of reservoir inflow and storage
capacity and also with the location of the impounding structures in
the river system. In the West n States, regulating capacity with
respect to total volume of streamfl0v, is greater than in the midd]e
and eastern sections of the country^ because of a need to conserve
the supply. The more humid climatic conditions and more abundant
precipitation farther east account for greater annual streamflow vol-
umes in the Mississippi River system and in the Gulf and Atlantic
Coastal drainage areas. But there ls orie common denominator which
Mr. Rainwater points out, that appiies to aU within.reservoi'r j
tions as well as to all downstream relations. He labels this- "Volume
of Flow." He mentions that the sample volume-concentration rela-
tion may be expressed as:
Concentration — ^JfoHution load
Volume of flow
He stresses that although this simple dilution mechanism is ap-
plicable to control of almost all types 0f pollutants ... it may not
be the most effective or most efficient means in all cases." It must
be conceded, however, that no matter what the method or mechanism
flow information is a primary parameter for study purposes as well
as for actual control activities whenever regulation is to be considered
Mr. Rainwater's analysis of the deficiencies encountered in the
use of hydrologic facts and figures is also of special interest. His
conclusive statements are that "Knowledge of streamflow characteris-
tics is of little use unless it can be associated closely with attendant
stream quality," and "For our purposes, each event in the past and
future time series of events comprising stream behavior has two
dimensions — quantity of flow and concurrent quality." Then he
makes a recommendation: "To overcome these deficiencies will re-
quire increasing effort on the part of agencies that collect and sum-
marize hydrologic data."
The U. S. Weather Bureau is one of these responsible agencies. In
addition to collecting, analyzing, and publishing basic data that may
be used for hydrologic studies, we also furnish streamflow informa-
tion and forecasts. Other types of forecasts include temperature and
quantitative precipitation data.
Now, if a dynamic study program or operating activity requires
that today's water quality conditions are to be correlated with today's
streamflow information, the next step of planning or regulating for
the improvement or maintenance of water quality for tomorrow must
also be correlated with a specific, corresponding forecast of stream-
flow. In this program of furnishing specific hydrologic information
for up-to-date, concurrent operations and studies, the U.S. Weather
Bureau has made important advancements and contributions through
their River Forecast Center public service programs.
The service furnished by the Weather Bureau River Forecast
Center at Cincinnati has been described(2) and is summarized below.
232
DISCUSSION
-------
Our hydrojogic service program is similar to the programs conducted
a: 10 other centers in the United States In the following discussion
the application of streamficw farac&sting as a tool :n the quality
management of water resources is emphasized. Examples of rela-
tively low flow conditions are given rather than of the flood conditions
that are also an important feature of the II. S. feather Bureau's
program.
Stream/5ow forecasting includes any of the predictable elements
that make up streamflow conditions such as volume, discharge, stage,
flood crests, temperature, velocity, and ice formation.
RIVER FORECASTING SERVICE DESCRIBED
The river forecasting service for the Ohio River and most of its
tributaries is divided among three Weatl:er Bureau forecasting cen-
ters; the Pittsburgh River District Office covering the Allegheny and
the Monongahela Rivers and the Ohio River as far as McMechen,
W. Va. (a main-stem distance of 96 miles); the River Forecast Center
at Cincinnati, Ohio, covering the Ohio River with all the intervening
tributaries between McMechen and Fords Ferry, Ky. (a main-stem
distance of 781 miles); and the River Forecast Center at St. "Louis,
Mo., covering the rest of the Ohio River (a main-stem distance of
103 miles).
The service program at the Cincinnati River Forecast Center was
established in September 1946 for the purpose of strengthening the
river forecasting activities of the Weather Bureau in the Ohio River
basin. The River Forecast Center works closely with and through
eight River District Offices (Parkersburg and Huntington, W. Va.;
Columbus and Cincinnati, Ohio; Louisville, Ky.; Indianapolis and
Ev-ansville, Ind.: and Nashville, Term.) that operate as collection and
dissemination centers for their individual areas, and also provide
sane local river forecasting service
COLLECTION OF DATA USED IN FORECASTING
The complex work of establishing and administering observa-
tional substations, selecting and training observers, collecting and
distributing weather and river data, as needed for the hydrolagic
program in the Ohio River Valley, is a cooperative effort. It involves
many federal, state, municipal, industrial, and private organizations.
The Division and District Offices of the Corps of Engineers, the Geo-
logical Survey District Offices, and the Weather Bureau participate
extensively in establishing, maintaining and financing this reporting
network on which the river forecasting service is based. The main
purpose is to provide current reports of precipitation and river condi-
tions at specified times. The River District Offices of the Weather
Bureau are chiefly responsible for collecting and relaying the reports
to the River Forecast Center at Cincinnati. Reports from 575 stations
are used to develop the forecasts for the Ohio River basin, a drainage
area of 132,433 square miles; 352 stations report precipitation only,
200 stations report precipitation and river conditions, and 23 report
river conditions only. Some of these are fringe stations outside the
Lundquist
233
-------
drainage area. Many stations report once every 24 hours as of 7 a.m.,
but report more frequently on reaching prescribed criteria. Others
report only on criteria. A very important group of 51 stations dis-
tributed over the basin report precipitation every 6 hours on a regular
basis.
THE FORECASTING SERVICE AT THE RIVER FORECAST
CENTER
The data received at the River Forecast Center are immediately
processed through various procedures into discharge and stage fore-
casts The precipitation reports include information on duration and
amount of rainfall, amount of snowfall, snow depth on the ground
and the water equivalent. Meteorological reports from first-order
Weather Bureau stations include information on air temperatures,
wind velocity, amount and types of clouds, and dew points. This
information is used to estimate the volume of runoff to be expected,
which is a fundamental problem in streamflow forecasting. The basic
details of the procedures and techniques used to predict runoff from
storm rainfall are discussed in Weather Bureau Research Paper No.
34(2) The unit hydrograph principle(3) is used for distributing
the runoff and developing the forecast streamflow hydrograph for
the 137 unit sub-areas that make up the total area served by the
Cincinnati River Forecast Center. The river-stage reports are con-
verted to discharge information by use of stage-discharge relation-
ships most of which are developed and furnished by the U. S. Geo-
logical Survey. These discharge values together with reported res-
ervoir releases are used to check the forecast hydrographs and as a
basis for hydrograph adjustment. A combination of streamflow rout-
ine and unitgraph procedures produces the discharge hydrographs in
successive reaches of the tributaries and main stem of the Ohio River.
Thprp are 145 points for which forecasts are made. The forecast
messages are sent out immediately by TWX to the River District
Offices the U S. Corps of Engineer Offices, the St. Louis River Fore-
cast Center and the Weather Bureau Central Office in Washington,
t-\ f othpr forecast summaries and special data are sent on the inter-
Weather Bureau nationwide teletype circuit called RAWARC. To
accommodate other users of the reported information and river fore-
casts the "Daily Ohio River Bulletin," Figure 1, is published and
mailed daily Monday through Friday. Monday's issue carries the
Saturday and Sunday data to help recipients keep a continuous rec-
ord, if desired.
A recent development in the forecasting service at the Cincinnati
River Forecast Center is the velocity of flow forecasts furnished to the
Ohio River Valley Water Sanitation Commission (ORSANCO).
USE OF STREAMFLOW FORECASTS IN
QUANTITY MANAGEMENT
The varying stages and flow conditions that occur in the rivers
and reservoirs of the Ohio River Valley affect the lives or property
of a great many people in some way. When drouth or retention up-
234
DISCUSSION
-------
U. S. DEPARTMENT OF COMMERCE - WEATHER BUREAU
719 POST OFFICE BLDG., CINCINNATI 2, OHIO DATE: Nov. 8, 1963
(Friday)
DAILY OHIO RIVER BULLETIN
RIVERS
AND
STATIONS
OHIO
Pittsburgh
Dam No. 1?
Dam No IS
Parte rifcuri
Dam No. 22
Galtipotb Low*
Graanup Lomt
PortanMuth
Dm Ht 39
Cincinnati
MarfcJafttf Ldwar
Dam N®. 45
Dim No- 44
Ht V
Dam Na 49
Oam No 90
Cake
MISSISSIPPI
St imh
S3 • STAGE ; 24-HR i
TODAY ; CHANGE j
» <16.7P! -0.1 •
se ,'16,3Pj -O.lj !
» |15.6P! -0.3 i
* :io.8Pj-o.lt '
** ii5.5P!-o.5 ¦
» :13.2P:+o.7 !
5k!n.UP: -0.2 ;
w ;12.6P;-0.3 t
» :i5.9P:+o.3 >
M |26 . IP i +o .2 :
5l!i2.6P;+o.3 I
» ii7.opt-o,2 :
ss ! 9.5Pt-0.2 ;
« :l7.8P:+o.li :
*i :i9.iP:-o.3 •
*2; 8.8p:-o.5 •
j# :12.5P- o.o :
V ;19.0P|-0.3 |
S4 :i9.iP{+o.i :
» il5.6Pj-0,i} !
«o :io.8 :+l.8 i
m -'-0.7Fi-l.l
*4 i-2.9F| -O.U
~i j-3.6 ; o.o
u ; 2.up:-o.i
TMBUTARIU
!
FnnMifrAM*wvRtor | 17 ! 2 . 5R ; +0.2
lastNft7-M«noiipMH»¦! »7 jl3<7Fj+2.6
McCkxhIw. Mu^hihh R. j II I M J M
K«»«ht Fato-KaimM Rl » | 3>2Rj+3«0
r. i m ; - j -
141 N I H
! i» i-o.is) 0.0
! m ; o.6s: o.o
! " i
» i
¦
: m
n
HmMRoh-MmI R.
LMkMo. 4-KMkI**.
Mla(OnM-aan«K*.
Ink No. 4-Omm R.
Uck M». 2-Qrmn *
Ml. CtmwMMaMi I.
0.7Ri 0.0
6.BR!+0.2
!
M
6.9
9.9
0.2
0.0
0.0
0.0
NmMM-CvmMXMi* R. ! 40 jl6.5Pj o.o
24-HR
SELECTED FORE
PCPN
9th
loth
•57
Pool
Pool
.02
P
p
.02
P
p
T
P
p
0
P
p
T
P
p
0
P
p
0
P
p
T
P
p
0
P
p
0
P
p
0
P
p
0
P
p
0
0
P
p
0
0
P
p
0
P
p
0
0
10,3
94
0
-0.7
-0.7
0
0
0
.37
.37
.01
7.9
7.5
0
0
0
0
0
0
0
0
0.3
0.3
0
*7.0
17.0
..0.....
.16*5
j CRCtt FORECASTS
—
li STABC
Pooi;
p I
p !
p i
p !
£ '
p i
* i
p i
p i
p
i
•
p
P i
p :
p i
t
t
9.9j
-0.8:
7.5
0.I4
17.0
.16J3-
As of 7 AMs Montgomery Dan, 1 ft. of gates open; Oalllpolla Dan, 3.5
ft. of rollers up; Oreenup Dam, 5 ft. of gate* op«n) Maryland Dan,
2 ft. of gates open, with U« ft. of gates subnerged; McAlpiaa Dan, Ho
dan down, 3 units running. Ohio River watar surface teaperaturei
Mar Id and Dam, 60°F.
Persistant rainfall yesterday over the upper Ohio tributaries, particu-
larly tho Kanawha ana Monongahela, has Increased streanflow substan-
tially and will affect the entire Ohio River flow as ths navigation
dans pass tho increase rapidly through their pools.
g-Day Outlooki Air tosneratures will average about 6° above normal and
rainraii will total 1/2 inoh or more as occasional rain about tho
first of next waok.
MEAN DAILY DISCHARGE FORECASTS in thousands of ouble feet per second:
Hklo D Stft. 0fcU ¦* 1 ^ — -- -- * - - — r
Ohio B.
Dan 15
Huntington
Clnolnnatl
Loulavlllo
8th 9th 10th 11th
6 16 23 17
15 214 2o 23
15 20 25 25
IS IS 22 26
Hth 9th 10th 11th Trlb. Sta. 8th 9th 10th 11th
Kanawha ?alls
Rlvorton.Ind.
Charl'erol,?*.
Oaten ont,Pa,
13
10
3
2
1
2
2
2
11
16
10
3
U
Figure 1 — Typical river foracait bulletin.
Lundquist
235
-------
stream reduces the quantity of flow to minimum ranges, hardship and
damage may result. When excessive rains drench the drainage basins,
the runoff may cause streams to overflow their banks, waste water
resources, contaminate water supplies, endanger property, or even
cause drownings and disaster. One way to lessen these extreme
effects, or at least to moderate the problems that occur when stream-
flows are near extreme limits, is to give advance notice of their occur-
rence.
One of the most important uses for streamflow forecasts is in
flood control management. The U. S. Corps of Engineers and others
engaged in flood control in the Ohio River basin have worked out
very effective manuals of operation that are put into action as soon
as certain conditions of stage or discharge occur. Streamflow fore-
casts (some of which are made by their own organization) are part
of the decisive information on which they base control measures or
preparations to initiate such measures. The stage forecasts of the
Cincinnati River Forecast Center provide official public information
on what the streamflow conditions may be at critical points above or
below the control areas. When criteria stages have been reached or
may be anticipated with a high degree of certainty, the various sys-
tems of control are coordinated as needed to accomplish a desired
objective as determined by the responsible authority. This action
takes into account streamflow forecasts, weather forecasts, river con-
ditions that influence downstream points, available control storage
in reservoirs, probable flood damage in the reservoir areas if water
levels are increased, probable damage at downstream points with
various releases, and other significant factors. It is a complex problem
of balancing damages and managing resources within limits of policy.
The key to the final decision is what may be expected, based on fore-
casts for the immediate future and on studies of past records and
seasonal patterns.
One of the obvious uses of streamflow forecasting as a tool in
water resources management is in connection with water power pro-
duction. Efficient utilization of available storage capacity depends
upon accurate forecasts of inflow into the reservoirs. Where multi-
purpose projects are built using storage capacity for conservation of
water resources, power operations, water supply, navigation, recrea-
tion, and flood control, proper and effective coordination of these
operations depends upon reliable streamflow forecasts.
Another program in which forecasts of river conditions serve
as an effective tool for management is in the obtaining of streamflow
measurements. Receiving advance information about the possibilities
of high water or crest stages on a given stream allows personnel and
equipment to be dispatched in time to make valuable discharge
measurements.
USE OF STREAMFLOW FORECASTS IN
QUALITY MANAGEMENT
There are at least two phases of activity in the quality manage-
ment program that are related to streamflow forecasting. The first
236
DISCUSSION
-------
phase is when studies or plans are being made to develop an operat-
ing program. The second phase is the implementation of actual opera-
tions. These phases may be repeated frequently.
In the study or planning phase, management is watching and
testing both river conditions and the forecasts, preparing operating
criteria for future action. In connection with actual operation, if the
river is falling and streamflow forecasts indicate a continued reduc-
tion of available dilution, management may hold back or reduce
discharge of waste materials until the flow forecasts indicate an in-
creasing quantity of dilution water will become available. Actual
discharge of wastes will occur when flow information and forecasts
assure adequate dilution to yield safe concentrations of the pollutants.
It may be additional waste treatment will be provided. If manage-
ment is concerned about concentrations of undesirable constituents
that often occur in raw river water during the low flow and high
temperature conditions, the forecast information will serve as a guide
to when and for how long additional treatment is needed and will also
help to govern the activities during the period of possible poor quality.
There are other situations that fit into this dual concept of using
the streamflow forecasts subjectively while studies are made and ob-
jectively during control operations. It is well-known by the public
as well as by sanitary engineers that quality conditions of the water
supply change with the seasons and various other circumstances. The
first period of runoff after a dry spell often brings substantial quan-
tities of pollutants into streams. Industrial operations and associated
waste discharges change according to the seasons and with fluctua-
tions in business. Mining activities and resultant wastes vary. Cli-
matic conditions of winter, spring, summer, and fall have character-
istic effects on river flow and water temperature. Ice cover on a river
may limit reaeration at times. The amount of navigation may influ-
ence quality, as may occasional accidental spills of deleterious mate-
rials. The composition of municipal sewage discharges is rather con-
stant on a daily average basis, but the quantity and quality of the
dilution water varies. The concentrations of population and industry
as well as the available water supply differ along the river's course.
This wide variety of circumstances that affect quality conditions in
rivers suggests that during some periods certain operations in quality
management may be relatively dormant, whereas during other periods
expensive, active programs of sampling and treatment must be initi-
ated to accomplish desired objectives in quality control. Sanitary
engineers are well aware that throughout the year the problems of
water quality in the Ohio River Valley change along with changes in
flow conditions in the rivers. Streamflow forecasts, therefore, provide
management with guidance and quantitative information that may
be used in all phases of quality control activities.
Because the Ohio River drainage system is so vast and complex,
quality- and pollution-control problems are complicated and create
the need for broad control programs such as that undertaken in 1948
by ORSANCO under an eight-state compact. While considering a
proposal to regulate discharge of chlorides and other pollutants into
Lundquist
237
-------
the Ohio River and its tributaries, ORSANCO found that streamflow
forecasts would be needed to accomplish their objectives in the plan
ning phase as well as later in the operational phase. Arrangements
were made with the Cincinnati River Forecast Center to furnish
streamflow forecasts, Monday through Friday, in terms of mean dailv
discharge and average velocity of flow for a period of 3 to 4 days in
advance for two tributary stations and four Ohio River stations
At first the accuracy of the streamflow forecasts had to be estab
lished to determine whether they would be satisfactory to use as a
basis for regulation and operation on a broad scale A study as to
how the forecast information could be coupled to current qualitv
data was also made. With the recent establishment of robot-monitor
ing stations to appraise water quality conditions, the streamflow fore
casts provided to ORSANCO have been expanded to cover 12 station*
with 4-day forecasts. As the ORSANCO projects on quality control
move toward operational programs, streamflow forecasts will be morP
significant and be given wider distribution.
The relative accuracy of the streamflow forecasts has been meas
ured against the standard of mean daily discharges published bv the
Geological Survey for the same or nearby points. In Figure 2 data
for the Ohio River at Gallipolis, Ohio, is compared with published
data for the nearby station at Huntington, W. Va. (32 miles down
stream), during low-water conditions in September 1960.
In an appraisal of the need for streamflow forecasts in aualitv
management published in Public Works(4), Edward J. Cleary Exec
utive Director and Chief Engineer of ORSANCO, wrote "Knowledge
of the availability of dilution water (streamflow) is fundamental to
the practice of pollution control. But if we regard pollution-control
practice as a dynamic exercise in the management of river aualitv
the need becomes apparent for daily information — and preferablv
advance forecasts - on the availability of dilution water "
CONCLUSION
Streamflow forecasting has been and will continue to be an im-
portant tool in quantity and quality management of water resources
There is need to get better acquainted with its possibilities and to
extend this service In the future, greater demands on our water
resources will inevitably generate the development of a more com-
plete program for supplying advance information of streamflow con-
ditions.
REFERENCES
1. Lundquist, R. E. Streamflow forecasting as a tool in the quantity
and quality management of water resources. Presented at Water
Resources Eng. Conf., ASCE, Omaha, Nebr., May 14, 1962.
2. Kohler, M. A., and R. K. Linsley, Predicting the runoff from
storm rainfall. Weather Bur. Research Paper No. 34. Sept. 1951.
3. Sherman, L. K. Streamflow from rainfall by the unit-graph
method. Eng. News-Record. 108(14):501-05. Apr. 7, 1932.
238
DISCUSSION
-------
4. Cleary, E. J. Streamflow forecasts — a new "tool" for river-
quality management- Public Works. 92:72-74. July 1961.
TFSSWTWTFSSMTWTFSSMTWTFSSMrwrF
15 10 15 ao 25 30
Figur* 2 — U.S. Weottiw (wtou {Cincinnati) foracadi for Ohio Riv«r itiwamKaw for
i«p>*drbar I960 '-or OalKpoHi, Ohio, lew«r ga-g*, compared with published US.G.S. dii-
chacgm at nearby Pvnlia&ton. W«*t YirgSafa mi!«» dowmtremui.
Lund quia*.
-------
DISCUSSION FROM THE FLOOR
Mr. Morriss, Atlanta Water Pollution Control: Dr. Velz, did I
understand you to say that below the Buford Complex there are no
reregulating facilities?
Dr. Velz, University of Michigan: That is my impression.
Mr. Morriss: On behalf of our taxpayers, I have to claim credit
for a half million dollars worth of reregulation facilities in conjunc-
tion with an equal amount supplied by the Georgia Power Company.
Our problem is not the lack of a facility. Perhaps it is not as big
as it should be — it is only 3200 acre-feet — and its capability to
meet dilution needs is not completely adequate — but this is not the
real problem. The problem is getting enough water out of Buford
Reservoir to operate the facility as it should be operated. This is a
problem that will come more and more into play in the area of our
discussion here. The question really is: Shouldn't reregulation facil-
ities be considered in the economic justification of a peaking-power
operation, particularly where there are strong pollution sources down-
stream including other steam-generation power facilities? And can't
we make this consideration ex post facto; in other words, give us our
money back and we will build a bigger one?
Dr. Velz: I tried to point out that we cannot put the hydropower
people out of business. Production of peaking power, while carrying
the base demand on steam plants, is the essence of the value of hydro-
power. But I maintain that by and large we have failed to recognize
that peaking-power operations are creating some real havoc on our
rivers, and I raise this issue here: Do the power people have the
inherent right to do this? Or, isn't there an inherent obligation in
hydro operation to equalize flow to iron out the hourly fluctuations,
and more important, the shutdown of flows on weekends? There are
many rivers in which for as long as 2 Vz to 3 days during the weekend
shutdown, flow is below the once-in-10-year drought flow. Yet we
invest large sums of public money, presumably in multipurpose de-
velopments, and we are no better off than we were under the natural
drought condition. This is an issue I would like to see out on the
table. I would hope that the Federal Power Commission might play
a role in this by recognizing that in giving a license to build a dam,
it should include reregulators as a basic part of the project.
Mr. Morriss: Thank you, Sir. This is my point. Don't let me
give the impression that we are not satisfied with our reregulation.
When this structure was built in 1957 it was to save "X" number of
dollars; however, the industrial and population growth in the metro-
politan area has made this a highly untenable statement today. We
have to have more flow, but under the law we have to have complete
treatment first. We have to get the treatment, then we have to go
to Congress, and then we may get the flow. Now, this is "backwards."
Dr. Velz: You need water right now, I believe, for steam-power
operation in Atlanta, don't you?
240
DISCUSSION
-------
Mr. Morriss: That is generally true.
Dr. Velz: So it isn't only a matter of pollution alleviation. In the
total power interest, reregulation would be a real asset and it ought
to be considered as an integral part of the deal.
Mr. Morriss: That is my feeling.
Mr. Moore, U. S. Corps of Engineers: I want to comment with
respect to Mr. Morriss' question. We are now designing and building
reregulating structures in connection with four hydropower develop-
ment projects in the Tulsa District. A part of the cost of these reregu-
lating structures is included as a cost to power; a portion is assigned
to water quality control. It is a federal expense and will be paid for
by the federal government.
Mr. Wigglesworth, U. S. Department of Interior: I have some
little knowledge about the operation of the Buford project. The oper-
ating schedule for Buford Reservoir includes a provision to provide a
650-cfs flow at Atlanta. The average critical-period flow at Atlanta
is some 1200 cfs, so we are releasing a substantial part of the critical
flow as a minimum flow release at Buford. The reason we have such
a substantial minimum flow on the Savannah River is that it was
planned as a navigation project. The Buford project was not planned
for the purpose of providing navigation flows in the upper Chatta-
hoochee River; it was planned primarily for flood control, power, and
pollution abatement. The power operation pays a substantial part of
the cost of the Buford Reservoir, and if we were not able to peak
there, as we do, we would not be able to pay this cost.
From the Floor
241
-------
THE NATURE OF THE STORAGE-YIELD
RELATIONSHIP
Dr. Myron B. Fiering
Assistant Professor of Engineering and Applied Mathematics
Harvard University, Cambridge, Massachusetts
INTRODUCTION
The storage-yield relation is a traditional tool in water resource
planning; it specifies the storage required to maintain a given flow, or
conversely, the yield associated with a given reservoir size. This
paper summarizes a technique for making more effective use of the
historical flow sequence in developing the storage-yield relation (1).
The statistical consequence of the proposed analysis is more
efficient utilization of the information inherent in the discharge rec-
ord. The essential point of this paper is that the numerical values
comprising the discharge record, and their particular permutation,
identify but a single sample from the population of all possible rec-
ords at the site; a different required storage can be evaluated as a
function of each possible record. This study focuses on the distribu-
tion of the population of required storages, since any member of the
population can be that storage required during the active economic
life of the project. The selection of a design value from the popula-
tion of storages must be made on the basis of certain criteria that
encompass the objective function for the project.
This study treats a wide range of possible data combinations
characterized by several inflow populations with different coefficients
of skewness and serial correlation, by several levels of regulation,
and by record lengths typical in hydrologic studies. Each combination
of population, record length, and level of regulation defines a separate
run; in all, 144 different runs were investigated. Without the aid of a
high-speed digital computer this research could not be undertaken.
HISTORICAL SURVEY
Mass curve (or Rippl diagram) analysis was introduced in
1883(2). Observed flows at a proposed site are used in t e mass
curve analysis to determine the storage capacity required for any
desired draft. For example, at 100 percent regulation (i.e., when tne
annual draft is a constant equal to the mean annual inflow), e
associated "ultimate" storage is given as the range of the cumulative
departures of the inflows from their sample mean. Graphical
adaptations of this technique permit one to compute the storage r
quired for any given level of regulation, or to evaluate e y*®
associated with any prescribed storage level. For iOO percent regu -
toon, the ultimate storage is the storage that spills and goes y
once during the active life of the structure.
H. E. Hurst(3), in his controversial paper in 1951, presented th
fiering
243
-------
theory that gives the average or expected value nf the r
lative departures of inflows from their sarrmle mean tt >, ° Cumu~
sample mean, He showed that
E(R*> - /f » - 1.25 „ vrr Eq J
where a is the standard deviation of annmi ^ at ,
record, and R* the adjusted range (j e th ' . length of
from their sample mean). He examined an pnnmr!""65 measured
natural phenomena, collecting data from trep rim** nurn^er °f
hydrologic records. Computations showed that tiE varves> and
proportional to the 0.72-povver of the record 86 is
0.5-power as postulated by Equation l Hp wn
-------
We postulate a linear regression model (Figure 1) of lag one;
that is, the flow in any season is a linear function of the flow in the
preceding season. Thus,
Qi+i = Qj+i -f- /?j (Qi ~ Qj) Eq. 5
wherein Qj+i and Qt are the discharges during the (i+l)Bt and ith
season, respectively, reckoned from the start of the synthesized se-
quence; Qj+i and Q] are the mean seasonal discharges during the
(j + l)st and jth season, respectively, within a repetitive annual cycle
of seasons; and /3j is the least-squares regression coefficient for esti-
mating the flow in the (j + l)9t from the jth season. Season is used
here to define any convenient subdivision of a year — months, weeks,
days, etc.
Figure 1 — General linear regreition model.
The expected value of Qi+1, written E(Q1+1), is"£jJ+i, thus the
first moment of the distribution of historical flows is preserved by the
model in Equation 5. It can be shown, however, that the variance
V(Qi+1) = pj2 ffj-fi2. In this formulation, trj+i2 is the variance of
discharges during the (j+l)st season and p} is the serial correlation
coefficient between flows in the jth and (j + l)Bt season. The model in
Equation 5 does not preserve the variance of the historical flows
unless pj = + 1; hence, some modification is necessary.
If ti is a standardized random variate, the recursion relation
Qi+i = Qj + i + Pi (Qi — Qj) + ti «jj +1 (1 — pj2)1'2 Eq. 6
preserves the mean and variance of historical flows. Equation 6 en-
tails the deterministic, or trend line, estimate defined by the linear
Fiering
245
-------
relation of Equation 5 and a stochastic, or random, component pro-
portional to the standard error of estimate „j + 1 (1 — We can
envision band widths of 1, 2, , n [ffj+1 (j — pj2)1/2], which contain
a corresponding proportion of all the points in the bivariate distri-
bution of Qj and Qj+i' t^le evaluation of these proportions depends
on the distribution of seasonal flows.
If a normal distribution is postulated, Equation 6 is a sufficient
algorithm, since the first two moments are preserved. In this case, the
tj are normally distributed. If the flows are characterized by the
gamma distribution, three parameters must be preserved and the
distribution of t( must refiectYjtthe skewness of the seasonal inflows.
It can be shown that, in order to preserve the skewness of the
distribution of historical flows, the skewness of the random variate
t, is
yt = — Pj'Vj
(1-pj2)3/2 Eq. 7
If tN is a standardized normal variate, ty can be computed from the
transform
ty yt 11 + 6 36 J yt E(l-8
It is interesting to note that more skewness must be built into the
gamma variate, ty, than exists in the original record.
If the historical record were characterized by a distribution re-
quiring four or more parameters, the kurtosis and higher moments
of the synthetic data could be preserved by further manipulation of
the distribution of t\.
The computation can be summarized as follows:
1. Analyze the historical data and compute the constants Qt R,
aj, p}, yj for each season. J'
2. Specify the distribution to be maintained.
3. Synthesize a flow sequence of desired length.
4. Examine the moments of the synthesized data to ascertain
that the model faithfully, but not necessarily exactly repro-
duces the given moments.
The essential property of every generated sequence is that, based
on all statistical tests of significance, the generated sequence cannot
be distinguished from the original historical record. If the synthetic
generating scheme contained in Equation 6 is valid, any sequence
generated therefrom defines a sample that might be drawn from the
total population of all possible records characterized by the moments
of the observed sequence. Moreover, the storage required in any
one synthetically generated sequence is just as likely to be the
storage required during the active life of the project as that asso-
ciated with any other generated sequence, including the storage
2
2
246
the storage-yield relationship
-------
computed from the historical record. That is, the historical record
defines a single storage that is but one member of the population of
all possible storages. If a large number of replicate sets of synthetic
data are generated, e.g. 100, the 100 resulting storages define a dis-
tribution that is an approximation to the distribution of all storages.
When 100 sets are generated and analyzed, the first two moments
of the distribution of required storage appear to be fairly reliable;
the third and higher moments of the distribution are not well defined.
Let us summarize the analysis. From any of the 144 combina-
tions of streamflow population, serial correlation, length of record,
and degree of regulation, 100 replicate sets of synthetic flow data are
generated. For each set, we compute the storage required to meet
the specified draft; this value is a member of a sample of size 100
taken from the population of all possible required storages. The
sample is used to deduce the distribution (i.e., the moments) of the
required storage corresponding to the given combination of param-
eters.
COMPUTER STUDIES
We consider four streamflow distributions, four record lengths,
three levels of annual serial correlation, and three degrees of regu-
lation. Since the number of parameter combinations increases in a
multiplicative fashion, 144 combinations define the sample space for
this investigation. In each instance, the annual flows are assumed to
be derived from a population with a mean of 1.0 and a standard
deviation of 0.25; the resulting coefficient of variation, 25 percent, is
typical for streams in several regions of the United States. The
computer is coded to generate a variable number of replicate sets
of synthetic data; we choose 100 for each combination. Thus, 14,400
sets of hydrology are generated and analyzed by the computer pro-
gram within a few minutes. The parameters assigned in the compu-
tation are:
Populations
Normal
Gamma with skewness —0.5
Gamma with skewness 0.5
Gamma with skewness 1.0
Record lengths
10, 25, 50, and 100 years
Annual serial correlations
0.0, 0.1, and 0.2
Degrees of regulation
100%, 90%, and 80%
A one-season model is utilized in these runs to minimize the
effects of within-year storage. The computer generates a synthetic
Fiering
247
-------
flnw ceoupnee of given length from Which the mean annual inflow is
The annual draft is computed from the sample mean (lem
mean annual inflow) whereupon 'he qui d torage is determined.
Thp tpohniaue for evaluating "^required storage is applicable to
zny annual draft pattern and > * relation; it is a gener
alization of the mass-curve analysis and ls called the sequent-peak
method.
The nrocram computes and reports the mean and standard devi-
sample of size 100 Qf the required storage. These
results are summarized in Tables 1 arid 2 and in Figures 2, 3, 4, and 5.
Figure 2 — Required storage v» length of record, 100% regulation.
248
THE STORAGE-YIELD relationship
opo ezi-7«o-»
-------
TABLE 1. MEAN AND STANDARD DEVIATION OF
REQUIRED STORAGE ®
Popula- M Regulation, %
tion P N> yr 100 90 80
Normal
0.0
10
0.7
(.23)
0.4
(.17)
0.2
(.17)
25
1.2
(.35)
0.6
(.22)
0.4
(.18)
50
2.0
(.47)
0.8
(.28)
0.5
(.18)
100
2.9
(.70)
1.0
(.25)
0.6
(.19)
0.1
10
0.7
(.24)
0.4
(.18)
0.2
(.17)
25
1.3
(.37)
0.6
(.25)
0.4
(.20)
50
2.1
(.54)
0.9
(.33)
0.5
(.21)
100
3.2
(74)
1.1
(.36)
0.6
(.22)
0.2
10
0.7
(.29)
0.4
(.20)
0.2
(.18)
25
1.4
(.43)
0.7
(.29)
0.4
(.21)
50
2.2
(.53)
1.0
(.39)
0.5
(.24)
100
3.2
(.74)
1.3
(.44)
0.7
(.24)
Gamma
0.0
10
0.7
(.23)
0.3
(.16)
0.2
(.13)
(y=-j-0.5)
25
1.2
(35)
0.5
(.20)
0.3
(14)
50
1.9
(.46)
0.7
(.25)
0.4
(.13)
100
2.9
(.73)
0.9
(.23)
0.5
(.14)
0.1
10
0.7
(.24)
0.3
(.17)
0.2
(14)
25
1.3
(.39)
0.6
(.23)
0.3
(.16)
50
2.1
(.55)
0.8
(.29)
0.4
(.16)
100
3.2
(.75)
1.0
(.32)
0.5
(.17)
0.2
10
0.7
(.30)
0.3
(.18)
0.2
(.15)
25
1.4
(.44)
0.6
(.26)
0.3
(.17)
50
2.2
(.54)
0.9
(.35)
0.4
(.19)
100
3.2
(.75)
1.2
(.40)
0.5
(.19)
Gamma
0.0
10
0.7
(.24)
0.4
(.20)
0.2
(.17)
(y= -0.5)
25
1.2
(.36)
0.6
(.24)
0.4
(.23)
50
2.0
(.49)
0.9
(.31)
0.6
(.24)
100
2.9
(.68)
1.1
(.29)
0.7
(.24)
0.1
10
0.7
(.27)
0.4
(.20)
0.2
(.17)
25
1.3
(.37)
0.7
(.29)
0.5
(.25)
50
2.1
(.52)
1.0
(.37)
0.6
(.27)
100
3.2
(.74)
1.2
(.41)
0.8
(.27)
0.2
10
0.7
(.29)
0.4
(.27)
0.2
(.22)
25
1.4
(.44)
0.7
(.32)
0.4
(.26)
50
2.2
(.56)
1.0
(.43)
0.7
(.29)
100
3.2
(.72)
1.4
(.48)
0.8
(.29)
Gamma
0.0
10
0.7
(.24)
0.3
(.17)
0.1
(.10)
(y—hl-0)
25
1.2
(.36)
0.5
(.19)
0.2
(.10)
50
1.9
(.46)
0.6
(.22)
0.3
(.09)
100
2.9
(.78)
0.8
(20)
0.3
(.10)
0.1
10
0.7
(.26)
0.3
(.18)
0.1
(.11)
25
1.3
(.41)
0.5
(.22)
0.2
(12)
50
2.1
(.58)
0.7
(.26)
0.3
(.11)
100
3.2
(.78)
0.9
(.30)
0.4
(.13)
0.2
10
0.7
(.32)
0.3
(.19)
0.1
(14)
25
1.5
(.47)
0.5
(.24)
0.2
(14)
50
2.2
(.57)
0.8
(.31)
0.3
(.16)
100
3.3
(.78)
1.0
(.37)
0.4
(15)
a Entries in the body of the table are the average storages, based on a
sample of 100; the associated standard deviation is in parenthesis.
Fiering 249
-------
TABLE 2. COMPARISON OF THEORETICAL AND
MEASURED STORAGE a
w E(storage) Range of calculated Range of calculated
'yr =1.25(r-\/N average storage Cv, %
10 099 0.7 — 0.7 33 — 36
25 1.56 1.2 — 1.5 28 — 32
50 2.22 1.9 — 2.2 23 — 27
100 3^13 2.9 — 3.3 23 — 27
s 100% regulation; 100 replicate sets.
100% REGULATION
2.5
2.0
Li
19
<
at
O
UJ
a
<
a:
UJ
>
<
1.5
1.0
0.5
90% REGULATION
80% REGULATION
SO
LENGTH OF FLOW SEQUENCE, yr
100
Figure 3 — Average required storage vs length of record,
p = 0.1, gamma distribution (y — 0.5).
250
THE STORAGE-YIELD RELATIONSHIP
-------
0 50 100
LENGTH OF FLOW SEQUENCE, yr
Figure 4 — Average required storage vs length of record,
gammct distribution (y = 1,0), 100% regulation.
SAMPLE MEAN * 2.14
SAMPLE STANDARD DEVIATION = 0.537
0.15
s
z
id
3
S
* 0.10
w
>
0.05
1.1 1.3 1.5 1.7 1.9 2.1
I I
2.3 2.5 2.7 2.9 3.1
N 50 YEARS, 100<7r REGULATION, n = 0.1, C = 0.25.
r *
100 REPLICATE SETS
Figure 5 — Distribution of required storage.
Fiering
251
-------
CONCLUSIONS
The principal results can be stated as follows:
1. The work of Hurst and Feller is verified. Skewness of
the data is far less important than the standard deviation
in determining the mean range. Agreement between
theoretical and measured values improves as N increases.
2. The effects of low values of the annual serial correlation
on the range are pronounced; presumably more substan-
tial variations (i.e., larger storages) would arise for
higher values of p.
3. The effects of small reductions in the yield are very
marked.
4. It is manifest that use of the ultimate storage, or any
other storage computed from a mass-curve analysis of
the historic flow sequence, is fraught with risk because
of the high degree of instability inherent in the statistic.
The use of synthetic sequences to define the sampling
distribution of the range is proposed as an alternative
design technique.
FUTURE WORK
Employing the statistical consequences of the distribution of the
range to forge a union between engineering and economics, which is
tantamount to writing an unambiguous objective function, we are
currently studying techniques to further the scope of rational deci-
sion making in engineering design. Work on the theory of the distri-
bution of the range is an essential link in the process of providing
rational designs to satisfy man's needs.
REFERENCES
1. Fiering, M. B. Statistical analysis of the reservoir storage-yield
relations. In: Operations Research in Water Quality Manage-
ment. H, A. Thomas, Jr., and R. P. Burden, eds. Harvard Water
Resources Group, Cambridge, Mass. 1963.
2. Rippl, W. The capacity of storage reservoirs for water supply.
Proc. Inst. Civil Engrs. 71:270. 1883.
3. Hurst, H. E. Long-term storage capacities of reservoirs. Trans
ASCE. 116:776. 1951.
4. Feller, W. The asymptotic distribution of the range of a series of
independent random variables. Ann. Math. Stat. 22:427. Sept
1951.
252
THE STORAGE-YIELD RELATIONSHIP
-------
5. Anis, A. A., and E. H. Lloyd. On the range of partial sums of a
finite number of independent variates. Biometrika. 40:35. 1953.
6. Yevdjevich, V. M. Some general aspects of fluctuations of annual
runoff in the Upper Colorado River Basin. Colorado State Univ.
Rept. 1961.
DISCUSSION
Dr. Richard L. Woodward*
U. S. Public Health Service, Cincinnati
Dr. Fiering has done an excellent job of honoring Rippl and
castigating his diagram. This paper points up very clearly the weak-
ness of the practice of basing reservoir designs on studies that con-
sider only the single historic trace of a streamflow record. The
use of the statistical methods outlined by Dr. Fiering in this and in
other papers can give the designer a much clearer insight into how
the reservoir he is designing is likely to function in meeting the
demands made upon it.
These studies emphasize that reservoir yield needs to be looked
upon as a statistical variate. This fact is probably appreciated intui-
tively by most engineers and hydrologists engaged in water resource
design, but the implications of the fact have not been very deeply
explored. Neither was there any good information on the variance
of our estimates of storage requirements for a given yield prior to
this paper. I doubt that many would have believed that these esti-
mates would have a coefficient of variation of more than 20 percent,
as Dr. Fiering's studies show. Although the variance of these esti-
mates decreases as the percentage of the mean annual flow developed
decreases, the coefficient of variation is even greater. Some of the
figures in Table 1 indicate coefficients of variation greater than one.
As Dr. Fiering points out in his final paragraph, the obvious
lesson from this is that the common design practice of selecting a
target demand for some use such as water supply and determining
the storage required to meet the demand leaves something to be
desired. Nevertheless, it is necessary to set some storage for design
purposes. Since there is some finite probability that this fixed storage
will fail to meet the target demand, we need to be able to estimate
the losses such failure will entail. This procedure is not uncommon
in the case of storage for power generation or for irrigation, but as
yet it has not been used in the case of storage for municipal or indus-
trial water supply. We need to develop the background information
necessary to make such estimates realistically.
*Now Senior Research Associate, Harvard School of Public Health,
Boston, Massachusetts.
Woodward
253
-------
DISCUSSION
Dr. Zane Spiegel*
Harvard University, Cambridge, Massachusetts
One of the important implications of Dr. Fiering's paper is that
the natural streamflow that we can expect to have in the future will
at times reach new minima that will be lower than any experienced
historically. Obviously the reason for this is that precipitation in
some years or seasons will be less at times than for similar periods
experienced in the past. The methods used by Dr. Fiering(l) can
be applied directly to precipitation data to give us an idea of the
potential severity of future precipitation droughts over any period
of time, in a stochastic sense.
The low flow portion of the streamflow distribution can be con-
sidered the result of the interaction of (1) a stochastic precipitation
element, having a low serial correlation coefficient, and (2) a deter-
ministic ground-water element, dependent upon the characteristics
of all the aquifers in the stream basin. Any stream basin can be
represented as an aquifer system, and the ground-water outflow that
sustains the perennial streams can be predicted from the time-
dependent boundary conditions imposed by precipitation, infiltration
capacities, direct runoff, and man's activities(2).
"Streamflow regulation for quality control," the title of this
Symposium, is needed principally during the times when natural
streamflow is entirely from ground-water sources. Therefore the
problem at hand is essentially a ground-water problem, not a surface-
water problem and should be called "base-flow augmentation" to
keep our thinking clear. Heretofore the approach to regulation has
been from the surface-water point of view alone. In many stream
basins, however, manipulation of ground-water storage in extensive
aquifers or systems of aquifers can provide a significant degree of
regulation for quality control and other purposes. This method of
streamflow regulation can be called "internal base-flow augmenta-
tion," in contrast to the "external base-flow augmentation" achieved
by the use of surface reservoir storage.
Now, two of the four symposium objectives, quoted from our
program, are:
1. Review available knowledge on the effects of streamflow
regulation on water quality.
2. Explore research needs in the field of streamflow regula-
tion for quality control.
In regard to the first of these two objectives, it should be empha-
sized that we already have a vast store of knowledge of theoretical
and field data on ground water that relates to base-flow augmenta-
*Now Manager, United Nations Special Fund Project, Ground-Water
Research in Northwest Argentina, San Juan.
254
DISCUSSION
-------
tion; the important feature that is lacking is a general understanding
of what is already in the literature and in the files of public agencies.
