EPA-600/3-77-004
January 1977
Ecological Research Series
A GUIDE TO AERATION/CIRCULATION
TECHNIQUES FOR LAKE MANAGEMENT
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Corvailis, Oregon 97330
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U,S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects. This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-77-004
January 1977
A GUIDE TO AERATION/CIRCULATION
TECHNIQUES FOR LAKE MANAGEMENT
by
Marc Lorenzen and Arlo Fast
Tetra Tech, Inc.
Lafayette, California 94549
Contract Number 68-03-2192
Project Officer
D.P. Larsen
Ecosystems Modeling and Analysis Branch
Corvallis Environmental Research Laboratory
Corvallis, Oregon 97330
CORVALLIS ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CORVALLIS, OREGON 97330
-------
DISCLAIMER
This report has been reviewed by the Corvallis
Environmental Research Laboratory, U.S. Environmental
Protection Agency and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendaton for use.
XI
-------
FOREWORD
Effective regulatory and enforcement actions by the
Environmental Protection Agency would be virtually impossible
without sound scientific data on pollutants and their impact
on environmental stability and human health. Responsibility
for building this data base has been assigned to EPA's
Office of Research and Development and its 15 major field
installations, one of which is the Corvallis Environmental
Research Laboratory-
The primary mission of the Corvallis Laboratory is
research on the effects of environmental pollutants on
terrestrial, freshwater, and marine ecosystems; the behavior,
effects and control of pollutants in lake systems; and the
development of predictive models on the movement of pollutants
in the biosphere.
This report provides a valuable contribution to knowledge
and practice in the field of eutrophication control with lake
aeration/circulation techniques.
A. F. Bartsch
Director
Corvallis Environmental Research
Laboratory
111
-------
TABLE OF CONTENTS
Page
Foreword .....................
List of Figures ................. ix
List of Tables .................. xiv
Units and Conversions .............. xv
Sections
I. INTRODUCTION ................. 1
BACKGROUND ............... 1
Reduce Nutrient Influx ....... 1
Reduce Nutrient Availability. ... 2
Manage the Consequences of Nutrient
Enrichment ............. 2
SCOPE AND OBJECTIVES .......... 2
LAKE MANAGEMENT ............. 3
AERATION/CIRCULATION .......... 4
Benefits .............. 4
Domestic Water Supply ..... 4
Downstream Water Releases. . . 4
Industrial Uses ........ 4
Fisheries Management ..... 5
Algal Production ....... 6
Application ............ 6
Problem Definition and
Selection of Goals ...... 7
System Selection ....... 7
II. DESTRATIFICATION ............... 9
BACKGROUND ............... 9
EXPERIENCE AND METHODS ......... 10
v
-------
TABLE OF CONTENTS (Continued)
Sections Page
Mechanical Mixing I2
Diffused Air Mixing 13
DESIGN CONSIDERATIONS 13
Aeration 13
Algal Bloom Control 15
System Selection 21
Induced Circulation 22
Effects of Circulation 22
Diffuser Design 26
Compressor and Supply Line 29
Potential Problems 29
III. HYPOLIM-NETIC AERATION 30
BACKGROUND 30
Experience and Methods 31
DESIGN CONSIDERATIONS 35
Hypolimnetic Volume Estimates. ... 35
Oxygen Consumption Rates 39
Oxygen Input Capacity 42
Aeration Devices 45
Air Required 45
Air Source 46
Diffuser Design and Bubble
Size 47
Potential Problems 47
Undersized Aeration Capacity. . 47
Unintentional Thermal
Destratif ication 47
Nitrogen Gas Supersaturation. . 48
High Free Carbon Dioxide. ... 49
Water Level Fluctuations. ... 49
VI
-------
TABLE .OF CONTENTS (Continued)
Sections Page
IV. EXAMPLES OF METHODOLOGY APPLICATION 50
INTRODUCTION 50
PAST EXPERIENCE 50
Kezar Lake 50
El Capitan Reservoir 52
San Vicente Reservoir 52
Hypolimnion Aeration 54
Oxygen Required 54
Minimum Air Flow .... 54
Air Flow Based on
Saturation 54
Destratif ication 55
Cost Estimating 56
Hypolimnion Aeration . . 56
Destratification .... 56
V. REFERENCES 57
VI. APPENDIX A 65
LAKE MODEL COMPUTATIONS 65
Physical Representation 65
Temperature 65
Algal Growth 66
Imposed Mixing 67
Application-Verification 67
Test Cases . 69
VII. APPENDIX B 88
HYPOLIMNETIC AERATION/OXYGENATION ... 88
Vll
-------
TABLE OF CONTENTS (Continued)
Sections Page
HYPOLIMNETIC AERATION/OXYGENATION
DEVICES 88
Mechanical Agitation System ... 88
Air Injection System 88
Partial Air Lift Designs . . 88
Full Air Lift Designs. ... 92
Downflow Air Injection ... 96
Other Air Lift Designs . . . 102
Oxygen Injection Systems 106
VIII APPENDIX C 115
AIR SUPPLY LINE AND COMPRESSOR
SELECTION 115
Design Parameters 115
Selection and Installation. . . . 117
Air Line 117
Compressor 119
IX. APPENDIX D 120
HYPOLIMNION AERATOR SIZING 120
Vlll
-------
LIST OF FIGURES
Number Page
II-l Destratification system installed at
El Capitan Reservoir, California 14
II-2 Generalized plot of peak algal biomass
as a function of mixed depth for both
nutrient and light limitation 18
II-3 Net specific growth rates of Amphani-
zomenon (from Lorenzen and Mitchell,
1975) 20
II-4 Total flow of upwelled water as a
function of air release rate and depth
of release 23
III-l Hypolimnion aerator used by Bernhardt
(1974) in the Wahnbach Reservoir 33
III-2 Suggested modifications to Bernhardt
hypolimnion aerator 34
III-3 Hypolimnion aerator degassing chamber
details 36
III-4 Hypolimnion aerator by Fast (1971).... 37
III-5 Temperature profiles in Ottoville Quarry
with and without hypolimnetic aeration. . 38
III-6 Example of oxygen depletion rates data. . 41
IV-1 Theoretical and observed peak biomass in
Kezar Lake, 1968 stratified; 1969 and
1970 artificially destratified 51
IV-2 Temperature and oxygen profiles in the
San Vicente Reservoir 53
A-l Diagramatic representation of air induced
vertical mixing 68
A-2 Comparison of computed and observed
temperature profiles in Kezar Lake. ... 70
IX
-------
LIST OF FIGURES (Continued)
Number
A-3 Comparison of computed and observed
temperature profiles in El Capitan
Reservoir 71
A-4 Computed effects of mixing on temperature
stratification in Kezar Lake 73
A-5 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 1 74
A-6 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 2 75
A-7 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 3 76
A-8 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 4 77
A-9 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 5 78
A-10 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 6 79
A-ll Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 7 «* 80
A-12 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 8 81
A-13 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 9 82
A-14 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 10 83
x
-------
LIST OF FIGURES (Continued)
Number Page
A-15 Computed effects of induced mixing
on temperature and algal stratification
in hypothetical lake no. 11 84
A-16 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 12 85
A-17 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 13 86
A-18 Computed effects of induced mixing on
temperature and algal stratification in
hypothetical lake no. 14 87
B-l Mechanical aeration system of hypolimnion
aeration from Lake Bret, Switzerland
(Mercier and Ferret, 1949; Mercier and
Gay, 1954; Mercier, 1955) 89
B-2 Exposed view of LIMNO hypolimnetic
aerator. This is a partial air lift
design since hypolimnetic waters are
upwelled only a short distance by air
injection (Fast, Dorr and Rosen, 1975a). . 91
B-3 A proposed partial air lift system of
hypolimnetic aeration (Speece, et al.,
1974) 93
B-4 A full air lift, hypolimnetic aerator
used in Hemlock Lake, Michigan (Fast,
1971). This is a full air lift design
since water is upwelled to the surface
before it returns to the hypolimnion ... 94
B-5 A full air lift, hypolimnetic aerator
used in Wahnbach Reservoir, West Germany
(Bernhardt, 1974) 95
B-6 A full air lift, hypolimnetic aerator
used in Lake Tullingesjon, Sweden
(Bengtsson, et a_l. , 1972) 97
XI
-------
LIST OF FIGURES (Continued)
Number Page
B-7 Plan view of Lake Tullingesjon, Sweden
showing the hypolimnetic aerator with
four outlet pipes (Bengtsson, et al.,
1972) 98
B-8 Full air lift, hypolimnetic aerator
used at Lake Jarlasjon, Sweden (Bengts-
son, et al. , 1972) 99
B-9 Plan view of Lake Jarlasjon, Sweden
showing the hypolimnetic aerator and
ten outlet pipes (Bengtsson, et al.,
1972) 100
B-10 Full air lift, hypolimnetic aerator
used in Mirror and Larson Lakes, Wis-
consin by Smith, Knauer and Wirth,
(1975) 101
B-ll A proposed downflow air injection system
which also incorporates the air lift
feature (Speece, 1970) ... 103
B-12 A proposed downflow air injection system
(Speece, et al. , 1974) 104
B-13 "Stand-pipe" hypolimnetic aerator used
at Wahnbach Reservoir, West Germany
(Bernhardt, 1967) 105
B-14 Schematic view of the hypolimnion oxygena-
tion system used at Ottoville Quarry, Ohio
and Attica Reservoir, New York (Fast,
Overholtz and Tubb, 1975b). The cross
hatched area represents the thermocline. 107
B-15 Proposed modes of bubble and water plume
interactions during deep oxygen bubble
injection (Speece, 1975b) 109
B-16 Proposed oxygenated water plume behavior
during deep oxygen bubble injection
(Speece, 1975b) 110
B-17 Downflow bubble contact aerator in
position for proposed hypolimnetic
oxygenation (Speece, et al. , 1974) . . . Ill
XII
-------
LIST OF FIGURES (Continued]
Number Page
B-18 Hypolimnetic oxygenation system used
at Spruce Run Reservoir, New Jersey
(Whipple, 1975) 113
B-19 Isteri oxygen injection system used to
oxygenate under the ice in Lakes Hem-
trask and Kiteenjarni, Finland (Sep-
panen, 1974) 114
D-l Definition sketch for air induced
flow 121
D-2 Ratio of water flow to air release for
different pipe diameters and depth of
release 125
xin
-------
LIST OF TABLES
Number Page
II-l Characteristics of Selected Previous
Destratification Systems 11
II-2 Approximate Air Flow to Mix Various
Lakes 25
III-l Observed and Calculated Oxygen Absorp-
tion and Energy Efficiency for Several
Hypolimnetic Aeration/Oxygenation Sys-
tems 32
III-2 Hypolimnetic Oxygen Depletion Rates from
Selected Eutrophic Lakes and Reservoirs . 43
A-l Geometry of Lakes Simulated with
Reservoir Model 72
xiv
-------
UNITS AND CONVERSION FACTORS
The use of mixed (English-metric) units was considered
to be desirable as well as unavoidable. Certain specifi-
cations and nomographs for solution of flow equations
were only available in English units. Consequently,
necessary computations were illustrated with English
units. Because much of the information contained herein
was taken from published sources it was decided to use
the same units used by the originator. This minimizes
errors in conversion as well as facilitates comparison
to the original work. Useful conversion factors are:
1 meter = 3.281 feet
1 cubic meter = 35.32 ft3
1 acre = 43,560 ft2; 4046 m2
1 acre-foot = 43,560 ft3? 1234 m3
°F = 9/5 (°C) + 32
xv
-------
ACKNOWLEDGMENTS
The authors wish to acknowledge the invaluable
assistance of Messrs. Paul Johanson and Don Smith in
performing the lake temperature and algal growth sim-
ulations as well as computation of water flow-air
release relationships.
xvi
-------
SECTION I
INTRODUCTION
BACKGROUND
"One of the most important problems in the pollution of
inland waters is the progressive enrichment with nutrients
concomitant with mass production of algae, increased water
productivity, and other undesirable biotic changes" (Stumm
& Stumm-Zollinger, 1972). Lakes that continue to receive
nutrients progressively increase their productivity and
deteriorate in water quality.
Excessive algal production which results from nutrient
enrichment has direct as well as indirect implications. The
algal cells can cause unsightly scums and reduce water trans-
parency. The decomposition of settling algae can result in
oxygen depletion and subsequent relase of reduced chemical
species including iron, manganese and sulfides. Lack of
oxygen can eliminate cold water fisheries in many lakes. High
rates of algal productivity can also increase rates of nutrient
cycling within a lake by providing a temporary sink to promote
nutrient relase from sediments. Subsequent algal decomposition
then releases some of the nutrients to the water column.
These problems associated with increasing eutrophication
of lakes have received increasing attention in recent years.
Lake Erie has received national recognition while thousands
of smaller lakes have caused more localized concern. The
Water Pollution Control Act Amendments of 1972 (PL 92-500)
authorized 300 million dollars to be expended for lake restor-
ation or pollution control measures related to eutrophication
of lakes during fiscal years 1973, 1974, 1975. As a result
of this growing concern and awareness of problems there has
been substantial activity related to remedial and preventative
measures.
"Ecological theory and a number of case histories indi-
cate that prospects for preservation and restoration of lakes
are good provided an array of remedial measures and technology
is applied" (Stumm & Stumm-Zollinger, 1972). The array of
preventative and remedial measures which is currently considered
feasible includes both procedures to limit fertility and in-
lake procedures to manage the consequences of eutrophication
(Dunst, e_t al. , 1974) .
Reduce Nutrient Influx
It is generally agreed that the most desirable long-term
lake management technique is to control nutrient input.
-------
Considerable data are available to relate lake nutrient
concentrations to algal production (Sakamoto, 1966; Dillon,
1974) . A number of models have been developed to predict
in-lake phosphorus concentrations based on loading rates,
physical parameters of lake morphometry and flow rates
(Vollenweider, 1969; Snodgrass and O'Melia, 1975; and
Lorenzen, e_t al. , 1976).
Reductions in nutrient influx can result from waste-
water treatment, diversion, modified land use practices,
treatment of inflow, or product modification.
Reduce Nutrient Availability
Several procedures to reduce the level of nutrients
within lakes without changing loading rates have been devised.
Dredging may be appropriate for lakes whose sediments are high
in nutrients. Chemical precipitation of dissolved nutrients
is possible. Dilution or .flushing is possible if large quan-
tities of low nutrient water are available. Harvesting of
both algae and macrophytes can remove nutrients. Lake sedi-
ments can be sealed to prevent nutrient release.
Manage the Consequences of Nutrient Enrichment
Chemical means including the use of algicides, herbi-
cides, and piscicides have been used to control certain
undesirable conditions in enriched lakes. Physical controls
such as lake deepening, harvesting, fluctuating water levels,
and artificial aeration or circulation techniques can also
be used to manage the consequences of nutrient enrichment.
Artificial aeration and circulation techniques can be
used to improve water quality for a wide array of beneficial
uses including domestic water supply, downstream releases,
industrial use, fisheries management, and algal bloom control.
Maintenance of aerobic conditions may also affect nutrient
cycling within a lake.
SCOPE AND OBJECTIVES
Although a variety of chemical and physical control mea-
sures may be appropriate for a specific set of circumstances,
it is beyond the scope of this manual to provide guidance on
selection and application of techniques from this wide array
of possibilities.
This manual concentrates on AERATION-CIRCULATION techniques
The types of problems amenable to solution or control by these
techniques and the procedures for selection of methods and ex-
pected results are described in detail. It is realized that
-------
our knowledge about lake aeration and circulation is far
from complete and that there is a great deal yet to be
learned. However, it is considered appropriate to provide
a guidance manual which is based on a synthesis of theo-
retical considerations and experience. This manual there-
fore represents an attempt to provide the benefits of
current knowledge to those who must proceed from problem
identification to implementation of solutions without
waiting for the results of future research.
It is not the purpose of this manual to provide another
literature review but rather to synthesize from a review the
important variables and relationships which affect perform-
ance of aeration/circulation systems. The manual does not
cover every situation or type of lake but provides examples
and procedures which can be interpreted in conjunction with
site specific conditions.
LAKE MANAGEMENT
A simple example of the decision making process in lake
evaluation is described below:
The first step is to determine the desired uses of the
lake. These may include: domestic water supply, industrial
water supply, warm or cold water fishery maintenance, non-
contact water sports, swimming, and/or aesthetic enjoyment.
The required water quality may vary greatly depending on the
intended use.
The second step is to determine the condition and exist-
ing problems associated with the lake or impoundment. The
determination of physical, chemical and biological properties
of a lake may require an in-depth survey of the lake and
watershed including bathymetry, oxygen profiles, nitrogen
and phosphorus concentrations, algal counts, land use prac-
tices and nutrient loading rates.
The third step is to determine if existing conditions
are compatible with desired uses. If water quality is suit-
able and there is no indication that it is deteriorating,
the evaluation can be concluded. If water quality is not
suitable, either desires can be modified or restoration
techniques can be considered. If aeration or circulation
techniques are appropriate the next step is to select and
design a system.
-------
AERATION/CIRCULATION
Benefits
The applicability of aeration/circulation techniques
to improvement of water quality for domestic and industrial
uses, downstream utilization and fisheries management is
briefly discussed below.
Domestic Water Supply --
Aeration can greatly reduce undesirable concentrations
of iron, manganese, hydrogen sulfide, carbon dioxide, ammonia
and other substances associated with anaerobic conditions.
These conditions contribute to offensive tastes and odors,
unsightly water, discolored basins and clothing, clogging
and scaling of pipes, corrosion and other undesirable condi-
tions. These conditions can be partially alleviated by modern
water treatment processes after the water leaves the lake.
However, it may be less costly and better to treat the lake
rather than the water after it is withdrawn. Destratification
or hypolimnetic aeration can be used to treat these conditions,
However, destratification will also eliminate the cold water
which could be maintained with hypolimnetic aeration.
Downstream Water Releases —
Many reservoir tailwaters contain valuable cold water
fisheries. These fisheries generally require water tempera-
tures of less than 22°.C (72°F) and oxygen concentrations of
5 mg/1 or more. The oxygen condition can be achieved through
destratification of the reservoir, but the temperature limit
may then be exceeded. Alternative solutions include either
aeration/oxygenation of the hypolimnion, or pure oxygen injec-
tion into the discharge only. The latter approach may be
less expensive than hypolimnetic aeration, but it does not
improve conditions within the reservoir. In some cases,
it may not be permissible if the discharge waters are
"highly" anaerobic. During anaerobic conditions, the waters
may still be toxic due to carbon dioxide, hydrogen sulfide
or other substances even though the oxygen concentration is
apparently adequate (Irwin, Symons and Robeck, 1966). Pure
oxygen injection does not result in significant stripping
of unwanted gases from the water. Air injection does strip
much of these gases, but it may also supersaturate the water
with nitrogen gas (N2). The water is normally 100% satur-
ated with nitrogen initially; and nitrogen gas concen-
tration of only 115% saturation can be toxic to fish (Rucker,
1972).
