REPORT TO CONGRESS
DAM WATER QUALITY STUDY
Prepared by:
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Water
Office of Water Regulations and Standards
Assessment and Watershed Protection Division
401 M Street, S.W.
Washington, D.C. 20460
March 1989
U.S. Environmental Protection Agency
Region 5, Library (PL-12J)
77 West Jackson Boulevard, 12th Flow
Chicago, IL 60604-3590
-------
ACKNOWLEDGEMENTS
EPA acknowledges the following organizations which provided input in
the preparation of this report:
American Rivers
Bonneville Power Administration
Department of Water Resources - State of California
Electric Power Research Institute
Federal Energy Regulatory Commission
Missouri Department of Conservation
National Wildlife Federation
New Mexico Health and Environment Department
Oak Ridge National Laboratory
Salt River Project
Tennessee Division of Water
Tennessee Valley Authority
U.S. Army - Corps of Engineers
U.S. Army - Engineers Waterways Experiment Station
U.S. Bureau of Reclamation
U.S. Department of Interior
U.S. Environmental Protection Agency Regional Offices
U.S. Fish and Wildlife Service
U.S. Soil Conservation Service - USDA
Utilities Water Act Group - Hunton & Williams
Western States Water Council
EPA especially appreciates the efforts of U.S. Bureau of
Reclamation, Tennessee Valley Authority, U.S. Army Corps of Engineers,
Soil Conservation Service, Utilities Water Act Group, Oak Ridge National
Laboratory, and Electric Power Research Institute, who provided very
comprehensive and useful comments on the report.
Contract support was provided by GKY and Associates, Inc., under
subcontract with Versar, Inc. through Contract No. 68-03-3339 with EPA's
Monitoring and Data Support Division.
-------
EXECUTIVE SUfflARY
The objective of this report is to identify water quality effects
attributable to the impoundment of water by dams as required by Section
524 of the Water Quality Act of 1987. This document presents a study of
water quality effects associated with impoundments in the U.S.A.
This Executive Summary provides the general observations of the
report followed by summaries of each of the six major report elements.
First, a generic description of water quality effects of dams based on
existing knowledge is presented. Four types of analyses are then used to
attempt to define the occurrence and magnitude of water quality effects
within and downstream of impoundments: a mixing analysis; an analysis of
dissolved oxygen concentrations in power dam tailwaters; a comparison of
upstream versus downstream concentrations of selected indicator
parameters; and an investigation of phosphorous enrichment. Case studies
are presented to illustrate some of the water quality effects and
mitigation measures. Next, a generic discussion of mitigation measures
for addressing some of the water quality effects is presented.
Assessments of water quality conditions at U.S. Army Corps of Engineers,
Tennessee Valley Authority, and the U.S. Bureau of Reclamation
impoundments are provided as a supplement to these analyses. Finally,
conclusions on impoundment effects on water quality are presented.
There are a large number of dams throughout the United States. The
Soil Conservation Service estimates there are over 2,000,000 dams
including farm ponds and recreational impoundments. The Corps of
Engineers (U.S. Army Corps of Engineers, 1982b) inventoried 68,155 larger
dams meeting minimum size criteria. These 68,155 dams are the basis of
the Environmental Protection Agency (EPA) analyses in this study. They
can be divided into three categories: large power dams (424), large
nonpower dams (1,701), and small dams (66,030). The large power dams
contain 62 percent of the total volume of water normally stored by dams;
large nonpower and small dams contain 33 and 5 percent of the total
volume, respectively. Dams have a variety of purposes which include
hydropower generation (including pump storage), navigation, flood
control, water supply, conservation, recreation, fish and wildlife
maintenance, and low flow augmentation.
This study is limited in estimating the national extent of dam water
quality primarily because of a lack of monitoring and descriptive data.
The STOrage and RETrieval data base (STORET) was used as the primary
source of monitoring data. Although quite extensive, data were not
available for many of the sites randomly selected for analysis. Other
descriptive data, such as the type of outlet structure, watershed land
ii
-------
use, and other influences on water quality, were also not available for
this study. Additional monitoring data, descriptive data, and a larger
random sample of dams would probably extend the study's findings. This
study is also limited in that it identifies water quality effects from the
impoundment of waters by dams, but does not address the effects on
biological habitat or wetlands, which may be substantial.
WATER QUALITY EFFECTS OF IMPOUHDMENTS
Impoundment of free-flowing water by dams may potentially create
several effects, both positive and negative, on water quality within the
pool and downstream. Although this report focuses on unwanted effects,
desirable changes, such as a reduced sediment load, may also result. The
potential effects are often interdependent. Altering one condition in an
impoundment may create a ripple of effects throughout the reservoir-
stream ecosystem.
Impoundments can modify the physical, chemical, and biological
characteristics of the free-flowing aquatic ecosystem. Physical and
chemical characteristics in impoundments are also related to depth,
volume, climate, watershed land use, geographic location, reservoir
siting, and the schedule of water releases. Biological characteristics
are related to the type of habitat. The magnitude of effect of the dam
on water quality of releases appears related to the type of reservoir and
to the design and operation of the impoundment.
Effects can be divided into three categories: stratification-
related effects, eutrophication, and other changes. Stratification, a
naturally occurring condition, results when warmer waters overlie cooler,
denser waters. Deeper impoundments with poor mixing tend to stratify.
Stratification water quality effects may include:
low dissolved oxygen in the hypolimnion (bottom waters);
increased dissolved iron and manganese concentrations;
hydrogen sulfide production;
production of nitrous gas; and
changes in water temperature.
Eutrophication is a naturally occurring process whece excess
nutrients (especially nitrogen and phosphorus) from watersheds flow into
an impoundment. These excess nutrients cause increased and sometimes
undesirable growth of algae and rooted plants. This situation, coupled
with stratification, may result in the depletion of dissolved oxygen in
the bottom layer of the impoundment and the release of soluble iron,
soluble manganese, and hydrogen sulfide.
n
-------
When there are stratification-related effects, possibly compounded
with eutrophication, impoundments with outlets accepting withdrawals from
lower impoundment depths transmit the water quality of the pool's lower
levels into the tail waters. Large power dams are more likely to have
low-level outlets than are small and large nonpower dams because project
purposes influence the type of outlet.
Other water quality effects of dams are generally considered to be
less predominant than eutrophication and stratification-related effects.
Other effects include gaseous supersaturation, salinity changes, sediment
deposition, sediment movement, flow regulation, reaeration denial (the
restriction of natural reaeration processes), fish entrainment (the
capture and passage of fish through turbine machinery), and toxics
accumulation. Downstream transport of toxic-bearing sediments from the
pool area may occur if sediments are disturbed by dredging, dam removal,
or dam failure.
RESULTS OF EPA ANALYSES
Information on small and large nonpower impoundments is limited, and
no quantitative conclusions are reached through the analyses of effects
of these two categories of dams, with the exception of the mixing
analysis for large nonpower impoundments. Data obtained for this study
on these two categories are presented in the appendices. Quantitative
findings, limited to a few broad conclusions for large power
impoundments, are presented below:
The mixing analysis using Froude numbers indicates that
stratification conditions ("poor mixing") are estimated to
occur in 40 percent of the large power impoundments and 37
percent in the large nonpower impoundments. The Froude number,
which considers only the kinetic and potential energy provided
to the flowing waters by gravity, is a limited predictor of
stratification. The influence of low-level releases on these
results is unknown.
Dissolved oxygen levels in tail waters in the dams of the Oak
Ridge National Laboratory study (Cada, et al., 1983) and the
dams in this study are similar. The data are not intended to
quantify the dissolved oxygen findings on a national basis, but
certain relationships are observed in the sample:
Dissolved oxygen in power dam tailwaters during the summer
has a much greater probability of not meeting a criterion
of 5 mg/1 than during winter.
Larger power-generating facilities show greater
probability of not meeting a dissolved oxygen criterion
than do smaller power-generating facilities.
IV
-------
Dissolved oxygen in downstream water showed a decrease in 22 to
50 percent of the waters below large power dams, based on a
comparison of upstream and downstream dissolved oxygen levels
of 40 large power dams. Similarly, 15 percent to 42 percent
showed an increase in dissolved oxygen, while 35 percent to 62
percent showed no change.
Large power impoundments are likely to experience phosphorous
enrichment, an indicator of potential eutrophication. The
sample for large power impoundments showed such potential in 58
percent to 78 percent of the sample. However, phosphorus
enrichment is only one of several factors, including climate
and the presence of other nutrients, resulting in
eutrophication.
CASE STUDIES
Case studies provide a detailed examination of several of the water
quality effects and mitigation measures presented in this report. The
fifteen case studies are not representative of all impoundments in the
nation. They do, however, represent impoundments specifically studied
because they exhibit, or are thought to exhibit, certain water quality
effects. In order to maintain perspective in the discussion of case
studies, one of the case studies included is reported to have no
undesirable effects. The case studies describe, in most situations,
efforts undertaken by operating Agencies to mitigate undesirable water
quality conditions within the operating and legal constraints imposed at
the time the dams are authorized and constructed. The fourteen case
studies showing adverse water quality effects exhibit one or more of the
following: low hypolimnetic dissolved oxygen, increased iron and
manganese, eutrophication, hydrogen sulfide, sediment movement, flow
regulation, thermal changes, and reaeration denial. All of the case
studies that exhibit increased iron and manganese, eutrophication,
thermal change, and hydrogen sulfide have low hypolimnetic dissolved
oxygen. Mitigation measures used to improve water quality in the pool
are typically reaeration or destratification by pumping and/or air
injection. Mitigation of water quality downstream of the impoundments is
typically selective withdrawals and turbine aeration.
MITIGATION MEASURES
This report identifies major mitigation measures (mainly oriented to
addressing tailwater improvements below power dams) that can address the
adverse water quality effects associated with certain impoundments.
Because each impoundment system is unique, the applicability of specific
mitigation measures must be evaluated on a case-by-case basis. This
evaluation must consider the adverse effects that require correction, the
present impoundment purposes, the measure cost, and the undesirable side
-------
effects of the measure itself. Oftentimes, multiple measures may be
necessary. Mitigation measures can be divided into three broad
categories: physical measures, operational measures, and structural
modification.
Physical measures include technologies that require specific
processes or equipment to be used to correct the problem. Physical
mitigation measures include the control of water quality in the
reservoir, selective withdrawal of reservoir water with acceptable water
quality, aeration of reservoir releases, and habitat modification.
Operational measures include changes to the present operating regime
of the reservoir modification. These include maintaining a minimum
discharge, limiting the maximum discharge, and altering the rule curves
for reservoir operations.
Structural modifications involve changes to the structure of the dam
and/or its outlet works; examples are the addition of ports, gates,
vents, or weirs to modify the depth or manner in which water is withdrawn
from the reservoir.
Mitigation measures can be applied to the pool, tailwaters, and/or
to the sources of runoff to the impoundment. Three pool mitigative
measures are induced mixing, aeration of the bottom layer of a stratified
impoundment, and dredging applied directly to the impoundment pool.
Induced mixing usually is the pumping of surface waters downward or
pumping of bottom waters upward, and is intended to reduce stratification
effects. Aeration through injection of air or oxygen raises dissolved
oxygen concentrations, and may induce mixing. Dredging, the physical
removal of sediment deposits, may lengthen the useful life of an
impoundment, but may result in short-term water quality problems through
the resuspension of nutrients and contaminants.
Three mitigative measures applied to tailwaters are aeration of
reservoir releases, selective withdrawal of reservoir water, and
improvement of tailwater habitat. Aeration of reservoir releases may be
achieved by turbine venting, air injection, and/or cascaded tailwaters.
Selective withdrawal is the ability to choose a withdrawal depth with
appropriate dissolved oxygen levels and temperatures. Habitat
improvement provides minimum flows and D.O. levels required to support
target fish populations in downstream pools.
Other mitigative measures involve watershed management and changes
in dam operations. Watershed management addresses reducing the nutrient
and contaminant sources (nonpoint source pollution) in the watershed of
an impoundment. One change in dam operations is to maintain a minimum
constant discharge where zero discharge periodically occurs. This helps
maintain a minimum flow to avoid rapid temperature fluctuations, reduce
the impact of low dissolved oxygen concentrations through natural
aeration (unless the minimum release itself has a very low dissolved
oxygen concentration), and increase habitats for fish and benthic biota.
A second change in dam operation is to limit discharges to a certain
vi
-------
maximum flow, reducing impacts on dissolved oxygen during periods of low
concentrations in the discharges through natural aeration in the
tailwaters.
FEDERAL AGENCY ASSESSMENTS
The U.S. Army Corps of Engineers (COE), the Tennessee Valley
Authority (TVA), and the U.S. Bureau of Reclamation (USBR) provided
supplementary information to this report on water quality conditions at
their dams. These three agencies requested that they be allowed to
contribute specific assessments of the dams they manage. None of the
other commenters offered similar assessments. Information provided by
each Agency includes:
Statement of policies and procedures followed by these
Agencies in the development and management of water resources.
Assessment of water quality with respect to the Agency's dams.
Dams managed by these three Agencies represent a broad range of
geography, climate, and operational situations. COE dams are typically
multipurpose, and may include flood control, navigation, hydropower, water
supply, water quality, recreation, and fish and wildlife enhancement. The
COE dams are concentrated along mainstem navigable rivers, coastal areas,
industrialized areas of the Southeast and the Ohio River Basin, and the
Pacific Northwest. These dams contain approximately 34 percent of the
total volume of water normally stored by all 68,155 dams. TVA operates
reservoirs in the Tennessee Valley primarily for purposes of navigation,
flood control, and electrical generation. The TVA reservoirs represent
mature reservoirs in a well developed and somewhat industrialized extended
river basin. Normal storage volume at TVA reservoirs is approximately 2
percent of the total volume. USBR reservoirs are primarily in the western
states, typically in arid areas. The USBR projects are often
multipurpose, and include: water supply, hydropower, irrigation, water
quality, flood control, river regulation, recreation, and fish and
wildlife enhancement. USBR reservoirs, at normal pool volumes, account
for approximately 22 percent of the total volume. The remaining 42
percent of normal pool volume is impounded behind dams owned by other
governmental entities or privately.
Information for the Agencies' assessments was obtained through a
questionnaire designed to collect information on project design,
operation, and water quality status. The questionnaires were completed
by agency personnel familiar with each dam. Subjective responses were
requested in regard to water quality problems of pool waters and
tailwaters. Where problems were indicated, the impact of each problem
upon user benefits is estimated and a rating assigned. User benefits vary
by individual project. Analysis of the results is limited to frequency
of occurrence of specific water quality conditions and their impact upon
user benefits.
vii
-------
Water quality problems at the Agencies' impoundments vary in
frequency and in degree of impact. Overall, physical conditions, such as
fluctuating pool and tailwater levels and high and low flows, appear to
be the primary conditions affecting user benefits. Data are lacking for
many chemical water quality parameters. For the TVA system, chemical
concerns are usually more serious than physical concerns, even though the
latter are not prevalent. Eutrophication and related water quality
conditions (e.g., algae, high nutrient levels, and low dissolved oxygen)
are noted in many reservoirs.
The COE survey evaluates 46 of their 700 impoundments.
Approximately half of the 46 projects have water quality data. Tailwater
problems are identified in 35 to 40 percent of the samples with data.
Flow fluctuation and high and low flows are the key problems. Pool water
problems in 40 to 50 percent of the samples with data are identified as
eutrophication and related problems (high nutrients, low dissolved
oxygen, algal blooms, and macrophytes).
The TVA survey evaluates all 33 of their large impoundments. Data
were available for dissolved oxygen, temperature, flow, pool levels, and
macrophytes at all projects. Data for the other parameters were
available for an average of 60 percent of the projects. Tailwaters
experienced problems with low dissolved oxygen, flow fluctuations, high
and low flows, and low temperature at 40 percent of their dams.
Significant pool water problems include level fluctuations at 50 percent
of the projects, bacteria at 30 percent, and turbidity, algae,
macrophytes, and sediment at 15 to 20 percent.
The USBR survey evaluates 250 of their 349 impoundments (41 percent
of the tail waters and 46 percent of the pool water sampled have water
quality data). High flow is the primary problem in tailwaters, affecting
21 percent of those with data. Drawdown and pool level fluctuations are
identified as the main impact-producing conditions in reservoir pools,
occurring at 36 percent and 35 percent, respectively.
Because of the lack of data on the majority of dams, the
questionnaire results do not present a complete picture of water quality
for each Agency's impoundments. Also, it is likely that existing water
quality data were only collected at projects with known or suspected
problems; the resulting picture may, therefore, be skewed toward
conditions at which problems are perceived. The data are insufficient to
support specific conclusions applicable to all dams; however, preliminary
evaluations indicate that although project operation plays a significant
role in determining water quality of reservoir releases, there are
pronounced regional patterns of water quality conditions associated with
dams. Regional attributes, such as geology and climate, together with
watershed processes and land use, play a major role in reservoir water
quality. The Agencies feel the limited analysis presented herein gives
an accurate picture of the known extent of given water quality conditions
vm
-------
across a broad range of geography, climate, and project operating
criteria, along with an assessment of the perceived impacts of these
conditions on user benefits.
CONCLUSIONS
The overall conclusions for this study are based on both the agency
assessments and the EPA analytical results. Impoundment of free-flowing
water by dams may potentially create several effects on water quality.
Effects can be divided into three categories: stratification-related
effects, eutrophication, and other changes, such as gaseous
supersaturation. Dam outlets at low levels transmit the water quality of
the pool's lower levels into the tailwaters. Poorly mixed or stratified
impoundments with low-level outlets (inhibiting reaeration) are likely to
exhibit low levels of dissolved oxygen and increased levels of reduced
iron and manganese concentrations in the tailwaters. Furthermore, dams
that create impoundments with long detention times have the potential for
nutrient enrichment when the upstream runoff includes significant
nutrients from point or nonpoint sources. Nutrient-rich impoundment
waters are an indicator of potential excessive eutrophication.
The results of the four EPA analyses conducted for this study
(mixing, tailwater dissolved oxygen, upstream/downstream comparison of
parameters, and phosphorous enrichment) cannot be directly related to the
findings of the COE, TVA, and USBR assessments due to differences in
analytical methods. However, a few complementary findings are noted:
A decrease in dissolved oxygen from the upstream to downstream
was found in 22 to 50 percent of large power impoundments in
the EPA analysis.
According to the other agency assessments, 20 percent of COE's
projects, 38 percent of TVA's projects, and 4 percent of USBR's
projects experience low dissolved oxygen in tailwaters.
Low dissolved oxygen occurs more frequently in eastern dams,
particularly in southeastern dams.
Phosphorous, a potential indicator of eutrophication, occurred
at levels above a guidance of 0.025 mg/1 in 58 to 78 percent of
large power impoundments in the EPA analysis. High nutrient
levels (presumably a mix of nitrogen and phosphorous) were
reported in 35 percent of COE's pools, 30 percent'T)f TVA's
pools, and 15 percent of USBR's pools in the other agency
assessments.
The study identifies water quality mitigation methods that can be
evaluated on a case-by-case basis:
induced mixing of the impoundment pool;
aeration of the bottom layer of a stratified impoundment and
aeration of impoundment releases;
ix
-------
dredging to remove sediment deposits;
selective withdrawal to provide a choice of withdrawal depth;
habitat improvement of downstream pools to support desired fish
populations;
watershed management to reduce upstream nutrient and
contaminant sources that drain to an impoundment;
constraining reservoir releases to maintain target minimums or
to be less than target maximums.
-------
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS i
EXECUTIVE SUMMARY i i
WATER QUALITY EFFECTS OF IMPOUNDMENTS i i i
RESULTS OF EPA ANALYSES iv
CASE STUDIES v
MITIGATION MEASURES v
FEDERAL AGENCY ASSESSMENTS vii
CONCLUSIONS ix
TABLE OF CONTENTS xi
LIST OF TABLES xvi
LIST OF FIGURES xvi i
I. INTRODUCTION 1-1
OBJECTIVES AND HISTORICAL CONTEXT 1-1
PURPOSES AND NUMBERS OF IMPOUNDMENTS 1-4
ANALYTICAL DEVELOPMENT 1-6
OVERVIEW OF REPORT CONTENTS 1-12
II. WATER QUALITY EFFECTS OF IMPOUNDMENTS II-l
STRATIFICATION II-2
Low Hypolimnetic Dissolved Oxygen (DO) II-3
Increased Iron and Manganese 11-3
Hydrogen Sulfide Production 11-4
Denitrif ication II-4
Thermal Changes 11 -4
EUTROPHICATION ~... 11-5
OTHER CHANGES II-6
Gaseous Supersaturation II-6
Salt Concentrations II-6
Sediment Movement 11 -7
Flow Regulation II-7
Reaeration Denial II-8
Fish Entrainment II-8
Toxics Accumulation II-9
xi
-------
TABLE OF CONTENTS (Continued)
III. EPA ANALYSES III-l
OBJECTIVES III-l
PRELIMINARIES 111-2
The Population and Associated Data III-3
Sampling 111-4
Obtaining Sample Dams 111-6
Other Logical Checks and Data Sources 111-6
MIXING ANALYSIS III-7
Relationship of Froude Number to Mixing III-7
Findings III-9
DISSOLVED OXYGEN IN DAM TAILWATERS 111-11
Oak Ridge National Laboratory Study III-ll
Fi ndi ngs 111 -12
UPSTREAM/DOWNSTREAM COMPARISON OF WATER QUALITY 111-16
Acquiring Water Quality Data 111-16
Limitations of the Water Quality Data 111-17
Statistical Comparison of Means 111-18
Findings 111-19
PHOSPHORUS ENRICHMENT ANALYSIS 111-20
Relationship of Phosphorus Enrichment and
Eutrophication 111-21
TVA and CE Results 111-24
Phosphorus Retention Regression Model 111-25
Findings II1-26
IV. CASE STUDIES IV-1
INTRODUCTION IV-1
LOW DISSOLVED OXYGEN AND/OR INCREASED
IRON AND MANGANESE IV-3
J. Percy Priest Lake and Dam IV-3
Old Hickory Lake, Lock, and Dam IV-10
Richard B. Russell Lake and Dam ._. .IV-11
Upper Bear Creek Reservoir and Dam .IV-13
Casitas Lake and Dam IV-15
Lake Cachuma and Bradbury Dam IV-16
EUTROPHICATION IV-17
Guntersvi 11 e Reservoi r and Dam IV-17
Boone Reservoi r and Dam IV-18
Eau Gal 1 e Lake and Dam IV-20
xn
-------
TABLE OF CONTENTS (Continued)
FLOW REGULATION/REAERATION DENIAL IV-21
Norris Reservoir and Dam IV-21
Mark Twain Lake and Clarence Cannon Dam IV-23
Fort Patrick Henry Reservoir and Dam IV-23
SEDIMENT MOVEMENT IV-25
Lake Red Rock and Dam IV-25
THERMAL CHANGES IV-26
Flaming Gorge Reservoir and Dam IV-26
NEUTRAL OR POSITIVE EFFECTS IV-28
McCloud Reservoi r and Dam IV-28
SUMMARY IV-29
V. MITIGATION MEASURES V-l
WATER QUALITY CONTROL IN THE RESERVOIR V-2
Induced Mixing V-2
Aeration of the Hypolimnion V-3
Dredging V-4
WATER QUALITY CONTROL OF TAILWATERS V-4
Aeration of Reservoir Releases V-4
Selective Withdrawal of Reservoir Water V-5
Habitat Improvement of Tailwaters V-6
OTHER MITIGATION MEASURES V-7
Watershed Management V-7
Operational Changes V-9
VI. FEDERAL AGENCIES' WATER QUALITY ASSESSMENTS OF IMPOUNDMENTS...VI-1
INTRODUCTION VI-1
CORPS OF ENGINEERS VI-2
Background VI-2
COE Water Quality Assessment VI-4
TENNESSEE VALLEY AUTHORITY VI-8
Background VI-8
TVA Water Quality Assessment VI-10
THE BUREAU OF RECLAMATION VI-16
Background VI-16
USBR Water Qual ity Assessment VI-17
OVERVIEW VI-24
xm
-------
TABLE OF CONTENTS (Continued)
VII. CONCLUSIONS VII-1
WATER QUALITY EFFECTS OF IMPOUNDMENTS VII-2
RESULTS OF EPA ANALYSES VII-4
AGENCY ASSESSMENTS VII-6
SUMMARY OF EPA ANALYSES AND AGENCY ASSESSMENTS VII-9
APPENDIX A. REFERENCES A-l
APPENDIX B. SAMPLE/CASE STUDY DATA BASE B-l
APPENDIX C. ANALYSIS SUPPLEMENT C-l
MIXING MODEL C-l
PHOSPHORUS RETENTION MODEL C-3
PHYSICAL ATTRIBUTE ESTIMATORS C-5
SIGNIFICANCE OF CORRELATION C-6
SAMPLE SIZE C-8
SAMPLE REPRESENTATIVENESS C-9
STORET WATER QUALITY RETRIEVAL SUMMARIZATIONS C-12
APPENDIX D. LARGE NONPOWER DAM SUPPLEMENT D-l
MIXING ANALYSIS D-l
DISSOLVED OXYGEN IN DAM TAILWATERS D-l
UPSTREAM/DOWNSTREAM COMPARISON OF WATER QUALITY D-l
PHOSPHORUS ENRICHMENT ANALYSIS D-3
APPENDIX E. SMALL DAM SUPPLEMENT J. .E-l
MIXING ANALYSIS E-l
DISSOLVED OXYGEN IN DAM TAILWATERS E-l
UPSTREAM/DOWNSTREAM COMPARISON OF WATER QUALITY E-l
PHOSPHORUS ENRICHMENT ANALYSIS E-5
REMARKS E-5
xiv
-------
TABLE OF CONTENTS (CONTINUED)
APPENDIX F. CORPS OF ENGINEER'S ASSESSMENT SUPPLEMENT F-l
GENERAL F-l
WATER QUALITY STATUS OF CORPS OF ENGINEER'S PROJECTS F-4
Introduction F-4
Methods F-5
Assessment. F-6
FUTURE DIRECTION F-10
APPENDIX G. TENNESSEE VALLEY AUTHORITY'S ASSESSMENT SUPPLEMENT..G-l
INTRODUCTION G-l
ASSESSMENT OF WATER QUALITY G-l
RESERVOIR WATER QUALITY MANAGEMENT 6-2
TAILWATER MANAGEMENT STRATEGY G-2
Reservoir Management Strategy G-3
Watershed Management Strategy G-4
POLICIES - WATER QUALITY MANAGEMENT G-5
Water Quality Policies, Codes, and Instructions G-5
RECOMMENDATIONS 6-7
Water Quality Research Recommendations G-9
Reservoir Research Recommendations G-10
Technology Development Recommendations G-ll
Operational Monitoring Recommendations G-ll
APPENDIX H. U.S. BUREAU OF RECLAMATION'S ASSESSMENTS SUPPLEMENT.H-l
AGENCY ASSESSMENTS H-l
THE BUREAU OF RECLAMATION H-4
Unique Features of the Bureau of Reclamation H-6
Reclamations of Salinity Control Program H-7
The Changing Nature of the Reclamation Program...._. .H-8
ASSESSMENT OF WATER QUALITY CONDITIONS IN BUREAU OF
RECLAMATION RESERVOIRS AND TAILWATERS H-10
CONCLUSIONS H-ll
REFERENCES H-12
RECOMMENDATIONS H-22
APPENDIX I. AGENCY QUESTIONNAIRE USED IN DEVELOPING ASSESSMENTS.1-1
XV
-------
LIST OF TABLES
Number Title Page
III-l Partitioning Criteria 111-4
III-2 National Potential Mixing Percentages for
Large Dams 111-10
III-3 Probabilities of Non-Compliance with 5 mg/1
Dissolved Oxygen for Power Dams 111-14
111-4 Upstream/Downstream Water Quality Changes in Large
Power Dams 111-19
III-5 Selected Summary of Nitrogen and Phosphorus in
the Pool 111-22
111-6 TVA Trophic Status Data 111-24
III-7 Summary Statistics for Total Phosphorus, Total
Nitrogen, and Chlorophyll-a in the Mixed Layer
During the Growing Season 111-25
II1-8 Phosphorus Enrichment Results for Large Power Dams...111-27
IV-1 Comparison of Principal Features of Case Study
Impoundments/Dams IV-4
IV-2 Case Study Profiles: Principal Effects of Dams
on Water Qua!ity IV-6
IV-3 Case Study Profiles: Mitigation Measures IV-8
C-l Sample Sizes for 95% Significant R2 C-7
C-2 Signi fi cance of El ementary Model s C-8
C-3 Summary of Geographic Distribution C-10
C-4 Results of the KS Test for Sample Distributions C-12
C-5 Water Qua! i ty Parameters C-13
C-6 Number of Dams Report! ng Data C-13
C-7 Average Number of Observations per Dam for Dams
Reporting Data ._.C-14
D-l Probabilities of Non-Compliance with 5 mg/1
Dissolved Oxygen for Large Nonpower Dams D-2
D-2 Tallies of Water Quality Changes in Large Nonpower
Dams D-3
xvi
-------
LIST OF FIGURES
Number Title Page
1-1 Population of All Dams 1-7
III-l Population of Large Power Dams III-5
111-2 Population of Large Nonpower Dams III-5
III-3 Population of Small Dams III-5
III-4 Stratification versus Densimetric Froude Number
for TVA Reservoi rs 111 -8
III-5 Geographical Breakdown of Regions for the Oak Ridge
National Laboratory Study 111-13
III-6 Distribution of 73 CE Hydropower Projects and
Their Tailwater Conditions 111-15
III-7 Vollenweider Model Performance 111-26
IV-1 Location and Name of Case Study Impoundments.. IV-2
VI-1 Geographic Distribution of Corps of Engineer's Water
Resources Projects VI-3
VI-2 Geographic Distribution of a Ten Percent, Stratified,
Random Sample of Corps of Engineer's Projects
for Which Questionnaires Have Been Received VI-3
VI-3 Frequency of Occurrence of Water Quality Conditions
in COE Tailwaters VI-5
VI-4 Frequency of Occurrence of Water Quality Conditions
in COE Impoundments VI-6
VI-5 The Tennessee Valley Authority's Multipurpose Water
Resource System VI-9
VI-6 Frequency of Occurrence of Water Quality Conditions
in TVA Reservoir Pools VI-11
VI-7 User Impact Assessment for TVA Reservoir Pools VI-12
VI-8 Frequency of Occurrence of Water Quality Conditions
in TVA Tailwaters VI-13
VI-9 User Impact Assessment for TVA Reservoir Tailwaters VI-14
VI-10 Geographic Distribution of Responses to Water
Quality Assessment of Bureau of Reclamation
Reservoirs by State VI-18
VI-11 Frequency of Occurrence of Water Quality Conditions
in USBR Reservoirs VI-19
VI-12 User Impacts of Water Quality Conditions on User
Benefits at USBR Reservoirs VI-20
VI-13 Frequency of Occurrence of Water Quality Conditions
in Tailwaters of USBR Reservoirs VI-21
VI-14 Impacts of Water Quality on User Benefits at USBR
Reservoirs VI-22
xvi i
-------
LIST OF FIGURES
(Continued)
VII-1 Percentages of Mixing Tendency VII-4
VI1-2 Probabilities of Non-compliance (PNC) with 5 mg/1
Dissolved Oxygen (DO) VII-5
VII-3 Summary of Quantitative Analyses of Large Power
Dams VII-7
C-l Volume Distributions (large power dams) C-ll
F-l Geographic Distribution of Corps of Engineer's Water
Resource Projects F-12
F-2 Geographic Distribution of a Ten Percent, Stratified,
Random Sample of Corps of Engineer's Projects for
Which Questionnaires Have Been Received F-12
F-3 Frequency of Occurrence of Occasionally Problematic,
Intermittently Problematic, Chronically Problematic,
and Non-problematic Water Quality Conditions in
Tailwaters F-13
F-4 Frequency of Occurrence of Occasionally Problematic,
Intermittently Problematic, Chronically Problematic,
and Non-problematic Pool Water Quality Conditions....F-13
G-l The Tennessee Valley Authority's Multipurpose Water
Resource System G-12
G-2 Rate of Recurrence and User Impact Assessment for
TVA Reservoir Pools for the Indicated Parameters G-13
G-3 Rate of Recurrence and User Impact Assessment for
TVA Reservoir Tailwaters for the Indicated
Parameters G-14
H-l Geographic Distribution of Responses for Water
Quality Assessment of Bureau of Reclamation
Reservoirs by State (Total-250) H-13
H-2 Frequency of Occurrence of Water Quality Conditions
in USER Reservoirs H-14
H-3 Impacts of Water Quality Conditions on User Benefits
at USBR Reservoirs H-15
H-4 Water Quality Conditions in Tailwaters of USBR
Reservoirs H-16
H-5 Impacts of Water Quality on User Benefits -
Tailwaters .H-17
H-6 Frequency of Occurrence of Water Quality Conditions
in USBR Reservoirs H-18
H-7 Impacts of Water Quality Conditions on User
Benefits at USBR Reservoirs. .H-19
H-8 Water Quality Conditions in Tailwaters of USBR
Reservoirs H-20
H-9 Impacts of Water Quality on User Benefits -
Tailwaters H-21
XVTM
-------
I. INTRODUCTION
OBJECTIVES AND HISTORICAL CONTEXT
The objective of this document is to fulfill EPA's responsibility to
provide a report to Congress on Dam Water Quality in response to Section
524 of the Water Quality Act of 1987. The report addresses the water
quality effects associated with impoundments and attempts to estimate the
character and national extent of these effects. The specific language of
Section 524 is:
Sec. 524 DAM WATER QUALITY STUDY
"The Administrator, in cooperation with interested States
and Federal agencies, shall study and monitor the effects
on the quality of navigable waters attributable to the
impoundment of water by dams. The results of such study
shall be submitted to Congress not later than December 31,
1987."
The effects of impoundments on water quality has been a topic of
interest for many years. At a 1963 symposium on impoundments, W.E.
Knight, from the state of North Carolina, summed up the situation (USHEW,
1965):
"The effects of storage on water quality have been the
subject of much study by Federal and State agencies.
Thermal stratification of lakes has boor ^ecognized by man
since he first dived into a lake as ound the water near
the bottom colder than that at the jr1 :•„?. Likewise, the
fact that at times water near' t ' ""a reservoir
contains little, if any, dissolved oxygen has long been
known. Studies of the concentration of dissolved o'vgenjn
water discharged from reservoirs indicate that du, ing the
critical summer period, dams with deep intakes discharged
water of very low dissolved oxygen content, dams with
intermediate intakes discharged water of higher dissolved
oxygen content, and dams with high-level intakes discharged
water with even higher dissolved oxygen content."
Furthermore, based on 26 years of data, eminent researcher Milo A.
Churchill of the Tennessee Valley Authority, concluded (USHEW, 1965):
1-1
-------
"If water is released through low-level outlets in the dam,
significant effects on downstream water temperature result
during the warmer months. If water is released through
high-level outlets, the reservoir may have little effect on
water temperature".
Thus, it is well known that deep reservoirs thermally stratify and
can have poor water quality in their deeper layers. It is also well
known that reservoirs retain plant nutrients, nitrogen, and phosphorus
and can become eutrophic or filled with excessive aquatic plants from
microscopic algae up to, and including, large plants, like water lilies.
Unwanted effects can be mitigated, if deemed necessary, in several ways
from the aeration of tail waters for oxygen restoration to the watershed
control of nutrients to avoid overfertilization of impoundments.
During the past fifteen years, federal agencies responsible for the
planning, construction, operation, and/or regulation of impoundments have
taken the initiative and attempted to address many of the emerging water
quality issues. In 1978, the Tennessee Valley Authority (TVA) released a
report that "identifies adverse impacts on water quality and stream uses
that are associated with water releases from dams operated as part of the
TVA water management system" (TVA, 1978). TVA identifies several
reservoirs as having dissolved oxygen concentrations in their releases
that do not constantly attain numerical criteria.
In 1976, the U.S. Army Corps of Engineers (CE) conducted a survey to
identify and assess the magnitude of environmental and water quality
problems at their reservoir and waterway projects, and to determine major
research needs to address these problems (Keeley, et al., 1978). This
information was incorporated into a research and development program,
Environmental and Water Quality Operational Studies (EWQOS), designed to
provide new or improved technology for addressing the environmental and
water quality problems associated with these projects, in a manner
compatible with authorized project purposes. The EWQOS program
represents an eight-year research effort (1977-1985), and included areas
such as predictive techniques to help describe, predict, and model
various aspects of reservoir hydrodynamics, ecology, and water quality
processes. Reservoir operational and management techniques were also
developed and evaluated.
The U.S. Bureau of Reclamation has increasingly addressed water
quality and other environmental issues for their projects since the late
1960's. Procedures have been pursued for the incorporation of water
quality factors in project planning, design, construction, and
operations. Supporting research has also been conducted including
special programs to address irrigation return flows.
1-2
-------
EPA currently addresses water quality issues pertaining to
impoundments as part of the nonpoint source program of State and local
water quality oversight and area wide management plans. In 1982, the
National Wildlife Federation pressed an unsuccessful suit against EPA to
regulate dams as point sources rather than continue their regulation under
areawide wastewater management plans. All parties to this suit
acknowledged various water quality changes brought about by impoundments.
What is missing in the transcripts is an estimate of the magnitude of the
situation nationally.
This report, with its ramifications, is of interest to the States
and Federal agencies that manage impoundments as well as those agencies
responsible for regulating impoundments. As requested by Section 524,
several States and Federal agencies were solicited for their suggestions
and ideas. Federal agencies have also provided case studies to this
report to illustrate water quality changes and mitigative methods to
counter change at a variety of sites.
With this background in mind, the following questions are addressed
in this report:
1. What are the effects of impoundments on water quality?
2. What mitigation measures can reverse unwanted effects?
3. What is the national extent of significant water quality
effects?
The answers to questions 1 and 2 are definitive and are supported
with case studies. The answers to question 3 are based on elementary
statistical tests using limited random sampling and therefore give a
preliminary estimate of the national scope of water quality effects
associated with impoundments. For some aspects, a national assessment
could not be made. More work and analysis are necessary to make these
answers definitive.
Section 524 of the Clean Water Act requires EPA to "study and
monitor the effects on the quality of navigable waters" due to water
impoundment. Thus, the scope of this report is limited to historically
documented ambient water quality effects. The report does not address
effects on biological habitat or wetlands, which may be substantial.
Further, the study is focused on the effects of existing impoundments and
does not attempt to assess the water quality or biological effects
associated with new impoundments of free-flowing waters in particular
geographic areas. Flooding to create new impoundments may cause
significant hydrologic changes to the river and destroy wetland and upland
habitat, and often raises profound environmental questions. These effects
and questions are evaluated by EPA on a site-by-site basis through the
Environmental Impact Statement process and as otherwise required by State
and Federal law, but are beyond the scope of this report.
1-3
-------
PURPOSES AND NUMBERS OF IMPOUNDMENTS
Impoundments are created for a variety of purposes that provide
important social, economic, and aesthetic benefits. Most projects serve
multiple purposes. Understanding these purposes is important for
providing a context in which to evaluate their water quality consequences.
Brief descriptions of the more common purposes are as follows:
• Hydropower Generation - Large hydropower projects tend to have
sizable volumes and high dams. Several large TVA projects, though
multi-purpose in design and use, have as one of their purposes
hydropower generation. However, low head small hydropower
projects are more common. There is a current interest in add-on
hydropower and this typically is associated with low hydraulic
heads. Developers of such projects seek licenses which give them
rights of eminent domain and guaranteed markets. Such private
development of add-ons to public projects can be a concern to
water control agencies because the associated purposes were not
included in the original design. Federal and state regulators do
review these add-on projects for water quality impacts.
• Pump Storage Hydropower - These projects can be sizable and
typically have two water storage pools—a lower pool and an upper
pool—with combination turbine/pump(s) between the pools. These
combined units are used to balance power generation with power
demand. In low demand periods, water is pumped from the lower
pool to the upper pool to create potential energy. In high demand
periods, the potential energy is converted back to power by
release through turbines to the lower pool.
• Navigation - There are numerous navigation projects, having locks,
run-of-river dams, and levees. These facilities provide a
reliable transportation route for commercial, military, and
private transport. Study of polluted navigation pools on the Ohio
River, by the Public Health Service nearly 75 years ago, led to
the Streeter-Phelps water quality model of dissolved oxygen
(Streeter, 1925).
• Flood Control - These projects have a permanent pool which is
small in comparison to the large volume set aside to trap floods.
A large number of "dry dams" have a very small or no permanent
pool. _
• Water Supply - Water is stored for public or private use and may
include irrigation of agricultural areas.
• Conservation - A large number of small private and public
projects are built to maintain ground water levels, trap sediment,
control gully erosion, provide livestock watering, and offer
recreational activities. The Soil Conservation Service (SCS)
frequently assists in the design of these facilities.
1-4
-------
• Recreation - Typically, recreation is one use in a multi-purpose
federal facility. For example, Corps of Engineers' flood control
projects often have a recreational component. In private
projects, recreation may constitute the only purpose. Private
real estate lakes, like Lake-of-the-Woods, a 500-acre lake that
is the centerpiece of a single lot land development in Virginia,
are purely recreational.
• Other - Many impoundments service other uses such as fish and
wildlife maintenance, water quality enhancement, and low flow
augmentation. A combination of these with the above uses is
common and play an important role in water resource management.
The development of water resource projects in the public sector
follows a procedure that seeks benefits, in terms of achieving stated
purposes, in excess of costs. Since passage of the National Environmental
Policy Act, this assessment process has explicitly included consideration
of environmental consequences. Development of private projects is
dependent upon generation of profits to justify their capitalization.
Regulated activities in the private sector also are required to satisfy
environmental concerns to secure development rights.
In addition to understanding impoundment purposes, an accurate
estimate of the preponderance of facilities is necessary to assess
national water quality effects. Identifying or counting all of these dams
is no small task. To expedite this effort, several existing inventories
were evaluated.
The U.S. Army Corps of Engineers compiled an inventory of dams in
the United States as part of its National Program of Inspection of Non-
Federal Dams (CE, 1982a). This inventory, which was completed in
September 1980, includes dams which are in excess of 6 feet in height and
have a maximum water impounding capacity of at least 50 acre-feet; or
which are at least 25 feet in height and have a maximum water impoundment
capacity in excess of 15 acre-feet. The inventory contains 68,155
entries. The Soil Conservation Service (SCS) estimated an additional
2,000,000 small farm pond and recreational dams with volumes less than 50
acre-feet, or dams less than 25 feet high. Approximately 24,000 of
entries in the inventory have received technical and/or financial
assistance from the Soil Conservation Service. These are the largest of
the SCS-assisted dams.
The 68,155 dams in the Corps of Engineers inventory collectively
store 490 million acre-feet of water at the normal pool volume. The Corps
of Engineers is associated with 166 million acre-feet (34 percent) and the
Bureau of Reclamation is associated with 107 million acre-feet (22
percent). The Tennessee Valley Authority is associated with 10 million
acre-feet (2 percent).