Thus, the second objective can be translated to:
2a. Explore methods of dissemination of knowledge of tne
importance of ground water to streamflow regulation for
quality control.
There is more usable ground water in storage than exists in all
the surface reservoirs ever built. Nearly every large stream basis
has at least one important aquifer, if not a whole system of inter-
related aquifers. Every stream on the Atlantic coast has not one, but
two major aquifer systems (one for the proponents of upstream con-
trol, and one for the proponents of downstream control). These
aquifer systems are, respectively, (1) the belt of limestone forming
the Great Valley in Pennsylvania, Maryland, and Virginia, and
extending as far south as Georgia and Alabama; and (2) the
coastal plain sediments that extend all the way to Florida and
the Gulf Coast as well. In general, current development of the
ground water alone is resulting in stream depletion rather than aug-
mentation, but if wells are located properly with respect to the
streams and if recharge facilities are constructed, ground-water utili-
zation and base-flow augmentation can be achieved simultaneously.
Current investigations of quality control, using the Lehigh River
in Pennsylvania as an example, lead to the conclusion that ground-
water storage in the limestone belt of the Lehigh Valley (an exten-
sion of the Great Valley) can be manipulated to increase base flow,
decrease stream temperature, and neutralize acid mine drainage more
effectively than any surface reservoir plan now under consideration.
A few figures will illustrate. The historical minimum flow of the
lower Lehigh River near Allentown has been less than 700 cfs and
can be expected to be much lower in the future, as has been implied
earlier. River temperatures currently go to more than 75°F and can
be expected to go even higher in the future. Surface reservoir sites
are too far upstream to be effective in lowering stream temperatures
in the lower Lehigh River. Installation of a system of injection (from
surplus winter flow) and production wells in the Lehigh Valley,
however, can furnish 416 cfs of water at a temperature of about 54°F
in the immediate area during the low-flow summer season when
base-flow augmentation is needed most.
A similar solution is possible in the Potomac and Susquehanna
Rivers and in fact in every stream basin in which the limestone aquifer
is present. Other aquifers can be utilized similarly, although artificial
recharge may be more difficult and hence more costly. Nevertheless,
large-scale aquifer management schemes may be more economical
than surface reservoirs in many areas.
Now back to the generalities. Some of you may conclude that I
came up here to present a concept of multipurpose water resource
design based on aquifer storage manipulation, rather than on surface
reservoir regulation. Well, this is in fact my aim. Current concepts
Spiegel
255
-------
of multipurpose use of surface reservnirc Wro j
from the need to add new function! haX ,PrinCipally
benefits, real or otherwise, to projects that were
the primary function alone The same logic can be appliedto the plan*"
ning of water resource control systemg baged imaril aquiferPstor'
age manipulation. Floodplain clearing to nrnvirtP t f
or artificial recharge facmtes has the additional benefits ofXod^tage
and damage reduction open space for recreational and aesthetic uses
or space for waste treatment facilities. Replacement of fixed channel'
obstructions with pump diverSions or flpx;ulfa ^amB a Channel
stage and damage reduction as we ? fl°°d
course aquifer storage manipmall0 w ,. fi ,, ^ anc* water-
in parks or along "green strip.,- '°"d ^ be constructed
distributed in landscaped open channels for some injectlon'o/watM
^ater^aeraUo^pu^li^ben'efits'th11!0^6 aeSthvided on a
muftipurtseUeflts o, JXurt™1'!'* <* '!*
cited, but only one more will be given jn nrHf.,, tr water can be
discussion of Dr. Fiering's paper. ' return to specific
It is apparent from Dr. Fiering's Figure 3 that the average stora,.
required increases very rapidly in order to achieve infreSlv
greater degrees of streamflow reeniat;™, 07 increasingly
the total runoff. This means thatTemenZJ j°f
must be made if maximum utilization nf haoir. aPltal investments
achieved by the construction of surface tpcp ¦W fr is to be
the uncertainty in the accuracy of proieetior>I°nf%at0ne' j view of
stream regulation, such largeTap^^^S,^ ***?* for
nomically justifiable. However, ground-water developments can^be
used to provide water for various purposes by well-by-weU or field
by-field construction at much less risk of capital (hence w h ereatt
reduced interest payments). Floodolain 7nnino a™,* • greatly
channel and floodplain rehabilitation can usually provide^lUhetene"3
(its allocated to reservoir detention, frequently at lower cost
references
1. Fienng, M. B Statistical analysis of the reservoir storage-yield
relation. In: Operations Research in Water Quality Management
H. A. Thomas and R. p. Burden, eds. Harvard Water Resources
Group, Cambridge, Mass. 1963. ources
2. Spiegel, Z. Hydraulics of certain stream-connected aquifer sys-
loTpp °° 6 Eng" SpeC< Rept- Santa Fe< N. M. 1962.
DISCUSSION FROM THE FLOOR
Dr. Yevdjevich, Colorado State University: There are three
groups around the world actively studying the storage-yield problem.
256
DISCUSSION
-------
One is the Harvard group of which Dr. Fiering is a part, the second
is the Australian group with Statistician Professor Moran and Hy-
drologist Alexander, and the third is the Moscow group with Pro-
fessor Kritskiy and Academician Menkel. Each of the three has a
different approach to the storage problem.
The Harvard approach is generally called the "synthethic hydrol-
ogy method." Another approach, called the "probability method,"
uses stochastic processes in describing hydrologic sequences.
Dr. Fiering noted the work on the expected value of the range of
cumulative departures of streamflow from the sample mean by Hurst
in Great Britain and by Feller in this country. It should be pointed
out that they were concerned with the asymptotic distribution of the
range of storage needed, i.e., with very large n. A substantial con-
tribution in this field was made by Annis when he computed the
theoretical moments of the range of the cumulative departure from
the mean (or the necessary storage) for normal independent vari-
ables. Dr. Fiering might be interested in comparing his results with
those derived by Annis. On general inspection of presented data, it
seems to me that Annis' values are higher, and that there might be
some differences in definition of the mean range.
The Australian group has been working very actively the last
couple of years to obtain a theoretical solution for the range of the
departure of independent flows having gamma distribution when the
mean annual flows are asymetrically distributed, or to find the
probability that a given amount of storage is required to regulate
flows over periods of several years.
These two approaches, one using computer simulation that is
now well advanced by the Harvard group, and the other that is
purely theoretical, are a good combination.
Professor Feller has indicated that it is practically impossible
to compute the exact distribution of the range of required storage
even for 3, 4, cr 5 years. We have succeeded in the last few years
in developing expressions for exact distribution up to n = 4. It
would be a good contribution to the solution of storage problems if
the exact distributions of the range could be developed for n = 5
to n = 10.
In Fiering's approach, the first serial correlation coefficient is
used for the correlation of successive seasons or periods. It is known
that the St. Lawrence River, for example, has a carryover of approxi-
mately 8 to 10 years, so that a wet year affects flows during the next
8 to 10 years. It might be useful to extend the correlation approach
to rivers that have long carryover periods, or for which the first
order or the second order linear autoregressive schemes may be
applied. Dr. Fiering's statement about the average values of first
serial correlation coefficients is correct. In a recent study that I did
with 446 river gaging stations in the Western United States and
Canada, I found that the average first serial correlation coefficient
From the Floor
257
-------
was 0.20; this might now be taken as an expected value for river
gaging stations and annual river flows.
By using these new techniques described by Dr. Fiering, we may
soon be able to describe river flows by stochastic hydrologic proc-
esses. These hydrologic processes have a structural part (seasonal
fluctuations that are cyclic) and a random part, while water carry-
over from month to month and year to year may be considered by a
moving average model.
It should also be possible to describe water quality by similar
mathematical processes, so that in the future it can be described in
much the same way as we are now attempting to describe the flow
sequence.
Dr. Fiering, Harvard University: I would like to comment on
one of Dr. Yevdjevich's points. The gentlemen from Russia have re-
quested permission to come to Harvard next year; I am going to
Australia in the year following. So we hope to get a fairly compre-
hensive look at all the attacks on this matter.
One other point is that in instances with long carryovers, such
as the St. Lawrence River, you don't really have much of a design
problem simply because the flows are, by definition, so nearly uniform
that extending autoregressive schemes to such long lags is, at least
from a design point of view, not a terribly useful practice-
Mr, Deininger, Northwestern University: Dr. Fiering, you are
synthesizing these streamflows from equation 6. Since equation 6
has a random term at the end, when you start synthesizing over a
number of years you are bound to get negative flows. What do you
do in this instance?
Dr. Fiering: You are quite right. In the actual simulation runs
that we studied in detail we did get negative flows. Let me answer
the question first by making the simple assumption that all the data
might be log transforms: then the negative values will cause no
trouble. In the more general case in which there is no log transform,
we made a study of exactly how much error is involved in truncating
the distribution at zero and throwing away all the negative values.
It turned out to be less than one-tenth of one percent, so that for
the particular values of mean and standard deviation that we en-
countered this was not a damaging practice. But you are quite right,
this is a conceptual difficulty with the model, particularly since one
must admit that the normal distribution goes to infinity at both ends
and can generate negative results.
Mr. Deininger: Well, if you truncate these low flows, you could
truncate some of the high flows, too.
Dr. Fiering: We tried that.
Mr. Deininger: The question is, how accurate is this means of
forecasting? I think you are really back to the old technique of
shuffling cards.
"58
DISCUSSION
-------
Dr. Fiering: We tried that, too. But when you shuffle cards,
you are constrained to getting the same numbers again and again
and you can never get anything lower than an ace or higher than a
king. The definitive answer for the very difficult problem of how to
reproduce, or synthesize, possible sequences from any distribution
that has extremes higher or lower than the recorded values has never
been found. I am merely suggesting this as a possibility, and an
empirical study of the errors introduced by truncation indicates that
our errors are extremely small. We wish we had the answer to the
general case.
From the Floor
259
-------
OBJECTIVES AND CRITERIA FOR
WATER POLLUTION CONTROL
James B. Coulter
Sanitary Engineer Director, Technical Services Branch, Division of
Water Supply and Pollution Control, U. S. Public Health Service
Washington, D. C.
Because water is a universal solvent, in situ it always contains
impurities. An impurity is a pollutant when it becomes objectionable
or in any other way interferes with man's use of water. Pollutants
may be physical, chemical, or biological; and they may stem from
manmade or natural sources. The Federal Water Pollution Control
Program is directed to the elimination of pollution and the mainte-
nance of surface and ground water in an unpolluted condition.
For the discussion that follows, a distinction is made between
water quality objectives and water quality criteria. Water quality
objectives are goals that take the form of statements of intent to
protect water uses and preserve the enjoyment of clean water. Con-
trary to thinking in technical circles, the objectives of the Federal
program are established by the Congress in response to the desires
of the voting public. Some of the objectives are expressed as public
wants and transcend economic analysis or professional preference.
The Federal Water Pollution Control Act contains the objectives
of the Federal program, and it is clear that instream as well as with-
drawal uses are to be protected. Instream uses include recreation,
propagation of fish and wildlife, and preservation of the natural
appeal of the stream. Benefits resulting from instream uses are
sometimes intangible from the dollar-value standpoint, but they are
nonetheless legitimate objectives with high priority in a water quality
management program. It is not uncommon that quality objectives
established to protect instream uses will be higher than those re-
quired to protect withdrawal uses. For instance, the bacterial quality
required of water for swimming is higher than that for a raw drinking
water supply.
Withdrawal uses include domestic supply, industrial processing
cooling, irrigation, and all other uses for which water is taken from
the stream. Withdrawal uses generally have an associated return
flow with altered quality characteristics.
Water quality objectives must be reduced to an operational pro-
gram; and, for that purpose, criteria are established. At times criteria
and objectives are synonymous but more frequently a criterion can
be measured while an objective represents a general condition. A
criterion should provide one connecting link between water uses and
the control of pollution. It should be operational in the sense that a
change in the criterion will produce a predictable change in oppor-
tunity or cost of water use.
Unlike objectives, water quality criteria are established by water
pollution control administrators and their technically trained staffs.
Coulter
261
-------
The profession must be constantly aware of two charges. First,
quality criteria must reflect the intent of water quality objectives;
and second, a cost-benefit ratio is imputed when a water quality
criterion is establishedO).
Water quality criteria are commonly selected through one of
three processes: edict, judgment, or (not so commonly) systems
analysis. Edicts follow clearly expressed public wants stemming from
widespread desires or fears. Fish kilis are unacceptable; therefore,
materials toxic to fish are banned without hesitancy. Propagation of
fish and a healthy As*1 environment ar^ expressed as public wants, and
desirable levels of beneficial substances as well as tolerable limits
for potentially harmful substances are established. Unfavorable com-
munity reaction to visual evidence of human contamination results
in edicts prohibiting floating refuse or recognizable objects of human
origin. Fear of death or disease is an over-riding motivation that
takes the form of highly restrictive criteria that limit concentrations
of toxic substances or other materials suspected of being a threat to
public health. Oil slicks, floating scums, and color- or odor-producing
compounds are usually controlled by edict, although the same results
might be indicated by economic analysis.
Judgment is by far the most common mechanism in current use
for establishing water quality criteria. In arriving at a particular
criterion, many things are considered. First of all, the water uses
that might be affected are identified and then the effect on each use
is examined. A search is made for existing information, and studies
are made to determine the threshold concentration at which the con-
tamination interferes with the water use under consideration. The
investigation is then extended to determine how frequently interfer-
ence with the use will be tolerated and the nature of damages that
will result as the period of substandard conditions is lengthened.
With these factors in mind, the examination turns to the possi-
bilities for control or treatment. Often the criteria are established
from a judgment of what can reasonably be expected in the way of
control. For instance, it is known that uranium in drinking water is
harmful to humans in any concentration. Therefore, water pollution
control agencies attempt to keep uranium in drinking water supplies
at the lowest practicable level. The extensive investigations of the
Public Health Service undertaken as an enforcement action at the
request of, and in cooperation with, the seven States of the Colorado
River Basin revealed that proper control measures at uranium mill-
ing plants would keep the radioactivity from radium-226 in the
Colorado River and its tributaries below 1 micromicrocurie per liter.
Even though the previously accepted level was 3 micromicrocuries
per liter, the seven States, the Atomic Energy Commission, and the
Public Health Service agreed to the limit of 1 micromicrocurie per
liter because in their judgment that level could be obtained without
undue penalty to the uranium milling industry.
Another example is that of pathogenic organisms. Without argu-
ing the merits of the relationship, if one accepts the philosophy that
OBJECTIVES AND CRITERIA FOR
262 WATER POLLUTION CONTROL
-------
indicator organisms are a parametric measure of the density of patho-
genic organisms, the risk to public health is reduced as the density
of coliform organisms is decreased. Water pollution authorities at-
tempt to maintain a low coliform density at bathing beaches. A
density of 1000 per 100 milliliters is commonly accepted because
judgment has demonstrated that this level can be met and that the
public health hazard is not too great at that level. However, few
health authorities would fail to require a higher quality and thus
secure greater protection if it were practical and possible to do so.
In effect, the quality criterion is established from a judgment of levels
that can be obtained through reasonable control measures.
Summarizing, the judgment process includes the effect of a con-
taminant on water use, the frequency of occurrence of substandard
conditions that will be tolerated, and the length or duration of any
one substandard period that can be tolerated without severe penalty.
This knowledge is balanced against a consideration of the effectiveness
and cost of practical control measures. Thus, a criterion for chlorides
might be expressed as a concentration no greater than 100 milligrams
per liter, not to be permitted for a period longer than 7 consecutive
days, with an expected occurrence interval of no less than once in 10
years. The prudent water pollution control authority would assure
himself that the criterion was reasonable and could be met without
excessive cost before he would attempt to put it into effect.
It was stated earlier that a cost-benefit ratio is imputed when a
water quality criterion is established. Thomas(I) illustrated this
important fact in an analogy using an "Animal Farm." In that illus-
tration, the farmer's daughter solved her father's dilemma — to treat
or not to treat — by deriving the equation:
V
X ==.
KBNU
Where:
X(. is the critical concentration (quality criterion).
V is the cost of treatment.
K is an infectivity parameter.
B is the discounting function used to bring a series of future
benefits to a present value.
N is the number of animals.
U is the profit per animal.
In the right hand member of the equation, the numerator is the
cost and the denominator is the benefit. The equation is general and
applies to a wide range of water quality decisions.
When an administrator uses judgment to establish a water qual-
ity criterion, he intuitively balances the reduction in damages against
the cost of preventing damages. Systems analysis provides a meth-
odology for quantitating the decision-making process. By skillful
analysis, the administrator can arrive at the optimum set of criteria
for increasing regional income gains, and he can demonstrate the
Coulter
263
-------
penalty of choosing something less than optimum to satisfy another
community objective.
Before a general procedure is described for a systems approach
to the selection of water quality cr1teria> some general comments are
in order. First, the technique for analysis need not be more refined
than the data going into the analysis; but, conversely, an outline and
general formulation °f the systems calculations will show the tvDe
of data needed, and, therefore, the kind of studies required This
point is of special importance to the state and Federal agencies en-
trusted with public funds to develop effective comprehensive water
pollution control programs A large Part of that effort shouJd gQ .
studies to obtain the operational data needed for systems analysis.
Another consideration is that the systems approach will fall into
disrepute (as well it should) if mstream uses and intangible benefits
are ignored. Because these uses and benefits are difficult to measure
there may be a tendency for the analyst to pay them lip service and
base his calculations and final conclusion on the easier-to-measure
withdrawal uses and other tangible benefits. One way to avoid this
is to accept quality standards that wm protect the propagation of
fish and wildlife and other mstream uses as constraints on the system
Another way is to expend more energy in water quality studies to
arrive at a good means of ascribing "shadow costs" or in some other
way of measuring the utility of intangible benefits.
One final note "the external economies and diseconomies of
changes in water quality are diffused and ill-defined. Effects on a
specific manufacturing process or a municipal water supply are often
apparent; but effects on, say, hot water heaters in tens of thousands
of homes may go virtually unnoticed. The effect on any individual
home owner might be small, but the multiplication factor is very
large. Furthermore, water quality changes affect the community of
water users as a whole. Because of the community nature of the
market, marketplace concepts specifically the willingness to pav
cannot be relied on as a method for establishing relative values.
The foregoing comments might lead to the conclusion that the
problem is so ill-defined and so full of imponderables that experience
and skilled judgment will never be replaced by systems analysis.
This is true; skillful analysis can only complement experience and
judgment. On the other hand, systems analysis is the only means
for quantitativeiy forecasting the changes in regional income gains
that will be brought about by changes in the selection of water qualitv
criteria. J
The selection of a water quality criterion through systems analy-
sis is a joint economics-engineering venture. The pattern is the same
countrywide, but the specifics can only be supplied with reference to
place, time, and season in a particular river basin. The steps are:
1. Determine potential water uses.
2. Construct a use-concentration spectrum.
264
OBJECTIVES AND CRITERIA FOR
WATER POLLUTION CONTROL
-------
3. Construct a net benefit - concentration chart.
4. Determine the marginal loss function.
5. Construct a cost of quality improvement-concentration chart.
6. Determine the marginal cost of quality improvement.
7. Determine concentration where marginal loss and marginal
cost are equal, i.e., indicated quality criterion.
8. Compare indicated criterion with constraint level and select
the one representing the better quality of water.
The determination of potential water uses is made after an eval-
uation of resources and other economic factors. The economic study
required to determine the potential for present and future water uses
is beyond the scope of this paper.
The use-concentration spectrum may be shown in tabular form,
but it is often convenient to show it on semilogarithmic graph paper.
Concentration is marked off so that it increases from left to right.
Each water use that is influenced by the contaminant is shown above
the concentration where the use first becomes practical. Thus, all
uses to the right of any point on the spectrum can be served by the
quality at that point.
The net benefit - concentration chart is constructed by moving
from right to left on the use-concentration spectrum and plotting the
cumulative net income increase above each use. The technique of
determining gains in net benefits is beyond the scope of this paper,
but a general discussion may be helpful.
One consideration is the scarcity of suitable quality water. For
instance, as an extreme example, consider an area with underdevel-
oped resources in a good economic location. Water is plentiful but
so highly saline that economic development is stymied. In this case,
the analyst is justified in adding total net benefits after adjustment
for opportunity shifts as he considers each new use moving from
right to left on the use-concentration spectrum. In the more com-
mon situation where alternative sources of good water are available,
he may take credit only for increments of net benefits added by in-
creased opportunity for instream uses and for savings that will result
from using the improved water rather than water from an alternate
source.
The cumulative net benefits at the left of the net benefit - con-
centration chart show the benefits that will accrue as water quality
is made very good with respect to the contaminant under considera-
tion. The difference between the maximum level of benefits and the
level of benefits at increasing concentrations represents a penalty or
loss that the community suffers as a result of poor quality water. On
this basis an overall loss function can be derived. The difference be-
tween the cumulative net benefits at any concentration and the total
benefits at the left of the chart is plotted at each concentration. This
will result in a curve showing small losses for good quality water and
increasing losses as the quality of water deteriorates.
Coulter
265
-------
The marginal loss function may be defined as the first derivative
of the net benefit curve with a change of sign. A generalized equation
can be fitted to the loss function and its first derivative plotted, or the
A Loss
change in loss, can Picke<* off the loss curve at a
number of concentrations and plotted above the concentration to give
the locus of a curve.
The cost of quality improvement - concentration curve is based
on engineering estimates of the effectiveness and cost of control
measures needed to bring water from the quality in the stream to a
target quality. The analysis is not concerned with the one control
measure that will provide the highest quality of water. Rather it is
concerned with the optimum combination of feasible control measures
that wiil produce the target quality with minimum total cost. Cost-
concentration curves require investigation and analysis with respect
to a particular year, season, and place in a specific river system.
Within the scope of the Federal Water Pollution Control program,
there are three basic measures for water pollution control: in-plant
modifications to keep waste out of water or to regulate the flow of
waste waters, collection and treatment of waste waters to remove
pollutants, and the regulation of stream flow for quality control. To
this list must be added a fourth possibility for cost studies, treatment
of water taken from the stream to make it suitable for a specific use.
Each of these has associated initial, operational, maintenance, and
replacement costs. The optimum arrangement for a given quality
objective is the combination that will give the smallest sum of the
present values of all costs.
The Federal Water Pollution Control Act places a constraint on
the use of flow regulation. The Act stipulates that storage in Federal
reservoirs for streamflow regulation for quality control will not be
provided as a substitute for adequate treatment at the source. This
restriction is in keeping with public policy recorded at the President's
Conference on Water Pollution Control(2). Many comments have
been made regarding the economic inefficiency of the constraint; how-
ever, it remains to be seen whether the constraint is inefficient. In
the first place, streamflow regulation is a costly control measure that
is most effective at low flows. The stream flow must be doubled to
halve the concentration of a stable pollutant. Furthermore, the ground
rule that each water user bear his own costs by providing treatment
before discharging waste water makes more than a little sense from
the cost distribution standpoint.
Under the Federal program the following scheme is used to arrive
at a combination of control measures. Treatment at the source and
in-plant modifications are required to the extent of the polluter's
ability to control pollution. Streamflow regulation is provided from
storage in Federal reservoirs to obtain further improvement if needed.
Process water drawn from the stream is treated to meet the specifica-
tions for a particular withdrawal use.
Within this scheme the cost of the optimum arrangement is found
266
OBJECTIVES AND CRITERIA FOR
WATER POLLUTION CONTROL
-------
by identifying the combination that will give the minimum total cost
to bring the water to a target quality. The total cost is the sum of the
initial cost, all recurring annual maintenance and operating costs,
and periodic replacement costs, all discounted to a present value.
CT-C, + C. [ ' ~ (1r+rrT ] +c-
where:
CT is the present value of all costs during time T.
Ct is the initial cost.
Cu is the recurring annual maintenance and operating costs.
r is the discount rate.
T is the economic time horizon.
Cr is the present value of periodic equipment and replacement
costs.
The cost of quality improvement - concentration curve starts at
zero at the concentration of the pollutant that would be found in the
raw river water if no control measures were applied. The locus of
the cost - concentration curve is established as the optimum cost of
control is found for each target quality level. The marginal cost curve
is the first derivative of the cost - concentration curve.
The marginal loss function and the marginal cost of optimum
measures for controlling pollution are compared to arrive at the indi-
cated water quality criterion. The indicated criterion is the concen-
tration where the added cost of another increment of improvement
would not provide an equal amount of additional net benefits, but a
reduction in cost of treatment would result in a loss in net benefits
greater than the reduction in cost. This is the quality where the first
derivatives of cost and benefits are equal.
The indicated criterion is then compared with the concentration
needed to protect instream uses or other intangible conditions that
have been established as public wants. The criterion representing the
higher quality is selected. If the constraint placed on the system by
public wants results in an economic penalty, the analyst must accept
the fact that optimizing net benefits may not be the over-riding
objective. He is, however, in an excellent position to demonstrate
to the public the cost of satisfying its stated want.
Thus far, the return period and the intensity and duration of
adverse conditions have not been brought into the picture. These
facets are considered in arriving at optimum costs of treatment,
especially with respect to streamflow regulation. The return period,
intensity, and duration of adverse conditions are related to water
uses. Their selection will depend somewhat on the severity of the
penalty when the water quality objective is violated. For instance,
it might take years for a stream to recover from a violation that
caused a general fish kill and destruction of other aquatic life. This
event would be assigned a much longer return interval than an event
Coulter
267
-------
that would cause a slight nuisance 0( short duratl0n similarty
damages were imposed on a larfier cit m. lndustri c0£',* ""f
longer return interval would be assigned than if the affected reach
of the stream were used less intensively leach
In summary, the elements 0f Investigation and analysis reauired
to establish a cr. tenon to ma*imi2e net bcnems have bM, brieflv
Establishing criteria that will SS
Experience and judgment must be fortified with h ' • analysis,
models and systems-type analyses thit r(1(,n„n' ecision-makmg
public wants and include both tangible and intangible"benem^ a°n
rr^^trcontrofst'udfes ^Ld'b^evTd' ?*" ^ T
j * ^ ^ i ^uvuia oe aevoted to assembly of Hntn
and perfection of analysis techniques that ran h* k„Z /
qlXtritfrt""'"40''5 eSlaWish COS'-b
-------
quate treatment or other methods of cantroiling waste at the source."
It also requires thai the value of the storage for such purposes be
determined and be taken into account in appraising the -economic
value of the entire project, and in the functional allocation of project
costs. Finally^ tlie Act states that the beneficiaries of such flow aug-
mentation stetage ar« la Iw identified., ar.d "if the benefits are wide-
spread ntr ruitioti£l in the costs of such features shall bfc noc-
reitnburssble." These several provisions have led to increased interest
in a quantitative approach to water quality control criteria.
Tba policy decisions pcse
-------
is much higher than the analyst can comprehend. This is one way of
saying that enactment of laws in a free society represents one type of
systems analysis.
Similarly, analysis might sh0w that optimum development is
considerably beyond the level of development that the public sub
sequently supports. Such a situation may not mean the analysis is
wrong or that society disagrees with the solution. Society may how
ever, be forced to settle for something less than optimum ' water
control, or change the time scale fQr achieving optimum control be-
cause it must forever grapple with the problem of optimizing its
capital resources among many needs of which water is but one
These remarks are not offered to oppose the author's plea for
quantitative analysis or to deny the potential advantages that systems
analysis may one day give to the administrator in this field Thev
are intended to strike a word of caution. We must always remember
that the high-speed computer can only search out the best solution
based on the many human and often nebulous value judgments that
were inserted as input data.
This leads to the second important qualification cited by the
author. There is need for a great deal of research effort before reli-
ably useful solutions can be developed. Recitation of just four of the
factors needed in the way of input (whether based on educated guesti-
mate, factual knowledge, or hope) to perform the analysis outlined
by the author will illustrate: u
1. Prediction of the water-use pattern.
2. Determination of the benefits that accrue from each use
category.
3. Establishment of the cost of improvements needed (in
turn dependent on the appraisal of the future use pattern).
4. Assignment of a time scale to each of the foregoing items.
Based on a recognition that there is still ample opportunity for
disagreement on such questions as the rate of population growth to
be expected in a given area, that shifts in employment opportunity
will take place with changing technology, and that advancements will
be made in waste treatment technology, the problem is rather awe-
some. It becomes extremely sticky and shot with fantasy when we
face the question of project analysis based on a 100-year life expect-
ancy. As an academic venture the problem is most intriguing but
until research improves present knowledge and techniques appre-
ciably apphca ion of systems analysis to such policy decisions as
allocation of storage and assignment of financial responsibility re-
mains highly dangerous.
Recognition of this situation provides little satisfaction to the
administrator who must make decisions today within the framework
of existing statutes. He has to establish some sort of perspective as
to the long-range objectives of the program he is administering.
Moreover, he has a responsibility to develop administrative policies
270
DISCUSSION
-------
and procedures that will lead toward attainment of these objectives,
but not to premature consecration of policy decisions in areas where
current appraisal is mast problematical.
Perhaps this is an appropriate place to summarize the possible
role cf dilutior. in waste disposal. Adequate treatment for dagradaale
wastes should be viewed with respect to downstream water us.es.
If, for example, the objective a low coliform count m the receiving
water, secondary treatment followed by chlorination will achieve
this gcal at nr.j^l. less cost than will diljticri
If oxygen balance is the critical factor, it seems doubtful whether
any water should be released to add oxygen until a]] wastes have
received treatment to remove at least 90 to 95 percent of the BOD.
Dilution water can be most easily justified where it is used to
dilute dissolved minerals. Constituents such as nitrates, boron, chlo-
rides, sulfates, phosphates, and carbonates may come from either
natural sources or manmade pollution. To an appreciable degree they
car be controlled at the point of origin in man's activities, but contri-
butions of these constituents from natural sources are often very
costly if not impossible to keep out of streams.
Also, mention should be made that one of the more recent meas-
ures of pollution is given by carbon filter chloroform and alcohol
extracts. These substances have a high taste and odor potential and
some may be carcinogenic in concentrated form. They represent a
class of toxic and organoleptic substances, and include such com-
pounds as the organic pesticides, waste components from organic
manufacturing processes, ar.d some organolytics of natural origin.
These substances are rot removed by eorventiajial waste treatment
processes. Until better methods are developed for determining their
actual nature, and until processes are designed to remove them from
waste streams, dilution may be the only practical solution to the prob-
lem of keeping levels reasonably law in areas of high chemical
manufacturing intensity.
How do the objectives of the 1961 amendments of the Federal
Water Pollution Control Act compare with this general physical
outlook? I have r.c way of knowing the exact objectives the drafters
cf :rn 1&63 amendments had in ,r.i id My interpretsticu af the sig-
nificance of fttese amendments, in terms of 3 public statement of wants
is as. (3.1;w. Water quality cantrcl is of auHteient importance that
some of the available water of good quality and some of the potential
storage resources in this country should be allocated to flow supple-
mentation needs in the interest of (1) achieving improved water
quality, (2) ensuring full development of the limited available sup-
plies of good water and effective reservoir sites, (3.) determining the
economic feasibility of multipurpose reservoir projects, and (4) es-
tablishing an additional basis for public investment in water resources
development programs. Each of these may prove to be a most desir-
able social objective if comprehensive development of the water
resources of this country is to be achieved. It is inconceivable, how-
ever, that anyone can comprehend what the ultimate cost of fulfilling
Smith
271
-------
these wants will be or what adjustments their fulfillment will require
in both intra- and inter-governmental institutional arrangements
regarding the control and use of water. However, if the implementa
tion of the Act is to be based on an objective of optimum economic
return as outlined by Mr. Coulter, if the restraint that dilution shall
not be a substitute for treatment is to be retained, if reasonable equity
among water users is to prevail, and if today's actions are to stand
the test of time, one can visualize the desirability of some rather
conservative approaches to initial policies. Some thoughts that come
to mind are as follows:
1. Adequate treatment for degradable wastes should be de-
pendent on stream uses, but the use of water for dilution
of such wastes should not be required until at least 90
percent BOU reduction is achieved. How much addi-
tional treatment is required should be dependent on
analysis of treatment costs versus storage costs.
2. Recommendations for the development of flow augmenta-
tion storage on a nonreimbursable basis for the dilution of
inorganic or nondegradable wastes should be limited to
that amoun of storage needed to dilute natural pollutants
plus 50 percent of the manmade contributions. (The 50
percent is arbitrarily chosen, the point being that some
sort of limi ation is needed if reasonable equity is to be
secured among water users m different geographic areas
and if the use of storage as a substitute for treatment or
other management techniques is not to be abused. Note
should also e made that such a policy would require con-
current action under the enforcement authorities of state
and federal pollution control laws.)
3. Projects recommended for construction when their bene-
fit-cost ratio, in the absence of flow augmentation benefits,
is less than unity should be assigned a priority in keeping
with their relative need for flow augmentation.
4. Whenever conservation storage is provided in a reservoir,
whether it be for flow augmentation or for other purposes',
such storage should be developed, within the physical
limitations of the site, to the point where the marginal
cost per unit of yield is no longer competitive with the
average of alternative storage opportunities in the same
general region.
5. Any authorization of storage for future flow augmentation
purposes should carry the provision that the storage may
be reallocated to other water uses provided such realloca-
tion will not adversely affect water quality and provided
the users thereof agree to repay appropriate portions of
the storage cost. Concurrently, all water users would
have to be prohibited from using releases made for flow
augmentation purposes unless they had assumed a pro-
portionate share of the storage costs.
272
DISCUSSION
-------
Some of these suggestions may be highly unpopular with the
public and some may even be contrary to the objectives of the 1961
Act. They are mentioned, however, to point up some of the ramifica-
tions of streamflow regulation for quality control that have not been
discussed at this meeting. They also suggest one way of seeking a
quantitative evaluation (in the democratic sense) of the public's
desire for flow augmentation. Their adoption would minimize the
need for compromising professional judgment and would help provide
opportunity to conduct the research needed to bridge some of the
present gaps in knowledge.
DISCUSSION FROM THE FLOOR
Mr. Lebosquet, U. S. Public Health Sermce: I thought 1 might
mention a few things to back up Jim Coulter in this particular ap-
proach. This is perhaps not as new as some of you might think. We
have been writing reports and estimating economic benefits from flow
augmentation since about 1936. The first report was in connection
with the Allegheny Reservoir, which finally is about to be built.
That report was signed by Streeter, Hoskins, and Crohurst — the
only document I have ever seen that was signed by these three
eminent engineers in the Public Health Service. From around 1941
until P. L. 87-88 was passed, probably a hundred reports were
written. Since then very, very many reports have been made. So,
there is quite a lot of background.
I would like to second Jim Coulter's remark that low-flow regu-
lation is a very expensive method of correcting pollution, even when
the cost of this water is based on the most economical project. Even
if there are 10 sources of pollution in a row so that 10 cities benefit,
it is still very expensive. In all of these analyses, water quality is
always considered with other multipurpose features. There might be,
for instance, a power development; we would be interested in whether
there is enough water from this power development for pallulion
control. Perhaps reregulation would be needed; the cost of reregula-
turn would be only a small part of the overall project cost.
While many cf the projects that have been built were partJy
justified by this low-flow augmentation, they have depended largely
on these other benefits for the basic justification. Nevertheless, low-
flow augmentation may add just that little bit the Corps of Engineers
or the Bureau of Reclamation might need to push a project over the
1:1 benefit: cost ratio that Congress thinks we should have on these
water resources projects. You should understand we are not trying
to justify projects for low-flow augmentation alone; there are gener-
ally many purposes. I think the project that came closest to being
strictly for low-flow regulation was Berlin Reservoir on the Mahoning
River in Ohio. This project was built during World War II. We
needed steel, and Youngstown needed cooling water to operate at full
capacity; so we were, what I call, sharpening the pencil. There were
From the Floor
273
-------
benefits from cleanliness, from hardness and acid reduction, from
cooling water use — this was the big point — and finally, from
organic pollution abatement.
These evaluations have become very complex. We have been
struggling with them for a long time, and have used various formulas.
Now we have a mathematical model — new words, but pretty much
the same story. Sometimes the computations can get quite complex,
and I am happy to turn over the future computations in this field to
younger people who understand computers.
Mr. Stoevener, Oregon State University: I would like to make
one brief comment on the model outlined by Mr. Coulter. I think the
difficulties with it, which were pointed out earlier by Prof. Velz and
now by Mr. Coulter and Prof. Smith, perhaps originate from the lack
of making one distinction. This model as it is outlined here has one
objective, that is, maximizing net benefits or, from a general economic
point of view, maximizing income. It does not make any allowances
for something else, which is very dear to most of our hearts, and that
is income distribution, which means, who is paying for what, and who,
in particular, is getting the benefits?
This also underlies the difficulties and the paradox pointed out
by Prof. Velz this morning in the Federal Water Pollution Control
Act. It is a part of the institutional framework under which we
operate and should not be confused with the objectives that a model
like this may point out. At best, if we disregard the institutional
framework, we can only come up with some opportunity cost of ful-
filling some of these criteria that may have been set.
Dr. Velz, University of Michigan: I want to make two very brief
statements. First, we can never charge all the costs of low-flow
regulation against waste treatment responsibilities. A lot of other
benefits accrue from flow regulation, and if we try to justify flow
regulation on the benefits of reduced treatment alone, we never will
get it. I firmly believe there are broad benefits to flow regulation.
If we can get around the dollar sign with which people insist on
evaluating these things, we will get out of this box. For the last 50
years we have been living under the bushel of natural drought con-
ditions and making all human activity adhere to the least flow that
goes down the river.
The second point concerns the four approaches suggested by
Mr. Coulter. I agree with every one of them, but I don't agree with
putting them into effect in layers. Let us put these together and
consider all possible ways of solution to arrive at the best combina-
tion of waste control at the source, waste treatment, flow regulation,
and water treatment for specific uses. It seems to me this is where we
fall down in our system analysis approach. The cross combinations
of these things must be brought into play; not one versus the other,
but rather an intelligent give and take among all of the practical
ways of getting at this problem. Let's not go whole hog in any one
direction, but see what we can get the most out of for the least
number of dollars.
274
DISCUSSION
-------
Mr. Smith, University of Kansas: I agree wholeheartedly with
some of the concepts outlined by Prof. Velz. I might point out that
this situation has existed in the State of Kansas in very recent years.
Since 1958 we have amended our State constitution to allow the State
to be a partner to development. As a result of a mathematical model
and analysis of flow considerations, we decided that we did need much
flow supplementation. We also concluded that this was not a cost
we could pinpoint too well, so it became a question, a broader ques-
tion, of political economy. We subsequently recommended to the
State legislature that the State take an initial hand in this, and by
concurrent resolution 2 years ago they indicated their desire to pro-
ceed. This past year a law was finally enacted to allow the Kansas
Water Resources Board to begin negotiation of contracts with Federal
agencies.
From the Floor
275
-------
Session 5
MEASURES FOR .IMPROVING
THE QUALITY OF RESERVOIR DISCHARGES
Moderator: R. S. Ingols
Georgia Institute of Technology
-------
IMPROVEMENT OF THE QUALITY OF RESERVOIR
DISCHARGES THROUGH CONTROL OF
DISCHARGE ELEVATION
W. E. Knight
Division of Stream Sanitation and Hydrology
North Carolina Department of Water Resources, Raleigh
Next to the air we breathe, water is the most important sub-
stance that we use. All plant and animal life is sustained by it. Men
have fought and died for water; civilizations have perished for lack
of it.
Man's demands for water have increased as our civilization has
progressed. Per capita use of water in 1900 was 50 gallons per day;
by 1940 it was 122 gallons per day; and by 1960, 150 gallons per day.
The use of water by agriculture and industry has increased at a more
rapid rate. Yet man's total supply of water is a relatively fixed
quantity.
The growing demand for water has directed more and more
attention to its conservation. One obvious means of stretching the
available supply is to prevent its waste during periods of high runoff
by storing it in reservoirs. The idea of storage is, of course, far
from new; however, the use of reservoirs as a means of conserving
water is certain to increase in the future.
The increased use of electric power has already resulted in many
of our rivers being developed to their potential for the production
of hydroelectric power. Reservoirs for flood control and low-flow
augmentation have been constructed and others are being planned.
Intelligent planning of water resources conservation and devel-
opment requires that releases from reservoirs be adequate in both
quantity and quality to serve essential uses.
When water is stored in reservoirs, changes take place in prac-
tically all aspects of its quality. Fortunately, most of these changes
result in improvements in quality. For some uses, however, the
changes are detrimental.
The effects of storage on water quality have been the subject
of much study by Federal and State agencies. Thermal stratification
of lakes has been recognized by man since he first dived into a lake
and found the water near the bottom colder than that at the surface.
Likewise, the fact that at times water near the bottom of a reservoir
contains little, if any, dissolved oxygen has long been known. Studies
of the concentration of dissolved oxygen in water discharged from
reservoirs indicate that during the critical summer period, dams with
deep intakes discharged water of very low dissolved oxygen content,
dams with intermediate intakes discharged water of higher dissolved
oxygen content, and dams with high-level intakes discharged water
with even higher dissolved oxygen content.
Knight
279
-------
One of the major uses of our streams is the carriage and absorn
tion of wastewater. This must remain a principal function of streams
but if the fullest development of the water resources is to be made
while adequate water quality is maintained, the degree of waste
carriage and absorption must be reduced through waste treatment.
Of the many and often conflicting uses made of our streams onlv
two, the discharge of water from deep reservoirs through low-level
intakes and the absorption of municipal and industrial wastewater
exert any significantly detrimental effects on the dissolved oxygen
assets of streams. Waste absorption requires the presence of dissolved
oxygen in the stream water, whereas the discharge of water from
reservoirs through low-level intakes reduces the quantity of dis
solved oxygen available for the absorption of wastes. From the stand
point of effects on dissolved oxygen assets, reservoirs that discharge
water low in dissolved oxygen are similar to sources of organic
pollution.
There are many well-established principles and practices, result-
ing from years of study and experience, that may be employed to
reduce the oxygen demand of organic wastes. In the case of a reser-
voir used to produce hydroelectric power, however, efficiency of oper-
ation requires that water enter the turbines from the lowest possible
level. Unfortunately, there are no well-established procedures re-
sulting from years of experience, for restoring the dissolved oxygen
content of the discharge immediately downstream from the dam. It
is possible, however, to discharge water relatively high in dissolved
oxygen content by controlling the level from which water is selected
for discharge without at the same time producing appreciable head
loss.
The Roanoke River in North Carolina has been developed to its
potential for the production of hydroelectric power. The river also
serves as the source of water supply for municipalities and industries
throughout its length. It is the vehicle for the disposal of municipal
and industrial wastes. It is an important fishing stream used exten
sively by sport and commercial fishermen and is the principal spawn"
ing ground for striped bass (Figure 1).
In 1952, the John H. Kerr Reservoir was completed and power
production was begun in November. Kerr Reservoir, a Corps of
Engineers development located in Virginia 18 miles above the North
Carolina State Line at river mile 179, is a multipurpose project that
operates primarily for flood control and peaking-power production
At maximum power pool elevation (300 feet), the lake is 100 feet
deep at the dam and has a surface area of 48,900 acres. Turbine in-
takes for the six generating units are from elevation 204.5 to eleva-
tion 265. The project is also responsible for maintaining a schedule
of minimum flows during off-peak hours (Figure 2).
Even though thermal stratification occurred during summer
months, resulting in the discharge of water low in dissolved oxygen
reaeration over some 44 miles of natural stream bed resulted in water
approximately 90 percent saturated with dissolved oxygen at the first
point of organic wastes discharge.
improvement through control of
280 DISCHARGE ELEVATION
C5FO 82 1—740—1 O
-------
Figure 1 — Lower Roanoke River.