Industrial Uses —
Some industrial processes require cold noncorrosive and
non-scaling water for cooling and other purposes. Hypolim-
netic aeration/oxygenation can be used to maintain such water
quality.
-------
Fisheries Management --
Thermal stratification and its associated hypolimnetic
oxygen depletion in eutrophic lakes is widely known to re-
strict fish and other biota to shallow depths (Dendy, 1945;
Bardach, 1955; Ziebell, 1969) . In some cases the fish may
be further compressed by warm water above into a narrow band
in the thermocline (Hile, 1936; Gebhart and Summerfelt, 1975)
Coldwater fish such as trout and salmon are often eliminated
from lakes in this manner by warm water above and anaerobic
water below. At some time during the summer there may be no
place in the lake with suitable living conditions for cold-
water fish.
Even if thermal and chemical stratification are not
lethal to the fish, they can severely stress the population.
Johnson (1966) attributed low survival of silver salmon
(Oncorhynchus risutch) in Erdman Lake, Washington to this
condition. Silver salmon survival increased 500% during
artificial destratification. Mayhew (1963) observed reduced
growth and catch rates of several warmwater fish species
in Red Haw Lake, Iowa when these fish were compressed into
the upper third of the lake by stratification and stagnation.
Artificial destratification is one means of reducing or
eliminating oxygen and temperature barriers to fish distribu-
tions. However, destratification is not always successful.
In Lake Roberts, New Mexico, destratification eliminated
thermal stratification but it also caused oxygen concentra-
tions to approach zero throughout the entire lake and killed
many fish as a consequence (McNall, 1971). In El Capitan
Reservoir, California, partial destratification during 1965
greatly reduced thermal stratification but oxygen concentra-
tions were still low in parts of the lake. Consequently,
warmwater fishes did not make much use of these areas and
were largely confined to areas with more than 3 nig/l oxygen
(Miller and Fast, in prep.). Gebhardt and Summerfelt (1975)
observed similar conditions in Lake of the Arbuckles, Okla-
homa, during artificial destratification.
Destratification when properly used can eliminate both
thermal and chemical barriers to fish distributions and there-
by greatly expand available habitat. However, destratifica-
tion may also eliminate the cold water required by coldwater
fishes. When a lake is thoroughly and continuously destrati-
fied, the entire lake's temperature will approach the temper-
ature of the surface waters before aeration began (Fast,
1968; Fast and St. Ament, 1971) . If the surface temperatures
were normally too warm for trout, then the entire lake may
be too warm during destratification.
-------
Hypolimnetic aeration/oxygenation if properly used will
create both suitable oxygen and temperature conditions for
coldwater fish. Fast (1971) first demonstrated the efficacy
of this management technique in Hemlock Lake, Michigan.
Before hypolimnetic aeration, the rainbow trout (Salmo gairdneri)
were confined to a narrow band within the thermocline by warm
epilimnetic waters above and anaerobic hypolimnetic waters
below. During hypolimnetic aeration, the trout distributed
throughout the hypolimnion and thermocline. Fast (1973a,
1973b) later created suitable yearlong trout habitat in Lake
Waccabuc, New York, using a different hypolimnetic aerator
(Fast, et aJL. , 1975a) .
In addition to creating suitable yearlong habitat for
trout, hypolimnetic aeration may allow trout stocking at a
much smaller size than usual. Many California lakes for
example have a winter trout fishery where "catchable" size
trout are stocked during the cooler months only (Butler and
Borgeson, 1965). Since these fish generally do not survive
through the summer they do not sustain much growth in the
receiving water. They typically are stocked at 8 to 9 inches
in length. With hypolimnetic aeration and yearlong survivalt
it may be possible and economical to stock fingerlings or
small trout and allow them to reach catchable size in the
lake. The lake, rather than the fish manager, would feed the
fish.
Algal Production --
Algal biomass, species composition, and rates of pro-
duction can \all be affected by artificial aeration/circula-
tion techniques. Changes in available light for photosynthe-
sis as well as pH and nutrient status may be responsible
for observed changes. These factors are discussed in detail
in Section II.
Application
The remainder of this guide discusses procedures to
determine if aeration/circulation techniques are appropriate
and design considerations related to implementation. The
broad range of possibilities included with aeration/circu-
lation techniques has been divided into two major groups:
Destratification and hypolimnion aeration.
Procedures which are designed to either mix the lake
or provide aeration without maintaining the normal thermal
structure are included in the "destratification" category.
Within this category systems may range from high energy
mixing devices to low energy aeration procedures. Both
mechanical pumps and compressed air can be used as mixxng
devices. Destratification systems can be used to control
excessive algal growth under certain circumstances and can
-------
maintain aerobic conditions. However, cold water cannot be
maintained when complete mixing is achieved.
Those systems which are designed to maintain the normal
thermal structure of a lake while adding oxygen are included
under the heading "hypolimnion aeration". Both air and oxy-
gen have been used in such systems. Hypolimnion aeration
can be an effective means to maintain aerobic conditions
without losing cold hypolimnetic water which may be necessary
for domestic or industrial use and is required for the main-
tenance of cold water fisheries.
Problem Definition and Selection of Goals --
Aeration/circulation techniques can be used to maintain
desired oxygen levels and, under certain circumstances, to
control algal blooms. Prior to deciding on a lake restora-
tion technique it is necessary to identify and quantify the
problem. Several surveys should be made to determine oxygen
and temperature profiles as well as algal species and abun-
dance. It is recommended that at least three surveys be
conducted during summer months to quantify the extent of
any problems. Samples should be collected from a sufficient
number of stations and depth intervals to determine spatial
variation (initially, approximately one station per 100 acres
(40 hectares) at 3 foot (-1 meter) depth intervals. Sampling
stations should include the deepest portions of the lake. Tem-
perature and dissolved oxygen should be measured in the field.
A good quality Temperature-Dissolved Oxygen Analyzer is of suffi-
cient accuracy for problem identification. Water samples should
be taken to a competent laboratory for phytoplankton analysis.
Only predominant species and total biomass as measured by chloro-
phyll a or ash free dry weight need be determined.
System Selection —
This manual addresses two possible management techniques:
destratification and hypolimnion aeration. Each technique
is appropriate for certain circumstances. Destratification
can maintain aerobic conditions and control algal blooms.
However, when a lake is destratifled the net heat input is
increased significantly and the cold water formerly contained
in the hypolimnion is heated. In general, a destratified
lake will have a fairly uniform temperature from top to
bottom which is slightly (.1-3°F) lower than the surface
temperature under stratified conditions. Hypolimnion aera-
tion can maintain cold, aerobic bottom water but does not
control algal blooms.
If it is determined that excessive algal growth is
not a problem but low oxygen levels are, then it must be
determined if cold hypolimnion temperatures are needed.
If low temperatures are required then hypolimnion aeration
should be considered. However, the cause of low oxygen
-------
should first, be determined to ascertain if the problem
could be solved by eliminating the source.
If algal growth is a problem and cold water is not
required, a high energy mixing system may be beneficial
but further analysis will be required to determine the
probable results. A high energy mixing system sufficient
to control algal growth will normally maintain high oxygen
levels.
If algal growth is a problem and cold water is required,
an aeration/circulation system will not solve the entire
problem. However, aerobic conditions can be maintained
in the hypolimnion without disturbing the temperature pro-
file by application of a hypolimnion aeration technique.
-------
SECTION II
DESTRATIFICATION
BACKGROUND
Destratification or mixing of lakes has been used for
two general purposes:
1. Aeration and
2. Algal Bloom Control
The need for aeration results from the fact that major
oxygen consuming processes occur in the hypolimnion of lakes,
Because very little oxygen is transported downward across
the thermocline during periods of stratification, the avail-
able oxygen can be depleted. Such anoxic water may then
require aeration for a variety of beneficial uses.
Water supply reservoirs may experience severe problems
associated with anoxic conditions such as high concentra-
tions of iron and manganese and odors associated with re-
duced compounds such as hydrogen sulfide. Fishery main-
tenance may require oxygen additions. High oxygen concen-
trations may be required for reservoir releases to maintain
desirable biological communities downstream.
It is possible to provide the required oxygen in situ
by artificially mixing the lake. It has been shown (Smith
et al., 1975) that the primary mechanism of oxygen transfer
is at the water surface even if compressed air is used as
a mixing device. Riddick (1957) concluded that "...an
aerator should be regarded as a cheap, uncomplicated and
relatively efficient device for pumping water". However,
in deep lakes gas transfer between the rising bubbles and
surrounding water may be important.
Destratification to control algal blooms is only appro-
priate under certain circumstances and requires high energy
systems sufficient to redistribute algal cells throughout
the water column. It is believed that mixing can control
algal blooms by limiting the amount of solar energy avail-
able for photosynthesis. Theoretical considerations in-
dicate that the depth available for mixing is a controlling
factor in limiting algal biomass by mixing (Lorenzen and
Mitchell, 1973, 1975). It has also been suggested that pH
changes induced by mixing can have a significant effect on
algal species dominance. Shapiro (1973) feels "that the
rate of supply of C0~ is an important factor in regulating
-------
the qualitative nature of the phytoplankton". Blue-green
algae may have a competitive advantage at high pH. It
is therefore possible that lowering the pH as a result of
mixing could favor non blue-green algal species over blue-
green species.
Mixing may also favor non blue-green algae by elimina-
ting the competitive advantage of blue-green algae which
are often buoyant and thus able to maintain themselves
at optimum light levels under quiescent conditions. The
long term effects of maintaining oxic conditions on nutrient
cycling have not been well established. However, the net
effect should be a reduction in sediment phosphorus release.
The following sections of this chapter provide a de-
scription of past experience and methods for destratifica-
tion followed by design considerations which include defi-
nition of goals, data requirements, procedures to estimate
results, and selection of methods. The chapter concludes
with a summary of potential problems.
EXPERIENCE AND METHODS
Dunst, e_t al. , (1974) list 123 individual lakes where
aeration and/or circulation techniques have been employed
as all or part of lake rehabilitation projects. There are
undoubtedly many more such projects which were not uncovered
in their survey. In 1971 the American Water Works Associa-
tion reported on "Artificial Destratification of Reservoirs",
Of 37 water utility managers, 86% considered their projects
a success. The great majority of destratification systems
have used air diffusers although several have used mechani-
cal devices.
Table II-l summarizes some of the pertinent character-
istics of selected previous destratification systems. All
of these systems used compressed air as a mixing device.
There is a wide range in amount of air supplied per unit
volume or unit area. The ratio of air release rate to
volume (Qa/V) is shown in standard cubic feet per minute
(SCFM) of air per million cubic feet of lake volume. The
ratio of air release rate to surface area (Qa/A) is shown
in SCFM of air to million square feet of surface area.
Those systems that provided good mixing generally used more
than one standard cubic foot per minute (SCFM) of air per
million cubic feet of lake volume, and more than 20 SCFM
per million square feet of surface area. Lake morphometry
in conjunction with weather conditions and wind exposure
will also effect the required air flow to induce mixing.
10
-------
TABLE II-l CHARACTERISTICS OF SELECTED PREVIOUS
DESTRATIFICATION SYSTEMS
Lake
dines Pond
University Lake
Eufaula Reservoir
Kezar Lake
Lake Roberts
El Capitan
Test Lake 1
Test Lake 2
Section Four Lake
Ottoville Quarry
Casitas Reservoir
Buchanan Lake
Valen Lake
Parvin Lake
Lake Wohlford
Cox Hollow
Wahnbach Reservoir
Hyrum Reservoir
Al latoona
Location Air Mean- Volume Area Power
RDeptahe D?fh 1Q6 ft3 1()6ft2 (HP)
ft.
Corvalis, Oregon 15 6.4 0.098 0.0153 1/4
North Carolina 30 10.5 91.5 8.714 3
Oklahoma 92.5 53 24,829 468
New Hampshire 25 10 76.2 7.62
New Mexico 30 14.3 42.7 2.986
California 1970 70 32 635.4 19.8 50
1971 93 31 743.4 23.9 50
Boltz, Kentucky 62 30 125.4 4.18
Falmouth, Kentucky 42 20 196 9.80
Michigan 60 32 3.9 0.121 20
Ohio 1970 55 29 2.2 0.076 3/4
1971 55 29 2.2 0.076 2
California 1972 140-160 85 8,790 103.4 150
1973 140-160 91 10,413 114.4 150
Ontario, Canada 44 16.2 15 0.926 2
Ontario, Canada 15 6.5 42.4 6.52 2
Colorado 32 14.4 30 2.08
California 45 19.2 109 5.66 50
Wisconsin 29 12.5 52.3 4.18 7.5
W. Germany 1961-62 140 63 1,470 23.08 15
1964 140 63 1,470 23.08 50
Utah 70 35 585 19 25
Georgia 140 31 16,170 522 300
Q . Q Q AT°C Comment Reference
xlO6 xlO6
1 10 65.3 ^l0 good mixing Maleug, et al . , 1973
14 0.15 1.61 1-2° 02 depletion Weiss, Breedlove, 1973
1200 0.05 2.56 9° Central J0ol 125 ^60^1 0/70^ '"
105 1.3 13.8 <1 15 psig - good mixing NHWSPCC , 197T
100 2.3 33.5 <1° EPA> 1970
215 0.34 10.9 2.5° Did not mix algae Fast, 1968
215 0.29 9.0 2.5°
112 0.89 26.8 Sym°ns*' ^.i!- 19b/
115 0.59 Ili7 Symons, eta]_., 1967
78 20 645 0° Compressor run only 8 Fast, 1971
hours a day
32 14.5 422 1° Non-continuous Gartman, 1973
78 35 1028 2 compressor operation
1 /
630 0.07 6.09 4°T/ Air injected about 80- Barnett (1971, 1975)
630 0.06 5.51 7° 100 feet off the bottom
10 0.67 10.8 10° Brown, el al. , 1971
10 0.24 1.53 o.5° MacBeth, et al_. , 1973
75 2.5 36 3° Helixor R Lackey (1972)
210 1.9 37.1
-------
Mechanical Mixing
Mechanical mixing of lakes and reservoirs has been used
to a lesser extent than diffused air mixing. However, several
systems have been used and may be appropriate under certain
circumstances.
The Metropolitan Water Board of London routinely uses
jet discharges to control stratification. Their reservoirs
are supplied by pumping river water so that modifications
could be made to the inlet system with little added pumping
cost. In one 500 acre (22 million square feet) 70 foot deep
reservoir the incoming water can be pumped into the reservoir
through a series of six jets placed a few feet above the
reservoir floor. ' The jet orifices are 36 inches in diameter
and discharge water at approximately 10 feet/second. Some
of the inlets are horizontal while others are inclined at
angles of 22-1/2 and 45° to the horizontal. The direction
of the jets relative to the peripheral embankment of the
reservoir was determined by extensive trials with scale
models (Ridley and Symons, 1972).
Symons (Ridley and Symons, 1972) used a 12 inch, mixed
flow pump driven by a gasoline engine to destratify Vesu-
vius Lake in Ohio and Boltz Lake in Kentucky. These lakes
were destratified by pumping water from the bottom to the
surface. Boltz Lake was effectively destratified by pump-
ing at a rate of 6.4 cubic feet per second. The lake volume
was approximately 1.3 million cubic feet.
Carton and Jarel (1975) have experimented with mechani-
cal destratification in Ham's Lake and Lake of the Arbuckles,
Oklahoma. A 42 inch diameter fan blade, a 1/2 horsepower
electric motor and belt driven reduction gear successfully
destratified Ham's Lake in one week. Ham's Lake has a
surface area of 4.4 million square feet and volume of 44
million cubic feet. Water was pumped from the surface
downward.
A much larger device consisting of a 16.5 foot Curtiss-
Wright propeller with a 40 horsepower Ford industrial motor
pumped approximately 670 cubic feet per second down from the
surface of Lake of the Arbuckles which has a surface area of
approximately 103 million square feet and volume of 3185 mil-
lion cubic feet at conservation storage level. The system did
not completely destratify the lake but reduced the maximum
temperature differential from 13°C to 4.5 C in 44 days.
12
-------
Diffused Air Mixing
Release of compressed air at the bottom of lakes and
impoundments has been found to be a fairly efficient mixing
technique. Ease of installation and simplicity of operation
add to the advantage of air as opposed to mechanical devices.
A typical installation, described by Fast (1968) is shown
in Figure II-l. This system was installed at El Capitan
Reservoir in California. The system uses a shore installed
air compressor (LeRoi 50-S-2) which was rated at 215 cubic
feet per minute, driven by a 50 horsepower electric motor.
A 1-1/2 inch nominal size galvanized steel pipe transports
air from the compressor to the reservoir. From this point,
a 300 foot length of 1-1/2 inch PVC plastic pipe extends
along the bottom. The plastic pipe is weighted by 15 con-
crete block anchors. The last 100 feet of the plastic pipe
are suspended almost horizontally above the bottom by
13 styrofoam floats and lengths of polyethylene anchor rope.
Thirteen sets of floats with anchors are evenly spaced along
the 100 feet of plastic pipe. This section is perforated
by 90 holes, 1/8 inch in diameter, and sealed at its distal
end. Clusters of three holes, spaced 120 degrees apart
around the circumference of the pipe, are located on this
section of pipe. The clusters are unevenly spaced. Be-
ginning 100 feet from the end of the pipe, the spacing
between the first six clusters is 5 feet, clusters 6 through
12 are 4 feet apart, clusters 12 through 21 are 3 feet
apart and clusters 21 through 30 are 2 feet apart. This
non-linear arrangement of air holes was intended to produce
a uniform air release over the length of the pipe.
DESIGN CONSIDERATIONS
Aeration
For those situations when aeration is needed, algal
bloom control is not necessary, and cold water is not re-
quired, a circulation device to eliminate the thermocline
can be used to provide sufficient surface aeration. Al-
though any specific lake may provide special problems, in
general, if there are no unusual oxygen consuming processes
it is believed that if the temperature profile can be main-
tained such that there are no sharp gradients and the max-
imum temperature difference between top and bottom is less
than 2°C, sufficient mixing will have been imposed to main-
tain aerobic conditions. Air requirements to provide suf-
ficient mixing are discussed under system selection.
13
-------
Figure II-l. Destratification system installed at El Capitan
Reservoir, California.
-------
Algal Bloom Control
Prior to designing a destratification system to control
algal production it is important to determine the suitability
of the impoundment. The procedure outlined below is taken
from Lorenzen and Mitchell (1975) and is designed to provide
a determination of the maximum algal biomass that could de-
velop under both stratified and completely destratified con-
ditions .