1-5
-------
The Corps of Engineers' listing was chosen as a realistic starting
point for a national assessment. Figure 1-1 shows the geographical
distribution of the dams contained in this 1980 inventory. This listing
contains information on hazard potential but also provides the following
basic information:
1. Location and name.
2. Volume of impoundment.
3. Maximum depth of water (at the dam).
4. Spillway design flow.
5. Length of dam crest.
6. Power generating capacity (if known and over 100 kW).
7. Purposes.
By itself, this is incomplete information for water quality analysis
purposes. But, it provides an exhaustive listing of dams existing in
1980. More complete data for specific dams are available in compendiums
of water quality information associated with Corps of Engineers dams.
These listings are not exhaustive and are not statistically representative
of all impoundments.
ANALYTICAL DEVELOPMENT
An approach to this nationwide assessment was developed through the
comment process as well as a review of the literature. Review of the
literature yielded numerous materials on specific case studies, generic
problem descriptions, mitigation measures, program reviews, and other
elements of the issues surrounding the water quality effects of
impoundments. A list of references cited in this report appears in the
reference section which includes additional materials reviewed but not
cited. Three relatively recent references (TVA, 1978; Cada, et al., 1983;
Kennedy and Gaugush, 1987) are representative of alternative approaches
to the task at hand. Each tends to be authoritative because of the
expertise of the authors and rigor of their methods. All three are
sponsored by Federal agencies with a responsibility of some sort for
impoundments. The Tennessee Valley Authority owns and operates
impoundments. Cada, et al., work for the Oak Ridge National Laboratory
(ORNL), which is under the Department of Energy. Finally, Kennedy and
Gaugush are with the U.S. Army Corps of Engineers (CE), which manages a
large number of dams. The development of the analytical approach for this
report to Congress is the result of adapting relevant features of each of
these works and making an application to a nationwide assessment.
TVA (1978) adopted a general approach to describing generic effects
and mitigation measures while taking a very detailed site-specific
examination of many of their reservoirs:
"Releases from Tennessee Valley reservoirs were evaluated with
respect to their adverse impacts on water quality and related
characteristics and on various uses of the water downstream.
The water quality characteristics examined were temperature,
1-6
-------
Alaska
Hawa i i
Guam
Puerto Rico
Trust Terr.
- 167
- 123
- 1
- 70
- 2
Virgin Islands- 8
Figure 1-1
Population of 68,155 Dams
(Source: CE, 1982b)
1-7
-------
dissolved oxygen, resolubilization of metals, streamflow, turbidity
and suspended solids, and gaseous supersaturation. The downstream
water uses examined were recreation, fisheries and other aquatic
life, water supply, and assimilative capacity. Evaluations were
based on available data on water quality and reservoir operations
from 1970 through 1977 in conjunction with biological and
engineering judgments of the significance of the impacts."
The TVA study has the following major features:
• Evaluation of a small population of large power-generating
impoundments,
• Detailed site-specific analysis of data to estimate effects,
• Analysis of pool and tailwater water quality,
• Comprehensive generic description of the "adverse impacts on water
quality and stream uses...", and
• Presentation of mitigation measures applicable to TVA's
impoundments.
Perhaps the most significant difference between the TVA study and
the one conceived for this report is the size of the population to be
analyzed. TVA operates a total of 47 reservoirs, of which 33 were
discussed in their study. On a national scale, one is faced with
thousands of reservoirs. TVA (1978) argues the following:
"There are two possible methods for identifying and
determining the extent of impacts from reservoir releases.
The first method compares the releases from all structures to
specific numerical criteria. This method simplifies problem
identification and monitoring. However, the application of
rigid, uniform numerical criteria may overlook the actual or
highest value uses and needs upstream and downstream from a
specific project. The second method determines the need for
improvements on a case-by-case basis and emphasizes providing
balanced protection of the water uses associated with each
specific project without unduly penalizing any given project
by the application of rigid numerical criteria. Although this
method makes problem identification and subsequent monitorang
efforts much more difficult, it provides the flexibility
needed to optimize resource management."
For the scope of TVA's study, a case-by-case approach is sensible.
However, for the purposes of this report to Congress, "problem
identification" is required as a necessary first step towards a national
strategy to "optimize resource management." Specific numerical criteria
1-8
-------
are used sparsely in this report to Congress. Also, recognizing the
importance of in-depth investigations of specific sites, several case
studies have been incorporated in this report to provide a high level of
detail.
Several features of the TVA study have been adopted for this report.
First, the water quality in both the impoundment itself and in the
downstream tailwaters are examined. This report takes the analysis one
step further and compares upstream water quality to downstream water
quality to assess the changes in the vicinity of, if not caused by, the
impoundment.
Second, the TVA study and this report present a generic description
of the water quality effects associated with impoundments. In fact, much
of the material included in this report's discussion of generic effects
is taken from the TVA study. TVA also discussed generic effects on water
uses. This was excluded from this report because, although important,
this pertains more to resource management rather than problem
identification. Furthermore, water uses are often designated based on
water quality, making the assessment of uses somewhat subjective.
Finally, TVA presented a generic discussion of mitigation measures.
This report also provides this information, much of which was taken from
the TVA study.
A second approach to the subject is presented by the ORNL study
(Cada, et al., 1983), which has the following abstract:
"One of the environmental issues affecting small-scale
hydropower development in the United States is water quality
degradation. The extent of this potential problem, as
exemplified by low dissolved oxygen concentrations in
reservoir tailwaters, was analyzed by pairing operating
hydroelectric sites with dissolved oxygen measurements from
nearby downstream U.S. Geological Survey water quality
stations. These data were used to calculate probabilities of
noncompliance (PNCs), that is the probabilities that dissolved
oxygen concentrations in the discharge waters of operating
hydroelectric dams will drop below 5 mg/1. The continental
states were grouped into eight regions based on geographic and
climatic similarities. Most regions had higher mean PNCs in
summer than in winter, and summer PNCs were greater for large-
scale than for small-scale hydropower facilities. Cumulative
probability distributions of PNC also indicated that low
dissolved oxygen concentrations in the tailwaters of operating
hydroelectric dams are phenomena largely confined to sites
with large-scale facilities."
The ORNL study has the following major features:
• Evaluation of a population of power generating impoundments based
on a survey of available data,
1-9
-------
• Analysis of dissolved oxygen data from STORE! (EPA's national
water quality data base) using a numerical criterion and
statistical methods,
• Analysis of tailwaters only, and
• Regional and seasonal evaluation of the data.
The ORNL study attempts to draw national conclusions based on the
pairing of a power dam with available dissolved oxygen data meeting
specified criteria. While national in scope, the study could not make the
claim that the sites analyzed were representative. It is desirable for
this report to Congress to be based on representative sites.
ORNL disaggregates available data geographically and seasonally.
If sufficient data exist to provide a complete picture, data
disaggregation can highlight the most pronounced results which may
otherwise be buried in an aggregated data set. This report to Congress
includes a revisitation of the ORNL study of dissolved oxygen to
illustrate some of the advantages of disaggregation and its importance to
the national perspective. Resource limitations precluded performing such
an approach on all parameters including the pool and the tailwaters.
However, this report does take the ORNL study one step further by
identifying a representative sample of sites which moves toward an
assessment of national scope.
A third approach to the subject is an ongoing study by the Army
Corps of Engineers. Intermediate progress in this effort has been
reported by Kennedy and Gaugush (1987). A brief summary of this study is
obtained from the paper's abstract:
"Increasing concern over the quality of freshwaters and
growing emphasis on the development of improved methods for
the management of its existing water resource projects, led
to the initiation of several major water research programs by
the U.S. Army Corps of Engineers. One such effort involves
the compilation and analysis of a Corps-wide water quality
database for reservoirs and tailwaters. These data are being
supplemented with subjective information concerning
enhancement needs solicited from field offices. The survey
involves over 750 reservoirs, locks and dams, and dry dams.
Most frequently cited during the survey were the need -to
improve tailwater conditions (particularly dissolved oxygen,
temperature extremes, and the presence of reduced metals),
reduce nutrient concentrations, and ameliorate conditions
associated with the eutrophication process. A preliminary
investigation of southeastern reservoirs indicated: 1) lower
total nitrogen concentrations than the national average, 2)
significant longitudinal gradients in water quality, and 3)
turbidity and flushing rate as probable constraints to primary
production."
MO
-------
The CE study has the following features:
• Evaluation of the CE water resources projects, which include
reservoirs, locks and dams, and dry dams built predominantly for
the purposes of flood control, recreation, and water supply,
• An analysis of dissolved oxygen as it compares to the physical
features of a dam,
• Analysis of nutrients, metals, and dissolved oxygen in
southeastern reservoirs by comparing discharges or releases and
inflows based on STORE! data,
• A growing season analysis as well as an investigation into water
quality gradients in the pool, and
• Qualitative presentation of water quality issues in the pool and
tailwaters.
Similar to the other two studies, Kennedy and Gaugush evaluate a
much smaller population of impoundments than is of interest for this
report to Congress. CE projects are also oriented towards flood control,
recreation, and water supply as well as power production. The Corps of
Engineers approach of comparing the physical features of a project to
water quality status of tailwaters is promising because if such
relationships can be found, nationwide assessments are greatly
facilitated. Much more is known (or is easily obtainable) about physical
features than is known about the water quality at unmonitored sites.
Therefore, this report to Congress endeavors to identify those
relationships where they may exist.
The Corps of Engineers, STORET comparison of paired inflow and
discharge concentrations of a given parameter also is expected to be quite
useful because it addresses the change in water quality that may be
caused, or at least facilitated, by an impoundment. Identified changes
may be both positive or negative. This report to Congress relies heavily
on both STORET data and an analysis of water quality changes between the
inflow and discharge. STORET is the EPA supported national computerized
repository for water quality data (USEPA, 1982a).
The CE study approaches seasonality by performing a "growing season"
analysis of nitrogen, phosphorus, and chlorophyll. The benefits of a
seasonal approach have already been discussed, although the scope of this
report to Congress has been limited to a seasonal analysis of dissolved
oxygen.
Lastly, the CE report is addressing the water quality issues
generically and reporting the frequency of occurrence of various types of
problems. This approach was also used in the TVA study.
Ml
-------
In sum, the three studies presented here by the TVA, ORNL, and CE
offer a representative range of approaches and current thinking available
for this national assessment. The analytical approach for this report to
Congress was developed to take the most appropriate features of each--
within budget and schedule constraints—in providing an informative
overview to Congress.
OVERVIEW OF REPORT CONTENTS
The major features of this report and their location in this
document are as follows:
• A generic description of water quality concerns related to
impoundments of all purposes including power production,
navigation, flood control, recreation, water supply, etc. (Chapter
II);
• A representative sampling methodology to arrive at a manageable
sample from the 68,155 population of impoundments for analysis
(Chapter III);
• A mixing analysis of impoundments to screen for stratification
potential based on physical characteristics of a site (Chapter
HI);
• A seasonal and regional examination of dissolved oxygen in the
tailwaters (Chapter III);
• An upstream/downstream comparison of concentrations of nutrients,
metals, and dissolved oxygen to assess the changes facilitated by
impoundments (Chapter III);
• An analysis of nutrient enrichment in the pool (Chapter III); and
• A presentation of case studies to examine site-specific situations
(Chapter IV); and
• A generic discussion of mitigation measures applicable to water
quality in the pool and/or the tailwaters (Chapter V).
• Agency assessments of approaches and need pertaining Jto water
quality issues for their own programs (Chapter VI).
Each of these concepts has been used in prior endeavors except for
the nationwide representative sampling. The purpose of sampling, if the
available data support it, is to extrapolate results of the analysis to
provide a national characterization of the issues to fulfill Congress'
information needs in this area.
1-12
-------
II. WATER QUALITY EFFECTS OF IMPOUNDMENTS
This chapter presents an overview of the potential effects of
impoundments upon water quality. Although attention is usually focused
on unwanted effects, desirable changes may also result. An "ecosystem
perspective" is necessary since issues are interdependent, in varying
degrees, upon each other. If one condition within the reservoir - stream
ecosystem is altered, its effects can ripple through the balanced system,
positively or negatively affecting other facets of water quality. The
following list is not exhaustive, but provides the major potential
reservoir-stream ecosystem changes:
STRATIFICATION
Low Hypolimnetic Dissolved Oxygen
Increased Iron & Manganese
Hydrogen Sulfide
Denitrification
Thermal Changes
EUTROPHICATION
OTHER CHANGES
Gaseous Supersaturation
Salinity Changes
Sediment Movement
Flow Regulation
Reaeration Denial
Fish Entrainment
Toxics Accumulation
The above effects do not occur in every impoundment system. Their
occurrence and magnitude depend on many factors, including depth of the
reservoir, climate, watershed land use, reservoir siting, and reservoir
features. Some—land use, siting, and reservoir features—are. at least
partially controllable, while others--climate--are uncontrollable.
Certain geographic regions have their own dominant water quality
concerns. A brief overview of each potential effect is discussed in the
following sections.
II-l
-------
STRATIFICATION
Water quality effects in the discharges of reservoirs can result from
seasonal warming and consequent thermal stratification of impounded
waters (Cada et.al, 1983). However, not all reservoirs stratify and
stratification, by itself, is not a water quality problem. The
occurrence of this effect depends on several factors, including reservoir
surface area, depth, volume, detention time, degree of protection from
the wind, and climatic conditions of the geographical area (Cada et. al,
1983). For example, run-of-river impoundments are located on main stream
rivers and are characterized by low head dams with impounded water not
extending far from the natural channel. Detention times are on the order
of a few days. Since water velocities are appreciable, significant
vertical stratification typically does not occur (USEPA, 1973).
In contrast, storage reservoirs are generally located on tributary
streams and are relatively deep, with the water surface extending far
beyond the natural river channel. These reservoirs have a large storage
capacity in relation to the drainage area and generally have a detention
time of several months. They can be characterized by thermal
stratification, usually of the classic three layer system, during the
summer warm periods (USEPA, 1973). The upper layer is termed the
epilimnion. A zone of rapid drop in temperature occurs below this and is
called the thermocline. Below the thermocline is a zone of fairly
uniform, cooler temperatures called the hypolimnion. In the fall,
surface temperatures cool to the same temperature as the hypolimnion and
the stratification is disrupted, with the impounded water completely
mixing. The warmer climate of the Southeast and Great Basin regions
results in thermal stratification occurring earlier in the year and
remaining longer than in the Northeast or Northwest (Cada et al, 1983).
A wind-driven turnover may also occur in the spring when there is no
vertical temperature gradient.
Thermal stratification often results in stratification of important
water quality chemical parameters, such as dissolved oxygen, metals, and
nutrients. These parameters are discussed in more detail in the
following sections. It is important to note that stratification is a
natural condition that occurs in many lakes. The important difference
between the lake and reservoir is that in a lake the epilimnetic waters,
generally characterized by good water quality and reflecting the
prevailing average ambient air temperatures, are released to an
outflowing stream. In a reservoir, the hypolimnetic waters, wjiich may
have poorer water quality (in terms of dissolved oxygen, iron, and
manganese) than the pre-impounded stream, can be released through low
level outlets. Some dams are constructed so that water can be withdrawn
at several different depths appropriate for discharge to the outflowing
stream. As a result, the dam may not have a negative effect on
downstream conditions. When reservoirs do not stratify and are mixed
from top to bottom, the water discharged does not represent a particular
zone and thus the depth of withdraw is not critical to water quality.
Water quality effects associated with stratification may include low
hypolimnetic dissolved oxygen, increased iron and manganese, hydrogen
sulfide, denitrification, and thermal changes.
II-2
-------
Low Hypolimnetic Dissolved Oxygen (DO).
During certain periods of the year, the denser waters of the
hypolimnion remain relatively free from turbulence or other significant
water motions, and have no opportunity for reaeration at the surface.
Demands on oxygen are present because of bacterial decay of organic
matter, resulting in oxygen-poor or anoxic waters. Anoxic bottom waters
in reservoirs cannot support fish and other aquatic life. Furthermore,
when large volumes of low DO water are released, downstream waters are
also adversely affected. Eventually, through turbulent reaeration
processes downstream, DO levels return to normal.
Low tail water DO concentrations generally occur where reservoirs
remain stratified for longer periods and contain relatively warm waters
(Cada, et al., 1983). The distance the stream needs to recover may be
great enough to lower the stream's assimilative capacity for oxygen
demanding wastes from downstream sources. The anaerobic conditions of
the hypolimnion are also harmful to water quality since it may cause
several chemical reactions to occur, such as release of metals,
phosphorus, iron and manganese from bottom sediments. Increases in iron,
ammonia, manganese, silica, phosphate and sulfide ions, and soluble
organic compounds have been observed in anoxic waters in contact with
bottom sediments. In the fall and spring, uniform vertical temperatures
result in periods of mixing called the fall and spring turnovers. During
these times, soluble materials entrapped in the hypolimnion, such as
inorganic nutrients, iron and manganese and organic material, are
returned to the biologically active surface waters. The nutrients become
available to support primary production and sometimes result in fall and
spring plankton blooms observed in many reservoirs and lakes (USEPA,
1973).
Increased Iron and Manganese.
A water quality problem directly tied to the anoxic waters of a
hypolimnion is the increased concentrations of certain metals, especially
iron and manganese. Reservoirs not stratified and thus without an anoxic
hypolimnicn usually act as sinks for these metals, which remain adsorbed
to bottom sediments. However, under anoxic conditions these insoluble
metals may become soluble and are released from the sediments, into
reservoir waters as well as downstream waters. Metals in significant
concentrations can be harmful to fish and other aquatic_life and
contaminate water supplies. When waters are aerated in the spillway,
iron and manganese become oxidized and precipitate out of the water,
being deposited upon and causing discoloration of structures and the
stream bed of the tailwaters. Soluble iron oxidizes rather rapidly, so
that the effect on downstream waters is limited, but soluble manganese
oxidizes much more slowly and therefore impacts a much larger area
downstream (TVA, 1978). High concentrations of either soluble iron or
manganese can adversely impact downstream water supplies by staining
plumbing fixtures and laundry, affecting the taste and aesthetic quality
of the water, and interfering with manufacturing processes.
II-3
-------
Hydrogen Sulflde Production.
Under anaerobic conditions of the hypo "limn ion, sulfate, a common
constituent of streams, is reduced to hydrogen sulfide (H^S). If
present in sufficient concentrations, h^S can reach levels toxic to fish
in the reservoir and downstream waters. Hydrogen sulfide can also be an
odor problem if enough is released into the atmosphere. Finally, where
reservoirs are used for supplying water for domestic use, sulfur
compounds can adversely affect water taste and lower water hardness.
Denitrification.
Denitrification only occurs in the absence of dissolved oxygen.
Facultative, anaerobic bacteria use nitrate and reduce it to produce
nitrite as an intermediate product, with the final principal end product
by most organisms.
Denitrification can
impoundments, where
summer months. The
being nitrogen gas--a nitrogen form not utilizable
Denitrification therefore acts as a nitrogen sink.
be an important process in stratified, eutrophic
anoxic conditions occur in the hypolimnion in the
loss of gaseous nitrogen changes the nutrient balance,
Thermal Changes.
Because stratification generates layers of warmer and colder waters,
the water withdrawn for
warmer or colder than
systems are themselves
fish and warm water fish.
nuclear power generation
energy or turbulence; the
release, depending upon the inlet depth, may be
pre-impoundment conditions. Some stratified
beneficial by providing habitat for cold water
Their use for heat exchange in fossil or
or for pump storage operations can add thermal
stratification regime could be altered.
In cold water streams an increase in temperature may be sufficient to
adversely impact the fish population inhabiting the system. This adverse
impact has a high potential for the cool water systems in the South,
since most of the streams are close to the temperature borderline for
classification as cool water streams (SCS, 1979). The elevated
temperatures may not be adverse to warm water fish (SCS, 1979). Fish can
acclimate to rising temperatures if the rates of increase remain gradual,
but death occurs once their absolute thermal ceiling is reached
(Schwiebert, 1984). The receiving stream will eventually approach or
reach the natural stream temperature, but the distance required depends
on many factors, such as canopy cover and groundwater flow (SCJS, 1979).
Sometimes the release of cold-water discharges of large reservoirs, like
the Ozark and Tennessee River tailwaters, can create new cold-water
fisheries downstream (Schwiebert, 1984). The Peapacton Dam on the east
branch of the Delaware River transforms a marginal trout river into one
of the best major trout streams in the East (Schwiebert, 1984).
Reservoir construction on the San Juan in New Mexico changes a coarse-
fish habitat into an excellent rainbow fishery due to cold water
discharges (Schwiebert, 1984). Impoundments and resulting tailwaters in
the southern states creates a trout fishery resource equal to hundreds of
miles of natural coldwater streams (Pfitzer, 1974). Some of these states
do not have a significant natural trout fishery.
II-4
-------
Fisheries created by impoundments may only be seasonal however, if
the volume of cold water stored in the reservoir is insufficient to allow
releases to occur until the fall turnover. At that time,warm water is
released into the tailwaters. As a result the tailwater is too cold in
spring and early summer for warm-water fish production and too warm in
late summer and fall to support cold-water fisheries (Pfitzer, 1974).
EUTROPHICATION
Eutrophication is a natural process that occurs not only in
reservoirs, but also in natural lakes and other water bodies,
particularly those which have low velocity rates. This process involves
increased growth and death rates of aquatic plants, usually resulting
of high levels of nitrogen and phosphorus. Since algae
of a complex food chain, there are numerous effects on
animals when the population and diversity of algae are
eutrophication and stratification can have
algae settle into the hypolimnion and increase
from the addition
are at the base
higher plants and
changed.
compounded
Furthermore,
effects. Dead
demands on scarce oxygen resources through bacterial decay. Algal blooms
degrade reservoir quality by triggering imbalances in the oxygen cycle to
which other organisms, such as fish and smaller food-chain organisms, are
sensitive. Excessive growths of certain blue-green algae can also cause
taste and odor problems in drinking water. During the day, the excessive
growths of algae produce DO through photosynthesis, but at night, algal
respiration depletes DO concentrations and produces carbon dioxide which
reduces pH. These daily fluctuations in pH and DO can have detrimental
effects on other biota. Once algae die, DO used in their decomposition
can significantly lower DO levels, particularly in the hypolimnion.
The anaerobic environment of the hypolimnion can increase
eutrophication by causing releases of phosphorus that would normally be
adsorbed to bottom sediments. This release can initiate a transfer of
soluble phosphorus from the sediment to hypolimnion to epilimnion and
provide nutrients during the growing season. When the lake destratifies
or completely mixes, the soluble phosphorus is also most available for
algal utilization although this typically occurs at the end of the
growing season.
Eutrophication can also include the excessive growth of rooted
aquatic plants, that detract from aesthetic qualities^ reduce
recreational opportunities, and also deplete DO resources when they die.
Eutrophication therefore may hinder recreational use of a reservoir and
impair its value as a low cost treatable water supply. Advanced
eutrophication is generally thought to be undesirable, but a moderate
amount of nutrient enrichment can result in a desirably productive
system, oftentimes with improved sport fishing opportunities.
II-5
-------
OTHER CHANGES
There are a number of other changes associated with impoundments that
can affect water quality. These are not directly related to
stratification or eutrophication and are, therefore, presented
separately. They include gaseous supersaturation, salinity changes,
sediment movement, flow regulation, reaeration denial, and fish
entrainment.
Gaseous Supersaturation.
In reservoirs, supersaturation of gases can occur as the result of
several processes. The most common manner in which impoundments may
cause supersaturation is the interaction of air-water mixtures under
pressure in spillways and sluiceways which can entrain air into deep
turbulent stilling basins where excessive gases are dissolved. The
injection of air, or reaeration of the hypolimnion may also contribute to
a state of supersaturation. A third cause is heating which lowers the
saturation point. An example of this occurs in the spring when a
reservoir receives gas saturated cold runoff water and is also recharged
with gases during the turnover. As the water is heated through,
seasonal warming supersaturation may occur (Bouck, 1980). Dissolved
oxygen is extremely bioactive and is not a problem, but dissolved
nitrogen is biologically inert in vertebrates and, therefore, can cause
gas bubble disease, a disease analogous to the "bends" experienced by
divers. The extent of this occurrence depends on vertical velocities in
the tailwater, bubble size, the depth to which the bubbles are carried,
turbulence, initial concentration of dissolved gases, and water
temperature (TVA, 1978). As the water flows downstream the nitrogen
concentration tends to equilibrate with those of the atmosphere. The
length of the stream reach before equilibrium depends on the rate of flow
and the physical characteristics of the stream channel (TVA, 1978).
Salt Concentrations.
When water evaporates, any salts present are left behind, causing an
increase in salt concentration. In arid regions and areas where salts
are relatively abundant in the watershed runoff, salinity levels can
reach critical proportions. Reservoirs, by virtue of their increased
surface area, increase losses in regional water budgets through greater
evaporation. Salinity may also increase in reservoirs receiving return
flows from irrigation. Many freshwater fish and other aquatic life that
are intolerant to relatively high or fluctuating salt concentrations
cannot survive such conditions. Water with critical salt levels is also
difficult and expensive to treat for human consumption.
Not all reservoirs have a negative effect on salt concentrations.
For example, some flood control reservoirs in the arid west store high
quality flood flows in the spring and release required flows in the fall
and winter, greatly improving salinity downstream. In Lake Mead,
dissolved materials normally kept in solution by carbon dioxide are
precipitated in the reservoir, thus decreasing salinity downstream.
II-6
-------
Sediment Movement.
Waters received by reservoirs generally carry higher sediment loads
than waters released from reservoirs. Suspended sediment received by the
stream from the upland erosion, along with channel erosion, is carried
downstream. When the streams enter a quiet body of water, the sediment
load drops out. Reservoirs, therefore, act as large settling basins.
Reservoirs may fill up as a result of this process which must either be
anticipated in design or mitigated by dredging.
Often, horizontally stratified currents within a reservoir, called
density currents, carry greater sediment loads than the rest of the
reservoir. These currents may then be tapped for release downstream by
regulating multiple-outlets, a feature common in many newer dams.
Regulation of these currents facilitates sediment releases and can reduce
the frequency of dredging and sluicing operations. Episodic intentional
releases of silt laden bottom waters to rid the impoundment of silt can
cause short term high turbidities, sudden unwanted depositions,
smothering of aquatic life, and unsuitable conditions for fish spawning.
Alternatively, low suspended solids releases can cause increased scouring
of stream channels as the stream seeks an equilibrium of sediment load.
Furthermore, sediment can be partially or completely composed of
organic material. The organic material can decompose by biotic action
and cause oxygen depletion in the water column over the sediment. This
oxygen depletion is called sediment oxygen demand. It can be significant
in the balance of oxygen within reservoirs. The organic material in the
sediment can come from point or nonpoint sources, organic growths in the
inflowing stream or from the organic processes within the reservoir
itself.
Sediment changes do not always have to occur. If a reservoir is
operated to convey sediment density currents through the reservoir, in
time an equilibrium can be achieved where the inflow and outflow of
suspended sediment are equal.
Flow Regulation.
Dams typically cause modified flow conditions when operated for water
supply, recreation, hydroelectric power, or other uses. The use of
reservoirs to regulate streamflow can have a positive effect on water
quality. If the dam is operated such that low flows are augmented, a
stream's assimilative capacity can be enhanced and the concentration of
pollutants already present downstream can decrease (provided the
reservoir water has lower pollutant levels). When dams are used for
flood protection, extremely high flows are avoided, as are associated
suspended sediment load problems and other serious water quality problems
related to flood water.
If, however, the dam is operated such that it produces a lower than
normal low flow, downstream assimilative capacity for downstream
discharges may be reduced. A lower than normal low flow can also greatly
affect aquatic habitats by increasing water temperatures, decreasing
II-7
-------
surface area, depth of riffles and pools, and stream width. There is much
less habitat, reduced food-chain organisms, reduced fish nursery and
spawning acreage, and fish are more exposed to predators and more
susceptible to disease (Schwiebert, 1984). Tailwaters without minimum flow
discharges greatly reduces aquatic insect production since major shoal and
riffle habitat areas are exposed (Pfitzer, 1974). A higher than normal
high flow can create a severe scour and bank erosion problem as well as the
possibility of flooding. High flows can produce velocities that some
aquatic biota cannot tolerate unless some "slack water" habitat is
available. Flow regulation can sometimes improve the composition of the
aquatic community. It has been suggested that the retarding of storm flows
is a primary reason for the much larger benthic populations and improved
species diversity found downstream of southern SCS flood retention
impoundments (SCS, 1979). For intermittent streams, the longer duration
of flow by the release of stored flood flow may cause a benthic population
improvement (SCS, 1979).
Impoundments also can have a chemical stabilizing effect on the
receiving streams because the impoundments reduce the inflowing peak
concentrations. Chemical concentrations would fluctuate more erratically
in the inflowing stream than in the receiving stream (SCS, 1979). One
positive use of dams that has been proposed for the Yakima River in
Washington is increasing the upstream storage of a reservoir for fish flow
enhancement. It is predicted that some 340 miles of spawning, rearing and
resting areas for steelhead and salmon in the Yakima River System can be
improved by this proposal (Dompier and Woodworth, 1980).
Reaeration Denial.
Dams slow waters down, decrease turbulence and thus decrease natural
transfer of oxygen from the air to the water by reaeration. At a non-power
dam with a spillway, the rapid flow of the water over the spillway and
subsequent tail water turbulence, promote reaeration of the tail waters.
In power impoundments, water reaches the release point by transport in a
closed conduit having given up some of its kinetic energy to the turbines.
Reaeration under this condition partially depends on the nature of the
point of release, with submerged releases providing less aeration than
free-falling releases. The retrofitting of existing unused mill ponds and
impoundments with add-on low head small hydropower turbines is a particular
example of where this effect may be noticeable.
Fish Entrainment.
Fish entrainment is the capture and passage of fish through turbine
machinery during power generation operations. In addition to the reduction
of fishery resources of the reservoir system, the discharge of dead fish
and their remains can cause water quality problems from an aesthetic
viewpoint. If discharged in significant quantities, the dissolved oxygen
demands of their decomposition may depress available oxygen. Because there
has been much recent interest in this issue, fish entrainment is mentioned
in this report even though it is not a water quality issue per se.
II-8
-------
Toxics Accumulation.
Sediments trapped behind dams can be contaminated with toxic metals
and/or organics. Normally, these toxic-bearing sediments are not
transported downstream unless the sediments are disturbed as a result of
dredging, dam removal, or dam failure. An example of this phenomenon
occurred in the Hudson River where large amounts of polychlorinated
biphenyls (PCBs) were transported downstream when the Fort Edward Dam was
removed in 1973 (Carcich and Tofflemire, 1982). Analysis of toxics
accumulation behind dams was not included in this report due to
insufficient data on a national level.
II-9
-------
III. EPA ANALYSES
OBJECTIVES
The previous chapter describes the generic effects that are possible.
The objective of this chapter is to make quantitative estimates of the
national scope of the effects that dams have upon water quality.
This chapter summarizes the methodology for estimating the scope of
effects. Despite the fact that a statistically valid, quantitative
approach is used, the estimates presented are highly qualified due to
limitations in data availability. These estimates are based upon
publications, written by experts in this topic, and upon monitoring data,
contained within the EPA STORET water quality data repository. Two
parallel paths are followed. What current publications are enlightening
about the scope of water quality effects of dams? What can be inferred by
examination of a random sample of the population of all 68,155 dams?
The objective is to make the estimation as simple, straightforward,
and accurate as possible. The goal is to communicate the essential and
basic aspects of dam water quality effects to nonspecialists.
Qualifications are made throughout about the preliminary nature of the
quantification.
Given the overall goal of quantification, a number of conditions are
imposed upon its achievement:
No new field data are involved in the analyses, although new
data would further reduce uncertainty.
The findings are to be unbiased and scientifically sound.
Results are to be reproducible.
Assumptions and limitations are to be stated.
The organization of this chapter is directed at answering the
following questions.
How many dams of various types are there?
Where are the dams?
III-l
-------
What elementary theories and models can be utilized, given the
information at hand, to assist the reader to understand some
basic aspects of dam behavior? Note that the appeal of simple
models is to provide generalities for problem identification and
enlightenment and not to solve site-specific problems. Models
appropriate for site-specific problems tend to be much more
complicated and out of reach of the available resources of this
report.
possibly
What is the extent of the tailwater dissolved oxygen effect?
How many impoundments have the potential to stratify,
facilitating water quality problems?
How many dams alter water
the impoundment?
quality as streamflow passes through
What is the extent of
requirement for imminent,
phosphorus enrichment, which is one
ongoing, or advanced eutrophication?
The first three questions are addressed in the following section on
"Preliminaries" or in Appendix C "Analysis Supplement." The last four
questions are addressed in four subsequent sections of this chapter
titled "Mixing Analysis," "Dissolved Oxygen in Dam Tailwaters,"
"Upstream/Downstream Comparison of Water Quality," and "Phosphorus
Enrichment Analysis" respectively. For each of these latter questions,
an attempt is made to identify any correlation between dam type and water
quality. The mixing analysis and tailwater DO analysis are related since
a poorly mixed (stratified) reservoir is much more likely to have
tailwater DO problems.
PRELIMINARIES
This section presents a logical and
reader in following the various analyses.
their associated data are discussed.
and the
modeling
qualified
associated
strategic context to
The population of
aid the
dams and
Ancillary data, not in the Corps of
Engineers' data base, but added to random samples to fill in missing
data, are also discussed. The general approach to sampling is presented
topic of appropriate sample size is addressed. Elementary
tools, which are used in the chapter, are discussed and
in Appendix C, as is the significance of correlations
with various sized samples.
III-2
-------
The Population and Associated Data.
The Corps of Engineers' data base includes 68,155 dams built up to
the year 1980. This data base includes the following basic data:
Coordinates: latitude and longitude; this allows a particular
dam to be isolated and ancillary data, such as for water
quality, obtained.
V = volume of normal pool (acre*feet).
H = hydraulic height of dam (feet).
Q = flow capacity of the spillway to pass floodwaters (cfs).
Installed hydropower of 100 kilowatts (kW) or more (100s of kW
if known).
Other identifiers: Corps of Engineers assigned number, state,
name, uses such as power, flood control, etc.
The limitations of the Corps of Engineers data base are the
following:
There is no information on water quality
Dams built since about 1980 are not included.
Dams with less than 100 kW of power generation capacity are not
labeled; there are numerous small hydroelectric projects in the
United States.
Small impoundments with volumes less than 50 acre*feet and
having dams less than 25 feet high are excluded.
The data base has no information on outlet level. Such
information would be difficult to develop for the large numbers
of dams in the United States.
To get numerical perspective on this population of dams, there are:
68,155 dams overall.
1,091 dams that are known to have more than 100 kilowatts of
installed power.
301 dams that have more than 30 megawatts (MW) of installed
power.
2,125 "large" dams with over 10,000 acre«feet of storage at the
normal pool elevation. Of these, 424 have 100 kW or more of
installed power, and 1,701 have no installed power.
66,030 "small" dams with less than 10,000 acre-feet of storage.
III-3
-------
The dam inventory is partitioned, as to potential for incidence of
water quality effects, using the knowledge contained in the generic
effect descriptions and recommendations to the authors of this report
from TVA experts:
Large dams
tend to
nutrients.
have potential
stratify and
for significant
have sufficient
effects because they
detention to trap
Dams with low-level outlets transmit water quality effects
downstream associated with the hypolimnetic impoundment layer.
It should be noted that power dams are most likely to have low-
level outlets. Large nonpower dams may also have low-level
outlets. Small dams are much less likely to have such outlets.
With this logic to support the partitioning, the Corps of Engineers'
Dam listing is divided into three parts based on the criteria summarized
in Table III-l. The partitioning is intended to focus resources on those
subsets of the dam inventory wherein significant water quality effects
are most likely. The geographic distributions of the three types of dams
are depicted in Figures III-l, 2, and 3.
Table III-l. Partitioning Criteria.
Type Criteria
Number
Large Power Dams
Large Nonpower Dams
Smal1 Dams
Over 10,000 acre-feet and over 424
100 kW of installed power.
Over 10,000 acre-feet and having 1,701
no installed power.
All dams under 10,000 acre-feet 66,030
(including 667 that have power).
Sampling.
The general approach is to conduct a random sample of each of the
partitioned data sets of dams in the Corps of Engineers data base. The
sample is then utilized to determine some or all of the properties of the
partitioned populations. For practical reasons of time and manpower
constraints, it was decided that a sample size of 40 dams in each of the
three partitioned data sets would suffice. The principal reason for a
combined sample of 3 x 40 = 120 dams, is a large effort requirement to
secure ancillary data for each dam.
Once a random dam is identified, ancillary data are sought. These
ancillary data elements include: impoundment surface area, length of
impoundment along the flow axis, and average annual inflow. These
geometric data are secured from USGS topographic maps and EPA's River
III-4
-------
Population of Large Power Dams
Figure III-l
Population of Large Non-Power Dams
Figure III-2
Population of Small Dams
Figure III-3
III-5
-------
Reach File. Also sought as ancillary data are water quality data above,
within, and below the impoundment associated with the dam. These water
quality data are obtained from EPA's STORE! data repository using methods
described in detail later in this chapter.
The partitioned random sample of dams is presented in Appendix B.
Within this appendix are dam data taken from the Corps of Engineers' data
base and ancillary data. Some, or all, of the following data elements
are presented: identification number, name, installed power capacity,
volume, hydraulic height, spillway capacity, latitude, longitude, mean
inflow, area, length, Froude number, retention coefficient, mean
phosphorus concentrations, and identification of dams having ancillary
water quality data. (The Froude number and retention coefficient are
explained later in this report.)
Obtaining Sample Dams
The sample of dams to be considered is randomly drawn from records
contained in the 68,155-dam data base of the Corps of Engineers National
Inventory of Dams. The various mathematical and statistical procedures
are performed using the Statistical Analysis System (SAS), a commercial
statistical software package (SAS Institute, 1982).
The dam population records are placed in a data base where analytical
and statistical operations are performed. In all, 68,155 records were
placed into the data base. The data base is sorted into the three dam
categories: large power, large nonpower, and small dams. Each of these
three categories represents a specific population. A random sample from
each of these populations is developed using a SAS uniform distribution
random number generating function. The SAS random numbers are used to
select 40 random dams from each sub-population for a total of 120 random
dams. A list of the random sample dams appears in Appendix B. The
sample is partitioned to allow sufficient representation of the large
power, large nonpower, and small dams. However, since the populations
are of different sizes, the confidence bounds surrounding each sample
vary.
Other Logical Checks and Data Sources
The sample size of 40, relative to its adequacy to represent the
partitional data sets of dams, is discussed in Appendix C. It is shown
that the theoretic confidence intervals are rather large, indicating that
larger samplings are desirable to further define water quality attributes
of dams.
After selection of each 40 dam random sample set, various checks are
performed to confirm the representative nature of the sample. The
geographic distribution is checked and the sample frequency functions are
compared to the population frequency functions. The hypothesis is tested
so that the sample and population are from the same distribution with 95
percent certainty. These checks are presented in Appendix C.
III-6
-------
The Corps of Engineers dam data contained in the data base and
associated with the random sample dams are supplemented with ancillary
data. These data are:
Estimates of pool area, impoundment length, and average inflow
using equations found in Appendix C.
Water quality data at upstream, pool, and downstream sites (as
described in a subsequent section).
In addition to the random dams, information on other dams studied by
the Oak Ridge National Laboratory, TVA, and the CE are incorporated into
this chapter.
MIXING ANALYSIS
The mixing analysis is the first of four major analytical efforts
undertaken for this report. An approach is developed to categorize the
stratification potential (indicative of little mixing) in the population
of impoundments. This analysis is based on the Froude number which is
suggestive of stratification. The potential for stratification should be
investigated as an indication of poor mixing which, when accompanied by
large oxygen demands, may result in low dissolved oxygen content and
accompanying water quality problems in bottom waters. The potential for
dissolved oxygen problems is estimated in two analyses (the Froude number
analysis and the Oak Ridge National Laboratory analysis).
Relationship of Froude Number to Mixing.
According to TVA, the potential for cold, deoxygenated hydropower
releases from the hypolimnion of a reservoir is due largely to the effects
of thermal stratification (TVA, 1987a). The degree of thermal
stratification depends on hydrologic and morphologic characteristics that
vary significantly across TVA reservoirs. Many strategies for release
improvement either influence or are influenced by thermal stratification.
When such strategies are under consideration for a wide range of
reservoirs, it is useful to have a system whereby reservoirs can be ranked
according to their stratification potential.
Thermal stratification insulates the reservoir hypolimnion from
warming, inhibits mixing with the epilimnion, and sharply reduces natural
reaeration of the hypolimnion. Without replenishment from the .surface or
tributary inflow, hypolimnetic oxygen can be depleted by organic
decomposition and by respiration of aquatic plants. The effects of other
important variables such as reservoir operations and inflowing organic and
nutrient loads are highly coupled with thermal stratification in producing
the ultimate water quality of the releases. In fact, these other variables
can interact to produce significant vertical oxyclines (oxygen gradients)
in the hypolimnion even in the absence of significant thermal
stratification. In the southeastern United States strong thermal
stratification generally produces a strong ovycline. Therefore, when used
in combination with measures of other important variables, some index of
stratification potential can be useful for ranking the potential for low
DO, as well as cold temperatures in low-level releases. In this sense, the
subsequent DO tailwater analysis and this analysis each support the other.
III-7
-------
The densimetric Froude number represents the ratio of inertia! forces
imposed by the longitudinal flow to gravitational forces within the
stratified impoundment; it is therefore a measure of the degree to which
flow can alter the internal density structure of the reservoir. Small
Froude numbers (less than about 0.3) indicate strong stratification
potential, while larger Froude numbers indicate weak or intermittent
stratification, progressing to completely mixed conditions for very large
Froude numbers (TVA, 1987a). Furthermore, the Froude number can
distinguish the greater stratification potential of a short, deep
reservoir as compared to a long, shallow reservoir, even though the two
reservoirs may have the same flow to volume ratio. For example,
comparing reservoir flow to volume ratios and Froude numbers for Cherokee
and Fontana Reservoirs, within TVA, indicates similar mean annual
flushing rates, yet considerable difference in Froude number resulting
from morphological differences.
The data in Figure III-4 represent a summer drought condition wherein
inflows are low and hydraulic mixing potential is also low; however, the
reservoirs with lower Froude numbers are also reservoirs with high top to
bottom temperature differences. Figure III-4 shows average vertical
temperature differential (surface to bottom) in TVA reservoirs at the
peak of the 1986 drought versus the densimetric Froude number.
TVA |987; Fiq. XII-1)
Median Temp.
Difference
Ol
i.ae«-eei i.ee*+eee i.ee.+eei
Densimetric Froude Number
i.ee«+ees
Figure III-4
Stratification versus Densimetric Froude Number
for TVA Reservoirs.
III-8
-------
For these reservoirs under these conditions, the densimetric Froude
number is a significant index of stratification potential. The square of
the correlation coefficient, R^, which equals the percentage of explained
variation, is 69 percent for this sample size of 22 TVA reservoirs under
summer drought conditions. In other words, the variation in temperature
differential between the top and bottom layers of TVA reservoirs is 69
percent explained by a linear relationship. TVA has suggested use of a
seasonal Froude number (e.g., using only summer flows, volumes, depths,
etc.) to improve the correlation, in lieu of annual average Froude
numbers that were used to develop Figure III-4. Also shown in Figure
III-4 is the F = 0.3 demarcation line between strong stratification and
weak to intermittent stratification to mixed conditions. The vertical
line defined by F = 0.3 and the horizontal line defined by the median
temperature difference divide the 22 dams into two shaded subsets shown
on Figure III-4. Eighteen of the twenty-two dams fall in the subsets. F
= 0.3 provides a reasonable demarcation. Also there is a cluster of dams
(6) around F = 10 that show the least temperature differential
substantiating that there is less tendency to stratify (more tendency to
mix) at high F values.