In 1953, in keeping with provisions of the North Carolina law
relating to stream sanitation (enacted in 1951), studies were con-
ducted and stream classifications were assigned. The organic waste
assimilative capacity of the segment of the river receiving organic
wastes was calculated based on an upstream dissolved oxygen con-
centration of 90 percent of saturation at 25 °C and the weekday
minimum flow of 2000 cfs as set forth in the minimum flow schedule
governing the operation of the Kerr project.
The waste assimilative capacity was allocated to the various
polluters and a comprehensive plan was developed for the abatement
of pollution in the lower river.
In 1955, the Virginia Electric and Power Company placed its
Roanoke Rapids project in operation (at river mile 137), and the
Company's application for a license to construct the Gaston Dam (at
river mile 146) was pending action by the Federal Power Commis-
sion. The schedule for minimum flow releases from the Roanoke
Rapids Dam was the same as that for the John H. Kerr Reservoir.
Unlike the John H. Kerr Reservoir, the Roanoke Rapids Reservoir
is relatively small. The dam, which is located just upstream from
the town of Roanoke Rapids, creates a lake approximately 9 miles
long with a surface area of about 4900 acres at normal full-power
pool (elevation 132). The reservoir is approximately 62 feet deep
near the dam.
Knight
281
-------
As in Kerr Reservoir dissolved oxygen deficiencies developed in
the lower depths of the Roanoke RaniHo j • vtloPea
months. The discharge of water through the low WpW S Aumn?er
takes resulted in a rigDiflrant reduction and substantial variations'^
dissolved1 oxygen concentrations downstream from the dam studies
conducted during the summer of 1955 and m(. lndi(;ated
dissolved oxygen content of the discharge water (measured at the
highway bridge approximately j miIe downstream from the daSl
fell as low as 3.2 milhgrams per liter. The average of the dissolved
oxygen data collected over the period was 5 4 milligrams cer ii£r
65 percent of saturation at 25°C. milligrams per liter or
The oxygen reduced the waste assimila-
live capacity below that which had been available at the licensees
fhTre^rr1—^^^
regarding the further effects of another hydroelectric project on'water
quality The of dissolved oxygen to downstream use"
caused the S ate agenc.es concerned with maintenance of water qua™
on'd"d oTygento satiSf^ ^ indUStria'
on dissolved oxyge xo satisty the oxygen demand of its wastes to
remies^ed bv Se the iSnse
requested by the O pany for the construction of its Gaston proiect
Sr,lPtUhePOpro0babfe "'id'h """"
adequately investigated and effective measures could be°tacorpoSted
340
320
_i 300
UJ
2 2801
2 260
w 240
z
220
T
UJ
s
co
<
H
Ui
UJ
200 -
180 -
POWER,
¦ 85,000 acre-ft
z
o
>
- 120 h
POWER, 400,000
acre-tt
I FLOOD
POWER,
1,046,000 acre
DEAD STORAGE,
484,100
acre-ft
140 150 160 ,170 180 190 200 210
MILES ABOVE MOUTH OF ROANOKE RIVER
220
240
Figure 2 — Condensed profile of Lower Roanoke River.
282
IMPROVEMENT THROUGH CONTROL OF
DISCHARGE ELEVATION
-------
in "the licerise :o assure that further seduction ia the dissolved oxygen
in the discharge from the lower reservoir would not occur.
After considerable investigation of possible methods ai remedy-
ing the detrimental effects of impoundments on dissolved oxygen
resources, the Virginia "Etecwic arid Power Company proposed to con-
struct a submerged weir around the Gaston turbine intakes as a
solution to the problem of low dissolved oxygen <55;charge frotn lew-
lev -A ints. The st-btrw rirsd weir would form an underwater barrier
to obstruct passage of deep water into the turbines so that waler
would be selected for discharge from levels above ihe crest oi the vreix
and thereby act as the equivalent of a high-level intake.
Since the concept of a submerged weir to act as a high-level
intake was new, it was believed that the performaEce of such a
device should be verified. Iti artier to determine the effectiveness of
the submerged weir in improving the dissolved oxygen discharged,
Virginia Electric arid Power Company constructed a submerged weir
around the turbine intakes in the Hoanokft Rapids Reservoir in the
summer of IS57.
The weir has two sides extending perpendicular to the face of
ihe dam for a distance oi about 275 fe^t connected by an upstream
side about 525 feet in length; tivjs an enclosed fcrebay is formed
around the turbine intakes. The weir, constructed of crushed stone,
extends upward to within 25 feet of the surface and in effect is an
mtegral pari of the iiytuka system (Figures 3 and. 4),
Rgufe 3 —¦ Pfan of submerged weir — Itocmoke Jtepids italic ri
A comprehensive survey was planned jointly by the interested
State and Federal agencies (Steering Committee for Roanoke River
Studies} and the Virginia Electric and Power Company aiad carried
out during summer and early fall of 1357. This was one of ihe
mcst intensive limnological studies ever performed anywhere; it in-
A
Knight
233
-------
NORMAL TAIL
WATER ELEVATION 55.7 ft
NORMAL Watfo LEVEL ELEVATION 132.0 ft
ELEVATION 107.0 ft
ELEVATION 70.0 ft ^
SUBMERGED ROCK WEIR
LOW TAIL WATER
LEVEL ELEVATION 45.5 ft
Figure 4 — Profile of submerged weir „
Roanoke Rapids hydroelectric station.
volved more than 20 engineers, cu. • . , . ,
samplers representing State and S ' i ^Trino^ogist, and
industries. e agencies as well as two
pies Testations ™
discharge, (2) reaeration over the reach nf H Reservoir
Dam and Roanoke Rapids Reservoir m the mil Tit # between Kerr
Roanoke Rapids Reservoir, (4) diUoi JJ »q y Tater enterine
patterns and water movement in Roanoke Raoid^F temperature
quality of water in the Roanoke Sds T?' (5) the
(6) the quality of water at Highway 48 brfdgl 6' ®nd
Upwards of 15,000 separate chemical and
of water quality were made, consuming spvpmI measurements
hours. No effort will be made here to present or ciimmS ?/ man"
data obtained during the study. Onlv a nart nf ^™manze a11 °f the
data will be discussed. y' Unly a part of tf* more pertinent
In order to understand the perform^™ ,
Roanoke Rapids Reservoir, it is necessary to in
upstream factors. For this reason this diseu^inn u .e cts of
summary of water quality to theVerr WM *
^onPatrSSthehibiSinf *
hypolimnion. This water is verv low in ^rimarily from the
tively low in temperature. Table i chnw
-------
TABLE 1. SUMMARY OF DISSOLVED OXYGEN AND TEMPER-
ATURE SAMPLES AT KERR RESERVOIR TAILRACE
Dissolved oxygen,
mg/1
Temperature,
op
Maximum
4.7
76.3
Upper quartile
3.0
72.6
Weighted average
2.5
71.9
Median
2.3
71.3
Lower quartile
1.7
69.5
Minimum
0.3
64.0
Number of samples
323
318
TABLE. 2. SUMMARY OF DISSOLVED OXYGEN AND TEMPER-
ATURE SAMPLES AT GASTON DAM SITE
Dissolved oxygen,
mg/1
Temperature,
Maximum
8.2
87.0
Upper quartile
7.1
75.7
Arithmetic average
6.6
74.4
Median
6.8
73.9
Lower quartile
6.0
72.9
Minimum
5.2
68.4
Number of samples
303
307
Water entering the headwaters of Roanoke Rapids Reservoir
varies considerably in quality, as observed at the Gaston Dam site
sampling station. Compared to the surface water in the reservoir,
the inflow water is relatively cool, especially during periods follow-
ing heavy discharges from Kerr Reservoir. In the extreme upper
reaches of the reservoir, high velocities and the resulting turbulence
prevent stratification. Eventually, the inflow reaches wider and
deeper sections of the reservoir of such cross-sectional area that
turbulence is no longer sufficient to prevent stratification between
the cool inflow and the warmer surface waters in the reservoir. The
cool water then sinks beneath the surface and becomes a density
underflow.
The inflow continues downstream as a density underflow until
it reaches its own density level, which varies in elevation, depending
on the temperature of the reservoir as well as the thermal structure
within the reservoir. Joining its density layer, the new inflow be-
comes the tail end of that layer. The hydrostatic head of the extreme
upstream end of the inflow, which is at a higher elevation, tends to
compress and thicken the downstream portion, displacing the warmer
waters above upward.
Knight
285
-------
The volume of water that can be discharged from Kerr Reservoir
in a single peaking-power period is large compared to the volume
of Roanoke Rapids Reservoir; therefore, the vertical displacement of
warmer water immediately above the stratum corresponding to a cool
inflow is considerable. Water thus displaced upward may be low
in dissolved oxygen, since it may have been stored at the intermediate
depth for a considerable time with little chance for reaeration. When
water with such low dissolved oxygen is raised to the elevation of the
weir crest or above, it may be withdrawn alone or mixed with other
water from above and below, depending primarily on the Roanoke
Rapids discharge rate. This fact accounts for the occasional observed
periods of relatively low dissolved oxygen in the Roanoke Rapids
Reservoir discharge.
The hydraulic performance of the submerged weir was a subject
of considerable study. This work, conducted by Dr. D. W. Pritchard
and J. H. Carpenter of the Chesapeake Bay Oceanographic Institute,
included measurements of the water movement in the vicinity of the
weir and lower reservoir. Four methods of direct measurement were
employed: (1) Gurley Current Meter, (2) confined drags, (3) free-
drifting drogues, and (4) fluorescent dye tracer.
Data from all four techniques of observing the pattern of water
movement were combined and velocity-versus-depth curves for vari-
ous locations and discharge rates were determined. On the basis of
these curves, the percentage contribution from each 5-foot layer to
the discharge was determined. Table 3, taken from Special Report
No. 1, Roanoke River Studies — A Report of a Study of the Effects
of a Submerged Weir in the Roanoke Rapids Reservoir Upon Down-
stream Water Quality, shows the percent contribution of each 5-foot
layer of the reservoir 500 yards upstream from the weir at the dis-
charge rates shown.
TABLE 3. PERCENT OF DISCHARGE CONTRIBUTED BY EACH
5-FOOT LAYER IN ROANOKE RAPIDS RESERVOIR
Depth interval, Discharge rate, cfs
ft 2,000 6,000 12,000
0-5
0.0
5.0
9.4
5-10
0.0
14.8
21.0
10-15
7.5
23.8
21.3
15-20
32.8
23.8
18.0
20-25
34.3
17.4
13.9
25-30
17.9
9.9
9.4
30-35
6.0
4.2
4.6
35-40
1.5
1.0
1.7
40-45
0.0
0.0
0.7
These data show (1) only a small percent of the flow originates
at depths below 35 feet; (2) the layer from which the maximum
contribution to the flow originates increases in depth with decreasing
286
IMPROVEMENT THROUGH CONTROL OF
DISCHARGE ELEVATION
-------
flow, for 12,000 cfs it is centered at 10 feet, for 6,000 cfs it is centered
at 15 feet, and for 2,000 cfs it is centered at 20 feet; and (3) there is
no significant change in percent contribution with flow for the layers
below 35 feet, but there is a marked increase in percent contribution
for the upper 5 feet with increasing flow.
The final check on the efficiency of the submerged weir in the
Roanoke Rapids Reservoir is provided by the data collected below
the impoundment at the bridge crossing Roanoke River at Highway 48
bridge. During the survey period, this station was sampled a total
of 1092 times, usually for dissolved oxygen, temperature, and turbid-
ity. The results for temperature and dissolved oxygen are summar-
ized in Table 4.
TABLE 4. SUMMARY OF DISSOLVED OXYGEN AND TEMPER-
ATURE SAMPLES AT HIGHWAY 48 BRIDGE
Dissolved oxygen, Temperature,
mg/1 °F
Maximum
8.6
84.2
Arithmetic average
6.3
76.2
Minimum
2.7
71.0
Number of samples
1092
1082
These data show water of relatively good average quality, but
with a wide range of values separating the maximums and minimums,
especially with respect to dissolved oxygen.
An analysis of the data shows that the dissolved oxygen content
of the waters selected by the weir for discharge from the reservoir
is closely related to the rate of discharge from the reservoir. At low
discharge rates, the dissolved oxygen in the discharge is less, other
contributing factors being equal, than for high discharge rates. This
is in keeping with the velocity-versus-depth curves. Under condi-
tions of minimum discharge at Roanoke Rapids Dam (2000 cfs), water
is selected from a narrow range of elevations (centers at 20 feet below
surface) immediately above the weir crest, with little mixing with
the surface layers. The quality of water at this elevation is influenced
by several factors, including the rate of discharge from Kerr Dam
for the preceding several days, the rate of wind mixing and reaeration
in Roanoke Rapids Reservoir in the preceding several days, the rate
of discharge from Roanoke Rapids Dam for several preceding days,
and the degree of thermal stratification above the weir-crest eleva-
tion at the time of discharge. The minimum dissolved oxygen values
observed at Highway 48 bridge represented the worst combination
of these factors.
As another measure of the effectiveness of the submerged weir in
selecting water of relatively good quality from the impoundment, the
data obtained at Highway 48 bridge during the study can be com-
pared with data collected prior to the installation of the weir. Avail-
able data are summarized in Table 5.
Knight
287
-------
TABLE 5. COMPARISON OF ROAivrrwr1
ITY AT HIGHWAY 48 BRIDGE S iflY WATER QUAL-
53, 1956' AND 1957
^^g°Xygen' T^mperature—
r—: "53 I95T1^
Maximum 9.0 c ~
Upper quartile 8.2 6 Q fi' j* 77 0 84.2
Arithmetic average 7.9 5 ' ' 75 2 77.8
Median 7.9 A " ™ 734 76.2
Lower quartile 7.5 4? ' ' 73,4 76-l
Minimum 7.0 3' " "-0 716 7^
Number of samples 47 „ ' ' 71*> 71.0
21 1092 45 21 1082
The 1953 data were collected dnZilTTI . . „ "
August 31; the 1956 data during the Jeriod+A"eust 9 through
tember 21; and the 1957 data during Z ^^ r 1 thr°Ugh Sep~
September 13. mg the penod July ™ through
Comparison of the dissolved owcr^v. i_
oxygen values except the minimum are stenifiSi l^u7 dissolved
corresponding 1956 values, but lower thin the 1Q^ f ^ the
data show a significant improvement in rhco ? H values. These
tration below the Roanoke Rapids Dam over thl T^n COncen~
weir period. The improvement can be attributed t?Ut Pre~
weir surrounding the turbine intakes anrf nm h n!6 su^merSed
of a high-level intake. The survey resulted inTh Tii qUiValent
sions: Y resulted ln the following conclu-
1. The submerged weir in Rnanni«» d
draulically effective in selecting for riiUh686™0^ *S hy~
reservoir, water primarily frnrn the 1 3r^G ,fr0m t*le
crest of the weir. he Iayers above the
2. The weir causes a si£taifi/»a«+ ~
water quality. Becaure lmPr°ve"lent average
capacity of the Roanoke^Lnw/p V<^ Sma" storaBe
the large releases from Sm nrnT''' compared to
leases cause occasional diinln™ *' t owever< such re-
(rom intermediate levelsT water
level of the weir. The i° ?8 /ere above the
becomes available for w?,hH '!"»! SP ^ "PWard 0,6,1
thus selected may °[ the Weir' Water
solved oxygen content undesirably low dis-
3' oTf watS qSy »™ly »Ived the problem
nigh flows below Roanoke Rapids Dam
4' we?i„°Vebpropa^SGa fT*6 RaPid>. voir, the
siened to extend ,,!f Gaston Reservoir should be rede-
instead of 25 ^
288
IMPROVEMENT through control of
DISCHARGE ELEVATION
-------
The proposed Gaston Reservoir (being constructed at river mile
146) will create a lake 34 miles long from the head of Roanoke Rapids
Reservoir to the toe of John H. Kerr Dam. The surface of the normal
full-power pool will be at elevation 200 and will have an area of
20,300 acres. Storage capacity of the power pool will be approximately
400,000 acre feet. The Gaston Reservoir will have a volume approxi-
mately 5.8 times that of Roanoke Rapids Reservoir.
Flow releases from Gaston Dam will be controlled by peaking-
power operation with no requirements for minimum flows and will
be discharged directly into Roanoke Rapids Reservoir. Under normal
operating conditions, the Gaston Reservoir level will vary less than
1 foot.
After Gaston is constructed with a submerged weir extending to
within 15 feet of the surface, water entering Roanoke Rapids Reser-
voir is expected to be of better quality than that available during
the study. With a higher level weir in Gaston Reservoir than in
Roanoke Rapids Reservoir, the quality of the discharge is expected to
be better than that presently observed below Roanoke Rapids Dam.
Should the quality of the discharge from Gaston Dam be no better
than the average observed below Roanoke Rapids Dam, however, this
would mean that water entering Roanoke Rapids Reservoir would be
at a temperature of about 75°F and would have a dissolved oxygen
content of about 6.0 milligrams per liter. Under these conditions, the
problem of poor-quality water being discharged from Roanoke Rapids
as a result of a large discharge of cool water from Kerr Reservoir
should be greatly reduced. After these facts were considered, it was
concluded that "the dissolved oxygen content in the waters released
from Roanoke Rapids will be at least 6.0 mg/1, and in most cases
higher."
Upon completion of the study, consideration was given provisions
that might be included in the Gaston Project license relating to mini-
mum flows and minimum water quality. Provisions acceptable to the
Virginia Electric and Power Company and the agencies concerned
were finally agreed upon. Upon reviewing the Gaston Project, the
Federal Power Commission concluded that "the proposed Gaston
development and the constructed Roanoke Rapids development shall
be considered as units of one complete project." The license for the
Roanoke Rapids Project was amended to include Gaston Dam with a
submerged weir. This license is unique in that it not only established
a schedule of minimum flows to be maintained downstream from
Roanoke Rapids Dam, it also specified minimum water quality in
terms of minimum pounds of dissolved oxygen to be discharged
per day.
The Gaston Dam has been completed, and power production will
start during the Spring of 1963. For this reason, it is too early to draw
conclusions concerning the effectiveness of the two submerged weirs
on water quality downstream from the Roanoke Rapids Dam. Need-
less to say, this will be the subject of considerable study during the
coming summer.
Knight
289
-------
The dissolved oxygen content Q* +>,„ j.
Rapids Reservoir has been monit0rpH . from the ^oanoke
dissolved oxygen analyzer was ittstal1S"JCe " June 1960» a Hays
bridge, and a continuous record °,n ;?e nver at Highway 48
until the analyzer was removed |n lsso*ved oxygen was obtained
minimum dissolved oxygen concernDunn£ this period the
per liter. The power company instant tamed was 5-6 milligrams
and has a continuous record of dis«^i ^a &n analyzer in the tailrace
the summer of I960. These records concentration since
resulting from large discharges frrim J density currents
result in occasional low dissolved r> Fr ^eservo^r continue to
charge from Roanoke Rapids Dam n~Sen c during periods of mini-
aerates the water through its vacuUlTl wQi, ro"fh two turbines and
the effects of the submerged weir Consequently, through
dissolved oxygen content of the 1•-wheel operation, the
6.0 milligrams per liter for extends J"' drops much below
that with Gaston in operation, with £ me' ** is e*pected
warmer water with a higher dissolve ^ 67el We*r se*ecting
water discharged from the Roanoke RaniJ- p* c e.nt> *he Quality of
as good as that required by the license S servoir wil1 be at least
Since completion of the RoanoWo j • ..
conducted on Catawba River by thP ni.v r>^' 3 S11*"^ar study was
study concluded that a submerged w„:r ¦ Jl ¦ °^er Company. This
intake. Accordingly, the Company's 3 *"gk-level
project includes a provision that a submprpp^01" Cowans Ford
around the turbine intakes. weir be constructed
In order that storage reservoirc moi,
conserving water, it is essential that consiLrat^ as, a means °f
methods of ensuring that the Qualitv nf th* Hi h ¦ £*ven to
meet the needs of downstream users From thp a*ge ,1S adeQuate to
serving dissolved oXygen assets. „ submerged wl^n elcZe
DISCUSSION
John M. Hester*
U. S. Fish and Wildlife Service, Atlanta, Georgia
highlights
,,MI' ?n'fht has Presented a valuable, informative paper on one
method of obtaining desired water quality below a thermally stratified
reservoir with low-level outlets. His paper on the hydraulic per-
*Now Field Supervisor, Branch of River Studies, U. S. Fish and
Wildlife Studies, Decatur, Alabama.
290
DISCUSSION
-------
formance of the submerged weir should serve as a useful basis for
future planning of projects.
Although it is evident the dissolved oxygen levels of the inflows
to Roanoke Rapids Reservoir are reduced within the reservoir, it is
also evident the top layers of the reservoir are capable of restoring
almost an equivalent amount through reaeration and photosynthetic
activity of phytoplankton. Based on this apparent situation and a
submerged weir that selectively discharges certain layers of the reser-
voir, the dissolved oxygen content of the discharges is adequately
maintained when higher discharge rates prevail.
It is interesting to compare the percentage of flow contributed
by each 5-foot layer of the reservoir under various rates of discharge.
Even though the Roanoke Rapids weir has not entirely overcome all
the dissolved oxygen deficiencies below the reservoir, the data indi-
cate the principle has merit in improving the dissolved oxygen con-
tent of waters discharged from certain types of thermally stratified
reservoirs. It shows considerable promise for other reservoirs that
will have limited water level fluctuation and relatively shallow depth.
The same principle will be employed in the newer Gaston Reservoir,
now under construction. The weir will, however, be raised to within
15 feet of the surface to ensure better water quality.
In the selection of the features for controlling the dissolved oxy-
gen content, the design and operations of the project were basic
considerations. I believe it is recognized that the engineering design
and operations of reservoirs limit the type of control features that
would be adaptable. In these two reservoirs, it was determined that
the most feasible means of providing releases with high dissolved
oxygen content was by a submerged weir surrounding the penstocks.
Hull (1) has made a review of the potential corrective measures that
have been used at other projects to remedy poor quality water result-
ing from impoundment.
Based on the findings presented, the submerged weir becomes
less effective in assuring a satisfactory dissolved oxygen content
when the magnitude and/or duration of cold inflows are increased.
This suggests that further increases in the volume of this type of
water entering the reservoir would increase the period of poor-
quality releases below Roanoke Rapids Dam. This illustrates the
necessity of providing means for water quality control at each devel-
opment unit added within the drainage system. The lack of means
to control water quality at an upstream project could tend to obscure
the effectiveness of a control device located in a downstream unit.
On the other hand, the downstream device may be made more
effective because of controls installed in a new unit. It is predicted
that this will occur below Roanoke Rapids Reservoir as the result of
the Gaston weir.
TRENDS IN WATER USE DEVELOPMENT
The recent report of the Senate Select Committee on National
Water Resources(2) indicates that by 1980 we will need to double
Hester
291
-------
the facilities we now have or conservation and use of water. Based
on this prediction, the Corps of EngjneerSt one of several planning
agencies for water use development, has recently formulated 20-year
estimates for their portion of the national program(3). The Corps
envisions the need of some 0 million acre-feet of storage in more
than 300 new reservoirs, plus inland waterways, harbors, levees, flood
walls, channel improvemen , and recreational facilities to accommo-
date perhaps 300 million visitors annually at the projects involved.
Fox(4) has pointed out that waste disposal and outdoor recreation,
two old uses of water resources, are growing with unusual rapidity!
He predicts in many parts of the country these two uses will tend
to dominate water management jn future years. Recently, Congress
has given expanded recognition to the need of additional outdoor
recreation areas and the need of Water quality control.
With this magnitude of reservoir development in prospect and
with increased competition by water users, it is of great concern
that adequate means be devised to provide optimum water quality
control for the entire stream system.
NEED FOR FLEXIBILITY To ACHIEVE WATER QUALITY
CONTROL
The Bureau of Sport Fisheries and Wildlife, in carrying out its
responsibilities of studying and reporting on water use projects, rec-
ognizes the importance of maintaining water quality control for
effective management of the fisheries affected. Since water quality
is subjected to detrimental changes because of impoundment, means
are sought whereby both the reservoir and downstream area can be
managed effectively under the changing conditions.
The Bureau is aware of the current limits of definitive knowledge
on precise management of water quality at individual units as well
as on total drainage system plans. Accordingly, our approach to date
has been to provide for as much flexibility as is feasible so that future
knowledge and understanding can be implemented as a management
tool without costly project modification. It should be emphasized here
that more future attention must be given to the total effects within
a drainage system.
A means that provides for some flexibility in management of
affected fisheries has been included in certain projects that have
either recently been placed in operation or under construction by the
Corps of Engineers(5). These projects generally provide for flood
control, low-flow augmentation, and raised water levels during the
recreational season. In providing for these purposes, the water levels
of the reservoir will be widely fluctuated.
Because the reservoirs will have widely fluctuating water levels,
multiple openings, such as are shown in Figure 1, are necessary to
provide for various levels of discharge so that the desired water can
be selected under the various conditions that occur as a result of
changing water levels. Although the dissolved oxygen content must
292
DISCUSSION
-------
be sufficient, it is also vital to the fishery that a favorable annual
pattern of water temperatures be maintained. The flexibility in select-
ing water temperatures afforded by the multilevel openings provides
opportunity for maintaining the desired temperature pattern (Figure
2).
SECTION OF UPPER INTAKES AND
REGULATION VALVE
Figure 1 — Multiple level intakes gated for selective withdrawal of water.
Figure 2 — Hypothetical temperature patterns with mulliple level intakes (Nolin Reservoir,
Kentucky).
Hester
293
-------
The use of multilevel openjn
managing the fisheries downstream wgreater Iatitude in
downstream area can be managed ./ . control provided, the
within limitations of other physi0al T °Pt"num fishery permitted
by the Kentucky Department of pj„v acj°^l'1t!idies are bei"g made
ate the effectiveness of the muitji a,n Resources to evalu-
the downstream fisheries (6). openings for management of
Concern has been expressed by some workers as to the effect
different levels of dUs*charges wilj. have on production and utilization
of the reservoir fishery ( j. I he most effective management of both
the reservoir and the downstream area should be the objective. For
now, there is need for flexibility 0f p]ant operations and basin studies
to resolve differences ana develop best management plans.
Mr. Knight has presented information on a technique that offers
a great deal of promise as a means of providing water quality control
when adapted to a set of conditions whereby the submerged weir can
effectively function. Although its limitations are recognized, the
findings reported provide useful information. From the fishery man-
ager's standpoint, *°eans are needed to provide sufficient latitude in
selecting levels of discharge so that the greatest benefits can be
achieved from management of both the reservoir and the downstream
area. Additional research is needed to determine the effects that
various levels of discharges will have on reservoir fish populations and
to explore potentials for development of highly specialized fisheries
below a dam.
In devising structural means to control the quality of water
discharged from reservoirs, it is necessary that project design and
operations for other purposes be recognized. Consequently, because
of the imposed limitations, each reservoir requires planning on an
individual basis.
REFERENCES
1. Hull, C. H. Discussion on effects of impoundments on oxygen
resources. In: Oxygen Relationships in Streams. Proc. of sem-
inar, Cincinnati, Ohio, Oct. 30-Nov. 1, 1957. Tech. Hep. W58-2
SEC. Mar. 1958. pp. 124-29.
2. Water resources activities in the United States. Committee Print
No. 1. Select Com. on Nat. Water Res., U. S. Sen. 1959. pp. 1-59.
3. Weber, E. W. (Office, Chief of Engineers, Corps of Engineers,
Washington, D. C.) Remarks. Presented at the 1963 Annual
Meeting of the National Wildlife Federation, Detroit, Michigan.
4. Fox, I. K. Benefit optimization in water resources management,
the 1960's and beyond. Interstate Conf. on Water Problems.
Resources for the Future, Inc., 1775 Massachusetts Ave., N. W.,
Washington, D. C. 20006. I960, pp. 1-16.
5. A detailed report on fish and wildlife resources in relation to
Nolin River Reservoir, Kentucky. U. S. Dept. of Int., Bur. of
294
DISCUSSION
-------
Sport Fish, and Wildlife, Branch of River Basin Studies. 1960.
15 pp.
6. Turner, W. R. Pre-and-post-impoundment surveys. Progress
report on project. Ky. Dept. of F. & W. Res. F-16-R-3. 1962.
pp. 1-45.
7. Sport Fishing Institute Bulletin. No. 98. Jan. 1960. pp. 3-6.
DISCUSSION FROM THE FLOOR
Mr. Coutant, Lehigh University: Mr. Knight, would you please
defend your reliance upon average dissolved oxygen values when it
is rather well known that the only datum of real significance to the
physiology of organisms is the oxygen minimum?
Mr. Knight, North Carolina Dept. of Water Resources: Appar-
ently I did not make myself clear. I thought I said it was the mini-
mum dissolved oxygen values that the people in the water pollution
control business had to concern themselves with, and that the mini-
mum values observed in this situation were the ones that required
some action for improvement. Under minimum flow conditions, when
dissolved oxygen values fall to below 6, I believe it is, the power
company splits the minimum flow between two of its turbines to
provide additional oxygen.
Of course, the circumstances that bring about these lower dis-
solved oxygen concentrations at minimum flow conditions should be
largely if not completely eliminated after the Gaston weir is in oper-
ation, since the density underflow conditions that occurred in the
system during the period of study and produced the observed low
dissolved oxygen values should be largely eliminated.
Mr. Gellman, National Council for Stream Improvement: Mr.
Knight, first, to what extent might the ability of the submerged weir
to select water from various levels be a function of its porosity?
Second, I note in your paper that the operating range of Roanoke
Rapids Reservoir is about 3 feet. From your studies do you have any
indication as to how much the effectiveness of this structure might be
increased by raising the elevation of the weir crest closer to the
surface than the present 25 feet?
Mr. Knight: We have not investigated the effect of the porosity
of the weir on the quality of the discharged water. It is my opinon,
however, that since the hydrostatic pressures on both sides of the
weir are approximately the same, the porosity of the weir will not
have a serious adverse effect on the quality of the water that is dis-
charged.
Regarding the second question, the effectiveness of the weir in
improving water quality might be increased by raising the crest to
a higher elevation. This installation was constructed, however, to
From the Floor
295
-------
form a basis for predicting what should be done at the Gaston Project.
The 25-foot submergence at Roanoke Rapids was arbitrarily selected
for its use as an experimental unit. The higher the weir, the higher
the velocity across the top of the weir and, from the standpoint of
power production, the greater the head loss. Perhaps some of the
engineers in the audience can contribute more on this point.
Mr. Peters, Virginia Electric and Power Company: I might be
able to add a little to the answers to both of these questions. While
studying the weir during the 1957 survey that Mr. Knight speaks of,
we made very careful temperature measurements on the inside face
of the weir, using a very sensitive thermometer. Since there was a
temperature differential in the water on the two sides of the weir
at the lower depths, we would have been able to pick up a lowering
of the temperature on the inside if there was any leakage. At no place
did we find any evidence of seepage through the weir.
In addition to this, a comparison of the contribution at various
levels (for which the temperature and dissolved oxygen were known)
with the temperature and dissolved oxygen of the water in the tail-
race also showed that there was no seepage of water through the weir.
On the question of the submergence of the weir, the Gaston weir
is being built to within 15 feet of the surface. Based on the head loss
across the weir, the length of the weir required at Gaston Dam is
just about 1,000 feet.
Mr. Ragone, Virginia Electric and Power Company: Figures 4
and 5 in our paper show the construction of the Roanoke Rapids weir.
The weir was actually installed in the old Coffer Dam area after the
reservoir was filled. The height was limited by the construction
method. It had to be rock fill because of the economic problem of
keeping the plant in operation. We could not lengthen the weir and
still stay on the Coffer Dam area. The base was such that 25 feet
was as close as we could come to the surface and not have excessive
head loss over the short length of weir.
Dr. Harleman*, California Institute of Technology: From my
experience with structures for control of stratified flow, the weir is
a very effective device. We have done some testing for TV A on an-
other type of structure you may be interested in seeing very briefly.
I think it may reinforce your confidence in this control of stratified
flow. In connection with the planning of a fairly large steam power
plant, an intake channel and outlet channel were to be constructed
on a section of a river. As you well know, when you pass a portion
of the water through a condenser you raise the temperature, which
results in a temperature stratification in the river. The condenser
return will move upstream, especially if the quantity of water used
is large compared to streamflow. Therefore, we planned a skimmer
wall structure, which is just the upside-down part of the structure
Mr. Knight discussed. This was, in effect, a submerged sluice gate
*Now with Massachusetts Institute of Technology.
298
DISCUSSION
-------
that held back the warm water so that the cold water could be taken
through. The water on the downstream side was essentially the same
level (very slight head loss), but the submerged sluice has been very
effective in preventing a thermal recirculation.
We conducted experiments to determine what maximum rate of
flow per unit width could be allowed for a given elevation of the
cold water without taking in the warm water above. Probably the
most interesting thing about the submerged weir is that one can
analyze it as a problem of critical flow analogous to surface flow. It
looks just like a flow over a weir in air, the difference being that the
effect of gravity is greatly reduced as a result of the reduced density
difference. This type of flow, which is analogous to free surface flow,
takes place in relatively slow motion. This is a further example of
selective withdrawal and the ability to control flows in stratified
water having very small density differences. The density difference
between two layers is about one-third of 1 percent when the tempera-
ture difference is about 10 degrees Fahrenheit.
This type of structure can obviously also be used if the river in
which the diversion is to occur is already stratified due to, let us say,
deep releases from a reservoir upstream.
Mr. Kittrell, U. S. Public Health Service: Dr. Harleman, what
velocities were used in your experiments?
Dr. Harleman: The velocities were about half a foot a second in
the model.
Mr. McLean, JJ. S. Public Health Service: Mr. Churchill, in one
of your recent papers you showed a graph of the annual change in
dissolved oxygen in the discharge from a reservoir that had a pro-
nounced dip in September or October. I think the outlet from this
structure was at an intermediate level in a deep reservoir. Do you
think that this could be explained in terms of a density underflow
of colder fall flows that came into the reservoir and raised the hypo-
limnetic water, which earlier was below the level of the outlet, into
the area of the outlet? This would suggest another reason for having
multilevel outlets,
Mr. Churchill, Tennessee Valley Authority: I could not say. Each
situation must be analyzed by itself. I would need to know more
about the situation before I would want to comment on it.
From the Floor
297
-------
IMPROVEMENT OF THE QUALITY OF RESERVOIR
DISCHARGES THROUGH TURBINE OR
TAILRACE AERATION
Theodore F. Wisniewski
Director, Wisconsin Committee on Water Pollution, Madison
Control of water pollution requires a continuing search for new,
practical methods of treating wastewater. The development and in-
stallation of processes capable of removing 90 to 95 percent of the
oxygen demanding components of sanitary and of industrial waste-
waters prior to discharge remains a necessary first objective in every
water pollution control program. But the tonnages of residual oxygen
demanding components from even these high efficiency processes add
to the ever increasing loading placed on the receiving waters of rivers,
lakes, and streams.
More than 20 years ago Tyler (1) suggested that practical means
of aeration might be developed to increase the natural treatment
capacity of a body of water. Stream studies have demonstrated re-
peatedly that moderately sized rivers have a large and important
working capacity for self-purification, which may involve the
processing of hundreds of tons of natural and man-made organic
waste matter in a relatively short stretch of the stream. Artificial
reaeration may often be required, however, to facilitate these treatment
processes in natural waters and to aid in rapid recovery to clean
stream conditions. The early concepts of stream reaeration processes
are now attaining the status of practical and workable auxiliary
methods of waste treatment of growing importance to the overall
objective of water pollution control. Reaeration of quiescent waters
in lakes and bays has received less study, but could be a promising
area for further research.
Development of methods for reaeration of rivers has been a
subject for continuing study by several cooperating organizations in
Wisconsin since 1943. Various aspects of this continuing state and
industry study have been published in a series of seven papers(2-8).
Hydroturbine aeration at power dams has been especially successful
as a practical method for introducing air into the water for large-
scale reoxygenation of waters depleted in dissolved oxygen. As of
October 1961, 18 different dams on the Flambeau, Wisconsin, and Fox
Rivers have been equipped for hydroturbine aeration. Several power
dams are also known to have been equipped for aeration in neigh-
boring states. Other installations have been under consideration or
are being tested on a preliminary basis.
Experience with hydroturbine aeration of rivers at some of these
installations covers 5 years or more, and many have had at least 3
years of operation during critical periods of dissolved oxygen deple-
tion in these rivers. Basic principles for successful operation of stream
reaeration are becoming established, and the limitations of methods
are becoming known. This paper is intended to review the present
Wisqiewski
299
-------
status of commercial_scale stream aeratinn nc *
ing the recovery of oxygen depleted waters and aT ^ acceIerat"
working capacity of rivers subjected to hen 1 increasing the
and ohemicaroxygen demand. heav>-load'"^ of biochemical
DISSOLVED oxvgen tension
There is a limitation on the effioipnr»v ^rt r^„+- i
reaeration in that the rates of solu^0n f oxwpnati Ue °*
slow at the higher levels of oxygen saturation^ watlr" V!ry
temperatures. During summertime critical nerinric higher
flow and high temperature, it is Seldom nraetical fn t ^ stream
waters when the dissolved oxygen saturation excee(Js 4yp"dr/eenaerate
oxygen level declines to 3 milligrams per liter nr loce ,,e ^
g™np^r o "m-diSS°IVed •*» -acheTTmlS
''-r-
waters of the Flambeau River at the PixW nam b h teri^med in
stallation. Very high rates of increase in thf> rficoni ^ ro ^me *n-
were acco mplished under condition"* 'hf. dissol^d oxygen level
tion of 10 percent or less, hut a" the X™ S8tUra-
saturation increased to the 50-percent level th& rlt Vf ' ,oxygen
oxygen increase dropped rapidly to less than' in np 6 +°f dissolved
perature of the water materially Xcts the %Jr J?'ZT ' tem"
thus the rate and efficiency of dissolved nxvupf ah n aturation and
peratures approaching the freezing p0jnt Qf water ah"' ~ ¦ tem"
oxygen may be highly efficient, even when thf> ,absorPtlon of
exceeds 6.0 milligrams per liter. On the other hsnrt Vle
solved oxygen in the range of 3 Sgrams per WeTn °'<*"
important that the degree of dissolve nJvo + + ' t^ere^ore>
sidered in establishing the level at whirh r^IJ ? sat"rat11°n be con"
taken lcvei at which reaeration should be under -
AERATION EXPERIENCE IN WISCONSIN
Flambeau River
w,EXPenem;eoWlth riV6r reaeration on the Flambeau River extend*
back over an 18-year period ct, j j i .¦ " extends
located just upstream from (St ,m""""1 Plxl6y Dam is
accelerated reaction rates and »t tif ° the summertime zone of
of depleted dissolved oxygen lev , , "fh t °f,Crltieal 20ne
river that extends dowSeam Ir„m th^tfnTpil' ot the
pulp and paper mill is located. allS where a
air *»« introduction o<
user plates that extended across the bottom
300
IMPROVEMENT THROUGH TURBINE AERATION
-------
~0 SATURATION BEFORE AERATION, %
Figure 1 — Relationship of DO saturation to DO absorption by turbine
aeration at Pixley Dam.
of the tailrace; later, diffusers were added in the headraces. Appreci-
able increases in the amount of dissolved oxygen were obtained by
this method of aeration, but at a rather low level of aeration efficiency
(on the order of 6- to 8-percent oxygen absorption for the air intro-
duced to the water). Many of the original questions about reaerating
a river were settled in the preliminary experiments with the diffuser
assemblies. Air introduced into the headrace and passing through
the turbines did not in any way seem disadvantageous from the
standpoints of vibration or corrosion problems.
A brief description of the flow pattern of the river may be helpful
in gaining an understanding of the conditions that had to be met to
obtain any reasonable advantage from stream aeration.
Wisniewski
301
-------
The Flambeau River receives no known man-made wastes above
Park Falls. The river flow is controlled principally by releases from
the Flambeau reservoir. The drainage basin at the reservoir outlet is
666 square miles, while at Park Falls the basin measures 720 square
miles. The reservoir outlet produces a controlled release of at least
609 cfs 50 percent of the time, and at least 150 cfs 95 percent of
the time.
At Park Falls, wastes from a sulfite pulp and paper mill, as well
as treated sewage effluent, reach the stream. The strong spent sulfite
liquor is either employed as roadbinder or disposed of by hauling for
land disposal. Pulp-washing wastes are sprayed on an extensive area
of bark fill and leach to the stream. Normal papermill wastes are
also discharged to the stream. All wastes, including treated sewage
effluent, reach the stream just below Mill Dam at Park Falls, the
first dam below the reservoir.
Wastes discharged at this point almost immediately enter the
flowage area formed by Lower Dam. Some added holding time is
thus provided and the dissolved oxygen decreases. After discharge
through Lower Dam turbines, the stream flows more rapidly for about
a mile before entering Pixley Dam flowage, an extensive body of
water providing considerable holding time. The level of dissolved
oxygen in the stream continues to drop through the flowage area and
reaches critical levels at Pixley Dam. Discharge from Pixley turbines
enters Crowley Dam flowage almost immediately, and this body
provides further time for waste decomposition. Below Crowley Dam
the river flow is more rapid and the stream is used for recreational
purposes, including fishing and canoeing. The objective of earlier
consideration of compressed air aeration at Pixley Dam was to im-
prove the section of stream below Crowley Dam down to Oxbow, a
resort area located about 25 stream miles below Park Falls. The
critical dissolved oxygen conditions developed at the Pixley Dam
outlet, 7 stream miles and two flowage areas below Park Falls, and
made this accessible point the likely location for aeration facilities.
In 1957 the first experiments with turbine aeration were initiated
at Pixley Dam, and when the better efficiency of this method of aera-
tion was proven, the diffuser assemblies were placed in standby.
Hydroturbine aeration has been studied each summer since 1957 and
during all seasons at that location on the Flambeau River. Two of
the three turbines at this dam were equipped for aeration and were
operated in excess of 250 days during 1958 and 1959. The aeration
facilities were in operation for 144 days in 1960 and for about the
same number of days in 1961. Dissolved oxygen levels above the
point of aeration at the dam have averaged from 3.1 to 4.3 milligrams
per liter for the entire year and from 1.1 to 3.3 milligrams per liter
during summer operation. Water temperatures were relatively low,
however, on the order of 20°C or less for most of these operating
periods. At these temperatures the degree of saturation was usually
less than 40 percent, so that aeration efficiency was good in terms of
oxygen pickup by the water. Increases in the dissolved oxygen con-
tent of the water were on the order of 1 milligram per liter or better.
I
IMPROVEMENT THROUGH TURBINE AERATION
-------
Lower Fox River
Five dams along the 39-mile stretch of the Lower Fox River
between Lake Winnebago and Green Bay are equipped for hydro-
turbine aeration. Operators at these dams are ready and quickly able
to start aeration through the turbines whenever oxygen levels along
the river drop below 3 milligrams per liter.
Three water wheels are equipped for aeration at Kimberly. One
turbine aeration unit was available and operating in 1958 when con-
ditions of low dissolved oxygen prevailed, but during 1959, 1960, and
1961 water flow conditions on the Fox River were at relatively high
levels and dissolved oxygen concentrations remained above 3 milli-
grams per liter at that location. No aeration was practiced in those
years. At Combined Locks, three hydroturbines for power production
are now equipped for aeration. One unit was operated for 35 days in
1959 and all three for 19 days in 1961. The dissolved oxygen pickup
was substantial in 1959 when the dissolved oxygen level dropped to
0.6 milligram per liter, but oxygen absorption was much less evident
in 1961 when the average dissolved oxygen was 2.62 milligrams per
liter ahead of the point of aeration. At Rapid Croche Dam farther
downstream, hydroturbine aeration was practiced with one turbine
for 28 days in 1958 and 85 days in 1959, but not at all in 1960 when
stream flows were high and dissolved oxygen levels were above 3
milligrams per liter. A second turbine was equipped for aeration at
this dam in 1961, and aeration has been in progress for 72 days with
the season of operating continuing. During periods of operation, the
dissolved oxygen levels have been shown to be increased substan-
tially by turbine aeration.