The evaluation procedure is based on a mathematical model
of algal production and respiration. The model considered
both nutrient depletion and light limitation as potential
biomass limiting factors. The two mechanisms were evaluated
independently and then combined to determine the upper limit
to biomass production as a function of mixed depth. Nutrient
limitation was considered as the capacity of the system
to produce biomass prior to some essential nutrient(s) being
exhausted. The total nutrient limited biomass that could
be produced in a water column was shown to be
C x Z = X x Z (II-l)
where C = algal concentration, mg/1
Z = depth of uniform algal distribution, m
X = capacity of system to produce algal biomass
before nutrient(s) is (are) depleted, mg/1
Biomass limitation by available light was determined by
evaluating the balance between photosynthesis and respiration
in a water column. By considering light attenuation to be
a function of depth and concentration of algal cells, it was
shown that at some maximum algal concentration, light would
be sufficiently diminished so that total photosynthesis would
equal total respiration and no further net production would
take place. This maximum algal concentration, multiplied by
the depth of uniform distribution (mixed depth) is the peak
light limited algal biomass (grams per square meter).
It was shown that peak light limited algal biomass in the
water column as a function of mixed depth could be repre-
sented by:
V(k/ ln^o(t) +U+ ["o't"^;-'-/
(ii-:
,
j -
15
-------
-1
where K = maximum specific algal growth rate, day
IllGL,X.
R = specific rate of respiration, day
3 = incremental light attenuation coefficient
caused by algal cells, m /mg/2,
AT = interval of integration
A = measure of alga's adaption to low light
levels [ 1 ]
I (t) = surface illumination as a function of
0 time [1]
a = attenuation coefficient of light in the
water, base e, m
n J Selection of appropriate units and devices for measuring
radiation available for photosynthesis is a complex and
difficult problem. Measures of illumination (foot-candle;
lux) have been commonly used. However, it has been shown
(Tyler, 1973) that considerable errors in estimated
available energy can result from illumination measure-
ments. The most appropriate measurement is the total
quanta within wavelength limits of 350-700 nanometers
in watts/cm^ (Booth, 1976). Unfortunately, past re-
search and easily available instrumentation do not
conform to this. The following table (from Westlake,
1965) provides some conversion factors so that the
best use can be made of existing work.
Active Light 390-710 nm
,2 2 2
Joule/m /sec ergs/cm /sec g cal/cm /min lux
Joule/m -sec 1 103 1.43 x 10~3 '^2.5 x 1Q2
2 3 _o 9
watt/m 1 10 1.43 x 10. 2.5 x 10
!la$Sfc 6.98X102 6.98 X105 1 1. 8 x 1C5
'candle^ **^ ^° 5.70 xlO'6
L lux = 0.0929 foot candles
-------
If CZ is plotted against Z, the light limited biomass is
represented by a line having an intercept of
AT
K f
max I , ,., / , 9 l/2\
PRAT / v <-\\ ' i I + IAI (t) } )dt CTT i\
r\pt-i I ,/o 0 nv ' -1 ' ^J-J- —o/
and a slope of -(a/3).
Figure II-2 shows a typical plot of both nutrient limited
and light limited peak algal biomass in a water column
(C x Z) as functions of mixed depth, Z.
To evaluate the suitability of a lake for artificial
destratification, a diagram such as Figure II-2 can be
constructed. The approximate peak biomasses in the strat-
ified and destratified cases can then be determined from
the thermocline depth (stratified case) and average lake
depth (destratified case). It should be noted that the
analysis for the destratified case assumes that sufficient
mixing takes place to maintain uniform vertical algal pro-
files .
To construct a diagram such as Figure II-2, the values
of the individual parameters must be determined. Although
various levels of sophistication can be used in these determi-
nations, a simple evaluation should suffice for planning
purposes. The following outline summarizes a procedure for
evaluating the needed parameter values and constructing the
diagram.
This procedure has been applied by Lorenzen and Mitchell
(1975) to Kezar Lake in New Hampshire. The observed results
were consistent with theoretical predictions.
1. Determine the value of X, the nutrient limited algal
concentration, mg/£.
a. Bioassay; Incubation of lake samples under
approximately 5000 lux for three to four weeks should
give an adequate representation of maximum nutrient
limited biomass. Replicate samples should be run and
slight agitation provided. Biomass measurements should
be made approximately every three days until no further
increase is observed.
b. Historical records: Observed peak biomass
levels could be used to estimate the value of X.
However, caution should be exercised to ensure that
the observed levels were limited by nutrient depletion.
17
-------
CNJ
O
M
O
orT
oo
<
S
O
QQ
<
LU
Q_
jlrAl0(t)
LIGHT LIMITED ALGAL BIOMASS
NUTRIENT LI Ml TED
ALGAL BIOMASS
MIXED DEPTH, z (METERS)
Figure 11-2.
Generalized plot of peak algal biomass as a function of mixed
depth for both nutrient and light limitations.
-------
2. Determine algal growth rate parameters, K , A and
Rmax
.
a. Laboratory studies: Ideally, algal growth rates,
as a function of light intensity in the presence of excess
nutrients, should be determined in the laboratory- The
maximum specific growth rate K , the rate of respiration
R, and the adaption to low light A can then be determined
from a plot such as Figure II-3.
b. Literature values : There are extensive data in
the literature relating algal growth rates to light in-
tensity for various algal species.
3. Determine illumination values, I (t) .
o
a. Field measurement : Values of illumination as a
function of time can be measured directly or obtained
from meteorological records.
b. Standard light day : For planning purposes a
standard light day as suggested by Vollenweider (1965)
may be used. For a standard 14-hr day with a maximum
noon intensity of 100,000 lux and A = 0.0005 lux" , the
term
AT
AT/ "" v"V' ' L' ' niov"' J 'uu (H-4)
1 / -in (AIQ(t) + [1 + AIQ(t)2]1/2)dt
has a value of approximately 2.4.
4. Determine the incremental attenuation, coefficient, 3.
a. Laboratory studies : The value of 3 can be
determined by measuring total light attenuation coef-
ficients with various concentrations of algal cells.
A plot of total attenuation coefficient versus algal
concentration, yields a line with slope equal to 3.
b. Literature values : Values of 3 from 0.17 to
0.20 m~ / mg/& dry weight have been used (Lorenzen and
Mitchell, 1975; Chen and Orlob, 1975).
5. Determine the light attenuation coefficient, a.
a. Field measurement: The value of a can vary
widely and should be determined for eacn lake. An
underwater photometer can be used to measure light
penetration as a function of depth. The total
19
-------
ro
o
o
o
o
LU
Q_
on
A -- 0.00045 LUX'1
I
I
2000
4000 6000 8000
ILLUMINATION, LUX
10,000 100,000
Figure II-3. Net Specific growth rates of Aphanizomenon (from Lorenzen and Mitchell, 1975)
-------
attenuation coefficient is given by (a + 3C) in the
expression:
Zd = Io exP[~(a + 3C)dJ (II-5)
where: Id = illumination at depth d, lux
I = surface illumination, lux
d = depth, m
a, 3, and C are as previously defined.
6. Construct peak biomass diagram.
a. Nutrient limited biomass: Plot a line through
the origin with slope X.
b. Light limited biomass:
(1) Evaluate intercept using the estimated
parameter values.
AT
K C
Intercept- ~^Jo ^(Al^t) + [1 MAIQ(t))2)1/2 dt (II_6)
Evaluate slope = a/6
If this procedure indicates that an acceptable degree of
algal bloom control can be achieved by mixing then it is
appropriate to select a mixing system.
System Selection
If it is determined that artificial mixing can achieve
the desired water quality then a system must be designed to
impose the required circulation. There are basically two
types of devices to provide induced water flow in impounded
waters. Mechanical systems can pump water from bottom to
top or top to bottom. Air diffusers have also been success-
fully used to provide vertical mixing.
Because air diffusers are by far the most commonly used
mixing device, considerable effort was devoted to analysis
of the relationships between air flow and water flow as well
as the effects of various mixing levels on resulting temper-
ature and algal profiles.
21
-------
Induced Circulation --
There has been little research on the relationships be-
tween air release rates and water flow rates. However, the
best available theory (Kobus, 1972) indicates that the amount
of water flow induced by a rising bubble plume is primarily
a function of air release depth and air flow rate. Kobus
has shown theoretically (and to some extent experimentally)
that the water flow as a function of height above an orifice
is given by:
- 35.6C(xt0.8)V'V"(1' h *10'3 tu-7)
where': Q (x) = water flow m /sec
x = height above orifice
C = 2V + 0.05
o
V = air flow m /sec at 1 atm.
h = depth of orifice
y, = 25V +0.7 m/sec
b o
This theory was used to prepare Figure II-4 which shows
the total flow of upwelled water as a function of air re-
lease rate for different depths of air release. These
results illustrate that one of the advantages of a rising
bubble plume is that it continues to entrain water and
induce mixing all along its path.
Effects of Circulation --
The relationships established between air release and
water flow were used to compute the effects of mixing on
temperature and algal distributions for a range of lake
sizes. The procedures used and results obtained are de-
scribed in detail in Appendix A.
Table II-'2 summarizes the lake characteristics and air
flow required to provide mixing that results in small
variations in vertical algal concentrations and temperatures.
In general, the results indicate that approximately 30 SCFM
of air per 106 square feet of surface area would be required
to achieve good mixing.
22
-------
ro
oo
"b
Q
IT
LJ
I
LJ
O
cc
=>
en
240
480 720
AIR FLOW (scfm)
960
1200
Figure II-4.
Total flow of upwelled water as a function of air
release rate and depth of release.
-------
-5
n>
o
o
<-*•
rs
CD
D.
1
L.
OL
U
UJ
O
o:
^
CO
2400
4800 7200
AIR FLOW (scfm)
9600
12000
-------
TABLE II-2
APPROXIMATE AIR FLOW TO MIX VARIOUS LAKES
Lake Avg. Depth Max. Depth Surface9Area Volume Air Required
(ft) (ft) (ft/) (ft3) (SCFM)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
9.7
9.7
9.7
9.7
26
26
26
26
50
50
50
50
93
93
21
21
21
21
51
51
51
51
100
100
100
100
200
200
7.47 x 106
1.49 x 107
2.99 x 107
7.47 x 107
1.71 x 107
3.42 x 107
8.55 x 107
2.14 x 108
4.26 x 107
1.07 x 108
2.66 x 108
1.07 x 109
2.07 x 108
8.28 x 108
7.23 x 107
1.45 x 108
2.89 x 108
7.22 x 108
4.41 x 108
8.80 x 108
2.21 x 109
5.52 x 109
2.10 x 109
5.24 x 109
1.31 x 1010
5.24 x 1010
1.92 x 1010
7.66 x 1010
250
600
1200
2400
600
1200
2400
5000
1200
2400
4800
>12000
5000
10000
-------
The results of these simulations should be used as
general guidance and interpreted in view of site specific
conditions compared to the cases presented here. There are
an infinite number of possible conditions that could be
evaluated. Many factors such as dispersion coefficients,
weather conditions and lake morphometry are extremely site
specific and should be evaluated on a site by site basis.
The examples presented cover a range of possible lake
sizes but the same weather conditions were used for all lakes
and vertical dispersion coefficients were selected from the
literature.
jDiffuser Design
Air diffusers can be selected from numerous possibilities
ranging from patented systems to simple holes drilled in
polyethylene pipe. Patented systems could not be evaluated
due to proprietary rights. However, the procedures to
design a simple system are given so that the user can com-
pare quotations and recommendations of various suppliers.
Once the total amount of air needed has been determined,
it must be decided how it should be distributed. This de-
cision is still very subjective and depends on local con-
ditions .
The following general comments can be made as a result
of past observations:
1) Induced mixing will normally only occur above the
level of air release.
2) If the aerator is located in the deepest part of a
lake, cooler water will flow toward it.
3) Because the purpose of the air release system is to
promote circulation it should be located to minimize
flow restrictions to a rising water column. For
example, it would be preferable to locate a diffuse^
perpendicular rather than parallel to a dam face.
4) Air release rates have ranged from 0.02 to 2 SCFM
of air per lineal foot of diffuser.
The design and layout of a diffuser system to promote
vertical circulation requires the specification of a number
of components including:
« Length of diffuser
• Configuration of diffuser
26
-------
• Pipe size and material
• Orifice size and spacing
The length and configuration of the diffuser system is
the most subjective. The primary goal is to induce circu-
lation so the air should be dispersed as widely as possible
As a general guide, for low exit velocities, the rising
plume will spread horizontally at a rate of 0.05 feet per
foot rise. Orifice spacing should therefore be at least
0.1 times the depth of air release. The total length of
diffuser is then determined by dividing the total air
required by the flow through each orifice.
The flow through an orifice can be computed from the
relation (CRANE, Technical Paper 410, 1973):
to - 0.525 d CAAPp (H-8)
where: w = weight flow rate, Ibs/sec
d = orifice diameter, inches
C = discharge coefficient, =0.65
AP = pressure drop across orifice, psi
p = weight density, Ibs/ft
As an example, the following flow rates were computed
for air at 70°F at a range of overpressures (AP, pressure
in diffuser - hydrostatic pressure) and operating pressures
(P, pressure in diffuser) for a 1/8 inch diamter orifice.
Air Flow Rate in SCFM
P psi
AP, psi
5
10
15
30
7.0
9.6
12.0
50
8. 4
12.0
15.0
100
11
16
19
27
-------
A uniform air flow along the length of the diffuser line
can be achieved when the pressure change is small compared
to pressure in the line. However, when pressure loss in the
diffuser is significant the hole spacing should be decreased
along the length of the diffuser. The following example
illustrates the computation for a 2 inch diameter line, 500
feet long, discharging 500 SCFM of air in a 100 foot deep
lake.
Inlet Pressure, P = 60 psi
P =
Pressure at end of line, P (Van der Hegge Zijnen, 1951)
2
3o +^f [°-9 - °'158 Reo"V4 IT §]( (H-9)
where: V = inlet velocity, ft/sec
Reo = Reynolds Number at inlet, VQ D/y
L = line length, ft
D = line diameter, ft
Y = density of air, Ib/ft
v = kinematic viscosity, ft /sec
|(500 SCFM) ^|yV ^(2/12) 2] ~ = 93 ft/sec
4 (11-10)
v2
^ = 136 feet
2g
V D
R - -5- = 600,000
eo v
R ~1/4 = 0.036
eo
L/D = 500/(2/12) = 3000
Y60 psi = °'381
V =
o
28
-------
P = (60) (144) + (136)(0.9-12.4) (.381) = 8640-596
= 8044 lb/ft2 - 56 psi
The internal pressure at the last orifice would be
approximately 56 psi. The ratio of air flow in the last
orifice to the first orifice can be computed from equation
II-8.
Air flow last hole (8) (0.34) 1. 65
J
Air flow first hole M.12)('0.38) 2.14
Uniformly spaced orifices would be adequate.
Compressor and Supply Line
Compressor selection and design of the air supply line
are applicable to both hypolimnion and total aeration. The
procedures to select compressor size and compute pressure
losses are given in Appendix C.
Potential Problems
The most significant potential problem is inadequate
performance resulting from underdesign. Anticipated re-
sults are based on achieving adequate mixing. If mixing
is not adequate then goals will not be achieved.
Caution should be exercised when beginning to destrati
fy a lake which is already stratified and low in oxygen.
A rapid turnover could cause the entire lake to become
anaerobic .
29
-------
SECTION III
HYPOLIMNETIC AERATION
BACKGROUND
The basic purpose of all hypolimnetic aeration/oxygena-
tion systems is to add oxygen to hypolimnetic_or bottom
waters without destroying the thermal stratification. Hypo-
limnetic aeration has been used in ice covered lakes where
it is desirable to aerate the water but not to create open
water conditions and in lakes which are stratified during
the summer where the objective is to preserve water which
is both aerated and cold. There are a variety of reasons
for these objectives and a number of ways of achieving
hypolimnetic aeration.
Hypolimnetic aeration was first used by Mercier and Per-
ret (1949) to aerate the hypolimnion of Lake Bret, Switzer-
land without disturbing the lake's thermal stratification.
Their system mechanically pumped water from the hypolimnion,
discharged the water into a splash basin on shore (where it
was aerated) and allowed the water to return by gravity
flow back into the hypolimnion. The purpose of their sys-
tem was to improve domestic water quality.
Most of the early hypolimnetic aerator development
occurred in Europe. By 1974, there were 11 hypolimnetic
aerator installations in Europe: Sweden (4) , Germany (2) ,
Finland (2), Switzerland (1), Norway (1), and Italy (1).
Interest was subsequently awakened in the United States,
and the first installation occurred in Hemlock Lake,
Michigan, during 1970 (Fast, 1971). Since then at least
8 hypolimnetic aeration/oxygenation systems have been
installed in the United States. These include air injection
systems and systems which use liquid (or pure) oxygen.
Following Lake Bret, probably the next most significant
breakthrough in hypolimnetic aeration was a system designed
by Bernhardt (1967) . His system was first used in 1966 to
aerate the Wahnbach Reservoir's (West Germany) hypolimnion
for domestic and industrial water quality control. Bern-
hardt has since designed two other aerators for the Wahn-
bach Reservoir. His most recent design is one of the
most efficient in terms of energy required per mass of
oxygen dissolved and in terms of operating expense.
30
-------
Beginning in 1969, Bengtsson, et al. , (1972) in Sweden
and Fast (1971) in the United States designed and tested
several hypolimnetic aerators. These all use air injection
with either a full air lift to the lake's surface, or a par-
tial air lift within the hypolimnion. More recently a hypo-
limnetic oxygenator was tested in Ottoville Quarry, Ohio,
and Attica Reservoir, New York (Matsch, 1973: Fast, et. al. ,
1975b) . This system uses liquid (or nearly pure)' oxycfen
and a mechanical water pump.
Experience and Methods
A synoptic survey of described and proposed hypolimnetic
aeration/oxygenation systems is presented in Appendix B.
These systems can be categorized as mechanical agitation
systems, pure oxygen injection systems, and air injection
systems. The air injection systems can be further subdivided
into full air life designs, which lift the water to the
surface and then return it to the hypolimnion, partial air
lift designs which aerate the water and discharge to the hypo-
limnion without transport to the surface, and downflow air
injection systems which use mechanical pumping and inject
air to the water returning to the hypolimnion. Thirteen of
the nineteen described systems have apparently been field
tested, although details of the tests are not available
for all systems.
A review of these systems indicates that two basic air
lift designs appear to have general application. These
designs are in essence modified versions of earlier designs
by Fast (1971) and Bernhardt (1974). Suggested improvements
involve construction and materials modifications that should
reduce capital costs, improve the appearance of the aerators,
and simplify assembly and installation.
The full air lift hypolimnetic aerator design by Bern-
hardt (1974) has the highest reported efficiency (Table III-
1). Bernhardt reported a 50 o oxygen absorption rate and an
energy efficiency of 2.4 Ibs 0^/kw-hr. His original design
(Figure III-l) is rather bulky~and may be difficult to install.
Suggested modifications include (Figure III-2): (1) use of corru-
gated pipe and flexible sheeting construction. Both the
upwelling and downwelling tubes may be constructed of plastic
or rubber sheeting, such as commonly used for pond lining.