Findings.
The national Froude number cumulative frequencies are utilized to
make national estimates of mixing potential based on F = 0.3. The Froude
number does not describe a condition of water quality - it does indicate
the tendency of a reservoir to thermally stratify. A thermally
stratified reservoir is a poorly mixed reservoir. A poorly mixed
reservoir may have poor quality in the hypolimnion. If a reservoir has
low-level outlets and poor quality water in the hypolimnion, the
tailwaters may have poor quality. Poor, in this context, generally
refers to low dissolved oxygen and the presence of iron and manganese.
The Froude number frequencies are estimated using the elementary
model, F = K (L/D)(q/V), where K depends upon units and F is
dimensionless. A threshold for theoretic flow separation is F = 1/n =
0.3. This theoretic value agrees with the suggested TVA threshold.
Above this level, there is a tendency to mix. At much less than 0.3, the
tendency to stratify is strong.
The mixing tendency classification scheme enables determination of
dam population percentages shown in Table III-2. Data for small dams are
presented in Appendix E. Impoundments that are strongly .mixed are
unlikely to have as severe tailwater quality effects as stratified
impoundments, provided other important factors such as inflow loadings
and sediment oxygen demands are roughly equal. Strongly stratified
impoundments may have water quality effects which may be transmitted
downstream by low-level outlets. Without field inspections or some type
of intensive polling, the incidence of such outlets is problematic.
However, low-level outlets are more typical with power dams, but may
occur in any dam.
III-9
-------
Table III-2.
National Potential Mixing Percentages for Large Dams.
Mixing Tendency
Potentially Strongly Stratified
F < 0.3
Potentially Weakly or Inter-
mittently Stratified or
Completely Mixed.
F > 0.3
Missing Data
TOTALS
Power
171
217
36
424
Nonpower
(40%) 631 (37%)
(51%) 742 (44%)
(9%) 328 (19%)
1,701
The F estimates utilized a large amount of large and
some small dam data to generate values for L, D, q;
however the linear estimators are forced through zero
to enable small dam extrapolations of F. Therefore,
the small dam tallies of potential strong
stratification may not be as valid as the large dam
tallies; these small dam tallies appear in Appendix E.
111-10
-------
DISSOLVED OXYGEN IN DAN TAILWATERS
The second of four major analytical efforts undertaken is an
evaluation of dissolved oxygen concentrations in the tailwaters below
impoundments. In particular, this effort reviews a study by ORNL and
provides analogous results for the random sample and case study data
collected for this report. Unlike the mixing analysis, this effort
identifies dissolved oxygen problems as defined by the exceedance of a
specific numerical criterion.
Oak Ridge National Laboratory Study.
The U.S. Department of Energy has supported a Small-Scale Hydropower
Development Program. Under this program, the Oak Ridge Laboratory
conducted a water quality study of hydropower dam tailwaters (Cada, et al.
1983). The objective of the study is to estimate the extent of the
problem of DO in tailwaters for small-scale hydropower development. This
is analyzed by pairing operating hydroelectric sites with dissolved oxygen
measurements from nearby downstream U.S. Geological Survey water quality
stations. These data are used to calculate probabilities of noncompliance
(PNCs), that is, the probabilities that dissolved oxygen concentrations
in the discharge waters of operating hydroelectric dams will drop below
5 mg/1. Incorporated within this study are several technical judgments:
A probability of noncompliance (PNC) is chosen as the
statistic of interest because it directly addresses the
question "What are the chances that discharges below a
hydroelectric dam will violate dissolved oxygen criteria?"
PNC is defined as the probability that concentrations of
dissolved oxygen will be less than some specified value.
Because thermal stratification and resultant oxygen depletion
are seasonal phenomena, two probabilities are calculated for
each site: one for the summer months (July, August,
September, and October); and another for the remaining months,
defined as winter months.
The EPA criterion of a minimum dissolved oxygen concentration
of 5.0 mg/1 is utilized to assess the potential for water
quality problems at small-scale hydroelectric projects
(defined by the U.S. Department of Energy (DOE) as-having a
potential capacity of 30 MW or less).
The data base for the U.S. Corps of Engineers National
Hydropower Study is used, containing 15,300 existing dams.
Dissolved oxygen data are acquired from the National Water
Data Storage and Retrieval System (WATSTORE), maintained by
U.S. Geological Survey. By cooperative arrangements, these
data are a large subset of EPA's STORET data base.
III-ll
-------
Selection of operating hydroelectric dams used in this study is
based on the existence of appropriate water quality data in the
USGS data base. A water quality monitoring station was
considered appropriate if it (1) was downstream from the dam,
(2) was within 4.8 km (3 miles) of the dam, and (3) had more
than two measurements of dissolved oxygen concentration. No
random sampling is included in the Oak Ridge study.
Of the 15,300 potential dams, 65 small-scale hydroelectric sites were
selected for determination of PNCs. The study showed effects associated
with season, geography, and whether or not the facility had greater or
less than 30 MW of capacity. The PNCs tend to be higher in the summer,
east of the Mississippi, and for facilities with greater than 30 MW.
Findings.
The method of analysis used in the Oak Ridge Study is applied in
this study. The purpose is to determine if the methods and procedures of
this study can reproduce the Oak Ridge results and, if so, to strengthen
their previous findings. A summary in Table III-3 presents the
comparison of the PNCs for the Oak Ridge Study and for this study.
The Oak Ridge PNCs are based on 139 power dams. Figure III-5
presents a geographical breakdown of the regions for the Oak Ridge study.
Table III-3
power dams
(Some sites
winter data.
power dams
data, which
Oak Ridge
summarizes the ORNL data as well as data available for 23
from the large power sample of dams and the case studies.
in both the ORNL and EPA analyses lacked either summer or
) The individual PNCs are presented in Appendix B for these
as well as for the nonpower dams. The STORET water quality
includes WATSTORE records, produce results comparable to the
Study. A significant finding is the confirmation of the
seasonality of PNC levels and their relative magnitudes reported in the
Oak Ridge Study. From the perspective of where and when dissolved oxygen
levels are below EPA criteria, the combined PNC results indicate:
PNCs vary regionally, seasonally,
facilities.
and with size for generating
In the Ohio Valley and the Southeast
probability of low DO - 0.31 to 0.56.
highest in summer months.
there is a significant
The probability is
The next section makes dissolved oxygen comparisons of annual means
to establish differences above and below dams. The use of the annual
mean tends to reduce the ability to detect significant differences.
Thus, the findings associated with mean annual effects should tend to be
conservative. For example, if 10 percent of a sample shows significant
difference on a mean annual comparison, the PNC seasonal results lead one
to the conclusion that the seasonal effect will be larger than 10
percent.
111-12
-------
D NE
O Great Basin
§ Lake States
D Ohio Valley
Pacific Coast
Rocky Mountain
Great Dlains
Figure III-5.
Geographical Breakdown of Regions for
the Oak Ridge National Laboratory Study.
111-13
-------
Table III-3.
Probabilities of Non-Compliance with
5 mg/1 Dissolved Oxygen for Power Dams.
Summer Season
Location
Great Basin
Great Plains
Lake States
Northeast
Ohio Valley
Pacific Coast
Rocky Mountain
Southeast
(n) Mean
Summer Season
Great Basin
Great Plains
Hawaii
Lake States
Northeast
Ohio Valley
Pacific Coast
Rocky Mountain
Southeast
(n) Mean
Winter Season
Great Basin
Great Plains
Lake States
Northeast
Ohio Valley
Pacific Coast
Rocky Mountain
Southeast
(n) Mean
Winter Season
Great Basin
Great Plains
Hawaii
Lake States
Northeast
Ohio Valley
Pacific Coast
Rocky Mountain
Southeast
(n) Mean
Oak Ridge
National Lab
n
3
6
-
3
16
19
6
18
(71)
3
1
1
5
15
3
7
9
17
(61)
3
6
-
3
18
19
6
18
(73)
3
2
1
6
16
3
6
9
18
(64)
Capacity >
PNC
0.004
0.182
-
0.144
0.404
0.039
0.052
0.308
0.162
Capacity <
0.373
0.000
0.000
0.043
0.066
0.111
0.003
0.027
0.131
0.084
Capacity >
0.000
0.008
-
0.005
0.096
0.000
0.000
0.039
0.021
Capacity <
0.0274,{
0.508
0.000
0.005
0.010
0.001
0.000
0.000
0.010
0.007
This
30 MW
n
_
-
-
-
5
4
2
2
(13)
30 MW
-
1
-
4
-
3
2
(10)
30 W
-
-
-
-
5
4
2
1
(12)
30 MW
1
-
3
-
3
2
(9)
Study
PNC
_
-
-
-
0.560
0.053
0.000
0.170
0.196
_
0.000
-
0.123
-
0.220
0.190
0.137
-
-
-
-
0.102
0.015
0.000
0.080
0.049
0.000
-
0.003
-
0.073
0.005
0.020
Presumed outlier.
111-14
-------
The Corps of Engineers has also performed an analysis of tail waters
for low dissolved oxygen (Kennedy and Gaugush, 1987). Their study
examined 73 Corps of Engineers hydropower projects concentrated in the
pacific northwest and southeastern regions of the U.S. As shown in
Figure III-6, low DO tailwaters are more frequently a problem in the
southeast. This figure is the result of a Corps of Engineers
questionnaire. A similar approach is presented in Chapter VI; the
Chapter VI presentation extends and broadens the Corps of Engineer
questionnaire approach to include TVA and Bureau of Reclamation projects.
• Severe DO Problems
o Minor DO Problems
A No DO Problems
Figure III-6. Distribution of 73 Hydropower Projects and
Their Tailwater Conditions. (Source: Kennedy and Gaugush, 1987)
111-15
-------
UPSTREAM/DOWNSTREAM COMPARISON OF WATER QUALITY
This section presents the third of four major analytical efforts
comprising this chapter. Water quality data retrieved from EPA's STORET
data base upstream and downstream of impoundments were collected and are
compared. Changes in water quality observed as the result of passing
through the impoundment are reported. The analysis is attempted on the
random sample for a number of water quality parameters.
Acquiring Water Quality Data.
The water quality data acquisition phase of this analysis is the
collection of ancillary water quality monitoring data for the sites
randomly selected from the population. The definition of the dam site
requires the manual study of USGS and hydrological unit maps to determine
the scope of the reaches to be studied. Reaches within the pool,
upstream of the pool, and downstream of the pool are determined. The
data search could reach out up to 20 miles from the dam, but such a
distant search was infrequent. Latitude and longitude are used to define
a "window" to be searched for water quality information.
If major changes in the hydrography occurred within these limits
(such as another dam five miles upstream) the "windows" are made smaller
to exclude extraneous, downstream backwater, or misleading data. The
types of water quality and related parameters retrieved for this study
are: mean streamflow, dissolved oxygen, phosphorus, nitrogen,turbidity,
water temperature, iron, manganese, hydrogen sulfide, chlorophyll-a, and
BOD5.
Retrieval of data from STORET requires the use of the latitude,
longitude polygon and the water quality parameters. The STORET system
retrieves water quality data inside the polygon "window". The water
quality data include station information (agency, location, name, depth
of sample) as well as data on the water quality parameters. The
upstream, downstream, and pool retrievals are placed into intermediate
data sets; which are combined into a unique dam site file. SAS can be
used to calculate the number of records, the maximum, minimum, mean and
standard deviation values for each parameter in this combined site file.
Information in Appendix B presents the incidence of ancillary water
quality data for each dam in the random sample.
The arithmetic mean, standard deviation, maximum, and minimum values
are calculated for each parameter at each dam. The resulting statistics
are then downloaded into an RBASE data base; RBASE is a commercial
software package that operates on personal computers. Of the original
sample containing 120 impoundments (40 large power [L-P]; 40 large
nonpower [L-NP]; and 40 small [S] dams), 65 impoundments (or 54%) had
data for one or more parameters (39 L-P, 21 L-NP, 5 S). The L-P dams
averaged 25 stations, L-NP averaged 8 stations (with the exception of
Lake Tahoe with 301 stations), and small dams averaged 8 stations
reporting data. STORET retrieval summarizations are presented in
Appendix C.
111-16
-------
Limitations of the Water Quality Data.
For many random sample dams, water quality data are not obtainable
with the methods described herein. For such dams, then, an effect is
either present or it is not, and this simple either/or provides a boundary
on the estimate. In some cases, a very large number of the dams in the
random sample have no water quality data.
Consider the following example of a hypothetical effect above and
below a sample of 40 dams:
10 dams have significant effect
18 dams have insignificant effect
12 dams have no data
To estimate the upper bound of the number of dams having a significant
effect, assume all dams without data (12) have a significant effect,
therefore:
10 + 12
Upper bound = = 55 percent
40
Conversely, a lower bound can be approximated by assuming all dams without
data (12) have an insignificant effect as follows:
10
Lower bound = = 25 percent
40
Thus, one can assume for this hypothetical example that between 25 percent
and 55 percent of the dams exhibit a significant effect.
The difference in means testing is based on all data available at a
site. The tailwater analysis of dissolved oxygen highlighted the fact that
summer dissolved oxygen depressions are more prevalent than winter
depressions. Therefore, there may be sites that show no significance in
dissolved oxygen differences by annual means testing that would be
significant if summer data alone were examined. This may be true for other
parameters, particularly iron and manganese.
A data limitation may be the fact that many agencies enter data into
STORET. Such data may have different levels of accuracy and quality
control. The probability of noncompliance study conducted herein gives
similar results to the Oak Ridge National Laboratory. Furthermore, an
audit of the agency codes that identify the STORET contributors gave the
following agencies: USGS, TVA, several Corps of Engineers districts, EPA,
and water quality state agencies in Florida, Michigan, Wisconsin,
California, Pennsylvania, and Iowa. These agencies' data are the data used
to support the tailwater dissolved oxygen work discussed herein and provide
insight into typical sources of STORET monitoring data.
111-17
-------
Statistical Comparison of Means.
This section discusses the statistical comparison of the means of
water quality samples collected above and below the dam. Samples from
the pool are excluded except for the phosphorus analysis presented later.
The analysis applies statistical significance testing to the difference
of means assuming the difference is normal.
There are three possible situations:
The difference is positive and statistically significant; in
this case, the average concentration is higher below the
impoundment than above the impoundment.
The difference is positive or negative and statistically
insignificant; for this case, there is no significant difference
above and below the impoundment.
The difference of means is negative and statistically
significant; in this case, the average concentration is lower
below the impoundment than above the impoundment.
If there is no change in water quality above and below the dams, the
expected distribution of differences^ would be 5% decrease, 90% the same,
and 5% increase. That is, if one explored a large sample of dams having
no effect on mean water quality, the distribution would be 5%/90%/5%.
If, on the other hand, there is a difference, other percentages would
appear for the increase and decreasing categories - say 10% or 20%.
The statistical approach is straightforward. The mean of a sample
from any distribution tends to be normal by the Central Limit Theorem.
The differences of two means, in this case, the mean upstream and the
mean downstream water quality, also tend to be normal because sums and
differences of normal variates are themselves normal. The variance, or
square of the standard error of the mean difference is the sum of the
variances associated with the individual means, each computed with their
respective sample sizes (Hoel, 1951). With this information, the
hypotheses are that:
the positive difference of means is significant, or
the negative difference of means is significant.
The hypotheses are accepted with 95 percent confidence if a positive
difference is greater than 1.65 times the standard error of the mean
difference or an absolute value of a negative difference is greater than
the same product.
•'•Applying a 95% confidence interval for a "one tailed" test on each
end of the distribution of the mean difference.
111-18
-------
Findings.
There are 1701 large nonpower dams having over 10,000 acre-feet of
storage. Twenty-five percent, or less, of the random samples of nonpower
dams have water quality data both upstream and downstream of the
impoundment. Tallies for those dams having water quality data are found
in Appendix D.
There are 66,030 "small" dams having less than 10,000 acre-feet of
normal pool volume comprising 96.8 percent of the population of Corps of
Engineers dams. Of the 40 small random dams in the sample, only 5 had
some form of water quality data. Therefore, there is very little water
quality monitoring evidence to indicate the effects of small
impoundments. What monitoring results there are, as well as descriptive
and site specific information on southeastern small dams are presented in
Appendix E.
The large power dams have installed power and over 10,000 acre-feet
of storage. The results of comparing water quality means are presented
in Table III-4. Over half the random sample of dams have data on
temperature, dissolved oxygen, phosphorus, and TKN both upstream and
downstream of the impoundment. Reasonable bounds can be stated for most
of the parameters.
Table III-4.
Upstream/Downstream Water Quality Changes
For Large Power Dams.
Parameter
Temperature
Dissolved Oxygen
Dissolved Oxygen*
Iron
Manganese
Phosphorus
TKN
Total Nitrogen
Dams
Lacking
Necessary
Data
11
11
11
23
30
16
17
31
Dams with Upstream and
Total Having
Data
29
29
29
17
10
24
23
9
Signif i
Increase
11
8
6
4
0
5
4
1
Downstream Data
cant
Decrease
5
9
9
3
1
12
10
2
Insignifi-
cant**
13
12
14
10
9
7
9
6
* nearest station to dam
** indicates no change in water quality
111-19
-------
An impoundment can either increase or decrease tail water temperatures
depending upon the level of the outlets. Surface outlets would release
warm surface waters and low-level outlets could release colder
hypolimnetic waters of stratified impoundments. Of the sample of 40
random large power dams, between 27.5 percent and 55 percent increase the
annual average downstream temperature and between 12.5 percent and 40
percent decrease annual average downstream temperatures. For large
nonpower and small dams, the data are too limited to provide reasonable
effect estimates.
For hypolimnetic low-level releases, the dissolved oxygen may be
lower than upstream with corresponding increases in soluble iron and
manganese. Of the sample of 40 random large power dams, between 22.5
percent and 50 percent show significant annual decreases in dissolved
oxygen. This effect is evidenced in the annual data; the numbers of
significant water quality changes would possibly be higher for a
comparison of seasonal means.
PHOSPHORUS ENRICHMENT ANALYSIS
This analysis is oriented to determination of the extent of possible
nutrient enrichment of impoundments. Over enrichment can lead to water
quality effects. This analysis is the fourth of four major analytical
efforts undertaken for this report. The approach is to estimate the
average phosphorus concentration in the pool. Then, a simple tally is
made to determine the numbers of dams having average concentrations
exceeding the EPA suggested "Gold Book" guidance value of 0.025 mg/1
(USEPA, 1986). There are several points to consider in evaluation of
this method:
The EPA guidance value is "suggested" and not a standard.
Phosphorus enrichment is only an indication of eutrophication,
although 50 to 90 percent of the variation of other trophic
state measurements are predicted with water column phosphorus
concentrations (Sobotka and Company, Inc., 1986).
The approach to "filling in" missing data uses Vollenweider's
model discussed in Appendix C. It is an elementary approach,
but it works with the available information and is appropriate
for screening.
Phosphorus data on large power dams are relatively plentiful,
but not for other dams. Thus, the results pertain to large
power dams. Data gathered for large nonpower and small dams
appear in Appendices D and E, respectively.
111-20
-------
Relationship of Phosphorus Enrichment and Eutrophication.
Ambient phosphorus concentrations do not measure eutrophication
although they do indicate a potential for eutrophication. EPA recognized
that a number of specific exceptions can occur which may reduce the threat
of phosphorus as a contributor to lake eutrophication:
naturally occurring phenomena may limit the development of
plant nuisances;
technological or cost-effective limitations may help control
introduced pollutants;
waters may be highly laden with natural silts or colors that
reduce the penetration of sunlight needed for plant
photosynthesis;
waters may have no history of plant problems due to various
morphometric features such as steep banks, great depth, and
substantial flows;
in some waters, nutrients other than phosphorus are limiting
to plant growth; the level and nature of such limiting nutrient
would not be expected to increase to an extent that would
influence eutrophication; and
in some waters, phosphorus control cannot be sufficiently
effective under present technology to make phosphorus the
limiting nutrient.
A brief analysis was performed on the nitrogen and phosphorus data
to confirm the hypothesis that phosphorus enrichment is a reasonable
screening indicator for overall potential impoundment enrichment. This
was accomplished by examining total nitrogen concentrations, total
phosphorus concentrations, and the N:P ratio for ten sites with both
phosphorus and nitrogen data for the pool as shown in Table III-5.
The assessment of nitrogen versus phosphorus limitation is often
performed using the N:P ratio (USEPA, 1985; USEPA, 1978b). If this ratio
is greater than a given value, the water body is considered to be
phosphorus limited. The appropriate ratio is dependent on the types of
algae and macrophyte growth that may occur and usually ranges between 7
and 15 (USEPA, 1985; USEPA, 1978b). Five of ten sites exhibit N:P ratios
greater than or equal to 15, and seven of ten sites show ratios greater
than or equal to 7 suggesting phosphorus limiting conditions.
111-21
-------
Table III-5
Selected Summary of Nitrogen and Phosphorus in the Pool
("Yes" implies enriched levels)
Site N:P Ratio P > 0.025 mg/1* N > 0.375 mg/1**
AK00001
AR00174
CA00813
CA10162
GA03742
ID00223
KY03048
MN00653
PA00924
TN03702
32.
3.
4.
37.
7.
4.
16.
16.
43.
12.
No (0.01)
Yes (0.03)
Yes (0.22)
No (0.01)
Yes (0.08)
Yes (0.09)
Yes (0.06)
Yes (0.04)
Yes (0.08)
Yes (0.06)
No (0.32)
No (0.09)
Yes (0.84)
No (0.37)
Yes (0.59)
No (0.36)
Yes (0.95)
Yes (0.66)
Yes (3.46)
Yes (0.70)
* Phosphorus measurements may be made down to 0.01 mg/1 using
the single reagent method and 0.001 mg/1 using the automated
colorimetric ascorbic acid reduction method (USEPA, 1974).
** A concentration of 0.375 mg/1 (as N) corresponds to a
balanced condition of nitrogen and phosphorus with
phosphorus at the 0.025 mg/1 criterion level using an N:P
ratio of 15:1.
The phosphorus data in Table III-5 show that eight of ten sites are
labeled as "phosphorus enriched" based on the 0.025 mg/1 guidance value,
yet only five to seven are -estimated to be phosphorus limiting. However,
the thrust of this analysis is to develop an indicator of enrichment.
Examination of the corresponding nitrogen data suggests that high
phosphorus concentrations may indicate enrichment whether or not
phosphorus concentrations are limiting. For example, the site CA00813
exhibits extremely high nitrogen as well as phosphorus concentrations,
but an N:P ratio of only 4. Therefore, the indication of phosphorus
enrichment was suggestive of overall enrichment in spite of the fact that
phosphorus is not limiting.
Furthermore, the practical consideration is how to achieve nutrient
reduction for situations that are enriched. It appears, based on point
source control, that phosphorus reduction is much less expensive than
nitrogen reduction. Thus, it is economically possible to make a
previously nitrogen limiting situation into a phosphorus limiting case
and to achieve enrichment control. In other words, phosphorus is more
cost effective to remove than nitrogen. It makes operational sense to
judge enrichment with high phosphorus levels since it may be feasible to
reduce them and make phosphorus either become the (or become the more)
limiting nutrient.
In sum, there are two sites (AR00074 and ID00223) of the ten which
show enriched levels of phosphorus combined with lower levels of nitrogen
as defined in Table III-5; and the nitrogen level for site ID00223 is
only slightly below the specified nitrogen threshold (0.36 mg/1 versus
111-22
-------
0.375 mg/1). Therefore, these data support the use of phosphorus
enrichment as a key indicator of eutrophication while at the same time
demonstrating the importance of qualifying its use as an indicator.
Note, however, that since these data are not statistically
representative, these data cannot be used to quantify error bounds.
Nutrients in the water column and eutrophication are related and this
linkage is discussed in the chapter on generic effects. TVA evaluated
the trophic status of their reservoirs (TVA, 1983). They mention that
eutrophication, from the Greek word for "well fed," refers to progressive
fertilization of a water body and the changes in water quality that
result. Natural eutrophication, over geologic time spans of thousands of
years, is a normal process of aging. During lake aging, plant biomass
accumulates until rooted aquatic plants cover the entire bottom of a lake
and the basin fills with organic and inorganic sediments. Rivers and
man-made reservoirs with short retention times do not age in the same way
as lakes, but added nutrients may increase their biological productivity
to levels traditionally identified as "eutrophic."
TVA also recommends that "Evaluation of TVA reservoirs based on
trophic state indices devised for classification of natural lakes should
be avoided", and further states that "Models predicting in-lake
phosphorus concentrations from phosphorus loads assuming steady state
conditions and continuous stirred tank reactor behavior are inappropriate
for the evaluation of most TVA reservoirs and should be avoided."
However, a report performed under contract to EPA (Sobotka and
Company, 1986) points out the generally acknowledged understanding that
"...ambient concentrations of phosphorus bear a strong but not perfect
relationship to eutrophication response...algal growth, water
transparency, DO depletion, species diversity, etc." The report points
out that correlation studies of cross-sectional samples of more
sophisticated trophic state measurements and ambient phosphorus
concentration are on the order of 0.5 to 0.9. This indicates a
percentage of explained variation, R2, of 25 percent to 81 percent. The
same report then states the operational conclusion:
"In our view, this smallish imperfection in the relation-
ship between P concentration and ultimate water quality concerns
is a reasonable price to pay in exchange for the advantages in
implementability of an ambient P standard. An ambient P
standard is much easier to translate into permit limits and
other control decisions than water quality standards expressed
in other items — as narratives, as chlorophyll limits, or
as trophic states."
The 1986 EPA "Gold Book" on quality criteria for water presents a
rationale to support guidance for ambient phosphorus. To prevent the
development of biological nuisances and to control accelerated or
cultural eutrophication, a suggested phosphorus guidance value for lakes
and reservoirs is published by EPA at 0.025 mg/1. Thus, the simplest
model or approach is to use 0.025 mg/1 as a threshold level. If ambient
111-23
-------
phosphorus exceeds 0.025 eutrophication is suspected, and, in fact, EPA
is considering "control" and regulatory options to cut levels back as
evidenced by their "Gold Book" suggestions. From a pragmatic screening
standpoint, one can assign measured or predicted reservoir phosphorus
concentrations of greater than 0.025 to indicate either potential or
actual eutrophic condition.
TVA and CE Results.
There are reservations about using the Vollenweider model, or similar
models, for control decisions. However, it is instructive to review TVA
trophic status data (TVA, 1987a) for several TVA tributary reservoirs
reproduced in Table III-6. The Vollenweider predictions calculated by
TVA are based on a Vs = 10 meters/year, apparent settling velocity. The
uncorrected Vollenweider model used by TVA uniformly overestimates
phosphorus cncentrations. A corrected model is discussed in the next
section.
The square of the correlation coefficient, R2, of these 11 samples,
between Vollenweider prediction and measurement, is 44 percent. This is
a significant R2, but probably not strong enough to make site specific
decisions. However, R2 is sufficient to provide credibility to the next
section of this chapter.
Table III-6.
TVA Trophic Status Data.
Phosphorus Concentration (mg/1)
Site
Blue Ridge
Boone
Chatuga
Cherokee
Douglas
Fontana
Hiawassee
Norris
South Holston
Tims Ford
Watauga
Vollenweider
.015
.063
.017
.148
.062
.061
.022
.032
.029
.017
.044
Measured
.007
.022
.009
.021
.024
.008
.010
.007
.008
.009
.006
Kennedy and Gaugush (1987) also report data for phosphorus, nitrogen,
and chlorophyll-a taken from STORET for 47 southeastern Corps of
Engineers reservoirs and the same parameters for 299 reservoirs Corps-
wide taken from Walker (1981). These data for the growing season (April-
September) in the mixed layer (0-3 meters) are reproduced in Table III-
7. Kennedy and Gaugush concluded that southeastern reservoirs are
111-24
-------
typical of Corps reservoirs in terms of phosphorus and chlorophyll-a, but
much different with respect to nitrogen. Kennedy and Gaugush suggest
that reservoirs in the southeast may experience nitrogen limitation of
primary production.
Table III-7
Summary Statistics for Total Phosphorus, Total Nitrogen, and
Chlorophyll-a in the Mixed Layer (0-3 m) During the Growing
Season (April through September).
Variable Mean Mean
(mg/1) (47 Southeastern Reservoirs) (299 Corps-wide Reservoirs)*
TP
TN
CHL-a
0.044
0.372
0.0093
0.048
1.00
0.0093
* "Corps-wide" refers to a Corps reservoir data set described in
Walker (1981).
Phosphorus Retention Regression Model.
Of the 80 dams in the random sample with 10,000 acre*feet or more, 22
have average phosphorus data for inflows and pool concentrations that are
used in a correlation analysis. Small dam sites were excluded because
none of the sample had upstream and pool phosphorus data. The analysis
proceeds through the following steps:
The inflow phosphorus is utilized with the Vollenweider model to
make a prediction of pool concentrations. The apparent settling
velocity, Vs = 10 meters/year, is utilized.
The 22 predictions are compared to the 22 observations.
The observed values are analyzed in a least squares regression
analysis as a linear function of the Vollenweider calculated
values. The line is constrained to pass through the origin.
The data for this analysis are found in Appendix B and the results
are presented in Figure III-7. The correlation between observed and
calculated data is 0.72 which implies a significant R2 of 0.52. This
correlation is slightly better than the TVA results presented in the
previous section. The slope of the regression line is 0.65 indicating
that Vollenweider estimates, that utilize Vs=10, make a high prediction
of pool concentration. Thus, if one reduces the Vollenweider prediction
by 35 percent (multiply by 0.65), one obtains a "corrected" least squares
estimator of the phosphorus in the pool.
111-25
-------
c
0
CJ
0
0
Q.
0.30 -y-
0 23 -I
0.26 -|
0 24 -i
0 22
0.20
0 13
0.16
0.14 -i
0.12
0.10
0 08
0.06 -)
0 04
0 02
0 00
R = 0.52
n = 22
slope = 0.65
• = power
n = non-power
0.04 0.03 0.12 0.1C 0.2
Calculated Pool Phosonorus Cone, (mg/l)
0.24
0.29
Figure III-7. Vollenweider Model Performance.
Findings.
An estimate of nutrient
sample of dams is presented.
data in the impoundment
"corrected" least squares
discussed above.
enrichment potential derived from the random
The estimate is based on average phosphorus
pool, or lacking actual data, from the
estimator of the phosphorus in the pool
The actual data, and the "corrected" estimates which fill in missing
in-pool values when upstream values exist, are presented in Appendix B.
Even using the regressions for large nonpower dams, insufficient data
were available, for a representative analysis of the large nonpower dams
sample, therefore, only data for large power dams are summarized in Table
III-8. Data for large nonpower dams can be found in Appendix D. Using
the suggested EPA guidance value of 0.025 mg/l phosphorus in the water
column for the sample of large power dams, between 57.5 percent and 77.5
percent are phosphorus enriched.
111-26
-------
Table III-8.
Phosphorus Enrichment Results
for Large Power Dams
Number of Dams
Data Source
Observed Pool Data
"Corrected" Estimate
Missing Data
Total
No Phos.
Data
-
-
8
8
P < 0.025
6
3
-
9
P > 0.025
16
7
-
23
111-27
-------
IV. CASE STUDIES
INTRODUCTION
Fifteen case studies have been selected for presentation in this
chapter as a means of describing some water quality effects resulting
from the impoundment of water and methods available for mitigating
negative effects. The selection of case study sites was meant to
enhance understanding by providing specific examples of situations which
have occurred in the past and not to be representative of the
distribution of effects. The case studies also illustrate the mitigation
action which has been taken by the responsible agencies.
Case study suggestions and selections were made during the course of
several meetings with interested Federal agencies, including the Bureau
of Reclamation (BR), the Army Corps of Engineers, and the Tennessee
Valley Authority. Through their knowledge of particular sites, the set
of case studies was developed and the original text for this report was
provided. The text was edited only to place each case study in a common
format. Therefore, the information supplied reflects the assessment of
the respective agency (BR, 1987b; CE, 1987; TVA, 1987b).
The case studies are distributed throughout the country, as shown in
Figure IV-1, to show the effects that climate and location may have on
water quality conditions. For example, Lake Casitas is located in
southern California where almost all of the precipitation occurs in the
winter. This results in a highly stable stratified condition, leading to
an anoxic hypolimnion in the summer. The climate of an area will also
affect the period of stratification and occurrence of mixing. For
example, in southern states longer summers result in a longer period of
stratification.
Physical characteristics of a reservoir, including volume, length,
depth, surface, and shoreline length, may all be related to the observed
water quality in a reservoir or downstream. For example, reservoirs with
a large storage area in relation to drainage area will have a long
detention time, with a resulting increased degree of settling of
sediments and organic matter. This may result in increased decomposition
in the hypolimnion and a possible increased tendency towards
eutrophication. The depth of the reservoir will affect annual heat
budgets and the impoundment's resistance to mixing. Shoreline length
gives an indication of the extent of the littoral zone (the interface
between the land of the drainage basin and the open water of the lake).
IV-1
-------
McCLOUD (PGiE)
FLAMING GORGE 3R)
JOCK (CE)
£AU GALLE (CE)
o PERCY PRIEST (CE)
OLD HICKORY (CE)
SOONE (TVA)
FORT P. HENRY (TVA)
MORRIS (TVA)
RICHARD B. RUSSELL (CE)
CASITAS (BR)
CACHUMA (BR)
LTPER 3EAR CREEK (TVA)
GCTTERSVILLE (TVA)
MARK TWAIN (CE)
NOTE:
OPERATING AGENCY
GIVEN IN ( ).
Figure IV-1
Location and Name of Case Study Impoundments
The littoral zone is often a very productive habitat and also contributes
organic detritus to the aquatic system. Table IV-1 lists each of these
characteristics along with the name of the impounded river, the major
purposes, and the year the impoundment was filled.
Impoundment date is also important because water quality conditions
often change with time. For example, Richard B. Russell Lake presently
has low DO conditions in the hypolimnion caused by the recent inundation
of a forested area to form the lake. This condition should improve as
the reservoir ages. The newer reservoirs, such as Casitas Lake, were
designed with multilevel intakes and other measures as a means to
mitigate potential water quality problems. Since reservoirs act as sinks
for substances as they age, sediment, organic matter, metals and other
pollutants entering the reservoir will accumulate in the benthos and, if
resuspended, can have a negative effect on water quality.
The uses of a reservoir may affect the water quality conditions in
the tailwater since they will be a major determinant of the method of
withdrawal and frequency of flow release. For example, the water may be
released and aerated by being discharged over the dam spillway or
constrained in turbines and taken from deeper waters for power
production. The uses may also constrain the type of mitigation measures
that can be implemented.
IV-2
-------
Table IV-2 gives an overview of the major water quality effects
found in the case study reservoirs, while Table IV-3 presents the
mitigative measures either planned or implemented. In the narrative that
follows, the reservoirs are grouped by the major type of water quality
effect. The major types of effects include: low hypolimnetic DO,
increased iron and manganese, eutrophication, sediment movement, flow
regulation, reaeration denial, thermal changes, as well as neutral or
positive effects. Several of the case studies exhibit multiple water
quality effects. The categorical listing is intended merely to highlight
one prominent effect.
LOW DISSOLVED OXYGEN AND/OR INCREASED IRON AND MANGANESE
J. Percy Priest Lake and Dam
Operating Agency: Corps of Engineers
Location. J. Percy Priest Dam and Lake are located at mile 6.8 on
the Stones River within the metropolitan area of Nashville, Tennessee.
Discharges from the dam flow into the Cumberland River at mile 205.9,
which is about 15 river miles upstream of the inner city ^rea.
Principal Features. J. Percy Prnest Lake is approximately 211,000
feet long and has a maximum depth of 103 feet. The summer recreation
pool level has a volume of 391,900 acre«feet, a surface area of 14,200
acres, and 213 miles of shoreline. The drainage area feeding the lake
covers 571,000 acres. The dam was completed in 1967.
Uses. The major uses of the lake and dam are flood control,
hydropower production, and recreation. The powerplant contains one
turbine with a capacity of 28,000 kW. The lake and the reach of the
Stones River downstream of the dam are classified by the state for all
stream uses, including: domestic water supply, industrial water supply,
fish and aquatic life, recreation, irrigation and livestock watering, and
wildlife habitat.
Water Quality Conditions. J. Percy Priest Lake is classified as
eutrophic as a result of nutrient loads directly attributable to
upstream land use activities. High nutrient concentrations occur
particularly during the growing season and result in heavy algal blooms,
and other plant growth, which occasionally impair water uses such as
water supply and recreation. The lake also exhibits other water quality
characteristics typical of highly productive lake systems.
Thermal stratification and the decay of organic material results in
an anoxic hypolimnion. High concentrations of dissolved iron and
manganese and the production of hydrogen sulfide result, causing the
water to have the appearance of black ink and a strong rotten egg odor.
During power production, water withdrawn from the lower layers of
the reservoir and discharged through the turbine becomes aerated while
flowing through the tailwater. This reoxidation causes the dissolved
IV-3
-------
w
E
Q
*^^
w
C
T3
0
&
hH
^
•o
3
4J
CO
0)
5)
<4H
rH O
HI fl>
rl
0) 3
rH 4-1
,0 tO
ft* /ll
(Q w
HI*
rH
P.
U
C
•rl
IH
PH
<4H
O
G
0
i/i
•H
cO
Q
73
0)
IH I-H
CO rH
01 -H
» b
^4H
•^_-«
rH ^~
O 4-1
> 0)
0)
a) u-i
0) OO •
> CO 0)
•H r4 rl
4-1 O CJ
O 4-1 Cfl
^
o^
•r^ Q)
** N-«
'O
0)
*o
SI*
.^
•g
o
1
M Q
CM CO CT^
IT) LTl LT)
O1* O^ O^
i— t rH i— (
O O O
o o o
0 O rH
O r** 0*
CO ""O ^0
r-t CM
•v
rH
CM CM rH
CM -41 <*)
rH
O O O
-HO O
CO rH r^
•V *\ *\
-^ ro CM
CM O O
CM 0^ *O
rH r4 CM
1 i i
rH" (3 C
O O O
o o o
rH O O
ON*1 C-T r-T
CO O m
rH CS CS
K
ftTs 3 3
r4 0)
O C
be £
o a)
54J IH cO IH 4-1 M
i/i 0) 4.) a) o oi
3rH > e > >,a)
o o *H nj -H o }-t
W 33 05 W Otf U O
4H,
^ §
^4 t<3 ^J Q
OK «3 rl < W W
> H 1 3 CJ « «
COI-N a)£73" Ol'H-H
OMB ^ CJ CO S ^ Ul W
ON
ON
rH
O
O
0
rH
^
in
•
CM
O
m
rH
O
o
o
co"
O
O
m
rH
«
b
0)
rH
rH
U IH
a)
W £
0)
(g ^
W rJ Q
CM
O
C\
f— (
O
§
o
0
CO
rH
rH
m
r^»
CO
O
o
0^
OJ
-*
O
o
o
in
r^.
^
0
o
o
*a
rH
m
CO
B
C h
01 a>
a) >
IH -H
U oi
^
•H •*
•H O) M Q
3 IH W 73
rH 0 CD g
CO
in
ON
rH
8
o
o
OJ
04
rH
^
CO
0)
r^
CO
Q
CO
o
o
ON
^*
m
o
o
ON
vo"
04
X
IH
0
&H C
o
54-1 V4
ui a)
3 rH >
0 O -H
CO P3 K
2
H
•H •>
• OS
PH > CO
>> IH Q
4-1 IH 0)
>H C W 73
0 01 01 S
b S K 3
ON
CO
ON
rH
8
o
o
o
in
rH
ON
ON
O
O
ON
("—
vO
O
o
8^
rH
O
j>
Q
o
o
o
rH
s{
b m
«
01
Ul
Ul
01 ^
Is
ai
rH
T-H
•H M
> -H
woes
IH > iO H
4J OI "3
§W 73
U-SS^
p^.
NO
ON
rH
O
o
o
•y
rH
p^
in
CO
rH
OJ
o
0
01
•y
_H^
rH
CO
O
rH
O
O
O
rH
rH
OJ
O
O
ON
rH
ON
CO
aT
&t< OH
Ul
0) rl
e o)
0 >
4J -H
a) g
(Q
0 g
^ 4J a.
O) 01 C
OH OI
•H 73
• VJ C
•n OH cc
rH
P< C
a, o
3 -H
«
IH 0)
0) IH
4-1 O
CO 0)
2 OS
1 1
3 Oi
o
4-1 C M
c o a>
O -H »
O 4-1 O
CO P.
73 60 O
O -H IH
O > 73
rH CO >1
b 2 IJH
1 t 1
b 2 33
..
S_(
g
o
C
01
60
nj
C
(0
JH
O>
P,
O
r^
£1
73
0)
73
•H
O
(H
P.
4J
O
11
p<
C
IV-4
-------
1
CO
—
U)
4-1
B
i
•o
3
%
M
•o
3
4-1
/-xc/5
4-1
- 0)
8.3
<4H
•H O
> VI
H 0)
tj) 3
rH 4J
CQ 01
H Cc,
g,
•H
U
-S
rl
P-
<4H
O
B
O
VI
•H
rH
CO
I*
3
•o
0)
rlrH
(ij ^H
O> -H
0)
00 '-H
SI/1
01
•H fl IH
CO 01 c.
Q <^-
01
B
•H f~
rH £ M
01 4J 01
tH OflrH
O B -H
£ CD 6
01 ^
o w
CO 0)
14H CO rl
rH 01 O
3 rH CO
VI < •—
1 £
•rl 5 4
X P< 0)
H 01 n.
5-
bQ Q)
C Q)
0) HH
,
rH /—
O 4-1
> 0)
01
01 >4H
01 00 •
> CO 0)
•H rl IH
4J O U
O 4J CO
< :/}•—
^_.
o-!l
•r-, 4)
ss
(U
T3
^ 3
01 O
^ p«
OS M
4-1
B
01
Irapoundm
Dam
cO
•j
P.
rl B
0
o
o
*.
o
r-.
rH
in
CO
CM
o
o
•v
oo
rH
Pi
B
O
l\
CM
CO
rH
O
O
0
*\
r-
33 (^
fe 3
rl
>
•H
OS
4-1
rH
CO
^
B §
•H CO
CO *O Q) Cl
0) O
CO to rH CO O
S J U O •£.
vO
Q«l
1— 1
O
8
•v
vO
•
P.
B
O
CM
m
P.
B
P.
B
0
O
CM
*\
in
ro
X
•o
O JH
rH 0)
o >
O -H
s as
^
u
14
McCloud
Reservoi
and Dam,
CO
o^
1—1
8
o
*,
o
vO
00
rH
O
in
r«^
O
o
CM
rn
o
o
CM
I
•V
m
00
ro
O
O
O
1
CM
•v
•z
taX
O M
B 01
•H >
rH -H
O DC
z
H
IH
Norris
Reservoi
and Dam,
-3-
in
o
rH
O
o
CO
00
a.