At Little Rapids, aeration is conducted through the water wheels
operating six grinding wheels in the groundwood mill. Two wheels
were equipped for aeration in 1959, four in 1960, and six in 1961.
Operating conditions at this aeration facility are unique in that the
low head turbines operate at tail-water level, and the water wheels
do not create enough suction to draw their own supply of air. Blowers
introduce the air to the wheels at this dam.
A fifth power dam at DePere on the Lower Fox was equipped
for aeration in 1961, but as yet this unit has not been operated for a
sustained period of time and no data are available.
Wisconsin River
Twelve dams along the Wisconsin River have been equipped for
hydroturbine aeration. Some of these dams are located in stretches
of the river that are below the critical zone of oxygen depletion
during warm weather periods of high reaction rates. These dams,
however, are located in stretches of the river that become critically
low in dissolved oxygen as the critical zones of low oxygen stretch
farther downstream to reach the dams. Preliminary trials have been
made and it is expected that hydroturbine aeration will be operated
at these dams during wintertime conditions. For those dams for which
summertime operating data are available, the oxygen pickup by the
Wisniewski
303
-------
water has been on the order of 1 milligram per liter or more in most
cases. This amount may at first glance seem insignificant, yet it
amounts to 5,400 pounds of oxygen per day per 1,000 cfs of stream
flow. The oxidation activity of the stream is considerably increased
with tonnage oxygen supplied at these levels.
Table 1 shows the percentage of dissolved oxygen increase and
the increase of dissolved oxygen in pounds per 1,000 cfs of daily
stream flow for those hydroturbine installations where analytical
data are available. A wide variation in the percent of dissolved
oxygen increase can be noted, ranging from 2.7 to 1,173 percent. The
value of this information lies chiefly in emphasizing the importance
of carrying out aeration during periods when the level of dissolved
oxygen saturation is less than about 40 percent. It is seldom profitable
to attempt aeration when the saturation level is above 40 percent
because efficiencies are low. It has been mentioned that most dam
operators have agreed to start their aeration facilities when the dis-
solved oxygen drops below 3 milligrams per liter during warm weather
operating periods. The information in Figure 1 and Table 1, there-
fore, suggests the importance of carrying this operational criterion
a step further. This would involve starting an aeration facility when
the dissolved oxygen saturation drops below 40 percent.
TABLE 1. DISSOLVED OXYGEN INCREASE BY REAERATION
Location
Year
DO
increase,
mg/1
DO
increase,
%
DO increase,
pounds/day/
1,000 cfs
Combined Locks
1959
1.2
200.0
6,480
1961
0.07
2.7
378
Rapid Croche
1958
1.5
Total
8,100
1959
1.8
233.0
9,720
Little Rapids
1959
0.35
23.8
1,890
1960
0.47
21.4
2,538
1961
0.67
40.4
3,618
Hat Rapids
1959
1.17
260.0
6,318
1960
0.71
51.0
3,834
1961
0.77
110.0
4,158
Tomahawk (Exp.)
1961
1.76
1,173.0
9,505
Rothschild
1959
1.80
176.5
9,720
1960
1.11
40.7
5,994
1961
1.45
114.2
7,830
Mosinee (Exp.)
1961
1.00
66.7
5,400
Pixley
1958
1.3
37.1
7,020
1959
1.2
27.9
6,480
1960
0.9
23.1
4,960
1961
1.0
32.3
5,400
304
IMPROVEMENT THROUGH TURBINE AERATION
-------
The value of aeration from the practical standpoint can be
shown in the increase in pounds of dissolved oxygen per day per
1,000 cfs of stream flow. Table 1 shows that this varies from 378
pounds of dissolved oxygen pickup per 1,000 cfs in a marginal period
of aeration upward to nearly 10,000 pounds of increased oxygen per
day per 1,000 cfs. It is apparent that most of the hydroturbine aera-
tion facilities installed in Wisconsin are capable of introducing about
2Vz tons of oxygen to 1,000 cfs of daily stream flow; as the methods
of introducing oxygen are improved, it is considered likely that most
installations can approach the point of introducing 5 tons of oxygen
per day per 1,000 cfs of stream flow.
COST OF STREAM REAERATION
A thorough comparative evaluation of capital and operating
charges for various methods of introducing oxygen into large streams,
rivers, and quiescent water bodies is much needed. Several prelim-
inary evaluations have been undertaken.
Capital Charges
The capital costs for pumping air and laying diffuser beds across
a stream or in lakes or bays to achieve direct aeration by diffusion
methods may be substantial. On the other hand, hydroturbine aera-
tion may, by use of existing facilities designed for other purposes, be
found to contribute substantially large amounts of oxygen with little
or no outlay for equipment. Hydroturbine aeration can and is being
achieved in several of the 18 installations reported in this paper with
little or no alteration of the existing turbine equipment. The simplest
turbine aeration system uses draft tube vents, which are often avail-
able to introduce air for control of vibration. These are merely opened
and used for aeration in some of the cases cited.
Not all hydroturbine installations are equipped for venting of air
to the draft tube or wheel. Modern propeller-type turbines usually
require special venting, and this can be a considerable undertaking
through massive concrete foundations.
Most hydroturbines can apparently benefit from installation of
carefully designed venting to fit individual operating conditions and
plant design. Opinion varies as to conversion costs under these con-
ditions, but apparently the average cost would be in the range of
$5,000 or less for each turbine, not considering down time. This esti-
mate is given merely to provide an order of magnitude.
Operating Charges
The costs of operating aeration facilities vary widely with the
method and the conditions that prevail at individual installations
(Table 2). The mechanical and diffusion methods of introducing
oxygen to the water may have substantial operating charges for
power, labor, and supervision.
Turbine aeration at power dams appears to be a low-cost method
under most conditions, with charges largely confined to the loss of
Wisniewski
305
-------
power production potential for anv .
of the loss in power production rfJP^0n turbine- The exact extent
swered on a quantitative basis at e question not easily an-
where preliminary evaluations hav/k the several Power plants
picture based on observations madp k n undertaken- The general
installations seems to favor a power opei^ors at the hydroturbine
less at most low-head dams (15 t the order of 5 percent or
stallations (exceeding 10O ft). n) and less at high-head in-
TABLE 2. COMPARISON OF Costq ,
dam aeration at pixley
Diffuse!-
aeration
880
2,875
3,270
1,320
0.46
$12.85
$2,900
$1,000
Turbine
aeration
~317
3,330
10,340
1,040
0.31
$8.45
$39,000
Stream flow, cfs
02 absorbed, lbs/day
02 absorbed, lbs/1,000 cfs/day
Power used, KWH/day
Power used, KWH/lb of Oo
Cost of Power*/ton 02 absorbed
Seasonal operating cost to absorb 5 tons of
O./day (45 days per year) $2 900
Annual maintenance cost $1 n '
Installation cost to give 5 tons
O2/day/lf000 cfs
*f(1.4^/KWH.
An example of the problem faced in determining costs for loss in
power production may be cited, la this case, panel board readings
showed power production from one turbine to be 677 7 kilowatts
with the aeration turned oft and 599.8 kilowatts with the aeratoTon
The apparent loss of_ 77.9 kilowatts amounted to 11.5 percent The
"rr? wlt°h th 'he tUrWne <3uring thcse tests ™ determined
to be 791.6 cfs with the aerator off and 752 cfs when the aerator was
on. Aeration thus reduced the amount of water going through The
turbine by 39.6 cfs, which is a 5-percent reduction in throughput.
u oc^T-i Production per unit of water flow is then calculated
to be 0-8561 kilowatt per cfs with aerator off and 0.7976 kilowatt per
cfs with the aerator on, or a loss of 6.8 percent, which appears afa
during the periid'^ofTesfeg.085 °' POWW "y the indivWual turbine
One must also consider that the restriction of 5 percent of water
l t^r „ S ft :" lvidual turbine actually saves this water for
later use by that turbine or for use by other turbines if they are avail-
able, as is usually the case when summertime operation of hydro-
£JeS is emPIo^ed durin§ the season of low water
flows. With this added consideration, it is readily apparent that the
actual loss m power in this example is substantially less than 5
percent of that when operating without aeration.
improvement through turbine aeration
-------
The overall costs of hydroturbine aeration appear reasonable and
economically supportable for the installations surveyed.
CONCLUSIONS
1. Reaeration of oxygen-depleted waters is being successfully
developed as a supplementary method of alleviating critically low
levels of dissolved oxygen and of increasing the self-purification
capacity of streams. As such, stream reaeration is proving to be an
important new tool for water pollution control.
2. Mechanical methods of aeration and diffusion of forced air
appear promising and probably will find specific applications, but in
general appear to be relatively expensive in terms of capital and
operating charges.
3. Of the methods of aeration studied and applied to large flows
of water, hydroturbine aeration has received widest application, with
18 dams having one or more turbines equipped for aeration in Wis-
consin in 1961. Other such installations have also been made and
operated on a commercial scale elsewhere.
4. Costs of hydroturbine aeration are considered to be practical
and economically feasible in the installations thus far made. Equip-
ment alterations are usually a low-order expense and operational
charges appear to be largely a matter of losing a certain percentage
of the power production of a turbine, with the loss estimated to be
usually 5 percent or less of normal rates for total power production
in a multiple turbine installation during seasons of low water flow.
5. The efficiency of introducing air into water has previously
been estimated to range to 25 percent with hydroturbine aeration.
Data collected for his report indicated that most hydroturbine aeration
assemblies now operating can add as much as 2% tons of oxygen to
a stream per day per 1,000 cfs; that several assemblies are now adding
as much as 5 tons of oxygen per day per 1,000 cfs; and that well-
designed installations for aeration could further increase this effi-
ciency.
6. Each ton of oxygen absorbed by the water from stream aera-
tion facilities greatly increases the capacity of the stream to handle
the residual oxygen demand from wastes discharged from the often
overloaded treatment plants of cities and industries.
REFERENCES
1. Tyler, R. G. Accelerated reaeration. Sewage Works J. 14(4):
834. July 1942.
2. Wiley, A. J., L. Parkinson, H. W. Gehm, T. F. Wisniewski, and
H. F. Bartsch. River reaeration. Paper Trade J. 124:123. Mar.
20, 1947.
3. Scott, R .H., T. F. Wisniewski, B. F. Lueck, and A. J. Wiley.
Organization of cooperative state-industry stream studies in
Wisconsin. Sewage and Ind. Wastes. 29:298. Mar. 1957.
Wisniewski
307
-------
4. Wiley, A. J., B. F. Lueck, R, h. Scott, and T. F. Wisniewski.
Cooperative state-industry stream studies — Lower Fox River
Wisconsin. Sewage and Ind, Wastes. 29:76. Jan. 1957.
5. Lueck, B. F., A. J. Wiley, r. H. Scott> and T F Wisniewski.
Determination of stream purification capacity. Sewage and Ind
Wastes. 29:1054. Sept 1957.
6. Scott, R. H., T. F. Wisniewski, B. F. Lueck, and A. J. Wiley.
Aeration of stream flow at power turbines. Sewage and Ind
Wastes. 30:1496. Dec. 1958.
7. Wiley, A. J., B. F. Lueck, R. H. Scott, and T. F. Wisniewski.
Commercial-scale operation 0f turbine aeration on Wisconsin
rivers. Sewage and Ind. Wastes. 32:186. Feb. 1960,
8. Scott, R. H., and T. F. Wisnieswski. Hydro-turbine aeration of
rivers with supplemental data on cascades aeration. Pulp Paper
Mag. Can. 61:T45. Feb. I960.
discussion
W. S. Lee
Engineering Manager, Duke Power Company, Charlotte, N. C.
Mr. Wisniewski is to be complimented for the pioneer work in
stream reaeration done under his direction in Wisconsin. With the
cooperation of many industries and other participating groups, he and
his associates have accomplished a formidable task in arranging for
reaeration in hydroturbines in about 30 dams on 3 rivers in Wisconsin.
This brief discussion compares his results with the results of
similar reaeration experiments conducted by my company at one
dam in South Carolina. These experiments are a part of a compre-
hensive program of water-quality monitoring and research on rivers
and lakes within the service area of Duke Power Company in North
and South Carolina. This program is directed by Mr. J. Ben Stephen-
son, Supervisor of Water Research for our company, who is here today.
Our aeration experiments were conducted at the Wylie Station
plant located near Rock Hill, South Carolina, on the Catawba River.
This station is one of a series of dams operated by Duke Power
Company on the Catawba-Wateree River in North and South Caro-
lina. Figure 1 shows a profile of this river on which there are 11 dams
with impounded reservoirs and 2 open stretches of free-flowing river
in the 221 miles of river shown on the profile.
Significant quantities of BOD load are brought into Lake Wylie
by tributary streams. Along the river downstream of Wylie Station
are substantial industrial developments involving waste effluents.
These developments include a textile finishing plant, a large syn-
thetic textile plant, a large pulp and paper mill, a bleachery, and
municipal effluents treated to various degrees. Because of the waste
assimilative work-load on this section of the river, Wylie Station was
selected as the most appropriate for reaeration experiments.
DISCUSSION
-------
CATAWBA RIVER DEVELOPMENT
FOHfNG CRKK
GREAT FALLS
ROCKY CREEK
WATEREE
10 20 30
0 30
PROFILE OF JOHNS RIVER
PROFILE OF SOUTH FORK
f gti! 1 — Fraffle al Catawba River.
Mr, WisTnew ski's work in Wisconsin was done on several types
of hydro turbines ranging from about 185 to 1,000 kilowatts in capa-
city. The dams had heads ranging from a minimum of 7 feet
to a. maximum of about 22 feet. In contrast, Wylie Station has 4
Francis turbines, each rated at 15,000 kilowatts, and operates under a
gross head of about 70 feet. Wylie dam impounds a reservoir with a
full-pond storage of ahout 2&0,000 acre-feet. During the late summer
months when reservoir stratification occurs, there is some degradation
of the dissolved oxygen in the bottom waters of this reservoir. This
variation in dissolved oxygen content with depth of the lake is shown
on Figure 2 which represents data taken on August 28, 1959, a day
when the hypolimnion was particularly low in oxygen. Figure 3
shows a cross section of the Wylie Station. By a series of velocity
measurements at various depthst we have determined that the plant's
turbines draw with about equal velocity from all reservoir depths
when operating from half load to full load. As the load is decreased
below half load, less of the oxygen-rich surface water contributes to
the flow through the turbines. The setting of these units with respect
to tail-water elevation is such that the vacuum breakers are not oper-
ative unless the units are at very low load- During medium to high
loads, no vacuum occurs in the draft tube, but, to the contrary, a
positive pressure exists.
Lee
-------
<
c7)
Ul
a
z
3 <
«A i_
—
O
m
<
z
o
5
5
76 78 80 82
DISTANCE, mi
figure 2 — Dissolved oxygen in Lake Wylie, mg/l. August 28, 1959.
Generator
Turbine
Tailrace
a
&
i
0)
>
c
4)
8
X
XJ
$E
Intake
lake wylie « »
Water Surface Elevation 566.2 s. a
Trash Racks
SAMPLE COMPUTATION TO DETERMINE TAILRACE DO
Depth Interval
0-5
5-10
10-15
15-20
20-25
25-30
30-35
35-40
40-45
45-50
50-55
OBSERVED DO
VARIED FROM
DO,
mg/l
8.4
8.4
7.8
4.8
1.4
0.3
0.2
0.0
0.0
0.0
0.0
r/r Contribution
10,200 cfs
DO Contribution
10,200 cfs
10.4
11.4
12.0
12.1
11.7
10.8
10.4
9.8
8.1
2.6
0.7
CALCULATED DO
0.87
0.96
0.94
0.58
0.16
0.03
0.02
0.00
0.00
0.00
0.00
3.56 mg/l
IN TAILRACE DURING SAME PERIOD
3.3 mg/l TO 3.7 mg/l
Figure 3 — Velocity patterns at Wylie Station, August 25 and 26, 1959.
310
DISCUSSION
-------
Tabie I gives results of the 1959 tests at Wylie Station to deter-
mine the effect of vacuum breakers on the discharged dissolved
oxygen. Two runs of 2 hours each were made on 3 days with all four
units running at a load of 1,000 kilowatts each. This is about 7 percent
of their rated load. The unit load was held constant throughout the
day. In the first run each day, the vacuum breaker valve was blocked
shut so that no air could rush in to satisfy the negative pressure in
the draft tube. The second run allowed the vacuum breaker valve to
open so that air entered the draft tube for reaeration.
TABLE 1. "RESULTS OF TESTS AT WYLIE STATION TO DETER-
MINE EFFECT OF VACUUM BREAKERS ON DISCHARGED
DISSOLVED OXYGEN 0
No. of Vacuum
Date of units breaker _ , DO
sampling, Discharge, Loading, oper- oper- discharged gain>
1959
cfs
kw
ating
ation
mg/1
% sat.
mg/1
Aug 28
2600
4000
4
No
2.0
37
—
Aug 28
2600
4000
4
Yes
5.7
73
3.7
Sept 12
2600
4000
4
No
4.9
62
—
Sept 12
2600
4000
4
Yes
7,7
98
2.8
Sept 13
2600
4000
4
No
5.5
69
¦—
Sept 13
2600
4000
4
Yes
8.4
102
2.9
Sept 14
1400
6000
1
No
4.6
54
—
Sept 14
1300
2000
2
Yes
7.4
86
2.8
Sept 16
1480
60OD
1
No
5.0
58
—¦
Sept 16
1300
2000
2
Yes
7.1
83
2.1
"The dissolved oxygen values are average values obtained
from a series collected at 15-minute intervals for 2 hours.
The values in the series showed consistency within plus or
minus 0.2 mg/I, verifying a stable flaw condition.
The gain in dissolved oxygen varies from 2.1 milligrams per liter
to a maximum of 3.7 milligrams per liter. In the August test, the
increase of 3.7 milligrams per liter of dissolved oxygen resulting from
the vacuum breaker operation is equivalent to a net gain of about
20,000 pounds of oxygen per 1,000 cfs per day. The gain of 2.1 milli-
grams per liter of dissolved oxygen on September 16th is equivalent
to about 11,000 pounds of oxygen per 1,000 cfs per day. These values
are about double the increase in dissolved oxygen experienced by Mr.
Wisniewski on "Wisconsin rivers. In the Wisconsin studies it was found
that reaeration was not efficient or productive at temperatures above
20°C when the dissolved oxygen exceeded 3 milligrams per liter before
reaeration. In contrast, all of the September tests conducted at Wylie
Station started with dissolved oxygen values of more than 40 percent
of saturation and showed dissolved oxygen increases as great as 2.9
milligrams per liter. It must be remembered, however, that all
of the Wylie tests were conducted at very low turbine loads in eom-
Lee
311
-------
parison to the machine ratings, as the vacuum effect is not present
at higher loads.
Mr. Wisniewski's work in "Wisconsin included several measure-
ments that we did not take during our experiments. The flow through
our turbines was estimated from the turbine and generator perform-
ance curves and not from flow measurements. We were not able to
measure the flow of air through the vacuum breaker pipe, and there-
fore no computations of absorption efficiency were possible. Mr. Wis-
niewski's work includes computations of kilowatt hours lost per pound
of oxygen added due to decrease in turbine efficiency from adding
air below the runner. At Wylie Station, when operating at sufficiently
low loads to cause the vacuum breaker to open, the turbines are at
their least efficient operating point with or without the vacuum
breakers. As a consequence, our kilowatt hour loss due to low load
operation exceeds the kilowatt-hour change that might be exclusively
effected by admitting air to the draft tube.
When additional work on hydroturbine reaeration is done in
Wisconsin, in South Carolina, and elsewhere, it may be possible to
better predict the empirical parameters of reaeration at a variety of
installations. It is suggested that reaeration effectiveness depends
upon the turbine characteristics, the load being carried, the setting
of the wheel above tail-water elevation, the vacuum-breaker arrange-
ment, and the shape and dimensions of the draft tube. These physical
relationships are in addition to the chemical factors suggested by Mr.
Wisniewski.
As we continue our work in this area, we are sure to find more
answers and achieve a better understanding of the phenomena in-
volved. It has been a pleasure to discuss the highlights of our work
on the Catawba River and to compare it with the notable work in
Wisconsin described by Mr. Wisniewski.
DISCUSSION FROM THE FLOOR
Dr. Ingots, Georgia Institute of Technology: In 1957, a paper was
submitted to the editor of the Journal of the AWWA, entitled "The
Pollution of Downstream Rivers by Hydraulic Power Installations."
The editor would not accept the paper with that title on the basis
that it merely described a pollution effect. Our meeting here today
is at least a partial confirmation of the author's thinking that possibly
one could define as pollution the effect of a power dam in degrading
downstream water quality.
Mr. Lee, Duke Power Company: There are some of us at this
symposium representing power companies, and a number of others
representing our sister agencies with the Government who generate
power. I do not speak for all of these folks, but I might comment
that my company's interest in the subject of pollution and waste
assimilation is a strong one. Several fingers have been pointed at us
at this symposium, but let me say we are endeavoring in the Carolinas
312
DISCUSSION
QPO 83 1—740"! I
-------
to create in these rivers the capacity for doing work that will allow
us to go to Michigan where Professor Velz lives, and to Cincinnati,
and to Atlanta, and say, "Look, Industry, we have the capacity down
here in the Carolinas; come and join us." So our interest is not en-
tirely altruistic.
Mr. Wigglesworthj U. S. Department oj Interior: I am concerned
with the loss of energy that occurs- with turbine aeration. Mr. Lee,
in view of the 65 percent loss of energy and the substantially greater
oxygen absorption that you cited, do you believe that the greater
oxygen ab&orption was due to the greater loss of energy, or perhaps
the reverse, that the greater loss of energy was due to the greater
absorption?
Mr. Lee: No, sir, our loss of energy is only very indirectly asso-
ciated with the aeration phenomenon. Our loss af energy results from
not operating at the optimum point on the efficiency curves of the
machines that are installed in the dam at minimum discharges, not
from admitting air to the turbine, The loads we are talking about
are very low with respect to plant capacity. On occasion we operate
at this low load on off-peak when water quality downstream indi-
cates it is needed, but this is at some sacrifice of energy. Some emer-
gency might occur in our system that would preclude such operation,
but insofar as are abje-. we try to he good neighbors and help out
when there is trouble downstream.
We are plagued with the characteristics of the machinery and not
the effect on efficiency due to the air entering1 the turbine.
Dr. Falk, E, 1. du Pout de Nemours and Company: Mr, Wisniew-
ski. if the loss in efficiency means that the power is produced at a
greater cost, who pays this extra cost and who do you think ought
to pay it?
Mr. Wisnietosfci, Wisconsin. Committee on Water Pollution: This
will vary with the power plant installation. For example, the Flam-
beau Paper Company owns the Pixley Dam and the downstream
dams, They figure that the turbine aeration is their contribution
towards helping improve the stream.
We also have installations on the Fox River where a power utility
will own a dam and its customer, a pulp and paper mill, will ask the
utility to provide reaeration at its dam at a critical time. The pulp
and paper mill will then reimburse the power utility for the power
loss.
On the Wisconsin River all the power plant operators are mem-
bers of the Wisconsin Valley Improvement Company and their oper-
ation is coordinated by the secretary of this company. During the
summer, daily dissolved oxygen determinations are made at every
power dam along the river and these are telephoned in to the secre-
tary of the Wisconsin Valley Improvement Company, Depending upon
his listing of the dissolved oxygen concentrations on that stream, he
then issues instructions indicating which dams should operate to pro-
vide reaeration. You all knowh as the temperature changes, the
From the Floor
323
-------
critical zone or the zone of most serious depletion moves up or down
the stream and may not necessarily be in the best position for reaera-
tion. The Wisconsin Valley Improvement Company is trying to add
oxygen as efficiently as possible, and to provide aeration at the
bottom of the oxygen sag curve if possible.
Mr. Grounds, U. S. Army Engineers; Artificial aeration of streams
is apparently one method of maintaining higher dissolved oxygen
levels in critical reaches. Would you comment, Mr. Wisniewski, on
what is being done by your State in the updating of sewage treatment
practices?
Mr. Wisniewski: Wisconsin is probably the leading state in up-
dating sewage treatment practices. We have over 400 communities
that are connected to about 375 sewage treatment plants. These
plants serve 99.9 percent of the sewered population, and provide
treatment that reduces the pollutional load from this population by
76 percent. I challenge anybody to match that!
In the field of industrial waste treatment, the pulp and paper
industry in Wisconsin has built more waste utilization facilities than
there are any place else in the world. The only two yeast fermenta-
tion plants in the western hemisphere are located in Wisconsin. We
are always proud to say that in Wisconsin we provide employment
for 200 people in the production of products from a waste that was
formerly thrown away: spent sulfite liquor. We have a number of
evaporators; there is more Type 316 stainless steel in the Wisconsin
pulp and paper industry than there is in all the rest of the industry
in the United States. The United States' average for recovery of pulp
mill solids, i.e., sulfite mill solids, is 20 percent; Wisconsin's average
is 40 percent. We have the Sulfite Pulp Manufacturers Research
League in Wisconsin. This was made up of the first 15 Wisconsin
pulp mills. Since their organization in 1939, they have spent several
million dollars on research and development of these utilization
processes.
We say in Wisconsin that there is no excuse for stream pollution
by cannery wastes. We worked for years with the National Canners
Association and the Wisconsin Canners Association, and the pollu-
tional load from cannery waste has been reduced by 95 percent. Most
of our methods of disposal involve discharge of wastes on land. Our
milk waste disposal facilities are providing about 67 percent reduc-
tion in the pollutional load.
Mr. Grounds: I did not ask that question to put you on the spot,
but rather to bring out that you are doing other things about trying
to improve your streams. I happen to enjoy fishing in Wisconsin
myself, so I know you do a good job.
Mr. Wisniewski: I wanted to make the point in my talk that
we do recognize that even when we get 90 and 95 percent waste
removal on some of the small streams in Wisconsin, we still have to
go to actual stream treatment in terms of stream reaeration to main-
tain the quality in those streams.
314
DISCUSSION
-------
Let us consider the situation that develops when 1 square mile
of what was formerly farmland becomes urbanized. In Wisconsin we
base the design of our treatment plants on the 7-day minimum flow
in the past 10 years; we can expect that this square mile will dis-
charge 0.1 cfs at its low-flow runoff. A square mile is 640 acres. If
we say 140 acres of that will go for roads, that will leave 500 acres
to build on. If we build four homes per acre that is 2,000 homes;
at an average of three people per home (and that's a low average)
that is 6,000 people. The 6,000 people on that 1 square mile will then
be producing 1,000 pounds of oxygen demand for a runoff of 0.1 cfs.
We must design to meet these conditions. We must depend on water
that is coming off from other parts of the watershed to provide dilu-
tion even for the effluent from a plant providing a high degree of
treatment. So you realize why we must go to high degrees of treat-
ment.
One of the studies we made on self purification of streams estab-
lished that at 24°C, 11 pounds of 5-day BOD would remove one part
per million of dissolved oxygen from a stream. This shows why the
loadings must be low.
Mr. Wilroy, Wiedeman and Singleton Engineers: Mr. Lee, in
dealing with low flows on the Catawba River at Rock Hill, has any
economic study been made on the reaeration that can be obtained in
the stream channel below that dam compared with the reaeration
obtained in the turbine itself? In other words, is your low-flow
operation suppling enough streamflow and additional oxygen to pro-
vide the dilution and oxygen needed downstream by the various
industries located there? And is it being provided in the most eco-
nomical manner?
Mr. Lee: No comprehensive economic study of the type that was
suggested to us yesterday has been made. This situation is very
complex. The low flows we are maintaining are determined by the
needs downstream. They happen to be about eight times the mini-
mum 90-day flow that would occur without regulation provided by
our upstream storage.
Mr. Wilroy: I just wanted to get across that there are a lot of
other factors involved in the provision of minimum flows.
Mr, Lee: We well recognize that.
Mr. Morriss, Atlanta Water Pollution Control: Mr. Lee, il you
discharged treated sewage effluent to a small reservoir with the pro-
posed purpose of offsetting some of the expense of treatment by
generation of power, would you anticipate that the turbine reaeration
itself could play a material part in treating the effluent prior to its
return to the stream?
Mr. Lee: It has occurred to us to build one of these waste storage
reservoirs and say, "All right, Public, let's have no water skiing, no
swimming, and no recreation: this is a lagoon." It is not a very
popular thought.
From the Floor
315
-------
We are putting in very large power installations in terms of
capacity today compared to what we did with the Wylie plant in 1925,
As a result, we are having to set our wheels lower and lower with
respect to tail water elevation to avoid cavitation, not only from a
metal loss point of view, but also from a wheel stability and vibration
point of view. As a result — and we have studied this in more recent
installations — the vacuum breakers will not operate as aspirators
so we must use blowers, as Mr. Wisniewski has done at some installa-
tions in Wisconsin. At one project, which will have a full load dis-
charge of 42,000 cubic feet per second, we found that in order to
increase the tail race dissolved oxygen by one part per million, we
needed some 14,000 horsepower in compressors to supply the neces-
sary air. As a result, we have constructed a submerged weir at that
plant.
With reference to your original question, the economics of power
generation are quite complex, and I cannot answer your theoretical
problem directly.
Mr. Wigglesworth: Mr. Wisniewski, can you account for the
difference in results that you observed compared with those that Mr.
Lee observed?
Mr, Wisniewski: It is difficult to account for the difference. I can
go over our own results and show you why they are accurate.
Here are some possibilities. There must be an increased time of
residence in the turbines, or what you could consider double passing
or something of that sort, which gives a longer exposure of the air
to the water. Also, larger amounts of air are introduced than we are
introducing. Our turbines are operating under normal operating con-
ditions, not under reduced flow conditions. The intention is to add
oxygen at a minimum power loss.
We can get a higher concentration of dissolved oxygen in the
tailwater, but only at a terrific sacrifice in power. We make it a
practice to add air at a rate only slightly higher than the solubility
of air in water, so that we can keep the efficiency of transfer in the
proper range. If, for example, we added three times the amount of
air that is needed, we would reduce the power output by considerably
more than three times the amount of power displaced by the single
unit of air. Our operations are carefully controlled, therefore, to con-
serve power while adding dissolved oxygen.
Mr. Lee: Mr. Wigglesworth, I have a reinforcing surmisal to
support Mr. Wisniewski's comment. I think we are getting multiple
opportunities for absorption of air. At low load the draft tube below
the runner, the elbow, and the exit are enormous structures compared
to the flow of water going through them. There is a relatively long
detention time there. Also, we are measuring the tail race dissolved
oxygen some 300 feet downstream from the power plant. You can
observe air bubbles breaking the water surface all around the sample
area; I think we are getting additional detention as well as aeration in
the tail race.
316
DISCUSSION
-------
IMPROVEMENT IN THE QUALITY OF RESERVOIR
DISCHARGES THROUGH RESERVOIR MIXING AND
AERATION
J. G. Bryan
Engineering Vice President
Aero-Hydraulics Corporation, Montreal, Canada
INTRODUCTION
The economic utilization of our water resources involves the use of
storage basins, both natural and man-made. In nature, a high degree
of purification takes place through the filtering action of the ground,
photosynthetic mechanisms deriving from sunlight and biological and
physical processes depending on oxygen carried in the water and
absorbed at the free surface.
In the various types of impoundments, be they for hydro-electric
power, drinking water, or waste disposal, the above processes of
natural purification are greatly hindered. Thus in deep hydroelectric
reservoirs in many months of the year, generally throughout the
summer, due to a high ambient air temperature and calm wind condi-
tions, the body of water stratifies thermally(1). Then, depending on
the size of the impoundment and the throughput, water may be main-
tained for prolonged periods away from exposure to sunlight and the
atmosphere and may become deoxygenated. Discharge of such water
through the turbines may lead to unacceptable conditions downstream
from the point of view of fish life or purification capacity.
The consequences of such deterioration depend entirely on the
purpose of the impoundment and its role in the overall water-resource
system(2). This concept of a system is beyond the scope of the present
paper, but as part of a program to establish a mathematical model of
a water-resource system, the author was led initially to examine the
economics of various types of reaeration. Enough quantitative data
to enable trends to be expressed analytically were needed to carry
out a computer optimization of the many design variables. The
optimization technique was developed in the nuclear power field
where, due to the background of the people involved, there is a keen
awareness of the quantitative approach. Details of the technique may
be found elsewhere (5); only brief mention is made here regarding its
application in the design of reservoir installations.
Most of the ideas described in this paper found their birth in an
attempt to recreate in a stagnant pool some of the conditions leading
to purification in nature. Aero-hydraulic guns, as used in ice-pre-
vention and aerobic sewage treatment (4), with their ability to circu-
late large volumes of a liquid within its own surface level are used to
stimulate the conditions found in a fast-fiowing river where natural
turbulence leads to mixing and aeration.
Bryan
317
-------
THE AERO-HYDRAULIC GUN
The aero-hydraulic gun consists of a vortical pipe between 6
inches and 3 feet 6 inches in diameter, into which bubbles having
approximately the same cross sectional area as the pipe are intro-
duced intermittently from an air distributor. Figure 1 shows a com-
mon arrangement in which the air distributor is located inside the
lower end of the stack pipe. The complete unit can be constructed
from plastic so that in water treatment applications the equipment is
virtually free from corrosion. In reservoir installations, a commonly
used method of fixing the guns in position is to anchor them to con-
crete sinkers via brackets Because of buoyancy forces, the guns are
designed to assume a vertical position when firing.
In operation, air is fed through the hose and into the air
distributor and lowers the water level in the middle and outer
chambers. This action continues until the water level in the middle
and outer chambers reaches the bottom of the inner chamber or
central standpipe. At this point, the air is free to travel from the
outer and middle chambers to the inner chamber. As soon as the
air breaks the water seal at the bottom of the inner chamber, an
inverse siphon motion takes place, and the accumulated air in all
chambers is suddenly released through the upper opening of the inner
chamber into the stack in the form of a large single bubble.
This rising bubble acts as an expendable piston, forcing the water
above it up the stack and out at the top. At the same time as the
bubble rises in the stack, it draws water behind it, through the ports.
This water continues to flow up the stack, drawn by the same bubble,
Although the bubbles are released intermittently, the flow is con-
tinuous because of the momentum of the water in the stack. The
stream, on leaving the top of the gun behaves as a free turbulent jet
and entrains further quantities of water in its upward movement
towards the surface.
The dimensions of these bubbles are very important. Whenever
the bubble fits loosely in the pipe, liquid slips past the bubble more
rapidly; on the other hand, friction at the pipe wall seems to increase
rapidly as soon as the cross-seetion of the bubble (if it were un-
confined) exceeds that of the pipe. In practice these huge bubbles
have the bullet shape shown in Figure 2.
As the bubbles rise, their volumes increase, and in a uniform
pipe it is thus not possible to preserve their optimum relation to the
pipe cross-section. However, as the linear dimensions of the bubble
vary as the cube root of the volume, these changes are relatively small
and, in fact, only become important for pipe lengths in excess of 50
feet. Then it becomes desirable to use a tapered or stepped pipe, if
pumping efficiency is a critical consideration.
In the determination of the frequency of bubble emission, the
inertia of the water column plays a dominant role. In general, at least
one bubble should always be in the pipe, otherwise a large portion of
the energy required to introduce the air against the hydrostatic
pressure, is used in accelerating the water column.
318
IMPROVEMENT THROUGH RESERVOIR MIXING
-------
Figure 1 — Aero-hydraulic gun.
The word "optimum" has been used several times, and it should
be noted that this must be related to the specific task in view. If an
installation is required for aerating a deoxygenated reservoir through
the production of high velocity surface currents, the design variables
must be combined to achieve this result. A different combination
would arise if the purpose were to destroy thermal stratification in
the reservoir, for, in such a case, the requirement would be for a high
Bryan
319
-------
Figure 2
— Bullet ihape of a p'uton bubble.
rate of volume mixing, irrespective of any velocities that might be
produced.
To carry out such studies, the interrelations of the design
variables have been determined for wide ranges of values. Figure
3 shows a performance chart covering the smaller-sized guns com-
monly used in reservoir installations. Many variables may be repre-
sented on such a chart. If one takes two points that represent condi-
tions with only one independent variable changed, the dependent
variable can be determined.
In Figure 3 the compressed air consumption is the volume of
air used at the pressure existing at the air distributor. With such
a variable, it is unnecessary to include the water depth as a variable.
The chart shown relates to the design of the gun shown in Figure 1
in which the siphon is located at the bottom of the vertical stack
pipe. In such a case, the hydraulic length is the distance over which
the bubble acts as a piston, i.e., approximately the length of the gun.
There are applications in which the pipe is extended considerably
below the siphon in order to draw water from a greater depth than
that to which the air is pumped. In such a case, the chart must be
corrected for the additional entrance losses.
320
IMPROVEMENT THROUGH RESERVOIR MIXING
-------
Figure 3 — Performance of aero-hydraulic unit* in water.
In an optimism study with many variables, the curves of Figure 3
are represented by polynomials, which are programmed for a digital
computer. The chart is useful, however, in showing the general
features at a glance. Thus in the lower right hand corner, where the
air supply rate to a large diameter gun is low, say, 0.02 cubic feet
per second, the bubbles are emitted very slowly, for example once
every 20 seconds. In these circumstances the large pulsations in the
flow absorb energy, causing the output to decrease with hydraulic
length.
AERATION CAPACITY AND MIXING
The oxygenation produced by an installation of guns occurs in
three distinct regions:
1. In the space between the mouth of the gun and the liquid
surface. Here active entrainment of water by the turbulent
jet is occurring and oxygen is absorbed from the large inter-
facial area that is produced when the piston bubbles burst
on emerging from the mouth of the gun. The absorption
process in this region follows the normal law(5) describing
the performance of bubble diffusers, namely,
Nr = KLaV(C, — C) (1)
where,
Nr = rate of oxygen absorption, mg/hr.
KLa = overall liquid film coefficient, per hr.
Bryan
321
-------
V = liquid volume associated with each aeration unit, 1
Ca = saturation concentration of oxygen, mg/1
C = existing oxygen concentration, mg/1
2. Directly above a gun, where the bubbles reach the surface,
in a small area in which the turbulence is of an isotropic
nature. The extent of this area depends on the depth of sub-
mergence of the mouth and may vary from a circle about
5 feet in diameter down to almost nothing. In this zone aera-
tion follows a relation of the form,
N, — KiAi (CH — C) (2)
where the liquid film coefficient, KL, is dependent on the
turbulent velocity, vt. Approximately KL = vt1/2 and Ai is
the surface area over which the isotropic turbulence extends.
3. As the water spreads outward on the free surface, the turbu-
lence becomes non-isotropic and aeration occurs through the
shearing of the surface layer. In this region, which is by far
the most extensive in reservoir applications, the aeration
follows the same form of expression,
Nn = KLAB(CB — C) (3)
but with KIy depending on the square root of the surface
dv
velocity gradient,——, and An is the area over which non-
dz "
isotropic turbulence extends.
Thus, finally, the aeration produced by a system of say, n aero-
hydraulic guns, is given by,
n
N = 2 [c^aVfC, — C) 4- C2As vt1/a(CH — C)
r = 1
+ Cj ("5r) '*A»(C.-C>]r <«>
where Ci, C2, and C3 are constants involving the diffusivity of oxygen
and the temperature dependence of the absorption phenomena.
The values achievable for oxygen absorption depend very much
on the operating conditions, particularly the existing level of oxygen
concentration, C. With deoxygenated water at 20°C, aeration efficien-
cies of 1.6 pounds per brake horsepower are obtained at an aeration
capacity of approximately 1 pound per hour. Of course these figures
are dependent on the values of the design variables, gun dimensions,
submergence, etc., which may not be chosen to maximize aeration
efficiency.
The flow in the vicinity of the mouth of a gun is extremely
difficult to describe mathematically. The zone of flow establish-
ment (6) is ill defined, no doubt on account of the periodic bursting
of the bubbles. At distances greater than five or six times the
diameter of the gun, however, a well defined jet exists. The surface on
322
IMPROVEMENT through reservoir mixing
-------
which the velocity in the axial direction is half the velocity on the
axis, is a cone of semi-angle about 6 degrees.
Over a wide range of flows and depths of submergence of the
mouth of the gun, the effect of entrainment can be represented satis-
factorily by the expression
-55T (5)
where, Qx = total rate of flow of water passing a section at distance x
from the mouth of the gun
Q(> = rate of flow of water emerging from gun
x == distance from mouth of gun, ft
D - diameter of gun, ft
C4 = a constant between 1 and 3 that depends slightly on x
and Q„.
The mixing performance of a gun may be described in terms
of simple mass and heat balances. Thus if an installation is effected
with the guns crossing a thermocline in a reservoir, the homogenizing
of the temperature, T, may be predicted by expressions of the form
(W, + Wa)T = WtT, +W,,T2 (6)
H = KhA(T —TJ (7)
where equation (6) represents mass mixing and equation (7) relates
to heat exchange with the surroundings.
Such a treatment has proved completely satisfactory in predicting
the performance of aero-hydraulic systems for ice prevention where
accurate calculation of the resulting temperatures is often required.
In these applications the vital feature is the total quantity of bottom
water raised to the surface. Pumping rates of about 4 million gallons
per day for the expenditure of each brake horsepower are often
achieved in a properly optimized system.
DESIGN PROCEDURE
In equations (4), (5), (6), and (7), the terms for aeration and
mixing can be related to the design variables, number of guns, their
length and diameter, submergence, spacing, and power supply, by
use of the performance chart and simple properties of jets. One can
then determine the combination of variables required to satisfy some
given criteria.
By having the computation programmed for a digital computer
and by using a modern technique of optimization(7) one can study
the simultaneous interaction of many independent design variables.
As a result of this procedure one is often lead to arrangements that
most probably would not have been noticed in an examination by
conventional methods. The effects of changes in design assumptions
can be studied realistically by comparing one optimum with another.
The criterion of merit is usually the cost of a particular operation —
lowering the temperature in the upper layers of a reservoir, aerating
Bryan
323
-------
a lagoon, and so on. Furthermore, the cost itself is a composite
variable depending on capital and running costs and it is interesting
to study the effect of changes in the relative importance of these
factors.
Figure 4 shows the steps in the optimizing sequence. Once a
mathematical model of the system has been formulated, the evalua-
tion of the criterion of merit is programmed for a digital computer.
This step merely speeds up the process of computation, albeit to a
fantastic extent, The optimization proper consists of a further set of
calculations in which a search is made for the optimum point. The
calculation proceeds automatically, without manual intervention. As
the search proceeds, information about the criterion of merit is stored
and is used to modify and speed up the process. In geometrical
language, one can think of a response surface in many dimensions
corresponding to the independent design variables. As points on this
surface are determined, a knowledge of its topography is gained and
this knowledge is used in further directing the path towards the
maximum or minimum, as the case may be.
Figure 4 — Optimizing sequence for system design.