This material is highly flexible, durable and comes rein-
forced with nylon. Its life expectancy in water is about
20 years. If the upwelling arm is constructed of this
sheeting, it must be fit over a rigid framework, otherwise
it will collapse during air injection and water flow. The
dcwnwellinq pipe should not need rigid reinforcement since
the pressure of the outflowing water should maintain its
form; i21 use of half-round corrugated pipe for the
31
-------
TABLE III-l OBSERVED AND CALCULATED OXYGEN ABSORPTION AND ENERGY EFFICIENCY FOR
SEVERAL HYPOLIMNETIC AERATION/OXYGENATION SYSTEMS
CO
ro
Hypo limnetic
Aerator
Installation
A. Full Air Lift
Wahnbach I
Wahnbach II
Mirror Lake
Larson Lake
Tullingesjon
Jarlasjon
B. Partial Air Lift
Lake Waccabuc
C. Oxygen Injection
Ottoville Quarry
Attica Reservoir
Spruce Run 1973
1974
Estimated Ibs.
oxygen dissolved
per kw-hr.
2.1
2.4
0.7
0.7
0.5
0.7
0.4
1.2
0.5
0.4
0.8-1.0
0.4-0.8
Percent
Oxygen
Absorption
50
50
9-14
14-23
8.1
10.3
5.3
10.6
>95
>95
30-41
18-30
Influent (to
Aerator) oxygen
concentration
<4
<4
0.0
<7.5
0.1
0.0
5.0
4
>8
<5
<0.5
<4
Source of observation or
data for calculations
Bernhardt (1967)
Bernhardt (1974)
Smith, et a]_., (1975)
Smith, et_ al_. , (1975)
Bengtsson, et al_. , (1972)
Bengtsson, et al_. , (1972)
Bengtsson, et aj,. , (1972)
Fast (1973a,b)
Matsch (1973)
Matsch (1973)
Whipple, .et al-, (1975)
Whipple, et al_. , (1975)
-------
Luff
Damm
CO
OJ
Epilimnion
Metalimnion
Hypolimnion
a^ufsteigendes
Luff - Vfassergemisch
Diffusor
Luftzufuhr
~ ^'^^*'~-'
Grund
Zch Nr 922-2 6GO
Figure III-l. Hypolimnion aerator used by Bernhardt (1974) in
the Wahnbach Reservoir.
-------
WOOD DECK
AIR
EPI LIMN ION
HYPOLIMNION
WATER OUT
STEEL
-DEGASSING
CHAMBER
WATER IN
COMPRESSED
AIR
Figure III-2.
Suggested modifications to Bernhardt hypolimm'on
aerator.
34
-------
degassing chamber. This will reduce weight, costs and
make the chamber more aesthetically appealing. If a
9 foot diameter degassing chamber is used and the water
level is at midpoint, then the chamber will emerge from
the water 4-1/2 feet if full round pipe is used but less
than one foot if half-round pipe is used; (3) the half
round degassing chamber can be covered with wooden
planking to further improve its appearance, utility,
and safety. It can be used for a recreation float; and
(4) flotation of the aerator can be easily provided by
extensions of the degassing chamber (Figure III-3). The
end chambers may be filled with polyurethane foam. In
addition, it is advisable to install ballast tanks inside
these end-chambers so that the aerator can be adjusted
vertically or leveled when in operation. Leveling can be
accomplished by pumping water into or out of the respective
ballast tanks.
The hypolimnetic aerator described by Fast (1971) is
also a full air lift design (Figure III-4). Although its
original efficiency was not as great as Bernhardt's, a
modified version (presented here) should be as efficient,
simpler and cheaper to construct, and easier to install.
Proposed modifications include the use of plastic materials
for the upwelling and downwelling pipes. The upwelling
pipe will need reinforcement.
DESIGN CONSIDERATIONS
Certain data are required for sizing any hypolimnetic
aerator to a given lake. The data needed will depend on
the specific lake, but should include at least monthly
oxygen and temperature profiles during one or two seasons,
and a determination of water volume as a function of
depth.
Hypolimnetic Volume Estimates
Water volumes within any depth interval are generally
calculated from a bathymetric map as discussed by Welch
(1948) . Volume estimates for each five foot (or less )
depth interval are desirable. The volumes of respective
depth intervals will be used to calculate oxygen depletion
rates. The total volume of the hypolimnion is also re-
quired and can be estimated from the volume-depth relation-
ships and temperature profile data.
The hypolimnetic volume can be estimated by reviewing
a set of temperature profiles during summer stratification.
These profiles must be interpreted in order to account for
thermocline erosion during aeration. For example, before
aeration began in Ottoville Quarry, the hypolimnion extended
from about 8 m to the bottom (Figure III-5), and contained
35
-------
END PLATE-)
WATER
FOAM FILLER
FILLER TUBE
BALAST TANK
PARTITION
FLOTATION
CHAMBER
DEGASSING
CHAMBER
DOWNWELLING OR
UPWELLING PIPE
X-SECTION THROUGH FLOTATION TANK
FLOTATION CHAMBER
BALAST TANK
END PLATE
DEGASSING
CHAMBER
PARTITION
DOWNWELLING OR
UPWELLING PIPE
Figure III-3. Hypolimm'on aerator degassing chamber details.
36
-------
WOOD DECK
WASTE AIR
STEEL FRAME
W/FLOATS
WATER
OUT
PLASTIC
SHEETING
AIR INJECTION
WATER IN
Figure III-4. Hypolimnion aerator by Fast (1971).
37
-------
TEMPERATURE (°C )
5 10 15 20
3 -
6 -
E
I
£ 9
UJ
Q
12
15
—
/
- /
_ i
/
/
i
_
•
i
i
i
1
No Air
Air
Figure 111-5. Temperature profiles in Ottoville Quarry with and
without hypo "limnetic aeration.
38
-------
3 3
24.1 x 10 m of water. This profile would be relatively
stable during a summer without artificial aeration. How-
ever, when the quarry's hypolimnion was artificially oxy-
genated, the thermocline was "sharpened" and pushed to
a shallower depth. During two summers of hypolimnetic
aeration the hypolimnion extended from 5 m to maximum depth
and contained 35.2 x 10^ m^. Increases in hypolimnetic vol-
ume during aeration have been observed elsewhere (Fast, 1975]
Since most reservoir basins are funnel shaped, a difference
of only a few meters in hypolimnetic depth can represent a
very large increase in the volume of water which must be
circulated and aerated during hypolimnetic aeration. In
many cases, the hypolimnetic volume during aeration may ex-
ceed the non-aerated hypolimnetic volume by 50% or more. If
this situation is not considered, then the aeration system
may be undersized and inadequate.
Oxygen Consumption Rates
The rate of oxygen consumption in water (g O^/m /day)
and by the sediment (g 02/m2/day) can be used to estimate
oxygen input requirements during aeration (g 02/day). The
oxygen consumption rates can be calculated from estimates
of sediment and water respiration using techniques described
by Edberg and Hofsten (1973) or Edwards and Rolley (1965).
These methods involve incubating a sediment core sample in
the lab or in a respiration chamber placed over the sedi-
ment, and incubating water samples from different depths.
The rate of oxygen depletion or consumption is observed,
and the respective values are multiplied by the profundal
sediment area and hypolimnetic water volume. The sum of
these products represents an estimate of the total oxygen
consumption rate of the hypolimnion. This technique has
the advantage that it can be used any time of the year,
even after the hypolimnion has become anaerobic or during
periods when the lake is well mixed such as at overturn.
However, experience indicates that this method tends to
underestimate the hypolimnetic oxygen consumption rate
during aeration.
The most accurate estimates of hypolimnetic oxygen con-
sumption are derived from observing the rate of oxygen
depletion following the onset of thermal stratification.
The hypolimnetic oxygen depletion rate following thermal
stratification in the spring-summer is probably preferable,
but the fall-winter stagnation may also be used in those
lakes with ice cover. These methods are limited to one or
two periods of the year. These periods can be very short
in those lakes with high depletion rates. For example,
if the depletion rate is 1 mg 02/&/day and the water is
39
-------
saturated with oxygen when thermal stratification develops,
then the oxygen will be essentially absent within two weeks.
If the lake is monomictic (i.e. one overturn per year) there
will be only one brief opportunity to observe the oxygen
depletion rates.
The rate of oxygen depletion is calculated by first
determining the initial oxygen concentration.
n
Total Hypolimnetic _ y*
Oxygen Content t—t i i
where: V . = water volume in the ith depth interval (m )
C . = oxygen concentration in the ith depth
x 3
interval (grams O^/m )
n = number of depth intervals within
hypolimnion
The total hypolimnetic oxygen content is then plotted
against time as shown in Figure III-6. This plot yields
a curve from which the rate of oxygen depletion can be
calculated. The depletion rate in kg/day is calculated
from the slope of a regression line through selected data
points. Generally, the points which give a maximum de-
pletion rate are chosen. The rate of depletion is concen-
tration dependent and decreases as the hypolimnetic oxygen
content approaches zero.
This method underestimated by 30% the oxygen consumption
rate during hypolimnetic aeration of Lake Waccabuc. The
oxygen depletion rate was about 2.1 mg/£/week during the
spring prior to aeration, but the lake consumed 3.0 mg/i/week
during steady state aerated conditions and hypolimnetic
oxygen concentrations of 4.0 mg/£. Overholtz (19«75) found
closer agreement between oxygen depletion rates in Otto-
ville Quarry and oxygen consumption during aeration. De-
pletion rates averaged 1.3 mg/£/week, while consumption
rates were 1.0 mg/£/week. However, Smith, et al., (1975)
found that the consumption rate in Larson Lake, "...
during aeration was as much as three to four times as great
as the normal depletion rate." Factors which could account
for greater oxygen consumption during aeration include: (a)
increased water circulation caused by the aerator, and thus
more renewal of the mud-water interface; (b) increased
40
-------
0-
APR.
MAY
JUN.
JUL
AUG.
Figure III-6. Example of oxygen depletion rates data.
-------
circulation of water through the sediments by burrowing
organisms under well aerated conditions. Midge larvae,
oligochaetes and other benthic organisms can circulate
substantial quantities of water. 2Brinkhurst (1972) re-
ports values of 600 ml of water/m /hour; (c) vertical
mixing of sediments, caused by benthic organisms and;
(d) increased respiration and decomposition due to temp-
erature increases. Hypolimnetic aeration has caused
temperature increases ranging from a few degrees to more
than 9°C.
Hypolimnetic oxygen depletion rates without aeration
can vary considerably from year to year within a given
lake. For example, depletion rates at Lafayette Reservoir
ranged from 0.35 to 0.32 mg 02/&/week (Table III-2).
Values for nearby San Pablo Reservoir were much more uni-
form. Smith, et al., (1975) observed substantial variation
in oxygen depletion rates without aeration in both Mirror
and Larson Lakes. This variation was greatest in Larson
Lake where depletions (without aeration) ranged from 0.08
to 2.31 mg 02/Vweek.
Unfortunately if the oxygen depletion is measured for
only one season then the amount of yearly variability is
unknown. In these cases, ample allowance should be made
for both yearly variation and for possibly increased con-
sumption during aeration.
Oxygen Input Capacity
The oxygen input capacity will depend on: (a) desired
oxygen concentration, and (b) expected oxygen consumption
rate during aeration. In some cases, hypolimnetic aeration
may be successful even though it does not increase the hypo-
limnetic oxygen concentration above zero. Lake Brunnsviken
in central Stockholm, Sweden is a case in point. During
winter and summer stagnation large amounts of hydrogen
sulfide accumulated in the deep waters. This gas was vented
to the atmosphere at spring and fall overturns to the cha-
grin of the surrounding inhabitants and businesses. Hypo-
limnetic aeration of Brunnsviken has prevented the accumu-
lation of hydrogen sulfide in the lake even though the
hypolimnion still has no oxygen. A larger system would be
required in order to maintain oxygen above zero.
If iron and manganese are a problem, then oxygen concen-
trations above 2 mg/£ should prevent these substances from
coming into or remaining in solution (Bernhardt, 1974).
42
-------
TABLE III-2
HYPOLIMNETIC OXYGEN DEPLETION RATES FROM
SELECTED EUTROPHIC LAKES AND RESERVOIRS
Lake or Reservoir Year Time
Year
El Capitan, CA.
San Vicente, CA
Lafayette, CA.
San Pablo, CA.
Waccabuc, NY.
Ottoville, OH.
Mirror, WI .
Larson, WI
1967
1973
1960
1961
1962
1963
1964
1965
1966
1969
1970
1971
1972
1973
1973
1973
1973
1973
1973
1971-72
1972
1972-73
1973
1973
1971-72
1972
1972
1972-73
1973
1973
1973
1973
1973
1973
S = Spring or Summer rates
F = Fall rates
W = Winter rates (under ice
A = During arti
(1) hypolimnion
(2) July 12 to
ficial aerati
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S-A
S
S-A
(1)
W
S-A
W-A
S
S-A
W
W
F
W
W
W-A
W
S
S-A
F
of Oxygen
Depletion
(mg02/£/wk)
0.43
0.25
0.82
0.62
0.62
0.75
0.35
0.67
0.77
0.42
0.40
0.38
0.38
0.44
2.1
3.0
1.32
1 .00
1.26
0.28
>7.0
>2.5
1.3
(2) 2.6
0.59
0.26
1.75
0.08
1 .54
0.35
0.77
2.31
7.0
1.82
Source of observation or
data for calculations
Fast (1968)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpubl ished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Unpublished data (Fast)
Fast,(1973a,b)
Fast,(1973a,b)
Overholtz (1975)
Overholtz (1975)
Overholtz (1975)
Smith, et al. (1975)1
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al (1975)
Smith, et IT. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
Smith, et al. (1975)
cover)
on
circulated but
July 21
(hypol imnetic)
no oxygen added
43
-------
Oxygen concentrations of less than 3 mg/5, will
be unsatisfactory for most fish in most instances,
but oxygen below this level may be suitable for fish forage
organisms such as benthic fauna and zooplankton (Fast, 1971,
1973b). Fish vary greatly in their need for oxygen depending
on such factors as temperature, species, activity level,
maturity and physiological state. Most fish can live long
periods at 2 mg 02/£ and low temperatures, while some can
survive at 1 mg 02/£ or less, providing that toxic concen-
trations of hydrogen sulfide, carbon dioxide, ammonia or some
other substance are not present (Doudoroff, 1957). Fast
(1973a) and Overholtz (1975) successfully stocked rainbow
trout into the hypolimnions of lakes which were being aerated
artificially even though the oxygen concentrations were only
4 mg/£.. The trout utilized the aerated hypolimnion which
were each at 10°C. However, Miller and Fast (in prep.) found
that warmwater species did not utilize the deep waters with
less than 3 mg 02/£ at El Capitan Reservoir while the reser-
voir was being destratified even though the reservoir was
nearly isothermal. Although fish can and do utilize low
D.O. water, fishery biologists generally accept limits of
5 mg ()„/£ for coldwater fish (Doudoroff and Shumway, 1967) .
These limits are generally accepted even though it has
been shown that any value below air saturation (at the
surface) can restrict activity, growth and reproduction
(Doudoroff and Shumway, 1967; Shumway and Palensky, 1975).
In practice, it is highly desirable to oversize the
aeration system to allow for unforeseen variations in
oxygen consumption rates, hypolimnetic volume increases,
temporary equipment shutdown, or other factors. For ex-
ample, if the observed rate of oxygen depletion is 3.5
mg/£/week and the hypolimnetic volume is 2000 acre-feet,
then the oxygen depletion rate is 1.9 Ibs/min. It would
be desirable to size an aeration system to inject between
2.5 and 3.0 Ibs 02/min. at the desired maintenance oxygen
concentration. The maintenance level is important since
the efficiency of most aerators is dependent on the ambient
oxygen concentration. The added costs to oversize a system
are generally a small portion of the total capital costs.
If the system is over capacity then it does not need to
operate continuously- Intermittant operation of an over-
sized system can result in satisfactory oxygen levels
and operating costs comparable to the continuous operation
of a smaller system.
44
-------
Aeration Devices
The following discussion describes procedures to select
and to size components of the aerator system. The discussion
is limited to those systems suggested earlier.
Air Required --
The required air supply is based in part on the computed
oxygen consumption rates. However, because air is used as a
pump to circulate the water, the air required to provide
circulation must also be considered.
Oxygen absorption efficiencies for compressed air have
been reported to vary from 5 to 50% (Table III-l). This
efficiency is a function of many factors including: depth
of air injection, bubble size, oil content of air, chemical
characteristics of the water, air to water ratios in the up-
welling pipe, and most importantly, oxygen deficit in the water
being aerated. The precise relationships between these
factors are not well understood. However, a simplified
approach to design can provide useful guidance.
It has generally been observed that upwelled water is
saturated with oxygen by the time it reaches the surface
of full air lift devices. If it is assumed that water
reaching the surface is saturated, then the amount of flow
required to maintain a specified oxygen level can be com-
puted. The air required to provide that flow can also be
calculated as a function of aerator geometry. The follow-
ing example is used to illustrate the procedure.
Given: 1) Design oxygen depletion rate = 0.1 mg/£/day
7 3
2) Hypolimnetic volume = 10 m
3) .Maximum depth = 110 feet
4) Required D.O. = 5 mg/&
5) Hypolimnic Temperature = 55 F
Solution: 1) The minimum oxygen requirement is
0.1 mg/£/day x 107m3 = 1 x 109 mg/day
= 1 x 106 g/day
air @ 21% 02 = 8.5 g 02/ft
, 1,000,000 _ -, ! p nnn f
therefore need —'—5—F - lib, 000 rt
o . D
= 82 cfm
45
-------
2) Required water flow if oxygen concen-
tration increased from 5 mg/& to 9.5 mg/£.
1 x 106 g 09 needed fi
_—= = 222 x 10 £/day
4.5 x 10 g/H
= 7.778 x 106 ft3/day
= 5401 cfra
3) Air flow to give 5500 cfm of water in
100 foot upwelling pipe (see Appendix D,
Figure D-l).
Qa
(SCFM)
451
150
70
36
21
Diameter
(ft).
4
5
6
7
8
An air flow rate of 150 standard cubic feet per minute
(SCFM) and a 5 foot diameter upwelling pipe would be select-
ed. This computation assumes that half of the potential
energy is needed for upflow and half for downflow. There-
fore, the downwelling pipe(s) should have the same or less
total head loss as the upwelling pipe.
Air Source --
The usual air source is an air compressor located on the
shore or a raft. There are a wide variety of compressor
designs and manufacturers. The most appropriate compressor
for a given aeration system will depend primarily on the
availability of electrical power, discharge pressure re-
quirements, and the volume of air required. Electrical
power is generally the most economical means of producing
compressed air. Not only are costs per volume o£ air
generally lower than with fuel operated compressors, but
electrically run compressors generally need less attention,
have fewer mechanical problems, and create less noise.
Noise generation by any compressor can be substantial and
may require soundproofing depending on location and local
46
-------
conditions.
A description of procedures for compressor selection and
supply line design is given in Appendix C.
Diffuser Design and Bubble Size --
There is some controversy over the proper design of
diffusers and .desirable bnbble sizes. Without further
experimental and theoretical work no specific design cri-
teria can be given. In general, small bubbles well dis-
persed in the upwelling pipe are desirable.