B
o
0
in
CM
CM
0
o
O
0
•v
00
CM
in
O
O
0^
O
CM
Z K
X fe
•o
^
rH
01 rH
-Q 0)
i >
3 -H
U OS
>!>! 2
rl U H
0 O
J«! t-3 •>
01
o i4 <3
S
O
rH
O
o
0
o
H
^
P,
B
0
O
*s
vO
O.
B
8
cs
CM
rv
o
OS
(t.
UI
B
• H
5« ij
0)
UI >
01 -rt
Q OS
13 T3 M
>
CO-H
W Qi
.
m
•o
T3 rH E <
rH rH (3 t3
r9 0)
J2 w 01 ~
o 3
co ui
01 0) •
0) 0)
P. 0) -0
P. rl C
D O (3
rH
§*-H
t/1 4J
CO
VH 0)
0) IH
4J 0
cfl 01
3 Qi
I i
3 «
rH
0
rl
4-1 B IH
BOO)
O -H S
O 4-1 O
CO P<
•o oo o
§-H !H
> -0
fc Z S
t 1 1
fc.55 SB
,,
SH
S
l^i
1
60
CO
60
•rl
4J
CO
S
01
a.
o
5j*
"O
01
73
•H
O
rl
P.
4J
O
fi
II
P.
B
IV-5
-------
I
w
0)
u
5
•H -H
«
Ag
as
•33
as
81
Late Cas
Casitas
-3B
Flaming
Reservo
Data, UT
•
sss
££l
ti&J
O V
t". a: ij)
Gunter
Reserv
AL & T
8
I
u
£
IV-6
-------
£
t
flS
•H «j
h. X
*J 14
« .3
SI2
II
.8
Jl
Ss
&g
5 I
s:
2 °
o -3
EJS
t
1
-I
I48
•65
£3
IV-7
-------
-r
41
3 r.
s
H
3
z
1
SMO11V3UK
t.
t~
5
i
f-
rt
i
8
CA
o
fe!
ll
4)
!o 0
1C
ft, ^
» C
M C
(/> -M
s
c
ft
1
OL
k.
C
1
c
•~ c
*J C
e •*•
4J C
M U
1
fc
C
U
A
1
v
1$
reduction being
and sludge proble
s«
a "u
£S
O
K
ft- *
ll
J
"
inn ion
c
c
f
1C
ftr
IB
IM M
•Si
^
3
Lake Cachuu
6 Bradbury Dm,
c
0
u O
!*• 4J
S 0
B U
|5
a
w
ss
tion in hypo 1 inn i
ion. Copper sulf
? •>
sl
->
-»
Lake Casitas 6
Casita* Daa, CA
0
o
o
tJ
c
g
f
E
I
«
o
ft)
f)
s
V
4J M
II
M
-
i
— a
"c
r
tc
f
a
->
Flaaing Gorge
Reserroir A
tm, UT
1
V
V)
S
k.
-
O
C «•* V
0 O ft)
*J O -*
fti *J C
K C 0
3 «•«
H i
s*s
treatments and g
ophication. Sedi
SediBent/renoval
•o u •
•38?
f s!
-
*•
Guntersvillp
Reierroir & Dna
AL A IN
e
c*
— ' U
o
te
k.
•« •»*
x •
O *H k<
ion of releases t
, and localized •
1 of surface wate
HI
si-
o
o
ft.
z «c
IV-8
-------
*
l/l
3
Z
0
H
CJ
[i-
g
w
1
C/l
z
0
H
5
g
a
H
3
3
z:
o
§
a
s
(/I
g
U
__,
O
t/1 tH
CU *J
2: c
o
CJ
60
c
H VI
4-t 4)
rt -»
1S
4)
5
O
ui
(d E
•H >
-Q 0
i-t
H ^
U VH
*-* 5
W -H
C
O
H
U
0)
LJ
01
5
00
c
rt
on
1
Aeration
strati- 1
cation
0> H
O u-i
UJ
H
t/i
•A
C
O
H
J_J
n
u
0)
ex
o
0) v)
on a)
fl n
V» 0)
c
9 o
0) a
4-) 3
Vi O
p.-—1
"O
'4-t O
O H
on *fl
c
H O
'a
TJ TJ
-C G
v) (0
0) rH
OH
0>
>> "
o <9
u
as 'a
O t-J
c
o
u
BJ
£
0)
c
o
u
c
0)
V)
0)
o
c
4_»
c
01
01
a.
(fl -H
0)
&o >
G 3)
pt
3 O
4-1 O
tn p.
a.
jg d
TJ J4
to u
C£ O
Qi
01
J£ T3
01
fl
u
c
•H i-l
£ 0
V) 01
•a
0) C
W 41
fl M
a) >,
VH X
0 0
c
H O
[/) *J
o c
m H
0) O
c p. .
MX 1)
0 M «]
cow
0 > C
H El H
4J (U
—i at a)
tt-< *j
V)
O t*-! >.
o o «
-
0)
V)
I/I <
s
PQ Q
Q
-o
o X
H m
a: J
c
C 0)
O — '
•H -O
13
O W
P. VI
Is
C 0)
S '
3 e c
w G H
>i nj 4-J
VI •»"(
cSl
0 0
H 41 u
i-l I/]
(Q m c
t-i a> o
0) l-t H
(fl U *-»
C B)
T3 H 0
0) 'H
— -< O ^M
3U
r- nl
W 00 M
COW
t-H r-t V)
-
->
^
B
U Q
nj
01 M
CO H
o
01 0)
P. VI
CX 01
tD lii
IV-9
-------
iron and manganese to precipitate, staining downstream areas. Iron
concentrations are reduced to low levels a short distance downstream.
Manganese, however, is very persistent with effects evident some distance
downstream. The manganese concentrations affect domestic and industrial
water supplies, particularly under the anaerobic summer conditions.
Mitigation/Enhancement Measures. A mitigation measure that has been
implemented is the coordination of releases with another dam. When
discharges are required (for power production) from J. Percy Priest Lake
during the stratification period, releases are coordinated with releases
from Old Hickory Lock and Dam on the mainstem of the Cumberland River so
that manganese concentrations will have a lesser effect on water
treatment plants downstream.
Another measure selected for testing in 1987 is the installation of
pumps in the forebay, upstream of the penstock, to cause localized
mixing. This measure will force entrainment of surface water into the
withdrawal zone of the turbine intake. Thus, a high percentage of high
quality surface water will enter the penstock and, in effect, prevent or
reduce the percentage withdrawal of the low quality waters within the
hypolimnion.
Old Hickory Lake. Lock, and Dam
Operating Agency: Corps of Engineers
Location. Old Hickory Lake, Lock, and Dam are located at mile 216.2
on the Cumberland River in Davidson and Sumner Counties, Tennessee, about
25 river miles upstream of Nashville, Tennessee.
Principal Features. The reservoir was completed in 1954. The lake
extends almost 100 miles upstream to Cordell Hill, at river mile 313.5,
and has a maximum depth of 70 feet. The surface area of the lake is
22,500 acres and the volume is 420,000 acre*feet. The drainage area is
898,000 acres below upstream dams. There are several large embayments
covering 5,000 acres in the downstream portion of the reservoir that are
fairly isolated from the main channel flow. Flows to the lake are
regulated by three upstream tributary storage reservoirs. The reservoir
is generally confined to the old river channel and is very narrow and
serpentine, with an average width of 1,500 feet, excluding embayments.
Uses. The main uses of the dam and lake are hydropower production
and provision of navigation upstream to the Cordell Hull Lock and Dam.
They also serve for flood control, recreation, and water quality. The
reach of the Cumberland River downstream of Old Hickory is classified for
public water supply and fish and aquatic life, and has a DO standard of 5
mg/1.
Water Quality Conditions. Analysis of data collected by the Corps
of Engineers shows that Old Hickory may be thermally stratified from
April or May to September. Dissolved oxygen concentrations during this
period become reduced in the downstream forebay of the lake as well as in
IV-10
-------
the releases to the tailwater. These conditions are most severe during
periods of drought or low stream flow.
Water quality routing studies have shown that inflows from one of
the upstream reservoirs, Cordell Hull Dam, are cooler and, therefore, are
routed beneath the epilimnion during stratification. They are
subsequently released in the tailwaters downstream. During times of
drought or low flow, such as June through August, these inflows are
reduced and the retention time in the hypolimnion is increased, lowering
DO concentrations in the releases of Old Hickory. A minimum observed DO
concentration of 2.1 mg/1 occurred in August-September 1975. Flows
during this period were unusually low due primarily to changes in
discharges from controlled releases upstream. Seasonal drought
conditions in the lake also contributed to these low DO values.
Mitigation/Enhancement Measures. Since Old Hickory Lock and Dam are
operated for navigation and hydropower generation, the pool elevation is
maintained within a tolerance of a few feet, in accordance with the
project water control plan. This regular operation is maintained in
concert with the discharges of dams upstream whose operations are also
specified by project water control plans. The DO release problems
associated with this system can be mitigated only if flows resulting from
the projects can be accurately forecast in time to reschedule operations.
The need for a predictive capability of flows and their effects led
to the development of a DO routing model between the uppermost storage
project, Wolf Creek, and Old Hickory Dam. Modeling results are used to
evaluate potential DO problems in the releases from Old Hickory Dam and
to test various operational changes which might be required to avoid low
DO concentrations. There is some flexibility in the operations of the
upstream storage impoundments. Simulations of these operations suggest
that a spring target allocation for storage in the impoundments upstream
will provide water in sufficient quantity for flows through Old Hickory
Lake during the summer stratification period. Other more detailed models
are under development which will be used to evaluate the effects of
particular point and nonpoint source discharges or in other situations
where more realistic and detailed simulations of reservoir dynamics are
required.
Richard B. Russell Lake and Dam
Operating Agency: Corps of Engineers
Location. Richard B. Russell Dam is located on the Savannah River
at river mile 275 Elberton, Georgia, on the Georgia-South Carolina
border. Two other Corps of Engineers reservoirs, Clarks Hill Lake and
Hartwell Lake, are located immediately below and above Richard B. Russell
Lake, respectively.
Principal Features. Richard B. Russell Lake has a surface area of
26,700 acres, a volume of 1,030,000 acre-feet, and a shoreline length of
550 miles. Mean and maximum depths are 38 and 150 feet, respectively,
IV-11
-------
with a theoretical hydraulic residence time of 102 days. The dam was
completed in 1984, but work on the dam and powerhouse, including the
installation of four turbines, will continue until late 1989. The major
inflow to the lake is the Savannah River, which is regulated immediately
upstream by the operation of Hartwell Dam. Two large embayments were
formed near mid-lake by the flooding of Rocky River to the east and
Beaverdam Creek to the west.
Uses. Richard B. Russell Dam and Lake were authorized by the Flood
Control Act of 1966 to provide power generation, incidental flood
control, recreation, streamflow regulation, and water supply on the
Savannah River. The installation of four reversible turbines for pumped
storage operation will continue until late 1989. Current operation
allows for power generation using four conventional turbines rated at 75
MW each. When completed, the powerhouse will have a total generating
capacity of 600 MW. The dam is currently operated to meet peak power
demand. Water from the Rocky River embayment also is the primary source
of drinking water for the city of Abbeville, South Carolina. Dissolved
oxygen standards require a minimum concentration of 6.0 mg/1 in the
tailwaters.
Water Quality Conditions. Water quality conditions in the lake have
changed markedly since its impoundment. During the initial stages of the
filling process, water quality conditions were strongly influenced by the
inundation and subsequent decomposition of terrestrial vegetation,
detritus, and organic materials contained in flooded soils. Nearly 9,000
acres of forested area were inundated during filling, resulting in the
inundation of an estimated 550,000 metric tons green weight of standing
vegetation and 5,100 metric tons of litter and detritus. While standing
vegetation was determined to have minimal direct impacts on water
quality, the decomposition of litter and detritus has had a pronounced
impact, particularly with respect to DO concentrations in the bottom
waters. Dissolved oxygen concentrations in bottom waters were severely
depressed during the summer of 1984, and elevated concentrations of iron
and manganese prevented the testing of turbines due to downstream water
quality impacts. Also during 1984, powerhouse construction necessitated
the release of water through tainter gates. This operational scheme
reduced flushing of the hypolimnion, contributing to the accumulation of
poor quality hypolimnetic waters. Following completion of the four
conventional generators, releases were made from the lower portion of the
water column. This greatly reduced the residence time of hypolimnetic
waters and increased flushing. Inflows from Hartwell Lake were observed
to enter Richard B. Russell Lake as an interflowing density current,
which further increased flushing of deeper strata.
Water quality conditions were also impaired in the two major
tributaries. Anoxic conditions developed in near-bottom waters by late
March and iron and manganese concentrations began to increase. By early
June much of the water column in each embayment was anoxic and elevated
concentrations of soluble nutrients and iron and manganese were recorded.
These conditions persisted until turnover in mid-November. On one
occasion during summer stratification, the withdrawal of anoxic
IV-12
-------
hypolimnetic water form the Rocky River embayment resulted in the short-
term closure of the Abbeville water treatment facility until the
relocation of the intake to a higher elevation was completed.
During 1985, the decomposition of litter and detritus accounted for
approximately 60 percent of the total hypolimnetic oxygen demand.
However, field and laboratory studies indicate that as organic materials
are decomposed, oxygen demand should decrease by approximately 70 percent
by 1988. These declines may have accounted for the less severe water
quality conditions observed during 1986 and 1987.
During the summer stratification of 1985 through 1987, water quality
conditions in the main portion of the reservoir greatly improved. These
changes are related to the following: decreases in the quantity of
labile terrestrial organic matter, changes in flow patterns in the lake,
and operation of an oxygen injection system in the forebay. Improvements
in the two major impoundments were far less pronounced, suggesting the
importance of changes in flow patterns in influencing water quality
conditions in the main portion of the reservoir.
Mitigation/Enhancement Measures. An oxygen injection system was
designed and installed in Richard B. Russell Lake for the purpose of
maintaining a minimum DO concentration of 6 mg/1 in the tailwater. This
is the DO standard for the tailwater and is based on the habitat
requirements of the fishery in the upper reaches of Clarks Hill Lake.
The injection system is composed of two independent components. A pulse
component, consisting of eight diffuser lines, is oriented perpendicular
to the intake section of the dam and is located approximately 10 to 16
feet above bottom grade. This component of the injection system provides
additional oxygen injection capability during generation and is capable
of delivering oxygen at a maximum rate of 80 tons/day. The second
component was designed for continuous injection of oxygen at rates up to
100 tons/day, and is located approximately 1 mile upstream from the dam.
During the summer stratified period, liquid oxygen is transported to the
site, evaporated, and delivered to the diffuser system through a system
of pipes.
Depending on conditions in the lake and tailwater, the system is
generally operated from early April until mid to late November.
Injection rates are varied periodically to insure adequate and economical
operation by routine field monitoring and through the use of a numerical
model. To date, the system has been operated successfully and problem
conditions have not been observed in the tailwater. The system has also
led to significant increases in the average summer oxygen concentration
of bottom waters in areas immediately upstream from the dam.
Upper Bear Creek Reservoir and Dam
Operating Agency: Tennessee Valley Authority
Location. Upper Bear Creek Reservoir and Dam are located on Upper
Bear Creek in northwestern Alabama in Marion County.
IV-13
-------
Principal Features. The reservoir is 37,000 feet long, with an
average volume of 37,400 acre-feet, and a maximum depth of 70 feet. The
total drainage area above the dam site is 72,300 acres.
Uses. The reservoir provides for fish and wildlife, recreation,
shoreline development, and water quality control, and serves as a water
supply for several communities. The reservoir was also designed to
provide water for the weekend operation of the Bear Creek Floatway
located downstream from the dam.
Water Quality Conditions. Low summer DO concentrations occur in the
Upper Bear Creek Reservoir. Anoxic conditions in the lower depths of the
reservoir provide a reducing environment, which results in the
resolublization of iron, manganese, and sulfur present in the sediments
and causing high concentrations in the water. One source of these
constituents is believed to be upstream mining activities, but they can
also occur naturally.
The water treatment plant that uses the reservoir as a water supply
often struggles with the removal of the iron and manganese. When water
is released from the lower depths of the reservoir, the iron and
manganese oxidize and precipitate in the creek, leaving it highly
stained, with large growths of iron bacteria and precipitates coating
much of the aquatic life below the stream. Also, concentrations of
hydrogen sulfide as high as 0.5 mg/1 have been detected in the creek.
Hydrogen sulfide in concentrations greater than 0.002 mg/1 can be toxic
to aquatic life. In 1986, iron and manganese concentrations of 6.9 mg/1
and 3.6 mg/1, respectively, were measured immediately below the dam,
while concentrations of approximately half these values were measured
three miles below the dam.
Another effect of the dam occurs when the location of the release is
changed. Water released from the surface overflow might be 10 to 15°C
warmer than the water released from the low-level release. Changing the
withdrawal point due to the decreasing reservoir elevation probably
causes thermal shock to downstream aquatic organisms.
A biological study of Upper Bear Creek Reservoir was conducted by
the State of Alabama during 1979 to 1983. The study concluded that
benthic macroinvertebrate levels were adversely affected by the reservoir
water quality. In addition, fish population data indicated a poor
fishery exists in the reservoir.
Mitigation/Enhancement Measures. In 1987, a diffused aeration
system was installed in the reservoir, consisting of four diffusers
stretched across the old river channel. Compressors supply the air and
are run continuously from about March to September. The aeration system
significantly warmed the hypolimnion, rendering the reservoir
stratification less stable. Dissolved oxygen levels have also increased.
However, the reservoir has remained thermally and chemically stratified.
This might be due in part to several compressor malfunctions which
IV-14
-------
resulted in the operation of only two diffusers for several weeks.
During the summer of 1987, all of the iron was converted to the oxidized
form, but most of the manganese was still in the dissolved form. Both
iron and manganese levels increased as the reservoir became more anoxic.
However, the 1987 concentrations of iron and manganese were lower than
they were before the aeration system was installed.
Casitas Lake and Dam
Operating Agency: Bureau of Reclamation
Location. Lake Casitas is located in southern California near Santa
Barbara. The two main tributary inflows are Coyote Creek and the Robles-
Casitas Diversion Canal.
Principal Features. The reservoir has an active capacity of 251,000
acre-feet, a maximum depth of 260 feet, a surface area of 2,700 acres,
and 31 miles of shoreline. The reservoir was filled in 1959. The Coyote
Creek watershed has a direct drainage basin of 21,100 acres and is former
ranch land that is now managed for recreation and water quality control.
The Robles-Casitas Diversion Canal brings in water from the Ventura
River, which drains 48,000 acres. This indirect drainage basin includes
numerous smal'i holdings, large ranches, public domain land, and a
population of about 10,000. Due to the climate of the area, the
reservoir is typically filled during the winter rains, and then is drawn
down steadily during the summer when there is little or no significant
precipitation. The dam was originally designed with a multi-level intake
to the outlet structure that allows selected strata to be withdrawn.
Uses. The Casitas Municipal Water District manages the lake
primarily for water supply. Secondary water uses include irrigation and
nonbody contact recreation.
Water Quality Conditions. From 1959 to 1967, water below about 60
feet from the surface and, at times, as little as 30 feet below the
surface became anaerobic during the summer thermal stratification.
Unacceptable concentrations of manganese and hydrogen sulfide accumulated
in this anaerobic zone. At the same time, blooms of taste and odor-
producing algae developed in the surface water. Under these extreme
situations, attempts to use the multi-level intake to withdraw water from
the narrow zone between the algal blooms and the anaerobic hypolimnion
often failed.
Mitigation/Enhancement Measures. In 1968, a diffused air injection
reaeration system was installed in the reservoir that injects compressed
air into the hypolimnion through a series of diffusers that are
positioned 80 to 100 feet from the bottom. The system has been refined
over the years.
The reaeration eliminated the water quality problems caused by
manganese and hydrogen sulfide accumulations in the hypolimnion. It is
now possible to withdraw water from 100 feet or more below the surface
IV-15
-------
during summer stratification, thus avoiding any taste and odor problems
from surface algae. This, in turn, allowed a reduction in the number of
copper sulfate applications needed to control algae. Aeration of the
cooler depths of the lake during the summer months made possible the
establishment of a "2-story fishery" with warm-water species, like bass,
in the upper waters and rainbow trout in the deeper waters.
Lake Cachuma and Bradbury Dam
Operating Agency: Bureau of Reclamation
Location. Bradbury Dam is located on the Santa Ynez River near
Ventura in southern California.
Principal Features. The reservoir has an active capacity of 202,000
acre-feet, a maximum depth of 190 feet, a surface area of 3,100 acres,
and a 42-mile shoreline. The drainage area is 267,000 acres, with most
of the drainage basin consisting of the Los Padres National Forest.
Runoff is extremely variable from year to year, with almost all of the
annual runoff concentrated in the winter months.
Uses. The major use of the reservoir is water supply. Most of the
water districts withdraw their supplies from the upper end of Lake
Cachuma through a multi-level selective withdrawal tower. However, the
Santa Ynez River Water Conservation District (SYRWCD) withdraws its
supply of water from the bottom outlet of the dam.
Water Quality Conditions. During the summer, thermal stratification
results in anoxic conditions occurring in the hypolimnion. As a result,
manganese and hydrogen sulfide accumulate. These conditions adversely
affect the SYRWCD domestic water supplies for users who withdraw water
from the bottom outlet of the dam.
Mitigation/Enhancement Measures. In 1981, a diffused air injection
reaeration system was installed near the outlet works of Bradbury Dam.
This system was designed to treat only the lower basin of the reservoir.
It consists of a compressor that pumps air to four diffusers suspended
about 30 to 40 feet off the bottom.
Before the aeration system was installed, water quality usually
deteriorated to extremely poor levels by early August. With the aeration
system in place and beginning in April, the lower basin of Lake Cachuma
was sufficiently aerated to extend the period of acceptable water quality
about a month to a month and a half. This extension was significant to
the SYRWCD water suppliers.
In 1985, the SYRWCD added a flexible extension to the bottom outlet
withdraw that allowed water to be withdrawn from the well-oxygenated
waters of the epilimnion. The new outlet was operated with the aeration
system in 1985, but the district operated it without the aeration system
during the summers of 1986 and 1987. So far, the SYRWCD has been very
satisfied with using the flexible outlet alone.
IV-16
-------
EUTROPHICATION
Guntersvi11e Reservoir and Dam
Operating Agency: Tennessee Valley Authority
Location. Guntersvilie Dam is located on the Tennessee River at
river mile 349.0 near the city of Guntersville, Alabama. Guntersville
Reservoir is located in Jackson and Marshall Counties, Alabama, and
Marion County, Tennessee.
Principal Features. The reservoir is 401,000 feet long. At the
normal maximum pool level, the reservoir has an average volume of 1.02
million acre-feet, a maximum depth of 60 feet, and a surface area of
67,900 acres. The dam was completed in 1939.
Uses. Guntersville Reservoir is the second largest of the multi-
purpose reservoirs operated by the TVA for navigation, flood control, and
power production. Recreation, water supply, and assimilative capacity
are significant secondary uses. Four turbine generators with a total
rated capacity of 102 MW are used for power production. Drawdowns are
used for flood control, aquatic weed control, and vector (disease
carrier) control. Eight major public and industrial users withdraw water
directly from the reservoir, while there are 30 major municipal and
domestic waste dischargers and 16 industrial dischargers to the
reservoir. The Alabama Department of Environmental Management and the
Tennessee Department of Health and Environment have identified the
following as appropriate uses of the waters of the reservoir: domestic
and industrial water supply, recreation, fish and aquatic life,
navigation, irrigation, and livestock watering. The Guntersville Lock
System is an integral part of the 650-mile water transportation channel
of the Tennessee River system.
Water Quality Conditions. Guntersville Reservoir is thermally
stratified in the deeper, downstream portion during the summer. The
stratification results in hypo!imnetic DO depression. The upstream third
of the reservoir is well-mixed, while the midsection is a transition
section. More than 80 percent of the nutrient budget to the reservoir is
contributed by the Tennessee River.
The morphometric, hydraulic, and nutrient characteristics of
Guntersville Reservoir provide ideal conditions for macrophyte and algal
growth. It has been classified as highly eutrophic by several commonly
used indices. Several biological and water quality parameters indicate
that Guntersville Reservoir's trophic state is increasing in a eutrophic
direction. Aquatic macrophyte growth, heterotrophic growth, total
organic carbon, and BOD are increasing. More than 25 percent of the
reservoir acreage is infested with macrophytes. Hydrilla continues to
spread, with a 30 percent increase in the past three years.
IV-17
-------
Many areas of the reservoir are of limited use due to macrophyte
growth. It also greatly diminishes aesthetic appeal for swimming and
boating. Algal-induced taste and odor complaints from public water
supply customers are common, and water treatment plants must continuously
adjust chemical concentrations to provide acceptable drinking water.
From 1974 to 1983, significant downward trends in fish biomass and
numbers occurred for one or more size classes of eight of the eleven
dominant fish species. However, in 1985 biomass estimates were over 2.5
times greater than in 1983 and were the greater than any other time in
the period of 1974 to 1983. The condition of several important game fish
species in the reservoir was better than the average for all mainstem
reservoirs.
Approximately 7 million recreational visits are made to Guntersville
Reservoir annually. Continued water quality degradation will result in
significantly fewer visits and hence revenue reductions to the local
economy. Aquatic macrophytes, excessive algal growth, and to some extent
declining fisheries are the most obvious problem as perceived by the
public. A recent survey indicated that respondents felt the reservoir
would be unusable by the year 2000 if water degradation continues at its
present rate. Conversely, a 13 percent rise in visits is expected if
water quality conditions remain stable.
Mitigation/Enhancement Measures. Several efforts to improve/
mitigate conditions in Guntersville Reservoir are currently being
undertaken. Herbicide treatment has been used for aquatic macrophyte
control, and recently grass carp were introduced. The cost-effectiveness
and weed control capabilities of the herbicide and grass carp are being
evaluated. The possibility of clearing boat lanes of macrophyte
vegetation to allow fishermen to reach embayment areas is also being
evaluated.
Sediment and nutrient controls are planned for selected
subwatersheds. When these controls are implemented, in-reservoir
techniques, such as sediment removal/covering and aeration, will be
implemented. When used in combination with reduced pollutant loads,
water quality improvements will be visible immediately. Water quality,
land, and fisheries management plans have been prepared for the
reservoir.
Boone Reservoir and Dam
Operating Agency: Tennessee Valley Authority
Location. Boone Reservoir is one of six TVA impoundments in the
Holston River Basin in northeast Tennessee. Boone Dam is located on the
South Fork Holston River at mile 18.6, approximately 1.4 miles below its
confluence with the Watauga River.
IV-18
-------
Principal Features. Boone Reservoir was completed in 1952, is
91,900 feet long, has a maximum depth of 122 feet, and has a mean depth
of 44 feet. It stores 189,100 acre«feet of water at the normal maximum
pool level, and has a surface area of 4,310 acres. It has an
uncontrolled drainage area of 428,000 acres, a total drainage area of
1,180,000 acres, and a 122-mile shoreline.
Uses. The primary purposes for the construction of Boone Dam and
Reservoir were flood control, regulation of flows for downstream
navigation, and to the extent consistent with the first two purposes,
hydroelectric power generation. Flood control rules require that
definite amounts of storage space be reserved from January to May. When
flood control and power generation constraints permit, the reservoir pool
elevation is controlled to accommodate other uses, including recreation
and fishing. During the fish spawning season, the reservoir water level
is held as stable as possible for a two-week period.
Water Quality Conditions. Boone Reservoir was studied and modeled
by the TVA in the development of a water quality management plan. The
Boone Management Plan identified use impairments due to bacterial
contamination, sludge deposition, and toxicity in portions of the
reservoir and its tributaries. In addition, metalimnetic oxygen
depletion, eutrophication, and litter were identified as serious
concerns. Boone Reservoir is the most eutrophic tributary reservoir in
the TVA system. Nutrient loadings of 232 g/m^/yr Nitrogen (as N) and
10.7 g/m2/yr Phosphorus (as P) produce chlorophyll at levels ranging from
12 to 16 mg/m3. Secchi depths (a measure of clarity) of 5 feet are
common during the summer. A pronounced metalimnetic oxygen reduction is
believed to result primarily from algal decomposition and/or respiration.
These problems are the result of both point and nonpoint sources of
pollution.
Mitigation/Enhancement Measures. Point sources are being dealt with
by state regulatory programs. Their efforts have resulted in
construction of two expanded municipal waste treatment facilities at
Johnson City, Tennessee. Facilities are under construction at Bluff
City, Tennessee, and Bristol, Tennessee/Virginia. In addition, Bristol
is separating their combined sewers to reduce the incidence of collection
system failure and hydraulic overload of the treatment plant.
Nonpoint sources of pollution to Boone Reservoir include runoff from
livestock operations, cropland, urban areas, landfill and dump sites,
mining areas, and construction activities. In addition, combined sewer
overflows, failing septic tanks, and roadbank and streambank erosion also
contribute. Nonpoint sources will be addressed by a variety of
nonregulatory approaches, including waste management demonstrations,
public education, and cooperative projects involving the State, EPA,
TDHE, USDA, and the public.
The TVA is carrying out several tasks in 1987-88 to address the
following: animal waste management, houseboat waste management,
productivity management, and biological evaluation of stream segments
IV-19
-------
impacted by toxics. During 1988, the installation of four animal waste
management systems is planned, as well as educating the public on the
need for animal waste management systems. Presently 19 of the worst case
operations have applied for the TVA cost share program. Four of these
have been selected for initial installation, and the effectiveness of the
systems for eliminating bacteria from the streams will be monitored. In
1987, pre-installation monitoring was initiated and will continue into
the beginning of 1988 to provide a preliminary data base on bacterial
contamination. Once the systems are completed, a monitoring program will
determine the effectiveness of the measures.
The TVA is evaluating methods of dealing with houseboat wastes and,
in cooperation with marina operators, will install one or more treatment
facilities. Public education activities will be used to encourage use of
these facilities.
Eau Galle Lake and Dam
Operating Agency: Corps of Engineers
Location. Eau Galle Dam is located on the Eau Galle River
immediately upstream from Spring Valley in west central Wisconsin,
approximately 50 miles east of St. Paul, Minnesota.
Principal Features. The lake is 3,300 feet long, has a volume of
1,500 acre«feet, a maximum depth of 30 feet, a surface area of 150 acres,
and a shoreline length of 25 miles. The watershed has an area of 41,000
acres, with land use primarily dedicated to dairy operations and
associated agriculture, pastureland, and woodlots.
The dam is a rolled-earth and rock-filled structure and was
completed in 1968. The outlet structure provides for both surface and
bottom releases. The reservoir was filled in 1969.
Uses. Eau Galle Dam and Lake were authorized by the Flood Control
Act of 1958 to provide flood control for the village of Spring Valley,
Wisconsin, and associated downstream areas. Additional uses include
recreation and fish and wildlife habitat.
Water Quality Conditions. Eau Galle Lake is thermally stratified
with seasonal anoxic conditions occurring in the hypolimnion. It is
considered to be eutrophic and exhibits excessive algal and macrophyte
growth. Seasonal high flow events dominate the external loading of
nutrients. The growth of algae and macrophytes, coupled with episodic
tributary loading, contribute to reduced transparency and the
establishment of nutrient-rich sediments.
The release of cool hypolimnetic water through the low-level gate
outflow increases the heat content of the lake and reduces the thermal
stability of the lake. The reduced thermal stability and wind-induced
mixing result in exchange between phosphorus-rich hypolimnetic waters and
the epilimnetic waters, suggesting that internal phosphorus loadings may
be significant in the lake.
IV-20
-------
Mitigation/Enhancement Measures. In 1986, hypolimnetic application
of aluminum sulfate was conducted to control internal phosphorus
recycling from anoxic sediments. Following application, phosphorus
concentrations in bottom waters, internal phosphorus loading rate, and
the abundance of algae were reduced relative to previous years. Prior to
treatment, summer internal phosphorus loadings (as P) averaged 15.6 mg/sq
m/day. Following application, internal phosphorus loadings were reduced
to 6.5 mg/sq m/day. Algal biomass remained relatively high after
treatment. The high levels are related to the proliferation of an algal
species, Ceratium sp., that has the ability to migrate vertically in the
water column. The algae may have obtained additional sources of
phosphorus from nutrient-rich river waters, which enter the lake in a
density current at the depth of the thermocline. The effectiveness and
longevity of the treatment, as well as mechanisms governing algal
abundance, will be more completely assessed during ongoing studies.
FLOW REGULATION/REAERATION DENIAL
Norris Reservoir and Dam
Operating Agency: Tennessee Valley Authority
Location. Norris Dam is located on the Clinch River at river mile
79.8 in northeast Tennessee.
Principal Features. The reservoir is 385,000 feet long. At normal
maximum pool level, it has an average volume of 2.04 million acre-feet,
a maximum depth of 200 feet, and a surface area of 34,200 acres.
Uses. Norris Reservoir is a multi-purpose reservoir, primarily
used for flood control, navigation, and hydropower production. Secondary
uses include recreation, fish and aquatic life, water supply, and
assimilative capacity. The rated capacity for power production is 101
MW, produced by two Francis turbines. Hydropower operations provide cold
water conditions suitable for "trout waters" and have been so classified
by the State of Tennessee. Cold water releases also benefit the steam
electric power generation plant located downstream on Melton Hill
Reservoir. The tail water extends 15 miles to the headwaters of Melton
Hill Reservoir, and is used for boat, bank, and wade fishing. Boat
fishing occurs when turbines are operating, while bank and wade fishing
are limited to primarily when the turbines are not operating.
Hydropower production causes a fluctuation in the tailwater level of
about 6 feet, with the river width changing from approximately 432 feet
to 310 feet. Until 1984, minimum streamflow downstream from the dam
consisted of leakage from the dam, and there were no hydropower releases
for an average of three weeks.
Water Quality Conditions. A study of the tail waters conducted from
1971 to 1977 identified low DO and inadequate minimum flow as the primary
factors limiting further development of the trout fishery. The benthic
IV-21
-------
fauna were dominated by tolerant organisms, while the fish condition
factor, a ratio of the weight of a fish to its length, decreased an
average of 11 percent each year during periods when DO levels decreased.
Until 1981, DO concentrations in the hydropower releases remained less
than 6 mg/1 for an average of approximately one-third of the year, and
remained less than 3 mg/1 for 55 days during periods of hydropower
releases. When DO concentrations were at minimum levels and both
turbines were operating, releases would flow about 13 miles before
natural aeration resulted in an increase to about 5 mg/1.
Mitigation/Enhancement Measures. In 1981, a hub baffle aeration
system was installed on the two turbines. It resulted in increases of
minimum DO concentrations of 2 to 3 mg/1 in released waters. The
aeration system was operated when DO in the hydropower scroll case
decreased in concentration to less than 4 mg/1, with a resulting minimum
DO of 3 mg/1.
Greater minimum flows have been maintained since 1984 by the
construction of a flow regulation weir approximately two miles downstream
from the dam. Water stored behind the weir, supplied by pulsed flow from
the hydropower units, has since provided a minimum downstream flow of
approximately 200 ft^/sec. Approximate one-half hour discharges from one
turbine is required to fill the pool behind the weir, which can then
supply flow downstream for 12 hours.
In an effort to speed recovery of tail waters, desirable benthic
invertebrates were transplanted to these waters. There has also been an
increase in the stocking of fingerling and catchable rainbow and brown
trout. There is an apparent trend toward improving trout condition
following aeration, although there are no statistically significant
differences between the pre-aeration and post-aeration years. Although
the tailwater benthic fauna continue to be dominated by tolerant species,
less tolerant organisms, which are also desirable food for trout, are
occurring more frequently. The 1987 benthic fauna samples showed a large
increase in mayfly abundance, which may signal the beginning of a more
rapid recovery. Caddisflies, crayfish, snails, and mayflies are also
beginning to influence the benthic community structure. The delayed, and
as yet incomplete, recovery of the tailwater benthic fauna community may
be because (1) the DO concentrations are still too low to allow survival
of some sensitive species, (2) there is a shortage of colonizers of
sensitive forms, or (3) because full recovery simply takes longer than
expected.
Together, all mitigative efforts have dramatically improved the
fishery of Norris tailwater. During 1980-83, when DO improvements were
made and public awareness of the fishery at Norris increased, a 17
percent increase in fishing effort occurred. Following the establishment
of minimum flow conditions, angling pressure increased 79 percent from
the 1970s and was also significantly greater than 1980-83. Total annual
trout harvest increased 77 percent in the 1980s compared to the 1970s,
while the average annual catch rates improved from 0.34 fish/hr in 1973-
74 to 0.42 fish/hr for 1980-85. These increases are attributable to a
IV-22
-------
number of factors, including public interest and awareness of efforts to
improve the fisheries; provision of more stable, aesthetically pleasing
and more "fishable" waters by the maintenance of a minimum flow; and
additional access created by the flow regulation weir.
Presently, the recovery of benthic fauna may not yet be complete,
and growth of individual trout has only been minimally affected, but the
fishery as a whole has dramatically improved. The fishery will continue
to be monitored to evaluate sufficient DO and flow conditions.
Mark Twain Lake and Clarence Cannon Dam
Operating Agency: Corps of Engineers
Location. Clarence Cannon Dam is located on the Salt River at river
mile 63 in northeastern Missouri.
Principal Features. The reservoir is 132,000 feet long, and has an
active storage volume of 457,000 acre-feet, a surface area of 18,600
acres, a shoreline of 285 miles, and a drainage basin of 1,470,000 acres.
Approximately 400 feet upstream from the main dam is a temperature
control weir that allows the withdrawal of epilimnetic water that is
warmer and higher in DO than the release of the hypolimnetic waters.
Uses. The uses of the reservoir consist of flood control,
hydropower generation, water supply, fish and wildlife conservation,
recreation, and incidental navigation for the Missouri River during low
flow periods. The power installation consists of two turbines, one
31,000 kW reversible and one 27,000 kW conventional.
Water Quality Conditions. When the reservoir is not stratified, or
when the thermocline-oxycline are below the temperature control weir, no
water quality problems are experienced in the releases during hydropower
generation. However, if the thermocline-oxycline is above the level of
the temperature control weir, as has been observed from July through mid-
September, poor quality water is pulled over the weir and released during
power generation. This occurs until sufficient force flexes the thermo-
cline downward, resulting in mostly epilimnetic water being pulled over
the weir.
Mitigation/Enhancement Measures. The problem of poor water quality
releases at power generation start-up is currently being studied. One
alternative being considered is a hydraulic or pneumatic destratification
system for the mini-lake area between periods of generation, and also
when minimum releases are being made and the tainter gates cannot be
used.
Fort Patrick Henry Reservoir and Dam
Operating Agency: Tennessee Valley Authority
Location. Fort Patrick Henry Dam is located on the South Fork
Holston River at river mile 8.2, near Kingsport in northeast Tennessee.
IV-23
-------
Principal Features. The reservoir is 54,900 feet in length and has
a maximum depth of 80 feet. At normal maximum pool the reservoir has an
area of 872 acres, a volume of 26,900 acre-feet, and a shoreline of 37
miles. The drainage area at the dam is 1,220,000 acres. The South Fork
Holston is a highly developed watershed with three major reservoirs
located upstream from the dam.
Uses. The reservoir is primarily used for hydropower production.
The rated capacity for power production is 36 MW, produced by two Kaplan
turbines. Significant secondary uses of the reservoir include
recreation, fish and aquatic life, water supply, and assimilative
capacity. A trout fishery was created four miles downstream from the
dam, at which point a thermal load is added to the river by a large
chemical plant that predates the dam.
Water Quality Conditions. Hydropower releases from Fort Patrick
Henry Dam have low DO concentrations. Dissolved oxygen was less than 6
mg/1 for about 102 days each year, less than 5 ing/1 for 62 days and less
than 4 mg/1 for 25 days. Periodically, the DO has reached 3 mg/1. The
releases naturally aerate from 3 mg/1 to about 4 mg/1 within 5 miles and
to about 5 mg/1 within 10 miles.
The reach of South Fork Holston River below Fort Patrick Henry Dam
and the upper reach of the Holston River receive waste from 13 industries
and the city of Kingsport. The quantity of waste discharged exceeds the
natural capacity of the river, and several dischargers provide treatment
beyond best practicable. The reduced DO conditions of the dam contribute
to the low assimilative capacity of these reaches. Lack of sufficient
streamflow was one limiting factor for future growth and development in
the Kingsport area.
The principal discharger to the South Fork Holston River, the
Tennessee Eastman Company, contracted with the TVA to provide a minimum
daily release of 750 ftVsec for supply purposes. This flow is therefore
available for assimilative capacity purposes.
Although DO concentrations are seasonally low in the Fort Patrick
Henry Reservoir, the primary focus of low DO concern has been in the
Moisten River below Kingsport, downstream from the dam. At this point,
the DO has been observed to be 3 mg/1 during concurrent low flows on the
North and South Forks of the Holston.
Mi ti gati on/Enhancement Measures. From December 1983 to May 1985,
the TVA participated in a joint study with EPA, the State of Tennessee,
and major industrial dischargers from Kingsport to assess current water
quality conditions and evaluate the cost-effectiveness of various DO
improvement strategies. Modeling results of the river indicated that
flow in the North and South Forks had the greatest effect on DO below
Kingsport, and the natural aeration, photosynthesis, and respiration had
the next greatest impacts. Several DO improvement strategies were
simulated and their cost-effectiveness determined. It was found that
IV-24
-------
using supplemental evening flow pulses could produce DO concentrations
beyond that achievable with more traditional treatments, even treatment
facilities upgraded to zero waste discharges. The evening flow pulses
use hydro power generation of varying durations, each strategically timed
to arrive at the DO sag to relieve low DO caused by plant respiration and
wasteloads during the early morning hours. Aeration of the releases,
instream aeration, and combinations of the three strategies were also
found to be effective strategies.
A two-year pulsing demonstration was to begin in June 1986.
However, at that time Fort Patrick Henry Dam would be required to provide
additional flow, on a regular basis, to meet new permit requirements for
sustained flow past John Sevier Fossil-Fired Power Plant. It was
therefore required to increase minimum pulsing from six one-hour pulses
per day to as many as 12 pulses per day, depending on the North Fork flow
and the number of units operating at the power plant. This indicated
that the DO improvement due to the flow requirement would exceed the
improvement expected (1 mg/1) in the reservoir releases demonstration.
With implementation of the power plant permit, significant
improvements have been achieved ^n DO concentrations downstream from
Kingsport. Additional field studies and modeling have confirmed this
improvement. The frequency and duration of outages of one or more units
of the power plant will be evaluated to determine if any incremental
pulsing plan could be performed, through the Reservoir Release Program,
to provide any additional treatment.
SEDIMENT MOVEMENT
Lake Red Rock and Dam
Operating Agency: Corps of Engineers
Location. Lake Red Rock is located on the Des Moines River in
south-central Iowa approximately 60 miles downstream of Des Moines, Iowa.
Principal Features. Lake Red Rock is approximately 42,200 feet
long, with a storage capacity of 90,000 acre«feet, a surface area of
6,300 acres, and a drainage area of 4,160,000 acres. It has a 7-day mean
residence time. The lake was impounded in 1969. Agriculture is the
predominate land use in the watershed, however, significant point-source
loadings to the Des Moines River occur upstream near the city of Des
Moines.