324 IMPROVEMENT THROUGH RESERVOIR MIXING
-------
Although the hydraulic characteristics of the equipment are rela-
tively straightforward, to determine the effect of completely altering
the flow pattern in a reservoir is vastly more difficult. Here, many of
the important factors involve biological and biochemical changes that
may require many months or even years to evaluate. Moreover,
during such prolonged periods, the surface conditions as determined
by the weather may change violently and make interpretation of
results very difficult. On several occasions after complex instrumenta-
tion had been installed for observations on the breaking up of strati-
fication or reaeration, a strong surface wind caused the reservoir to
overturn, and the results were completely masked.
A program to evaluate the biological effects of thoroughly mix-
ing bodies of water that had been prone to stratification was begun in
1959 and is still under way. Tests are being carried out in Britain,
Ireland, and South Africa and it is hoped that geographical coverage
will enable us to assess the influence of a wide range of conditions.
There are many purely chemical or physical aspects of great im-
portance, however, that have been studied sufficiently to provide a
basis for the optimized design of installations for specific tasks such
as physical mixing, temperature control, and aeration.
PRACTICAL APPLICATIONS
Blelham Tarn
To study the large-scale hydraulic and chemical effects of a
system of aero-hydraulic guns under as controlled conditions as one
is likely to find in nature, an installation was made in a field-
laboratory lake belonging to the Freshwater Biological Association of
Great Britain.
The lake, Blelham Tarn, (Figure 5), is at an altitude of 140 feet
and is close to the western shore of Lake Windermere in the Lake
District. It is 733 yards long, 27 acres in area, and has a maximum
depth of 44 feet and a mean depth of 21.3 feet. Over the past 16
years the Freshwater Biological Association has carefully observed
temperature, dissolved oxygen, chemical nutrients, and the plankton
in the lake.
Between July 31 and August 14, 1961, to provide data for assess-
ment of the physical and biological effects of mixing such a lake
artificially at a time of year when it would normally be stratified,
five aero-hydraulic guns were used to effect its overturn.
The immediate physical effects of the installation during the 14-
day period it was operating are described in Figures 6, 7, and 8 in
terms of the temperature, oxygen content and ferrous iron content of
the water.
The five guns used were 17 feet long and 12 inches in diameter,
and under normal conditions were capable of transferring water from
the hypolimnion to the surface at a rate of 10 million gallons per day.
Bryan
325
-------
To spread the operation over 10 days so as to make observation easier
and to avoid disturbing the bed of the lake, the combined output of
the guns was restricted to 3.5 million gallons per day by reducing the
frequency of bubble emission from the normal rate of 4 seconds to
9 seconds in three guns, and to 6 seconds in the other two. The guns
were located in two pockets deeper than the rest of the lake. Fre-
quent measurements at the 12 stations were consistent, and showed
that records for the single station presented in Figures 6, 7, and 8
represent adequately the changes that took place in the lake as a
whole.
Figure 5 — Blelham Tarn showing positions of the five aero-hydraulic guns and the com-
pressor (depth in meters).
When operations began on July 31, the lake had a marked
thermocline between 5 and 9 meters and a subsidiary one at 3 meters.
As the hypolimnetic water was pumped up and mixed with the
epilimnion, the thermocline moved progressively down, extending from
6 to 10 meters on August 3, and from 7 to 11 meters on August 7. By
August 14, all the water below 13°C, except small amounts in the
deepest hollows, had been removed, and temperature over almost the
whole of the bottom of the lake had risen from 8.7° to about 13°C.
This was somewhat influenced, however, by rain and wind during the
first week.
improvement through reservoir mixing
-------
7 8 9 10 11 12 13 14 15 16 17
TEMPERATURE, °C
Figure 6 — Effect of mixing by aero-hydraulic guns on temperature in Blelham Tarn
(Freshwater Biological Association).
DISSOLVED OXYGEN, m*/l
Figure 7 — Effect of mixing by aero-hydraulic guns on dissolved oxygen in
Blelham Tarn (Freshwater Biological Association).
The distributions of dissolved oxygen and of ferrous iron, whose
presence in quantity is an extremely sensitive indicator of oxygen-
lack, indicate in another way the same sequence of events. On July
31, almost no oxygen was present below 7 meters. By August 7, water
rich in oxygen from the epilimnion had penetrated nearly to 10
meters, and on August 14, there was water 60 percent saturated with
oxygen at 12 meters.
Inniscarra Reservoir
The Inniscarra Reservoir, in County Cork, Ireland, was completed
in 1957 by the Irish Electricity Supply Board. It is situated on the
River Lee, above Cork.
Bryan
327
-------
0
2
4
£
£
E 6
X
£ 8
Q
10
12
'HON (Fe++), mg/l
Figure 8 Effect of mixing by a«ro-hydraulics gunt on iron in Blalham Torn
(Frethwafer Biological Attociation).
The River Lee is an important salmon fishery, and the Board
operates a hatchery designed to produce 1.5 million fry each season.
A salmon liftls built into the dam.
The catchment area is 540 square miles, the average annual
rainfall is inches, and the average annual flow of the river is 970
cubic feet per second. The Reservoir is long and narrow — about 8
miles in length. It has an area of 2,000 acres, and impounds 12,500
million gallons of water. The dam is 800 feet long and 140 feet high,
and the power station has one 15,000 KW set and one 4,000 KW set.
The depth of water at the dam face is 103 feet and the turbine intakes
are between 65 feet and 75 feet. The minimum turbine discharge is 95
million gallons per day. There is a second reservoir of similar area at
Carrigadroghid, 11 miles above Inniscarra.
The Reservoir is subject to a normal spring and autumn overturn,
and, although it is sensitive to high winds, for an average of some 16
weeks between June and October it suffered from stratification in 1959
and I960. This, combined with the products of anaerobic decomposi-
tion, resulted in toxic water passing through the turbines and poison-
ing fish downstream. It is estimated that 16 percent of the water was
hypolimnetic during these periods of stratification; such conditions
are expected to worsen from year to year because of the eutrophic
characteristics of the reservoir.
During a period of characteristic stratification in early August
1959, the temperature variations were between 20°C at the surface
and 16°C at 24 meters, and the 13° isotherm was above 27 meters.
At the same time the oxygen content at the surface was 9.0 milligrams
per liter, but only 3.0 milligrams per liter at 18 meters. High winds
in mid-August caused a complete overturn, but the reservoir re-
stratified to the same pattern early in September. The fishery author-
ities of the Supply Board regard 6.0 milligrams of oxygen per liter
as the desirable minimum and 5.0 milligrams per liter as the per-
missible minimum.
^(V.31, 1961
* issuTir -
(August
_L
August
,U5t 1A. W61
1000 2000 3000 4000 5000
328
IMPROVEMENT through reservoir mixing
-------
In early May 1961, a system consisting of six 12-inch aero-
hydraulic guns, each 44 feet in length, was installed, with a combined
capacity for transferring hypolimnion water to the surface through
their muzzles of 20 million gallons per day. The guns were sited
2,000 feet above the dam, their intake ports being at a depth of 86
feet. Weekly oxygen and temperature readings were taken at the
dam and 1 mile above the dam.
Figures 9 and 10 show the oxygen content reading in the vicinity
of the dam for 1959 and 1961. The 5.0 milligrams per liter lines in
each case have been emphasized, and the depths of the turbine intakes
are shown.
1 "-T 1 1 1 r
JUNE JULY 1 AUGUST ' SEPTEMBER1 OCTOBER
Figure 9 — Dissolved oxygen in Inniscarra Reservoir (in mg/l), in 1959.
MAY
JUNE
JULY ' AUGUST ' SEPTEMBER1 OCTOBER
Figure 10 — Dissolved oxygen in Inniscarra Reservoir in 1961, mg/l.
(Bectric Supply Board, Ireland.)
Bryan
329
-------
May, June, and July 1961 were relatively dry, and the turbines
were off load between May 24 and July 15. The only discharge during
this period was 70 cubic feet per second compensation spilling. The
rainfall recorded at Cork, as a percentage of the 1916-50 average for
the corresponding months, was 40 percent in May, 67 percent in June,
and 60 percent in July. Under these conditions, the reservoir was
relatively static and stratification could have been severe. The read-
ings from June to October 1961 while the aero-hydraulic system was
in operation, however, show that the oxygen content of the water
delivered down-river from the turbine intakes (located at a depth of
21 to 24 meters) was increased considerably.
Accordingly, the results show how such an aero-hydraulic system
can correct a condition of de-oxygenation within a limited area with-
out overturning the whole reservoir.
No fish poisonings were reported during the summer of 1961.
The installation ceased operation on October 27, 1961, when the
autumn overturn occurred. The guns were left in position, to be
operated annually thereafter during the spring, summer, and autumn.
During 1962 the guns operated satisfactorily and there was no
stratification. The installation has required no maintenance and is
ready to operate again this year.
Loch Turret Reservoir
Loch Turret (Figure 11) is a natural lake at an elevation of 1100
feet above sea level in the foothills of the Grampian mountains
near Perth in Scotland. The lake is used as a reservoir for drinking
water and at present has an area of 165 acres. To raise the capacity
impounded, a new dam is being constructed about three-quarters of
a mile below the present outlet. This will increase the area of the
reservoir to about 450 acres and the capacity will be raised to 4
billion gallons.
When the reservoir is extended to the new dam, the top water
level will be raised by some 36 feet and the existing reservoir will
form a pool (Figure 12a). While stratification has not created prob-
lems in the present reservoir, probably because of the natural circu-
lation produced by strong winds from Glen Turret, stratification
possibly will occur in the pool once the reservoir is extended.
If freedom from stratification and anaerobic conditions can be
assured, the basic purity of the water will permit the use of a filtra-
tion plant based on micro-straining and oxidation with ozone. Thus
the high cost associated with full filtration involving chemical co-
agulation can be avoided. In this particular case, a danger arising
with anaerobic conditions is the possibility of iron and manganese
being brought into solution. The maximum permissible concentration
of manganese when treatment is based on the use of ozone is only
0.1 milligram per liter.
330
IMPROVEMENT THROUGH RESERVOIR MIXING
-------
The destratification system proposed is designed to counteract
such harmful effects by continuously pumping large quantities of
bottom water to the surface in the existing reservoir. This is achieved
by self-buoyant aero-hydraulic guns sinkered to the bottom and
supplied with compressed air through submerged feeder lines as
shown in Figure 12. These guns are constructed of low density poly-
ethylene and are approximately 40 feet long and 18 inches in diam-
eter. They can be raised and lowered individually to vary the height
of intakes.
Area 164 acres
Max. Depth 79 ft
Mean Depth 32 ft
Figure 11 — Lock turret (from Bathymetrical Survey of June 9, 1903).
-1170'-
(a) Plan of layout of Aero-Hydraulic guns
Recovery Buoy
/
• Aero-Hydraulic guns
(18" diam x 40' long)
¦ Compressor
— Air Line
18" diam, 40' long gun
(b) Section through installation.
2" diam. air supply
>4" diam. air supply to gun
Figure 12 — Proposed layout of destratification equipment in Loch Turret.
The pumping capacity of the installation is based on rate of rise
of the thermocline in cases where stratification occurs regularly. Bad
climatic conditions similar to those at Loch Turret occurred at Innis-
carra. There, as shown in Figure 9, the thermocline rose at a rate of
about 1 foot per day until the onset of the autumnal overturn. Greater
rates of rise have been noted under hot, still conditions, such as those
Bryan
331
-------
uledtraKS in ®h<^esia- The figure of 1 foot per day has been
iUs Mt thlf f".? on for the Turret installation, although
stratification would not occur at this rate.
den JHemr°nieneSS 0f any Particular body of water to stratify must
wind configuration, the ambient temperature, and
water arfo i is relatively easy to characterize the stability of
mnnLa fmgir0m«thermal gradients alone, but the problem of ac-
donp ln th°r +e f.Cts of wind is much more difficult. Work being
ever £ t? fractions of the atmosphere and the oceans (8), how-
honP thi °Wmg, ?uCh light on this question and there is distinct
hope that an analytical treatment may be useful.
bv JSf/1 r6+SerI?^ is extended> the water level will be raised
present in T t u^h stagnant conditions are not found at
limit for stacrnnnt urr^t, when the level is raised an extreme upper
iT atIn , ?n?n T W°Uld be a layer 36 feet above the bott°m:
.e., at about 1090 average ordinance datum (AOD).
tion aPProximate volume impounded as a func-
or the urmer W 5T ^ If we assum* that the thermocline
per dav the ritJi hypolimnetic water, rises at a rate of 1 foot
gallons per dav the ,hyPolimnion is about 12.9 million
this pumping capacity 3 ™£~hydraullc installation is designed to have
90 cfm free at about ^oIsf«°0mPreSSed 3ir requirement is then ab°
1050 1060 1070 1080 1090 1100 1110 11
ELEVATION OF WATER SURFACE, ft
Figure 13 _ Volume impounded in Loch Turret v« water curiae* elevation.
The total quantity of water brought to the surface by the action
of the system will be many times greater than the above 13 mgd
because of the entrainment in the jets issuing from the guns. In the
system proposed here, about 150 mgd will be brought to the surface.
Although in Loch Turret most of this water will be from the epilim-
332
IMPROVEMENT THROUGH RESERVOIR MIXING
-------
, .. _ tfnns such a continuous rate of
nion lying above the tops of the g ' nt conditions extremely
turnover will make the occurrence
Unlikely- . f fte aero-hydraulic guns. The
Figure 12a shows the layout o 2-inch-diameter poly-
compressor is located at the new dam, reservoir for about 1%
ethylene pipe extends along the be guns are connected to
miles until the deep pools are reac e • (trigure 12b). The whole
this header by 3A-inch-diameter fee e , ^en SUnk by filling
installation will be floated into position ana
certain air chambers with water.
CONCLUSION
the quality of reservoir
In many cases, it is economical to improve ^ q{ compreSsed air
discharges by promoting mass mixi g
and aero-hydraulic guns. interacting design
By a proper consideration of digital computing tech-
variables that is made possible by mo .fic Unction. With this
niques, a system can be optimized i taken into account th
ap^ach to design, many more lactorsc^betaK ^ ^
would otherwise be possible. One may -tfalls 0f oversimplifica
ventional arrangements, and some ulariy important in the d -
tion may be avoided. This may be p . ere there are many in "
sign of a multipurpose impoundmen emjcai, and physical p
acting variables relating to biological, cne
nomena- ovtremely reliable in
The systems discussed here have PJ£J^e> nD excessive strams
operation. If carefully installed lnv;ng - a trouble-free i
are imposed on the pipework during
to io years can be expected. before any definite
Much data is still to be collected the biol°gic^Iraulic
pronouncements can be made regar The large scale y
of turning-over reservoirs a ^ w:<;hed however, to ma
effects have been sufficiently established, n
applications possible.
references and res.
1. Thompson, R. W S. StratWc*tta» ^^ttSlondon, S.W. 1.
ervoirs. Presented at I.C.fci.,
England, Dec. 4, 1953. H. A. Thomas, S. A.
2- Maass, A., H. M. Hufschmidt, R- Dor water_resource systems.
Marglin, and G. M. Fair Dejagn ^ 620pp.
Harvard Univ. Press, Cambridg , lear engineering design.
3- Bryan, J. G. A modern appr0^*^^
Nuclear Energy. PP- 206-16. treatment by total ®*1"
4- Fielding, H. R. New aP?r°a^d digesting. Can. Munic. 11
dation effective in ponding a
101:22-23, 49. Jan. 1963.
333
-------
5. Downing, A. L., R. W. Bayley, and A. G. Boon. The performance
of mechanical aerators. Inst. Sewage Purif. J. Proc. Pt. 3, pp.
231-47. I960.
6. Albertson, M. L,, Y. B. Dai, R. A. Jensen, and H. Rouse. Diffusion
of submerged jets. Trans. ASCE. 115:639-64. 1950.
7. Bryan, J- G., and T. J, Darwent. The shortest route to optimum
design. Engineering. 191:860-61. June 23, 1961; 192:14-15. July
7, 1961.
8. Eckart, C. H. Hydrodynamics of oceans and atmospheres. Per-
gamon Press, New York, N. Y„ 1960.
DISCUSSION
Dr. L. L. Falk
Engineering Service Division, Engineering Department
E. 1. du Pont de Nemours and Company, Wilmington, Delaware
We can predict reasonably well the grossly obvious effects of
stratification on impounded waters. Dr. Bryan points out some of
these effects in his paper: in the deeper layers oxygen may become
depleted, temperatures do not reach high levels, iron and COa con-
centrations rise, etc. And it is true that some of these effects are not
always desirable for subsequent use of the impounded water.
I would like to touch brefly on certain aspects of overturning this
natural stratification. There are, perhaps, some advantages not
touched on by Dr. Bryan, and there are some disadvantages, too. I
shall talk about the significance of overturning in regard to waste
assimilative capacity, heat load dissipation (sometimes called "thermal
pollution"), and biological productivity. These and other factors you
may think of can be particularly significant in multi-purpose im-
poundments.
WASTE ASSIMILATIVE CAPACITY
Mixing of impounded waters can increase waste assimilative
capacity in several ways. The most obvious is to increase the rate of
oxidation of organic pollution loads. This occurs because:
1. Turnover increases reaeration. More water is exposed to
reaeration from the atmosphere as well as to the bubbles
of aeration devices or Dr. Bryan's Aero-Hydraulic Gun.
2. Turnover raises the average water temperature. This
raises the overall reaeration rate coefficient, commonly
called K2, for the entire body even though the tempera-
ture at the surface is somewhat lower than in the strati-
fied condition. Of course, higher temperatures also raise
the overall deoxygenation rate constant, K,, so that oxy-
334
DISCUSSION
-------
gen is used up at a faster rate. However, I think that the
overall effect will be to raise the organic load assimilative
capacity of the impounded water, given a specific mini-
mum dissolved oxygen criterion.
Another way that turnover increases waste assimilative capacity
is by reducing the effects of density currents. Fry, Churchill and
Elder (I) found density currents in Cherokee Reservoir on the TVA
system. Water entering the reservoir in the fall was heavier because
of lower temperatures and higher chloride concentrations. A density
current traveled along the bottom for the entire 47-mile length of
the reservoir. One can visualize that mixing of the reservoir contents
in its upper reaches would destroy this density current.
HEAT LOSS
The atmosphere is the heat sink for the processes of human civili-
zation. Excess heat from industrial processes is dissipated either
directly to the air (e.g., air coolers, exposed equipment, etc.) or in-
directly with water as the heat transfer medium. When water is used,
its release to a stream may result in thermal pollution. This happens
when heat release to the receiving water is higher than the heat
transfer capacity of the stream to the atmosphere; the temperature
rises to a higher equilibrium level and may affect water quality for
a subsequent use. Many people do not recognize that this would be
a more serious problem if it were not for the fact that the heated
cooling water is lighter than the receiving stream. It floats on the
top where contact with the atmosphere allows rapid heat dissiptation
by evaporative cooling, convection, and radiation. This loss of heat
may be even faster in impounded waters because of less turbulence
and mixing to deeper depths. If the need to maintain the warm water
on the surface is not recognized and the impoundment is artificially
mixed for other reasons, the rate of heat release from the water will
be less, and undesirably high temperatures throughout the entire
impoundment may occur.
There is another aspect to the heat problem. Power plants on
large impoundments often take advantage of the cooler hypolimnion.
A curtain wall prevents warmer, upper waters from reaching plant
intakes; cooler, deeper waters are preferentially used with consider-
able economic advantage. Mixing such a reservoir would result in
an economic penalty for any cooling process.
PRIMARY PRODUCTION
Multi-purpose impoundments serve many uses: flood control,
pollution control, power production, recreation, wildlife development,
etc. I wonder what effect complete mixing would have on biological
productivity. The production of desired fish life depends ultimately
on the primary production of algae — the starting point of the bio-
dynamic chain or cycle. A mixed reservoir will have a lower temper-
ature in the upper layers where most algae grow (solar energy avail-
ability). Will this lower temperature reduce algae productivity?
Falk
335
-------
Algae convert carbon dioxide to cell material. In an unmixed
stratified impoundment the carbon dioxide concentration in the deeper
waters is several fold higher than in the upper layers. Some surely
diffuses upward for use by the algae. If the waters are completely
mixed, will the loss of carbon dioxide by diffusion to the atmosphere
reduce the algae productivity?
On the other side of the coin, the mixing of reservoirs can expose
more water to the solar energy zone. More nutrients become available
to algae. This factor would increase algae productivity. One also
wonders whether this would favor algae known to create taste and
odor problems in reservoirs.
The literature is not clear on the effect of mixing on algae pro-
duction. Riddick (2) studied the effects of forced circulation of a 103
million-gallon reservoir by aerators at Ossining, New York. He found
that this seemed to inhibit the algae.
CONCLUSION
Mixing of impounded waters on a large scale has several advan-
tages, as Dr. Bryan's paper points out. Where the impoundment is
used for many purposes, however, the effects of the mixing must be
looked at from many points of view. What may be an advantage
from one standpoint may turn out to be a disadvantage from another.
The total credits and liabilities must be balanced in determining the
benefits of reservoir mixing.
REFERENCES
1. Fry, A. S., M. A. Churchill, and R. A, Elder, Significant effects of
density currents in TVA's integrated reservoir and river system.
Proc. Minn. Intern. Hydraulics Conv. IAHR and ASCE, Minne-
apolis, Minn., Aug. 1953. p. 335.
2. Riddick, T. M. Forced circulation of reservoir waters. Water &
Sewage Works. 104:231. 1957.
DISCUSSION FROM THE FLOOR
Dr. Ingols, Georgia Institute of Technology: I could not help but
notice the temperature of the water at the surface of the lakes Dr.
Bryan showed in his figures, and wonder whether one would ever
have a recreational lake if 14° and 15°C water temperature were
common to the Atlanta area. It chilled me just to look at those figures.
Dr. L. Williams, U. S. Public Health Service: I have been in
charge of the plankton program of the National Water Quality Net-
work for nearly 5 years, and in that capacity have been examining
waters from the major waterways throughout the nation. I am there-
fore familiar with the plankton counts at various locations and have
some idea as to what algae, plankton, and microorganisms can do to
336
DISCUSSION
-------
water. I have been very much impressed with the fact that water
coming from the bottom of impoundments contains very few micro-
organisms; on the other hand, samples from many of our impounded
stations contain total counts of from 80,000 to 120,000 per milliliter.
If these organisms contribute oxygen in large amounts, much of it
must be dissipated back to the air so that it is not available for maxi-
mum self-purification of streams. It would seem that nature can pro-
vide the best way to handle organic material, both that contributed
by productivity of the algae themselves and that contributed by
wastes from domestic, industrial, and other sources. The discharge
of some of the surface waters from an impoundment not only pro-
vides dissolved oxygen, but also provides seeds that make it possible
for the stream below the impoundment to recover more rapidly by
making its own oxygen while producing organic material needed in
the food chain for fish.
For example, in the St. Paul area, at Red Wing, Minnesota, there
is a large impoundment. The algal counts there are enormous, yet
the fish productivity is also very great; as long as the blue-green algae
do not come in, the dissolved oxygen never reaches low concentrations
and taste and odor problems do not occur. We need to make better
use of the aquatic organisms available in the stream. When an or-
ganism provides oxygen, essentially all of it can be absorbed, while
much of the oxygen injected into water as a jet of air cannot be
absorbed.
To provide for further photosynthesis downstream, therefore, it
seems desirable to use sufficient of the surface water to mix with the
water discharged from the lower levels of the reservoir.
As a further consideration, in an impounded situation algae
themselves become BOD. They eventually settle out as detritis and
then contribute to the depletion of the available oxygen in the bottom
of the impoundment. This is highly undesirable, and should be con-
sidered mismanagement. By the discharge of some of the surface
waters we can reduce the deoxygenation of the hypolimnion and use
the organisms to provide oxygen and food; thus we can perhaps use
nature to help solve the problem of pollution and still have fish and
wildlife.
Dr. Symons, U. S. Public Health Service: Dr. Bryan, since you
can mix and reaerate an impoundment through simple pumping with
a large-volume low-head pump, I wonder if you would express an
opinion as to the relative importance of your device as a high-capacity
pump, or as a diffusion device.
Dr. Bryan, Aero-Hydraulics Corporation: I think the merit of
the device is almost entirely due to its high pumping capacity.
In relation to what Dr. Williams said, the basic idea in the devel-
opment of this equipment was to try and recreate some of the puri-
fication processes that take place in nature. The idea was, in fact, to
lift water from the hypolimnion to the surface, so as to expose that
water to natural processes of aeration. Most of the aeration takes
From the Floor
337
-------
place on the surface; very little takes place in the zone between the
mouth of the pipe and the surface.
Dr. Symons: If that is true, I wonder how your device compares
economically with an ordinary pump.
Dr. Bryan: I think it compares favorably, especially in view of
the tremendous simplicity of the equipment. There are no moving
parts submerged, which is, of course, a great advantage in many sites
that are very remote. If the dimensions of these large piston bubbles
are properly proportioned to the pipe, the slip is very low, indeed,
and most of the work that is done in getting the air down to the bot-
tom is, in fact, recovered. This, of course, is not so with the ordinary
airlift pump that works with a continuous supply of air and largely
viscous drag.
Mr. Wilroy, Wiedeman and Singleton Engineers: I would like to
point out that there are definite economic considerations in reservoir
aeration. In Georgia we are concerned with a small water supply
reservoir where we have a manganese problem that persists for about
3 or 4 months during the fall. Removal of the manganese from the
two mgd used each day costs between $1,500 and $1,800 a month.
Therefore, if we could turn over this reservoir for something less
than the $4,500 to $7,500 expended each year for treatment we would
consider it economical. This is definitely a possibility for preventing
the formation of the manganese. We are considering the cost of using
either an airlift pump or a mechanical mixer.
Dr. Ingols has been working fairly closely with this problem.
We have been concerned with just the removal of manganese alone.
Other benefits may accrue, but we can definitely tie down the eco-
nomics of the manganese removal.
Mr. Summers, Missouri Water Resources Board: Dr. Ingols, I
have been slow in putting all this together, but it has occurred to me
that no mention has been made of turbidity in any of the papers
presented. I am wondering if perhaps the turbidity, especially the
colloidal turbidity, might be of value in getting the oxygen into the
water. Here we would have the benefit of adhesion, which would
allow a longer time for the oxygen to dissolve in the water. This
has come to our attention in Missouri, particularly in the Missouri
River where the recovery rate is more rapid than we had anticipated.
The only reason for this we can think of is that it contains a lot of
colloidal turbidity. Is it possible, therefore, that the difference in
results between Wisconsin and the Carolinas might be related to
turbidity?
Dr. Ingols: I am not qualified to answer that question. I will be
glad to entertain an answer from someone in the audience.
Mr. Wilroy: I would like to comment on the effect of turbidity.
On the Chattahoochee River below Buford Dam, we have had tre-
mendous reductions in turbidity as a result of the construction of
the dam. Prior to the dam, the turbidity ranged from a low of 600
338
DISCUSSION
-------
units to several thousand. At the present time it drops to the range
of 30 to 40 units and persists for months at a time. Simultaneously
with this reduction in turbidity, we have experienced a tremendous
algae bloom in the river each spring. This creates problems for the
waterworks along the river, occasionally from a taste and odor
standpoint, but more importantly, from a pump-clogging standpoint.
The algal blooms have been severe enough to result in clogging,
which required that the pumps be pulled down and several hours
spent in getting the algae growths out. In 1961, we estimated that,
with an average flow of around 2,000 cfs, between 200 and 300 tons
of algal material was passing down the river each day.
Dr. Bryan: I would like to say a few words about the problem
of manganese. In the Loch Turret Reservoir, shown in my paper,
the water is of a fairly good quality, but there is a slight color prob-
lem. We thought this could be oxidized out, using ozone. The pres-
ence of even minute quantities of manganese (perhaps 0.1 to 0.2
milligram per liter), however, rendered treatment by ozone relatively
ineffective. One of the main incentives for mixing in this particular
installation is to keep aerobic conditions at the bottom of the deep
pool that will be formed by the existing reservoir to prevent manga-
nese from coming into solution.
Dr. Falk's comments raised some most useful points that add to
my already very long list of unsolved problems.
Mr. Swiggart, U. S. Army Engineers: Yesterday Mr. Moore of
the Tulsa District mentioned that we are building reregulating struc-
tures below some of our hydroelectric multiple-purpose projects. One
example is the Keystone Reservoir about 17 miles up the Arkansas
River from the City of Tulsa, Oklahoma. The purposes of the project
include hydropower, flood control, and flow regulation to improve
water quality conditions at Tulsa or at least to maintain the condi-
tions existing prior to construction of the dam. Since we are talking
about a 5 to 10 percent load-factor plant, both daily and weekend
fluctuations will occur. We have designed and are building a reregu-
lating structure to take care of the flow sag due to hydroelectric
power generation.
Now that we are taking care of the flow sag, the problem that
concerns us is, "What quality of water is going to be coming out of
the reservoir?" So far during this conference I have not heard how
we can forecast whether or not we will have a problem with the
water discharged? In the White River Basin in Arkansas we have
the Bull Shoals Project, which has a head of from 170 to 200 feet
for hydroelectric power generation (a 34-foot drawdown) and the
Norfolk Project, which has a head of 170 feet. Our main problem
below these reservoirs is keeping the trout fishermen out of the tail
race.
Trout have been introduced below Tenkiller Ferry Reservoir on
the Illinois River in Oklahoma, but we have no fish life in the first
8 miles below the reservoir.
From the Floor
339
-------
Can't we forecast whether we are going to have a problem with
the water discharged? It will not do us much good to build reregulat-
ing structures if we are still going to have water that will not support
the desired uses. Since most of our hydroelectric projects are mar-
ginal, financially, anyway, we cannot spend a lot of money building
some kind of a skimmer when we do not need it. Further, I do not
know what kind of a skimmer we could build where we are going
to have 30 to 40 feet of drawdown.
Mr. McLean, U. S. Public Health Service: I think the design of
the Smith Mountain pumped storage project being built by Appa-
lachian Power Company in Virginia provides a logical method for
the correction of low dissolved oxygen. There, one of the intakes
extends to a higher level providing an opportunity for mixing surface
water with the hypolimnetic water when needed.
Mr. Swiggart: My primary question is, how can you forecast in
advance whether you will have a problem of stratification and dis-
charge of bad water? All of the projects in our Division are good
except one. If we had known we would have the problem of pro-
viding dissolved oxygen maybe we would not have built the project
to begin with, because it would not be justified economically consid-
ering the money that would have to be spent in providing these
features.
Mr. McLean: We have not been able to forecast this in the past.
I hope, from our experience with the dams that have been built, we
will be able to forecast the quality of water that will be discharged.
The Public Health Service has initiated research in this area under
Dr. Woodward, Mr. Weibel, Dr. Symons, and Dr. Irwin in the Basic
and Applied Sciences Branch. Several Corps of Engineers offices have
done a considerable amount of work in observing water quality
changes in and below their reservoirs, and we are discussing ways of
working with them on it. At this time, however, we can not say
exactly what will happen in a specific proposed reservoir.
Dr. L. Williams: I believe that whether or not the organisms
will develop can be predicted from a survey in advance of the con-
struction of an impoundment. If the enrichment of the water is such
that it is very highly productive, one can predict that large crops of
algae will be produced, and that when they settle to the bottom they
will decompose, using up the oxygen in the lower levels of the
reservoir.
Dr. Ingols: I have noticed that during the spring turnover in
Bufford Reservoir the water is completely uniform and the Chatta-
hoochee River below the dam is extremely green. Once stratification
sets in, all the algae trapped below the thermocline are going to con-
tribute to the organic matter in this layer; we have to consider this,
as well as how much organic matter is going into the reservoir, in
order to evaluate the productivity.
Mr. Whitehouse, Central Electricity Research Labs, England:
I would like to comment on this problem of forecasting the possibility
340
DISCUSSION
-------
of biochemical stratification developing, even if thermal stratifica-
tion does not. In the planning of the Chew Valley Lake Reservoir
for water supply storage the biologist carried out a survey of the
river prior to impoundment. Based on the mineral composition of
the water and the biota present, he predicted with almost absolute
certainty that biochemical stratification would set in, given certain
physical conditions. These conditions were that there be high insu-
lation rates in summer and low wind velocities. When these condi-
tions occurred, sure enough, biochemical stratification developed along
with all its obnoxious effects. But the biologist had had the foresight,
as a result of investigating the river in advance of impounding the
water, to suggest to the engineers involved in the project that they
build a multiple-level intake. On the basis of his advice, they built
an intake tower so that they could draw from the surface, the middle,
or the bottom of the reservoir. Depending on the depth to which bio-
chemical stratification had risen from the bottom, they were able to
determine which level intake to use. Except in one instance where
blue-greens developed because the nutrient level was sufficiently high
to allow them to grow, they have not had a problem of taste and odor
in the drinking water delivered to the residents of the city supplied
by this reservoir.
Dr. Ingols: This is one of the parts of the program that I think is
tremendously significant. I would like to inject two points, if I may.
At a luncheon recently I talked with somebody who said, "We
do not have any problem with the hypolimnion, we simply inject
chlorine into it." Ray Derby, from Los Angeles, reported they are
introducing chlorine into the hypolimnion to control the quality of
water so that it will not require taste and odor treatment before it
goes into the distribution system. This is a direct reservoir supply
with no flocculation or filtration.
The other point is that I know of several reservoirs that are re-
ceiving wastes without interference with recreation because of the
presence of a thermocline and the discharge of the waste into the
hypolimnion. The surface waters are being used for recreation even
above the point of waste discharge. It has always been interesting
that one can watch fish being taken from the lake above the outlet
because this is where the fish accumulate. This is apparently the most
productive point in the lake. The waste is discharged after treatment
on a trickling filter. I hardly need say that a trickling filter effluent
contains sensible numbers of worms, larvae, and so forth, and that
this, of course, makes excellent fish food, particularly for the scav-
enger fish.
From the Floor
341
-------
Session 6
OPERATION AND ADMINISTRATION
OF FLOW REGULATION PROJECTS
Moderator: W. E. Bell
U. S. Public Health Service
-------
WATER QUALITY MONITORING FOR
WATER QUALITY CONTROL
Stanley Ragone and B. J. Peters
Virginia Electric and Power Company, Richmond, Virginia
Peaking hydroelectric power is a great asset to the modern elec-
tric utility that generates the major portion of its system electric load
with steam. The fact that hydroelectric generators can be started and
brought up to full load in a relatively few minutes provides excellent
protection to the power system in the event a large steam unit sud-
denly fails in service. Its main purpose is, however, to provide gen-
eration for the peak demand periods, thereby reducing the daily
starting, loading, and shutting down of older steam-generating units
with the lowest efficiencies in the power system.
On the Roanoke River in North Carolina, the Virginia Electric
and Power Company now has two peaking hydroelectric stations
that are operated in conjunction with the U. S. Corps of Engineers'
John H. Kerr and Philpott hydroelectric projects, which are also
peaking stations.
Prior to any discussion of our experience in flow regulation and
monitoring for water quality control at our Roanoke River projects,
it is necessary to briefly explain the operation of these peaking hydro-
electric stations in conjunction with our steam stations. Figure 1
shows a typical winter daily load pattern for the Virginia Electric
and Power Company, with our steam energy carrying the major
portion of the system load and with the available hydroelectric energy
taking -care of the peak portion of the load cycles. The Roanoke River
does not have sufficient flows to provide energy around the clock
throughout the year, and it is necessary to schedule the water flows
for maximum utilization of the energy available. The figure also
shows that some energy is used each hour (small rectangles at top of
steam energy generation) to maintain minimum streamflow require-
ments from the station farthest downstream. This variation in river
flows, from a minimum that will satisfy the Federal, State, and local
agencies to the flows necessary to provide the most economical elec-
trical energy production and at the same time provide maximum
dissolved oxygen content, was the prime reason that the Virginia
Electric and Power Company undertook elaborate studies of reservoir
and river flow conditions, oxygen monitoring, and hydroelectric
station design and operation. The following discussion includes the
more important phases oi these studies pertaining to water quality
monitoring and flow control of the Hoanoke River.
The Roanoke River is an interstate stream rising in the mountains
of Virginia, entering North Carolina in the Piedmont Plateau, and
terminating in the Albemarle Sound, Figure 2 shows the Roanoke
River Basin and the following hydroelectric stations, (1) Philpott —
14 MW (U. S. Corps of Engineers), (2) John H. Kerr — 204 MW
(U. S. Corps of Engineers), (3) Gaston — 200 MW (Virginia Electric
and Power Company), (4) Roanoke Rapids — 100 MW (Virginia
Ragone and Peters
345
-------
E 1/1300
OJ-S
ujO g 200
Sioo
wZ Q)
Figure 1
12 N 12 N 12 N 12 N 12 N 12
PM PM PM PM PM PM
MONDAY TUESDAY WEDNESDAY THURSDAY FRIDAY SATURDAY SUNDAY
NOV. 26 NOV. 27 NOV. 28 NOV. 29 NOV. 30 DEC. 1 DEC. 2
Typical daily load pattern of Virginia Electric and Power Company for week
ending December 2, 1962.
Electric and Power Company), and (5) Smith Mountain dumped
Storage Project — 480 MW (Appalachian Power Company — now in
the final stages of construction). The drainage area of the Roanoke
River above the USGS gage at Roanoke Rapids is approximately 8,400
square miles. The average river flow at this gage prior to regulation
was 8,364 cfs, with a maximum discharge of 261,000 cfs and a mini-
mum of 458 cfs. The 7-day once in 10-year minimum flow was 925 cfs.
In August 1950, the U. S. Corps of Engineers' multipurpose John
H. Kerr Dam began regulation of flows in the lower Roanoke River
Figure 3 shows the section of the Roanoke River from Kerr Dam to
Roanoke Rapids. The minimum flows provided by Kerr Dam were
never clearly defined in the Definite Project Report; however, they
closely approached the minimum flows later provided by the Vir-
ginia Electric and Power Company's Roanoke Rapids project under
its license from the Federal Power Commission.
The Virginia Electric and Power Company had planned for de-
velopment of their Gaston and Roanoke Rapids hydroelectric sites
on the Roanoke River for a number of years and in 1948 applied for
a license to construct the Roanoke Rapids Dam and hydroelectric
station. Litigation ensued for 4% years and finally, in 1953 the
Supreme Court confirmed the license for the project and construction
was started immediately. The Roanoke Rapids Dam assumed the
regulation or the lower Roanoke River upon its completion in July
346
MONITORING FOR WATER QUALITY CONTROL
-------
w
•4 Figure 2 — Location map of Roanoke River Basin hydroelectric projects.
-------
-------
of 1955. The instantaneous minimum flow requirements initially
established by the Federal Power Commission license are shown in
Table 1.
TABLE 1. INSTANTANEOUS MINIMUM FLOW REQUIREMENTS
ESTABLISHED BY FEDERAL POWER COMMISSION FOR ROA-
NOKE RAPIDS DAM
Month
Instantaneous minimum flow, cfs
Weekday
Weekend
January
500
500
February
500
500
March
500*
500®
April
500»
500a
May
1250a
600*
June
2000*
1000a
July
2000
1000
August
2000
1000
September
1600
800
October
1250
600
November
500
500
December
500
500
aStriped bass spawning flows: minimum of 2,000 cfs for 75
days to begin no earlier than March 15 and to end no later
than June 15, at the request of the North Carolina Wildlife
Commission.
In the 3-mile stretch immediately below the Roanoke Rapids
Dam, an untreated industrial and domestic waste load, with a popu-
lation equivalent of 590,000(1), is discharged into the river. The
reduced weekend flow was based on the premise that a reduction in
this waste load would occur over the weekend. This was not the
case, however, and at the request of the Division of Water Pollution
Control, North Carolina State Board of Health, the Virginia Electric
and Power Company agreed to modify these flows, with the consent
of the Federal Power Commission, to discharge more water on week-
ends for stream protection and to lower weekday minimums. The
following revised schedule of minimum flows for the critical hot
weather months has been observed since June 1056.
In 1951, prior to issuance of the Roanoke Rapids license, the
Virginia Electric and Power Company petitioned the Federal Power
Commission for a license for the Gaston project to be located 8 miles
upstream from the Roanoke Rapids project.
In 1953, the Virginia Electric and Power Company applied for
a license change in the design and electrical capacity of the Roanoke
Rapids Project. During the hearings concerning these modifications,
interveners brought up the new problems of stratification in reser-
Ragone and Peters
349
-------
voirs, low dissolved oxygen, density currents, iron, and manganese.
The interveners claimed that the low-level intakes in the Roanoke
Rapids and Gaston Projects would allow the cold oxygen-deficient
discharge of Kerr Reservoir to move as a density current through the
lower reservoirs and emerge from the Roanoke Rapids Reservoir with
such a low dissolved oxygen content that waste being dumped in the
river could not be assimilated and the lower Roanoke River would
become a dead stream. This brought severe protests from the sport
and commercial fishing interests, because the Roanoke Rapids —
Weldon area is the major breeding ground of the Albemarle Sound
species of the striped bass.
TABLE 2. REVISED INSTANTANEOUS MINIMUM FLOW
REQUIREMENTS FOR ROANOKE RIVER DAM (June 1956)
Instantaneous minimum flow, cfs
Month Weekday Weekend
June 1500" 2000
July 1500 2000
August 1500 2000
September 1200 1600
October 900 1500
"Spawning flows are the same as in Table 1.
Faced with this new problem, the Company immediately hired
consultants and started studies of the Kerr Reservoir and the Roanoke
River. This testing was put on a regular schedule and involved deter-
mination of dissolved oxygen, temperature, pH, iron, and manganese
at 10-foot intervals of depth in Kerr Reservoir and at five stations
over a 7 7-mile stretch of the river below the Kerr Dam.
Based on the data obtained by Virginia Electric and Power Com-
pany on the Roanoke River and information obtained on other reser-
voir systems, our consultants in 1955 recommended the construction
of a submerged weir or curtain wall around the turbine intakes as part
of the proposed Gaston Project. This weir was to function as a high-
level intake that would discharge the upper-level high-dissolved-
oxygen waters from the project and prevent density current flow-
through in the reservoir system.
Severe doubts and objections were raised to this plan by inter-
veners to the Gaston license application, and finally, the Company
agreed to construct a weir to within 25 feet of the surface in front of
the Roanoke Rapids turbine intakes to allow testing of such an
installation. Figure 4 illustrates the weir construction and configur-
ation. Since the Roanoke Rapids reservoir had been filled 2 years
before and the station was in operation, the construction of this weir
was a difficult engineering feat. Figure 5 shows how the rock fill
was handled.
350
MONITORING FOR WATER QUALITY CONTROL
-------
6*
W
Figure 4 —
Plan arid elevation of Roanoke River Rapids hydroelectric station.
-------
figure 5 — Placement of rock fill for Roanoka Rapidi w«ir.
This work was completed in July of 1957 at a cost of approxi-
mately $210,000 and an intensive survey of the performance of the
weir was conducted. This survey, cooperated in by personnel of the
Division of Water Pollution Control, North Carolina State Board of
Health, North Carolina Wildlife Commission, U. S. Public Health
Service, the U. S. Fish and Wildlife Service, and Virginia Electric
and Power Company and its consultants, extended from July 15
through September 15, 1957, and had up to 24 men working in the
field at times. Dissolved oxygen and temperature profiles were con-
ducted throughout the Roanoke Rapids Reservoir, and dissolved
oxygen and temperature analyses were made in the Kerr and Roanoke
Rapids tailraces. In addition, water current studies were conducted
around the weir to determine its effect on flow patterns and hydro-
electric generation.