Potential Problems
There are several potential problems with hypolimnion
aerators which are briefly discussed below.
Undersized Aeration Capacity —
It is possible to undersize the system either by under-
estimating the oxygen consumption rate of the hypolimnion
and/or by overestimating the amount of oxygen the system
will dissolve. In all cases i it. is highly desirable to de-
sign a larger system than is needed based on best estimates.
Oversizing the system will increase its capital costs only
slightly, and the operating costs should be equal to a
smaller system if the oversized system is operated inter-
mittently.
Unintentional Thermal Destratification --
It is possible to thermally destratify the lake during
hypolimnetic aeration if precautions are not taken. Fast
(1971) caused Hemlock Lake, Michigan, to destratify early be-
cause his hypolimnetic aeration system leaked water. His
aerator was constructed of corrugated steel plates which
were riveted together. Caulking should have been placed
between the plates before fastening. This leakage caused
a dense algal bloom since the upwelled water was high in
nutrients.
Attica Reservoir, New York, was rapidly destratified
by the Side Stream Pumping (SSP) hypolimnetic aeration
system even though only 2.6% of the hypolimnetic volume
was pumped per day (Fast, 1973b; Matsch, 1973). However,
the discharge velocity was very high and this apparently
imparted an intolerable momentum to the hypolimnion. The
volume pumped per day was later reduced to less than 1%
per day but the reservoir still destratified too rapidly
(Terry Haines, personal communications, 1976).
Satisfactory thermal stratification was maintained at
Ottoville Quarry, Ohio, with the SSP system when 2.6% of
the hypolimnetic volume was pumped per day. Nevertheless,
47
-------
Ottoville's hypolimnetic temperature increased greatly
during hypolimnetic oxygenation: 9.0°F in 3.5 months during
1973, and 14.4°F in 5 months during 1974 (overholtz, 1975).
Ottoville Quarry was 55 feet deep, compared to 31 feet for
Attica Reservoir.
Even though most full air lift designs move larger volumes
of water compared with the SSP, we know of none other than the
Hemlock Lake case where they have caused unacceptable destrat-
ification. This is undoubtedly due to their much lower dis-
charge velocities.
Nitrogen Gas Supersaturation --
Certain air injection systems may supersaturate the
water with nitrogen gas (N2) relative to surface hydro-
static pressures. The entire water column is normally at
100% saturation relative to the surface (Hutchinson, 1957),
and even a small increase will result in Supersaturation.
Nitrogen gas saturation in rivers of only 115% or less can
cause substantial mortalities of salmon and trout (Rucker,
1972).
According to Speece (1975a) even the full air lift hypo-
limnetic aerator and destratification air injection systems
may cause N2 Supersaturation relative to the surface of 160%
or more. We doubt this, but have no evidence to confirm or
reject the hypothesis. The only indirect evidence we have
is that oxygen concentrations at the top of hypolimnetic
aerators (full air lift designs), and oxygen concentrations
in the bubble plumes of lake destratification systems sel-
dom if ever exceed 100% saturation. Even if the "influent"
water were near zero in dissolved oxygen, we would not ex-
pect a very large amount of N2 to be absorbed . Some N2 almost
certainly would be absorbed as the 02 content of the air bubble
decreases relative to its N2 content. This change would in-
crease the partial pressure of N2 slightly (initially 78%)
and thus tend to put more N2 into the solution. This is
one of the most important questions which must be answered
before any form of air injection is used in conjunction with
tailwater releases.
The Limno hypolimnetic aeration system (partial air
lift design, Appendix B) can cause nitrogen gas Supersaturation
in the hypolimnion of more than 150% relative to surface pres-
sures (Fast, et al., 1975a). Special caution should be used
with this system even if there are no tailwater releases.
Under some situations, the system might cause N2 Supersaturation
even within the hypolimnion or base of the thermocline.
The principal reason for this is that the aeration occurs
in the hypolimnion and relatively high hydrostatic pressures
are maintained at all times.
48
-------
We know of no fish deaths caused by gas supersaturation
with any hypolimnetic aerator/oxygenator, but precautions
should be considered.
High Free Carbon Dioxide --
Systems which inject pure oxygen, and in which essentially
all the gas is dissolved in the hypolimnion, do not appear
to remove much gas such as CC>2. If the water is soft, and
is weakly buffered, there could be an excessive accumulation
of free CO2- The added oxygen may increase organic decompo-
sition and thus the generation of CC>2. There was not an un-
acceptable increase in free C02 at Ottoville Quarry even
during 5 months of oxygenation by the SSP process, but the
quarry was a former limestone quarry with well buffered,
hard waters.
Water Level Fluctuations --
Many of the hypolimnetic aerators described in Appendix
B are not well suited to accomodate large vertical changes in
the water levels. Some can be designed with telescoping
parts, but they are most efficient and least costly to build
in lakes with relatively stable water levels. If the water
level fluctuation is too drastic, there may not be much of
a hypolimnetic volume at the lower level, and no system would
be appropriate.
49
-------
SECTION IV
EXAMPLES OF METHODOLOGY APPLICATION
INTRODUCTION
The purposes of this section are to provide a comparison
of results with the procedures described here and also to
illustrate the application of the procedures to an example
lake.
PAST EXPERIENCE
Although a number of aeration/circulation systems have
been used in various parts of the world, there are very few
for which sufficient data are available to evaluate the ade-
quacy of procedures described in this document. However,
the procedures were found to be consistent with observations
from those few systems for which sufficient data were readily
available.
Kezar Lake
Kezar Lake in New Hampshire is a eutrophic recreational
lake of 182 acres (8 x 106 ft2) with a maximum depth of 30
feet and mean depth of 10 feet. As a result of nuisance
algal blooms the New England Regional Commission funded a
project to demonstrate the control of algae by artificial
mixing. The project was carried out by the New Hampshire
Water Supplytand Pollution Control Commission. Kezar Lake
was effectively destratified (nearly uniform vertical algal
profiles maintained) with the release of approximately 100
SCFM of air at a depth of 27 feet. The air was discharged
at 13 psi from four shore placed Curtiss CW-808 air com-
pressors through approximately 1000 feet of 2 inch PVC
pipe and 12 stone diffusers.
The design "rule of thumb" of 30 SCFM per 10 square
feet of Lake surface would have indicated 240 SCFM would
be needed for total mixing. Good mixing was probably
achieved with the smaller air flow rate due to shallow
depth and exposure to wind mixing.
The effects of mixing on algal production were found
to be consistent with the procedures described here.
Lorenzen and Mitchell (1975) provided an evaluation of
this project and compared observed values of peak algal
biomass to predicted values. The results, shown in Figure
IV-1 were compatible.
50
-------
60
CM 50
o
£ 40
o
5 30
o
< 20
10
NUTRIENT LIMITATION
LIGHT LIMITATION
OBSERVED
VALUES
2 3
MIXED DEPTH, METERS
5
Figure IV-1. Theoretical and observed peak biomass in Kezar Lake, 1968
stratified; 1969 and 1970 artificially destratified.
-------
El Capitan Reservoir
The chemical and physical characteristics of El Capitan
Reservoir, California have been described in detail by Fast
(1968). The reservoir impounds the intermittently flowing
San Diego River.
The reservoir area and volume fluctuate with changes
in surface elevation. However, the mean depth is approx-
imately 30 feet and the surface area 21 x 10^ square feer.
Thermal stratification usually begins in Ilarch or April and
disappears in November or December.
The reservoir was artificially aerated during 1965 and
1966 with approximately 215 SCFM of air discharged at approx-
imately 60 psi from a shore installed Le Roi 50-S-2 compressor
driven by a 50 horsepower electric motor. The 'air was re-
leased through 90, 1/8 inch diameter holes in 100 feet of
1-1/2 inch PVC pipe at a depth of approximately 80 feet.
This system did not destratify the reservoir and rates
of algal production were increased. No biomass measurements
were made. This increase was apparently due to a lessening
of the mixed depth which is consistent with the theory used
here. It is estimated that significantly more air (30 x 21
= 630 SCFM) would be required to adequately mix El Capitan
Reservoir.
San Vicente Reservoir
An aeration system has not been installed in San Vicente
Reservoir. However, some cost estimates have been made
(Fast, et al., 1976) and some data are available. This
example was chosen to illustrate the procedures for evalua-
tion .
San Vicente Reservoir is the major domestic water supply
reservoir for the City of San Diego. The reservoir stores a
small amount of direct runoff, but most of its water is imported
primarily from the Colorado River. It is a warmwmonomictic
lake. Thermal stratification typically begins in March or
April, and may extend through December or even into January.
The maximum volume, depth, and surface area are 90,000 acre-
feet, and 190 feet, and 1000 acres respectively. Oxygen
depletion rates are known to vary considerably from year
to year. However, data taken during 1973 will be used for
this example.
Thermal and chemical stratification developed in San
Vicente during April, 1973 at which time surface tempera-
tures were 59°F and bottom temperatures were 55°F (Figure
IV-Z). The thermocline was not well defined, but oxygen
52
-------
OXYGEN (mg/n
TEMPERATURE (°F)
en
CO
Q_
UJ
0
20-
40-
60-
80-
100
120
140
160
0
50
NOV.J;JUL
APR,
APR.
NOV\
70
^^fLi-*~^j^j^Jjf<~*~*~
JUL!
Figure IV-2. Temperature and oxygen profiles in the San Ficente
Reservoir.
-------
concentrations decreased sharply from 11 mg/£ at 20 feet
to 7 mg/£ at 50 feet. During July the thermocline was
well defined and extended from 25 to 50 feet. Hypolimnetic
oxygen had decreased from 7.5 mg/£ in April to less than
3 mg/£. During November surface waters had cooled to 67 F,
but the thermocline was still well defined between 30 and 50
feet. Hypolimnetic oxygen had decreased to an average
0.5 mg/&. These data yield an oxygen depletion rate of
slightly less than 1 mg/Vmonth, and a thermocline range of
25 to 50 feet during mid-summer.
Water depths ranged between a low of 170 feet in
January to a high of 184 feet in August. Maximum water
volume was 84,021 acre-feet. The base of the thermocline
was 50 feet without aeration during 1973. However, we will
assume that it would be raised to the 40 foot depth during
aeration. The hypolimnetic volume was approximately 47,720
acre-feet (62.9 x
Hypolimnion Aeration --
Following the procedures outlined in Chapter III, the
oxygen requirements are estimated to be;
Oxygen Required --
0.5 mg/Vmonth) 62>9 1Q9 £ = 3 145 x 1(J6 /(j
30 days/mo.
Minimum Air Flow (100% 0 transfer efficiency) --
Air at 21% 02 - 8.2 g02/ft3
3.145 x TO6 g/day = 370jOQO ft3/day = 256 SCFM
8.2 g/f-T
Air Flow Based on Saturation —
Water temperature = 55 F
Required 02 = 5 mg/l
Saturation at 55°F =*9.5mg/A
Increase = 9.5 - 5.0 = 4.5 mg/£
54
-------
Water = 3.145 x 10° g/da.y
°W 4.5 x 10"J g/£
Q = 283 cfs
w
1Q £/day
See Appendix D, Figure D-2 (for 150 foot depth)
Diameter
6 feet
7
8
Q3/Q
W c
7.5 x
1.4 x
2.4 x
i
io5
io6
io6
Q
a
>> 1800 SCFM
^
"*"*^
An eight foot diameter upwelling pipe with 565 SCFM of
air supplied at the 150 foot depth should be sufficient.
Based on a 300 foot, 2 inch diameter supply line, the
pressure required at the compressor discharge (following
filter and oil removal devices) would be (See Appendix C):
P = h + AP + depth
14.7
£
(IV-1)
where:
h
£
AP = 5 psi
14.7
150
33
6 psi (CRANE nomographs)
(trial value)
= 67 psi
P=6+5+67= 78psi
A compressor to deliver 565 SCFM at 80 psi would require
approximately 115 horsepower. Approximately 50, 1/8 inch
diameter orifices would be required to relase the 565 SCFM
of air at the bottom of the upwelling pipe (see Section II).
Destratification —
San Vicente Reservoir has a surface area of approximate-
ly 1000 acres (44 x IO6 ft2). According to the rule of thumb
established in Section II, approximately (30 x 44) = 1320
SCFM of air would be required for destratification. There
are not sufficient data to determine the effects of destrati-
fication on algal production. However, if thorough mixing
to a depth of even 60 feet could be achieved, very low algal
concentrations could be maintained.
55
-------
Cost Estimating --
Hypolimnion Aeration —
The major capital cost items for hypolimnion aeration
systems will be:
« Device
9 Compressor
• Air Supply Lines and Diffusers
The actual device could be a simple upwe 11 ing-down-well-
ing pipe such as shown in Figure B-5, in Appendix B. Fast,
et al., (1975c) estimated capital costs to be on the order
of $~200,000 for such a system. Various patented systems
described in Appendix B could also be used. Local sup-
pliers should be asked for prices on the described simple
systems to provide a comparison to price quotations for
patented systems.
Operating costs will be primarily for electricity. The
example system with a 115 horsepower motor operated 6000
hours per year would consume approximately 575,000 kw-hours.
At $0.03 per kw-hour this would be $15,500 per year.
Destratification --
The major capital cost items associated with a destrati-
fication system would be:
« Air Compressors
• Supply Line
• Diffuser
Two complete air compressors, motors, and associated
appurtenances to deliver 1200 SCFM of air would cost on
the order of $50 , 000-$60,000. Supply line and diffusers
would be approximately $5,000.00.
Power costs would be approximately double those for the
suggested hypolimnion aeration system or $31,000.00 per
year.
56
-------
SECTION V
REFERENCES
Annonymous. 1973. Seeking more fish. News in Engineering
Ohio State University, Columbus Ohio. pgs. 20-21.
Atlas Copco, Brochure 5SC1-1273. Atlas Copco, Inc., 930
Brittan Avenue, San Carlos, California 94070.
Bardach, J.E. 1955. Certain biological effects of
thermocline shifts. Hydrobiologia. 7:309-324.
Barnett, R.H. 1971. Reservoir destratification improves
water quality. Public Works. 102 (6) : 60-65.
Barnett, R.H. 1975. Core study of reaeration of Casitas
Reservoir. ASCE Speciality Conference on Reaeration
Research, Gatlinburg, Tennessee.
Bengtsson, L., H. Berggren, O. Meyer, and B. Verner. 1972.
Restauerering av sjor med kulturbetingat hypolimniskt.
Limnologiska Institutionen, Lund Universitet, Centrala
Fysiktaboratoriet, Atlas Copco AB.
Bernhardt, H. 1967. Aeration of Wahnbach Reservoir without
changing the temperature profile. J.Amer.Water Works
Assoc. 59:943-964.
Bernhardt, H. 1974. Ten years experience on reservoir
aeration. Presented at the 7th International Con-
ference on Water Poll.Res., Paris, France.
Bjork, S. 1974. European lake rehabilitation activities.
Presentation at the Conference on Lake Protection and
Management. Madison, Wisconsin.
Booth, C.R. 1976. The design and evaluation of a measure-
ment system for photosynthetically active quantum scale
irradiance. Limnology and Oceanography. 21 (2) : 326-336
Brinkhurst, R.O. 1972. The role of sludge worms in eutro-
phication. Ecolog.Res. Series, U.S. Environmental
Protection Agency, Washington, B.C.
Brown, D.J., T.G- Brydges, W. Ellerington, J.J. Evans,
M.F.P. Michalaski, G.G. Hitchin, M.D. Palmer, and
D.M. Veal. 1971. Progress report on the destrati-
fication of Buchanan Lake. Ont.Water Res.Comm., AID
for Lakes Program (Artificially Induced Destratifi-
cation).
57
-------
Butler, R.L. and D.P. Borgeson. 1965. California's "catch-
able" trout fisheries. California Department of Fish
and Game, Fish Bull. 127. 44 pp.
Chen, C.W. and G.T. Orlob. 1975. Ecologic simulation for
aquatic environments in systems analysis and simulation
in ecology- Academic Press, N.Y., San Francisco, and
London. 111:475-588.
Confer, J.L., R.A. Tubb, T.A. Haines, P. Blades, W. Over-
holtz, and C. Willoughby. 1974. Hypolimnetic aeration
with destratification: zooplankton response in three
lakes with normal clinograde oxygen curves. Presented
at 37th Annual Meeting Amer.Soc.Limn.Ocean., Seattle
Washington.
CRANE. 1973. Technical Paper 410. CRANE, Co., Chicago
Illinois 60632.
Dendy, J.S. 1945. Distribution of fish in relation to
environmental factors, Norris Reservoir. Report
Reelfoot Lake Biological Station. 9:114-135.
Dillon, P.J. 1974. A manual for calculating the capacity
of a lake for development. Ontario Ministry of the
Environment.
Doudoroff, P. 1957. Water quality requirements of fishes
and effects of toxic substances. In: The Physiology
of Fishes (M.E. Brown, Ed.). Academic Press, Inc.,
New York. 1:403-430.
Doudoroff, P. and D. Shumway. 1967. Dissolved oxygen
criteria for the protection of fish. Trans.Amer.
Fish.Soc., Special Publication No. 4. pp. 13-19.
Drury, D.D. 1975. The effects of artifical destratifica-
tion on the water quality and microbial populations
of Hyrum Reservoir. Utah Water Research Laboratory,
College of Engineering, Utah State University, Logan,
Utah 94322. 174 pp.
Dunst, R.C., S.M. Born, P.O. Uttormark, S.A. Smith, S.A.
Nichols, J.O. Peterson, D.R. Knauer, S.L. Serns, D.R.
Winter, and T.L. Wirth. 1974. Survey of lake restor-
ation techniques and experiences. Tech.Bull. 75,
Department of Natural Resources, Madison, Wisconsin.
179 pp.
Edberg, N. and B.V. Hofsten. 1973. Oxygen uptake of bot-
tom sediments studied in_ situ in the laboratory.
Water Resources. 7:1285-1294).
58
-------
Edwards, R.W. and H.L. Rolley. 1965. Oxygen consumption
of river muds. J.Ecol. 53(1):1-19.
Environmental Protection Agency. 1970. Induced aeration
of small mountain lakes. EPA 16080—11/70.
Fast, A.W. 1968. Artificial destratification of El
Capitan Reservoir, Part I, effects on chemical and
physical parameters. Fish.Bull. 141, California
Department of Fish and Game. 97 pp.
Fast, A.W. 1971. The effects of artifical aeration on
lake ecology. Environmental Protection Agency.
Water Pollution Control Series 16010 EXE 12/71.
470 pp.
Fast, A.W. 1973a. Effects of artificial hypolimnion
aeration on rainbow trout (Salmo gairdneri Richardson)
depth distribution. Trans.Amer.Fish.Soc. 102(4):715-
722.
Fast, A.W. 1973b. Review of three hypolimnetic aeration
projects. News Release, Union Carbide Corporation,
New York, New York.
Fast, A.W:. . 1975. Artificial aeration as a, lake restoration
technique. Symposium: Recovery of damaged ecosystems.
Virg.Poly.Inst.State University, Blackburg, Virginia.