Uses. Lake Red Rock is a Corps of Engineers flood control
reservoir. It also provides recreation.
Water Quality Conditions. The Des Moines River, in general, is
highly turbid, nutrient rich, and high in dissolved and suspended solids.
The high suspended solids load contributed to the lake, coupled with
reductions in velocity in the broad headwater area, have resulted in the
deposition of large quantities of sediment and the formation of an
IV-25
-------
extensive submerged delta. Sediment accumulation varies longitudinally
and laterally within the lake, with the greatest amounts occurring in the
headwater area and the submerged river delta. Data collected from 1968
to 1976 revealed that the submerged delta extended approximately 3 miles
into the lake from the inflow point and covered about 2,200 acres.
Depths of deposition in the old river channel ranged from 4 feet at the
dam to 20 feet at the lake headwater. The distribution of sediments also
has impaired the use of the lake for recreational boating.
Mitigation/Enhancement Measures. In 1979, the permanent pool level
was raised to elevation 728 to compensate for conservation storage lost
to sediment accumulation. As of 1985, more than 39,000 acre*feet had
accumulated below elevation 725, occupying about 45 percent of the
storage originally reserved for the 100-year project life. An additional
33,000 acre«feet had been deposited between elevation 725 and elevation
780, the top of the flood control pool, resulting in a loss of about 4
percent of the total flood control storage. The current estimate for the
sedimentation rate is approximately 3,500 acre-feet per year, about four
times the original estimate.
The Rock Island District of the Corps of Engineers is currently
studying a permanent increase in the conservation pool level to elevation
742 to provide 400,000 acre«feet of combined sediment/conservation
storage. This proposal would decrease flood control storage by 20
percent. However, it would not prohibit controlling the project design
flood. Implementation of this proposal is contingent upon completion of
environmental analyses, public coordination, and approval of a revised
water control plan.
THERMAL CHANGES
Flaming Gorge Reservoir and Dam
Operating Agency: Bureau of Reclamation
Location. Flaming Gorge Dam is located on the Green River in
northeastern Utah, about 32 river miles downstream of the Utah-Wyoming
border.
Principal Features. The reservoir is 475,000 feet long, with a
maximum depth of 440 feet, an active capacity of 3,516,000 acre-feet, a
surface area of 42,000 acres, and a 375-mile long shoreline. It drains
an area of approximately 11,300,000 acres. The dam was completed in
1962.
Uses. Flaming Gorge Reservoir is a major unit of the Bureau of
Reclamation's Colorado River Storage Project. It is operated primarily
for power production, although it also serves the purpose of storing
irrigation water. The power plant is located at the downstream toe of
the dam and it houses three 36-MW generators driven by three Francis-type
turbines. Reservoir releases are made almost exclusively through the
three penstocks, whose intakes are located halfway up the dam. Hourly
IV-26
-------
releases are determined by the power system loads, and during the summer
months by maximum and minimum operating criteria for downstream
recreational uses.
Water Quality Conditions. Flaming Gorge is a reservoir, with direct
thermal stratification in the summer, an ice cover over most of the
surface in the winter, and periods of complete mixing in the spring and
fall (dimictic). However, as the reservoir filled in the late 1960s and
early 1970s, a chemocline (water quality gradient) developed in the area
immediately behind the dam. This chemocline was generally located below
the turbine intakes at a depth of about 200 feet, while the thermocline
was usually at a depth of 30 to 35 feet. The zone below the chemocline
did not mix during spring and fall overturns, and the water was cold,
anoxic, and poor in quality. After the selective withdrawal
modifications were installed in 1978, the chemocline gradually weakened
each year and finally disappeared during the spring overturn of 1982. It
has not been observed since.
Water quality on the lower, canyon section of the reservoir is quite
good and could be classified as being relatively oligotrophic (not
enriched). A lake trout fishery had been established. In the more
upstream section of the reservoir toward the tributary inflow areas, the
trophic status becomes mesotrophic and finally eutrophic in the summer
months in both the Black's Fork and Green River arms. Summer algal
blooms are common in these warm water reaches.
After closure of Flaming Gorge Dam, the downstream aquatic
environment of the Green River changed radically. Temperatures before
dam closure ranged from 32°F to about 67°F. High runoff flows with heavy
sediment loads occurred in late May and June. After dam closure,
temperature fluctuations were reduced to a range of 39°F in March to
about 50°F in November. Flows were stabilized to a power demand cycle.
Sediment loads were reduced, as reflected in a drop in turbidity values,
from 5,000 Jackson Turbidity Units to around 60 Jackson Turbidity Units.
The biota also changed. Indigenous fish species were replaced by
rainbow trout in the 26 miles below the dam. This sport fishery quickly
became a valuable recreational resource for the State of Utah. However,
as the reservoir filled and stratified, the penstocks began to draw water
from deeper and colder strata, reducing summer water temperatures in the
tailwater area below the range necessary to maintain a self-sustaining
trout fishery. By the mid-1970s, trout production in this area declined
markedly.
Mitigation/Enhancement Measures. The Bureau of Reclamation, with
the cooperation of the Utah Division of Wildlife Resources and the U.S.
Fish and Wildlife Service, investigated various methods of controlling
reservoir discharges in order to raise downstream temperatures to allow
the growth and production of the trout fishery. As a result, shutter-
type selective withdrawal structures were retrofitted to the three
penstock intakes. The structures are square towers, each about 200 feet
high, with a series of four shutters along the length of one side. They
IV-27
-------
effectively extended the penstock intakes vertically upward into higher
and warmer reservoir strata so that warmer water could be released
downstream. The shutters are raised and lowered by hoists on the dam to
selectively withdraw water of a given temperature. Modification of the
penstock intakes was completed in 1978.
The selective withdrawal modification was successful in meeting the
fishery water temperature criteria downstream. Fish production and
condition improved. A large increase in numbers of fish food organisms
in the tailwaters was also noted. Other benefits included fewer cases of
hypothermia among river runners and an upstream movement of warmer water
fish communities.
NEUTRAL OR POSITIVE EFFECTS
McCloud Reservoir and Dam
Operating Agency: Pacific Gas and Electric Company
Location. The McCloud Dam is located on the McCloud River in
northern California.
Principal Features. The McCloud Reservoir has a gross capacity of
35,200 acre-feet, and a surface area of 520 acres. The drainage area
covers approximately 269,000 acres and is primarily used for timber
production and recreation. The dam was completed in 1965.
Uses. The McCloud Dam's primary use is to supply water to the
James B. Black hydroelectric powerhouse. Up to 2,000 ftVsec is diverted
through approximately 11 miles of conduit and an intervening reservoir to
the powerhouse, producing 172 MW at normal operating capacity.
Water Quality Conditions. Mud Creek, a tributary to the McCloud
River near the upper end of the reservoir, carries a significant sediment
load originating from the Konwakiton Glacier. Before the dam was built
in 1965, elevated turbidity levels resulting from this glacial runoff
occurred from midsummer to late fall or winter, depending on weather
conditions.
A water quality study was performed by Pacific Gas and Electric
Company due to concerns of resource management agencies regarding the
effect of dam releases on levels of turbidity and temperature downstream
of the reservoir. The water quality data indicate that the reservoir
appears to act as a settling basin for a large portion of the material
that drains to it. Turbidity of release water is generally much reduced
from the levels that would have occurred if the dam had not been
constructed. Although high sediment inputs from Mud Creek cause
turbidity levels and total suspended solids to increase throughout the
reservoir, the majority of the material that flows into the reservoir
remains near the bottom.
IV-28
-------
The study also concluded that the temperature of the river
immediately downstream of the McCloud Dam showed little change from that
of the river upstream of the dam. The thermal structure of the
reservoir, even during seasonal changes in reservoir elevation and inflow
temperatures, is such that little negative effect on temperatures
downstream can be expected as long as mid to lower elevation intakes are
used as sources of water for downstream releases. Reservoir data also
indicate that DO concentrations will remain above standards necessary for
the maintenance of aquatic life. This applies regardless of the intake
elevation used for downstream release.
Mitigation/Enhancement Measures. No mitigation/enhancement measures
are necessary.
SUM1ARY
The fifteen case studies illustrate several water quality effects
and some of the implemented and planned mitigation measures, though they
are not intended to be statistically representative.
J. Percy Priest Reservoir, Old Hickory Lake, Richard B. Russell
Lake, Upper Bear Creek Reservoir, Casitas Lake, and Lake Cachuma were
presented as case studies of low hypolimnetic DO and/or increased
concentrations of iron and manganese in the hypolimnion and tailwaters.
(Old Hickory Lake does not have elevated concentrations of iron and
manganese.) These effects are greatly facilitated in each case by a
seasonal stratification of the impoundment. Aeration of the impoundment
or tailwaters and selective withdrawal were reported as successful
mitigative efforts.
Guntersville Reservoir, Boone Reservoir, and Eau Galle Lake were
presented as examples of eutrophication, with phytoplankton and
macrophytes causing problems. Herbicide treatments, introduction of
grass carp, and hypolimnetic application of aluminum sulfate to control
internal phosphorus recycling are some of the mitigation measures
applied. Watershed management - including best management practices for
nonpoint source and point source controls - is becoming increasingly
important for reducing eutrophication.
Norris Reservoir, Mark Twain Lake, and Fort Patrick Henry Reservoir
are given as case studies of flow regulation and/or reaeration denial.
In these cases, management of low flows is an important issue. Flow
regulation weirs and operational changes are among the mitigative
measures employed.
Lake Red Rock and Flaming Gorge Reservoir were provided as examples
of a sediment accumulation problem and a cold tailwater problem,
respectively. The former has primarily been mitigated by altering the
operational philosophy, while the cold discharges have been reduced by
selective withdrawal.
IV-29
-------
Finally, McCloud Reservoir was included to demonstrate that not only
do some impoundments not have any water quality problems, but they may
actually improve water quality. In this case, the reservoir tends to
reduce downstream turbidities caused by high sediment loads generated by
a glacier.
All mitigative measures employed in the case studies were reported
to be successful to some extent. In some cases, several seasons were
required to completely see their effect, and in other cases, more than
one measure was needed to rectify poor water quality conditions. In all
cases where improvements in water quality conditions have occurred,
there have been significant measurable benefits to either the public, in
terms of improved fishing conditions and other recreational
opportunities, or to dischargers, in terms of increased assimilative
capacity.
IV-30
-------
V. MITIGATION MEASURES
This chapter identifies major mitigation measures that can be used
to address the adverse water quality effects associated with certain
impoundments. Mitigation is also discussed in specific situations
associated with the case studies presented in Chapter IV. Because each
reservoir system is unique, the applicability of specific mitigation
measures must be evaluated on a case-by-case basis. This evaluation must
consider the effects that need to be corrected, the present uses of the
reservoir system, as well as the benefit-cost relationship of the
mitigation measures. Due to the complexity of the reservoir-tailwater
system, mitigation measures have to be carefully evaluated before
implementation. Implementation of more than one measure may be necessary
to improve or maintain the water quality in a reservoir system. The
application of one measure may correct one adverse effect, but create or
intensify another water quality effect. For example, reaeration of
releases to increase DO may cause a problem with nitrogen supersaturation
(TVA, 1978). While mitigation measures are intended to minimize or prevent
certain undesirable water quality effects, they themselves sometimes cause
other adverse environmental effects. For example, gaining access to the
impoundment to implement a mitigation measure may require cutting a road
through a wetland. Another example might be a mitigation measure intended
to control algae growth which also causes fish mortality.
When considering the appropriateness of specific mitigation measures,
most operating agencies must insure that authorized project purposes are
met. If the selected mitigation measure negatively affects or restricts
authorized project purpose(s), its implementation often cannot be justified
without a modification to the authority, regardless of its potential
positive effect on water quality.
Mitigation measures can be divided into three broad categories:
physical measures, operational measures, and structural modification.
Physical measures include technologies that require specific processes or
equipment to be used to correct the problem. Physical mitigation measures
include the control of water quality in the reservoir, selective withdrawal
of reservoir water with acceptable water quality, aeration of reservoir
releases, and habitat modification. Operational measures include changes
to the present operating regime of the reservoir system. Operational
measures include maintaining a minimum discharge, limiting the maximum
discharge flow, and altering the rule curves for reservoir operations as
well as selective withdrawal. Structural modifications involve changes to
the structure of the dam and/or its outlet works; examples would be the
addition of ports, gates, vents, or weirs to modify the depth or manner in
which water is selectively withdrawn from the reservoir.
V-l
-------
The measures in this chapter are grouped according to whether they
primarily affect the pool, affect the tailwater, or are of a more general
nature. Frequently, mitigation measures will have an effect on both the
pool and tailwater because of the complex interrelationships. Depending
on site specific circumstances, all of, the mitigation measures are
candidates for retrofit as well as new construction.
WATER QUALITY CONTROL IN THE RESERVOIR
Several mitigation measures are available which may be specifically
applied to improve water quality in the impounded pool. These include
measures for induced mixing, hypolimnetic aeration, and dredging.
Induced Mixing.
Induced mixing is frequently applied to stratified impoundments to
mitigate water quality concerns related to stratification such as low
hypolimnetic DO, increased iron and manganese, and thermal changes. Mixing
may be induced by the use of low head, high volume mechanical pumps pushing
epilimnion waters into the hypolimnion or vice versa (TVA, 1978). Mixing
may also be induced by air injection in the hypolimnion, which will tend
to move upward with the gas bubbles. If sufficient air is introduced, the
process may continue until a stratified impoundment experiences nearly
uniform temperature and DO distributions (Fast, 1979).
Induced mixing may be applicable to reservoirs of any size (USEPA,
1973). However, it is not always necessary to destratify an entire
reservoir; only the area near the outlet structure may require mixing.
TVA is experimenting at Douglas Dam with three high-volume, low-speed axial
pumps just upstream from turbine intakes; the pumps force oxygenated
epilimnetic water into turbine intakes during stratified periods (TVA,
1987a).
In addition to increasing DO by the introduction of air and/or the
movement of anoxic hypolimnetic waters to the surface where they can be
reoxygenated, mixing can also reduce the occurrence of dissolved iron and
manganese and hydrogen sulfide that are soluble under anaerobic conditions.
If this occurs, the taste and odor characteristics of the impoundment
waters can be improved.
Induced mixing may also result in increasing the total energy (heat)
content of the water body during the summer months, since the average
temperature of the mixed impoundment will more closely approximate the
epilimnion temperature before mixing. If destratification is successful,
the cold hypolimnetic waters may be eliminated. Destratification is
generally considered beneficial for warm water fish, since there is an
V-2
-------
increase in their depth distribution, and therefore, an increase in
available habitat. An increase in food supply is often observed since
benthic fauna production often increases greatly during mixing (Fast,
1979). This is, in part, a result of reduction of barriers to the
distribution of fish, zooplankton, benthic fauna, and other biota (Fast,
1979). However, in some cases, this effect may be undesirable because it
can eliminate cold water fish species. Destratification also increases
the temperature of the water released into the tailwaters. It therefore
may not be appropriate where cold water fisheries are supported, but may
support management of warm water fisheries that in the past have been
impaired by cold water releases of the hypolimnion.
Destratification, through mixing, is notably less effective in
reducing algal densities and primary production. It can upwell nutrients
into the euphotic zone and thereby stimulate algal growth.
Aeration of the Hypolimnion.
Aeration of the hypolimnion can be performed with the intention of
mixing, as described above, or simply to provide additional oxygen to
targeted regions of an impoundment. In reservoirs where the cold water
of the hypolimnion and releases are desirable, but the water quality is
poor, aeration of the hypolimnion may be applicable. Aeration strips
undesirable gases such as carbon dioxide, hydrogen sulfide, and ammonia
(Fast, 1979), and lowers the concentration of iron, manganese, phosphorus
and other conditions associated with anaerobic conditions. Small changes
in temperature usually occur, but cold temperatures are maintained.
Artificial aeration will not affect external loadings of nutrients, but
it may affect the rates and directions of nutrient cycling once the
nutrients are in the reservoir, and in cases where the internal loading
of nutrients is significant, aeration may alleviate some of the symptoms
of eutrophication. It can increase species diversity by increasing the
suitable habitat available for cold water species such as trout, salmon,
zooplankton, and benthic fauna. Like mixing, aeration is not always
successful in reducing the algal standing crop.
Aeration of the hypolimnion of small water supply reservoirs has
been demonstrated to be effective (TVA, 1978). However, it may not be
economical for large reservoirs that are used for power production due to
the large volume of hypolimnetic waters that is continually released
through the turbines. Another drawback of aeration is the possibility of
supersaturation of nitrogen gas occurring in the hypolimnion and in the
tailwaters. Use of pure oxygen gas for aeration is one solution to this
problem.
Methods that control DO within the reservoir generally result in
better overall water quality than those that only increase DO levels in
the releases. However, mitigative measures applied in the reservoir may
not ensure a targeted concentration of DO in the tailwaters and may
require supplemental aeration. Full-scale projects by the Corps of
Engineers and EPA have demonstrated that one disadvantage of the method
is that DO levels in the tailwater releases cannot be regulated very
readily (TVA, 1978), and aeration of the tailwaters may still be
necessary.
V-3
-------
Dredging.
Dredging is the only mitigative measure that directly removes the
accumulated products of degradation from the reservoir, thereby
increasing depths and removing potentially recyclable nutrients
(Peterson, 1979). Increase in water depth is especially important for
reservoirs since they serve as sediment traps and therefore become
shallower with time. Dredging can remove aquatic vegetation, which is
especially important where unwanted species have invaded a reservoir
system. It can minimize the role of the sediments in recycling nutrients
and can lower the oxygen demand of the sediments by removing organic
matter. In Long Lake, Michigan, dredging was successfully used to
improve the fish habitat by increasing lake size and depth and decreasing
the amount of organic sediment (Peterson, 1979).
The disadvantages of dredging relate to environmental concerns as
well as economics. Dredging causes resuspension of bottom sediment and
toxic substances, resolubilization of chemicals, and can cause oxygen
depletion by resuspending settled organic matter. Resuspension of
sediments may reduce primary production rates due to decreased light
penetration in turbid waters, and nutrient levels may increase due to
cheir liberation from deep anaerobic water areas during dredging.
Dredging removes a large number of benthic organisms and can reduce fish
production by decreasing their food supply and spawning areas. Dredging
is costly and requires a disposal site.
WATER QUALITY CONTROL OF TAILWATERS
Additional mitigative measures are directly applicable for
improvements of water quality in the tail waters. These measures include
aeration of reservoir releases, selective withdrawals, and habitat
improvement.
Aeration of Reservoir Releases.
For a reservoir to discharge water with sufficient DO concentrations
to meet water quality standards, aeration of reservoir releases may be
necessary. This has been successful in tailwaters that receive
hypolimnetic waters. Most tailwater water quality effects are centered
on dam releases and their oxygen status.
Under contract to the Corps of Engineers EWQOS program, TVA reported
upon techniques for reaeration of hydropower releases (CE, 1983).
Although aeration may be applied in the tailrace or immediately
downstream, TVA found that most of the research and development
activities of the prior decade have been directed toward turbine venting
or aeration in the reservoir itself. Imbedded within the TVA report is a
translation, from German, of a review by Dr. Peter Volkart that also
V-4
-------
deals with turbine venting as well as diffused air, cascade, surface, and
pure oxygen aerators. Such processes could play a role in situations
where dissolved oxygen is needed and turbine venting is not an option.
Dr. Volkart points out that surface aerators are efficient oxygen
transfer devices, which explains their common use in wastewater treatment
applications.
Turbine venting is a process in which water is aerated as it passes
through a hydroelectric turbine. Venting is used not only for
oxygenation, but as a means to control cavitation and vibration. The
process is one in which air is aspirated or drawn into partial vacuum
regions which occur naturally or are created below the turbine in the
draft tube. A vacuum is created through the installation of baffles or
deflector plates near vent holes which cause a flow separation and
localized low pressure areas near the vent openings. There are
similarities between this process and the lift associated with an
airplane wing created as air flows over the wing causing a vacuum on the
underside.
The TVA team (CE, 1983) reported on turbine venting in the United
States and Europe, including efforts by Duke Power, Alabama Power, Union
Electric, TVA itself, and others. Various hub baffle schemes retrofitted
as field modifications to existing turbines are presented and discussed.
Configurations included oxygen diffusers in the turbine flow, aspiration
into the draft tube below the turbine wheel, and mechanical injection by
compressors. As a process, turbine venting can greatly increase
dissolved oxygen, and many schemes may only be operated as needed. A
disadvantage is a slight drop off in power production when venting is
underway.
TVA is evaluating a number of strategies for improving reservoir
releases (TVA, 1987a). At Appalachia Dam, flow reversion from the
powerhouse tunnel to the stream reach below the dam is being studied.
Turbine baffles or extra aspiration piping are being studied at Cherokee,
Norris, and South Holston dams. At Tims Ford, an air compressor is used
to inject air into dam releases. Turbine pulsing is being tried at Fort
Patrick Henry and Norris Dam; this strategy alternatively increases and
decreases the flows in the reach below the dams. As part of these site-
specific tests and evaluations, TVA measures dissolved oxygen
improvements and considers the associated costs.
Selective Withdrawal of Reservoir Water.
Thermal stratification of a reservoir can allow for the selective
withdrawal of strata with the best water quality for tailwater
conditions. For example, a water level that contains an acceptable DO
concentration and temperature could be released. Since the level at
which appropriate water quality conditions may occur changes, multi-level
intake structures may be necessary. The success of selective withdrawal
also depends on the required volume of water that has to be discharged
compared to the volume of water available in each strata. Fishery
problems can arise when the volume of cold water released in a reservoir
V-5
-------
runs out in mid-summer and warm water must suddenly be released. This
situation precludes support of both cold water and warm water fisheries.
Construction of multi-level intake structures may be too expensive for
existing dams in comparison to the benefits that will be achieved. It is
possible to predict the water quality and hydraulic effects that
selective withdrawals may have on reservoir waters using mathematical
models to determine if such expense is warranted.
Submerged weirs have been used to allow only well-oxygenated surface
water to pass through power units. All submerged weirs now in use are
permanent structures located upstream from the base of the dam (TVA,
1978). These structures are not well-suited for reservoirs whose pool
levels fluctuate considerably. The crest of the weir must be at a
relatively low depth to allow water to pass over it during all times of
the year. Hypolimnetic waters can then pass over the weir when the
reservoir is at normal operating levels during the summer (TVA, 1978).
Habitat Improvement of Tailwaters.
Tailwater management to support fisheries is a relatively new
science developed over the past thirty years. The first time it was
noted that warm water species of fish did not reproduce below a dam was
in 1943 (Pfitzer, 1975). In the years that followed, this phenomenon was
repeated each time a high dam was completed and discharged cold water
into formerly warm water (Pfitzer, 1975). However, with proper
management many tailwaters can support excellent fisheries.
Tailwaters otherwise suitable for fisheries, but limited by
inadequate minimum flow and low concentrations of DO, might be improved
by physical changes to the environment. One method of habitat
improvement related to the problem of minimum flow is increasing the area
of continuously wet substrate and the extent of reaches with deep water.
This can be accomplished by multiple rock structures, small dams, and
wing walls, which all result in formation of a series of small pools
separated by reaches of fast, turbulent water. These structures can also
serve to increase turbulence, and therefore DO. Unfortunately these
structures would not solve the problem of low DO during periods of
maximum discharge.
Another method to improve the habitat of tailwaters is flow
regulation of the discharge waters. A fairly accurate habitat
maintenance flow can be determined by studying shoal and riffle areas at
different flow regimes. There is an ideal minimum flow that would
maximize fisheries (Wiley and Mullen, 1975). Usually a compromise flow
is implemented as a result of conflicting needs of other uses of the
reservoir. The problem of control of releases is also discussed in the
Operational Changes section.
The Fish and Wildlife Service has developed a detailed methodology
to evaluate habitat (FWS, 1982). Their method is termed the "Instream
Flow Methodology" and is one of the methods used in Federal Energy
Regulatory Commission (FERC) licensing proceedings to evaluate the
V-6
-------
downstream impacts of hydroelectric projects on aquatic habitats.
Physical changes to the tailwaters (artificial spawning areas) and
operational changes are sometimes required to receive a license to
generate power. The method can be applied in a variety of situations.
OTHER MITIGATION MEASURES
Other mitigation measures of a more general nature may be applied.
These measures have the potential for improving water quality in the pool
and tailwaters. These measures are predominantly of a management nature
and include watershed management and changes in dam operations.
Watershed Management.
The two major sources of pollutants entering a reservoir are
nonpoint sources and point sources. Since adequate treatment measures
for point sources of pollution have already largely been installed, this
section will focus on controlling nonpoint sources of pollution in the
watershed. There are numerous and divergent sources of nonpoint source
pollution, and each type of source has its own type of treatment. For
example, sediment transport to a large reservoir can be dispersed by
smaller upstream impoundments. Major land uses in the watershed, such as
agricultural, urban development, and suburban development, contribute to
nonpoint source pollution in different ways. Therefore, the controls
implemented depend on the watershed land use.
Watershed management for nonpoint source pollution also requires the
cooperation and interaction of citizens, local governments, and state and
Federal agencies. One mechanism that has aided the implementation of
nonpoint source controls has been the formation of watershed districts
that serve as focal points for the identification of pollution problems
and for allocating funds for improvements. For example, at White Clay
Lake in Wisconsin, the creation of a Lake Protection District allowed
local citizens the opportunity to assist in the development and
implementation of land management plans (Peterson, 1979). It can also
help increase the public awareness of the possible controls and the
contribution individuals can make in reducing pollution.
Watershed management for reservoir water quality is extremely
important, yet difficult to effectively institutionalize. Reservoir
management agencies often have little or no administrative or regulatory
responsibility or authority in the watershed. Thus, they are faced with
the difficult task of dealing with the symptoms rather than the cause of
poor water quality. If the water quality of reservoirs is to be
adequately managed, greater emphasis must be placed on the development of
cooperative management approaches on a watershed-wide basis. This will
often require interagency cooperation. An example of this recent level
of concern is the Chesapeake Bay Agreement, reached in 1987 by the
governors of the states with watersheds draining into the Bay. The
governors formally - and very publicly - agreed to establish state-wide,
V-7
-------
watershed management programs, to reduce nutrient inputs to the Bay by 40
percent over the next ten to twenty years. Improved quality of the Bay
is the overriding goal. In a similar attempt, TVA is also organizing
county-level activities designed to control nonpoint source inputs
through watershed management.
One of the most visually evident sources of nonpoint source
pollution is soil erosion, resulting in sediment loads transported to,
and deposited in, reservoirs and the streams that feed them. Also
transported along with the sediments are solid mineral and organic
matter, and absorbed chemicals such as pesticides, herbicides, and
nutrients. There are numerous methods to help control soil erosion.
Examples include: limiting the extent and exposure time of bare ground;
keeping bare ground covered with mulches or protective matting during
construction activities; limiting construction to periods of minimal
precipitation; diverting runoff around exposed areas; utilizing settling
basins and silt retaining fences to reduce runoff velocity and to trap
suspended sediments; sloping reservoir banks to facilitate vegetation;
seeding exposed banks and revegetating banks with natural trees and
shrubs for erosion and thermal protection; and in areas where this is not
possible, installing rip rap (USEPA, 1977). Construction activities in
the watershed can greatly impact the sediment load to the reservoir.
Enforcing the implementation of the above measures during construction
activities can greatly reduce erosion.
Since dissolved pollutants such as nutrients will not be removed by
the above methods, different approaches must be utilized for their
control. Optimum application rates of pesticides, fertilizers, and
herbicides, as well as timing considerations are important, along with
suitable disposal of wastes with their application (USEPA, 1977).
Education of the public in the proper use of fertilizers, herbicides, and
pesticides can be especially important. In areas where these impacts are
critical, regulation of the time and extent of their use may be
necessary.
Depending on the location of the reservoir watershed, urban or
agricultural nonpoint sources may be major contributors of pollution.
There are numerous programs already designed to help farmers control the
loss of sediment, nutrients, and chemicals from their cropland and
pasturelands. These are not only effective in controlling pollution, but
also help reduce farming costs and increase farming efficiency. Examples
of agricultural management practices include contour farming,
conservation tillage, livestock waste management systems, and crop
residue management. There are also numerous methods to control urban
runoff that either reduce runoff or delay runoff, such as increasing the
extent of pervious areas, and ponding and detention measures for
impervious areas. Frequently, however, these options are limited because
the operating agency for a dam may lack the authority and/or the
necessary cooperation to implement these measures.
V-8
-------
Operational Changes.
Water quality impacts may be lessened by changes in the operational
procedures of a dam. However, the majority of dams are not, and cannot
be, operated solely for the purpose of achieving water quality
objectives. Conflict of interest can occur if water quality objectives
are added to the other primary operational purposes of the reservoir,
such as flood control or power generation. The losses due to changes in
operation for the primary uses would have to be compared to the benefits
of protecting or enhancing stream uses.
While the majority of reservoirs are not operated for the specific
purpose of water quality, most Corps of Engineers reservoirs are operated
to achieve water quality goals within operational constraints. Water
control plans for Corps of Engineers projects are developed to meet the
authorized project purposes. Deviations from these plans can be
accommodated only if there is no adverse impact on the authorized
purposes; otherwise, additional authority is required. Most other
involved federal agencies have similar constraints. Changing operations
requires justifications and authorizations.
The requirement to maintain a minimum, constant discharge would
greatly benefit tailwaters where zero discharge occurs periodically.
The advantages of maintaining a minimum flow is that it helps avoid rapid
temperature fluctuations, reduces the impact of low DO concentrations
through natural aeration in the tailrace, and increases the continuously
wet surface area of the stream, therefore increasing habitats for benthic
biota and fish. The drawbacks of this change in operation are that it
causes a loss in the flexibility of peak-power operation and a decrease
in power operation efficiency. Discharges of less than minimums required
to operate turbines in power dams may have to be sluiced, leading to a
complete loss of the energy potential of this water. Reservoir water
levels may also be affected, adding to the complexity of river management
and flood control (TVA, 1978, 1987a).
One compromise type minimum flow is a time-volume release, in which
relatively large volumes of water are released in short pulses during
periods of otherwise no discharge. For example, at a project where two-
or three-day periods of no-flow may frequently occur from March to
October, a release schedule could be adopted that would allow a brief
discharge of perhaps one full load on one generator. TVA is evaluating
such turbine pulsing (TVA, 1987a). Such a discharge schedule may
occasionally result in a minor loss in total system power production
capability. The discharge of water following this schedule would permit
electrical power to be generated and at the same time provide fresh
volumes of cold water in the tailwater (Pfitzer, 1975).
Limiting discharge to a certain maximum flow reduces impacts on DO
during the period of low concentrations in the discharges through natural
aeration in the tailwaters. This method involves a loss of flexibility
in peak-power generation and may place unacceptable constraints on flood
control operations (TVA, 1978).
V-9
-------
VI. FEDERAL AGENCIES' WATER QUALITY ASSESSMENTS OF IMPOUNDMENTS
INTRODUCTION
This chapter presents the results of assessments by the Corps of
Engineers (COE), Tennessee Valley Authority (TVA), and the Bureau of
Reclamation (USBR) of water quality at their impoundments. These three
agencies specifically requested that they be allowed to contribute
individual assessments of the dams they manage. None of the other
commenters offered similar assessments.
Water impoundments operated by these three Agencies represent a wide
variety of geography, climate, and operational situations. COE dams are
concentrated in the industrialized areas of the Southeast and the Ohio
River Basin, coastal areas, the Pacific Northwest, and along mainstem
navigable rivers. TVA dams are located in a well-developed and partially
industrialized extended river basin. USBR dams are located solely in the
17 western states and Hawaii. These Agencies were invited to prepare their
Agency assessments as a supplement to this Report to Congress in
recognition of the importance of regional effects and site-specific
characteristics on dam water quality. The original submittals by the
Agencies are provided in the appendices (Appendix F, COE; Appendix G, TVA;
and Appendix H, USBR). These original submittals provide the following
information specific to each Agency's area of interest:
(1) Statement of policies and procedures followed by the agency in
the development and management of water resources.
(2) Assessment of water quality with respect to the agency's dams.
The three Agencies adopted a questionnaire developed by Kennedy,
Gunkel, and Gaugush of the U.S. Army Corps of Engineers Waterways
Experiment Station (WES) at Vicksburg, Mississippi, as a uniform instrument
for collecting dam-related water quality information on their respective
projects. A copy of the questionnaire tailored to the USBR, with its
instruction sheet, is attached as Appendix I.
Basically, the questionnaire presents field personnel with a
comprehensive list of water quality attributes. They were asked to
subjectively rate each attribute in the tributary, pool, and tailwater of
each reservoir on the basis of the extent to which this attribute is a
problem, the level of impact of the attribute on user benefits, and the
reliability of the data upon which the rating is being made. A computer
program, also developed at WES, compiles the data from the questionnaires
into a SAS data file. Statistical analyses can then be performed on the
various attributes and their ratings using the SAS software. In the
VI-1
-------
short time allowed for this study, the three Agencies limited their efforts
to frequency analyses of each attribute's extent and impact in the pools
and tail waters of their reservoirs.
The Agencies feel the limited analysis presented herein gives an
accurate picture of the known extent of given water quality conditions
across a broad range of geography, climate, and project operating criteria,
along with an assessment of the perceived impacts of these conditions on
user benefits. This information is presented from the point of view of
field personnel who are directly responsible for daily dam operations and
the delivery of promised project benefits.
The remainder of this chapter contains summaries of the results of
each Agency's assessment of water quality related to its dams and
recommended research needs, followed by an overview of the data presented.
Material submitted by the Agencies has been incorporated directly into this
chapter.
CORPS OF ENGINEERS
Background.
The Corps of Engineers has constructed and now operates more than
700 water resource projects having a total surface area of nearly 10,000
square miles. The geographic distribution of these projects, as depicted
in Figure VI-1, reflects regional differences in water resource development
requirements, water control agency responsibilities, and topographic
requirements for cost-effective construction. Impoundments providing
navigation benefits, which comprise approximately 26 percent of all Corps
projects, are located along major inland waterways. These include the
Mississippi River and its major tributaries, the Arkansas and Red Rivers
draining from the west, and the Ohio and Illinois Rivers draining from the
east. Other waterways of importance include the Alabama River and the
Tennessee-Tombigbee Waterway in the mid-south, and the Columbia River in
the northwest. Twenty-one percent of all projects are dry dams or projects
which, by design, provide minimal permanent water storage during nonflood
periods. These projects are most prevalent in the arid southwest, where
flooding conditions are associated with intermittent periods of excessive
runoff, and in the New England states.
Reservoir projects providing short- and long-term storage of water,
but not navigation benefits, comprise the remaining 53 percent of all COE
water resource projects. These projects can be broadly categorized based
on reservoir morphometry and tributary type. Deep, storage reservoirs are
formed by the impoundment of higher order streams and rivers, and are
frequently located in deep, steeply-sloped river valleys. These projects
tend to be deep, narrow, and highly dendritic in shape. Mainstem
reservoirs are located on lower order (i.e., larger) rivers and tend to be
shallower, wider, and less complex in shape.
VI-2
-------
FIGURE VI-1. - Geographic distribution of Corps of Engineers water resource
projects.
FIGURE VI-2. - Geographic distribution of a ten percent, stratified, random
sancle of Corps of Engineers projects for which questionnaires
have been received.
VI-3
-------
Water control management programs provide the means for operating
Corps' projects to meet their authorized purposes. Most of the Corps'
projects are authorized for multiple purposes (e.g., flood control,
navigation, and recreation), and over 60 Corps of Engineer projects
include water quality as an authorized project purpose. Flow
augmentation for industrial and municipal pollution abatement, acid mine
drainage abatement, and other purposes which relate to water quality, are
often included in flood control and navigation projects. Water quality
management objectives have been developed for each project and
incorporated into Corps water control programs wherever possible.
COE Water Quality Assessment.
For the purposes of this report, information was obtained through
the use of a questionnaire designed to solicit information concerning
project design, operation, and water quality status. The questionnaires
were completed by COE personnel familiar with each project and its water
quality characteristics. With regard to water quality status, subjective
responses to questions concerning water quality were requested. In
general, these responses indicated the presence or absence of water
quality problems. In situations where problems were indicated, graded
responses allowed assessment of the severity of the problem and the
quality of the information upon which the assessment was based. To date,
questionnaires for approximately 470 of 700 projects have been completed
and compiled.
Since questionnaires for all projects have not yet been completed, a
sample of questionnaires was randomly drawn and analyzed by the COE for
the purpose of this report. The sample size was set at 10 percent (46
projects), and samples were drawn from strata based on project type
(reservoir, lock and dam, and dry dam) and COE District. The geographic
locations of sampled projects are presented in Figure VI-2 for comparison
with the distribution of all projects (Figure VI-1). Results presented
below are based on these analyses.
Figures VI-3 and VI-4 present the water quality status of tailwaters
and pools associated with the sampled projects, respectively. A
shortcoming of the data upon which these figures are based is the fact
that reliable information concerning water quality status is lacking for
approximately 40 to 50 percent of the projects. Thus, a degree of
uncertainty and/or bias exists for data discussed here, and
extrapolations of data compiled for the sampled COE projects to all COE
projects are not possible. The data do, however, provide a general
assessment of the types of water quality concerns associated with COE
water resource projects and some indication of their relation to other
project attributes.
As depicted in Figure VI-3, approximately 60 to 65 percent of those
sampled projects for which evaluations of the water quality status of
tailwaters were available were considered not to exhibit problematic
conditions. For those projects indicated as exhibiting problematic
conditions, several categories of water quality concerns are apparent.
VI-4
-------
o
o
te
0.
CM
Q
O
s
LLJ
CO
O
QC
Q.
O
Z
O
CO
o
o
o
H
Z
I-
H
S
QC
UJ
O
QC
Z
o
111
X
111
c
o
C
O
O
rt
3
O
3 W
O >-
O 0)
M- 5
o =
IS
D-O
0)
uI-E
CO
0)
VI-5
-------
o
o
LO
n
O
c
o
o
rt
3
O
k_
,£
is
crO
-------
Most prevalent are concerns related to flow, the release of waters low in
dissolved oxygen concentration, and the erosion and transport of
sediment.
Extremes in flow and/or excessive changes in flow, which result from
operational procedures required to meet authorized project purposes
(e.g., flood control, power generation, etc.), may impact downstream
uses.
Flow fluctuation, although not a water quality problem per se, is an
important factor affecting water quality and use in tailwaters, and is
frequently a key water quality management issue. Flow-related problems
for tailwaters include higher than normal flows following flood events as
retained flood waters are released, lower than normal flows during
periods when pool storage is being increased, and daily fluctuations
resulting from the operation of hydropower facilities, particularly when
power is produced to meet peak-load requirements.
The loss of dissolved oxygen in the hypolimnia of reservoirs
potentially results in direct and indirect impacts for tailwaters. For
projects which, because of their structural or operational
characteristics, do not allow for complete reaeration of release water
as, dissolved oxygen concentrations below saturation may occur throughout
part or all of the summer stratified season. Such is the case for
approximately one-third of the COE projects inventoried here. However,
the COE notes that only one of the 46 sampled projects experiences
periodically severe dissolved oxygen conditions in its tailwater and that
this project is a newly-filled reservoir where such occurrences are
predictable and short-lived.
The occurrence of elevated concentrations of metals and nutrients in
tailwaters is indicated for approximately 30 to 40 percent of the sampled
projects for which such evaluations were made. And, as reported by the
COE, these projects are primarily those for which reduced dissolved
oxygen concentrations were reported. These projects are also reported to
receive relatively high inputs of metals and nutrients from their
surrounding watershed.
The transport of suspended sediment from reservoir to tailwater
and/or the erosion and resuspension of bank and bed materials impacts
tailwater areas below approximately 40 to 50 percent of the projects for
which evaluations were provided. In most cases, impacts are minor and
result from increased turbidity. In other cases, degradation of
immediate downstream areas is indicated. Preliminary evaluation of
information by COE suggests that, while project operation plays a
significant role in the determination of release conditions, pronounced
regional patterns in the distribution of such conditions are apparent.
In general, reservoirs located in regions dominated by highly-erodible
soils experience higher inputs of suspended sediment and, therefore,
often release turbid waters.
VI-7
-------
An evaluation of water quality conditions in pools, also based on
sampled responses to the questionnaire, are presented in Figure VI-4.
Most prevalent were problems related to the eutrophication process.
These include excessive nutrient concentrations, algal blooms, reduced
water clarity, macrophyte infestations, and the loss of dissolved oxygen
in bottom waters. Other conditions of concern include excessive
concentrations of reduced iron and manganese in bottom waters, and the
accumulation of sediment and contaminants. As was discussed for
tailwaters, problematic conditions were identified for approximately 40
to 50 percent of the pools for which evaluations were available. And,
again, varying degrees of severity in problematic conditions are
apparent.
TENNESSEE VALLEY AUTHORITY
Background.
The TVA system of multipurpose dams encompasses more than 11,000
miles of shoreline and 940 square miles of surface water (see Figure
VI-5). Construction was largely completed by the late 1950's. The
primary purposes of TVA's projects are navigation, flood control, and
electrical generation. The TVA system includes 33 large dams (29
hydropower and 4 nonpower) and additional smaller dams. Due to their
age, the TVA impoundments are representative of mature reservoirs.
With the completion of the dam construction program, TVA has turned
its attention to managing the reservoir system and promoting the proper
growth, conservation, and management of the agency's natural resources.
As part of this continuing effort, the TVA Board of Directors authorized
in September 1987 the broadest reassessment in 50 years of the operating
policies of its dams and reservoirs. The central issues being addressed
VI-8
-------
LU
J—
OO
>-
OO
UJ
o
C£.
o
OO
LU
o;
LU
oo
o
0.
LU
OO
OO
a:
VI-9
-------
by the study are whether water quality and recreation should be added as
primary purposes of TVA reservoir operations to the statutory purposes of
navigation, flood control, and electrical generation. The study is being
conducted in accordance with the procedures of the National Environmental
Policy Act and will determine the long-term policies that should direct
TVA efforts in reservoir system operations and river management into the
next century. The current schedule calls for presentation of the final
report and Environmental Impact Statement and the results of public review
and comment of the recommendations in 1989.
TVA Water Quality Assessment.
Water quality conditions for reservoirs in the Tennessee Valley were
assessed using the approach applied by the Corps of Engineers to assess
its projects. Thirty-three projects were assessed, including all the
hydropower projects and four nonhydro projects. Those projects not
included are small projects for which no data were available. The results
are presented in Figures VI-6 and VI-7 for pools and Figures VI-8 and VI-9
for tailwaters.
Water uses were severely impacted at several sites. Low dissolved
oxygen, hydrogen sulfide, iron, and manganese are considered to be at
sufficient levels that the fishery at Upper Bear Creek Reservoir is
practically nonexistent. Other reservoir projects having severe use
impairment are the three Ocoee River projects where sediment accumulation,
iron, manganese, turbidity, and metal contaminants are adversely impacting
aquatic life and recreation, primarily in Ocoee Number 3, with less
impairment in Numbers 2 and 1. Finally, the Nolichucky Reservoir has been
filled with sediment to the point that it is no longer considered a
reservoir.