Based upon the data obtained during the 1957 survey, the efficacy
of a weir was proven. The Gaston Project weir was designed to rise
to within 15 feet of the surface in order to draw the uppermost layers
of reservoir water into the turbines. The installation of a weir can
create an appreciable hydraulic head loss, with a corresponding re-
duction in generation. Because of this a thorough study was made to
determine the proper length of weir for full utilization of the avail-
able hydraulic head. This study indicated that the length should be
approximately 1000 feet, and the weir was designed accordingly
Figure 6 shows the design of the weir and its configuration. The
Virginia Electric and Power Company was able to bring the top of
352 MONITORING FOR WATER QUALITY CONTROL
-------
this weir to within 15 feet of the normal water level of the Gaston
lake because variations in level for this project during normal power
operations will be less than 1 foot. This was not true of the Roanoke
Rapids lake, and with wider variations in lake level, the weir could
only be brought to within 25 feet of the surface. Figure 7 illustrates
the overall construction of the Gaston weir upon completion, and
Figure 8 is a closeup view showing the rock-filled concrete cribbing
that was a more economcal installation than the construction of a
completely rock-filled weir of the same height. The Gaston weir cost
approximately $650,000, which was considerably more than the ^oa"
noke Rapids weir. This was primarily due to the greater height an
length of the Gaston weir.
Figure 6 — Plan and election of Gc.ton hydroelectric station.
Virginia Electric and Power Company continued its bi-monthly
surveys of the Kerr and Roanoke Rapds reservoirs and the river below
these reservoirs by graduate chemists and engineers. Figure 9
illustrates a typical survey report. In 1958 it was decided to add
routine around-the-clock testing of the Roanoke Rapids tailrace for
dissolved oxygen. To accomplish this, it was necessary to provide
facilities and train the power station operators at Roanoke Rapids
in the necessary techniques of Winkler tests for oxygen. In addition
to this, arrangements had to be made for the operator to obtain a
representative sample of the tailrace discharge at any time. This
required the installation of a submersible pump suspended from a
raft located in the tailrace of the Roanoke Rapids project from which
a continuous sample was piped into the Roanoke Rapids station.
Figure 10 is a close up of the raft showing its construction. This
installation cost approximately $500, but it assured the availability
Ragone and Peters
353
-------
of a continuous sample of the discharge of Roanoke Rapids for the
operators conducting the tests. Oxygen tests were run and recorded
every 4 hours by the operator unless conditions indicated analyses
were required more often. During minimum flow conditions and/or
during the critical summer season, tests were run hourly. Two years
of testing indicated that reliable determinations of oxygen in the
discharge from Roanoke Rapids could be made by station operators,
and we concurred with the issuance of a license in 1960 for the
Gaston project. This license included requirements for minimum
instantaneous and minimum poundage per day of dissolved oxygen
in the Roanoke Rapids discharge, in addition to minimum flow
requirements. To our knowledge, this is the first license to contain
such a provision. Article 25 of the Federal Power Commission
license for Gaston, which pertains to these minimum flows and
dissolved oxygen, is shown in its entirety in Appendix I. Con-
struction was started almost immediately on the Gaston project, with
completion planned in 1963.
Figure 7 — View of completed Goiton weir looking upitream.
With the dissolved oxygen provisions of the license, the Virginia
Electric and Power Company wanted to provide an even more reli-
able and continuous means of monitoring the oxygen content of the
Roanoke Rapids discharge. Considerable experience had been ob-
tained in the Virginia Electric and Power Company steam stations
with dissolved oxygen analyzers and recorders on condensate in the
steam cycle and with flue-gas oxygen analyzers and recorders of
354
MONITORING FOR WATER QUALITY CONTROL
-------
various manufacture. The problem of continuous recording of dis-
solved oxygen in river water was discussed with several instrument
manufacturers. It was decided to install a newly designed Hays dis-
solved oxygen analyzer and recorder on the same Roanoke Rapids
discharge sample being piped into the station for the operator
analysis. The installation of the Hays equipment was completed in
1960 at a cost of approximately $4200. Figure 11 shows the com-
pleted installation and the small test bench used by the operators
at Roanoke Rapids. It was necessary to increase the size of the sub-
mersible pump suspended from the raft to supply the water flow
requirements for the analyzer and for operator analysis. The cost of
this larger pump is included in the total cost of $4200.
Figure 8 — Rock-filled concrete cribbing in (he submerged weir in Gallon Reservoir.
Winkler tests have been run by the operator every 4 hours from
March through October and every 8 hours the remainder of the year
since the installation of the analyzer and recorder to check the
dependability and accuracy of the recorder. During the 3% months
following the installation of the continuous analyzer in July 1960,
630 analyses were run by the operator and compared to the dissolved
oxygen determinations by the continuous analyzer. These tests indi-
cate that 85 percent of the recorder readings were within 0.2 milli-
gram per liter oxygen and over 95 percent were within 0.3 milligram
per liter of the corresponding operator's oxygen determination. Be-
cause of the difficulty with analyzing for oxygen by relatively un-
skilled personnel, it is believed that analyzer-recorder determination
Ragone and Peters
355
-------
of dissolved oxygen is more reliablp _
analyses. rename and accurate than operator's
Kmtr R*«wf»oif I
Ro R. Rturvoir |
10840
Scroti Cam*
Toll Raci
U.3. ini iw.iaon) South
U^JOI iW.Uon) Ulddb
Uj.1^ 301 IW.Ida.1 Norlh
Pm»» lr.1.1 IFlll.. Plan I
Plain CHIo.nl (FII1.1 Pk.nl
Jul* 16. 1M2 K.rr R.M"0lr *ur cl«r. >urr.c. MU - tot. ,mns sW> „r taBtritu„ M„r „ i00 „„
Jul, !7. >Nz l»»o« WW. *«.rvo,r: ««,, turtw. =lMpp7 . ^ ^ . „r ^ ^ ^ ^ ^
Figure 9 — Typical Roanoke River data report.
Virginia Electric and Power Compant
ROANOKE RIVER DAT*
•" \ -«2
Figure 10. Raft in Roanoke River tailrace.
356
Monitoring for water quality control
-------
I
Figure 11 — Hays dissolved oxygen analyzer at Roanoke Rapids power house.
Except for a few maintenance problems soon after installation,
primarily due to shipping and wiring damage, the analyzer and the
recorder have required very little maintenance. In fact, no technical
personnel are available any closer to Roanoke Rapids than Richmond,
Virginia, and the maintenance problems associated with this equip-
ment have been such that only a few trips for servicing the instru-
ment have been required in the past 2 years. Maintenance costs have
been less than $200 per year. The station operators have been in-
structed in the proper procedures for regularly zeroing and calibrating
the analyzer and recorder. After calibration, if their analysis deviates
by more than 0.3 ppm from the analyzer, they are instructed to call
the chief chemist and start running oxygen analyses hourly until the
instrument can be checked and repaired if necessary.
The chart from this analyzer and the daily dissolved oxygen log
sheet for August 9, 1962, are shown in Figure 12. A marked increase
in dissolved oxygen can be noted at 6:15 a.m. This increase is due to
the placing of a second water turbine in service while remaining at
a low total flow and thereby bringing vacuum breaker operation into
play for aeration of the discharge. (This operation will be discussed
more fully later.) It can also be noted that as the discharge increased
to 12,000 cfs, the dissolved oxygen increased considerably. This in-
crease was caused by the drawing of upper level waters over the
weir into the water turbines.
Ragone and Peters
357
-------
nj
/
\
Dissolved Oxygen and'
Temptratur»
| Room** ftapfcfi Tailrace
August 14 ,1962
L-4
or
MSii
a"°ir*tirfTnif
to.
of
JnlU
tar
Inning
Bi»ch*rf*
Mtt
Cfl
kJUL
Tlir«*
tlO*
T«^.
F
to
i
4 1
i
fI#
4 1
i
two
i
ISM
•2*
*>
*>
i
ISM
»•
* *
i
ItM
¦1°
~A
J
itM
I?*
4
-------
The Virginia Electric and Power Company has been thoroughly
satisfied with the performance of the dissolved oxygen analyzer and
recorder. The 2%-year test period that the analyzer-recorder has
undergone has convinced us that such an instrument is sufficiently
reliable to monitor the discharge of the Roanoke Bapids project for
compliance with our Federal Power Commission license provisions.
Several times during the period that we have been continuously
monitoring the project discharge we have been able to absolve the
Project discharge from blame in downstream fish kills.
Because of the very dependable performance of the analyzer
and recorder in the Roanoke Rapids discharge, it was decided to in-
stall a similar instrument in the Gaston project discharge. This was
a more difficult task because the Gaston flows were almost double
that of Roanoke Rapids and the tailrace was not as narrow as that
at Roanoke Rapids. The raft and pump installation required large
heavy piers to position the raft so that a representative continuous
sample could be obtained. Figure 13 shows the outer pier above the
water line in the tailrace. The other pier is near the shoreline and
cannot be seen in the figure. Gaston's four units went in service in
January and February 1963, but the raft and pump installation will
not be completed until April. Figure 14 shows the raft with pump
and strainer prior to its being placed in the discharge. Figure 15
shows the Gaston analyzer and recorder installation. The analyzer
and recorder were placed in a small enclosure on the bank of the
tailrace. This was the most economical installation, since there was
no need to take the sample into the Gaston Station because it is re-
motely controlled from the Roanoke Rapids Station and the oxygen
record is transmitted by microwave to the Roanoke Rapids operators.
This analyzer and recorder was in service during the filling of the
Gaston Reservoir in November 1962; however, it has been taken out
of service until the raft and pump can be installed in the tailrace.
It is planned that this equipment will be in service by April 1963.
The cost of the Gaston installation was considerably more than
the Roanoke Rapids installation because of the necessity of trans-
mitting the signal by microwave to Roanoke Rapids and because of
the more complex pier arrangement to position the raft properly.
The analyzer, recorder, and weatherproof cabinet installed cost
$7,500; the microwave faclities and miscellaneous wiring $6,500; and
the raft, piers, pump, and cable $23,000.
There are many benefits to operation of a hydroelectric project
from such analyzer installations. Of course, the main benefit of the
installation is that it permits the maximum utilization of available
water for generation while complying with the license provisions for
the project. An additional benefit of continuous monitoring and re-
cording oxygen in the discharge is the permanent record provided in
case fish kills occur downstream.
There are other hydroelectric operations that are benefited by
the continuous analysis for dissolved oxygen. These operations in-
volve the most economical methods of increasing the oxygen content
Ragone and Peters
359
-------
of the discharge from a hydroelectric project when indications are
that the oxygen content is low. A brief description of these pro-
cedures illustrates the complexity of peaking hydroelectric operations.
Figure 13 — Heavy pier for anchoring pump raft in Gaiton tailrace.
VACUUM BREAKER OPERATION
Hydraulic turbines are equipped with spring-loaded
vacuum breakers in the head cover. These breakers draw
in air when the turbines are operating at low load. The air
thus introduced into the turbine is intimately mixed with the
water flowing through the turbine and a considerable
quantity of oxygen is absorbed.
By this procedure, the Virginia Electric and Power
Company has been able to increase the dissolved oxygen in
the project discharge during periods of minimum flow by
splitting the minimum flow between two turbines to bring this
vacuum breaker operation into play. In this manner, the
same minimum flow is made to contain appreciably more
pounds of dissolved oxygen. It must be realized, however,
that this method of dissolved oxygen improvement is not
without cost to the Virginia Electric and Power Company,
since a loss in generation occurs when waterwheels operate
near their minimum capability and cavitation can occur,
causing increased maintenance.
MONITORING FOR WATER QUALITY CONTROL
-------
P"MP electrical control a
PIPING HOUSING
PUMP HOUSING ft STRAINER •
Figure 14 — Raft with pump and (trainer for Gaston tailrace.
SPILLING OF WATER OVER PROPERLY DESIGNED
SPILLWAYS
During the winter months when minimum flows are
500 cfs, the Virginia Electric and Power Company found it
best to spill this flow through the skimmer gate rather than
put it through a turbine. This 500-cfs flow is barely sufficient
to maintain turbine speed, and no generation is realized.
The spilling of this water ensures its complete saturation
with dissolved oxygen and provides the best stream condi-
tions at this flow. Figure 16 illustrates spill from the
Roanoke Rapids project and the white water produced.
By the use of a continuous monitor in installations such as the
Roanoke Rapids and Gaston Hydroelectric projects, which have both
minimum flow and minimum dissolved oxygen rate requirements
from the lower project, it is possible to use various methods to reach
an optimum operating condition. If the quantity of available water
is low and dissolved oxygen is also low, it is possible to employ
vacuum breaker operation that splits minimum flow between two
wheels rather than to increase flow. It may be advisable to spill
minimum flows to ensure the highest degree of saturation. If water
supplies are plentiful but oxygen is low, it may be better to increase
Ragone and Peters
361
-------
Hows sufficiently to ensure meeting total oxygen requirements. It is
impractical to expect a station operator to balance such factors by
repeated Winkler oxygen analyses; however, by referring to con-
tinuous flow and oxygen recorders, he can quickly establish the
optimum operating condition that will satisfy the requirements.
Although the Virginia Electric and Power Company provides the
oxygen content prescribed in the license for Roanoke Rapids and
Gaston, cooperation by all downstream users is essential if fish kills
from excessive river contamination are to be avoided.
nitrogen supply
Oxygen a temperature I
STRIP CHART RECORDER
OXYGEN ROUND CHART
recorder primelyJB
CALIBRATION PURPOSES
figure 15 Gaston dissolved
oxygen analyzer-recorder installation,
362
Monitoring fob water quality control
-------
Figure 16— Discharge from Roanoke Rapids spillway for minimum flows.
If other users monitored the river continuously, the information
on streamflow and dissolved oxygen content would allow polluters,
by simple lagooning, to discharge their wastes in relation to the
assimilative capacity of the stream at any time and ensure against
lowering the river below predetermined limits at the downstream
sag point. The dissolved oxygen continuous analyzer-recorder in-
stallations are relatively expensive; however, the maintenance costs
are low and the dependability and availability are excellent, and it
is our opinion such analyzers could be of inestimable value for the
proper control of stream conditions,
REFERENCES
1. Report on studies below the John H. Kerr Reservoir —
August-September 1953 and April-May 1954. USPHS,
Southeast Drainage Basin Office, Atlanta, Ga. Mar. 1955.
APPENDIX I.
FEDERAL POWER COMMISSION LICENSE PROVISION
ARTICLE 25 — PROJECTS 2093 (GASTON) AND
PROJECTS 2009 (ROANOKE RAPIDS)!
Effective with commencement of operation of the Gaston Devel-
opment, the Licensee shall release sufficient water from the Roanoke
Ragone and Peters
363
-------
Rapids reservoir to maintain minimum flows and pounds of oxygen
in the Roanoke River as measured, at the point of project discharge
to meet the following requirements:
Requirement A; Minimum Instantaneous Flows
The above flo^s are subject to the following special provisions:
(1) These flows may be reduced on weekdays, Monday
through Friday, during off-peak hours by not more
-than 20 percent, bat such reduction shall not dispense
with compliance with Requirement E,
(2) The average flow for any day shall equal or exceed
the minimum instantaneous flow specified for that
month,
(3) A minimum instantaneous flow of 2,000 cubic feet per
seeotid -will be furnished by the Licensee for the period
requested by the North Carolina Wildlife Resources
Commission, to begin as early as April 1, but not later
than April 15, and to continue for at least 60 days but
not longer than 75 days in any one year, in accommo-
dation to annual variations in the time and duration
of spawning activities of the striped bass.
(4) The reduction in instantaneous flows permitted above
shall not apply when special spawning flows are being
passed for the benefit of striped bass.
Requirement B; Minimum Oxygen
Discharge from the Boanoke Rapids Development shall he main-
tained to provide dissolved oxygen at an instantaneous rate of not less
than 78,000 pounds per calendar day during the months May through
October, except as is permitted under the following" condition?:
(1) A reduction not in excess of 34 percent of the instan-
taneous rate of 78,000 pounds per day will be per-
mitted for periods not exceeding 14 consecutive hours.
(2) Any oxygen deficit so created shall be offset by greater
discharges so that a cumulative average rate of dis-
charge of 7S.0C0 pounds per day will be attained
within a period up to but not exceeding 18 hours from
the beginning of the oxygen deficient flows, and this
condition {2) shall again begin to operate as soon as
the instantaneous rate again falls below 78,000 pounds
per day.
Month
Cubic feet ser second
January, -February, March
April
May, June, July, August, September
October
November, December
i,twsa
1,500
2,000
1,500
1,000
364
MONITORING FOR WATER
-------
The Licensee shall not be required to compensate through in-
creased flows or other changes in operations for any lessening in the
quality of water entering the Gaston Development which may after
the effective date of this article be created by Acts of God, or through
acts of others than the Licensee, and beyond its reasonable control.
Requirement C:
In the event of temporary emergency conditions arising in per-
formance of Requirements A and B the Licensee will cooperate in
good faith with the North Carolina State Department of Conservation
and Development, North Carolina State Stream Sanitation Committee
and/or North Carolina Wildlife Resources Commission to take such
reasonable steps in regard to the reductions permitted by item (1)
under Requirement B as may be proper to meet the emergency condi-
tions and in the interest of maintaining during such emergency condi-
tions water quality in keeping with the standards required of class
"C" waters by order of the North Carolina State Stream Sanitation
Committee effective September 1, 1957. But nothing herein is in-
tended to compel the Licensee to correct conditions which it has not
created and no permanent change of operations shall be made with-
out the approval of the Federal Power Commission.
DISCUSSION
Robert K. Horton
Ohio River Valley Water Sanitation Commission, Cincinnati, Ohio
May I, first, compliment Mr. Ragone and Mr. Peters on their
paper. It is stimulating to read of the ingenuity used in solving a
difficult problem. And the straightforward manner in which the story
is told — with simplicity and clarity — is refreshing.
The paper illustrates several principles in the water-quality-
management field that are worthy of comment and elaboration.
First of all, the paper demonstrates that "obvious" problems must
be solved promptly and on the basis of "today's" needs.
The Virginia Electric and Power Company had a real and imme-
diate problem. It wanted to develop a natural resource in order to
expand the company's hydro-generating facilities. The problem was
that the proposed development might destroy another valuable re-
source, the breeding ground of the Albermarle-Sound species of
striped bass.
Giving immediacy to this conflict was the fact that the power
company wanted the additional generating capacity "today."
We do not know what mental reservations those concerned might
have had at the time, or still might have, about the manner in which
the problem was resolved. Certainly, the solution is acceptable to the
Horton
365
-------
power company and, obviously, requirements of State and Federal
authorities regarding the breeding grounds have been met.
But what would have happened if at the time the Federal Power
Commission was on the verge of issuing its precedent-making license,
someone had come forward and said, "This solution is fine for the
time being, but what assurance can be given that the solution as now
formulated will maintain adequate quality conditions in the Roanoke
River 50 years from now?" And the questions might have continued:
If additional reservoirs are built, will the patterns of stratification
and density currents be so changed as to render invalid the tests on
which the license for the Gaston project is based? Has enough re-
search really been done on breeding requirements of striped bass?
What changes will occur in waste loads downstream from the Roanoke
Rapids dam? How many people and new industries will move into
the area by the year 2010?
We can only guess at the amount of time the Gaston project
might have been delayed if all such questions had to be answered
with some degree of finality. The point is that so much effort may be
devoted to side issues that the immediate job becomes subordinate.
In the words of Dr. Edward J. Cleary, Executive Director of the Ohio
River Valley Water Sanitation Commission (ORSANCO), "There is
failure to do the obvious."
Nothing is implied here that in any way minimizes the importance
of planning for the future. The fact is fully appreciated that if water
resources are to be managed properly there must be adequate plan-
ning, both short-term and long-range. Planning involves forecasting,
studies, surveys, and research. All of these are essential.
The problem is to see that administrators do not become so inter-
ested in forecasting, surveys, and research that knowledge already
acquired is not applied simply because answers to all aspects of a
problem have not been developed. Delays and failures have resulted
from such misdirected interest. Let me cite an example, based on an
experience of ORSANCO, having to do with the establishment of
control measures for acid mine drainage.
For years there was an attitude of defeatism throughout the coal
industry regarding mine drainage. Operators believed that control
work could not be started until a complete understanding of the com-
plex reactions and inter-relationships involved in the formation of
mine acid had been established through research. The idea of waiting
for research to solve the whole problem had become so entrenched
that as late as 1955, when ORSANCO adopted basic requirements for
control of industrial wastes, mine drainage was specifically exempted
from control "until such time as practical means are available for
control." The irony of it is that even at that time practical methods
were available for ameliorating the effects of mine drainage. These
methods were simple in principle and had been proved to be effective,
although they were by no means a complete solution to the mine-
drainage problem. Fortunately, a change in attitude was gradually
366
DISCUSSION
-------
brought about, and in 1960, the exemption was removed and control
measures based on existing knowledge and experience were adopted.
Adoption of the control measures represented the establishment of a
new attitude regarding the mine-drainage pollution problem in the
Ohio Valley, This change did not result in a stoppage of research
activities. It merely opened the door for taking control action "now."
Another example of how delays may be created can be found
in the emphasis that is given to long-term surveys (6 to 8 years, for
example). This is not to say that surveys and comprehensive plans
are not essential; they are.
But unfortunately, someone is always ready to seize on a survey-
and-planning program as an excuse to delay action. The plea is
made: "Why should I do anything now. Let us wait until the
survey-and-plan are completed. Then we'll know what has to be
done," This is reasonable up to a point; it ignores the fact that in
many cases some of the things that will have to be done, sooner or
later, are already known.
In recognition of just such a problem, ORSANCO recently saw
fit to take a stand regarding Federal survey-and-planning activities
in the Ohio Valley. These activities are mandated by law, and ¦—• I
repeat — there is no question regarding their necessity or desirability.
The Commission deemed it prudent, however, to issue a policy state-
ment to the effect that development of a Federal comprehensive plan
will not be tolerated as an excuse for municipalities and industries
to delay construction of pollution-control works.
RIVER MONITORING AND INSTRUMENTATION
Another principle brought into focus by the paper by Mr. Ragone
and Mr. Peters is that river monitoring is an essential part of water-
quality management and that instrumentation can be of great value
in this endeavor.
In view of the great popularity of monitoring programs and data-
collection networks today, the paper serves as a reminder that the
objectives of these programs should be as specifically understood and
as clearly enunciated as those of the Roanoke Rapids project.
The paper illustrates with clarity the advantages of instrumenta-
tion. The great advantage in this instance, of course, is that "it per-
mits the maximum utilization of available water for generation." As
pointed out by the authors, various methods are available to the
power company for achieving its objectives: vacuum-breaker opera-
tion, discharge over spillway, increasing flow through the power
plant, etc. Because of the complexities involved in water-quality
management projects, the following conclusion, therefore, has par-
ticular pertinence: "It is impractical to expect a station operator to
balance such factors by repeated Winkler oxygen analyses; however,
by referring to continuous flow and oxygen recorders, he can quickly
establish the optimum operating condition that will satisfy the re-
quirements."
Horton
367
-------
The experience of the Virginia Electric and Power Company is
in harmony with that of ORSANCO regarding the use of automatic
equipment. Since 1960, ORSANCO has installed automatic analyzers
— known as robot monitors — at strategic locations in the Ohio
Valley for continuous reporting of water quality. Today the system
comprises 11 monitor stations and a central receiving and data-
logging station.
Important as the development of hardware is, however, there are
other aspects of river monitoring that are just as important to the
proper management of water quality. In the Ohio Valley, air and
boat surveillance have become vital factors in furthering compliance
with pollution-control regulations. This type of surveillance involves
the ferreting out of unsightly conditions (scum, oil, unusual color,
etc.) that are in obvious violation of clean-stream standards.
In connection with river monitoring, mention should be made of
a new tool that has recently been placed in the hands of water-
quality-control administrators. This tool is the development of tech-
niques by the U. S. Weather Bureau by which streamflow and
velocity can be forecast several days in advance.
An important application of this tool is its use in connection with
"hazard-alert" operations — those operations by which downstream
water users are notified of potential hazards caused by some unusual
happening upstream. The value of advance flow-information is illus-
trated by the following incident: A truck carrying cyanide over-
turned above 20 miles above the Louisville, Kentucky, waterworks
intake. Radio and television stations were broadcasting bulletins
saying the river was poisoned.
Thanks to advance flow-information, it was possible to advise the
city promptly that no hazard existed. About a thousand pounds of
cyanide had been spilled. Flow in the Ohio River, however, was
sufficient to accommodate 1,750,000 pounds of cyanide before creating
a hazard. Furthermore, velocity of flow was such that it would take
48 hours for water at the scene of the spill to reach the city's intake.
The fact that no hazard existed was confirmed subsequently by on-
the-spot analyses.
PROPORTIONATE DISCHARGE
Still another principle highlighted by the paper is that the con-
cept of proportionate discharge can be incorporated as an effective
part of water-quality-management operations. How clearly the
authors of the paper state the principle: "If other users monitored
the river continuously, the information on streamflow and dissolved
oxygen content would allow polluters, by simple lagooning, to dis-
charge their wastes in relation to the assimilative capacity of the
stream at any time . . . ."
An example of how the principle of proportionate discharge can
be applied is the program for the control of chloride-waste discharges
now in effect in the Ohio Valley. Study of the chloride problem indi-
388
DISCUSSION
-------
cated that the most practical way by which concentrations could be
kept within satisfactory limits was to take advantage of the diluting
effects afforded by river flow. The control regulation promulgated
calls for the construction of impoundment basins and the development
of discharge schedules for producers of chloride waste. The discharge
schedules show the amount of waste that can be discharged directly
to the river for any river flow. Additional chloride must be im-
pounded for later release at higher flows. As the authors indicate, it
seems reasonable to anticipate that the principle of proportionate
discharge will be applied with increasing frequency.
CONCLUSIONS
In conclusion, then, the paper by Mr. Eagone and Mr. Peters
illuminates several important principles for better administration and
operation of water-quality-management projects.
Among these principles are the following:
1. Research, surveys, and long-range planning cannot be tol-
erated as an excuse for delaying obvious jobs and for failing to apply
knowledge already at hand.
2. Surveillance programs must include on-the-site observations
as well as collection of laboratory data.
3. Instrumentation offers greatly expanded opportunities for
understanding river characteristics and for improving water-quality
control.
4. Streamflow forecasting is a new tool that has application in
hazard-alert operations and in the management of waste-control
facilities.
5. The principle of proportionate discharge can be an effective
part of water-quality-management projects.
REFERENCES
1. Cleary, E. J. Safeguarding water resources through pollution
control. Presented at ASCE Water Resources Eng. Conf., Omaha,
Nebr., May 14, 1982.
DISCUSSION FROM THE FLOOR
Mr. Bell, U. S. Public Health Service: I think it is quite signifi-
cant that the Virginia Electric and Power Company, a private power
company, has gone to such lengths and expense to install dissolved
oxygen and temperature recording equipment and to maintain it in
this fashion.
Mr. Ragone, Virginia Electric and Power Company: Thank you,
Mr. Bell. When Mr. Peters and I were first invited to present a paper
at this symposium, we felt a little like Daniel must have felt when
From the Floor
369
-------
he was informed he was to be placed in the lions' den. We thought
we might be the only representatives of a private power project on
the program. Mr. Lee, of course, strengthened our ranks. After
reviewing the objectives of the symposium, we welcomed the oppor-
tunity to tell what can be and has been done on water quality control
and monitoring of the Roanoke River through the cooperative efforts
of Federal, State, and local agencies, and private enterprise.
Mr. McDermott, U. S. Public Health Service: In October 1953,
prior to construction of Roanoke Rapids Dam, the Public Health Serv-
ice surveyed the Roanoke River in cooperation with Mr. Hubbard's
office. At that time we found a diurnal fluctuation in the dissolved
oxygen downstream from Roanoke Rapids and the Halifax Paper
Company that corresponded to the diurnal change in discharge from
Kerr Dam. This diurnal fluctuation in flow, together with wastes
discharged in the area, caused dissolved oxygen concentrations as low
as 1.9 milligrams per liter at the critical point of the dissolved oxygen
sag about 17 miles below Roanoke Rapids.
I note that the control for the regulation of dissolved oxygen has
been placed at Roanoke Rapids Dam. It seems more logical that the
control should be at the point of minimum dissolved oxygen, and that
a considerable reduction in the loss of power could be attained by-
making use of the aeration in the rapids between Roanoke Rapids
and Weldon and oxygenating only the water corresponding to low-
flow discharges so as to increase the dissolved oxygen downstream
at the minimum point rather than continuously meeting some dis-
solved oxygen requirement at the dam. Mr. Ragone, could you com-
ment on whether consideration was given by VEPCO to placing the
control at the oxygen sag critical point rather than at the dam?
Mr. Peters, Virginia Electric and Power Company: That is a very
difficult problem, The possibility of establishing some criteria down-
stream was considered during the studies. From our own standpoint,
however, we would rather control what we have on our own property
and let the downstream users do the same.
Mr. Ragone: You know that a large waste load is discharged to
the river about 8 miles downstream from the dam. Dissolved oxygen
observations at or below that point would depend on the operation
of that plant as well as ours. We surely would not want to be in-
volved with another company's operation. We are in the power
business, hope to stay in the power business, and do not want to get
involved in any other business.
Dr. Ingols, Georgia Institute of Technology: Wouldn't the time
lag between the dam and the sag point be significant in trying to
control the dissolved oxygen concentration at a point downstream
rather than at your tail race.
Mr. Peters: Yes, you might have a boat down there, moving all
the time, because the location of the sag point varies up and down
the stream depending upon the flow and, of course, upon the pollu-
tional load discharged to the river.
370
DISCUSSION
-------
Dr. Yevdjevich, Colorado State University: I have a question on
the engineering features of those submerged weirs. "When we have a
submerged structure the difference in pressure on two sides should be
very small. Why does this require such a tremendous investment?
I question the necessity of building such an expensive dam just to
separate two different types of water.
Mr. Ragone: The cost of the first submerged weir in Roanoke
Rapids Reservoir was high because the reservoir was filled, and the
only way we could build the project was to dump rock from a barge.
The cost of caissons was prohibitive. There is a tremendous velocity
effect, especially as you get into high flows, so the strength has to
be there.
On the Gaston Project, we considered a thin concrete wall, but
it would not work. It was more expensive to reinforce a thin con-
crete wall than to use the rock fill for the major portion of the
structure and put the cribbing on top. I was not personally involved
with the engineering studies on the different dams, but our consult-
ants assured us that these were the most economical structures that
would provide the strength needed for the velocity over the top. Since
the Gaston weir was within 15 feet of the surface the length had to be
1,000 feet. We had originally started out to make it 600 feet long,
but had to increase it just because of velocity.
Dr. Yevdjevich: Did you conduct hydraulic model studies on the
impact of the crest velocities?
Mr. Ragone: No, but very elaborate studies were made on the
Roanoke Rapids weir by our consultants. We also consulted people
that were well versed in the problems of flow over this type of
configuration, and their data were used in the design of the Gaston
project to determine the necessary strength and length.
Mr. Lee, Duke Power Company: Duke Power Company has just
finished a submerged weir that is a little higher than the VEPCO
weir (80 feet) and almost the same length (about 900 feet). We had
the opportunity to build it in the dry before the reservoir was filled.
It is made of low-grade earth, secondary material that is not highly
compactible and could not be used in the main earth dam. Our weir
was very much less expensive than the Gaston weir.
One section of the weir has precast concrete stop logs running
the full height of the weir to equalize the pressure on the inside of
the weir so that there would not be a full head against the outside
of the weir as the reservoir was filled. The weir was designed to
resist the velocity head with 10 feet of water flowing across the crest
of the weir under our maximum discharge conditions.
I did not notice in your slides, Mr. Ragone, whether you provided
a way to fill the pool inside of your weir?
Mr. Ragone: They do not show, but two small doors, about 8
feet tall by 6 feet wide, were used in the filling so that the pressures
From the Floor
371
-------
were balanced on both sides. We used precast concrete stop logs on
top, as you can see.
Concerning the use of rock fill for the Roanoke Rapids weir, we
had to blast an 8,000 foot tailrace through the rock of the river bed
so there was rock available; in fact, we stored it on the side for
a while.
Mr. McLean, U. S. Public Health Service: The establishment of
the minimum flows for Roanoke Rapids is very interesting. I am
under the impression they were established in about 1946 or 1948,
before the expansion of the paper company and its waste loads. The
importance of considering the future growth of industry in an area
and the changes that may come about is very apparent in this situa-
tion.
372
DISCUSSION
-------
COOPERATION IN THE SOLUTION OF WATER
QUALITY PROBLEMS ASSOCIATED WITH
FLOW REGULATION
Willis King and Edward C. Kinney
Bureau of Sport Fisheries and Wildlife
V. S. Fish and Wildlife Service, Washington^ D. C.
WHY COOPERATION IS NEEDED
The years ahead will be characterized by a sharp rise in human
Population, a concentration of people in metropolitan areas, and in-
creased demands for water to supply man's daily requirements, to
toeet industrial needs, to remove waste products, and to provide recre-
ational opportunities (Tables 1 and 2). At present, supply and de-
mand for water of good quality are about equal in the United States,
The demand is increasing faster than the supply, and by 1980 will
probably be double the I960 standards. Statistics are available show-
ing that more water is being used every year in the home, in industry,
in agriculture, in waste removal, and for recreation. By the year
2000, which is within the lifetime of many of us, our population will
have doubled and the demand for water will be several times that
of 1960. Man's ingenuity, as well as nature's supply, will be taxed
to adequately meet the Nation's requirements. These needs can be
met only if all agencies concerned with the development and manage-
ment of water resources are willing to cooperate in seeking equitable
solutions to the many problems that must be solved, Cooperation in
this undertaking will be necessary to maintain our American standard
of living and our prominent world position.
The job ahead requires that administrators, planners, engineers,
and biologists work together more closely than in the past. The skills
of those in research, as well as those in management, must be prop-
erly directed and applied to the problems that are even now with us.
Competition for water and conflicts in its use must be resolved on a
basis of public need and willingness to accept "the principle that com-
patible and essential uses of water must be allowed, even encouraged.
Water quality must be maintained at acceptable levels, even though
this means less financial income to some.
COOPERATION IN PLANNING FOB FLOW REGULATION
All water development projects start with an existing need, an
anticipated requirement, or an idea. As soon as this stage arises, it is
time to determine what the effects of such a project would be on the
volume of stream flows, on the quality of water that results from the
anticipated use, and the steps that may be necessary to restore water
quality should it be damaged. The effect on uses already existing
cannot be overlooked.
The legal support for cooperation in water resource planning is
reasonably well established. Administrative machinery in the form
King and Kinney
373
-------
of Federal departments and the various State water boards and
agencies is already in existence. Federal and State laws usually say
what must be done before water can be impounded, diverted, pol-
luted, consumed, or treated with chemicals. Sometimes the laws are
inadequate or poorly enforced, but we will not attempt to analyze
their weaknesses in this paper.
TABLE 1. PROJECTED WATER REQUIREMENTS IN THE
UNITED STATES" — BILLIONS OF GALLONS PER DAY
Type of use
1954
1980
2000 Change, %
Waste dilution
512
332
447
—-13
Waterpower
374
616
637
70
Steam generation
74
259
429
478
Navigation
281
238
221
—15
Irrigation
176
167
184
4
Sport fish habitat
78
171
241
210
Manufacturing
32
102
229
613
Municipal
17
29
42
147
Mining
2
3
4
100
"From Senate Report No. 29, 87th Congress, Report of the Select
Committee on National Water Resources. 1961.
TABLE 2. CHANGE IN WATER WITHDRAWALS IN THE
UNITED STATES® — BILLIONS OF GALLONS PER DAY
Type of use
1950
1955
1960
Increase, %
Waterpower
1,100
1,500
2,000
fi2
Industrial1'
77
110
140
82
Irrigation
103
110
110
2
Public supplies
14
17
21
50
Rural
3.6
3.6
3.6
0
¦From Geological Survey Circular 456, "Estimated Use of Water
in the United States," 1960. Washington,. D. C. 1961.
•¦Includes steam generation.
NOTE: Table 1 figures were projected from the 1954 base. At the
time the study was undertaken, the 1954 use figures were the most
recently available data common to all of the uses listed. The require-
ments for waste dilution are based on anticipated 70 percent treat-
ment by 1980 and 80 percent treatment by the year 2000,
The data in the two tables differ considerably in the withdrawals
for water power. The figures in Table 2 are higher as they include
the amounts used at each hydroelectric installation. In Table 1, only
the amount used once at the largest plant on any one stream was
counted. That is, if there were several plants on one stream, only the
amount used by one plant was included.
374
COOPERATION IN STREAMFLOW REGULATION
-------
LEGAL BASIS TOR COOPERATION
Fish and Wildlife Coordination Act
One of the major purposes of Public Law 85-624, the Fish and
Wildlife Coordination Act, is to assure that wildlife conservation
receives equal consideration and is coordinated with other features
of water-resource development programs (Appendix I contains ex-
cerpts from the Act). The term "wildlife," as used in the Act, includes
fishes and aquatic life. Coordination is accomplished through ef-
fectual and harmonious planning.
The Act requires that the Fish and Wildlife Service and the head
of the State wildlife agency be notified by any Federal agency or any
public or private agency under Federal permit or license whenever
a water development project is proposed or authorized. This specifi-
cally applies to impoundments, diversions, channel improvements,
and other modifications including drainage and navigation improve-
ments.
The Secretary of the Interior is authorized to provide assistance
and to cooperate with Federal, State, public, and private agencies in
the development of all species of wildlife and their habitat. This in-
cludes quality of water as well as flow regulation.
After project notification has been received, the Fish and Wildlife
Service arranges for a joint meeting of Federal and State wildlife
biologists and personnel of the agency in charge of the development.
Following this briefing session, Federal and State biologists arrange
for whatever joint surveys and investigations are required to deter-
mine:
1. Possible damage to wildlife resources.
2. Proposed means and methods of preventing losses or mit-
igating damages.
3. Possible developments for the improvement of wildlife
resources (enhancement).
After the investigations, a report is prepared describing any
damages to wildlife attributable to the project and measures propose''
for mitigation. Specific recommendations are made for wildlife con-
servation and development, lands to be utilized or acquired, and the
type of management required to achieve anticipated results. Other
interested agencies and groups are consulted during the preparation
of the fish and wildlife report.
For Federal projects, the Secretary of the Interior submits a copy
of the report to the agency responsible for construction. The con-
structing agency includes the wildlife report in any project report
submitted to Congress or authorizing agency. The State fish and game
department may also submit a report to the constructing agency.
Usually, however, the State report is included in the coordinated
report of the Secretary of the Interior.
King and Kinney
375
-------
The Federal agency responsible for the construction is required
to give full consideration to the report and recommendations of the
Secretary of the Interior and to any report of the States. The con-
structing agency may modify or add to the structures and may acquire
land for the conservation, maintenance, and management of wildlife
resources, as directed by Congress in appropriating funds for the
project.
The use of such waters, land, or interests for wildlife conserva-
tion purposes shall be in accordance with general plans approved
jointly by the head of the department or government agency exercis-
ing primary administration, by the Secretary of the Interior, and by
the head of the State wildlife agency.
Impoundments of less than 10 surface acres are exempted from
provisions of the Act, as are developments carried out by Federal
agencies on lands under their jurisdiction. This Act forms the back-
bone of cooperative effort in planning, developing, and operating
water storage projects, where fish and wildlife resources are involved,
throughout the United States.
Projects for hydroelectric developments, proposed by private
interests, must be submitted to the Federal Power Commission, where
all applicable information is reviewed, weighed, and considered
before a license is granted for the construction and operation of the
desired facility. These reviews are especially thorough when im-
portant public interests are involved. The Commission has, in recent
years, shown itself to be receptive to recommendations from other
private and public interests having legitimate concern over what the
proposed development may do to an existing water supply. Needless
to say, it is better to have taken these related effects into account
before filing the license application. Likewise, objectors or repre-
sentatives of conflicting interests should also have their case well
justified and documented if they are to influence the terms of the
license granted by the Federal Power Commission.
The Watershed Protection and Flood Prevention Act
The Watershed Protection and Flood Prevention Act (Public Law
83-566 as amended by Public Law 87-703) authorizes the Secretary of
Agriculture to cooperate with other Federal agencies, with the States,
and with local units of government in planning and carrying out
works for soil conservation and other purposes. The Secretary is
authorized to cooperate in making investigations and surveys of
watersheds and proposed waterways as a basis for the development
of coordinated programs.
After the Secretary of Agriculture approves the request from a
local agency for planning assistance, the Secretary of the Interior
is notified. The latter may then make surveys and investigations and
prepare a report with recommendations concerning the development
of wildlife resources. Full consideration is given to the wildlife rec-
ommendations during the final drafting of the local plan. The wild-
life section of the plan is a cooperative effort developed by the Bureau
376
COOPERATION IN STREAMFLOW REGULATION
GPO 821-740-13
-------
of Sport Fisheries and Wildlife, the State fish and game agency, the
Soil Conservation Service, and local fish and game interests.
Many of the potential fish and wildlife developments are de-
scribed in Bureau of Sport Fisheries and Wildlife Circular 100 titled,
"Better Hunting and Fishing on Small Watershed Projects."
Federal Water Pollution Control Act
The Federal Water Pollution Control Act (Public Law 84-660,
as amended by Public Law 87-88) provides for water pollution con-
trol activities in the Public Health Service of the U. S. Department of
Health, Education, and Welfare and for other purposes.
The Secretary of Health, Education, and Welfare, after investiga-
tion and in cooperation with other Federal agencies, State and inter-
state agencies, and with municipalities and industries involved, pre-
pares and develops comprehensive programs for eliminating or re-
ducing the pollution of interstate waters and their tributaries, or
navigable waters, including coastal waters. In this development, due
regard is given to the improvements that are necessary to conserve
such waters for public water supplies, propagation of fish and aquatic
life and wildlife, recreational purposes, agricultural, industrial, and
other legitimate uses.
In recognition of the common interests in the field of water pol-
lution and to provide an effective program of inter-agency coopera-
tion, a Memorandum of Understanding, July 1958, was developed
between the Departments of Interior and Health, Education, and Wel-
fare. The Memorandum resulted in the establishment of close liaison
at various levels between the two Departments. It also provided for
cooperative water quality studies involving the well-being of fish and
fisheries and joint investigations of major fish kills.
The Act provides that in the surveys or planning of any reservoir
by any Federal agency consideration be given to provision for storage
of water so that streamflow may be regulated for the purpose of
water quality control.
State Laws
Most States have laws designed to maintain the quality of their
water resources in a condition so as to protect public health and
permit industrial growth. Sometimes recreational uses are neglected
or ignored, but public sentiment is growing rapidly, calling for full
understanding and cooperation in the development and use of water.
More will be said in this connection in the discussion of water quality
standards.
BASIC CONCEPTS IN COOPERATION
Cooperation in water quality control is largely dependent on the
free exchange of information. Recognition of the primary objectives
of a project must come early in its history. Likewise, secondary pur-
poses or what is generally considered a better concept, "compatible
King and Kinney
377
-------
uses," should also be defined. Rarely does any project confine itself
to a single purpose. The point here is that all necessary and potential
uses must be recognized and considered when planning is started.
Field surveys must be made to determine the level of use that
can be provided, the capacity for future growth, and the effects one
type of use may have on the quality and quantity of water passed
downstream or made available for other uses. This information and
often more must be in the hands of the designers of the project before
their work has gone very far.
Frequently, operating plans for dams or other water control
structures control the type of uses that a given body of water may
serve. The extent of drawdown, seasonal and daily variations in flow
pattern, and the effects of these on water quality, as well as quantity,
must be understood in the light of available knowledge. The biologist
must be able to think in terms of generating capacity, kilowatt hours,
density currents, operation of the rule curve, and waste assimilative
capacity. Engineers must not be content with understanding their
own jargon, but must know what is meant by photosynthetic produc-
tion, such terms as ecosystem, and the basic physiological require-
ments of fishes and other aquatic organisms. For the sake of brevity,
I will not even attempt to mention the terminologies of the economist
and the biometrician, which nowadays are frequently encountered.