Fast, A.W. , J-.A. St. Ament. 1971. Nighttime artificial
aeration of Puddingstone Reservoir, Los Angeles
County, California. California Fish and Game
57(3):213-216.
Fast, A.W., V.A. Dorr, and R.J. Rosen. 1975a. A submerged
hypolimnion aerator. Water Resources Res. 11(2):287-
293.
Fast, A.W., W.J. Overholtz, and R.A. Tubb. 1975b. Hypo-
limnetic oxygenation using liquid oxygen. Water
Resources Research. 11 (2) : 294-299.
Fast, A.W., M.W. Lorenzen, and J.H. Glenn. 1975c (M.S.).
Hypolimnetic aeration/oxygenation: a comparative
study with costs. Presented at Symposium on Reaeration
Research, Gatlinburg, Tennessee.
Gartman, O.K. 1973. Artificial aeration and destratifica-
tion of Ottoville Quarry. Ph.D. Thesis, Ohio State
University, Columbus, Ohio. 93 pp.
59
-------
Carton, J. and Jarrell, H.R. 1975. Lake Arbuckle —
A core study of destratification. ASCE Specialty
Conference on Reaeration Research, Gatlinburg,
Tennessee.
Gebhart, G.E. and R.C. Summerfelt. 1975. Effects of
destratification on depth distribution of fish.
Presented at Symposium on Reaeration Research,
ASCE, Gatlinburg, Tennessee.
Haines, Terry. 1976. Personal communications. State
University of New York at Brockport.
Hile, R. 1936. Age and growth of the cisco, Leucichthys
artedi (Le Sueur), in lakes of the Northeastern
Highlands, Wisconsin. Bull.U.S.Bur. of Fish. 48:211-
217.
Hirshon, B.E. 1974. Method and apparatus for aerating
bodies of water. U.S. Patent No. 3,794,303.
Hutchinson, G.E. 1957. A Treatise on Limnology. John
Wiley & Sons, Inc., New York. 1015 pp.
Irwin, W.H., J.M. Symons and G.G. Robeck. 1966. Impound-
ment destratification by mechanical pumping. Proc.
Amer.Soc. of Civil Engineering, J.Sanitary Eng.Div.
92:21-40.
Johnson, R.C. 1966. The effects of artificial circulation
on production in a thermally stratified lake. Wash.
Dept.Fish., Fish.Res. paper. 2(4):5-15.
Knauer, D.R. 1974. Personal communications. Department
of Natural Resources, Madison, Wisconsin.
Koberg, G.E. and M.E. Ford. 1965. Elimination of
thermal stratification in reservoirs and the resulting
benefits. U.S. Geol.Surv., Water Supply Paper 1809M.
Kobus, H.E. 1968. Analysis of the flow induced by air
bubble system. Coastal Eng.Conf., London. 11:1016-
1031.
Lackey, R.T. 1972. A technique for eliminating thermal
stratification in lakes. Water Res.Bull. 8(l):46-49.
Leach, L.E., W.R. Duffer, and C.C. Harlin, Jr. 1970-
Induced hypolimnion aeration for water quality
improvement of power releases. Environmental
Protection Agency, Water Pollution Control Series
16080. Ada, Oklahoma. 32 pp.
60
-------
Lorenzen, M.W. and R. Mitchell. 1973. Theoretical effects
of artificial destratification on algal production in
impoundments. Envir.Sci.Tech. 7 (10): 939-944.
Lorenzen, M.W. and R. Mitchell. 1975. An evaluation of
artificial destratification for control of algal
blooms. J.Amer.Water Works Assoc. 67 (7): 372-376.
Lorenzen, M.W., D.J. Smith, and L.V. Kimmel. 1976. A long
term phosphorus model for lakes: application to Lake
Washington. In: Modeling Biochemical Processes in
Aquatic Ecosystems, Raymond P. Canale, ed., Ann Arbor
Science Publishers, Inc., Ann Arbor, Michigan. pp.
75-91.
Lyon, D.M. 1975. Personal communication. University of
California, Berkeley, California.
MacBeth, S.E. 1973. Some observations on changes in water
chemistry and algal growth in artificial destratified
lakes in Ontario. Presented at the 36th Annual
Meeting of the Amer.Soc. of Limnol. and Oceanogr.,
Salt Lake City, Utah.
Maleug, K.W., J.R. Tilstra, D.W. Schults, and C.F. Powers.
1973. Effect of induced aeration on stratification
and eutrophication processes in an Oregon Farm Pond.
Geophysical Monograph Series. 17:578-587.
Matsch, L.C. 1973. Description of oxygenation systems
for hypolimnia. News Release, Union Carbide Corpora-
tion, New York, New York.
Mayhew, J. 1963. Thermal stratification and its effects
on fish and fishing in Red Haw Lake, Iowa. Biological
section, State Conservation Comm. 24 pp. (mimeo).
McNall, W.J. 1971. Destratification of lakes. Federal
AID Project F-22-R-11, J of C-8, Job program report.
31 pp (mimeo).
Mercier, P. 1955. Aeration partielle sous-lacustrine d'un
lac eutrope. Berh.Int.Ver.Limnol. 10:294-297.
Mercier, P. and J. Perret. 1949. Aeration station of Lake
Bret, Monastbull, Schweiz. ver.Gas.U. Wasser-Fachm.
29:25
Mercier, P. and S. Gay. 1954. Effets de 1'aeration artifi
cielle sous-lacustre au lac de Bret, Schweizer z.f.
Hydrol. 16:248-308.
61
-------
Miller, L.W. and A.W. Fast. (In prep.). Effects of artificial
destratification on fish depth distribution in El Capitan
Reservoir.
N.H.W.S.P.C.C. 1971. Algal control by mixing. Staff report
on Kezar Lake. New Hampshire Wtr.Sply. and Pollution
Cont.Com.Concord, New Hampshire.
Overholtz, W.J. 1975. An ecological evaluation of hypo-
limnetic oxygenation by the side stream pumping process
at Ottoville Quarry, Ottoville, Ohio. M.S. Thesis,
Ohio State University, Columbus, Ohio. 100 pp.
Raynes, J.J. 1975. Case Study -- Allatoona Reservoir.
ASCE Specialty Conference on Reaeration Research,
Gatlinburg, Tennessee.
Riddick, T.M. 1957. Forced circulation of reservoir
waters yields multiple benefits at Ossining. N.Y.
Water and Sewage Works. 104 (6) : 231-237.
Ridley, J.E. and J.M. Symons. 1972. New approaches to
water quality control in impoundments. Water
Pollution Microbiology, Ralph Mitchell, ed., Wiley
Interscience, New York. 389-412.
Rucker, R. 1972. Gas bubble disease: a critical review
Bur. of Sport Fish, and Wildlife, Tech. Paper No. 58.
U.S. Department of the Interior.
Sakamoto, M. 1966. Primary production by phytoplankton
community in some Japanese lakes and its dependence
on lake depth. Arch.Hydrobiol. 62:1-28.
Seppanen, P. 1974. Lake aeration in Finland. Presenta-
tion at conference on Lake Protection and Management,
Madison, Wisconsin.
Shapiro, J. 1973. Blue green algae: why they become
dominant. Science. 179:382-384.
Shumway, D.L. and J.R. Palensky. 1975. Influence of dis-
solved oxygen on freshwater fish. Presented at ASCE
Symposium on Reaeration Research, Gatlinburg, Tennessee.
Smith, S.A., D.R. Knauer and T.L. Wirth. 1975. Aeration
as a lake management technique. Tech.Bull. 87, Wiscon-
sin Department of Natural Resources, Madison, Wisconsin.
39 pp.
62
-------
Snodgrass, William J. and Charles R. O'Melia. 1975. Pre-
dictive model for phosphorus in lakes. Environmental
Science and Technology. 9 (10) : 937-944.
Speece, R.E. 1970. Downflow bubble contact aeration
apparatus and method. U.S. Patent No. 3,643,403.
Speece, R.E. 1971. Hypolimnion aeration. J.Amer. Water
Works Assoc. 64(1):6-9.
Speece, R.E. 1973. Alternative considerations in the
oxygenation of reservoir discharges and rivers.
In: Applications of Commercial Oxygen to Water
and Wastewater Systems, R.E. Speece, ed. University
of Texas, Austin, Texas.
Speece, R.E. 1975a. Personal communication. Drexel
University, Philadelphia, PA.
Speece, R.E. 1975b. Oxygen restoration to waters released
from Clark Hill Reservoir. Final report to U.S. Army
Corps of Engineers, Savannah, Georgia.
Speece, R.E. 1975c. Oxygenation of Clark Hill discharges.
Presented at ASCE Symposium on Reaeration Research,
Gatlinburg, Tennessee.
Speece, R.E., F. Rayyan and C.A. Givler. 1974. Invited
discussion of ten years experience of reservoir
aeration by H. Bernhardt (1974).
Stumm, W. and E. Stumm-Zollinger. 1972. The role of phos-
phorus in eutrophication. Water Pollution Micro-
biology, Ralph Mitchell, ed. Wiley Interscience,
New York. pp. 11-42.
Symons, J.M., W.H. Irwin, E.L. Robinson, and G.G. Robeck. 1969
Impoundment destratification for raw water quality
control using either mechanical or diffused-air
pumping. In: Water Quality Behavior in Reservoirs,
James M. Symons, ed. U.S. Department of Health
Education and Welfare, Paper No. 12. pp. 389-427.
Tyler, J.E. 1974. SCOR Working Group 15 recommendations.
Van der Hegge Zijnen, E.G. 1951. Flow through uniformly
topped pipes. Applied Science Research. A3:144-162,
Vollenweider, R.A. 1965. In: Primary Productivity in
Aquatic Environments. (C.RT Goldman, editor).
University of California Press, Berkeley and
Los Angeles, California. 425 pp.
63
-------
Vollenweider, R.A. 1969. Possibilities and limits of elemen-
tary models concerning the budget of substances in lakes.
Arch.Hydrobiology. 66(1).
Water Resources Engineers, Inc.
for aquatic environments.
Research.
1972. Ecologic simulation
Office of Water Resources
Weiss, Charles M. and Benjiman W. Breedlove. 1973. Water
quality changes in an impoundment as a consequence
of artificial destratification. ESE Publication No.
322, University of North Carolina at Chapel Hill.
216 pp.
Welch, P.S. 1948. Limnological Methods. McGraw Hill
Book Company, Inc., New York. 381 pp.
Westlake, D.F. 1965. Some problems in measurement of
radiation underwater — A review. Photochemistry
and Photobiology. 4:849-868.
Whipple, W., Jr. 1975. Personal Communication.
University, New Brunswick, New Jersey.
Rutgers
Wirth, T.L. and R.C. Dunst. 1967. Limnological changes
resulting from artificial destratification and aeration
of an impoundment. Wisconsin Conserv.Dep., Fish.Res.
Rep.No. 22.
Ziebell, C.D. 1969. Fishery implications associated with
prolonged temperature and oxygen stratification.
J.Arizona Acad.Sci. 5(4):258-262.
64
-------
SECTION VI
APPENDIX A
LAKE MODEL COMPUTATIONS
The basic model used for computation of temperature and
algal profiles for various levels of induced mixing is based
on the Lake Ecologic Model originally developed by Chen and
Orlob (1975) . The model was modified for this application
to compute only temperature and algal growth. Two types of
algae were simulated; one neutrally bouyant and one with a
rise velocity of 2 feet per day.
The purpose of the model application was solely to sim-
ulate the effects of imposed mixing on the vertical profiles
of temperature and algae.
Physical Representation
Each lake simulated was idealized as a number of hori-
zontally mixed layers. Natural vertical mixing is computed
by the use of dispersion coefficients in the vertical mass
transport equation. Values of the dispersion coefficients
for different size lakes were estimated from previous
studies (Water Resources Engineers, Inc., 1969).
Temperature
Temperatures as a function of depth were computed ac-
cording to equation A-l.
• (AD-I)- - (QT) * (, - v
where: T = the local water temperature
c = specific heat
p = fluid density
A = cross-sectional area at the fluid
element boundary
D = the eddy diffusion coefficient in the
vertical direction
Q - advection across the fluid element
boundaries
65
-------
A = cross-sectional area of the surface
3 fluid element
yA = coefficients describing heat transfer
across air water interface
0 = sum of all external additions of heat
to fluid volume of fluid element
v = element volume
To assure comparability between simulation the same
weather conditions were used for all cases.
Algal Growth
Because only the vertical distribution of algae was
of interest, algal growth rates were assumed to be only
a function of available light and temperature. Nutrient
dependence was not considered.
Algal concentrations as a function of depth were
computed according to equation A-2.
-§7 (PAZ) PS (A-2)
where: P = the algal concentration
PG = the algal growth rate
PG = PMAX ( L* )
J_i ^ ~r lj JL
PMAX= the maximum specific growth rate of
algae at 20°C
LI = the available light energy
L = the half saturation constant for algae
utilizing light energy
PR = the algal respiration rate
PS - the algal settling rate
Q = vertical advective flow
66
-------
Light intensity as a function of depth was computed
from the relationship:
LI = I e
o
(-a + 3P)d
(A-3)
where: I = surface light
o ^
a = extinction coefficient
B - incremental extinction coefficient due
to algae
d = depth
The algal growth rate parameters used were:
Parameter
PMAX
PR
PS
L2
a
3
Value
Alga 1
1.0
0.15
0
0.03
0.7
0.2
Alga 2
1.0/day
0.15/day
-0.5 m/day
0.03 langley
0. 70 m"1
0. 2 m~1/mg/£
Imposed Mixing
The theory of Kobus (1972) indicates that upwelling
flow can be computed if the air release rate and depth are
known. His relationship was used to compute an incremental
vertical flow for each layer. The total flow reaching the
top element was then evenly distributed from the top element
down to the element with a density equal to the mixture of
upwelled water. This procedure is diagramatically illu-
strated in Figure A-l.
Application-Verification
The model was applied to Kezar Lake in New Hampshire
and El Capitan Reservoir in California to verify that arti-
ficial mixing could be adequately simulated.
67
-------
Figure A-l. Diagramatic representation of air induced vertical
mixing.
68
-------
Computed temperature profiles are compared to observed
values in Figures A-2 and A-3. The model performance was
judged to be good for the intended purpose of providing guid-
ance. However, it is recommended that for any large scale
application, a simulation model be applied to the specific
lake in question.
Test Cases
This modeling procedure was applied to fourteen hypo-
thetical lakes listed in Table A-l. The range of depths and
volumes was intended to cover most probable situations. For
each lake, the model was run using the same weather conditions.
The temperature and concentrations of each alga in each layer
were computed on a daily basis for 180 days. The maximum
difference between top and bottom temperatures and algal con-
centrations that occurred during the simulation period were
then used to assess the degree of mixing. Figure A-4 shows
the results obtained for Kezar Lake with and without aeration.
Each lake simulation was repeated with different air
release rates. The concentration and temperature differences
were then plotted as a function of air release rate for each
lake. Figures A-5 through A-l8 show the results. The value
of AT represents the maximum difference between surface and
bottom temperatures computed over the 180 day period. The
value of AC represents the maximum difference between sur-
face and bottom algal concentrations for each of the two
algae.
It was found that, for the conditions simulated, the
air required in standard cubic feet per minute increased
nearly linearly with area and was nearly independent of lake
depth (of course, the energy is greater than that required
for a shallow lake).
As a "rule of thumb" it was found that approximately
30 (20-40) SCFM of air per 106 ft2 of lake surface area should
provide good mixing. This number could be reduced for shallow
lakes exposed to extensive wind mixing and should be increased
for warmer climates which tend to increase stability.
69
-------
10
TEMPERATURE (°C)
20 0 10
20
30 V
3-
6
12
I5-I
18-
21-
0
3 JUL 1968
I5JUL.I968
\~-~-Simulated
Figure A-2. Comparison of computed and observed temperature profiles
in Kezar Lake.
70
-------
EL CAPITAN 1964 -NO Ml XING
TEMPERATURE (°C)
10 15 20 10 15 20 10 15 20 25 U
14—
X
1—
Q.
LlJ
Q ?O-
4O-
60-
Of).
f
/X
>*•
• I
X
./
/
*T™
:/
/
y
/
Simulat
-Prototy
ISO DA YS 165 DAYS ISO DAYS
EL CAPITAN 1966 - WITH AERATION
TEMPERATURE (°C)
10 15 20 10 15 20 10 15 20 10 15 20
100
6ODAYS 9ODAYS I2ODAYS I80DAYS
Figure A-3. Comparison of computed and observed temperature profiles
in El Capitan Reservoir.
71
-------
TABLE A-l. GEOMETRY OF LAKES SIMULATED WITH RESERVOIR MODEL
Lake
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Avg. depth, (ft)
9.7
9.7
9.7
9.7
26
26
26
26
50
50
50
50
93
93
Max. depth, (ft.)
21
21
21
21
51
51
51
51
100
100
100
100
200
200
2
Surface area, (ft )
7.47 x 106
1 .49 x 107
2.99 x 107
7.47 x 107
7
1 .71 x 10
3.42 x 107
7
8.55 x 107
2.14 x 108
7
4.26 x 10
1.07 x 108
2.66 x 108
1.07 x 109
2.07 x 108
8.28 x 108
Volume, (ft3)
7.23 x 107
1 .45 x 108
2.89 x 108
7.22 x 108
R
4.41 x 10°
8.80 x 108
Q
2.21 x 10
5.52 x 109
Q
2.10 x 10s
5.24 x 109
1 .31 x 1010
5.24 x 1010
1.92 x 1010
7.66 x 1010
-------
CO
s
Q
CM
at
s-
3
•t-J
to
S-
(U
Q.
O)
O CD
_i£
O) (13
X S-
•r- (O
E N
01
M- ^:
o
(/) T-
+j
o c:
O) O
q- T-
<+- 4J
cu to
o
-a -i-
cu <*-
Q- (O
E S-
O -4->
O tn
O)
I- o
< o
73
-------
32--
LAKE NO. I
Mean Depth = IOFT.
Volume = 72xl06FT.3
--16
AC
(mg/l)
AT
960
AIRFLOW (scfm)
Figure A-5. Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 1.
74
-------
LAKE NO. 2
Mean Depth = IOFT.
Volume = 145 x!06FT.3
-r20
--16
--12
AC
(mg/l)
AT
--8
--4
600
1200
AIRFLOW (scfm)
1800
Figure A-6. Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 2.
75
-------
LAKE NO. 3
Mean Depth = IOFT.
Volume = 289xl06FT.3
--16
AC
(mg/l)
T-20
--12
AT
--8
--4
1200
2400
AIRFUOW (scfm)
3600
-0
4800
Figure A-7. Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 3.
76
-------
Mean Depth = IOFT.
Volume = 722 x I06 FT.3
AT
1200
2400
AIRFLOW (scfm)
Figure A-8. Computed effects of induced mixing on temperature
and algal stratification in hypothetical lake no. 4.
77
-------
LAKE NO. 5
Mean Depth = 26FT.