The results indicate that in reservoir pools the most significant
impacts were pool level fluctuations and bacteria (about 50 and 30 percent
of the reservoirs, respectively). The next most significant user impacts
were turbidity, algae, macrophytes, sediment accumulation, and shore
erosion, all of which occur at 15-20 percent of the reservoir project).
Minor impacts, occurring at 20 percent or more of the reservoir projects,
were related to the following parameters; iron, manganese, low dissolved
oxygen, turbidity, low temperature, high nutrients and algae, macrophytes,
sediment accumulation, pool level fluctuation, shore erosion, pH/acidity,
bacteria, and fish parasites.
Several items worth noting for the analysis on TVA pools are:
(1) data on hydrogen sulfide were limited and the results of this
analysis may change when more data becomes available;
(2) several parameters had high rates of recurrence with the
potential to impact uses in the future: i.e., high nutrients,
sediment accumulation, and shore erosion;
VI-10
-------
<
Q
O
z
HI
_l
m
O
DC
Q.
S\
V
V
<
°
°
O
o
I
o
8S
-•i
o 00 o)
l«2£
Q oc
2S>
E •=
0
0) Q.
C/5
o .S
U_ li.
.$275
HI
I-
X
HI
V>
C
o
c
o
o
>»
15
O
i_
O
CO
"
o
C CO
-2
»- o
3°-
o ±
00
>^- v.
O O
£.=
on
^
CD
vr-ii
-------
ca>o<<1)<1) -. -;r co co o
2
•o
c c
eo oj
c c
E S
CO <0
*- *-
c c
O O
CO (U
c ro
SS
r~ f~ o '••'
* r w —
3 •?. W m
co
o>
S-3
•= o
« c
CO fl,
= «
LL O
^ ui
o^
8|
T3 <
4) =
o OT
m «
CO h.
CQ (0
^o
O
'E
O
O
0)
CO
u. u_
u) in
VL iZ
(0 CO
.Si 75
£5
<
<
Q
U
<
Q.
5
O
z
I
DC
O
tu
LU
V)
O
<
Q.
O
v.
O
0)
O
cc
O
O
<0
*- CO
03 II
0) O
(0 O
DO.
a:
Z3
C3
VI-12
-------
i .2 .2 3 3 E ~
IJ- "- "• OB 03 « O
»_ k. H- CO
$ £ D) CD CD
O O) C Q. Q.
h- i •- •= c c ;= «
—* T ** C E c CD
2 0) CD C >
<
Q
O
Z
111
_l
m
O
oc
QL
Y////////A
OCCASIONAL
O
cc
I
u
Z
HI
I-
X
UJ
CO
c
o
c
o
o
CO
a
i_
o
-------
ij*>5l
s = H°x-
™ 0) -^ CO
S O) 3
2 o
•o .E
ooo^£"°S£o~coa>oO r •£ CD w w
!Tirir23|=£2--S o>S/55 ^ -o cB £ ® ®
^"•^SS^Sllsl^loOo^Jgu.".
$£0)0)0) ££33 CLmT3.E S-S-^^SStSz 2£TfflJ?-S-S
0) 0)
Q. CL
E E
0) 0)
O O)
Jsi
*- _>
3 tn
O CO
CO CO
£c w
o o
o o
^ ° CO
o-cSSS
-••i. *££
g CL
£ CO
CO .«
CO CO
.S2 75
CO £
0)5
<
o
u
<
Q.
O
Z
I
cc
O
o
<
Q.
O
w
0)
DC
c
0)
0)
II
o
(0 0)
°-^
c ®
-
CD
VI-14
-------
(3) low dissolved oxygen seasonally occurred in about 70 percent of
the reservoirs, but user impacts were considered minor for most
projects because the condition was restricted to the
hypolimnion or bottom waters of the reservoir; and
(4) more data are needed on organics and metals in fish flesh.
For reservoir tailwaters, the results indicate that the most
significant impacts resulted from low dissolved oxygen, streamflow (high,
low, and fluctuating), and low temperature. The next most frequent user
impacts were associated with iron, manganese, hydrogen sulfide,
turbidity, metal contaminants, streambank erosion, bacteria, and fish
parasites. All of these parameters occurred at about 15-20 percent of
the projects. It should be noted that the questionnaire approach for
this assessment did not differentiate between the significance of
physical and chemical parameters; therefore, it did not reveal that
chemical problems generally have more serious impacts on uses. The most
frequent minor impacts, occurring at 20 percent or more of the projects,
were related to the following parameters: low dissolved oxygen,
turbidity, high flow, fluctuating flow, low temperature, high
temperature, fluctuating temperature, streambank erosion, and parasites
in fish.
VI-15
-------
THE BUREAU OF RECLAMATION
Background.
The Bureau of Reclamation of the U.S. Department of the Interior is
responsible for the development and conservation of the Nation's water
resources in the western United States and Hawaii. In most areas of the
17 western states, which constitute the main area served by the USBR,
less than 20 inches of moisture fall each year. However, several
important rivers, fed mainly by the melting snow packs in the mountains,
flows through these states. A basic function of USBR is to harness these
streams and to store their surplus waters in times of heavy runoff for
later use when the natural flow is low. USBR water impoundment project
purposes cover a wide range of interrelated functions. These include
providing municipal and industrial water supplies; hydroelectric power
generation; irrigation water for agriculture; water quality improvement;
flood control; river regulation and control; fish and wild'Mfe
enhancement; and outdoor recreation.
Reclamation project facilities in operation during 1986 included:
349 storage reservoirs; 50 hydroelectric power plants; 288 circuit miles
of transmission lines; 15,804 miles of canals; 1,382 miles of pipelines;
276 miles of tunnels; 37,263 miles of laterals; 17,002 miles of project
drains; 240 pumping plants; and 254 diversion dams. Over 20.5 million
people receive municipal and industrial water, 13.8 million kilowatts of
installed hydroelectric power capacity exists, nearly 10 million acres of
western farm land receive full or supplemental irrigation, and 53.2
million visitor days of recreation are recorded annually.
VI-16
-------
Completed water service facilities are transferred to local water
user organizations for operation and maintenance as soon as the
organizations become capable of assuming these functions. USBR operates
and maintains hydroelectric power plants and some water storage and
supply works on multipurpose projects.
USBR Water Quality Assessment.
Information on water quality conditions and user impacts for all
USBR storage reservoirs and tail waters was solicited by distributing the
questionnaire to all six Bureau regional offices. All regions responded,
and a total of 250 questionnaires were returned. The geographical
distribution of this response by state is shown in Figure VI-10. Since
there are approximately 349 USBR storage reservoirs in 17 western states
and Hawaii, this response represents nearly 72 percent of the total.
Information obtained on the frequency of occurrence of various water
quality conditions in USBR reservoirs and their impact upon user benefits
are summarized in Figures VI-11 and VI-12, respectively. Tailwater
conditions and their impacts are shown in Figures VI-13 and VI-14,
respectively.
What is most immediately apparent in these four figures is that in
an average of 54 percent of the reservoir cases and 59 percent of the
tailwater cases, there are no data upon which to make an evaluation of
the conditions or their impact on user benefits. Water quality data are
usually only collected on a particular USBR project when some problem is
noted or suspected, or when some change in the structure or operation is
contemplated. Consequently, the picture of water quality conditions in
Bureau reservoirs and tailwaters given by the available information is
probably somewhat skewed toward those situations where some problem is
perceived or an impact is felt. The following assessment is, therefore,
probably conservative.
USBR data suggest that the main conditions affecting user benefits
in Bureau reservoirs are drawdown, pool fluctuation, turbidity, sediment,
and shore erosion. These conditions arise from the way water storage
reservoirs are operated in an arid climate, where spring snowmelt or
winter rains are the major source of runoff, and drawdown 1s cofltlmwtrs
throughout the long dry season. Drawdown was rated as having a severe
impact on user benefits in six USBR impoundments, and a significant
impact in 33 others, out of a total sample of 107 reservoirs with
information available. Thus, the cumulative percentage of reservoirs
with data in which drawdown was rated as having at least a significant
impact on user benefits is 36 percent. Corresponding cumulative
percentages for the other four conditions are: pool fluctuation
35 percent, turbidity 13 percent, sediment 13 percent, and shore erosion
10 percent. The last condition, shore erosion, had no severe impact
ratings out of a total of 108 impoundments rated.
VI-17
-------
O
3
O)
CO
»*-
O
Q)
i/t
I/)
-------
<
Q
UJ
_l
m
o
oc
Q.
O
z
I
OCCASIONAL
CO
c
0
c
o
o
CO
D
a
0>
75
!f
o
o
§5.
52
8g
o t
+- s
o ^
II
om
o
o
UJ
J-
X
UJ
VI-19
-------
<
Q
a
2
O
O
<
a.
tr
o
CO
+••
«?
o
c
o
m
i_
o
(0
D
c
o
c
o
O
= 10
CO CM
rt II
CO CO
"5 ^
CO CO
f o>
IQ:
iCD
D to
CM
!—I
I
f—H
s>
LU
VI-?0
-------
<
Q
01
_i
m
O
oc
Q.
I
OCCASIONAL
U
LU
h-
X
LU
(0
c
o
c
o
o
3 I!
(0
o
W CO
o o
Of£
OOC
tm
o
0
o .2
= •(5
^
oo
CD
VI-21
-------
<
Q
o
Q.
c
0)
o
0)
Q.
NOR
DC
CQ
CO
CO
"o
c
0)
CQ
0>
c
0
co
v-CV
n"
o
LL
Z
g
CO
o ro
ca 5
CL=
O
-------
The second most important set of reservoir water quality conditions
affecting user benefits are related to eutrophication: algae, high
nutrients, low dissolved oxygen, and taste and odor problems. Cumulative
percentages of reservoirs with data where these conditions were rated as
having at least significant impacts on user benefits were: algae 17
percent, high nutrients 15 percent, low dissolved oxygen 14 percent, and
taste and odor problems 12 percent. There were no severe impact ratings
for high nutrients or taste and odor problems, however.
Although iron is often present in USBR reservoirs, it was rated as
having at least a significant impact on user benefits in only 4 percent
of the rated reservoirs. In fact, it should be noted that only drawdown
and pool fluctuation were perceived as having really significant impacts
on reservoir user benefits.
Tailwater conditions and user impacts are depicted in Figures VI-13
and VI-14, respectively. Here again, the major impact-producing
conditions seem to cluster around the mode of operation of water supply
reservoirs in an arid region: high flow, low flow, turbidity, and high
temperature. Of these, high flow was rated as having significant user
impacts in 21 percent of the tailwaters with data, but in no case was it
rated as having a severe impact. The other three conditions were rated
as having at least a significant impact in 13 percent, 12 percent, and
11 percent of the rated tailwaters, respectively. Taste and odor
problems were rated as significant in 13 percent of the tailwaters with
available data, but were not considered severe in any case.
VI-23
-------
OVERVIEW
The Corps of Engineers (COE), Tennessee Valley Authority (TVA), and
U.S. Bureau of Reclamation (USBR) each prepared an assessment of known
water quality conditions at their water impoundment projects. These
assessments are based on a questionnaire completed by Agency personnel
familiar with the water quality conditions at each project. The result is
a subjective data base for each Agency, presenting the frequency of
occurrence of specified water quality conditions and their relative degree
of impact to user benefits. Parameter-specific data were not available for
a large percentage of the projects responding to the questionnaires. Also,
it is likely that existing water quality data were collected primarily at
projects where some problem was noted or suspected. Consequently, the
available data on water quality conditions may be skewed, introducing a
pessimistic bias. It is, therefore, difficult to project results of these
assessments to each Agency's population of impoundments.
The water quality conditions described in the results of the
agencies' questionnaire survey are a mix of conventional water quality
parameters (e.g., low dissolved oxygen, pH, turbidity, taste and odor,
bacteria, nutrients, dissolved solids, metals, and organics) and chemical
and physical parameters of specific interest to impoundments (e.g., iron,
manganese, hydrogen sulfide, temperature, algae, sediment accumulation,
pool level fluctuations, and shore erosion). The physical parameters are
reported as occurring with greater frequency and at higher levels of impact
than the chemical parameters. Pool and tailwater fluctuations; high and
low flows; erosion, sediment transport, and sediment accretion were the
primary physical water quality problems reported. It is not known whether
the physical parameters are actually more problematic or whether, in view
of the lack of hard monitoring data on chemical water quality, the more
visible physical parameters are simply easier to detect and report
subjectively. For the TVA system, even though physical concerns are more
prevalent, chemical concerns are generally more serious when they occur.
Low dissolved oxygen in pool hypolimnia appears to be a seasonal
concern primarily in projects located in the eastern half of the country.
Low dissolved oxygen occurred as at least an intermittent problem in
47 percent of the COE reservoirs, 70 percent of TVA's reservoirs, and only
9 percent of USSR's reservoirs. Tailwaters experienced similar results,
with low dissolved oxygen reported as at least an intermittent problem in
25 percent of COE's tailwaters, 38 percent of TVA's tailwaters, and only
4 percent of USSR's tailwaters. The USBR assessment showed that problems
of low DO, dissolved iron and manganese, and hydrogen sulfide are
relatively rare at Bureau dams, although summer thermal stratification is
nearly universal and low level outlets are not uncommon. Some major
differences between these reservoirs and southeastern reservoirs, where low
DO problems may be more prevalent, are that the western reservoirs are, on
the average, less eutrophic (i.e., lower BOD), more rapidly flushed, and
subject to shorter thermal stratification periods. In general, the factors
which are involved here include an arid climate, relatively low inflow
nutrient concentrations, a lower intensity of development in the
watersheds, and a wider range of seasonal water temperature fluctuations.
The size and depth of most Bureau reservoirs are not substantially
different from those in the southeast, however.
VI-24
-------
Coincident with the loss of dissolved oxygen from pool hypolimnia is
the increased release of dissolved materials from bottom sediments. Iron,
manganese, and hydrogen sulfide often appear in elevated concentrations in
the tailwaters of projects with low dissolved oxygen; the COE assessment
reports a clear link in these parameters where data were available. TVA
noted that iron, manganese, and hydrogen sulfide appeared in 50 percent of
their projects with low dissolved oxygen.
The COE assessment reports pronounced regional patterns in the
distributions of some water quality conditions. For example, reservoirs
located in areas with highly erodible soils experience greater sediment
loads and generally experience greater problems with turbidity in tailwater
releases and sediment accretion in the pool. Similarily, reservoirs
experiencing high nutrient and/or metals inputs from the surrounding
watershed typically reported more problems with eutrophication, low
dissolved oxygen, and other related issues in both pool and tailwaters.
Many of the results reported by TVA and USER can also be explained by
regional/local influences of land use, geology, topography, and climate.
VI-25
-------
VII. CONCLUSIONS
This report addresses general issues of water quality effects
associated with impoundment of water by dams. It attempts to estimate the
character and national extent of these effects through a literature review
and analyses conducted on a random sample of a partitioned data base
population of 68,155 dams. Insufficient data were collected to draw
quantitative conclusions pertaining to small dams (less than 10,000
acre-feet normal storage volume) and to a large degree, large nonpower dams
(at least 10,000 acre-feet of normal storage volume and no reported power
generating capacity). Quantitative and qualitative conclusions are drawn
for large power dams (greater than 10,000 acre«feet of normal storage
volume and 100 kilowatts or more of installed power). General conclusions
regarding likely water quality effects are derived from the literature
review. Specific conclusions are based on the results of the four EPA
analyses. Three federal agencies, the U.S. Army Corps of Engineers (COE),
the Tennessee Valley Authority (TVA), and the U.S. Bureau of Reclamation
(USBR) conducted independent assessments of water quality conditions at
their respective water resource projects. Due to the complex
interrelationships of the potential water quality effects, the numerous
variables affecting impoundment water quality, and the lack of sufficient
detailed information, it is difficult to draw accurate conclusions
regarding the national extent of water quality effects attributable to
impoundments.
This study is limited in estimating the national extent of dam water
quality primarily because of a lack of monitoring and descriptive data.
The STOrage and RETrieval data base (STORET) was used as the primary source
of monitoring data. Although quite extensive, data were not available for
many of the sites randomly selected for analysis. Other descriptive data,
such as the type of outlet structure, watershed land use, and other
influences on water quality, were also not available for this study.
Additional monitoring data, descriptive data, and a larger random sample
of dams would probably extend the study's findings.
Impoundments are created for a variety of purposes that provide
important social, economic, and aesthetic benefits. Most projects serve
multiple purposes, and it is important to recognize impoundment benefits
and purposes in evaluating their water quality consequences. Among the
project purposes recognized in this report are: hydropower generation
including pump storage, navigation, flood control, water supply,
conservation, recreation, fish and wildlife maintenance, water quality
enhancement, and low flow augmentation. Operating impoundments to
achieve multiple purposes is often complicated by conflicting
VII-1
-------
requirements for water flow and quality. Water quality within the
reservoir is dependent upon watershed land use, point sources, project
design, depth, season, and climate. Water quality in the tailwater
depends on the depth of water withdrawal, project design, configuration
of the tailwater channel, and local atmospheric conditions.
During the past 20 years, there has been a growing awareness of the
importance of water quality for water resources development and
management. This has resulted in major changes in the policies and
practices of Federal agencies, state and local agencies, private water
developers, and the related professions. Active research programs on
water quality have been initiated and carried out and coordinated. As a
result, the planning, design, and operations of dams show an enhanced
consideration of water quality. This trend should be encouraged to
continue.
WATER QUALITY EFFECTS OF IMPOUNDMENTS
Impoundment of free-flowing water by dams may potentially create
several effects, both positive and negative, on water quality within the
pool and downstream. Although this report focuses on unwanted effects,
desirable changes, such as a reduced sediment load, may also result. The
potential effects are often interdependent. Altering one condition in an
impoundment may create a ripple of effects throughout the reservoir-
stream ecosystem.
Impoundments can modify the physical, chemical, and biological
characteristics of the free-flowing aquatic ecosystem. Physical and
chemical characteristics in impoundments are also related to depth,
volume, climate, watershed land use, geographic location, reservoir
siting, and the schedule of water releases. Biological characteristics
are related to the type of habitat. The magnitude of effect of the dam
on water quality of releases appears related to the type of reservoir and
to the design and operation of the impoundment.
Effects can generally be characterized in three categories:
stratification-related, eutrophication, and other changes. Thermal
stratification of reservoirs results in warm waters of the epilimnion
(surface waters) overlying cooler and therefore denser waters of the
hypolimnion (bottom waters). Deeper impoundments with poor mixing
characteristics tend to stratify. Compared to waters upstream of the
impoundment, the waters of the epilimnion may tend to have slightly
higher temperatures and somewhat lower nutrient concentrations. Waters
of the hypolimnion, on the other hand, tend to have much lower
temperatures and lower dissolved oxygen levels.
When low concentrations of dissolved oxygen (e.g., anoxic
conditions) occur in the hypolimnion of reservoirs, this can result in
the formation of reduced forms of iron, manganese, sulfur, and nitrogen.
The reduced forms of these compounds can adversely affect water quality
and may be detrimental to aquatic life. These compounds are converted to
VII-2
-------
more assimilable compounds in oxidizing environments, such as the
epilimnion and tailwaters with adequate dissolved oxygen concentrations.
Well-mixed, unstratified reservoirs seldom experience problems with iron,
manganese, sulfur, or nitrogen compounds.
Eutrophication is a naturally-occurring process involving increased
growth and death rates of aquatic plants as well as sediment
accumulation, and is typically associated with reservoir aging. Excess
nutrients (especially nitrogen and phosphorous) in an impoundment can
increase eutrophication to undesirable levels. The settling and decay of
excess aquatic vegetation can deplete dissolved oxygen levels in the
hypolimnion, leading to anoxic conditions.
Water quality conditions in impoundment tailwaters are determined by
water quality in the reservoir and design and operation of the project
outlet works. For stratified impoundments, a primary determinant of
downstream water quality is whether a dam's outlet releases waters from
the epilimnion or the hypolimnion. With respect to temperature, release
of cool hypolimnetic waters, if done consistently all summer, can have a
desirable effect on downstream fisheries in many parts of the country.
However, low dissolved oxygen concentrations in the released hypolimnetic
waters may limit its ability to support some aquatic life.
Thus, the effects of stratification, possibly compounded by
eutrophication, are passed downstream via the discharge conduits. If the
outlets are at a low level, colder, possibly low dissolved oxygen, waters
are released. If the outlets are at a surface spillway, or downpipe,
warmer waters are discharged and the dissolved oxygen levels tend to be
higher because of turbulence and splash and also because of possible
algal photosynthesis in the summer. Power dams frequently have low
outlets. Large nonpower and small dams are less likely to have such
outlets because project purposes do not require their use. If a low-
level outlet exists as part of a large nonpower dam, reaeration typically
takes place in the process of energy dissipation associated with
reservoir releases.
Other water quality effects are generally considered to be less
predominant than eutrophication and stratification-enhanced effects.
Supersaturation of gases can occur as the result of rapid pressure
changes in spillway discharges plunging into deep stilling basins, and
may cause fish to suffer from gas bubble disease. Reservoirs, by virtue
of surface evaporation, and in some cases the acceptance of return flows,
may experience elevated salinity concentrations. The capture of sediment
behind an impoundment and changes in erosion patterns downstream may also
have an effect on water quality. Dams and their operations usually alter
the "natural" flow patterns. This effect may be desirable or
undesirable. Reaeration denial, where it occurs, deprives downstream
waters of dissolved oxygen which would have been generated in the absence
of the impoundment.
VII-3
-------
Reservoirs with short hydraulic residence times have reduced impacts
on tailwaters because the water is discharged before the effects of
impoundment are well established. Reservoirs with long hydraulic
residence times act as settling basins, removing suspended material from
inflowing waters. Pollutants and nutrients adsorbed to the sediments are
also removed, settling to the bottom of the reservoir.
RESULTS OF EPA ANALYSES
Four analyses were conducted in an attempt to estimate the national
scope of the water quality effects. Information on small and large
nonpower impoundments was limited, and no quantitative conclusions could
be reached through the analyses regarding the effects of these two
categories of dams except for the mixing analysis for large nonpower
dams. The investigative information that is available pertaining to
large nonpower and small dams is presented in the appendices. The
results of the four analyses - mixing, tailwater dissolved oxygen,
upstream/downstream comparison, and phosphorus enrichment - are presented
in the following paragraphs.
The entire study population of 68,155 impoundments were analyzed for
their mixing potential. Froude numbers were used as an index to
determine an impoundment's tendency to thermally stratify. Impoundments
that are strongly mixed are unlikely to have adverse water quality
effects on tailwaters, whereas unmixed or stratified impoundments may be
vulnerable to adverse water quality effects, which may be transferred
downstream by low-level outlet structures. It is estimated that 40
percent of the large (over 10,000 acre-feet) power impoundments are
potentially stratified as shown in Figure VII-1. For large nonpower
impoundments, 37 percent are potentially stratified. Of these
potentially stratified impoundments, some will experience water quality
effects and some will not.
The second analysis was the comparison of dissolved oxygen levels in
tailwaters below power impoundments. Dissolved oxygen levels were
compared with a criterion of 5 mg/1 for winter and summer data and
regional data. Data were not statistically representative so that it was
not possible to estimate dissolved oxygen levels on a national basis, but
for the sample as shown in Figure VII-2, two relationships were
established:
Dissolved oxygen in power dam tailwaters during the summer have
a much greater probability of not meeting a criterion of 5 mg/1
than during winter.
Larger power generating facilities show a greater probability
of not meeting a dissolved oxygen criterion than do smaller
power generating facilities.
However, these are general relationships and should not be applied
directly to individual impoundments without recognition of site-specific
conditions.
VII-4
-------
37%
40%
19%
44%
LARGE - NONPOYER DAMS
(n =1701)
51%
LARGE - POWER DAMS
(A = 424)
LEGEND: Potentially Stratified
Yearly Stratified or Mixed
Missing Date
a - Population
Figure VII-1
Percentages of Mixing Tendency
SUMMEI/DAMS>3tMW WINTE1/DAMS>3«MW SUMMEI/DAMS<30MW WINTEl/DAMS<30m
k\\\\XN OAK IIDCE NAT'L LAB
THIS STUDY
Figure VII-2
Probabilities of Non-coatpliance (PNC)
with 5 mg/1 Dissolved Oxygen (DO)
VII-5
-------
The third analysis was an attempted comparison of upstream and
downstream concentrations of temperature, dissolved oxygen, iron,
manganese, phosphorous, and nitrogen. The analysis focused on a random
sample of 40 impoundments from the 424 large power impoundments. Because
only half of the sample had data on temperature, dissolved oxygen,
phosphorous, and TKN, findings are limited to wide ranges applicable to
the sample as follows:
Between 28 percent and 55 percent of large power impoundments
are likely to increase downstream temperatures and between
13 percent and 40 percent are likely to decrease annual average
downstream temperature.
Between 23 percent and 50 percent are likely to cause decreased
dissolved oxygen levels in the tailwaters. Between 15 percent
and 42 percent are likely to cause increased dissolved oxygen
levels in tailwaters.
Data were also collected for large nonpower and small dams, but
success was limited to a minority of sites. The results of
these efforts are summarized in the appendices.
The fourth analysis estimated phosphorous enrichment within
impoundment pools, which is a potential indicator of eutrophication.
This indicator is limited by a lack of information on light availability,
hydraulic retention time, and other site-specific characteristics.
Phosphorous levels were—both observed and estimated—compared against a
guidance value of 0.025 mg/1 as a means of describing high and low
potential for enrichment. This analysis was conducted with a random
sample of 40 large power dams (population of 424). The sample of large
power dams showed 58 to 78 percent with phosphorous levels above 0.025
mg/1. Data were insufficient for similar analyses of large nonpower and
small impoundments. A summary of some of the more important results
appears in Figure VII-3.
AGENCY ASSESSMENTS
The U.S. Army Corps of Engineers, the Tennessee Valley Authority,
and the U.S. Bureau of Reclamation were invited to submit assessments of
water quality conditions in their own impoundments. Dams from these
three agencies represent a wide range of geography, climate, and
operational situations. The agencies surveyed their existing
impoundments using a questionnaire developed by COE, asking field
personnel what the dam-related water quality problems are and what
impacts are being felt. The resulting data, although subjective and
incomplete, gives a picture of the extent of known water quality problems
along with an assessment of these problems on user benefits. Agency
assessments based on visual observations indicate that physical water
quality conditions, such as pool level fluctuations and high and low
flows, may more frequently affect user benefits than poor chemical water
quality conditions. For the TVA system, chemical concerns are usually
more serious, even though physical parameters are more prevalent
VII-6
-------
M
O'
•0
M
\o
H
9*
e.
« O
5 -
* i ,
i 1l-
O « u
M
~O
M
to
O
- 2
M
H
O
O
VII-7
-------
concerns. It is possible that data were primarily collected at projects
where problems are known or suspected, thereby introducing a pessimistic
bias.
The COE operates more than 700 impoundments and has collected
questionnaires on 470. A sample of 46 (10 percent) was selected based on
project type and COE District. Data were available for approximately 50
to 60 percent of the projects sampled. Approximately 35 to 40 percent of
sampled projects with data had problematic conditions in the tailwaters.
Most prevalent were flow fluctuations and high and low flow issues, low
dissolved oxygen, and erosion and transport of sediment. Problematic
conditions in impoundment pools were identified in 40 to 50 percent of
the sampled projects with data. Most prevalent were eutrophication
problems (high nutrients, low dissolved oxygen, algal blooms, and
macrophytes). Additional conditions of concern in pools were excessive
concentrations of reduced iron and manganese in bottom waters and
accumulation of sediment and contaminants.
TVA assessed all 33 of their large projects, including all of their
hydropower projects and four nonhydro projects. Small impoundments were
not assessed. Data were available for dissolved oxygen, temperature,
flow, pool levels, and macrophytes at all projects. Data for the other
parameters were available for an average of 60 percent of the projects.
Water uses were severely impacted at several sites. In reservoir pools,
the most significant impacts were pool level fluctuations (50 percent of
the projects with data), bacteria (30 percent), and turbidity, algae,
macrophytes, sediment accretion, and shore erosion (all at 15 to
20 percent). Tailwaters primarily experienced significant problems with
low dissolved oxygen, flow (high, low, and fluctuating), and low
temperature, occurring at 15 to 20 percent of the projects with data.
USBR maintains 349 impoundments in the 17 western states; of these,
250 (72 percent) responded to the questionnaires. In approximately 54
percent of the reservoir pools and 59 percent of the tailwater cases, no
data were available. The main impact-producing conditions identified at
USBR reservoirs are drawdown and pool fluctuation, which are associated
with a mode of operation that combines rapid spring filling of reservoirs
with a steady withdrawal of water to satisfy irrigation, municipal, and
industrial demands during the long dry season. These two conditions were
rated as having at least significant impacts on user benefits in 36
percent and 35 percent, respectively, of the reservoirs with available
data. High flow was the main impact-producing condition noted in USBR
tailwaters, probably reflecting the high spring inflows and spillway
discharges of the mid-1980's. This condition was rated as having a
significant impact on user benefits in 21 percent of the rated
tailwaters. The next two significant impact-producing conditions cited
were low flow and taste and odor problems, each with a cumulative rating
of at least significant in about 13 percent of the tailwaters with
available data.
VII-8
-------
SUMMARY OF EPA ANALYSES AND AGENCY ASSESSMENTS
The results of the four EPA analyses conducted for this study
(mixing, tail water dissolved oxygen, upstream/downstream comparison of
parameters, and phosphorous enrichment) cannot be directly related to the
findings of the Agency assessments due to differences in the analytical
methods and parameters. However, a few simple comparisons can be made
which illustrate the likely range in which certain dam water quality
conditions occur. The study assessment of large power impoundments found
that low dissolved oxygen in tailwaters occurs more frequently in summer
than winter and at large power impoundments over small power impoundments.
The COE reported that approximately 33 percent of the projects assessed had
dissolved oxygen concentrations below saturation during the summer months,
while TVA reported seasonal low dissolved oxygen in 70 percent of their
reservoir hypolimnia or bottom waters. USBR experienced at least
intermittent low dissolved oxygen levels in only 9 percent of their
reservoirs.
The upstream/downstream comparison found low dissolved oxygen in 23
to 50 percent of large power impoundment tailwaters. Assessments by the
COC and TVA reported similar results, although their data were not
restricted to large power impoundments. COE reported 20 percent of their
project tailwaters experienced at least intermittent problems with low
dissolved oxygen, while TVA reported 38 percent with at least intermittent
problems. USBR reported only 4 percent of their projects as experiencing
at least intermittent problems with low dissolved oxygen in tailwaters.
Both the data analyzed by this study and the information provided by the
Agencies indicate that low dissolved oxygen is influenced, in part, by
climate and is therefore more likely to occur in eastern dams, and
particularly in southeastern dams.
The phosphorous analysis estimated phosphorous enrichment within
impoundment pools of large power dams as a potential indicator of
eutrophication. The analysis showed 58 to 78 percent of the large power
dam pools had phosphorous above a guidance value of 0.025 mg/1. The Agency
assessments subjectively estimated the frequency of occurrence of high
nutrient loads (presumably a combination of nitrogen and phosphorous) in
pools. Agency impoundments were not restricted to large power dams. High
nutrient loads were reported in 35 percent of the COE's pools, 30 percent
of TVA's pools, and 15 percent of USSR's pools.
VII-9
-------
APPENDIX A
REFERENCES
Allen, H.E. and J.R. Kramer, eds., Nutrients in Natural Waters, John
Wiley and Sons, New York, NY, 1972.
Barnett, R.H., "Case Study of Reaeration of Casitas Reservoir,"
Proceedings of the ASCE Symposium on Reaeration Research, Gatlinburg, TN,
1976.
* Bouck, G.R., "Air Supersaturation in Surface Water: A Continuing
Engineering and Biological Problem", Proceedings of the Symposium on
Surface Water Impoundments ASCE, Minneapolis, MN, June 2-5, 1980.
Bowker, A.H. and G.J. Lieberman, Engineering Statistics, Prentice-Hall,
Inc., Englewood Cliffs, NJ, 1972.
* Cada, G.F., K.D. Kumar, J.A. Solomon, and S.G. Hildebrand, "An Analysis
of Dissolved Oxygen Concentrations in Tailwaters of Hydroelectric Dams
and the Implications of Small-Scale Hydropower Development", Water
Resources Research, Volume 19, No. 4, pp. 1043-1048, 1983.
Carcich, I.G., and T. J. Tofflemire, "Distribution and Concentration of
PCB in the Hudson River and Associated Management Problems," Environment
International. Vol. VII, 1982.
Cassel, C.M., C.E. S rndal, and J.H. Wretman, Foundations of Interence in
Survey Samplings, John Wiley and Sons, Inc., New York, NY, 1977.
Clark, J.W., W. Viessman, M.J. Hammer, Water Supply and Pollution
Control. 3rd edition, Harper and Row, New York, NY, 1977.
* Dompier, D. and J.R. Woodworth, "Rehabilitation of Salmonid Fish Streams
Through Upstream Storage", Proceedings of Wild Trout II, Yellowstone
National Park, September 24-25, 1979, Trout Unlimited and Federation of
Fly Fishermen, 1980.
Electric Power Research Institute, Robustness of the ANOVA Model in
Environmental Monitoring Applications. Technical Report EPRI EA-4015,
Palo Alto, CA, 1985.
* Fast, A.W., "Artificial Aeration as a Lake Restoration Technique", Lake
Restoration, Proceedings of a National Conference, August 22-24, 1978,
U.S. EPA, Minneapolis, MN, 1979.
Federal Reporter 693, Second Series, "National Wildlife Federation, et
al. v. Anne Gorsuch, Administrator, Environmental Protection Agency, et
al.", pp. 156-183.
A-l
-------
Federal Energy Regulatory Commission (FERC), Hydro Resources Assessment
(approx. 15,000 entries), personal correspondence, 1987.
Golze, A.R., Handbook of Dam Engineering, Van Nostrand Reinhold Company,
New York, NY, 1977.
Higgins, J.M., W.L. Poppe, and M.L. Iwanski, "Eutrophication Analysis of
TVA Reservoirs", Proceedings of an ASCE Symposium on Surface Water
Impoundments, H.G. Stefon, ed., Minneapolis, MM, June 1980.
*Hoe1, P.G., Introduction to Mathematical Statistics, John Wiley and Sons,
Inc., New York, NY, 1951.
Johnson, P.L., "The Influence of Air Flow Rate on Line Diffuser
Efficiency and Impoundment Impact," Proceedings of the Symposium on
Surface Water Impoundments, Minneapolis, MN, 1980.
*Keeley, J.W., et al., Reservoirs and Waterways: Identification and
Assessment of Environmental Problems and Research Program Development,
Technical Report E-78-1, U.S. Army Corps of Engineer Waterways Experiment
Station, Vicksburg, MS, 1978.
*Kennedy, R.H., R.F. Gaugush, "Assessment of Water Quality Enhancement
Needs for Corps of Engineers Reservoirs," Presented at Seventh Annual
International Symposium, North American Lake Management Society, Orlando,
FL, November 1987.
*Kendall, M.G. and A. Stuart, The Advanced Theory of Statistics, Volume 1,
"Distribution Theory," Hafner Publishing Company, New York, NY, 1963.
*Kerr, K.M., "Effects of Bottom Withdrawal on Thermal and Chemical
Stratification of a Small Artificial Reservoir," Thesis submitted in
partial fulfillment of the requirements for the degree of Master of
Science, University of Arkansas, Fayetteville, AR, 1977.
*Larson, D.P. and H.T. Mercier, "Phosphorus Retention Capacity of Lakes,"
J. Fish. Res. Board Canada, 33:1742-1750, 1976.
Miller, J.B., "Intermountain West Reservoir Limnology and Management
Options," Proceedings of the Third Annual Conference, North American Lake
Management Society. Knoxville, TN, 1983.
*Moore, J.W., "Water Quality Investigation of a Small Artificial
Reservoir," Prepared for Department of Commerce, State of Arkansas,
Little Rock, AR, 1973.
National Academy of Sciences, "Eutrophication - Causes, Consequences,
Corrections", Proceedings of a Symposium, National Academy of Sciences,
Washington, D.C., 1969.
National Wildlife Federation, NWF V. Consumers Power Co., No. G85-1146,
Kalamazoo, MI, 1987.
A-2
-------
National Wildlife Federation, Appellee's Connected Brief - August 1982.
Peters, J.C., "Modification of Intakes at Flaming Gorge Dam, Utah, to
Improve Water Temperature in the Green River," Proceedings of an
International Symposium, Knoxville, TN, 1978.
* Peterson, S.A., "Dredging and Lake Restoration", Lake Restoration,
Proceeding of a National Conference, August 22-24, 1978, U.S. EPA,
Minneapolis, MN, 1979.
*Pfitzer, D., "Tailwater Trout Fisheries with Special Reference to the
Southeastern States", Proceedings of the Wild Trout Management Symposium,
Yellowstone National Park, September 25-26, 1974, Trout Unlimited, 1975.
Reid, G.K., Ecology of Inland Waters and Estuaries, D. Van Nostrand Co.,
New York, NY, 1961.
Ruttner, Franz (translated by D.G. Frey and F.E. Frey), Fundamentals of
Limnology, third edition, University of Toronto Press, 1970.
* SAS Institute Inc., SAS User's Guide: Basics, 1982 Edition, SAS
Institute Inc., Gary, NC, 1982.
Sawyer, C.N., Chemistry for Sanitary Engineers, McGraw-Hill Book Co.,
Inc., New York, NY, 1960.
*Schweibert, E., Trout, E.P. Dutton, Inc, New York, NY, 1984.
* Sobotka and Company, Inc., "Regulatory Approaches to Productivity
Management in Lakes and Reservoirs," prepared for U.S. Environmental
Protection Agency, Office of Regulatons and Standards, Washington, D.C.,
1986.
*Soil Conservation Service, Highlights of Studies Which Assess the
Influence of Flood Control Impoundments on Water Quality, USDA-SCS,
Auburn, Alabama, addendum 1987.
* Soil Conservation Service, Water Quality Effects of Impoundments, South
Technical Service Center, Series No. 802, USDA-SCS, Forth Worth, TX,
1979.
* Soil Conservation Service, An Evaluation of Water Quality and Related
Biological Parameters of Four Reservoirs and Their Inflowing and
Receiving Perennial Streams, Contract No. AG28565-00876, USDA-SCS,
Jackson, MS, 1978.
* Streeter, H.W., and E.B. Phelps, A Study of the Pollution and Natural
Purification of the Ohio River, Public Health Bulletin No. 146, U.S.
Public Health Service, Washington, D.C., reprinted 1958.
A-3
-------
Symons, J., S.R. Weibel, and G.G. Robeck, Influence of Impoundments on
Water Quality - A Review of Literature and Statement of Research Needs.,
US Department of Health, Education and Welfare. PHS Publication 999-WP-
18, R.A. Taft Engineering Center, Cincinnati, OH, 1964.
Taubert, B.D., "Changes in the Trout Fisheries of the Lower Colorado
River and Arizona", Wild Trout III, Proceedings of the Symposium,
Yellowstone National Park, September 24-25, 1984, Trout Unlimited, 1984.
* Tennessee Valley Authority, Improving Reservoir Releases, Report
TVA/ONRED/AWR-87/33, Knoxville, TN, 1987a.
* Tennessee Valley Authority, Case Studies (Draft), 1987b.
* Tennessee Valley Authority, Trophic Status Evaluation of TVA Reservoirs.
Report TVA/ONR/WR-83/7, Chattanooga, TN, 1983.
* Tennessee Valley Authority, Impact of Reservoir Releases on Downstream
Water Quality and Uses, Chattanooga, TN, 1978.
Thomas, H.H., The Engineering of Large Dams, John Wiley and Sons, Inc.,
New York, NY, 1976.
* U.S. Army Corps of Engineers, Case Studies, 1987.
* U.S. Army Corps of Engineers, Techniques for Reaeration of Hydropower
Releases, Technical Report E-83-5, Washington, D.C., 1983.
* U.S. Army Corps of Engineers, "National Program of Inspection of Non-
Federal Dams, Final Report to Congress," USAGE, HQUSACE, Washington,
D.C., 1982a.
* U.S. Army Corps of Engineers, National Inventory of Dams Data Base in
Card Format, Accession No. ADA118670, (9 track tape), NTIS, Springfield,
Virginia, 1982b.
U.S. Congress, 92nd Congress, "Federal Water Pollution Control Act
Amendments of 1972," Public Law 92-500, Washington D.C., 1972.
U.S. Department of Agriculture, SCS, Engineering Division, Dam Inventory
System User's Guide, Fort Collins, CO, 1984.
U.S. Department of Energy, Federal Energy Regulatory Commission, PURPA
Benefits at New Dams and Diversions, Docket No. EL87-9, Washington, D.C.,
1987.
*U.S. Department of Health* Education and Welfare, Symposium on Streamflow
-Regulation for Quality Control, April 3-5, 1963, Public Health Service
Publication 999-WP-30, 1965.
A-4
-------
U.S. Department of Health, Education and Welfare, Water Quality Behavior
in Reservoirs, Public Health Service Publication 1930, Washington, D.C.
U.S. Department of the Interior, Bureau of Reclamation, Limnological
Effects of Artificial Aeration at Lake Cachuma, California. 1980-1984,
Technical Report REC-ERC-87-10, Denver, CO, 1987a.
*U.S. Department of the Interior, Bureau of Reclamation, Case Studies,
1987b.
U.S. Department of the Interior, Bureau of Reclamation, Statistical
' Compilation of Engineering Features on Bureau of Reclamation Projects,
Denver, CO, 1984.
U.S. Department of the Interior, A Mathematical Model for Predicting
River Temperatures - Application to the Green River Below Flaming Gorge
Dam, Technical Report REL-ERC-76-7, Denver, CO, 1979.
U.S. Department of the Interior, Bureau of Reclamation, Design of Small
Dams, Washington, D.C., 1977.
* U.S. EPA, Quality Criteria for Water 1986, Technical Report EPA 440/5-
86-001, Office of Water Regulations and Standards, Washington, D.C.,
1986.
* U.S. EPA, Rates, Constants, and Kinetics Formulations in Surface Water
Quality Modeling, second edition, Report EPA 600/3-85/040, Office of
Research and Development, Athens, GA, 1985.
*U.S. EPA, Water Quality Control Information System (STORET), User
Handbook, Volumes I and II, Office of Information Resource Management,
Washington, D.C., 1982a.
U.S. EPA, Appellees' Corrected Brief - National Wildlife Federation, et
al. v. Anne M. Gorsuch, Administrator, Environmental Protection Agency,
August, 1982b.
U.S. EPA, Reply Brief for the Federal Appellant - National Wildlife
Federation, et al. v. Anne M. Gorsuch, Administrator, Environmental
Protection Agency, et a!., (7 counts - suit for permitting impoundment
releases as point sources), July, 1982c.
U.S. EPA, Brief for the Federal Appellant - NWF v. EPA, National Wildlife
Federation, et al. v. Anne M. Gorsuch, Administrator, Environmental
Protection Agency, et al., (7 counts - suit for permitting impoundment
releases as point sources), June, 1981.