For those engaged in the development and management of the Nation's
water resources, it is no longer enough to understand only the ter-
minology and requirements of a single field of interest.
WATER QUALITY STANDARDS
Water quality standards have been established by many of the
States and interest compacts formed for the various uses of water.
California adopted water quality criteria instead of standards. It was
believed that standards were too specific to apply to natural waters.
It is often difficult to set minima to many pollutants because other
chemicals in the water may nullify or intensify their effects. Most of
these standards are established by State pollution control agencies.
Some are established by public health agencies. Most agencies have
consulted with their fish and game organizations in setting the
standards.
Our purpose in reviewing the generally understood water qual-
ity levels described below is to emphasize that different uses require
different standards and any use that would tend to downgrade the
quality of water so that other uses cannot exist must be closely
scrutinized and evaluated.
REQUIREMENTS FOR VARIOUS WATER USES
Drinking Water
The standards for drinking water are usually taken from the
current Public Health Service Drinking Water Standards.* Most
*Latest revision — 1962.
378
COOPERATION IN STREAMFLOW REGULATION
-------
natural waters that meet these standards will support aquatic life.
Although sufficient dissolved oxygen to support aquatic life is not
required for drinking purposes, the established minima are usually
above 5.0 milligrams per liter. Most specifications require that no
materials toxic to fish life be present.
Industrial Water Supply
There is considerable variation in the water qualities required
by various industries. Water used in the food and beverage indus-
tries are in the higher drinking water class. Water used for cooling
purposes should be cool, soft, but not too acid. Water used in the
manufacture of steel should be soft, almost neutral, low in organic
matter, and cool. Pulp and paper mills require soft, almost neutral
water. Soft water with a higher pH can be used in rayon manufacture.
Irrigation
Because of the differences in soil types, precipitation pattern, and
vegetation tolerances, it is difficult to establish critical limits for irri-
gation water quality. Highly acid waters (below pH 3.0 to 4.0) and
highly mineralized waters are not suitable for irrigation because of
their harm to the soil and to agricultural crops.
Fish and Other Aquatic Life
A tremendous amount of research has been done on the effects
of pollutants on aquatic organisms. Much of this work has been
summarized by Ellis, Westfall, and Ellis(l), by the California Water
Pollution Control Board, and by various reports from the Department
of Health, Education, and Welfare's Robert A. Taft Sanitary Engineer-
ing Center. Fish production is favored in near-neutral to slightly
alkaline, moderately hard waters that are nearly saturated with oxy-
gen; contain sufficient nutrients for good phy to plankton production;
are free of injurious amounts of chemicals; have low turbidities; and
are subject to not more than moderate fluctuations in water levels
and flow velocities.
Navigation and Water Power
It is desirable that water used for these purposes be low in
corrosive substances; debris and silt; objectional odors; and algae,
fungi, and other substances that clog passageways or cling to vessels.
Swimming, Water Skiing, Boating, and Esthetic Enjoyment
Waters used for these activities should have a low coliform count,
be clear, and be free from objectionable odors, pathogenic organisms,
and irritants.
Water Dilution
Even after waste products are treated, some water is usually
required for dilution. For organic wastes, the water should have a
King and Kinney
379
-------
high oxygen content and have temperature and pH characteristics
that favor decomposition bactena. It should be free of bactericides
Sufficient volume is required so that bacterial action will not result
in low oxygen conditions. Inorganic wastes should be sufficiently
neutralized or diluted so as not to be harmful to aquatic life.
EXAMPLES OF THE COOPERATIVE APPROACH
TO SOLVING PROBLEMS OF WATER QUALITY CONTROL
Discussions of this broad subject of water quality control throueh
regulation can easily be lost in generalities and platitudes unless
we get down to actual cases. We will, therefore, cite situations where
water quality control has been sought through flow regulation and
where both success and failure have resulted.
The Roanoke River, North Carolina
It is worthwhile to describe the approach taken in dealing with
problems on the Roanoke River in North Carolina and the solutions
that were reached Many attending this symposium are familiar with
this story and participated in its development. For those who are not
acquainted with it, I hope the resume given here will be of interest
Like many similar situations, the problems were not all of sudden
origin, nor were the answers quick to come, immediate in their appll"
cation, or favorable to all interests.
The headwaters of the Roanoke River arise in the Blue Ridee
Mountains of Virginia. The mainstem of the river is formed by the
confluence of the Dan and Staunton Rivers, and some 200 miles down
stream, * empties into the western end of Albemarle Sound (Fieurl
1). Two-thirds of the watershed is in Virginia, and the river is very
important to the health, industry, and recreation of residents of the
s^LminNoIth cZlina3"16 thm°re W&tG« by far than any other
about 8500 cfs ~~ average flow through the State is
The river is historic as an avenue of transportation and com-
merce, since the days of the early settlers of the region. Its watSs
are«S the largest spawning population of striped bass
south of Chesapeake Bay. It offers many opportunities for recrea
tional developments that are largely unrealized at this time The
first two major developments that affected the life of the river were
the establishment of kraft pulp and paper mills at Roanoke Rapids
ami at Plymouth in the 1930's and the construction of the BugS
Island Dam (now called the John H. Kerr Dam) by the U. S Army
Corps of Engineers, starting in 1946. y
men^becfme^w^that. Sfpartment of Conservation and Develop.
lraSt-ii aware that the river was changing in the early 1940's
Fishkillswere reported; sportsmen and commercial fishermen cQm-
? "? f J stnPed bass and other fisheries were being destroyed
inV^PvX°and00?dthe PUlp ^ NeW indUStries wlreco^
y and old ones were expanding. Several cities became
380
COOPERATION IN STREAMFLOW REGULATION
-------
concerned about the quality of their water supplies. At that time,
the State did not have the funds or personnel to study the situation
intensively.
Figure 1 — The Roanoke River jyjtem, Virginia and North Carolina (Ref. 2).
The Federal government was brought into the picture when
hydroelectric developments on the river were under consideration
by the Corps of Engineers. The U. S. Public Health Service and the
Fish and Wildlife Service supplied recommendations in 1942 and
1943 in which minimum flows, thought to be needed to make the
river habitable for fish and to dilute the already evident pollution
load, were proposed. When the Buggs Island Dam at river mile 179
was authorized by Congress, the Fish and Wildlife Service was asked
to recommend minimum flows for the protection of the striped bass
fishery. The study and report that followed were the first to point
out that the striped bass spawning populations might be adversely
affected by proposed hydroelectric developments on the river. A
minimum of 5 milligrams of oxygen per liter was considered neces-
sary to protect the striped bass, and a minimum average daily flow of
2000 cfs was proposed for the fish spawning season. There were other
recommendations incorporated into the operating procedure for the
Kerr Dam (then called the Buggs Island Dam). The first electrical
power was generated in 1952, and flood control provisions of the
project were also effected that year. It soon became apparent that the
water discharge pattern greatly affected the migrations and spawning
of the striped bass.
Private power entered the picture when the Federal Power Com-
mission granted a license to the Virginia Electric and Power Company
in February 1951 to construct and operate a hydroelectric plant at
Roanoke Rapids (river mile 137). The Company was ordered to
maintain the same minimum flows specified for the Federal project,
King and Kinney
381
-------
with certain additional features and with the stipulation that its
Roanoke Rapids Dam should serve to reregulate water discharged
so that the water requirements designated for pollution abatement
and fish conservation could be met. The Commission left the way
open for further protective steps by inserting the following clause in
Opinon and Order No. 204, Project 2009:
"The Licensee shall construct, maintain, and operate
such fish protective devices and shall comply with such
special conditions in the interests of fish life as may be
prescribed hereafter by the Commission upon recommenda-
tion of the Secretary of the Interior."
Observations made by State fishery biologists and by State
Stream Sanitation Committee workers in 1952 and in 1953 indicated
that the minimum flows designated in the Roanoke Rapids license
and in the operating plans for the Kerr Dam were not enough to
protect the striped bass and to reduce the pollution entering the river
to safe levels. Some field studies were conducted by the State in the
spring of 1953 and continued in 1954, but just what flows were re-
quired to meet the conditions could not be determined. The U. S.
Public Health Service participated in these studies and expanded its
efforts in the spring of 1955, the first year that the Roanoke Rapids
project was in operation.
It was apparent to all agencies concerned that an expanded and
coordinated effort was urgently needed if some of the basic problems
of flows and water quality were to be solved and if the very life of
the river were to be saved. An appeal was made to Congressman
Herbert C. Bonner for assistance. Congressman Bonner arranged a
meeting of the representatives of all interested Federal and State agen-
cies, industries, municipalities, and private citizens. The group met
at Weldon, North Carolina, on May 2, 1955. The discussions there led
to a recommendation for the establishment of a "Steering Committee
for Roanoke River Studies." The senior author was chosen as chair-
man of the Committee and held that assignment throughout the study
until the Committee terminated its activities late in 1959.
The Steering Committee was organized with one member from
each of the following agencies or interests: State agencies — State
Stream Sanitation Committee, Department of Conservation and De-
velopment, Wildlife Resources Commission; Federal agencies — Fish
and Wildlife Service (U. S. Fishery Laboratory, Beaufort, North
Carolina), Corps of Engineers (Office of the District Engineer, Nor-
folk), Public Health Service (Atlanta Regional Office), Southeastern
Power Administration (Regional Office); Private industries — Halifax
Paper Company, Roanoke Rapids, and North Carolina Pulp Company,
Plymouth; General public — the State senator from the area, sport
fishermen of the area, and commercial fishermen of the area.
The representative for the commercial fishermen failed to attend
any meetings and their interests were handled by the Department
of Conservation and Development. The Virginia Electric and Power
382
COOPERATION IN STREAMFLOW REGULATION
-------
Company was invited to have a representative in the industrial group
but declined. The Company did, however, attend all meetings of the
Steering Committee and participated in and supported the study
throughout. In addition to the efforts of the official members of the
Committee and their staffs, active support was given to the program
by the National Council for Stream Improvement (a research affiliate
of the pulp and paper industries), by the Department of Zoology
and Wildlife of North Carolina State College, by the Institute for
Fisheries Research (University of North Carolina), by the Branch
of River Basin Studies of the Bureau of Sport Fisheries and Wildlife,
and by the Robert A. Taft Sanitary Engineering Center.
Members of the Steering Committee soon found that it would
not be possible for everyone to become fully informed on all aspects
of the problem or to contribute toward the solutions. Three task
forces were formed; their assignments were as follows:
Task Force 1: Water quality and engineering studies.
These included determination of the quantity, nature, and
pattern of discharges of pulp mill wastes and municipal sew-
age, together with evaluation of the resultant effects upon
water quality in the river. The influences of impoundment
and flow releases upon the physical and chemical qualities
of the water, and its effectiveness in neutralizing pollutants
were studied as a basis for recommending year-around flows
to prevent deterioration of water quality.
Task Force 2: Studies related to the spawning of the
striped bass, the survival of eggs and fry under existing river
conditions. These studies included the determination of the
quality and quantity of water necessary to protect the
striped bass fishery. This involved both field observations
and laboratory studies.
Task Force 3: Studies on the status of the striped bass
fishery in the Roanoke River, Albemarle Sound, and tribu-
tary streams. Included were the location of spawning areas;
migration pattern; survival of young and adult striped bass;
collection of data on the catch by sport fishermen and by
commercial fishermen; determination of the regulations nec-
essary to conserve the fishery; and habitat improvements
needed to assure continuance of acceptable conditions.
Each of the Task Forces promptly held initial meetings to draw
up programs of investigation and field study. The Task Forces re-
mained active throughout the life of the Steering Committee or until
their final reports were submitted. Each was chaired by the repre-
sentative of the appropriate State agency on the Steering Committee.
Within each Task Force, efforts were pooled whenever possible, and
the participating agencies called on for manpower, equipment, and
funds. Only one man was employed full-time. The U. S. Public
Health Service provided the services of Dr. Frederic F. Fish and the
Fish and Wildlife Service provided funds for his expenses. Dr. Fish
King and Kinney
GPO 821-740-14
383
-------
provided field coordination on certain phases of the studies and de-
voted his personal efforts to Task Force 2. He assisted the Committee
by preparing special reports when requested and prepared the final
report of the Steering Committee.
One further level of organization was found necessary — a Sub-
committee for Operations. This was an executive group composed of
the three Task Force Chairmen, the Chairman of the Steering Com-
mittee, and Dr. A. F. Bartsch, who represented the Robert A. Taft
Sanitary Engineering Center. Dr. Bartsch chaired the Subcommittee.
The Subcommittee met four to six times each year to review progress,
plan technical phases of the work, process reports, and prepare rec-
ommendations for presentation before the Steering Committee. All
members of the Subcommittee gave far more of their time and effort
than could be readily spared from their other official duties. Dr. Fish
also looked to the Subcommittee for general direction of his activities.
The Steering Committee held 11 meetings over a 5-year period
and a full record of its transactions is on file with each of the par-
ticipating agencies. From 25 to 50 people attended each meeting.
Many more were present at the public meetings held at Weldon,
North Carolina.
In this report, the organization, objectives, and program of the
Steering Committee for Roanoke River Studies have been empha-
sized because little of this information has been generally available.
The basic problem faced by the Committee was "How can the water
which nature has provided in the Roanoke River be used to meet
the varied and increasing demands placed upon it, so as to assure
the rights of all legitimate interests and still protect the fishery re-
sources with which the River was so specially endowed?" This was
a large order.
In working out solutions to the many problems, the Committee
recognized that no one philosophy or special interest could control.
There would have to be give and take and above all a willingness
to talk and negotiate differences even when at first sight it appeared
that there was no common ground for reaching solutions. The Steer-
ing Committee found itself in difficult situations many times, and
some of the debates held before it make highly interesting and
informative reading.
Once the studies were underway, it became apparent that what
was done at the Gaston Project, proposed by the Virginia Electric
and Power Company as a new hydro-development at river mile 146,
could well hold the key to the future of the Roanoke River. This new
dam and reservoir occupied all space available for development be-
tween the Roanoke Rapids project and the Kerr Dam above. Both
Roanoke Rapids and Kerr Dams have bottom-level discharges and are
operated as peaking-power plants for the generation of electrical
energy. The water coming out of the Kerr Dam was sometimes
devoid of oxygen, and often held less than 2 milligrams of oxygen
per liter. If the intervening stretch of the river were to be im-
pounded, would it then be possible to provide water of acceptable
384
COOPERATION IN STREAMFLOW REGULATION
-------
quality below Roanoke Rapids, as well as the flows needed for a
variety of uses? Most of the studies of Task Force 1 were applied
in seeking the answers to this problem. It was not enough to observe
and understand existing conditions. The Committee had to know
what would happen when these conditions were altered, and recom-
mend measures that could be expected to maintain or even improve
water quality for downstream uses. Unless this were done, there
would be no opportunity for future development of the river, exist-
ing industries would suffer, and the striped bass fishery would be
seriously threatened.
One of the study projects proposed and carried out by Task Force
1 was to learn what would occur if a curtain wall or submerged weir
were built around the turbine intake at Roanoke Rapids so that water
could be taken from the top 25 feet of the reservoir. A comprehensive
study plan was developed and carried out during the summer of 1957.
Utilizing the old coffer dam as a base, the power company constructed
a curtain wall of crushed stone. A full-scale monitoring project was
organized to record dissolved oxygen, temperature, and turbidity
over a 3-month period. A mass of data was gathered and analyzed
by a special committee of experts. The curtain wall did improve the
quality of water passed downstream, but certain structural modifica-
tions were indicated as to its height, stability of materials, and rela-
tion to the flow pattern.
Some of the best brains in the business reviewed the data and
gave their recommendation for solving the major problem. The group
included Dr. D. W. Pritchard and associates of the Chesapeake Ocea-
nographic Institute as consultants for the Virginia Electric and Power
Company; Mr. Milo Churchill, Chief, Stream Pollution Control Sec-
tion, TVA; Mr. Francis W. Kittrell, In Charge of Stream Sanitation
Studies, U. S. Public Health Service; and Mr. Thomas M. Riddick,
Consulting Engineer and Chemist for the Halifax Paper Company.
The findings guided Virginia Electric and Power Company in
revising and redesigning their proposal for the Gaston project and
provided a basis for providing the quality of water Task Force 1
found to be essential. Without this intensive study in which all agen-
cies cooperated it is doubtful whether a solution satisfactory to the
principal interests could have been found for the problems of the
Roanoke River. The findings led to the construction of a permanent
curtain wall or weir in the new Gaston project to assure that water
of acceptable quality and temperature would be passed downstream
to meet the necessary requirements.
Task Force 1 also completed a comprehensive pollution abate-
ment plan for the Roanoke River and applied its water quality
standards in a State classification system to appropriate sections of
the stream.
Task Force 2 studied the hatch and survival of striped bass eggs
and fry in varying concentrations of waste, drawing samples from the
individual mills and from the river where all known pollutants were
King and Kinney
385
-------
mixed. The passage of striped bass eggs at Weldon was recorded for
he peak of the 1958 and 1959 spawning season. These jobs required
that new equipment be designed and tested and took at least three
seasons of work before reliable quantitative data could be gathered
and analyzed The relation of stream flows to fish spawning was also
observed. Task Force 2 was not able to complete its studies during
the field program of the Steering Committee but continued them
under State guidance and issued its final recommendations in 1960(3).
Task Force 3 tackled a variety of problems associated with sport
and commercial fishing for the striped bass, as well as with its life
istory. The staff and facilities of the Fishery Laboratory, Bureau of
Commercial Fisheries Beaufort, North Carolina; the Institute for
Fisheries Research at Mcrehead City; and biologists from North Caro-
lina State College supported by grants from the pulp and paper com-
panies, carried on field work for three full seasons. Their findings
were summarized for the Steering Committee, and separate reports
were prepare or publication by the participating agencies.
*"ecornmendations of the Steering Committee
with the history of the development of the Roanoke River, were com_'
bined in a rep° ° the Steering Committee for Roanoke River studies
and issued under date: of June 1959(4). Unfortunately, the report has
not been Pu ¦ *nal recommendations for stream flows related
to the bM8 flsh.ery were not completed at the time that the
Steering Committee found it necessary to disband; however, year!
around flo tream sanitation purposes and the provisions be-
lieved necessary for melUsion in the license for the Gaston project
were completed and accepted.
, f°r 'strearn flows and water quality in
relation to the Gaston project were presented to the Federal PowS
Commission m November i960, in connection with its review of the
irginia Afw0 + °^e j ^omPany's application for a license for
meet^operating^requt^ni^r^s,
Commission11 A^opy^S
be pointed ^co^m^dation^Tnc^uded increased
flows over previous license provisions fPmiont 9nn J I' rec0§mzed the
license. The license1forth? , °WCT Commission
vision that a perm^ P*^ also included the pro-
of the power pool level b "fh ™ ^ be bmU within 15 '«*
surance of passing y i*e 'instruction agency, as a further as-
downstreaw uses. The rese 'ofr »™P"ature and quality ,or
only time and sciemjJL *Sefvolr xs now mim% (spnng of 1963) and
of the biologists 3nd the "engine1 ^ whe1Jer the rerommendations
a the engineers, arrived at jointly, were sound
Cooperation in stbeamflow regulation
-------
and will accomplish the desired objectives. Certainly, the demands
placed upon the reservoir will increase, and any action short of that
stemming from the best thinking and available knowledge would be
inadequate.
Inasmuch as the John H. Kerr dam holds the key to the seasonal
discharge of water into the downstream reservoirs, the Steering
Committee, early in its program, recommended that water be stored
in the Kerr dam between elevations 300 and 302 feet for the benefit
of the striped bass. This would be made available to augment mini-
mum flows that would occur during peaking-power operations and
over weekends. The critical period was found to be that time when
the striped bass approached their spawning grounds in the Roanoke
River. This period begins sometime after May 1 and extends from
2 to 4 weeks, depending upon flows, temperatures, and rainfall
(Figure 2). The volume of water provided by the Kerr Dam would
be scheduled by the power company to meet the fishery requirements
with a minimum loss of generating energy. The plan proposed was
found successful and was recommended as one of the permanent
operating features.
C 320|
TOP OF SPILLWAY GATES
V/////////,,,
£
310
3
«
z
a
5
LU
5
3 ///
300'
^ LkL' 1 ¦ ,!j! ,^-v
¦; nv:.v:^
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Figure 2 — Proposed and interim use of woter storage tpaea, John H. Kerr Reiervoir. Cour-
te«y, U.S. Army Engineer Diitricf, Norfolk (Ref. 2).
The White River, Arkansas and Missouri
The basin of the White River contains about 28,000 square miles,
60 percent of which is in Arkansas and 40 percent in Missouri. The
White River, which is 720 miles long, rises in the Ozark Mountains
in northwest Arkansas, flows northward into Missouri, then south-
ward back into Arkansas and empties into the Mississippi River
(Figure 3).
King and Kinney
387
-------
The upper half of the White River and tributaries contained
some of the most famous float fishing waters in America. The prin-
cipal species fished for were: smallmouth, spotted, and largemouth
bass, walleye, rock bass, other panfish, and suckers. The lower half
of the river supported moderate commercial fisheries. The species
taken most commonly were: largemouth bass, crappies, channel and
flathead catfish, buffalo, freshwater drum, carp, and gar.
After the 1937 flood on the Ohio and Mississippi Rivers, Congress
approved several flood control acts that resulted in a comprehensive
plan for the White River Basin. The plan consisted of a series of
reservoirs in the upper reaches of the river and various channel
improvements in the lower reaches.
The Corps of Engineer studies were begun in 1938. Some 15
reservoir sites were considered. Some sites could not be justified
economically- Others were strongly opposed by various conservation
groups. The 1945 plan proposed 12 reservoirs for flood control, power,
recreation, and other uses. The 12 reservoirs would have a total
flood pool area of about 350,000 acres and would inundate 751 miles
of streams in the heart of the float fishing trip area.
To date, Bull Shoals, Clearwater, Norfork, and Table Rock reser-
voirs have been completed. Greers Ferry Reservoir is nearing com-
pletion and Beaver Reservoir is under construction (Figure 3).
The Norfork Reservoir, which was the first to be constructed, was
completed in 1944. A coordinated fish and wildlife study by the Fish
and Wildlife Service and the States of Arkansas and Missouri was
initiated at about the time the reservoir began filling. Because of
personnel shortages during World War II, a comprehensive study
was not undertaken until 1946. Two studies were conducted. One
considered the values of the reservoir fishery as compared to the loss
of stream fishery in the inundated sections. The annual fish and wild-
life losses were estimated to be about $40,000 greater than the gains
from the impounded waters.
The second study considered the effects of the cold discharges
from the hydroelectric reservoirs. It was found that, except for the
sucker population, the warm-water fishery was practically eliminated
below the dam on the Norfork River.
In 1948, the Norfork Reservoir tail water was stocked with trout.
The stocking was very successful. This resulted in an increased de-
mand for more trout by the fishermen and local interests. When Bull
Shoals Reservoir began operation in 1952, it was found that about
30 miles oi suitable trout water was provided by the cold discharges
of this impoundment (Figure 4) (5),
In order to supply the demands for trout for these new waters
cold enough to support trout, it was necessary to construct a hatchery.
The Norfork National Fish Hatchery, which is located below the dam,
was completed in 1956. The water supply comes from the reservoir.
Trout from this hatchery are stocked principally in reservoir tail-
388 COOPERATION IN SXBEAMFLOW REGULATION
GPO 821—740—f S
-------
Springfield
,./i
Fayetteville
Scale in Miles *
Figure 3 — The White River Waterthed, Arkansas and Missouri.
NORFORK
J>eserr voir
"*.t Mountai
C, Home
BULL SHOALS f '
RESERVOIR
X
I
N
LEGEND
WHITE RIVER FISHERY ZONES
TROUT WATER 30 Miles
MARGINAL WATER 86 Miles
BARREN WATER 50 Miles
. „ WARM WATER
it*^"
, QatesviMe
10
20
Scale in Miles
Figure 4 — While River below Bull Shoals and Norfork Reservoirs showing the effects of
the colder discharges on the fisheries (Ref. S).
King and Kinney
3S9
-------
Wat.!5S anM-ln 6 ~ ~boul strea™ in northern Arkansas and
southern Missouri. The present annual production is about 200 000
pounds. The hatchery is: now being expanded and b i964 ( >u™
capable of producing 400,000 pounds of trout.
By 1957, trout £e»ig stocked in 91 miles of tailwater
streams that afforded 47,792 man-days of fishing(6).
During the Period from 1946 to 1956, investigations conducted
on the effects of bottom diScharges from reservoirs in Arkansas
Tennessee, and Kentucky revealed that several downstream warm-
water fisheries had been virtually eliminated by cold water dis-
charges. It was found that summer water temperatures were averag-
ing from 10 to 20 degrees Fahrenheit below pre.impoundment tem-
peratures. Management of these modified reaches of the White River
still present difficult problems. An extension of the zone of trout
stocking has been fairly successful, although expensive. Waters
designated as barren in Figure 4 by the 1957 report do support some
fish and may be improving. ume
Based on earlier studies and cooperative studies on the proposed
Greers Ferry Reservoir, the Pish and Wildlife Service recommended
to the constructing agency in 1957 that multi]eve] outlets be provi^°
at the Greers Ferry Dam. It was believed that it would be necessary
to discharge sufficient quantities of the warmer, near-surface waters
to protect the downstream warm-water fishery. The annual value of
the downstream fishery was estimated to be $123,000.
The constructing agency estimated that the additional cost for
TJ~¦$800,000 and re<3uested that fishery b i-
ologists Provide a justification for this additional expenditure As
aZeS?^l?s SuH? m+ .^on?mun^cations, funds were not made avail-
a y until about 2 years after the request was made.
Study was competed, work had begun
°n afirst?™ of th/S ~ the study alonS w^h a request for the
modification of the dam structure to provide for warm water releases
^ ^ ^ agenCy' The instructing agency re-
plied that to incorporate the modifications requested would require a
^ns'o?dolLTUld in loss o, LeJ
a P^iic learned that multilevel outlets were not planned
bliter contr„versv XlTm
-------
about 10 to 15 miles of good trout water, 10 miles of fair trout water,
30 miles of marginal water, and 30 miles of possibly barren water.
It will be necessary to construct an $800,000 hatchery to provide
trout for anticipated stocking needs in the area.
The most favorable outcome was the assurance by the construct-
ing agency that fish and wildlife matters would be coordinated with
the agencies concerned in the future at the earliest feasible stage and
continue through the survey, planning, construction, and operation
of the project.
The Assistant Secretary of the Army in a letter to the Secretary
of the Interior, dated January 9, 1961, stated that, "The difficulties
inherent in coordinating fish and wildlife as well as recreation pur-
poses in the development of multiple-purpose water resources projects
will continue to require the full cooperation of all concerned."
The Potomac Rivet
The Potomac River basin presents many unusual problems partly
because it encompasses mountainous areas, the rolling Piedmont, and
the coastal plain, including the Chesapeake Bay estuary. It drains
land from four States and passes through the District of Columbia.
It is visited by large numbers of people and in many respects will be
looked upon as a model watershed development. It has received a
great deal of attention from many agencies and must serve a large
variety of purposes, with water supply and recreation becoming of
increasing importance.
The Potomac River Basin Study(7) is an excellent current
example of what the Corps of Engineers call "Partners in Planning."
This study, which will be completed in June 1963, "will consider all
the beneficial uses of water and must provide a balanced program
to satisfy to the maximum possible extent all present and anticipated
water requirements in the most efficient and economical manner."
An examination of the various objectives of some of the previous
Potomac Basin studies illustrates the evolution of the multiple pur-
pose concept. The first major study was made in the 1920's and
specified river development for navigation, water power, flood con-
trol, and irrigation. The second study, which was published in 1946,
recommended a series of reservoirs for flood control and the genera-
tion of hydroelectric power.
In 1956, the Corps of Engineers was directed to prepare a basin-
wide plan for flood control, recreation, the development and con-
servation of municipal and industrial water supplies, and water
quality control. The present study should result in a maximum mul-
tiple-use development program. It will require considerable effort
and cooperation among the various agencies and groups on Federal,
State, and local levels.
Federal Agencies. — Department of the Army, Corps of Army Engi-
neers — is in overall charge of the planning, coordination, and major
impoundment studies.
King and Kinney
391
-------
Department of Agriculture — is developing a watershed protec-
tion and management program and is supplying data on agricultural
and rural use in the Basin. The Forest Service is preparing develop-
ment plans for water projects that will be on National Forest lands.
Department of Health, Education, and Welfare — is making
studies of water supply, sanitary water quality, and pollution abate-
ment needs.
Department of the Interior —
1. The Geological Survey is supplying data on sedimen-
tation, streamflow, chemical quality, and ground
water resources.
2. The Fish and Wildlife Service is studying the effects
of the proposed water development projects on fish
and wildlife resources and is recommending meas-
ures to gain maximum fish and wildlife benefits.
3. The National Park Service has supplied data on the
recreational resources and has evaluated the recre-
ational aspects of the project areas.
Department of Commerce — the Office of Business Economics
prepared an economic base study of the Basin.
Federal Power Commission — will supply power value estimates
and make hydroelectric studies.
State Agencies. — The States of Maryland, Pennsylvania, Virginia,
and West Virginia contain waters that are in the Potomac River
system. Various State agencies have certain responsibilities in regard
to water developments. In most cases, the State agencies work closely
with their Federal counterparts.
Local Agencies, — In addition to the various Federal and State agen-
cies, an advisory group, chaired by the District Engineer, was formed.
This group includes representatives from the four States in the basin
and the District of Columbia, the National Capital Regional Planning
Council, and the Interstate Commission on the Potomac River Basin.
The Interstate Commission has been interested principally in pollu-
tion control in the Potomac River Basin. The Potomac may well be
our most studied river.
THE BUREAU'S PROGRAM IN SPORT FISHERY RESEARCH
The Bureau of Sport Fisheries and Wildlife conducts a fishery
research program designed to provide answers to some of the prob-
lems encountered in planning and managing reservoirs where fishery
resources are significant. The maintenance of satisfactory recreational
frequently on® the objectives in reservoir development.
The decline in sport fishing, which is often encountered 5 to 10 years
following impoundment, is not inevitable, as has been demonstrated
392
COOPERATION IN STREAMFLOW REGULATION
-------
on some of the TVA reservoirs and in other locations where man-
agement programs have helped maintain good fishing.
The Bureau's research program includes projects related to
changes in productivity and water quality following impoundment.
The requirements of sport and forage fishes for spawning and growth
are being studied, as well as the life histories and behavior of some
of the principal species. Information is being gathered that ultimately
will be useful in advising anglers on characteristics of reservoirs that
increase the probability of fisherman success. Methods of gathering
and analyzing population data are being developed. The effects of
polluting substances will be observed, and the relationships that
occur as a result of various water uses will be examined. Limnolog-
ical studies will be undertaken to learn the effects of water level
fluctuation, rate of water exchange, water withdrawal at various
depths, and the flooding of standing timber on basic productivity and
Ash production. The chain effects of a series of reservoirs, which are
Presently not well understood, will be examined.
Field work is presently underway on Upper Missouri River basin
reservoirs, with a project hGadQuartered at Yankton, South Dakota.
Another phase of the reservoir study is concerned with the reservoirs
and streams of the White River in Missouri and Arkansas. The latter
Will be headquartered at Fayetteville, Arkansas, where some work
is being done in cooperation with the University of Arkansas.
The Bureau's Branch of River Basin Studies reviews plans pro-
posed by the Federal Government or by public or private agencies
Under Federal permit or license for impoundment, diversion, and
other modifications of natural waters, to determine possible effects
on fish and wildlife resources. It develops, in cooperation with the
State conservation agencies, recommendations for the steps necessary
to compensate for damage to fish and wildlife resources by proposed
developments and can frequently suggest ways that the projects can
be modified to benefit fish and wildlife resources. The maintenance
of the required water temperatures, oxygen, and other quality char-
acteristics to meet the needs of fishes, other aquatic species, and
recreational interests is of high importance.
The States are giving increasing attention to w^ter quality
studies associated with flow regulation. A list of the Federal Aid
Projects active in this field in 1961 is included as Appendix III.
CONCLUSIONS
Demands for water will undoubtedly double by 1980, and most
river systems will be subject to extensive development. Continuing
cooperation by all interests will be required in each water develop-
ment project where flow regulation is involved if the desired water
Quality is maintained and maximum benefits to the public are to be
achieved.
Five cardinal principles of cooperation where flow regulation
for quality control is involved are stated as follows."
King and Kinney
393
-------
1. The basic problems must be defined in terms of public need,
presently and in the future.
2. The specific requirements of controlling interests and most
important uses in terms of water quality must be known and
set forth early in project planning.
3. There must be a willingness on the part of all agencies to
meet together, to understand differing requirements, and to
negotiate agreements on project features.
4. There must be a full exchange of research information as soon
as it can be made available.
5. There must be continuing evaluation of recommendations and
management methods within broad policy decisions.
REFERENCES
1. Ellis, M. M., B. A. Westfall, and M. D. Ellis. Determination of
water quality. Research Report No. 9, U. S. Fish Wildlife Service.
1948. 121 pp.
2. U. S. Army Engineer District, Norfolk. John H. Kerr Dam and
Reservoir. Information Bull. 1962. 50 pp.
3. Fish, F. F. The minimum river discharges recommended for the
protection of the Roanoke River anadromous fishes. Issued by
the North Carolina Wildlife Resources Commission, Raleigh,
N. C. Dec. 1, 1960. Mimeo, unpublished. 51 pp.
4. Fish, F. F. Report of the Steering Committee for Roanoke River
Studies, 1955-1958. Raleigh, N. C. Mimeo, unpublished. 1959.
292 pp.
5. U. S. Fish and Wildlife Service. An evaluation investigation of
the fisheries resources in the downstream segments of the White
River below Bull Shoals and Norfork Dams, Arkansas. Office of
River Basin Studies, Region 4, U. S. Fish Wildlife Service. Feb.
1957. 14 pp.
6. Baker, R. F. Historical review of the Bull Shoals Dam and Nor-
fork Dam tailwater trout fishery. Proc. 13th Annual Conf. South-
eastern Assoc. Game and Fish Commissioners, 1959. pp. 229-236.
7. U. S. Army Engineer District, Baltimore, Potomac River Basin
Study. Information Bull. 1961. 30 pp.
APPENDIX I.
EXCERPTS FROM THE FISH AND WILDLIFE COORDINATION
ACT, AS AMENDED, 1958. (16 U.S.C. 661 ET SEQ.)
"Sec. 2 (a) Except as hereafter stated in subsection (h) of this
section, whenever the waters of any stream or other body of water
394
COOPERATION IN STREAMFLOW REGULATION
-------
are proposed or authorized to be impounded, diverted, the channel
deepened, or the stream or other body of water otherwise controlled
or modified for any purpose whatever, including navigation and
drainage, by any department or agency of the United States, or by
any public or private agency under Federal permit or license, such
department or agency first shall consult with the United States Fish
and Wildlife Service, Department of the Interior, and with the head
of the agency exercising administration over the wildlife resources
of the particular State wherein the impoundment, diversion, or other
control facility is to be constructed, with a view to the conservation
of wildlife resources by preventing loss of and damage to such re-
sources as well as providing for the development and improvement
thereof in connection with such water-resource development.
*******
"(f) In addition to other requirements, there shall be included
in any report submitted to Congress supporting a recommendation
for authorization of any new project for the control or use of water
as described herein (including any new division of such project or
new supplemental works on such project) an estimation of the wild-
life benefits or losses to be derived therefrom including benefits to be
derived from measures recommended specifically for the development
and improvement of wildlife resources, the cost of providing wildlife
benefits (including the cost of additional facilities to be installed or
lands to be acquired specifically for that particular phase of wildlife
conservation relating to the development and improvement of wild-
life), the part of the cost of joint-use facilities allocated to wildlife,
and the part of such costs, if any, to be reimbursed by non-Federal
interests.
*******
"(h) The provisions of this Act shall not be applicable to those
projects for the impoundment of water where the maximum surface
area of such impoundments is less than ten acres, nor to activities
for or in connection with programs primarily for land management
and use carried out by Federal agencies with respect to Federal lands
under their jurisdiction.
*******
"Sec. 5. The Secretary of the Interior, through the Fish and
Wildlife Service and the Bureau of Mines, is authorized to make such
investigations as he deems necessary to determine the effects of
domestic sewage, mine, petroleum, and industrial wastes, erosion
silt, and other polluting substances on wildlife, and to make reports
to the Congress concerning such investigations and of recommenda-
tions for alleviating dangerous and undesirable effects of such pollu-
tion. These investigations shall include (1) the determination of
standards of water quality for the maintenance of wildlife; (2) the
study of methods of abating and preventing pollution, including
methods for the recovery of useful or marketable products and by-
products of wastes; and (3) the collation and distribution of data
King and Kinney
395
-------
on the progress and results of such investigations for the use of
Federal, State, municipal, and private agencies, individuals, organi-
zations, or enterprises.
*******
Sec. 8. The terms 'wildlife' and 'wildlife resources' as used
herein include birds, fishes, mammals, and all other classes of wild
animals and all types of aquatic and land vegetation upon which
wildlife is dependent."
*******
APPENDIX II.
PRESENT AND FUTURE MINIMUM-FLOW REQUIREMENTS
STIPULATED IN FEDERAL POWER COMMISSION LICENSE 2009,
REVISED I960
The Federal Power Commission reviewed the application of the
Virginia Electric and Power Company to construct and operate a
hydroelectric plant at Gaston, on the Roanoke River, North Carolina.
The Commission issued their findings and order relative to the Gaston
Project on March 25, 1960, to become final 30 days thereafter. The
document was issued as an amendment to F.P.C. License 2009, com-
bining the Roanoke Rapids and Gaston Projects.
The new provisions as recommended by the Steering Committee
for the Roanoke River studies, and concurred in by the State of North
Carolina, the U. S. Fish and Wildlife Service and the Virginia Electric
and Power Company, are as follows:
Article 25. Effective with commencement of operation of the
Gaston Development, the Licensee shall release sufficient water from
the Roanoke Rapids reservoir to maintain minimum flows and pounds
of oxygen in the Roanoke River as measured at the point of project
discharge to meet the following requirements:
Requirement A: Minimum Instantaneous Flows
Month Cubic Feet Per Second
January, February, March 1,000
April 1,500
May, June, July, August, September 2,000
October 1,500
November, December 1,000
The above flows are subject to the following special provisions:
(1) These flows may be reduced on weekdays, Monday
through Friday, during off-peak hours by not more
than 20 percent, but such reduction shall not dispense
with compliance with Requirement B.
396
COOPERATION IN STREAMFLOW REGULATION
-------
(2) The average flow for any day shall equal or exceed the
minimum instantaneous flow specified for that month.
(3) A minimum instantaneous flow of 2,000 cubic feet per
second will be furnished by the Licensee for the period
requested by the North Carolina Wildlife Resources
Commission, to begin as ear?y as April 1, but not later
than April 15, and to continue for at least 60 days but
not longer than 75 days in any one year, in accommoda-
tion to annual variations in the time and duration of
spawning activities of the striped bass.
(4) The reduction in instantaneous flows permitted above
shall not apply when special spawning flows are being
passed for the benefit of striped bass.
Requirement B: Minimum Oxygen
Discharge from the Roanoke Rapids Development shall be main-
tained to provide dissolved oxygen at an instantaneous rate of not
less than 78,000 pounds per calendar day during the. months May
through October, except as is permitted under the following condi-
tions:
(1) A reduction not in excess of 34 percent of the instan-
taneous rate of 78,000 pounds per day will be permitted
for periods not exceeding 14 consecutive hours.
(2) Any oxygen deficit so created shall be offset by greater
discharges so that a cumulative average rate of dis-
charge of 78,000 pounds per day will be attained within
a period up to but not exceeding 16 hours from the
beginning of the oxygen deficient flows, and this condi-
tion (2) shall again begin to operate as soon as the
instantaneous rate again falls below 78,000 pounds
per day,
The Licensee shall not be required to compensate through in-
creased flows or other changes in operations for any lessening in the
quality of water entering the Gaston Development which may after
the effective date of this article be created by Acts of God, or through
acts of others than the Licensee, and beyond its reasonable control.
Requirement C:
In the event of temporary emergency conditions arising in per-
formance of Requirements A and B the Licensee will cooperate in
good faith with the North Carolina State Department of Conserva-
tion and Development, North Carolina State Stream Sanitation Com-
mittee and/or North Carolina Wildlife Resources Commission to take
such reasonable steps in regard to the reductions permitted by item
(1) under Requirement B as may be proper to meet the emergency
conditions and in the interest of maintaining during such emergency
conditions water quality in keeping with the standards required of
class "C" waters by order of the North Carolina State Stream Sani-
King and Kinney
397
-------
tation Committee effective to September 1, 1957. But nothing herein
is intended to compel the Licensee to correct conditions which it has
not created and no permanent change of operations shall be made
without the approval of the Federal Power Commission.
Article 26. The volume of controllable releases of water from
the Roanoke Rapids reservoir shall not be permitted to increase to
double, or to decrease to half, of any prevailing discharge in less than
one hour during the spawning period of striped bass. The spawning
period shall be the period from April 1 to June 1 of each year, except
as modified upon the recommendations of the North Carolina Wildlife
Resources Commission as provided for in Articles 24 and 25.
Article 27. The normal operating level of the Roanoke Rapids
reservoir is established at elevation 132 above sea level. (U.S.C. and
G.S. datum).
Article 28. The Licensee shall construct, maintain and operate
such protective devices and shall comply with such reasonable modi-
fications of the project structures and operation in the interest of fish
and wildlife resources, provided that such modifications shall be
reasonably consistent with the primary purpose of the project, as
may be prescribed hereafter by the Commission upon its own motion
or upon recommendation of the Secretary of the Interior, the North
Carolina Wildlife Resources Commission, or the Virginia Commission
of Game and Inland Fisheries, after notice and opportunity for hear-
ing and upon a finding based on substantial evidence that such modi-
fications are necessary and desirable and consistent with the provi-
sions of the Act; Provided, however, that no modifications of
project structures shall be required unless recommendations are made
by the North Carolina Wildlife Resources Commission, the Virginia
Commission of Game and Inland Fisheries, or the Secretary of the
Interior prior to November 30, 1960, or such further time as the
Federal Power Commission may prescribe by its order issued prior
to November 30, 1960.
Article 29. The Licensee shall consult with the U. S. Fish and
Wildlife Service, the North Carolina Wildlife Resources Commission
and the Virginia Commission of Game and Inland Fisheries, and on
the basis of this consultation shall undertake appropriate measures
for preserving and improving fish and wildlife resources of the
project area as may be prescribed by the Federal Power Commission
upon the recommendations of the Secretary of the Interior, the North
Carolina Wildlife Resources Commission, and the Virginia Commis-
sion of Game and Inland Fisheries, after notice and opportunity for
hearing.
Article 30. The entire project area, except for such portions as
may be reserved for reasons of safety and efficient operation, shall
be open to free use by the public for the purposes of fishing, hunting,
and other recreational pursuits; and the Licensee shall allow to a
reasonable extent for such purposes the construction of access roads,
wharves, landings and other facilities on project lands the occu-
398
COOPERATION IN STREAMFLOW REGULATION
-------
pancy of which may in appropriate circumstances be subject to pay-
ment of rent to the Licensee in a reasonable amount: Provided, that
the Licensee's consent to the construction of access roads, wharves,
landings and other facilities shall not without its express agreement
place upon the Licensee any obligation to construct or maintain such
facilities.