Volume = 441 x!06FT.3
+ 16
+ 12
AC
(mg/l)
AT
CO
+8
+4
600
1200
AIRFLOW (scfm)
1800
Figure A-9. Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 5.
78
-------
LAKE NO.
+ 16
+12
AC
(mg/l)
Mean Depth = 26FT
Volume
AT
CO
600
1200
AIRFUOW (scfm)
Figure A-10.
Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 6.
79
-------
Mean Depth = 26FT.
Volume = 222lxl06FT.3
--16
--12
AT
--8
2400
4800
AIRFLOW (scfm>
9600
14400
Figure A-ll. Computed effects of induced mixing on temperature
and algal stratification in hypothetical lake no. 7.
80
-------
LAKE NO. 8
Mean Depth = 26 FT.
Volume = 5520xl06FT.3
32+
+ 16
+ 12
AC
(mg/l)
AT
2400
4800
AIRFLOW (scfm)
9600
14400
Figure A-12,
Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 8.
81
-------
20-r
LAKE NO. 9
Mean Depth = SOFT.
Volume = 2IOOxl06FT.3
T20
+ 16
+ 12
AC
(mg/l)
+8
1200
AT
CO
+4
2400
AIRFLOW (scfm)
3600
Figure A-13.
Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 9.
82
-------
LAKE NO. 10
Mean Depth = SOFT.
Volume = 5240 x I06FT.3
-H7
AC
(mg/l)
CO
CO
AT
1200
2400 3600
AIRFLOW (scfm)
4800
I
6000
Figure A-14.
Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 10.
-------
LAKE NO. I!
Mean Depth = SOFT.
Volume - 1310 x I06FT.3
-r 6
--12
AC
(mg/l)
oo
AT
4800 7200
AIRFLOW (scfm)
Figure A-15. Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 11.
-------
LAKE NO. 12
Mean Depth = 50 FT.
Volume = 5a40xl06FT.3
-rl6
AC
(mg/l)
Co
en
AT
2400
4800 7200
AIRFLOW (scfm)
9600
Figure A-16.
Computed effects of induced mixing on temperature
and algal stratification in hypothetical lake no.12.
-------
16
LAKE NO. 13
Mean Depth = 93 FT.
Volume = 19200 x I06FT.3
16
12--
--12
AC
(mg/l)
00
--8
AT
2400
4800 7200
AIRFLOW (scfm)
9600
12000
Figure A-17. Computed effects of induced mrxtng on temperature and
algal stratification in hypothetical lake no. 13.
-------
LAKE NO. 14
Mean Depth = 93FT.
Volume = 76600 x I06FT3
-r-16
AC
(mg/l)
CO
AT
4800 7200
AIRFLOW (scfm)
Fiaure A-18.
Computed effects of induced mixing on temperature and
algal stratification in hypothetical lake no. 14.
-------
SECTION VII
APPENDIX B
HYPOLIMNETIC AERATION/OXYGENATION
A wide variety of hypolimnetic aeration/oxygenation
systems have been proposed and/or tested. At least 21 de-
signs have been proposed and 13 of these have been subjected
to full scale testing. These systems add oxygen to the hypo-
limnion by mechanical agitation, air injection or injection
of high purity oxygen. Although this variety of systems
exist, no system is in widespread use. There are not suf-
ficient data to thoroughly evaluate all of the systems. The
purpose of this appendix is to provide a summary of these
systems with performance data when available.
HYPOLIMNETIC AERATION/OXYGENATION DEVICES
Mechanical Agitation System
Hypolimnetic aeration was first reported for Lake Bret,
Switzerland, where hypolimnetic waters were aerated by mechan-
ical means (Mercier and Perret, 1949; Mercier and Gay, 1954;
Mercier, 1955). Water was withdrawn from the hypolimnion
through an intake pipe, and discharged into a splash basin
on shore where it was aerated (Figure B-l). The aerated
water then returned to the hypolimnion through a discharge
pipe by gravity flow. This system is relatively inefficient
in terms of oxygen absorbed/kw-hr, but it successfully re-
duced hypolimnetic iron concentrations in Lake Bret and
greatly improved water quality.
Air Injection Systems
Partial Air Lift Designs --
Partial air lift designs are those systems where hypo-
limnetic water is aerated and circulated by an air injec-
tion system, but where the air/water mixture does not upwell
to the lake's surface. Instead, the water and air separate
well below the lake's surface. The water is returned to the
hypolimnion and the air is wasted to the atmosphere. Two
partial air lift designs have been described.
The partial air lift designs appear to be much less ef-
ficient than the full air lift designs. Although the partial
air lifts have greater effluent oxygen concentrations, they
aerate less water volume and the total oxygen dissolved is
less than full air lift designs. Most of the energy used
-------
SPLASH
BASIN
WATER PUMP
00
DISCHARGE
INTAKE
Figure B-l. Mechanical aeration system of hypolimnion aeration from
Lake Bret, Switzerland (Mercier and Ferret, 1949; Mercier
and Gay, 1954; Mercier, 1955).
-------
to compress the air is "lost" in the waste air discharge
line, and the air does not expand greatly while it is in
contact with the water.
The Limno System of hypolimnetic aeration is the most
widespread hypolimnetic aerator in present use (Figure B-2).
Five Limno Systems are in European use (Bjork, 1974) , while
one system is in North American use (Fast, et al., 1975a).
The original Limno aerators were constructed of steel, but now
the design is standardized and they are constructed of molded
fiberglass. Each chamber measures only 15 feet high by 8
feet in diameter (28 feet diameter overall with outlet arms).
Compressed air is released from a diffuser at the bottom of
the aerator and rises within the aeration chamber. Upon
reaching the top of the chamber, the air and water separate.
The waste air is vented to the atmosphere through a
small diameter pipe and the water reverses its flow and
returns to the hypolimnion through six outlet pipes. A
valve on the wastes gas discharge pipe maintains pressure
on the waste gas and thus creates a gas "pocket" at the top
of the aeration chamber. This prevents the escape of water
through the waste gas discharge pipe. A shore based air com-
pressor can supply air to one or more submerged aeration
chambers. One air compressor (280 SCFM, 40 H.P-, single
stage) provided air to two Limno aerators at Lake Waccabuc,
New York (Fast, 1973b). This system had an oxygen absorp-
tion efficiency of 1.2 Ibs 02/kw-hr. The water to air ratio
was 3.8:1 and only 10.6% of the injected oxygen was absorbed.
A substantial increase in dissolved nitrogen gas occurred in
the hypolimnion, and after 80 days of continuous aeration the
hypolimnion was 150% saturated with N2 relative to surface
pressures (Fast, et. al. , 1975a) . Hypolimnetic phosphorus
concentrations were apparently not greatly reduced by aera-
tion, even though hypolimnetic oxygen concentrations were
typically greater than 2 mg/£ during the second summer of
aeration (Fast, 1975). Hypolimnetic ammonia concentrations
increased greatly during aeration, and nitrification apparent-
ly was not greatly accelerated by the increased oxygen con-
centrations (Confer, et al., 1974b). There was no apparent
effect of the aeration on algal or zooplankton densities (Fast,
1975; Confer, et al., 1974a). Rainbow trout (Salmo gairdneri
Richardson) survived in Lake Waccabuc with hypolimnetic aer-
ation, whereas they did not prior to aeration.
Registered Trademark of Atlas Copco, AB. A U.S.
patent was allegedly applied for circa 1972, on
this system, but its present status is unknown.
90
-------
WATER OUT
WATER OUT
WEIGHT
AIR LINE
Figure B-2. Exposed view of timno hypolimnetic aerator. This is
a partial air lift design since hypolimnetic waters are
upwelled only a short distance by air injection (Fast,
Dorr and Rosen, 1975a)
91
-------
Speece, et al., (1974) proposed another partial air
lift system (Figure B-3) , but this system has not been tested.
It is essentially the same design as the Limno except that
the waste gas discharge pipe does not have a valve, and a
gas pocket may not be created. Consequently, gas and water
may be pumped through the gas discharge line. Provisions
may have to be made to return this water to the hypolimnion.
Otherwise, substantial quantities of hypolimnetic water may
be pumped into the epilimnion.
Full Air Lift Designs --
Full air lift designs are those systems where compressed
air is injected near the bottom of the aerator and the air/
water mixture rises to near the lake's surface. The air then
separates from the mixture and the water is returned to the
hypolimnion.
Fast (1971) described one of the first full air lift,
hypolimnetic aerators (Figure B-4). This aerator consisted
of concentric upwelling and downwelling pipes. Air was
released within the center pipe and rose to the lake's sur-
face. The efficiency of this system was greatly reduced
because of the restricted outflow, but proposed modifications
to this system should greatly increase its efficiency and re-
duce its capital cost (Fast, et. al. , 1975c) . With this system
in Hemlock Lake, Michigan, hypolimnetic oxygen concentrations
were increased from 0 to over 8 mg/1 and rainbow trout distrib-
uted throughout the hypolimnion (Fast, 1973a). Phytoplankton
and zooplankton densities varied greatly during aeration.
This was due in part to leaks in the aerator which allowed
nutrient rich waters to mix with the epilimnion and thus
promote algal growth (Fast, et al. , 1973).
Bernhardt (1974) described another aerator which incorpor-
ates individual upwelling and downwelling pipes, separated by
a horizontal degassing chamber (Figure B-5).(2' This aerator
was used in Wahnbach Reservoir, West Germany, and has one of
the greatest efficiencies yet reported. Its efficiency was 2.5
Ibs 02/kw-hr, with absorption of 50% of the injected oxygen.
Bernhardt injected 247 SCFM of air at the 120 foot depth in
a 5.5 foot diameter upwelling pipe and observed a wa1?er:air
ratio of 31:1. The diameter of the downwelling pipe was 9.8
feet. This system greatly reduced the dissolution of iron,
manganese and phosphorus from the sediments while at the same
time maintained cold aerated waters in the hypolimnion.
Two similar full air lift hypolimnetic aerators were
used to aerate lakes Tullingesjon and Jarlasjon, Sweden
A U.S. patent was issued in 1974 (Hirshon, 1974) which
appears to embody many of the features of the Bernhardt
design. However, this patent does not reference some of
the more pertinent literature, and it may not withstand
a legal test of its claims.
92
-------
EPILIMNION
o o
-BUBBLE HARVESTER
WATER
OUT
-AIR DIFFUSOR
WATER IN
Figure B-3. A proposed partial air lift system of hypolimnetic
aeration (Speece, et^ al_., 1974)
93
-------
air
water surface
water high
in oxygen
Length
Width
Wt.
Air vol.
60ft.
6ft.
4.5 tons
150cfm
current deflector
air injected at
this point
water low
in oxygen
Figure B-4. A full air lift, hypolimnetic aeration used in
Hemlock Lake, Michigan (Fast, 1971). This is a
full air lift design since water is upwelled to
the surface before it returns to the hypolimnion,
94
-------
Metalimnion
Hypolimnion
Damm
a^ufsteigendes
Luff- Wfassergemisch
Diffusor
Luftzufuhr
Zch Nr 922-2 660
ei B-5. A full air lift, hypolimnetic aerator used
in Wahnbach Reservoir, West Germany (Bernhardt, 1974).
-------
(Bengtsson, et al., 1972). These systems were nearly identi-
cal except for the placement of the upwelling pipe, and
the number of outlet pipes.
The Lake Tulligesjon aerator had one 1.3 foot diameter
upwelling tube which discharged horizontally into the box-like
degassing chamber (Figure B-6). It had four 1.6 foot diameter
outlet pipes which distributed the aerated water throughout the
hypolimnion (Figure B-7). About 530 SCFM of compressed air were
injected. Hypolimnetic oxygen concentrations were not increased
above 0.0 mg/ even though Lake Tullingesjon was aerated for
more than 2-1/2 months. Hypolimnetic ammonia concentrations'
were decreased slightly during aeration, but hypolimnetic nitrate
concentrations were not significantly changed by aeration.
The Lake Jarlasjon aerator differed from the Lake Tulling-
esjon aerator primarily in the placement of the upwelling tube
and the number of outlet tubes. The Lake Jarlasjon aerator had
one 2 foot diameter upwelling pipe which extended through the
bottom of the degassing chamber and discharged "concentrically"
into the chamber (Figure B-8). Ten 1.6 foot diameter outlet
pipes distributed the aerated water back into the hypolimnion
(Figure B-9). About 789 SCFM of air were injected into the
upwelling tube. Lake Jarlasjon was aerated one summer. Hypo-
limnetic ammonia was greatly reduced and nitrification was
apparently increased by aeration. Nitrate concentrations were
more than 1.4 mg/£ with aeration and less than 0.1 mg/£ with-
out, whereas ammonia concentrations were correspondingly less
than 0.1 mq/l with aeration and 1.6 mg/£ without aeration.
Phosphorus was slightly reduced during aeration.
The efficiencies of the Lakes Tullingesjon and Jarlasjon
aeration systems were greatly reduced by the very large air
injection rate and small diameter upwelling pipe. These con-
ditions for the Swedish aerators reduced oxygen absorption
efficiencies and water volumes pumped per unit volume air
injected. The upwelling plume from the Lake Jarlasjon aer-
ator shot about 6 feet above the lake's surface.
Smith, et al., (1975) described a design similar to
Bernhardt's (Figure B-5) (Figure B-10). It is constructed
of inexpensive materials such as plywood, has flexible out-
let tubes and styrofoam flotation. The upwelling pipe con-
tained a patented helical insert which was supposed to increase
oxygen absorption. However, Smith, ejt al. reported that it
may have reduced absorption since the bubbles tended to be
channeled up the center of the helix (Knauer, pers. comm.,
1974; Smith, et a^L. , 1975).
Dg_wnflow Air Injection --
Speece (1970) and Speece, et al. (1974) have proposed
two hypolimnetic aeration devices which use a mechanical water
96
-------
AIR INJECTION
WATER IN
//w%Mww/$^^
Figure B-6. A full air lift, hypolimnetic aerator used in
Lake TuTlingesjb'n, Sweden (Bengtsson, et al., 1972),
97
-------
TULLINGESJON
OUTLET ARMS
AERATOR
Figure B-7. Plan view of Lake lullingesjon, Sweden, showing the
hypo!imnetic aerator with four outlet pipes
(Bengtsson, et al., 1972).
98
-------
AIR
COMPRESSOR
DEGASSING
CHAMBER
Figure B-8. Full air lift, hypolimnetic aerator used at Lake
Jarlasjon, Sweden (Bengtsson, et al_., 1972).
-------
JARLASJON
o
o
-OUTLET ARMS
0
I Km
Figure B-9. Plan view of Lake Jarlasjon, Sweden showing the
hypolimnetic aerator and ten outlet-pipes
(Bengtsson, et al_, , T972)
-------
SEPARATOR BOX
EPILIMNION
THERMOCUNi
HYPOLIMNION
39 FT.
Figure B-10.
Full air lift, hypolimnetic aerator used in Mirror
and Larson Lakes, Wisconsin, by Smith, Knauer and
Wirth, (1975).
101
-------
pump to force a flow of water down a vertical pipe (Figures B-ll,
B-12) . Air is injected below the pump, but the downward velocity
of the water is greater than the rise velocity of the bubble.
Consequently, the air bubbles are swept downward and separa-
tion of the air/water mixture takes place in the hypolimnion.
The advantages of this technique are that the air/water contact
time is increased for a given length pipe, the bubbles are ex-
posed to a continuously increasing oxygen saturation gradient,
and a low pressure air compressor is needed. These factors
tend to increase the percentage of oxygen absorbed from the
air. However, the large water velocity required would also
complicate air/water separation, and additional energy is
required to mechanically pump the water. Furthermore, dis-
solved nitrogen gas concentrations will be greatly increased
by these systems, and unacceptable N2 concentrations may occur.
Speece's first system incorporates both the downflow
air injection and air lift pump features (Figure B-ll) (Speece,
1971). This should increase its efficiency since some of the
energy required to compress the air is recovered as the bub-
bles rise in the upwelling tube.
Speece's later version does not make use of the potential
energy of the waste gases (Figure B-12) (Speece, et al. , 1974).
Instead, the waste gases are vented to shallow water, and hypo-
limnetic water is hopefully confined to the hypolimnion. In
practice, hypolimnetic water may be pumped through the waste gas
vent unless sufficient pressure is maintained therein.
Other Air Lift Designs --
One of the first successful hypolimnetic aerators was
the "standpipe" aerator described by Bernhardt for Wahnbach
Reservoir, West Germany (Figure B-13) (Bernhardt, 1967).
Superficially, this aerator looks like a full air lift design.
However, most of the water does not rise to the top of the
aerator. Instead, it separates from the air at the level
of the outlet arms and returns to the hypolimnion. In
essence, it is nearly identical in concept to the partial
air lift design shown in Figure B-4.
Bernhardt reportedly injected 142 SCFM of air into the
aerator and observed an oxygen absorption of 50%, or an
efficiency of 2.1 Ibs 02/kw-hr. This is only slightly less
efficient than his newer design (2.5 Ibs 02/hr) which has
since replaced this aerator at Wahnbach Reservoir (Bernhardt,
1967). The original aerator reduced the rate of oxygen
depletion during thermal stratification, which in turn
reduced the manganese and phosphorus concentrations in
the hypolimnion.
102
-------
PROPELLER
PUMP
EPILIMNION
THERMOCLINE
HYPOLIMNION
t
SUMP
-INTAKE
PIPE
O O
0° O °
o o
AIR
BLOWER
-DISCHARGE
PIPE
Figure B-ll. A proposed downflow air injection system which also
incorporates the air lift feature (Speece, 1970).
103
-------
MOTOR
BUBBLE
HARVEST
VENT-
O " o c
o o o
o
00°
o °
o o
o ° o
o o
o
o o
-AIR
Figure B-12.
A proposed downflow air injection system (Speece,
et al., 1974).
104
-------
o
en
Air— Wa ter — Mixture
D iffusor
Hypol i m i on
Ai r
4(M
4.'J
Figure B-13 "Stand-pipe" hypolimnetic aerator used at Wahnbach
Reservoir, West Germany (Bernhardt, 1967).
-------
Oxygen Injection Systems
One of the first successful systems of hypolimnetic
oxygenation which uses pure oxygen was the side stream
pumping (SSP) system (Fast, et al. , 1975b).(3) It is
conceptually uncomplicated (Figure B-14) . Water is drawn
from the hypolimnion through an intake pipe by a shore
based water pump. The water passes through the pump and
is returned to the hypolimnion through a high pressure
discharge line. Since nearly pure oxygen and high pressures
are used, the oxygen is almost completely dissolved before
it leaves the discharge pipe. This system was successfully
used at Ottoville Quarry, Ohio, where hypolimnetic oxygen
concentrations were increased from zero to 8 mg/Ji during
1973, and to 21.5 mg/£ during 1974 (Overholtz, 1975).
Hypolimnetic temperatures increased 5°C and 9°C during 1973
and 1974 respectively due to the induced water currents.
The system had a 5 H.P. water pump and maximum oxygen input
capacity of about 50 Ibs/day (Fast, 1973b; Matsch, 1973) .