*U.S. EPA, (Authorship: Kenneth H. Reckhow, Michigan State University),
Quantitative Techniques for the Assessment of Lake Quality, Technical
Report EPA-440/5-79-015, Washington, D.C., 1979a.
A-5
-------
* U.S. EPA, Lake Restoration, Proceedings of a National Conference, August
22-24, 1978, Office of Water Planning and Standards, Washington, D.C.,
1979b.
U.S. EPA, Quantitative Techniques for the Assessment of Lake Quality,
Technical Report 440/5-79-015, Washington, D.C., 1979c.
* U.S. EPA, (Project Officer: Norman A. Whalen), Non-point Source Control
Guidance Hydrologic Modifications, Washington, D.C., 1977.
U.S. EPA, Impact of Hydrologic Modifications on Water Quality,
Washington, D.C., 1975.
U.S. EPA, Chapter VI, "National Eutrophication Survey," National Water
Quality Inventory, 305(b) Report to Congress, 1975b.
U.S. EPA, Methods for Chemical Analysis of Water and Wastes,
Environmental Monitoring and Support Laboratory, Cincinnati, OH, 1974.
* U.S. EPA, The Control of Pollution Caused by Hydrographic Modifications,
Washington, D.C., 1973.
* U.S. EPA, (Authorship: Water Resources Engineers, Inc.), Mathematical
Models for the Prediction of Thermal Energy Changes in Impoundments,
Technical Report 16/30-EXT-12/69, Washington, D.C., 1969.
* U.S. Fish and Wildlife Service, A Guide to Stream Habitat Analysis Using
the Instream Flow Incremental Methodology, Instream Flow Information
Paper No. 12, Report FWS/OBS-82/26, Washington, D.C., 1982.
U.S. House of Representatives, 99th Congress, 2nd Session, Amending the
Clean Water Act, Report 99-1004, Washington D.C., 1986a.
U.S. House of Representatives, 99th Congress, 2nd Session, Water
Resources Development Act of 1986, Report 99-1013, Washington D.C.,
1986b.
* Vollenweider, R.A. and P.J. Dillon, The Application of the Phosphorus
Loading Concept to Eutrophication Research, Report 7, NRCC 13690, Ottawa,
Ontario, 1974.
Vollenweider, R.A., "Input-Output Models (With Special Reference to the
Phosphorus Loading Concept in Limnology," Conference on Chemical-
Ecological Considerations for Defining the Goals of Water Pollution
Control, Kastanienbaum, Switzerland, April 19-21, 1972.
* Walker, W.W., "Empirical Methods for Predicting Eutrophication in
Impoundments, Report 1, Phase I: Data Base Development," Technical
Report E-81-9, U.S. Army Engineer Waterways Experiment Station,
Vicksburg, MS, 1981.
A-6
-------
Walker, W.W., "Empirical Methods for Predicting Eutrophication in
Impoundments, Report 2, Phase I: Model Testing," Technical Report E-81-
9, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS, 1982.
Walker, W.W., "Empirical Methods for Predicting Eutrophication in
Impoundments, Report 3, Phase II: Model Refinements," Technical Report
E-81-9, U.S. Army Engineer Waterways Experiment Station, Vicksburg, MS,
1985.
Ward, J.C., S. Karaki, Evaluation of Effect of Impoundment on Water
Quality in Cheney Reservoir, Research Report 25, Bureau of Reclamation,
Fort Collins, CO, 1971.
Water Quality Division Committee on Nutrients in Water, "Chemistry of
Nitrogen and Phosphorus in Water", J. American Water Works Association,
Volume 62, No. 2, pp. 127-140, 1970.
Water Resources Engineers, Inc., Prediction of Thermal Energy
Distribution in Streams and Reservoirs, prepared for the Department of
Fish and Game, State of California, Water Resources Engineers, Walnut
Creek, CA, 1967.
* Wiley, Robert W. and James W. Mullen, "Philosophy and Management of the
Fontenelle - Green River Tailwater Trout Fisheries", Proceedings of the
Wild Trout Management Symposium, Yellowstone National Park, September 25-
26, 1974. Trout Unlimited, 1975.
* All references with an asterisk have been cited within the body of the
report and appendices.
A-7
-------
APPENDIX B
SAMPLE/CASE STUDY DATA BASE
This section contains a compilation of the various data reflected in
the 120 sample and 15 case study facilities. The 135 dams have been
broken into four tables. Each table reflects a category of facility;
large power dams, large nonpower dams, small dams, and case study dams.
Every page of the tables contains the category (i.e., random sample or
case study), the COE National Inventory of Dam identification number, and
the name of the facility. Although these tables include a significant
amount of the data and statistics used in the report, the individual
STORET monitoring data are excluded.
The sample tables consist of three pages for each partition. The
first page contains physical information taken from the COE National
Inventory: the location, power rating of the hydroelectric installation
(if any), normal volume, hydraulic dam height, and maximum spillway
capacity. Morphologic data, including estimates based on the regression
analysis discussed in the body of the report, are also included. These
data include area, annual inflow, and length of the reservoir. Note that
where data were obtained from other sources, (i.e., EPA IMS flow file,
agencies, or by direct measurement from USGS topographical maps), they
replace the calculated estimates. This is done throughout the tables.
Finally, the first page of the tables provides the calculated Froude and
Phosphorus Retention Coefficient values.
The second page of the sample tables contains information relevant to
the phosphorus and eutrophication analyses discussed in the body of the
report. Data used in the application of the Vollenweider model are
checked. The upstream phosphorus, inpool phosphorus, and downstream
phosphorus data are averages based on the tabulated sample size. The in-
pool phosphorus estimate is based on the actual data, a Vollenweider
estimate, or usage of the downstream level as the best estimate. Also
shown on this page of the table is the probability of noncompliance
statistics for summer and winter for the appropriate stations.
The third and final page of the sample tables consists of information
on where, in relationship to the facilities, data for 15 water quality
parameters were found. (The Case Studies category does not have this
information.) The data were categorized by being upstream (U), in-pool
(P), or downstream (D). The table reflects that data may have been
sampled at all three locations, for example, 6005 at the Beaver Dam
(AR00174), or found at none, such as TKN at the Cooper Dam (AK00001).
Notice that the location of the phosphorus information on the second page
of the tables are reflected under the column PHOS on page three.
The case study table consists of one page and combines information
similar to that on the first page of the sample tables with the
noncompliance data from the second page described above.
B-l
-------
8 •
•3
vz
235
U M
?
ii
H
^S
8 § 8
88
Srt e
in f
O O
~f ^
O 0s
O ^ o -^ co -* i
«-i «M -? O J1 r"
fH CO O vO •*> -^ I
o m i-« c*j i-<
rH i—I fM
i 88
j 1-4 «J I-H
Sr*.
eo
CM ^-« r»J
§QQ@?°S 2 5 °{
OOO9OC O O O <
rt O f* M •* -9 !
« jj f^ rt f
OO
OO
OQ
OO
OOO
OOO
588
• cOf^^C-u-i^pn^H^OJ-'
5^ 00^»HfMCim.
O M -H .
) O O O O C
) —. —t o O <
o i-< o o **i e
B-2
-------
O CO M CO CO
°
d
8 88S , 88388
o odd ddddd
8
d
O O 8S O
o o o o o
•* ""a "*$ *
N 888 m ^?58lS
d odd d ddddd
o
d
M f-t O Q n
mo OO O
dddd d
r-t -O 3 v
M C* -J P4 rH M 4
o o o o o o o
dddd d dddd ddddd
O O O W O rH O
o o o o ooo
s
o
d
o
d
iHir)tHOO-*O^^OC^-t
Or^iHOrlOOOrH
OOOOOOOOO
O O O O
o o o o
Ul Qi O li
o o o c
OOOOOOOOO
) vO -* *O PJ
) r-4O O O
o o oo o
o
l-t
o
d
oo o r- «
O O O <
d d ooo o
O rHO
d do*
o> *o ^
O O rH
do d
O O^ I/I ^ O
ooo So
ooo o o
o
d
in vo o
l/l Jl «H rH
O O 00
o o o o
•4 -t P4
o -»o
o o o
OOOOOOOOO
OOOOOOOOO
B-3
-------
1
B
i
£
E
N02N03
|
z
g
P
2
CO
§""
o c
O.QQ 0.
1 3 1 BB33 1
Q OOO
till
o. a. o« o.
13133331
o oaoo
00.0,0.0.0.
i i i i i i
0.333 = 3 = 0
O. O.OO. O.
1 1 i 1 t
O. 1 3 30.3 3 1
OO O
0. 0.0.0. 0.
0.333333 1
O.
0.1 1 33 1 3 I
0
O. O.
I < 1 B 3 I 3 1
O.
O. 1 3 I i I3O
Q
3 1
i
O.
O
0.
3 >
O
1
o.
3 1
O
O
3 i
0
3 G*
O
C O
1 3 Q 33
O
O.Qa.e.
13333
O OO
O. O 0, O.
a
00.0
333
0. O
330.
00
i i
QO.O.
C
1 3
O
1 3
O
O 0.
i i
3 3
CO OO O
0.00.0.00.0.00.
0333333333
0 00
o. o o. o.
13333
O O 0.
1=31 =
O
O OB.
i 3333
O Q 0.
0 Q
OO.OOO.
B 3 O.3B
O
I
O.O
1 =0.
Q o
Q 0.0.
333
O
3 I OS 3 1 3336.
B 1
1 Q
O
1 B
O
O O. O
1 3 B 0.3 3
o
O Q OO.O
1 330.33
D QO
OO.Oa.0.
1 3 BO.B 3
O O
0. OO.Q
1 1 3 0.33
O O
O O. O O. O
1 3 3 0.3 3
Q O O
1 3 I 0.33
O O
O QO.O.
1 3 B 0.3 3
Q O
0, O.
i 1 30.3 B
O
1 1 1 1 B B 3
OOO.
1 i 1 1333
0.0. Q 00.
> 33 I 333
O
O O
( 1
0.0.
0 3
O Q
O. O.
33
O OO
O.O.QaOB.a.0.
1 1 i t i i i i
1 33333333
B. O OOOB.
1 30.3333
0
I I i 1313
0. O O
1 3O. 1 333
O
1 1 1 1 3 OB
O O O
333
Q
3
Q
B
O
3
Q O
B. O. O
i i 1
333
O O
B 0.
0
O 0.
B
O
333
O
O 0.
O
333
O
1 3
O
O O.
3 3
1 i
o
1 B
O O
1 3 0.3
Q
> B
Q
1 3
1 i
1 0.
33
B 3
1 3
1 3
i
! t I
a o. o o
! 33 I 3 OB a\
a ao
oa oo. a0.0.
B33BB3B3
00 C
O OO.O. O. OO OO.O
300.33 > 30. I BB33B3O I
OO. O O O
0. IBIBBBCBIOBB I
O.
O O
I I
Q O O. O.
31 I 3 0.3 3 i
O
O O O O. O
I BB3333OO. I 3
O OO QQ O O OO DO O
0.00.0.00.0.00. 00.00.0,00.0. 000.0.0.000.0.0
O O OOO C
i i t i i i
00.0.0.0.0.^. __ _
3333330^100333333333 I 330.33333 I 3=333333 = 3
C O O
131 >33
O
o, a
3 3
O OOO
3113133
O
o._
I t
OOOB.OO.OOO =O O O. OO OOO
BBCB I CBBB333B3B > 3 = 0. I B I 33 I 33 = 333 I
O
3
o5-*?£
H 53 «-i r^ —
KE |3IS*«H
lg!2giBHSg-
2^3i|£<=i
(CtJQHHZWKO
^ *P90 ^.
O "^ *^ *v *
jmoo^popppoooe
SPPPPOPPOOMOC
5o>-fiM"-«MZOHC
; u. o o i-»»-» * :
>«-4OO*"*~«<*^«*iO
>PPPPOO — O
>uuuu2zz<
:ZOVIV>MHHS
8§§
B-4
-------
Sri
r*,r*cOO^COO>P -TOh- P
«-4PPOOOOOO«CC f* •-. CO CO
\n***G*O&r^r*r^ P CO r»- -^
oooooooo ooo
PPPPPPPPPPOP
OOoPPOPo f*i o P co mm^f-Hp^H.-HpooprMp
OOOOOOOP POP *-> OPPPPPPPPPPP
OPPPPP"-*-*
§*G t** P P CO O **J
r* r*. p p «-. '-*POpfocM-*cM»o-*fi*-pp-*i/.«»i
r-t CM r*. r-
\o fn p
r- f> -J-
m «c
op p p 4np*oinpppinpmm<
2§ i-j 2 W
_ _5i3u2g
-~t >H (ri BJ tri
< SBH B- <
3 3 3 » => U
-------
3 3
8 83 8 ,
o o o o
8 8
o o
o o
o o*
8
o
8. 83 8 5
o do o o
S 8
d d
88
d d
O iHO
d do
O
d
1S
(*> O
O
C)
NO
do*
O O
do
sO r^
-------
i
CM
i
I
i i i i i i t i a i
o
0. O
Q i a a
o
0.0
o a
a. a.
a a i
c
a. a.
i i i i i a. i
a. a.
i i
I a a i
a a c a
i iii
iaiaiiiiiii'i taiaaioia
a a a a
a. a. a, a o, a
it i i i i
ia>a i i i a i a. a i i i a
o c aaa
a. a o. a. a, a. a
a.3 i a. i i i i i a. a. i i i i i a a a i i a
co a Q a
&. a. a. o a. a. a
o, 3 3 3 i i < 10.0.0.1 • i 3 i o. a a » ia
o o
0.0, o. a. aa. o. a
0.3 33 i ' > 1310,1 , laicLaai ia
I I I I I O t a
a c
3 a &.
a
o, a
a
c. a
a ao.
a a a
a. o-Q a. a
i i=ii=>aaaia
a
a.
i a i
a
a i a
i " a o i a a c-
c c
0. Q C. C
o i a i j i &. c.
a
o. a
0.3 a o.
i i a a o. i i i i
a
o. o
a a
o, a. a a
a i aaa a i a
c
£.0. o
i i a a i Q i a
a aaa
L a. 0,0, a
a i a a a i 13
c. o a
a > a a i a i a
o: ^> CM o at
O §0 O S
> ii « > Z
pc ^ in >c o p£ <
UroO^ ^HU a U
WWO^N LO UW
!5£oge0H£ 1= I
a u <
O 00 r-1 c
—< *^ FN Z
< u
« « (K c u;:
u u u < i
eg pa m p u. j £
< < X X U
"il:
! a .
niSuj? E>»>«J2a
• —jitf-JpepeuiceuHtt
2uuo,u. u>-. .
B-7
-------
•-> W
4J H
& <
a> r
^O r«J -3 O •*• O f-»
CT* CO O O O ^O ^
•> O CO tn O CO ••
^ O* O CO O CO (
) O*' O4- 0> O r*- t
O O O O
-H £) -1 #
*H CO r- O
in r* pg e\]
O O -J O
OO--IOOO OOO-HOOO OOOOC3OO O
s in h.
) C1 ff-
> m o
O O >C O « -H
^e
CM P^ rt «H
7|
O H
t*-J r-J r-l
CO fl O J^ O
fl O J^
fO r^ CO
8O O CO
n -* in
-f J" rH
m m o ^O fJ f*J
O ^ CJ d ^ u*i
•-• --i 00 * *n -J
. rt o
) O O
1°
O^^Minu^iOO^O'^*D^^OvOrHpnin^-ij>r>JOOiAeMOOJ-insOf-^COf^ino«-'O(O«3'C*rs)O
«n fi J- (^1 .
1 H8~
H H X H
H
H VJ 0£
i o: a.
lie!
: w c>
•* ^ w
: 2^<
lid"
§
s B
:»i«l
oS-"'
6.^N!
e^s|
> N ua [
:^2w(
M m —• i
f^ m m i
•& <*^ »• '
j rJ rJ *o
> r- CO ON
- en & &•
H M u:
;§.
U < < cJ Z C
^3Q SSt
-3 3 CO U &
; e- &. 3 w -3 \o
I Crt <
o z a:
iiii
W W t/1 t
*: ^ &c :
sSdS
: H 3:
3-5
|^
la
< oc
i^gi
j —t W 0-
!SS
) c
•5r
• O u.t
'£ a
ibX.
5i!
> w o
iii
:u5
u o
. 3 5 Z
ilii
§OOi™"O — 'HO'^OO'^O'^C
000*^0000000 — M<
J2^s252£»lEii£z";>H*'^-*'
!SS2
;>^3
B-8
-------
8
d
5
o o o
o o o
o
o
&BI
B 1;
O*X tnr*4Mr*m4
gSNirti-(or^ eoo*c
SBr^nin^-i*) a>o»
JC U-»n-<^mj-»j
3iUHMMMUHu:
.:>$;
1S3 «!
• H J t-t «, C
sa:
33
3 M M t
la^se;
3
g
a
9
1 ^'
iS
;»
SB
O OT
iiipi;
' 9 Q M •
iiiiiii
4 r*. t» in •*it
I M C1 r4 >-* <
B-9
-------
i
rJ
i
U.
w
IIQ|IIIIO*I3PII I I I I I It It
Q
=J3 i i i
t i i i a
t-H *
i»>Si
JSS3K!
! a. c. a w .
: a a z
•'
:5
; a
3 w --
: a. H
5S§
OO'-*O'-'^-'O'~'OO'—O-HOO — -H<*JrOf^—"^10
OOOMOOOOOOO-^(^OOOOOOOOOO
'>-->->-|£*:^ai
-------
O Uj
OOCT*00\OO» *O r^ON f^-CO
J (-1CM ^ CM(M i-HcM
CO <-* f» 00 CO
§i-HO'-*O r-» J^T-t fMO
OOOO O OCM fH^H
(DC3OOO O OO OO
00 00 *£ CO ON
O O O O O
O O O O O
in \o
o o
CDC3OOC5OOOOOOOOOO
) O -4 O ^
i o 1-13 ^
§88 888
MOO O> O* O (
ro e»* tn iH •* tn »
OOO OOOOOOONOO
ONO OOOOOOTHI^Oin
CJf-fOOOOOOSOOOr-om
ONONtnin-^oocootMf~-incMONi-H
sssss:
) f-J O f- f» (
00 00 CM iH i-4
rHrHCOtn* iH* O O O O O CM CO
^ j r-t IH* o* m r« r-*. n
-------
APPENDIX C
ANALYSIS SUPPLEMENT
This appendix summarizes some of the analysis methods employed in the
technical analyses including a mixing model, a phosphorus retention
model, physical attribute estimators, and a discussion on the
significance of correlation, a discussion of sample size, a discussion of
how representative are the random samples with respect to the population,
and STORET water quality incidence summaries.
MIXING MODEL
The densimetric Froude number represents the ratio of inertial forces
imposed by the longitudinal flow to gravitational forces within an
impoundment and can be written for a reservoir as (USEPA, 1969):
where:
F =
L • q
g • 3
L = reservoir length,
q = average yearly discharge through the reservoir,
D = mean reservoir depth,
V = reservoir volume,
o = reference density,
3 = average density gradient in the reservoir, and
g = gravitational constant.
In deep reservoirs, the fact that the isotherms
indicates that the inertia of the longitudinal flow is
disturb the overall gravitational static equilibrium
reservoir, except possibly for local disturbances in the
are horizontal
insufficient to
state of the
vicinity of the
reservoir outlets and at points of tributary inflow. Thus, F would be
expected to be small for such reservoirs. On the other hand, in
completely mixed reservoirs, the inertia of the flow and its attendant
turbulence is sufficient to completely upset the gravitational structure
C-l
-------
and destratify the reservoir. For reservoirs of this class, F would be
expected to be large. In between these two extreme classes lies the
weakly-stratified reservoir in which the longitudinal flow possesses
enough inertia to disrupt the reservoir isotherms from their
gravitational static-equilibrium state configuration, but not enough to
completely mix the reservoir.
For the purpose of classifying reservoirs by their Froude number, (3
and dp may be approximated as 0.001 kg/nr and 1000 kg/m^, respectively.
Substituting these values and g leads to an expression for F as:
F = 320 •
L • q
V
where L and D have units of meters, q is in cfs, and V has units of nr.
From this .equation it is observed that the principal reservoir parameters
that determine a reservoir's classification are its length, depth, and
discharge to volume ratio (q/V).
In developing some familiarity with the magnitude of F for different
reservoir situations, it is helpful to note that theoretical and
experimental work in stratified flow indicates that flow separation
occurs in a stratified fluid when the Froude number is less than about
1/n (0.318). For example, for F < 1/n, part of the fluid will be in
motion longitudinally while the remainder is essentially at rest.
Furthermore, as f becomes smaller and smaller, the flowing layer becomes
more and more concentrated in the vertical direction. Thus, in a deep
reservoir, it is to be expected that the longitudinal flow is highly
concentrated at values of F « 1/n. While in the completely mixed case,
F must be at least greater than 1/n, since the entire reservoir is in
motion, and it may be expected in general that F » 1/n. Values of F for
the weakly-stratified case would fall between these two limits and might
be expected to be on the order of 1/n.
The Froude number, written in the units found in the Corps of
Engineers data base (feet, acre-feet, cubic feet per second) becomes:
F = 0.00735 •
L • q
D • V
The only variable in the Corps of Engineers data base that can be
directly used in this formula is the reservoir volume, V. The other
variables, L, D, and q, are ancillary variables and must be estimated.
Estimates are derived in a following section that calculates L as a
function of V/H, D as a function of V/H, and q as a function of Q. V, H,
and Q are basic variables in the Corps of Engineers data base: normal
volume, hydraulic height of dam, and spillway capacity, respectively.
The estimators are shown to be statistically significant.
C-2
-------
PHOSPHORUS RETENTION MODEL
Although TVA recommends against using "models...assuming steady state
conditions and continuous stirred tank reactor behavior" for estimating
ambient phosphorus to judge TVA reservoirs, such models are common,
available, and seem useful for screening on a national scale (TVA, 1983).
One would not use such a model to make control decisions without
considerable extra site specific study and refinement. One could use
"tank reactor models" to identify the possible presence of effects, and
if so, the extent of effects.
One such "tank reactor" model, attributed to Vollenweider
(Vollenweider, 1975), and presented by Reckhow (USEPA, 1979), has the
operational feature that it can be utilized with the Corps of Engineer's
data base.
The Vollenweider model presented here can be theoretically derived
from the annual phosphorus mass balance to any impoundment:
Pp = M
(V
where:
Pp = the average annual phosphorus concentration in the pool of
the impoundment,
M = the annual total phosphorus input to the impoundment, and
Vs = the "apparent" settling velocity,
A = the area of the bottom of the impoundment, which is assumed
equal to the normal pool surface area, and
q = the average annual inflow to the impoundment.
Vollenweider implicitly identifies a constant settling velocity of Vs
equal to 10 meters/year. A criticism of Vollenweider's approach is that
it is empiric, because Vs is estimated, and applies to northern temperate
situations. Reckhow observes that Vollenweider's model will probably
overestimate Pp when high surface overflow rates, Q/A, are present, and
underestimate Pp in lakes that are highly enriched. Reckhow discusses
model refinements that allow Vs to vary.
However, the Vollenweider model is dimensionally sound and is only
empirical with respect to selection of Vs. Furthermore, it explains a
significant amount of the variance, R2, associated with TVA samples (44
percent) and associated with sampling data in this report (51 percent).
C-3
-------
These R^ values are discussed in detail later in this appendix. Thus,
although elementary, the Vollenweider has a place in screening reservoirs
and evaluation of national eutrophication potential.
The Vollenweider model is transformed to a more useful form by
dividing M by q and multiplying the terms in bracket by q. This approach
results in an expression of the Vollenweider model that multiplies the
input average phosphorus concentration (M/q) by a dimensionless retention
coefficient, e. The transformed model is given as:
PP=-
M
q
(v
Substituting Pj, the average annual phosphorus concentration of input
flow to the impoundment, for the quotient M/q yields:
Pp = Pi
(Vs • A) + q
The retention coefficient, e, is given as:
e =
(Vs • A) + q
The formula of the retention coefficient, e, for q (units of cfs) and
A (units of acres), assuming Vs = 10 meters/second, is
e =
(0.0453 • A) + q
Thus, if Vs is unbiased, a value of e = 0.33 implies that the average
inpool lake phosphorus concentration is 33 percent of the phosphorus
concentration of reservoir inflow waters. The possible bias is a
concern, but not a fatal flaw of this model, as long as it is identified
and adjustments are made.
To sum up, one can estimate Pp directly or by using the Vollenweider
retention coefficient times the input phosphorus concentration. The
estimate uses a prediction that captures a significant amount of the
variation. The prediction is biased by the use of northern temperate
C-4
-------
situations to establish Vs, and the bias can be dealt with by quantifying
it in the application to reservoirs in the United States. Once Pp is
estimated, the EPA criterion of 0.025 mg/1 for phosphorus can be used to
evaluate the potential for eutrophication. Furthermore, this approach is
useful only for screening purposes. As such, the progression to more
detailed analyses is necessary to evaluate specific sites and to specify
need for and type of mitigation.
PHYSICAL ATTRIBUTE ESTIMATORS
The Corps of Engineers data base contains the following basic data
elements:
V = the normal pool volume (acre-feet),
H = the hydraulic height of the dam (feet), and
Q = the spillway capacity of the dam (cfs'.
As discussed earlier, the mixing model estimates the Froude number,
F, using the following independent variables, L, D, q, and V. Volume -is
the only variable contained in the basic data. The Vollenweider model
estimates the phosphorus retention coefficient, e, using the independent
variables, q and A, neither of which are contained in the basic data.
Therefore, to estimate F and e, using the Corps of Engineers data base,
estimates of L, D, q, and A as functions of V, H, and Q are necessary.
To secure such estimates, ancillary data for L, A, and q are
developed for subsets of the random sample of dams. For reasons of
confidence, a sample of at least 30 out of the 120 random dams, plus case
study dams given in Appendix B, is desired. The following methods are
used:
for L and A, the length and surface area of the impoundment are
measured using a USGS topographic map. The variables are
converted to units of feet and acres.
for q, the average annual flow is taken from the Reach File mean
flow data. The Reach File is linked to STORET and is an EPA in-
house data file. The unit of q is cfs.
D, the average depth, is computed by dividing V by A. The unit
of D is feet.
The ancillary data for L, A, and q, for the several sites for which
these data are estimated, are presented in Appendix B. The site
selection involved application of engineering judgement to spread the
sites around the United States and across power and nonpower dams.
Once obtained, least squares, linear regression analysis is performed
to estimate the fit. The models are specified, a priori, to be
C-5
-------
dimensionally sound and are constrained to have zero intercepts to give
logical functional relationships that are applicable to small dams.
Thus, q is estimated from Q. A is estimated from V/H; and L is
estimated from the square root of V/H. The linear regressions, sample
sizes, and percentages of explained variation, R, are:
q = 0.0647 • Q (n = 64, R2 = 0.61),
A = 2.88 • - (n = 30, R2 = 0.59), and
H
L = 6.61 •
43560
0.5
(n = 30, R2 = 0.47)
The units are in cfs for q and Q, acres for A, and feet for L and H.
The three regressions are statistically significant and are used to
estimate F and e. A discussion of the significance of R2 follows.
SIGNIFICANCE OF CORRELATION
The correlation coefficient, R, measures the strength of a linear
relationship between two variables. For R = 1, the relationship is
exactly linear whereas R = 0 indicates no relationship. Therefore, as R
approaches unity, a linear relationship approaches exactitude.
The square of the correlation coefficient, R2, measures the
percentage of the variation of a dependent variable that is explained by
its linear relationship with an independent variable. Variation is
defined in statistical terms as variance which is the square of the
standard deviation.
Therefore, if SD(y) is the standard deviation of a random variate, y,
and SE(yjx) is the standard error of predictions associated with a least
squares regression of y on x, a dependent variable, then the percentage
of explained variation is expressed as:
, [SD(y)]2 - [SE(y|x)]2
R2 =
CSD(y)]2
C-6
-------
The difference in the numerator is the variance or variation that is
conditionally explained by the least squares regression of y and x. The
square root of R2 is considerably less than R.
How does one judge the significance of R2 with respect to sample size
n? Kendall presents a logarithmic transform of R that approaches
normality (Kendall and Stuart, 1963). The transform, attributed to
Fisher, is given as:
z = 0.5 • loge
whose standard deviation is:
SD(z) = 1 / (n - 3)°-5
Because z tends to normality, the normal probability distribution can
be used to test the hypothesis that R is rot equal to zero, and hence, a
relationship is significant. At the 95 percent confidence interval, the
following values of R2, depicted below in Table C-l, are necessary to
infer that a relationship is significant using Fishers' z.
Table C-l. Sample Sizes for 95% Significant R2.
n, sample size
R2 above which R can be
deduced to be nonzero, and
hence, significant, with
95 percent confidence.
10
15
20
25
30
40
50
100
40%
26%
20%
16%
13%
10%
8%
4%
To place correlation results presented in this report into context,
the various R2s and sample sizes are summarized in Table C-2.
C-7
-------
Table C-2. Significance of Elementary Models.
Relationship
q vs Q **
A vs [V/H1 c **
L vs [V/H]°-5 **
STemp vs Froude F*
Measured P vs
Vollenweider *
Measured P vs
Vollenweider **
R2
61%
59%
47%
68%
44%
51%
n
64
30
30
22
11
22
95% Confidence
Significant
Significant
Significant
Significant
Significant
Significant
TVA Data.
Relationships derived
this report.
in various sections of
SAMPLE SIZE
Is a sample of 40 dams adequate to represent the finite population of
424 large power dams? For estimating the effects of dams, the goal is to
find how many dams have significant effects and how many do not. The
answer depends upon statistical reasoning. The statistician asks the
question: What sample size will provide given confidence intervals?
In other words, once you take your sample of n, you want to be 95
percent certain (that is, you want the estimate to be right 19 out of 20
times if you would repeat the random sample 20 times) that your answer is
within a particular error bound.
This problem can be theoretically formulated, like a coin tossing
experiment, with the Hypergeometric distribution. The error bound at 95
percent certainty is approximately plus or minus two standard deviations
of the standard deviation of the Hypergeometric distribution because it
approaches normality (Kendall and Stuart, 1963).
The standard deviation of the Hypergeometric distribution is:
SD =
n(p) ' (1 - P)
0.5
C-8
-------
where:
n = number of random samples,
p = fraction of the population having a particular attribute; for
example, p = 0.5 indicates 50% of the population possesses the
attribute, and
N = the size of the finite population.
Applying this Hypergeometric formulation to the sample of 40 large
power dams out of the total population of 424- of such dams, assume the
attribute of interest has a frequency of 50 percent or p = 0.5. Thus,
the assumption is that low dissolved oxygen in tailwaters, for example,
is present in 50 percent of all large power dams. For this assumption,
the standard deviation of the Hypergeometric distribution is 3 dams.
Twice the standard deviation is 6 dams. The expected number of dams is p
x n which for a sample of 40 with p = 0.5 is 20.
Therefore, out of a sample of 40 of a population with an attribute
with 50 percent frequency, one would expect between 20-6 and 20+6
dams (the expected value plus or minus two standard deviations) to
possess the attribute with 95 percent confidence. In other words, the 95
percent confidence range is 14 to 26 dams. A random sample of 40 would
fall within this range with 95 percent certainty. With the same logic a
random sample of 40 would fall within a range of 15 to 25 with 90 percent
certainty.
With regard to the need for partitioning the population into the
three subsets, large power, large nonpower, and small: assume that an
attribute is frequent in a small subset and infrequent in the entire
population. For example, let the small subset be the 424 large power
dams and the population be the 68,155 dams. Further, assume the
attribute frequency is 50 percent of the large power dams and 2.5 percent
of the population; the attribute of significant dissolved oxygen change
in dam tailwaters is a good example having high subset frequency and
presumed low population frequency. Thus, the partitioning places the
attribute in a subset sample range of 14 to 26 dams with n = 40, but a
representation of the attribute in the larger population, with 95 percent
confidence, implies a range of 0 to 3 with a high probability on 0. To
sum up, the number of 40 random subset dams possessing the attribute
would fall in the range 14 to 26, and the number of 40 random population
dams in the range 0 to 3. One would stand a sizable chance of not
identifying the attribute at all in a sample of 40 of the population.
SAMPLE REPRESENTATIVENESS
Do the forty dams distributed in each of three samples adequately
reflect the locational distribution of the associated population? The
samples are classified by geographic region. The first region,
Northeast, is defined as EPA regions I, II, III, and V. The second
C-9
-------
region, designated as South, contains EPA regions IV and VI. The final
region, West, is defined as EPA regions VII, VIII, IX, and X. The
comparison of the samples and the populations are illustrated in Table C-
3. Generally, it can be seen that the proportion of the observations in
sample regions are approximately equal to those of the population
regions.
Table C-3.
Summary of Geographic Distribution.
CLASSIFICATION
Large Power
Sample
Population
Large Nonpower
Sample
Population
Small Dam
Sample
Population
REGION
Northeast
8 (20%)
110 (26%)
16 (40%)
589 (35%)
11 (28%)
15478 (23%)
South
15 (38%)
139 (33%)
11 (28%)
488 (29%)
12 (30%)
25320 (38%)
West
17 (42%)
175 (41%)
13 (32%)
624 (36%)
17 (42%)
25232 (39%)
Total
40
424
40
1701
40
66030
Do the sample frequency distributions represent the population
frequency of all dams in each partitioned subset? The Kolmogorov-Smirnov
(KS) test is applied to answer this inquiry (Kendall and Stuart, 1963).
Figure C-l shows the cumulative distributions of V (normal pool volume)
of the population and the sample for large power dams. On the vertical
axis is the cumulative frequency or probability. The cumulative
frequency times 100 represents the "percent equal to or less than." The
horizontal axis for Figure C-l represents the volume.
The KS test involves determining the maximum vertical distance
between two frequency curves such as illustrated in Figure C-l. For
example, for the large power sample and population distribution, the
maximum vertical distance is 0.15 cumulative probability units. The
value 0.140 is the KS test statistic. If this statistic is less than the
value 0.210 (the cumulative probability units associated with a sample
size of 40 and a significance level of 95 percent), the sample and
population distributions are statistically similar.
In other words, the KS test compares the continuous distribution of a
population to the continuous distribution of a sample. The maximum
difference between the two distribution functions is itself a
distribution as a function of sample sizes and significance levels. The
value of the maximum difference distribution is compared to tabulated
C-10
-------
values of this distribution for various sample sizes and levels of
significance. If the level of significance value is greater than the
maximum difference, the sample is a representative, significant sample of
the population. The distributions are checked with the KS test for
volume, Froude number, phosphorus retention coefficient, and installed
power. The results are depicted in Table C-4.
Volume Distributions
(large power dams)
Cumulato.ua Probability
.98
.98
. 88
.78
.68
58
KS Statistic =0.15 !
n = 48
N = 424
i.88«+88i i.aa«»882 i.ae«+ea3 i.
Figure C-l.
Volume Distributions (large power dams)
C-ll
-------
Table C-4.
Results of the KS Test for Sample Distributions.
Distribution
Normal Volume (V)
1. Large Power
2. Large Nonpower
3. Small Combined
Froude Number
4. Large Power
5. Large Nonpower
6. Small Combined
Phosphorus Retention
Coefficient (e)
7. Large Power
8. Large Nonpower
9. Small Combined
Installed Power
10. Large Power
KS Test
Statistic
KS 95 Percent
Significant Level
0.15
0.14
0.17
0.13
0.15
0.14
0.21
0.21
0.21
0.21
0.21
0.21
0.12
0.10
0.12
0.12
0.21
0.21
0.21
0.21
Note: All 10 distributions are significant at 95 percent level.
STORE! WATER QUALITY RETRIEVAL SUIflARIZATIONS
This section presents information on the STORET parameters and the
incidence of data at the three classes of dams and the incidence of water
quality observations at dams for which data are present. Table C-5
presents the water quality parameters.
Tables C-6 and C-7 illustrate the distribution of the data for each
of the three categories, (L-P, L-NP, S), at upstream (U), downstream (D),
and Pool (P) locations. Table C-6 shows the number of dams reporting
data. The large power dams have the largest average number of
observations per parameter. The large nonpower dams are next. The data
for small dams are sparse. Table C-7 shows the average numbers of
observations per dam.
C-12
-------
Table C-5.
Water Quality Parameters.
Parameters
Fishkill
Stream Flow
Mean Daily
Dissolved Oxygen
Probe
Winker
Phosphorus
Total (as P)
Nitrogen
Total (as N)
TKN
N03 + N02
Turbidity
Jackson Candle
Storet #
1340
60
299
300
665
600
625
630
70
Units
dead fish
cfs
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
mg/1
Parameters Storet #
Water Temperature
Celsius
Iron (Fe)
Dissolved
Total (as Fe)
Manganese
Dissolved
Total
Hydrogen Sulfide
as H2S
Chlorophyll A
as Chi -a
BOD5
Dissolved Solids
10
1045
1046
1056
1055
71875
32230
310
70301
Units
degrees
ug/1
ug/1
ug/1
ug/1
mg/1
mg/1
mg/1
mg/1
Table C-6.
Number of Dams Reporting Data.
Parameter*
DO
Manganese Total
Iron Total
Nitrogen Total
TKN
Phosphorus
Temperature
Upstream (U)
L-P
33
22
24
19
30
28
33
L-NP
10
9
9
4
10
11
12
S
2
2
2
0
1
2
3
Pool (P)
L-P
23
12
11
7
19
22
24
L-NP
14
8
9
3
12
13
14
S
1
0
0
0
1
1
1
Downstream (D)
L-P
33
19
22
12
27
27
33
L-NP
12
9
10
2
11
11
13
S
3
1
2
0
2
3
3
*Data were found for other parameters, but much less frequently.
C-13
-------
Table C-7.
Average Number of Observations per Dam
for Dams Reporting Data.
Parameter*
DO
Manganese Total
Iron Total
Nitrogen Total
TKN
Phosphorus
Temperature
Upstream (U)
L-P
596
330
310
292
269
358
656
L-NP
232
34
63
35
122
288
491
S
6129
102
110
-
2
16
171
Pool (P)
L-P
693
104
102
83
147
157
693
L-NP
464
82
94
49
90
89
995
S
45
-
-
-
3
3
45
Downstream (D)
L-P
331
57
61
56
86
142
343
L-NP
215
71
895
22
85
115
209
S
111
2532
66
-
95
39
120
*Data were found for other parameters, but much less frequently.
C-14
-------
APPENDIX 0
LARGE NONPOWER DAM SUPPLEMENT
The large nonpower dams classification consists of 1701 dams with
over 10,000 acre*feet of storage at the normal pool elevation with no
reported installed power. These dams may or may not have low level
outlets which transmit water quality effects downstream.
Four analytical efforts were performed on large nonpower dams for
this report: Mixing Analysis; Dissolved Oxygen Concentrations in Dam
Tailwaters; Upstream/Downstream Comparisons; and Phosphorus Enrichment
Analysis. The mixing analysis is presented in Chapter III, but the other
three efforts are presented here because insufficient data were available
to draw conclusions based on these analyses.
MIXING ANALYSIS
These results are presented in Chapter III based on numbers of dams
with F < 0.3.
DISSOLVED OXYGEN IN DAM TAILWATERS
This method of analysis, described in detail in Chapter III, is based
upon a U.S. Department of Energy supported study done by the Oak Ridge
National Laboratory. Probabilities of non-compliance, i.e., the
probabilities that dissolved oxygen concentrations in the tailwaters of
dams will drop below 5 mg/1, were determined by the Oak Ridge Laboratory
for TVA dams, and using the same methodology, the results of this
analysis for large nonpower dams is presented in Table D-l.
UPSTREAM/DOWNSTREAM COMPARISON OF WATER QUALITY
This analysis, described in detail in Chapter III, involved the
collection of water quality data on sample dams from STORET, and the
comparison of up and downstream data for each site, when available. The
results of comparing water quality means for large nonpower dams are
presented in Table D-2. Twenty-five percent, or less, of the random
sample have water quality data both upstream and downstream of the
impoundment.
D-l
-------
Table D-l.
Probabilities of Non-Compliance with 5 nig/1 Dissolved Oxygen
for Large Nonpower Dams.
Summer Season
Location
Great Basin
Great Plains
Hawaii
Lake States
Northeast
Ohio Valley
Pacific Coast
Rocky Mountain
Southeast
Mean
n
_
1
-
-
2
3
7
1
1
PNC
_
0.020
-
-
0.000
0.033
0.073
0.070
0.000
0.033
Winter Season
Great Basin
Great Plains
Hawaii
Lake States
Northeast
Ohio Valley
Pacific Coast
Rocky Mountain
Southeast
Mean
_
1
-
-
2
3
7
-
1
_
0.010
-
-
0.010
0.000
0.027
-
0.000
0.009
D-2
-------
Table D-2
Tallies of Water Quality Changes
In Large Nonpower Dams
Parameter
Temperature
Dissolved Oxygen
Dissolved Oxygen*
Iron
Manganese
Phosphorus
TKN
Total Nitrogen
Dams
Lacking
Necessary
Data
30
31
31
34
36
31
31
38
Dams
Total
with Upstream and
Having Signifi
Data Increase
10
9
9
6
4
9
9
2
4
3
0
2
2
2
1
1
Downstream Data
cant
Decrease
3
4
2
2
0
2
2
0
Insignifi-
cant**
3
2
7
2
2
5
6
1
* nearest station to dam
** indicates no change in water quality
PHOSPHORUS ENRICHMENT ANALYSIS
The results of the phosphorus enrichment analysis, described in
detail in Chapter III, for large nonpower dams are as follows. Of the
forty sample dams, 24 had no phosphorus data whatsoever, seven had inpool
phosphorus concentration less than the suggested EPA "Gold Book" guidance
value of 0.025 mg/1, while nine had concentrations greater than 0.025
mg/1.
D-3
-------
APPENDIX E
SMALL DAM SUPPLEMENT
The small dam classification consists of 66,030 dams with less than
10,000 acre-feet of storage. Four analytical efforts were performed for
this report: Mixing Analysis; Dissolved Oxygen Concentrations in Dam
Tail waters; Upstream/Downstream Comparison; and Phosphorus Enrichment
Analysis. The results of these analyses follow.
MIXING ANALYSIS
The mixing analysis, explained in detail in Chapter III, categorizes
the stratification potential in the impoundments. The spillway design
for small dams may be based on different criteria than large dams. One
criteria utilizes the maximum probable flood (U.S. Dept. of the Interior,
1977). Furthermore, it appears that many designs of small dams use
standard designs (such as a 24-inch outlet pipe) for a wide range of
hydrologic conditions. Design criteria could result in highly variable
maximum spillway discharges which would, in turn, result in an
undesirable variation in the Froude number. These small dam Froude
numbers may not accurately predict the mixing potential of small
impoundments.
However, the linear estimate of average flow is dimensionally sound
and has been constrained to a zero intercept to allow estimates for small
values. Also there are several small dams in the data utilized to
develop the linear estimate. With these various qualifications, the
numbers of small dams with F < 0.3, the TVA suggested criteria, is 9954
out of 66,030 dams; data are missing (F could not be calculated) for
18,143 dams.
DISSOLVED OXYGEN IN DAM TAILWATERS
This analysis was not performed for small dams.