Article 32. The Licensee shall be responsible for reregulating ad-
ditional water discharged from John H. Kerr reservoir either specifi-
cally for the protection of the striped bass fishery during the migra-
tion and spawning period, or specifically for stream sanitation pur-
poses, in accordance with an agreement or agreements to be entered
into by the Licensee and the affected State and Federal agencies
subject to the approval of the Commission."
APPENDIX III.
STATE RESEARCH PROJECTS DEALING WITH WATER QUAL-
ITY IN RELATION TO FISH AND WILDLIFE RESOURCES
About $1,600,000 annually of Dingell-Johnson Federal Aid Funds
are programmed for reservoir research. In addition, most States that
have reservoirs, large lakes, and coastal waters, are conducting non-
Federal aid research on water quality.
Following are brief descriptions of the 1961 Federal Aid Projects
which include water quality studies.
Arkansas F-3-R-6. A survey of possible development of trout
waters. To obtain physical, chemical, and biological data on tail-
waters of multiple-purpose dams and to collect pre-impoundment
fish population data.
California F-18-R-1. Experimental management of reservoirs.
Florida F-8-R-6. River Basin Fisheries Investigations. To eval-
uate effects on fishery resources of proposed water control structures
of the Corps of Engineers in Florida flood control projects.
Idaho F-34-R-4. Water Quality Investigations. To study the
effects of pollutants on water quality, formulate water quality stand-
ards, and to determine methods of improving water quality.
Kentucky F-16-R-3. Pre- and Post-Impoundment Population
Surveys.
F-19-R-1. Pre- and Post-Impoundment Limnological Studies
and lake investigations.
Louisiana F-7-R-2. Lake D'Arbonne and Bussy Lake Investiga-
tion. To study the biological dynamics of two new reservoirs.
Maine F-8-R-10. State-wide Lake and Stream Investigation.
To survey lakes and streams for physical, chemical, and biological
characteristics.
King and Kinney
399
-------
Minnesota F-13-R-5. Investigation of Production of Natural Fish
Foods. To investigate the relationships of fertility of water and soils
to the production of natural fish food organisms with special emphasis
on lethal limits for various fertilizers.
Mississippi F-6-R-6. Fisheries Investigation on Flood Control
Reservoirs.
F_8_R_4, State-wide Lake and Stream Survey. To obtain in-
formation relative to physical, biological, and chemical characteristics.
F-9-R-3. To detect sources of stream pollution and its effect
upon bottom organisms, fishes, and chemical qualities of the water.
Montana F-9-R-10. To determine the physical, chemical, and
biological characteristics of important fishing waters.
F-23-R-5. To compile data on age and growth, and to study the
relation of pollution on bottom organisms.
New Mexico F-22-R-3. To conduct basic surveys; make pre-
impoundment studies; . . . cooperate in River Basin Studies ....
New York F-22-R-3. Trout: Water Chemistry Relationships.
North Carolina F-14-R-1. State-wide Fish Management Investi-
gations. To survey and classify significant freshwater streams; in-
vestigate pollution as it affects sport-fishing waters ....
South Carolina F-9-R-2. An Evaluation of Stocking Striped Bass
In Reservoirs. To study results of stocking striped bass in selected
reservoirs and effects of physical and chemical factors on success.
Tennessee F-5-R-9, Reelfoot Lake Investigations. To evaluate
effect of restricted commercial fishing and water level fluctuations;
conduct creel census, growth, food, movement, and limnological
studies.
F-12-R-7. Large Impoundment Investigations. To determine
populations; rate of exploitation; movement; depth distribution of
game fishes; seasonal oxygen and temperature profiles; and effect of
brush shelters.
Texas F-2 to 9 and 12-R7-9. Fisheries Investigations. Includes
pollution surveys.
F-16-D-1. Caddo Lake Circulation Channels. To direct water
flows into stagnant areas of Caddo Lake.
Utah FW-5-R-4. River Basin Investigations. To determine fish
and wildlife implications of private- and government-sponsored river
basin projects, and recommend features for mitigating losses.
Washington F-24-R-7. Environmental Control. To determine
the nutrient deficiencies of various bodies of water, methods of adding
organic fertilizers, and to study mineral metabolism.
Wisconsin F-60-R-1. Cold Water Fishery Research. To study
effects of angling regulations; evaluate habitat development; docu-
400
cooperation in streamflow regulation
-------
ment water quality studies; and to determine survival, growth and
yield of various streams.
Wyoming FW-3-R-9. Game and Fish Laboratory Research.
(Part) To investigate fish toxicants, water pollutants, and reaction
of trout to salinity.
DISCUSSION
W. W. Towne
Director, Columbia River Basin Project
U. S. Public Health Service, Portland, Oregon
It is a pleasure for me to return to Cincinnati. As Joe Flanagan
said, these symposia have grown. I recall the Dissolved Oxygen
Symposium, with which I had some little connection, that we held
some 6 years ago in our small auditorium at the Sanitary Engineer-
ing Center. We were crowded, but we were able to conduct the
meeting there. I am sure some of the things we have been discussing
here were discussed then; we are still talking about them and will
be for some time to come.
Dr. King outlined some of the legislation that assigns certain
responsibilities in the field of water resources development and water
quality control to Federal and State agencies. Since the Public Health
Service is one of these agencies, I thought my best contribution here
might be to briefly outline some of the cooperative steps that have
been taken in the Columbia River Basin. Most of these were initiated
and in operation long before I was assigned there, but I am happy
to be taking part now in some of these activities, which I think are
very progressive steps.
In the Columbia River Basin, as in any other basin, there are
some 20 Federal agencies that are interested in water resources
development as a result of authorizing legislation. There are at least
an equal number of separate State agencies. A Federal agency is
authori2ed by Congress to undertake some activity under specific
legislation. To carry out their responsibilities they receive a budget.
I think all of us here today recognize the interplay between the areas
of responsibility of the several agencies; if one agency, such as the
Fish and Wildlife Service or the Public Health Service, is responsible
for giving information to some of the other agencies, particularly
agencies responsible for the engineering design and construction of
these structures, the information must be given in such a form that
it can be utilized on a timing schedule that meets the need.
The Columbia River Basin Inter-Agency Committee is composed
of the Federal and State agencies concerned with the development of
land and water resources in the Pacific Northwest. This area in-
cludes the coastal waters of Oregon and Washington as well as the
Columbia River Basin because of its interrelated economic and devel-
opment picture.
Towne
401
-------
The Columbia River Basin Inter-Agency Committee has been
very active. It has about 20 active subcommittees. It had a Subcom-
mittee on Comprehensive Planning, on which were representatives
from the State and Federal agencies concerned with water resources
planning operations. In September 1962, the parent committee recom-
mended that the name of the subcommittee be changed to the
Coordinated Planning Subcommittee to emphasize the coordinating
aspects of its activities. I would like to read to you a brief excerpt
from the report of the Executive Committee recommending this
change.
"The function of the Subcommittee on Comprehensive
Planning be broadened to include the coordination of plan-
ning for the Columbia River Basin Inter-Agency Committee;
further, that this Subcommittee be directed to establish a
coordinating task force for each major river basin or group
of related basins as the need is determined."
The latter item relates specifically to intensive studies resulting
from Federal appropriations to one agency or another to carry out
broad, comprehensive planning operations. The Corps of Engineers
has just received authorization and budget from Congress to expand
their plan and studies on the Willamette Basin. A task force was set
up to coordinate the working relationships of the water resources
agencies in developing this plan.
"The objectives of the Coordinated Planning Subcom-
mittee are to provide a forum for the free exchange of
information concerning the planning activities of member
agencies; work towards the establishment of mutually satis-
factory priorities and schedules for water and related land
resources planning, and towards the development of fiscal
requirements to meet these coordinated planning schedules;
We think that this is a key item — to coordinate our fiscal
requirements.
". . . work towards the selection of comparable criteria
and procedures, uniform basic data, and consistent projec-
tions of future conditions and needs; provide and supervise
separate task forces as required for major basin and sub-
regional studies to effect detailed coordination of planning
at the working level; and to keep the parent committee
fully informed concerning the status and needs of coordinat-
ing and planning activities."
The Coordinated Planning Subcommittee has established sub-
basins into which the various studies fall. We have devised a reporting
procedure for the agencies so that we have some idea of the mag-
nitude of their present and future plans. These studies are categorized
into five general types:
(1) Comprehensive Framework Plans. These are area studies
that provide (a) long-range projections of economic development,
402
DISCUSSION
-------
(b) a translation of such projections into demands for water and
related J and-j~es012r.ee jzes, (c) hfdrslcgic projections of water avail-
ability both B5 -to quantity and quality, and {d) projections of the
availability of related land resources, Thes& studies enable lis to
outline the characteristics of projected water and land resource
problems and the general approaches that appear appropriate for
their solution,
(2) Comprehensive Detail Plans. Extension of the Framework
Plans to the detail necessary for the Bureau of Reclamation, Corps
of Engineers, or the Soil Conservation Service to request authoriza-
tion of one or more multipurpose reservoir projects in a subbasm.
(3) Special Project Plans. Plans oi a narrower geographic and
analytic scope, which usually relate to the request for appropriations
for one multipurpose reservoir.
(4) State Survey Plans, to bring the states, which are a part
of this committee, into the planning picture -we must know their
resource and land development agencies, their economic agencies,
and their plans so that they can be made a part of the overall water
resources planning.
(5) Related Water-Resources Activities. In addition to the agen-
cies with responsibility far broad planning, many of the other agen-
cies have supporting types of operations. The Public Health Service
is not only a planning agency, but is also responsible for giving
certain supporting information on needs for water quality storage
to the construction agencies. The U. S. Geological Survey, the
Weather Bureau, and the Bureau of Outdoor Recreation also provide
supporting information. We need to know their plans, their budget,
and what each can contribute to the planning effort. The Coordi-
nated Planning Subcommittee is attempting ta compile information
within these various categories an the fiscal year in which studies
have been or will be initiated, the expected completion date, the
percent completed, and the estimated cost of completion. We would
like as much estimated budget and scheduling data as possible for
the anticipated life of the project. We realize that these estimates
are subject to the directives of Congress, but if we are to plan on a
comprehensive basis, we must have something to stand with, Prac-
tically all agencies have some long-range plan, some idea related
to what they hope to do and what they feel is necessary in the next
few years in the field of water resources development.
Just a closing word: We that are associated with comprehensive
planning for water quality control in the Columbia River Basin feel
that this is one basin where planning can result in maximum divi-
dends. Much of this area is still relatively unpolluted so that pre-
ventive measures can ensure continued good water quality. All of
us recognize that prevention is cheaper than cure. Furthermore, it is
easier to keep an undegraded resource at a high level and put into
it some reserve for the future than it is to bring a downgraded water
resource back to where it might have been if preventive actions had
been taken.
Towne
103
-------
DISCUSSION FROM THE FLOOR
Mr. Bell, U. S. Public Health Service• Ac
cannot miss the opportunity to put in a i native Arkansan, I
of the state, even though I regret thrf ,P , for the fishinS industry
the U. S. Fish and Wildlife Service an* *1 °f cooPeration between
the White River Basin Reservoirs If a f ^ons^ruction agencies on
good fishing place, g0 to north Arkansas It i^ reali^ t0 to a
Mr. Morris, U S. Public Health Service- Th« r t *
are here and that we have heard ¦ °f people who
Service, Corps of Engineers, Federal ? includes the Public Health
Wildlife Service, Weather Bureau Dona,0wer Commission, Fish and
of Outdoor Recreation, Coast akd r ^f111 of AgricuZture> Bureau
Home Finance Agency, Bureau of Bp] Survey, Housing and
Federal commissions such as ORS A MCi®matlon', State governments
interests, public and private utility municiPa3ities, industrial
U Uies' and private land owners
I wonder, Mr. Towne, have wp „ *
some formal coordinating asenrv i« reached the point where
often overlapping interests undpr « Cessary *° put the diverse and
house and coordinate these water agency to act as a clearing
water resources activities?
Mr. Towne, V. s Public w
Congress now directing a centra^zatio^06'' J1*** is a bil1 before
traiization of water resources planning
I think — and this is pn+- i
Public Health Service — j y pers°nal, I do not speak for the
development is so broad anH Wh°!e field of water resources
operation under a single Dlanm'ncr S dlversified interests that
The Public Health Service fn "'? agfncy would be very difficult
development of facts relative P ' 18 basicaI1y responsible for
people who determine how the w^ter'?«T fr°m which those
judgments. 6 Water 15 to be used can arrive at sound
that there1" is no onelnSvidul" "nd^ m* h^' * somewhat fearful
organization, that can objecWlv ay ?,no.small staff within one
various segments 0f water rPc Vjew e imPortance of all the
to have s/ecialists 1^"^° d«elopment. Maybe it is better
Portance of their activities | gencie.s .who can speak on the im-
emphasized if they Were under nn af i Ts migbt be inadequately
person to develop his 0* ""^er one head. It is human nature for a
a single water resources aepnm interests, and administration by
another. agency WouId lean, I think, to one area or
agencies may be slower *houfht- Coordination between
autocratic method, but in a n somewhat less efficient than a more
decisions are made by the deij\°.cratic country such as ours the final
doubt whether a smalf the Public and they need the best facts I
the broad picture as w « Up 50uld marshal all the facts and present
pollution control, itl ^^pe^ialist^in flood control, in water
404
discussion
-------
Mr. McDermott, U. S. Public Health Service: I would like to
speak as a representative of the Advanced Waste Treatment Unit of
the Basic and Applied Sciences Branch at the Robert A. Taft Sanitary
Engineering Center. I have been considering some cost figures while
listening to the talks. The picture obtained is interesting enough
that I want to use it to present a point and pose a question. Assume
a situation where the BOD of the stream is 15 milligrams per liter
and it is necessary to reduce this to 7.5 milligrams per liter. Assume
further that the BOD of the water released from storage is 5 milli-
grams per liter. Under these assumptions, for each 1,000 gallons of
water in the stream, 3,000 gallons of dilution water would be re-
quired to produce a mixed water with a BOD of 7.5 milligrams per
liter. If the cost of providing the dilution water is assumed to be
2 cents per 1,000 gallons, then the cost of upgrading the quality of
the original river water would be 6 cents per 1,000 gallons.
I would like also to point out a problem in storing water for
waste water dilution in the Western states. In many areas there is a
shortage of irrigation water and even of domestic water. I can foresee
difficulty in getting people to accept release of water for quality
control if water for their livelihood is in short supply. These com-
ments lead to a question as to what cost levels or range of costs are
generally encountered in streamflow regulation; in other words, what
are the costs that advanced waste treatment methods must meet to
be competitive?
Mr. Towne: I fully agree with your comment on the objections
that will probably be raised to streamflow regulation in the West,
but we do not claim that low flow augmentation, or flow regulation,
is the most economical means of achieving a degree of quality con-
trol that we can reasonably accomplish by waste treatment now. The
fact that money is being made available for advanced waste treatment
research shows that the Public Health Service recognizes the im-
portance of furthering treatment capabilities rather than providing
dilution indefinitely. It may be some years, however, before effective
application of advanced waste treatment methods can be realized.
In Oregon, the Governor and the Water Resources Board have
stated that Oregon does not have enough water to use it for pollu-
tion abatement, i.e., in lieu of treatment. We do not propose flow
regulation in lieu of treatment. The Water Pollution Control Act
requires that we utilize treatment insofar as possible. As advances
are made in research and new knowledge assures higher degrees of
treatment, some of the water that is being stored at this time for
quality control can well be used for other purposes. This has been
the history of any water resources development. Until the Water
Supply Act of 1958 permitted the inclusion of storage for anticipated
municipal and industrial water supply in multiple-purpose reservoirs
we were not able to provide for future needs. Before that time
someone had to enter into a contract for repayment before capacity
could be provided in a project. We have all seen reservoirs built that
created a water demand that could not be paid for in advance. Then
the water could not be taken from the reservoir because it was allo-
From the Floor
405
-------
the state°waterSresourhmk the State water pollution control agency
allocating water should agency' ,or the state engineer, whoever ^
as it is needed. I do ^ +^frVe water for quality control as 1
flow supplementation wK any us are advocating the use
a"on where treatment can be effective.
tion of competition^or Engineers: This rather sticky
me out of my chair Th taS ansen and> I'm afraid, it has gotte
here on problems assnSt,! has been a lot of technical discussio
essentially nothing until QUality contro1' but 1 haVG ?!iue
and how you choose w ROW on how y°u determine its val '
storage forCOmpetine *** ™ the allocation of
a few months ago. We'wprt0-^^'68 with a Bureau of the Budget
me the question: "Sinre ..m.volved in basin planning, and he as
have sufficient storage sitf>c .ls commonly accepted that y°u d°
the needs, how will vou H +n °f Michigan to satisfy
. nmnj , m,ne the storage allocations for * ,
P rpose reservoirs in your basin planning •
• « J ? _ • 1a
. ' *»
various uses of your
vuus in your Dasjn -
us data in a form that°^ 1 the cooPerating agencies
including each purpose on J¦ the determination of the valui
other purposes we are .mcremental basis in competition w
projects." going to do the best we can to optimlze *
best quality of wate^ ^chnical discussion on how to achieve the
question as to the true VZ a°T/egulatio^ * would like to raise a
hility, and the probability£? m reSulation in light of the P^
or at least limit, other uses lny cases> that it is going to prec
uses of that sort? h as water recreation, irrigation, an
being, like motherhood tK°k-a!; the interests of our own agency *>s
raised above all others' w! \ st tllat has to be preserved a
is fully considered in thp ry to ensure that everything ir1v
of the poor man who has °f the Reiving agency. But tb«J*
tions in the form that wo v, a^e our evaluations and recommen
to the Public Health Servipp - them to him ~ and r11 direcCt Te
sponsored symposium — Ik w n°e this is a Public Health Servi ?
evaluations of the benefit to take the Public Health Servi
scheme for optimizing th* >J? £°W reeulation and fit them ltit0
poses You report that certain « ^ aU the P°ssible pr°jeCt t£e
investigated possible alternS fl°WS are required, that you
quality control, that you eSa? ?easures of Providing equivale"
quality control, and that it ? wil1 cost so much to ProVlde r„u
have a B/C ratio of one to nS Now, does this mean y
you have a B/C ratio of infinUy?°r QUality contro1- or does
from using a unifofTtorL^V*7 to comPare the value of the benefits
storage for water quality control with the value
406
DISCUSS!^
-------
of using it to maintain a fishery, or for water ^P|y> or for flood
control, or for any of the other uses, we are m
If we have to reduce or eHminatej°me uses ^use of^
storage capacity or limited runoff, how c ^ of storage for
of a unit of storage for quality c0Jltr.° . ter quality control
some other purpose? Is flow regulation ^ regardless
to be placed on a pedestal and considere ^ fits from fishing are
of the value of competing uses. Recreat , i e they present
similarly presented by the Fish and Wildli e storage being worth
fishery benefits in terms of a certain amount of storage
so much. . gs
Senate Document 97 considers ^^construction agencies
of life from floods. It seems to me that values and to consider
are being pressured to consider these kinds o have a common
secondary benefit values, it is imperative go ^hat when we
denominator in our consideration of these the combination
apply a project formulation procedure we people in the
of allocations that will maximize the benefits for the p
basin and in the country as a whole.
Perhaps this should be the subject of Competition for
can we work together, realizing that we _omDeting uses should
water and that determining how a project ^ aSency says, "This is
be formulated is almost impossible^ if ev y ^ „^e don't care
what you should put in your Project, saym of Engineers has ^
about the other possible uses. Yet th allocations of
report to Congress as to what combmai , 'to a particular
should be ascriDea way
uusin ana in me cuuhu^ ao o
Perhaps this should be the subject of ^^^^g^mpetition for
can we work together, realizing that we _omDeting uses should
water and that determining how a project of P g) »This is
be formulated is almost impossible if eye^ fffprt <«we don't care
what you should put in your project saymgm ^ E'ngineers has to
about the other possible uses. Yiet 1tlp allocations of
report to Congress as to what combmai ^d to a particular
facilities, and resultant benefits should be ascribed to p
project at a given site. .
Mr. Bell: Thank you, sir. ™at *
and I can assure you it could be the J ^
Mr. McLean, V. S. Public Health ^en^7 ^nsiderations. We do
another symposium that will include econ recognize the
not feel the time is quite right fortha; n • several desirable
necessity of means to discriminate be meeting is directed
purposes for which storage can be ProY\ ' jr1 water quality control,
toward better answers to technical pro H_tter answers to the prob-
Subsequent meetings may contribute o provided by quality
lems of economic evaluation of the
nn*. _ 1
From the Floor
407
-------
WHERE WE STAND ON STREAMFLOW
REGULATION FOR QUALITY CONTROL
K. S. Krause
Chief, Technical Services Branch, Division of Water Supply and
Pollution Control, Public Health Service, Washington, D. C.
This meeting is unique inasmuch as it is the first symposium on
streamflow regulation for quality control. The willingness of such a
high proportion of those attending to stay through this final presenta-
tion indicates a very vital and continuing interest in this subject.
I would like to ask just one simple question: Did the symposium
accomplish what it set out to do? First of all, its stated purpose was
to facilitate exchange of information on the effects of streamflow
regulation on water quality; secondly, to review available knowledge
on the effects of streamflow regulation on water quality; thirdly, to
discuss studies presently in progress on quality control problems asso-
ciated with streamflow regulation; and, finally, to explore research
needs in the field of streamflow regulation for quality control. I hope
you will agree with me that we can answer this question with a
resounding "Yes."
Mr. Stein referred to some of the early discussions of certain
legal matters as furnishing more heat than light. I believe in this
symposium we have had more light than heat. I do not intend to
review each paper in trying to sum up this particular symposium.
I am reminded of the student psychologist who taught his pet flea
to jump every time he said, "Jump!" But this psychologist had a
research bent, so he decided to see how well the flea would jump if
he removed some of his legs. He removed the hind pair of legs and
told the flea to jump, and the flea jumped. "Well," he thought, "this
is fine. What will happen if I remove another pair?" So he took
off another pair of legs, and he told the flea to jump, and again
the flea jumped — rather feebly, but he jumped. "Ah," he said, "what
will happen if I take the last pair off?" So he removed the last pair
of legs from the flea. He again told the flea to jump, but the flea just
quivered a little bit and stayed put. He shouted a little louder; still
the flea did not move. Then the student psychologist reached this
conclusion: "Fleas must hear through their legs!"
I am sure our conclusions would be on much firmer ground if
our systems analysis were more complete or we had Dr. Fiering to
push the buttons. But philosophizing a bit, one of our famous edu-
cators made the observation that we can never expect to know every-
thing, and a man must make decisions based upon the knowledge
that he has. Certainly knowledge and experience prepare each of
us to make faster and better decisions, but in our complex society
the cost of errors in decision is great and ever increasing. I am sure
a child could make decisions on almost any subject, and some would
be good. I believe, however, that we must strive to push the pro-
portion of correct decisions as much towards the hundred percentile
Krause
409
-------
as we can at this moment in history. The farther we proceed towards
this goal, the tougher each increment of progress becomes.
We are dealing with a complex subject, one that has far-reaching
technical, economic, and social ramifications. We cannot afford not
to examine our knowledge before making decisions. If there is one
earth-shaking conclusion I have reached here at this meeting, it is
that even problems have problems.
But what have we accomplished?
Forty persons have presented topical discussions covering a vast
array of subject matter, all pertinent to streamflow regulation for
water quality control. These might be grouped generally as dealing
with administrative, policy, hydraulic, hydrological, chemical, bio-
chemical, thermal, physical, and engineering evaluation types of
problems. We have touched on both the legal and economic ramifica-
tions of streamflow regulation. Both are sufficiently important to be
the subjects of separate symposia in the future. We have touched on
decision-making tools that we hope will put all of the former in-
formation into proper perspective. I am sure we have not heard
the last of this.
Now to some observations:
1. Fluctuations in water quality are serious economic and pro-
ductive factors to industrial, municipal, agricultural, fishery, and
recreational water users.
2. Reservoirs now in operation for quality control purposes have
shown themselves to be useful tools for adding oxygen, for thermal
control, for damping fluctuating inorganic loads, and for neutralizing
mineral acids both in the reservoir and downstream. Reservoirs have
also effectively reduced microorganisms at downstream water intakes.
3. Reservoir management for maximum water quality im-
provement shows promise for stabilization of thermal, organic, and
inorganic quality.
4. Flow augmentation can have adverse as well as beneficial
effects on downstream quality.
5. Reservoir storage and its resultant effects on water quality
must be studied on a case by case basis. We are cautiously optimistic
about the potential usefulness of flow regulation.
6. Reservoir outlet design features play an important role in
the quality of water emerging from reservoirs.
7. Proper control of land runoff and preparation of reservoir
storage sites may markedly improve water quality in storage.
8. Simulated hydrology and mathematical techniques for esti-
mating quality will help us to identify our deficiencies and permit us
to concentrate on overcoming the most significant weaknesses in our
knowledge.
410
WHERE WE STAND ON STREAMFLOW REGULATION
-------
9. Analysis of the effects of storage and flow control should be
made on an entire river basin to properly evaluate the accumulated
advantages and disadvantages, since time and location of events are
important and often controlling factors in whether the benefits out-
weigh the liabilities.
10. Streamflow regulation may be our only practical method of
improving water quality in some areas and instances.
Dissolved oxygen and its association with water quality has been
given major attention in this meeting. But, let us not forget that
there are other and perhaps equally important quality parameters.
A new science of water resources seems to be emerging. It is a
strange looking creature, but so is a camel; yet he functions very
effectively.
There are some challenges ahead, too. First, we need to be able
to simulate conditions in an impoundment prior to construction so as
to be able to predict results of storage on water quality. Second, we
need to improve our techniques for measuring water quality, espe-
cially our techniques for automatically monitoring the quality. Third,
we need to resolve problems of conflicting interests associated with
water storage and releases; some of these problems are legal. Fourth,
we need to develop operating criteria that will give the greatest
efficiency to river basin control systems to achieve the maximum
water use and re-use for all legitimate purposes. Fifth, we need to
improve our use of economic forecasts as a basis for estimating future
water needs. Sixth, we need to develop more effective structural
designs for hydropower intakes that can select the water, and hence
the quality, that will be discharged. And finally, by way of a little
thought teaser, we need to seek a common denominator for compar-
ing efficiencies of not just one system, but several systems. I am sug-
gesting that we look to basic energy in the physical and chemical
area as a possibility for this common denominator. We need some-
thing to give us a benchmark by which our choices can be effectively
made.
Gentlemen, it has been a pleasure to be here with you, to hear
so many fine papers, and to have so much active participation in the
discussions. I am sure that the symposium proceedings will provide
a wealth of information for you; I know they certainly will for us.
Krause
411
-------
BANQUET SESSION
Master of Ceremonies: Maurice LeBosquet
U. S. Public Health Service
-------
PUBLIC EXPECTATIONS
Dr. Maurice K. Goddard
Secretary, Pennsylvania Department of Forests and Waters
Harrisburg
Your program chairman is a very clever man! He advised me
that my topic would be of my own choice, and then, to make sure
that I would not get too far off-base and discuss something like
"Streamflow Regulation for Quality Control in Outer-Space," he
suggested that a broad view of the objectives of the general public,
since these objectives are the basis of our establishment and desire
to maintain quality objectives, might be a good subject for a stimu-
lating address. So be it. I can take a hint!
In my mind the general public really has but a single objective
insofar as water quantity and quality are concerned. And, it is not
even strictly an objective, but rather, an expectation. The general
public expects that someone, his government or his water company,
will see to it that he is delivered adequate supplies of good water for
all of his uses and to meet all of his demands. Pure water, and plenty
of it at all times, is considered by the American public as its right
and heritage. About the only time the general public becomes actively
concerned is when, for some reason, that delivery is curtailed, or
the quality of the water deteriorates to a point where it becomes
distasteful or unusable.
When either or both of these things happen, the general public
usually asks just three questions: Why did it happen? Who or what
was responsible? What is being done, and by whom, to correct the
conditions? Somebody is then expected to come up with quick and
satisfactory answers!
That somebody is usually the head or administrator of the agency
or agencies charged with the responsibility of ensuring the availabil-
ity of, or the protection of, water supplies. His answers, of course,
come from those unsung heroes — the scientists, engineers, and
technicians on his staff.
The general public's lack of concern about the services it expects
until those services are interrupted is well known, and that lack of
concern is not confined to things pertaining to water. The trait is
simply inherent in human nature.
The public, then, is only aroused collectively when a change in
some existing condition, either natural or man-made, disturbs its
comfort or status quo. Do not misunderstand me, I am not deriding
this trait — I have a profound respect for the general public. After
all, I serve it, and I am a part of it.
Now, let us divide the general public into its components —
you and I — the John Q. Citizens who make up the whole. Taken as
an individual, John Q. has many facets — his thinking and viewpoint
Goddard
415
-------
on any subject may be guided by either his vocation or his avocation,
or may be colored by both. He, too, may be aroused when his normal
comfort and status quo are disturbed, but, when his avocation is
directly affected in any manner, he is even more likely to make his
desires known. In order to do this effectively, he bands together
with other, similarly affected individuals having the same avocation,
and a new organization is born.
This point is well illustrated in the following statements ex-
tracted from a report, in pamphlet form, recently issued to the people
of Pennsylvania by the Pennsylvania Department of Health, and
entitled, "People and Water, Water and People."
"Like the torrent of an onrushing stream that cannot be stopped,
when people speak, they must be heard.
"Although clean water is the goal of the people, they may seek
action for different reasons. The sportsman wants a stream he can
fish in, and if the stream is polluted, he must know the reason why,
and what can be done about it. When nature is destroyed needlessly,
the sportsman feels like a kettle of water boiling over.
"But, he is not alone in his deep feelings.
"Industries that need water for their livelihood cannot counte-
nance a clean stream becoming abased, nor can polluting industrial
plants withstand the ire of the people, when the people understand
that water pollution can be stopped. Industrial groups in Pennsyl-
vania have taken concerted action to protect streams.
"Civic groups, women's clubs, the community councils see the
polluted stream from a different point of view. Many of the groups
act for clean streams simply because the cause is right. Others see
the threat of beauty destroyed, still others recognize that water
pollution is a health and economic threat."
In Pennsylvania, those of us who are charged with the responsi-
bility of assuring that adequate supplies of good water are available
at all times to the public are always on the look-out for new weapons
in our fight to clean up our streams and keep them clean. We con-
sider streamflow regulation for quality control to be a promising new
weapon. It is a comparatively new term, representing a change or
equalization of basic emphasis.
While improved quality was inherent in the concept of low-flow
augmentation for downstream water supply, the basic emphasis in
providing low-flow augmentation was to increase quantity. We rec-
ognized, of course, that such releases would, under most conditions,
also improve downstream water quality during low-flow periods.
Storage and streamflow regulation for water quality control simply
means, then, that now extra quantities of water will be stored for
subsequent release for further purification of the river after maxi-
mum waste treatment. We realize that it is a supplementary measure
to be used in conjunction with, and not in lieu of, necessary waste
treatment.
416
PUBLIC EXPECTATIONS
-------
Some 30 years ago, the Commonwealth of Pennsylvania con-
structed the Pymatuning Dam and Reservoir on the Shenango River
in western Pennsylvania for water supply and flood control, the
former being furnished to downstream communities and industries
through low-flow augmentation. The project has since become a
major recreation center as well. More recently, the amended decree
of the United States Supreme Court, dated June 7, 1954, in the
Delaware River Case, provided for a River Master to supervise the
diversions of water from the Delaware River Basin by New York City
for water supply, and the compensating releases from the New York
City water-supply reservoirs to augment the flow of the Delaware.
Even more recently, the U. S. Army Corps of Engineers' compre-
hensive study and plan for the development of the water resources
of the Delaware River Basin to the year 2010, which was completed
in 1960, provided for meeting downstream water supply needs
through low-flow augmentation. And, about the same time, my own
Department completed water resources development plans for the
Brandywine and Codorus Creek Basins in Pennsylvania, covering the
same period and providing upstream storage for release to meet
downstream water supply needs.
In all of these cases, the first consideration was to make increased
quantities of water available during low-flow periods. Possible bene-
fits from improved quality of water during the same period, while
recognized, were considered secondary to the main quantity objective.
Figure 1 shows the components of flow in the Delaware River
at Montague, New Jersey, from the latter part of May to October,
1962. This Figure is reproduced from the Delaware River Master's
1962 report. Incidentally, Montague is across the river and just a
little downstream of Milford, Pennsylvania, the headquarters of the
Delaware River Master,
You will note from the Figure that we had drought conditions in
the Delaware River Basin during the summer months and that about
64 percent of the total flow of the Delaware measured at Montague
during July, August, and September was made up by the compen-
sating releases from the New York City water supply reservoirs.
During the month of July, 72 percent of the total flow came from
those releases. It does not take much imagination to picture what
the condition of the Delaware River would have been, both quan-
tity- and quality-wise, if this water had not been released to the river
during those critical periods.
While Pennsylvania and New Jersey fought hard to prevent
the out-of-basin diversion of Delaware River water to New York
City, we have since found that the cloud did have a silver lining.
Without the upstream storage provided in the New York City reser-
voirs and the compensating releases, the Delaware would have been
in grievous trouble more than once since 1955, when the Pepacton
and Neversink reservoirs went into operation. Even more improve-
ment can be expected when the third reservoir, Cannonsville, is
completed and operating.
Goddard
417
-------
»*
ci
Dd
r
M
X
"fl
m
n
>-3
>
H
1
GO
o
«»-
o
<0
Ul
o
Od
<
X
o
<0
JUNE
N.Y.C. 13.9 BG
Power 4.2 BG
Uncontrolled 15-2 BG
TOTAL 33.3 BG
JULY
N.Y.C. 21.2 BG
Power 1 0 BG
Uncontrolled 7.1 BG
TOTAL 29,3 BG
AUGUST
N.Y.C. 19.2 BG
Power 3.4 BG
Uncontrolled _g.6 BG
TOTAL 31.2 BG
Figure 1 — Components of flow in the Delaware River at Montague, N.J. 1962.
-------
As you all know, one of Pennsylvania's major pollution problems
stems from acid mine drainage, a problem shared by a number of
our sister states. We have, in fact, approximately 2,300 miles of acid
streams. Our summary statement of present and anticipated water
problems to the Senate Select Committee on National Water Re-
sources in 1959 pointed out that while we have been relatively suc-
cessful in preventing further deterioration of our streams by mine
drainage — we have been about holding our own since 1945 — the
problem of discharges from active and inactive mines located along
already acid streams had certainly not been solved.
One of our recommendations in that statement is very interesting:
"On watersheds where acid mine drainage cannot be eliminated
or adequately reduced at the source, study is needed to determine
the practicability of, and the advantage to be gained by (a) the im-
poundment of such drainage during peaks of high acid flow with
subsequent regulated release of the acid waters so that downstream
adverse effects will be eliminated, or (b) the impoundment of clean
waters in dams on clean streams, tributary to acid-polluted streams,
and the release of such waters to augment the flow and neutralize the
effects of acid mine drainage in the receiving stream."
The waters of the West Branch Susquehanna River are highly
acid from its headwaters to the vicinity of Lock Haven, Pennsylvania,
where they are usually neutralized by the flow of a large, non-acid
tributary, Bald Eagle Creek. You will note that I qualified that state-
ment by saying that they are "usually" neutralized at that point.
The fact is that when heavy rains over the headwaters of the
West Branch have missed the Bald Eagle watershed we have had
heavy fish kills almost as far down the West Branch as its confluence
with the North Branch at Sunbury. Upon the completion of the last
two reservoirs of the four-reservoir system for flood control in the
West Branch Basin, we will have a unique opportunity to change
this situation.
The Corps of Engineers is considering the storage of acid waters
in the Curwensville Reservoir, now under construction on the upper
main stem of the West Branch, and plans releases to aid in pre-
venting downstream fish kills and in neutralizing water sufficiently
to make it suitable for many industrial uses. At the Blanchard Dam
and Reservoir on Bald Eagle Creek, now under planning, provision
will be made for storage of nonacid waters for water quality control.
Coordinated operation of the two reservoirs for water quality control
promises great improvement in the quality of the waters of the
West Branch Susquehanna River.
It was, of course, the 1961 amendment to the Federal Water
Pollution Control Act that officially added this new aspect of pollu-
tion control by recommending that consideration be given, in the
planning of any reservoir, to the inclusion of storage for the regula-
tion of streamflow as a means of water quality control. Obviously,
too, that amendment is directly responsible for this symposium —
Goddard
419
-------
we are here because the public expects us to make the best possible
and most efficient use of our new weapon, and we are exploring the
means of doing just that. Previews of the Corps' comprehensive study
and plan for the development of the water resources of the Potomac
River Basin, soon to be released, indicate that for the first time in a
major river basin plan consideration is being given to storage of
water for quality control. Estimates of the quantity of water that
will be needed in the various segments of the basin by 2010 will
include the quantities needed to ensure water of good quality through
streamflow regulation. The estimates, of course, presuppose that
maximum waste treatment will be provided.
Earlier, you will recall, I mentioned that when John Q. Citizen's
avocation is directly affected he is likely to make his desires known.
In closing, therefore, I wish to present the plight and complaint of
one Angus, a citizen of Scotland, as it appeared in a news-story from
the town of Largs, Scotland, on March 14 of this year.
Angus complained to the Town Council that the taste of the
town's water supply was so bad that he was forced to travel 30 miles
to Glasgow for bottled water to mix his drinks. The harried munici-
pal water engineer explained that the heavy snow melt was carrying
the taste of heather from the hills into the town reservoir, and then
added that the taste would soon disappear and that, in any case, it
would not harm the man's whiskey.
So you see, Ladies and Gentlemen, while John Q. expects us in
America to keep his water clean, Angus Q. expects our counterparts
in Scotland to keep his whiskey tasty!
420
PUBLIC EXPECTATIONS
GPO 821-740-16
-------
BIBLIOGRAPHIC: SYMPOSIUM ON STREAMFLOW
REGULATION FOR QUALITY CONTROL, CIN-
CINNATI, OHIO, APR. 3-5, 1963. J. E. McLean,
Chairman. Robert A. Taft Sanitary Engineering Cen-
ter. PHS Publ. No. 999-WP-30. June 1965. 420 pp.
ABSTRACT: This symposium on streamflow regula-
tion for water quality control, held in Cincinnati
in April 1963, was jointly sponsored by the Tech-
nical Services and the Basic and Applied Sciences
Branches of the Division of Water Supply and Pol-
lution Control of the Public Health Service. The
Proceedings contains 23 papers, plus prepared and
floor discussions. Papers describing the relation of
streamflow regulation for quality control to the
major Federal water resource development programs
(Public Health Service, Corps of Engineers, Bureau
of Reclamation, Soil Conservation Service, and Fed-
BIBLIOGRAPHIC: SYMPOSIUM ON STREAMFLOW
REGULATION FOR QUALITY CONTROL, CIN-
CINNATI, OHIO, APR. 3-5, 1963. J. E. McLean,
Chairman. Robert A. Taft Sanitary Engineering Cen-
ter. PHS Publ. No. 999-WP-30. June 1965. 420 pp.
ABSTRACT: This symposium on streamflow regula-
tion for water quality control, held in Cincinnati
in April 1963, was jointly sponsored by the Tech-
nical Services and the Basic and Applied Sciences
Branches of the Division of Water Supply and Pol-
lution Control of the Public Health Service. The
Proceedings contains 23 papers, plus prepared and
floor discussions. Papers describing the relation of
streamflow regulation for quality control to the
major Federal water resource development programs
(Public Health Service, Corps of Engineers, Bureau
of Reclamation, Soil Conservation Service, and Fed-
BIBLIOGRAPHIC: SYMPOSIUM ON STREAMFLOW
REGULATION FOR QUALITY CONTROL, CIN-
CINNATI, OHIO, APR. 3-5, 1963. J. E. McLean,
Chairman. Robert A. Taft Sanitary Engineering Cen-
ter. PHS Publ. No. 999-WP-30. June 1965. 420 pp.
ABSTRACT: This symposium on streamflow regula-
tion for water quality control, held in Cincinnati
in April 1963, was Jointly sponsored by the Tech-
nical Services and the Basic and Applied Sciences
Branches of the Division of Water Supply and Pol-
lution Control of the Public Health Service. The
Proceedings contains 23 papers, plus prepared and
floor discussions. Papers describing the relation of
streamflow regulation lor quality control to the
major Federal water resource development programs
(Public Health Service, Corps of Engineers, Bureau
of Reclamation, Soil Conservation Service, and Fed-
ACCESSION NO.
KEY WORDS:
Aeration
Flow regulation
Hydrology
Hydroturbines
Mixing
Monitoring system
Nutrients
Photosynthesis
Power plants
Reservoirs
ACCESSION NO.
KEY WORDS:
Aeration
Flow regulation
Hydrology
Hydroturbines
Mixing
Monitoring system
Nutrients
Photosynthesis
Power plants
Reservoirs
ACCESSION NO.
KEY WORDS:
Aeration
Flow regulation
Hydrology
Hydroturbines
Mixing
Monitoring system
Nutrients
Photosynthesis
Power plants
Reservoirs
-------
in,
ersl Pcxe" Con.rrxssir»rO, ta State acid federal water
rights, and ta public expectations in water resource
deve«GpKt2ryl programs^ were presented Other p-apers
¦cQBsjtfpred ib? effects f t ir-tj.'oi_c-inn",rjvs on ihe-rrn-ai
slrijjficabort. biocfieirfeal cxids-tfofl, pfiotosyirttvetic
oxygen production and ch&.-nical section an'g«c production, and chertiirai} sc utiar aad a«-
dotation in resecting J-.e effects cf r reg-Jjatloti
on wasi? ESi ir;jla:; >r. mineral .jja'ir;... temperature,
and r.utiiervts: iiyiroiogic data needs, statistical
studies ef storage - yield relationships, and estab-
lishment of quality objectives for Saw regulation
projects; measures Iot improving the quality of
reservoir discharges through selection ot discharg-e
level, turbine aeration, and reservoir Kiisictg and
aeration; and the place ot monitoring arid coopera-
tive effort in the solution of water quality* problems
through flow regulation.
Stratification
Waste assimGsti<3n
Wat&r law
Tester -€-swjrw£
davstopjcerit
VPaier quality
COTlttoi
Water quality
Wicalws
Water quality
objectives
r
era; Pimiec Comtnisatasn.;-, to Slate and Federal water
to ftxpoos if"_ tv~3tej' r&s-cuLrce
development programs, were presented. Other papers
considered the effects of impoundments an therm&l
stratification, biochemical oxidation. tahc'jasyrthe'-jLr
cvygen prcdmitice s.k; c&wik&I sulxion and pre-
tpisati'-c _r -aser.a.r:; tae affi&cas of Bow insulation
on "waste assimilation, miceral quaJii*, teicperaluis.
aad nutrients: tfdr*dagie data rjmls, 5ta:j5*jJa-'
studies of storage - jMd relationship an(j e5jab'
lishir.ent of quality objectives for Bow iegulatkw>
projects; measures for ^proving the quality of
reservoir discharges thrcaugt sstectian oi Sschar&>.
level, turbise »&rttitoc: arrd r^ejrvoir *oixUtg and
aeration; and ti# clace ci manibctfo-.g and coopera-
tive effort ta the solution ai water quality orofeierars
threats h flow i-egulaUcri.
Waste assimilation
Water law
WattriwoWMs
CBVsiaf ncett
"Water aucabt.'
c-antrcil
Water Quality
indicators
Vf&ter •'fWtliTj
object! k;
------- |