This indicates an oxygen absorption efficiency of 1.5 Ibs
02/kw-hr, assuming an energy consumption of 8~00/kw-hr/ton of
liquid oxygen (Fast, et al . , 1975c) ; Lyon, personal comm. ,
1975) . As a consequence~of SSP oxygenation, rainbow trout
survived yearlong at Ottoville Quarry (Annon., 1973; Over-
holtz, 1975). This greatly increased recreational use of
the quarry, but special tenchinques were required to stock
the fish during the summer (Overholtz, 1975).
A second SSP test at Attica Reservoir, Attica, New York,
was unsuccessful (Fast, Overholtz and Tubb , .M.S.; Haines,
1974; Matsch, 1973). The SSP system at Attica was larger
and the reservoir was much shallower than at Ottoville.
Although only 1.3% of the hypolimnetic water volume during
1973, and 0.7% during 1974 was pumped through the system
each day, the reservoir was rapidly destratified each year.
In Ottoville Quarry, 1.3% of the hypolimnetic volume was
pumped per day but thermal stratification was maintained.
This indicated that a much reduced water flow rate is
required to maintain thermal stratification in shallow
lakes. The Attica system was designed to dissolve 60 Ibs
of oxygen/day, for an efficiency of 0.17 Ibs O
Speece has proposed a deep oxygen bubble injection
method of hypolimnetic oxygenation (Speece, 1971, 1973,
1975) . This technique involves the injection of high
purity oxygen from coarse bubble diffusers at some depth
within the hypolimnion. If the hypolimnion is deep
enough (e.g., 60 feet of hypolimnetic height), then most
(3) "
A U.S. patent was allegedly applied for during 1974
on this system, but its present status is unknown.
106
-------
o
Figure B-14.
Schematic view of the hypolimnion oxygenation system used at
Ottoville Quarry, Ohio and Attica Reservoir, New York (Fast,
Overholtz and Tubb, 1975b). The cross-hatched area represents
the thermocline.
-------
of the bubble will be dissolved before it reaches the
thermocline. If the gas volume is small, then according
to Speece, the bubbles will "uncouple" through the thermo-
cline to the lake's surface while the oxygenated hypolim-
netic water will remain in the hypolimnion (Figure B-15 and
B-16). The oxygenated hypolimnetic water will partially
mix the warmer thermocline water and form a "sandwich
layer" at the base of the thermocline. The sandwich layer
will spread out from the plume area. If the rate of gas
input is too great, then the upwelled hypolimnetic water
may penetrate the thermocline and excess mixing with shallow
water may occur.
If in fact uncoupling occurs and oxygen can be econom-
ically and reliably produced, then this technique may have
practical application, especially for meeting tailwater
oxygen-temperature discharge requirements. Some form of
pure oxygen injection may be required in these situations
since air injection can in some cases cause unacceptable
dissolved nitrogen gas concentrations. The full extent of
this potential problem is still undocumented. Even with
pure oxygen injection, nitrogen gas can become supersatur-
ated due to hypolimnetic warming (Fast, Overholtz and Tubb,
M.S.) .
Deep oxygen bubble injection may not be feasible if
uncoupling is not complete, and/or if oxygenated water does
not mix throughout the hypolimnion. This situation could
lead to a sandwich layer with high oxygen content, and an
oxygen depleted zone near the lake bottom. On-site gener-
ation of oxygen is feasible in some large volume situations,
especially if the oxygen can be continuously injected with-
out liquification. Importation and/or storage of liquid
oxygen can be prohibitively expensive and unreliable. As
far as we know, deep oxygen bubble injection has not been
used full scale to oxygenate a lake's hypolimnion other
than for discharge purposes.
Speece also proposed the use of downflow bubble-contact
aerator (DBCA) for hypolimnetic oxygenation (Speece, 1970,
1971, 1973; Speece et al., 1974). This cone shaped device
could be suspended within the hypolimnion (Figure B-17).
A pump forces water downward and pure oxygen is diffused
below the pump. Bubble contact time is lengthened by this
device as with the downflow air injection. The high water
velocity in the throat (8 ± 2 ft/sec) prevents gas bubbles
from escaping through the top of the funnel. Instead, the
bubbles are forced under the lower rim after about 15
seconds contact time. By then, much of the oxygen has dif-
fused out of the bubble and into the water, while nitrogen
gas has diffused into the bubble. The waste bubbles then
rise to the surface. This waste stream will entrain water
108
-------
\7
~. . t MI r\j i_i
\ /
\ 1
r / I t
/ / V \
X / \ X^
s \ -
"* ODIFFUS*OR"
mv/w/wmy/www/w/w^w/^
(A) COMPLETE UNCOUPLING BETWEEN
PLUME WATER AND GAS BUBBLES
VI IN 1 WIN 1 •
| 1
| 1
1 1
\ 1
\ ^/
\ / /
\ * /
^r
^9^^W^^/W^
(B) PARTIAL OR
m i
n i
I i
\ '
\* /
\ v /
x /
ODIFFUSOF?
^m*mmm&
NON-UNCOUPLING
BETWEEN PLUME WATER AND
GAS RIIRRI FS
Figure B-15.
Proposed modes of bubble and water plume interactions
during deep oxygen bubble injection (Speece, 1975b).
-------
EPILIMNION
I
""x
\
1
1
HYPOLIMNION [
1
LU
ID
Q_
CC
Ld
^1
^
••.•:•:•:•:•:•:•:•:•:•;(
1 i|
SANDWICH
LAYER
Wi^WW^^
Figure B-16.
Proposed oxygenated water plume behavior during deep
oxygen bubble injections (Speece, 1975b).
110
-------
OXYGEN
EPILIMNION
^v\\\\\V^\\\\\\\\\\\v
THERMOCLINE
x\\\\\\\\\\\\\\\\\\\\\\^^
HYPOLIMNION
DAM
WATER OUT
Figure B-17.
Downflow bubble contact aerator in position for
proposed hypolimnetic oxygenation (Speece, et
aj_. , 1974).
Ill
-------
and cause an upwelling. This upwelling may penetrate the
thermocline in some cases, and therefore be unacceptable.
As far as we know, the DBCA has not been used for full scale
oxygenation of a lake's hypolimnion.
Whipple (1975) has designed and tested another oxygen
injection system (Figure B-18). His system is constructed
of wood and is suspended just above the lake's bottom by a
raft. He tested it at Spruce Run Reservoir, Clinton, New
Jersey, during 1973-74. it has many features in common with
the partial air lift designs (Figures B-2 and B-3) since
hypolimnetic water is circulated by the "air lift" principle
and the water is not upwelled to the surface. According to
Whipple, the aerator did not function as well as he had
expected. This may have been due in part to the equipment
design, and in part to the experimental design. The experi-
mental design involved an attempt to oxygenate only one
hypolimnetic arm of the reservoir. Experience elsewhere
indicates that this may not be possible since the water cur-
rents generated by the oxygenation equipment will tend to
distribute the oxygenated water throughout the hypolimnion.
SeppSnen (1974) described the design and operation of
Isteri's aerator in lakes Hemtrask and Kiteenjarni, Fin-
land (Figure B-19). This device was used to oxygenate
lake waters during ice cover . Seppanen intended to use the
aerator for hypolimnetic oxygenation as well. It is similar
in concept to downflow air injector systems previously
discussed. Water is withdrawn from the hypolimnion and dis-
charged to the hypolimnion at high velocities by a mechan-
ical water pump. Oxygen is injected above the pump and
the water forces the bubbles into the hypolimnion. Seppanen
reported a problem with a gas pocket forming at the top of
the tube's arch. This may have been due to bubble coalescence
within the downflow tube and the upwelling of these large
bubbles within the tube.
112
-------
OXYGEN
DEFICIENT
WATER
OXYGEN
RICH WATER
Figure B-18. Hypolimnetic oxygenation system used at Spruce Run
Reservoir, New Jersey (Whipple, 1975).
-------
PUMP MOTOR
OUTLET
Figure B-19.
Isteri oxygen injection system used to oxygenate
under the ice in Lakes Hemtrask and Kiteenjarni,
Finland (Seppanen, 1974).
114
-------
SECTION VIII
APPENDIX C
AIR SUPPLY LINE AND COMPRESSOR SELECTION
Following determination of total air requirements and
diffuser design (discussed in Chapters II and III), it is
necessary to design the supply line and select an appropriate
compressor.
Design Parameters
Air pressure requirements depend on depth of water at
which the air is released, length and diameter of the air
line between the compressor and the point of injection,
type of diffuser used and type of filter used to remove
oil from the air. Of these conditions, the depth of air
injection is usually the most important. It is highly de-
sirable to inject the air as deep as possible to maximize
the water flow and assure oxygen saturation at the surface.
The pressure required to deliver air from the compressor
to the diffuser depends on the depth of release and losses
in the supply system. The system should be designed to pro-
vide from 5 to 10 psi at the diffuser. The number of dif-
fusers or hole spacing can be adjusted to provide the
required air flow rate and a uniform distribution of bubbles.
Head losses in the air supply line are a function of
pipe material and diameter, flow rate and operating pres-
sures. The pressure loss is directly proportional to the
square of the flow rate and inversely proportional to
the fifth power of the pipe diameter. Nomographs for
computation of pressure changes in compressible flow lines
are available in "Flow of Fluids through Valves, Fittings
and Pipes, Technical Paper 410, CRANE CO., 4100 S. Kedzie
Avenue, Chicago, Illinois, 60632. As an example, for air
flowing through a 2 inch nominal diameter schedule 40 steel
pipe at 100 psi and 60°F the pressure drop (h£) per 100 feet
as a function of air flow is shown below:
Free Air Flow
(SCFM)
50
100
200
300
400
h£/100 ft
(psi)
.019
.070
.264
.573
1.00
115
-------
Free Air Flow
(SCFM)
500
600
700
800
900
1000
h£/100 ft
(psi)
1.
2.
3.
3.
4.
6.
55
21
00
90
91
06
The gauge pressure required at the compressor discharge
(following any filter) is given by:
P . , = AP + h0 + Depth (1^:7) (C-l)
required & ^ 33 ^ -1'
where: AP = pressure drop across orifice
h = pressure drop in supply system
x/
The pressure required should be computed for a number
of pipe sizes so that alternative compressor costs can
be compared to pipe costs.
It is important not to use a compressor which far
exceeds the pressure required to inject the air. Too
often a double stage air compressor with a working pres-
sure in excess of 100 psi is used to inject air at a 50
foot depth (22 psi) or less. This wastes energy and
greatly increases operating costs. For depths less than
50 feet a single stage reciprocating air compressor is
usually most satisfactory (working pressure of 50 psi
or less). Depths greater than 75 feet usually require
a double stage reciprocating air compressor. The dis-
charge pressure should be selected to produce maximum
efficiency.
Figure C-l shows the approximate horsepower required
per standard cubic foot per minute of air delivered at
various pressures. These data were extracted from Atlas
Copco brochure 5SC1-1273 for a variety of compressors.
It should be noted that reciprocating compressors are more
amenable to adjustment of pressure and flow rate than are
rotary type compressors. For example, the Atlas Copco
BE23-1160 can be adjusted to operate at from 20 to 50 psi
with free air flow rates ranging from 101 to 119 SCFM.
The horsepower required per SCFM of air at the different
pressures is shown in Figure C-l. For pressures less than
20 psi, commercial air blowers may be more efficient.
116
-------
0.0
0
25 50 75
DISCHARGE PRESSURE (psi)
100
Selection and Installation
Air Line --
The sizing, choice of materials, methods of assembly
and installation are all important considerations. They
all affect capital costs, operating costs and problems
such as line breaks. The proper design, construction and
installation of the air line extending from the air com-
pressor into the lake will greatly reduce future problems
and increase operating efficiencies. Some of the consider-
ations are discussed below.
Small diameter, light weight lines should be avoided on
lakes with recreational boating. Some of these lines can
be hooked by boat anchors, pulled to the surface, dislocated
and even broken. A variety of pipe line materials may be
used, such as steel, polyethylene and PVC plastic. Each
material has its advantages and disadvantages.
Generally, it is advisable to use galvanized steel pipe
from the compressor to the water. Unless this section is
buried or covered, it will be most exposed to vandalism and
liability problems. If the compressor does not have an
aftercooler for the compressed air, then this section of
pipeline must be heat resistant. Discharge temperatures
can exceed 200°F and will melt some plastics.
117
-------
If the pipeline is on an incline, it should be adequately
anchored and braced near the compressor, but not on the com-
pressor discharge port. A flexible connection should be pro-
vided between the discharge port and the pipe. Steel air
lines are often used in the lake as well as from the compres-
sor to the lake. However, steel pipe is generally not the best
choice of material. Steel pipe is expensive, difficult to
install underwater and subject to breakage on an irregular
lake bottom. It is very difficult to repair if a break does
occur. Repairs may require the use of divers or lifting
the pipe back to the surface. The inflexibility of the
steel, and its weight are primarily responsible for its
tendency to break either during installation or after it is
on the bottom. This problem can be partially overcome by
using flexible joints between each length of pipe.
Plastic pipe is generally the best choice for underwater
use. It is cheaper, lighter and much easier to install.
PVC plastic pipe has been widely used. It comes in lengths
of 20 feet or more, which can be glued together at the site.
However, it is subject to breakage. Flexible, polyethylene
plastic pipe is probably the best choice for most underwater
air lines since it comes in much longer lengths. It can be
purchased on large diameter spools. The polyethylene pipe
is more rigid when cold, and it may be necessary to discharge
heated air (e.g. from a compressor) or hot water through
the pipe in order to unwind it from the spool.
The installation procedures will depend on a variety of
factors including the pipe material, diameter, weight and
whether it will be suspended at the lake's surface or sub-
merged and anchored at the bottom. If the lake is closed
to the public, at least in the area used by the aeration
system, it may be possible to simply suspend the air line
by floats at the surface, extending from the shore to the
injection location. This may block some boat traffic, but
it is the easiest and cheapest installation. The pipe is
accessible for repairs or modifications, and it caa be
easily removed. However, in many cases this arrangement
is undesirable or impossible and the line must be submerged.
If a light weight, flexible plastic pipe is used the
first step in installation is to assemble or unwind it
from a spool, and stretch it out on the surface of the
lake from the shoreline to past the point of air injection.
This profile can be measured by stretching a rope from
the shore at the lake's surface to a float in the lake.
Soundings can be made along the rope with a weighted line,
or recording sonar.
118
-------
After the air line is stretched out on the lake's sur-
face and securely attached at the shoreline, weights or
anchors can be added to sink it to the bottom. The weights
should be added starting at the shore and working into deep-
er water. If the end of the air line is attached to a
hypolimnetic aerator or some other device, then it is desirable
to use a flexible connection. A flexible rubber pressure
hose works well. This will allow for some shifting of the
device without putting excessive stress on the connection.
Compressor --
After determination of required air flow rates and
pressures, it is recommended that a number of compressor
manufacturers be contacted to obtain detailed information
on price and performance characteristics.
119
-------
SECTION IX
APPENDIX D
HYPOLIMNION AERATOR SIZING
The following analysis provides a simplified method to
estimate air requirements and pipe diameters for simple
hypolimnion aerators. The analysis is based on the assump-
tion that the theoretical head available results from the
difference in density between the air-water mixture in the
pipe and the outside water. It is further assumed that one-
half the theoretical head would be used to pump water up
to the surface and an equal amount would be used to pump
the aerated water back down to the hypolimnion.
Figure D-l is a definition sketch of the system. The
theoretical lift, AH is obtained by solving the equation:
* rw L
(D-l)
AH
m
where: r
n
L
r
w
rwL
density of air-water mixture
depth of air release
density of water
by:
The average density of the air-water mixture is given
Q r + Q r
= w w a a
Q,., + Q,
(D-2)
where: Q = water flow rate
w
Qa = volumetric air flow rate
Qa = mean volumetric air flow rate
120
-------
* ^ii<' * _**V*'-•^•---A.
<—£->
./ ^—^ \
Qair
Q
w
Figure D-l. Definition sketch for air induced flow.
121
-------
r (Q + Q )
AH = W W a L - L (D_3)
^w rw + Qa ra
for small ratios of air flow rates compared to water flow rates
AH . wL. L . L - L
Q r gw
^W W
The mean volumetric air flow rate, Q can be shown to equal
a
L + 34.
Qa 34 in 34 (D_5)
where Q, = free air delivery, in SCFM.
a
therefore:
(L + 34.
34 Q In l 34 ;
AH - ~^
Neglecting the effect of the air bubbles interspersed
with the water (probably valid for low airrwater ratios)
there are three major head loss terms.
1. exit loss, h
L
2g \A / X 2? = TTZv?-
122
-------
where: v = velocity (fps)
9
g = acceleration of gravity (ft./sec )
>•>
A = cross-sectional area of pipe (ft )
D = pipe diameter (ft)
9
2. friction loss, h,
2 . Ik x Y! . fL x « - « (D-8,
L D ^ D ~
3. entrance loss, h.
3 2 K x 8Q
hL = KL X =
IT D g
where: K. - entrance constant
From our initial assumption:
AH = 2 x (h.1 + h2 + h?) or
L L L
(L±Ji) 2
34Q In k 34 ; 2 x 8Q^ f|
x
rearranging to isolate Q and Q :
_
Qa
/L + 34 x 2 4 . (L + 34)
34 In1 34 ' Wg = 673 D4 In l 34 ' (D-ll)
123
-------
This relationship was used to prepare Figure D-2 which
shows the ratio Q^/Qa for various pipe diameters and lengths.
A friction factor f, of 0.02 and K value equal to 0.5 were
assumed for this plot.
124
-------
8
cc.
LLl
I
L=2OOFT.
L-IOOFT.
L=4OFT.
o
5 6
DIAMETER, D (ft.)
Figure D-2. Ratio of water flow to air release for
different pipe diameters and depth of release.
125
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-77-004
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
A Guide to Aeration/Circulation Techniques for
Lake Management
5. REPORT DATE
January 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Marc Lorenzen and Arlo Fast
8. PERFORMING ORGANIZATION REPO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Tetra Tech, Inc.
Lafayette, California 94549
10. PROGRAM ELEMENT NO.
1BA031
11. CONTRACT/GRANT NO.
68-03-2192
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Corvallis Environmental Research Laboratory
U.S. Environmental Protection Agency
200 S.W. 35th Street
Corvallis, Oregon 97330
final
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
The application of aeration/circulation techniques to lakes are reviewed
from a theoretical and practical viewpoint. The effect of destratification
on algal production is related to the mixed depth with the use of a mathematical
model. Procedures are given to determine air required to mix lakes of different
sizes and shapes. It was found that approximately 30 scfm of air per 10^ square
feet of lake surface area can be used as a rule of thumb.
Hypolimnetic aeration systems that have been used are described in
detail. Procedures for design are given.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Lake Management
Lake Respiration
Eutrophication
Aeration/Circulation
Destratification
Algal Bloom Control
8H
S. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
142
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
126
U 5. GOVERNMENT PRINTING OFFICE 1976— 796-460 / 28 REGION 10
------- |