UPSTREAM/DOWNSTREAM COMPARISON OF WATER QUALITY
There are 66,030 "small" dams having less than 10,000 acre-feet of
normal pool volume comprising 96.8 percent of the population of Corps of
Engineers dams. Although the maximum normal volume is 10,000 acre-feet,
the median normal volume is 70 acre-feet, and one of the lower limits
(the other being dam height) of the Corps of Engineers 1980 census is a
pool volume of 50 acre-feet. Thus, about half the small dam population
have volumes less than 70 acre-feet. On the average, these dams are 199
E-l
-------
the size of the largest "large" dam in the partitioned sample.
That is, the "large" dams start at 10,000 acre-feet, and the small dams
are typically 100 acre-feet - a 100-fold difference.
Of the 40 small random dams in the sample, only 5 had some form of
water quality data. Therefore, there is very little water quality
monitoring evidence to indicate the effects of small impoundments. Since
monitoring resources are usually focused on problem areas, lack of
monitoring may suggest that either a site has no water quality problems,
or that the water quality extent of the site effects is limited.
One can ask the question, how do small impoundments behave? This
question arose in the 1970s relative to concerns for filing EIS documents
to support new dam construction. Several intensive SCS studies in the
southeast (Moore, 1973; SCS, 1978; SCS, 1987) provide site specific
answers.
In Arkansas, a 4000 acre-feet water supply reservoir 20 feet deep, on
the average, is strongly stratified. The Prairie Grove Lake showed
extensive periods of zero dissolved oxygen in the hypolimnion combined
with resulting solution of iron and manganese (Moore, 1973). The concern
is the utility of the impoundment for municipal water supply. A masters'
thesis (Kerr, 1977) followed up by experimenting with bottom withdrawals
to unsuccessfully promote impoundment mixing. Apparently, however, low
dissolved oxygen hypolimnetic waters are aerated in the splash following
freefall down a morning glory spillway - essentially a vertical pipe with
an elbow at the base leading to a horizontal discharge pipe which passes
through the dam. Kerr fabricated a shroud to fit over the vertical pipe
to allow bottom withdrawals. The withdrawals splash at the elbow.
This form of small dam release detail, a vertical pipe with an outlet
elbow, is common in dams built following SCS suggestions, as is the elbow
splash and resultant aeration. Thus, in general, small morning glory
spillways, whether or not they handle bottom or surface waters,
apparently provide aeration to tailwaters.
Extensive data gathering at four small impoundments in Mississippi
provide insight to small impoundment behavior (SCS, 1978). The four
projects have average water depths of 6 to 40 feet and volumes of about
200 to 8000 acre-feet. All of them stratified creating anoxic
hypolimnions. If they had surface releases (3 out of 4), the tailwaters
are warmer than inflows and have higher dissolved oxygen. The behavior
is consistent with a theory of surface warming and either algal oxygen
production or splash aeration of discharges, or both. The one
impoundment with low level outlets had low dissolved oxygen, iron, and
manganese in solution.
Thus, it appears that small impoundments can and do have poor quality
hypolimnetic waters. If they have low level outlets, these waters are
transmitted downstream. In the Mississippi studies, 1 out of 4 dams had
low level outlets. A further insight is the 4 Mississippi sites all had
high levels of phosphorus in their inflows and impoundments. This may be
E-2
-------
a characteristic of rural, agricultural regions having small
impoundments. In a watershed management context, such small impoundments
may filter phosphorus out of runoff and aid the phosphorus control in
downstream dams.
In Alabama, a study of the influence of flood control impoundments on
water quality by the SCS provides additional insight into small dams
(SCS, 1983). In Alabama, there are more than 80 structures which
discharge only surface waters, but there are only two which normally
release flows from the cooler bottom waters. The incidence of low level
outlets is 2 out of 80 for small Alabama flood control dams. The Alabama
impoundments tended to be stratified with a noticeable difference in
dissolved oxygen and temperature - top to bottom. The surface releases
tended to be warmer, as reported for the four Mississippi dams.
Specific details of the five random small dams provide the reader
with additional insights into the small dam situation. For example, one
of the random dams is a "dry" dam holding water only during floods. At
least one percent (742) of the small dams in the Corps of Engineers data
base are of this type. Up to an additional two percent (1137) could also
be dry dams because normal storage volumes are not recorded in the Corps
data base. Dry dams and, for that matter, many navigation-^ projects, are
not expected to cause significant downstream water quality problems.
Additional details of the five random small dams follow (water quality
data are summarized in Appendix B):
Parizek Pond Dam: CT00646
Parizek Pond dam, completed in 1870, impounds Parizek Pond. The dam
is located at 41°52.9' latitude and 72°17.3' longitude in Willington
County, Connecticut. The dam is of earthen, gravity construction with a
structural height of 15 feet and a hydraulic height of 12 feet. The
length along the crest of the dam is 400 feet with an uncontrolled
spillway 11 feet wide. A maximum flow of 30 cfs empties into Conant
Brook from the impoundment. The maximum storage capacity of the
impoundment is 61 acre-feet with a normal volume of 56 acre«feet. The
major purpose of the dam is to provide recreational opportunities. The
dam and impoundment are privately owned and maintained by William and
Mary Parizek. The data on record pertaining to Parizek Pond comes from 8
stations, 3 downstream and 3 upstream. The data indicate a low level of
dissolved oxygen downstream below 5 mg/1 and nutrient enrichment (mean
IA total of 130 dams in the Corps data base are denoted as having
one or more locks, while a total of 329 dams are reported to serve a
navigational purpose. The data base appears to be incomplete and
inconsistent on this point. The Corps of Engineers (Kennedy and Gaugush,
1987) state that lock and dams comprise 26 percent of 783 Corps water
resources projects, or approximately 200. Furthermore, the 130 locks and
dams in the Corps safety inventory are not a subset of the 329 reporting
a navigation purpose.
E-3
-------
phosphorus concentration of 0.12 mg/1) may be found downstream. No
statistical difference between upstream and downstream temperatures is
observed at the 90 percent confidence level.
KSNONAME 1335: KS01335
The dam, located at 38°26.7' latitude and 96°20.5' longitude in Chase
County, Kansas, was completed in 1967. The earth construction dam has a
structural height of 16 feet and a hydraulic height of 15 feet,
restraining an unnamed small stock or farm pond. The crest length of 620
feet contains an uncontrolled spillway discharging a maximum flow of 636
cfs into Beaver Creek. The maximum storage capacity of the pond is 83
acre-feet with a normal volume of 42 acre-feet. The dam is owned by
Kellam and regulated by DWR. The recorded data comes from 12 downstream
stations. The data indicate a mean dissolved oxygen content downstream
above 5 mg/1 and possible nutrient enrichment downstream. Iron and
manganese levels are high downstream, well over the maximum threshold.
Tyner Lake Dam: KY00271
Tyner Lake Dam, located at 37°22.6' latitude and 83°54.8' longitude
in Jackson County, Kentucky, impounds the Flat Lick Creek Reservoir. The
dam was completed in 1969 and is of earth construction with a structural
height of 69 feet and a hydraulic height of 67 feet. The dam has a crest
length of 1030 feet with an uncontrolled spillway 100 feet wide,
discharging a maximum flow of 3207 cfs into Flat Lick Creek. The maximum
storage capacity of the reservoir is 3250 acre-feet with a normal volume
of 2365 acre-feet. The system is intended to provide a source of water
as well as serving as a recreational facility. It is owned and
maintained by the Jackson County Water Association. The data recorded
comes from 1 station in the pool. These data indicate a dissolved oxygen
content marginally below 5 mg/1 and low phosphorus levels.
Hodges Village Dam: MA00967 (Dry Dam)
The Hodges Village Dam, completed in 1959, impounds Hodges Village
Pond. The dam, located at 42°07.2' latitude and 71°52.8' longitude in
Oxford County, Massachusetts, is primarily of earth construction with
rockfill and gravity design, with a structural height of 55 feet and a
hydraulic height of 50 feet. The crest length is 2050 feet with an
uncontrolled spillway 145 feet in width discharging a maximum flow of
25,800 cfs into the French River. The maximum storage capacity of the
impoundment is 26,000 acre-feet with a normal volume of 0 acre-feet.
This dam was constructed for flood control purposes as indicated by the
normal storage volume.
Sanitation Dam: MI01262
The dam, located at 42°31' latitude and 84°39' longitude in Ingham
County, Michigan, was completed in 1918. It is a gravity-type structure
with a hydraulic height of 2 feet. The crest length is 150 feet with a
130 feet spillway which discharges into the Grand River. The unnamed
E-4
-------
impoundment has a normal storage capacity of 8 acre-feet. This dam is
apparently below Corps of Engineers thresholds for safety reporting, and
its presence in the data base may indicate overzealousness on the part of
the census taker. The dam is owned and regulated by the City of Eaton
Rapids. Three upstream stations provide data on water quality. These
data indicate an upstream dissolved oxygen content above 5 mg/1,
phosphorus enrichment upstream, and high iron and manganese content.
The situation for small dams is water quality data are only present
for 12.5 percent (or 5) of the random sample of small dams. However,
site specific studies in Arkansas, Mississippi, and Alabama all indicate:
Small dams stratify and develop anoxic epilimnions.
Small dams with surface outlets tend to have higher dissolved
oxygen and temperature in tail waters than in inflows.
Small dams with low-level outlets show decreasing dissolved
oxygen downstream and the presence of iron and manganese.
The incidence of low-level outlets is low in the site specific
situations investigated: 1 low-level outlet in 4 Mississippi
sites and 2 low-level outlets in over 80 Alabama sites.
The typical morning glory vertical outlet with 90° outlet elbow
seems to generate a splash at the elbow that aerates the
tailwaters.
PHOSPHORUS ENRICHMENT ANALYSIS
The results of the phosphorus enrichment analysis as described in
Chapter III are rather sparse. For small dams, the random sample of 40
only had 4 dams with pool estimates of phosphorus. Of these four, 3 had
phosphorus levels over the 0.025 mg/1 critical threshold. The 4
Mississippi dams with site specific data all had high pool concentrations
of phosphorus. They also had high inflow concentrations of phosphorus.
Small dams are higher in watersheds and nearer to nonpoint sources of
enrichment. They also tend to filter or "screen out" the phosphorus
concentrations with the result that downstream inputs of phosphorus to
larger dams is less than the setting when small dams are absent.
REMARKS
The analysis of small dams is hampered by a
However, the following items are enlightening:
lack of monitoring data.
The small dams are only half as likely to stratify and thus be
poorly mixed based on population estimates.
E-5
-------
Site specific studies in the southeast do show that small dams
stratify. Low-level outlets from small dams are associated with
tailwater water quality effects. Surface outlets exhibit
tail waters with increased dissolved oxygen and temperature.
The incidence of low-level outlets in small dams appears to be
low in small dams - on the order of less than 3 percent for a
sizable sample in Alabama.
Insufficient data are obtained to draw quantitative conclusions
pertaining to small dams. What evidence exists quantitatively, points to
a situation of widespread enrichment of small dam reservoirs.
E-6
-------
APPENDIX F
CORPS OF ENGINEERS
WATER RESOURCE PROJECTS
WATER QUALITY ASSESSMENT
SUPPLEMENT
GENERAL
The U.S. Army Corps of Engineers, as one of the principle Federal
water resource development agencies, owns and operates over 700 water
resource projects consisting primarily of reservoirs and locks and
dams. The Corps civil works responsibilities began with an Act of
Congress in 1824 for the improvement of rivers and harbors for
navigation. Other acts expanded the legislative basis for Corps
participation in the various functional areas of water management.
Today, the Corps carries out a comprehensive water resources
planning, engineering, construction and operations effort in close
cooperation with government agencies at all levels and a wide variety
of civic and private interests. This program involves coordinated
management of water resources in a manner that addresses all
water-related requirements, both immediate and long-range. These
requirements include flood control, navigation, hydropower, water
supply, water quality, recreation, and fish and wildlife enhancement.
The allocation of storage and authorities to regulate these projects
are specified in legislative authorization acts for specific Corps
projects, as well as project and reservoir system documents.
Inceased public concern for the environment and recognition that
natural resources are both interrelated and finite, resulted in the
incorporation of considerations other than economic efficiency into
legislation for water recource development and management.
Environmental considerations such as water quality improvement and
management, and fish and wildlife requirements, were manadated by
legislation, and are incorporated into authorized project purposes and
operating plans, as well as Corps of Engineers regulations.
F-l
-------
Water control policy and management in the Corps of Engineers has
undergone significant changes in response to evolving environmental
concerns. Many of these changes have been directly in response to
Federal environmental legislation. Reservoir design and operation
activities which once focused solely on quantitative aspects, are now
sensitive to broader environmental considerations. Water quality is an
important consideration during all phases of Corps project development,
and an, integral part of the decision processes related to reservoir
operations.
Water control management programs provide the means for operating
Corps projects to meet their authorized purposes. Most of the Corps'
projects are authorized for multiple purposes, (e.g. flood control,
navigation and recreation) and over 60 Corps of Engineer projects
include water quality as an authorized project purpose. Flow
augmentation for industrial and municipal pollution abatement, acid
mine drainage abatement, and other purposes which relate to water
quality, are often included in flood control and navigation projects.
Water quality management objectives have been developed for each
project and incorporated into Corps water control programs wherever
possible.
The Corps recognizes that project regulation often plays a
significant role in influencing downstream environmental and water
quality conditions. Sensitivity to the effects of water control
activities on the environment is the cornerstone of the Corps'
initiatives for integrating water quality and environmental
considerations into water control management.
Experience gained over the last decade, through water control
management activities and Corps research and development efforts, has
contributed to a growing information base for designing and operating
reservoir projects to manage water quality. A large portion of these
research efforts were conducted as part of the Environmental and Water
Quality Operational Studies Program, an eight year effort which was
initiated by identifying technology and information needs for solving
environmental problems and meeting the responsibilities of legislative
and executive directives.
F-2
-------
This work, which was completed in 1985, produced a bettei
understanding of the effects of reservoirs on water and envmararential.
quality. Some of the efforts under this program addressed processes
involved in inflow mixing, internal reservoir mixing and suspended;
solids distribution. Methods for predicting environmental changass and
alternatives for managing water and environmental quality were=
evaluated. Predictive techniques were developed for describing.
reservoir hydrodynamics, loadings to reservoirs and for detecmi'niTMg iffee
ecological effects of reservoir project operation. A variety of otflnst
areas were also included in this program, and the resulting teehnolicagy
and criteria has been incorporated into Corps regulations and1 technical
guidance.
In making water control decisions,the Corps must take a vs-ci&tiy osBS> arad
the potential frequently exists to regulate the project(s) for
additional benefits. This flexibility depends on the compati&Mtty off
each water use, the characteristics of the project or river sg
water use requirements and other factors.
Water levels in impoundments may be regulated to providte
storage space to control floods, as well as to store water fear, a
range of uses. Releases may be regulated to achieve requitaientes ffinr
public use, recreation, and to support fish and wildlife needb
downstream. System regulation for water quality is most
F-3
-------
during low flow periods when available water must be used efficiently
to avoid degrading lake or river quality. However, even this worthy
objective often conflicts with other water needs such as water supply
or hydropower storage.
The melding of all of the above considerations usually requires
some degree of compromise to achieve water management goals. For each
project or system, the use requirements are evaluated to assure the
greatest project benefits. This balancing of water use demands,
priorities and project capabilities is the overall goal of the Corps
water control management program.
Finally, water quality management activities must compete with a
large number of other efforts for limited manpower and funding
resources. Given a particular level of resources, agencies prioritize
their programs, and carry them out at a level of effort consistent with
each program priority. The heightened public concern for the
environment, and the resulting legislative and executive mandates, have
raised the visibility and priority of water quality management programs
in the Corps of Engineers which resulted in the accomplishments
previously described.
WATER QUALITY STATUS OF CORPS OF ENGINEERS PROJECTS
a. Introduction
The Corps of Engineers has constructed and now operates more than
700 water resource projects having a total surface area of nearly
27,000 square kilometers. The geographic distribution of these
projects, as depicted in Figure 1, reflects regional differences in
water resource development requirements, water control agency
responsibilities, and topographic requirements for cost-effective
construction. Impoundments providing navigation benefits, which
comprise approximately 26 percent of all Corps projects, are located
along major inland waterways. These include the Mississippi River and
its major tributaries, the Arkansas and Red Rivers draining from the
F-4
-------
west, and the Ohio and Illinois Rivers draining from the east. Other
waterways of importance include the Alabama River and the
Tennessee-Tombigbee Waterway in the mid-south, and the Columbia River
in the northwest. Twenty one percent of all projects are dry dams or
projects which, by design, provide minimal permanent water storage
during non-flood periods. These projects are most prevalent in the
arid southwest, where flooding conditions are associated with
intermittent periods of excessive runoff, and in the New England
states.
Reservoir projects providing short- and long-term storage of water,
but not navigation benefits, comprise the remaining 53 percent of all
Corps projects. These projects can be broadly categorized based on
reservoir morphometry and tributary type. Deep, storage reservoirs are
formed by the impoundment of higher order streams and rivers, and are
frequently located in deep, steeply-sloped river valleys. These
projects tend to be deep, narrow and highly dendritic in shape.
Mainstem reservoirs are located on lower order (ie. larger) rivers and
tend to be shallower, wider, and less complex in shape.
b. Methods
Growing public concern over the quality of freshwater resources and
the desire to continue to provide responsible management of the
valuable environmental resource provided by its water resource projects
led the Corps of Engineers to institute several water resource research
programs. The purposes of these research programs were to expand the
understanding of processes influencing the environmental quality of
reservoirs and tailwaters, and to improve management technologies. As
a continuing effort, the Corps has established a number of technology
transfer programs as a means of distributing water quality information
and technology.
Currently, the Corps is compiling and analyzing water quality
information as a means of providing an up-to-date assessment of the
water quality status of all its water resource projects. Sources of
F-5
-------
information include annual water quality reports prepared by District
and Division offices, water quality data retrieved from EPA's STORET
database system, and detailed evaluations by project management
personnel.
For the purposes of this report, portions of the above mentioned
information have been reviewed and analyzed. Specifically, this
information was obtained through the use of a comprehensive
questionnaire designed to solicit detailed information concerning all
aspects of project design, operation, and water quality status. The
questionnaires were completed by personnel familiar with each pioject
and its water quality characteristics. With regard to water quality
status, subjective responses to questions concerning water quality were
requested. In general, these responses indicated the presence or
absence of water quality problems. In situations where problems were
indicated, graded responses allowed assessment of the severity of the
problem and the quality of the information upon which the assessment
was based. To date, questionnaires for approximately 470 projects have
been completed and compiled.
Since questionnaires for all projects have not yet been completed,
a sample of questionnaires was randomly drawn and analyzed for the
purpose of this report. The sample size was set at 10 percent (46
projects) and samples were drawn from strata based on project type
(reservoir, lock and dam, and dry dam) and District. The geographic
locations of sampled projects are presented in Figure 2 for comparison
with the distribution of all projects (Figure 1). Results presented
below are based on these analyses.
c. Assessment
Figures 3 and 4 present the water quality status of tailwaters and
pools associated with the sampled projects, respectively. A
shortcoming of the data upon which these figures are based is the fact
that reliable information concerning water quality status is lacking
for approximately 40 to 50 percent of the projects. A number of
F-6
-------
reasons for this lack of information are possible. The collection,
analysis, and reporting of water quality information at many projects
is performed by agencies other than the Corps. This is particularly
true for projects at which recreation and other water-based resources
are managed and maintained by local authorities. Thus, while water
quality data may be collected, such data may not be readily available
to Corps personnel.
Funding and manpower constraints have a significant impact on the
quality and quantity of water quality information collected at many
Corps projects. To overcome these constraints, priorities are often
established which provide for the collection of appropriate water
quality data for those projects at which water quality concerns are
deemed to be of highest priority. Other projects, because of
historical data or informal knowledge concerning their water quality
status, are sampled IPSS intensively or less frequently. Thus, a
degree of uncertainty and/or bias exists for data discussed here and
extrapolations of data compiled for the sampled projects to all
projects are not possible. The data do, however, provide a general
assessment of the types of water quality concerns associated with Corps
water resource projects and some indication of their relation to other
project attributes.
As depicted in Figure 3, approximately 60 to 65 percent of those
sampled projects for which evaluations of the water quality status of
tailwaters were available were considered not to exhibit problematic
conditions. For those projects indicated as exhibiting problematic
conditions, several categories of water quality concerns are apparent.
Most prevalent are concerns related to flow, the release of waters low
in dissolved oxygen concentration, and the erosion and transport of
sediment.
Extremes in flow and/or excessive changes in flow, which result
from operational procedures required to meet authorized project
purposes (e.g., flood control, power generation, etc.), may impact
downstream uses. While such conflicts in uses are frequently
problematic, every attempt is made to enhance non-authorized benefits
without incurring unacceptable impacts on authorized purposes.
F-7
-------
Flow-related problems for tailwaters include higher than normal flows
following flood events as retained flood waters are released, lower
than normal flows during periods then pool storage is being increased,
and daily fluctuations resulting from the operation of hydropower
facilities, particularly when power is produced to meet peak-load
requirements.
The loss of dissolved oxygen in the hypolimnia of reservoirs
potentially results in direct and indirect impacts for tailwaters. For
projects which, because of their structural or operational
characteristics, do not allow for complete reaeration of release
waters, dissolved oxygen concentrations below saturation may occur
throughout part or all of the summer stratified season. Such is the
case for approximately one-third of the projects inventoried here.
However, it should be noted only one of the 46 sampled projects
experiences periodically severe dissolved oxygen conditions in it's
tailwater and that this project is a newly-filled reservoir where such
occurrences are predictable and short-lived.
Coincident with the loss of dissolved oxygen from reservoir
hypolimnia is the potential for the release of dissolved materials,
particularly iron and manganese, from bottom sediments. The
accumulation of these materials in reservoir waters, in turn, leads to
their potential release to downstream areas. The occurrence of
elevated concentrations of metals and nutrients in tailwaters is
indicated for approximately 30 to 40 percent of the sampled projects
for which such evaluations were made. And, as would be expected, these
projects are primarily those for which reduced dissolved oxygen
concentrations were reported. It is also important to note that these
projects are also reported to receive relatively high inputs of these
materials from their surrounding watershed.
The transport of suspended sediment from reservoir to tailwater
*
and/or the erosion and resuspension of bank and bed materials impacts
tailwater areas below approximately 40 to 50 percent of the projects
for which evaluations were provided. In most cases, impacts are minor
and result from increased turbidity. In other cases, degradation of
immediate downstream areas is indicated. Preliminary evaluation of
F-8
-------
information suggests that, while project operation plays a significant
role in the determination of release conditions, pronounced regional
patterns in the distribution of such conditions are apparent. In
general, reservoirs located in regions dominated by highly-erodible
soils experience higher inputs of suspended sediment and, therefore,
often release turbid waters.
An evaluation of water quality conditions in pools, also based on
sampled responses to the questionnaire, are presented in Figure 4.
Most prevalent were problems related to the eutrophication process.
These include excessive nutrient concentrations, algal blooms, reduced
water clarity, macrophyte infestations, and the loss of dissolved
oxygen in bottom waters. Other conditions of concern include excessive
concentrations of reduced iron and manganese in bottom waters, and the
accumulation of sediment and contaminants. As was discussed for
tailwaters, problematic conditions were indentified for approximately
40 to 50 percent of the pools for which evaluations were available.
And, again, varying degrees of severity in problematic conditions are
apparent.
Complex interactions between biotic and abiotic components of the
reservoir ecosystem make simple, meaningful evaluations of data
difficult; however, several general patterns emerge from existing
information. Most notable is the impact of watershed processes and the
transport of material from watershed to reservoir. The loss of
nutrients and sediment from watersheds and their accumulation in
reservoirs fosters the growth of aquatic plants, primarily algae, and
losses in storage volume. These effects, in turn, lead to reduced
water clarity, reduced dissolved oxygen concentrations in bottom
waters, and the potential for internal material cycling between
nutrient-rich sediments and the overlying water column.
An understanding of the linkage between watershed and reservoir is
critical to our understanding of water quality processes and the
control of water quality problems. This is of particular concern to
agencies such as the Corps of Engineers since primary control of water
quality conditions must target processes occurring in that portion of
the reservoir-watershed ecosystem in which the water control agency has
F-9
-------
little or no regulatory authority. As an indication of the potential
importance of watershed processes, it should be noted that watershed-area
to lake-area ratios are higher for reservoirs than for natural lakes.
Obviously, as this ratio increases, so does the potential for increased or
excessive material loads to the reservoir.
A review of landuse patterns in the watersheds of the sampled
reservoir projects indicates that, on the average, natural, agricultural,
and urban/residential areas account for 45, 45, and 10 percent of the
landuses in the reservoir watersheds, respectively. Since agricultural and
urban/residential areas often contribute excessive nutrient and sediment
loads to streams and rivers, the control of point and nonpoint sources are
clearly indicated as the primary means by which reservoir water quality can
be protected or improved through reduced loading. This is underscored by
the observation that sampled projects for which eutrophication-related
problems were indicated also receive nutrient loads deemed by reservoir
water quality personnel to be excessive and problematic.
FUTURE DIRECTION
The following are areas where additional emphasis would benefit both
Congress and the federal water resource agencies in understanding and
managing water quality at federal reservoir projects.
a. Extensive research has been conducted to develop improved
techniques for analyzing river and reservoir water quality dynamics,
reservoir and reservoir system operations and associated
interrelationships. However, many unknowns and problems remain. Agency
water quality related research programs will be prioritized to address
urgent needs.
F-10
-------
b. Applications of technologies to ameliorate water quality
conditions at reservoirs have been undertaken by the Corps of Engineers.
Varying degrees of success have been achieved, and it has become evident
that additional direct field application and evaluation of water quality
enhancement techniques is needed to fill the void between research and
successful use in the field. The Corps intends to continue to conduct
demonstration programs to improve design guidance and define limitations,
capabilities and ancillary effects. Interagency cooperation in this
program would extend the limits of applicability and facilitate sharing of
agency expertise. This effort, in concert with the previously outlined
initiatives, would greatly enhance the federal agencies' abilities to carry
out Congressionally mandated responsibilities pertaining to reservoir water
quality management and the achievement of national water quality goals.
F-ll
-------
Figure 1. Geographic distribution of Corps of Engineers water resource
projects.
Figure 2. Geographic distribution of a ten percent, stratified, random
sample of Corps of Engineers projects for vchich questionnaires
have been received.
F-12
-------
Bacteria
Dissolved Solids
Fluctuating Flow
Fluctuating Temp
High Flow
High Nutrients
High Temperature
Hydrogen Sulfide
Iron
Low DO
Low Flow
Low Temperature
Mocrophytes
Manganese
Metals
Organics
Parasites
Sediment
Shore Erosion
Supersaturation
Taste and Odor
Turbidity
pH/Acidity
50
Percent
75
100
Figure 3. Frequency of occurrence of occasionally problematic (coarse
hatching), intermittently problematic (fine hatching), chronically
problematic (cross hatching), and non-problematic (no hatching) water
quality conditions in tailvaters. Difference between accumulative
frequency and 100 percent indicates percentage of projects for which no
evaluation was made.
Algoe
Bocterio
Dissolved Solids
Drowdown
High Nutrients
Hydrogen Sulfide
Iron
Low DO
Mocrophytes
Manganese
Metals
Orgonics
Parasites
Pool Fluctuation
Sediment
Shore Erosion
Taste and Odor
Turbidity
pH/Acidity
100
Figure 4. Frequency of occurrence of occasionally problematic (coarse
hatching), intermittently problematic (fine hatching), chronically
problematic (cross hatching), and non-problematic (no hatching) pool water
quality conditions. Difference between accumulative frequency and 100
percent indicates percentage of projects for which no evaluation was made.
F-13
-------
APPENDIX G
TENNESSEE VALLEY AUTHORITY'S ASSESSMENT
SUPPLEMENT
INTRODUCTION
This chapter provides an assessment of water quality conditions in the
Tennessee Valley Authority's (TVA) system of multipurpose water resource
projects (Figure 1). It responds to the Environmental Protection Agency's
(EPA) request for information for its report entitled "Dam Water Quality
Study--A Report to Congress under Section 52A, Clean Water Act of 1987."
It is organized in three main sections: Assessment of Water Quality;
Policies - Water Quality Management; and Recommendations.
ASSESSMENT OF WATER QUALITY
Water quality conditions for reservoirs in the Tennessee Valley were
assessed using the approach applied by the Corps of Engineers to assess
its 783 projects. First, a questionnaire was completed for each reservoir
by an individual knowledgeable about water quality in the reservoir. Then
information was obtained from personnel familiar with the primary reservoir
uses and reservoir operations. Finally, in several group settings, the
questionnaire responses were reviewed for uniformity in defining the
parameters and rating each project. Thirty-three projects were assessed,
including all the hydropower projects and four nonhydroprojects. Those
projects not included are small projects for which no data were available.
The results are presented in Figures 2 and 3 for pools and tailwaters,
respectively. Definitions are as follows: "continuous:" a chronic or
continuous problem; "seasonal:" an intermittent problem occurring on a
seasonal basis; "infrequent:" an occassional problem ocurring infrequently
on an annual basis; "no problem" and "no data:" self-defined; "severe:" a
severe impact resulting in the long-term loss of one or more user benefits;
"significant:" a significant impact that restricts but does not eliminate
user benefits; "minor:" a minor impact which does not restrict user
benefits.
Water uses were severely impacted at several sites. Low DO, hydrogen
sulfide, iron, and manganese are considered to be at sufficient levels that
the fishery at Upper Bear Creek. Reservoir is practically nonexistent.
Other reservoir projects having severe use impairment are the three Ocoee
projects where sediment accumulation, iron, manganese, turbidity, and metal
contaminants are adversely impacting aquatic life and recreation, primarily
in Ocoee number 3 with less impairment in Numbers 2 and 1. Finally, the
Nolichucky Reservoir has been filled with sediment to the point that it is
no longer considered a reservoir.
The results indicate that in reservoir pools the most significant impacts
were pool level fluctuations and bacteria (about 50 and 30 percent of the
reservoirs, respectively). The next most significant user impacts were
turbidity, algae, macrophytes, sediment accumulation, and shore erosion,
all of which occur at 15-20 percent of the reservoir projects. Minor
impacts, occurring at 20 percent or more of the reservoir projects, were
G-l
-------
related to the following parameters; iron, manganese, low DO, turbidity,
low temperature, high nutrients and algae, macrophytes, sediment
accumulation, pool level fluctuation, shore erosion, pH/acidity, bacteria,
and fish parasites.
Several items worth noting for the analysis on pools are: (1) data on
hydrogen sulfide were limited and the results of this analysis may change
when more data becomes available; (2) several parameters had high rates
of recurrence with the potential to impact uses in the future: i.e., high
nutrients, sediment accumulation, shore erosion; (3) low DO seasonally
occurred in about 70 percent of the reservoirs, but user impacts were
considered minor becasue the condition was restricted to the hypolimnion
or bottom waters of the reservoir, in which case "no data" was designated;
and (4) more data are needed on organics and metals in fish flesh.
For reservoir tailwaters, the results indicate that the most significant
impacts resulted from low DO, streamflow (high, low, and fluctuating), and
low temperature. The next most frequent user impacts were associated with
iron, manganese, hydrogen sulfide, turbidity, metal contaminants,
streambank. erosion, bacteria, and fish parasites. All of these parameters
occurred at about 15-20 percent of the projects. The most frequent minor
impacts, occurring at 20 percent or more of the projects, were related to
the following parameters: low DO, turbidity, high flow, fluctuating flow,
low temperature, high temperature, fluctuating temperature, streambank
erosion, and parasites in fish.
Several items worth noting on the tailwater assessment are: (1) even
though data were not available on the user impacts of gas supersatura-
tion, the impacts are probably minor; and (2) more data are needed on
hydrogen sulfide, bacteria, and fish flesh contaminants.
Reservoir Water Quality Management
TVA's programs on reservoir water quality management are focused on three
areas: tailwater management, reservoir management, and watershed
management. The long-range strategy is to achieve proper management of all
TVA reservoirs and tailwaters; however, at present the activities have been
directed toward reservoirs and tailwaters that have the greatest user
impacts.
TVA actively pursues reservoir water quality management through the
formation of "partnerships". To the extent practical, joint projects are
conducted with State agencies, EPA, and with public interest groups. These
partnerships are invaluable in ensuring that appropriate analyses are
conducted so that results are implemented in a timely manner. The
partnerships also reduce TVA costs for conducting studies and implementing
solutions to priority problems.
Tailwater Management Strategy—TVA's tailwater management strategy
concentrates on improving the habitat for fish and aquatic life,
recreational floating, assimilative capacity for treated wastewaters, and
providing for municipal and industrial water supply. The typical water
quality parameters include DO, streamflow, temperature, iron, manganese,
hydrogen sulfide, pathogens, sediment, and total dissolved gases.
G-2
-------
At the present time TVA is concentrating its efforts on improving DO
concentrations and enhancing stream flows. Elements of this effort
include:
1. Assessment of existing water quality, biological conditions, and
potential uses of the tailwater, followed by a plan that identifies
needed improvements and a monitoring strategy to assess its
effectiveness.
2. Aeration and/or oxygenation of the hydropower releases to about
4mg/L during those portions of the year when DO fails to meet this
value.
3. Provision of minimum or increased flows downstream from
hydroelectric power plants, tailored to meet specific needs as
determined through field studies and data analyses.
4. Investigation of impacts from iron, manganese, and hydrogen sulfide
and evaluation of methods to reduce these substances within the
reservoir as well as the tailwater.
The chapter on case studies provides more information on the activities
at the Norris and Upper Bear Creek, projects.
Reservoir Management Strategy—TVA's reservoir management strategy is
directed toward achieving objectives for fish and aquatic life, recrea-
tion (primarily boating and swimming), assimilative capacity for treated
wastewaters, and municipal and industrial water supplies. The water
quality parameters that are addressed include DO, temperature, iron,
manganese, hydrogen sulfide, pathogens, sediment, turbidity, nutrients,
algal concentrations, toxic substances, and aquatic weed growths.
TVA develops water quality management plans for selected reservoirs where
water uses are suspected of being affected and significant interest is
expressed by state agencies, EPA, and the public and private sectors.
Compared with free-flowing streams, reservoirs are very complex water
bodies that are affected not only by the retention of the water behind
the dam, but by operational procedures for the dam and environmental
processes that occur within the water body itself.
Because a number of agencies and water users have various authorities and
responsibilities associated with reservoirs, joint efforts are pursued to
achieve objectives and maximize the benefits of the reservoirs. The
following are the elements that are included in TVA's reservoir water
quality management plans:
1. An issues analysis is performed to define reservoir uses and
objectives, and key issues.
2. A task force is organized for the reservoir that includes State and
Federal agencies, universities, lake users, municipalities, and
industry, as appropriate.
3. Data is collected and analyzed to address key issues and explore
management alternatives.
G-3
-------
4. A plan is developed with recommended actions, including additional
data/analysis needs, point and nonpoint source actions, in-reservoir
enhancement methods, water quality improvement demonstrations, and
citizen involvement.
5. Activities committed to by TVA are implemented.
6. Follow-through actions are reevaluated to determine their
effectiveness in achieving objectives.
Reservoir management plans have been developed and are being implemented
at 5 projects. Special reservoir studies are also being conducted for
other reservoir projects where issues with limited scope need to be
addressed. More information is provided in the chapter on case studies
for the Boone and Guntersville projects.
Watershed Management Strategy—TVA's watershed management strategy is
primarily directed toward the reduction of nonpoint sources of pollution
from agriculture, abandoned mines, urban runoff, and land development
activities. The objective is to improve water quality conditions in
streams as well as reservoirs and tailwateis. Parameters of interest
include nutrients, toxic substances, sediment, pathogens, ammonia, and
carbonaceous biochemical oxygen demand. IVA's efforts in watersheds are
directed towards demonstrating nonpoint source management solutions and
encouraging full-scale implementation actions by other agencies and
private landowners. Even though reservoirs are the ultimate recipients
of nonpoint source contamination from the watershed, cause/effect
relationships between contaminants from watersheds and water quality in
reservoirs are poorly defined.
The elements of TVA's watershed management strategy are as follows:
1. Pertinent information is gathered from all available sources and
issues that need to be addressed are identified.
2. An aerial inventory of nonpoint sources is made.
3. A "delivery analysis" is conducted in which parameters from
contaminant sources to point of reservoir inflow are noted.
4. Water quality and aquatic life (e.g., index of biotic integrity) are
monitored at selected points to better define issues and assess
trends.
5. Nonpoint source control demonstrations are conducted and successful
technology is transferred to appropriale agencies and individuals.
6. Institutional arrangements are developed for full-scale
implementation of priority point and nonpoint source controls.
G-4
-------
POLICIES - WATER QUALITY MANAGEMENT
The TVA system of multipurpose dams with its more than 11,000 miles of
shoreline and 600,000 acres of water was largely completed by the late
1950s. With the completion of the dam construction program, TVA has
turned its attention to managing the reservoir system and promoting the
proper growth, conservation, and management of the Agency's natural
resources. As part of this continuing effort, the TVA Board of Directors
authorized in September 1987 the broadest reassessment in 50 years of the
operating policies of its dams and reservoirs. The central issues being
addressed by the study are whether water quality and recreation should be
added as primary purposes of TVA reservoir operations to the statutory
purposes of navigation, flood control and electrical generation. The
study is being conducted in accordance with the procedures of the
National Environmental Policy Act and will determine the long-term
policies that should direct TVA efforts in reservoir system operations
and river management into the next century. The current schedule calls
for presentation of the final report and Environmental Impact Statement
and the results of public review and comment of the recommendations in
1989.
Water Quality Policies. Codes, and Instructions
TVA has adopted the following operational policies or codes related to
water quality. The most comprehensive policy is contained in TVA Code IX
ENVIRONMENTAL QUALITY. This code states that:
TVA will ensure that its programs, projects, and activities protect
and enhance the quality of the human environment, including the air,
water, and land resources of the TVA region and other areas impacted
by its operations through compliance with applicable Federal, State,
and local laws and related regulations and through the
implementation of more rigorous controls or practices where
practical, beneficial, and cost effective.
TVA's planning procedures provide for early involvement of approp-
riate governmental agencies and the public in decisionmaking
related to activities which significantly affect the quality of the
environment. In addition, TVA prepares reports on the status of
environmental quality in the Tennessee Valley.
The code further states that:
implementation of this policy is a basic management responsibility
and TVA expects all line managers to provide positive environmental
leadership in carrying out agency operations and activities. TVA
conducts monitoring and auditing activities to measure and evaluate
the extent to which environmental quality standards and commitments
are met.
G-5
-------
TVA's specific policy on water quality is contained in TVA Code IX WATER
QUALITY MANAGEMENT. This code states that:
TVA's water quality management activities have as their primary
purpose the restoration or maintenance of suitable water quality
throughout the Valley to permit optimum use of surface and ground
waters for municipal, industrial, and agricultural water supplies;
for propagation of fish and wildlife; for aesthetic enjoyment; for
water-contact recreation; and for future development of streams and
reservoirs in the public interest. The goal is to keep all waters
clean and free of pollutants It applies existing or, where none
exist, develops water quality criteria for use in its water resource
development projects; where there is no practical alternative to
disposal of hydrowastes into streams, it encourages the imposition
of regulatory controls designed to obtain the highest degree of
waste treatment available under existing technology, within
reasonable economic limits, to protect the Valley's surface and
groundwater resources.
TVA's policy also recognizes that:
its statutory program of impoundment and streamflow regulation
produces significant changes in streamflow and in the physical,
chemical, and biological characteristics of affected waters in the
Valley. Overall, these alterations are highly beneficial, but some
changes have adverse water quality effects and may reduce the
capacity of the streams to assimilate wastes. TVA conducts studies
and field investigations to identify and evaluate the interrelation-
ship of water resource development and water quality, utilizing
research findings of other agencies and institutions to the maximum
extent feasible. Consistent with the primary purposes for which TVA
projects are operated under the TVA Act, TVA operates its system of
reservoirs to minimize adverse water quality effects and to give due
account to State-designated downstream uses. TVA controls or treats
wastes from its own operations in accordance with the stated primary
purpose of this policy, cooperating with Federal and State pollution
control agencies. In the administration and disposal of TVA lands
and in the licensing or regulation of water-use facilities con-
structed on TVA reservoirs, TVA incorporates pollution control
provisions, including requirements for using best management
practices to control nonpoint-source pollutants, in deeds, leases,
licenses, permits, and other documents as appropriate.
In addition to these policies, water quality considerations are also
included in TVA Code IX FLOODPLAIN MANAGEMENT AND PROTECTION OF
WETLANDS, TVA Code IX STREAM MODIFICATION, and TVA Code XII RESERVOIR
OPERATION. Of particular interest to this study is the policy which
specifies how TVA regulates reservoir levels and streamflow (where
consistent with statutory purposes) to:
o Minimize detrimental water pollution effects and produce water
quality benefits for the public health and public use of the
reservoirs, such regulation to be provided without accepting
liability for the regulation and without relieving the polluter from
full responsibility for such pollution.
fi-6
-------
Make the greatest possible contribution to the sustained control of
hazards to health in the Tennessee Valley region and to maintain
standards for control of mosquitos which are fully as effective as
those required for privately owned river impoundments under
prevailing public health regulations.
Enhance fisheries management and other recreation uses of these
waters. Under these conditions, it attempts to provide pool levels
favorable for recreational uses and to minimize fluctuations during
the prime recreation season and spawning season on tributary
reservoirs. Also, wherever primary operating requirements permit,
it attempts to regulate sufficient discharges from dams to provide
for boating and fishing activities on certain streams.
Minimize possible adverse effects on propagation of fish life and
enhance the value of Valley fisheries.
Provide a water supply for domestic, industrial, or agricultural
uses.
Accommodate individual navigators, ferry operators, farmers on river
islands and shore lands, and others who may otherwise suffer
discomforts of inconveniences or may need special flow regimes to
accommodate short-term scientific research or management
demonstrations. Such regulation does not entail acceptance of
responsibility for serving such incidental comforts or conveniences
or infringe on TVA's acquired right to flood the reservoir margins
whenever or wherever required for the major purposes of this program
without regard to any secondary land use which may have been
undertaken subject to such flowage rights.
RECOMMENDATIONS
As TVA reported to the U.S. Senate's Subcommittee on Environmental
Pollution in 1985, 7 of the 10 most critical water quality problems in
the Tennessee Valley are related to nonpoint pollution. TVA said then
and still believes nonpoint pollutants are not being adequately
addressed, and the Clean Water Act goals of "fishable, swimmable" waters
will not be met in many parts of the Valley unless additional point and
nonpoint pollution control measures are implemented. Damage is not only
occurring to natural resources but public health, economic development
efforts, water supplies, and the Nation's $1.53 billion investment in the
TVA reservoir system are affected.
Both in the region and nationally, there is increasing public recognition
of the interrelationships between land-based activities, surface water
pollution, and groundwater contamination. Many of the most intractable
environmental challenges currently facing the nation involve impacts that
involve two or more of these areas. The major constraints to addressing
these problems are limited resources and the difficulties inherent in
dealing with multimedia impacts through existing medium-specific
programs. Recent legislation offers some encouragement. The
conservation title of the Food Security Act of 1985 established major
national programs (Conservation Reserve, sodbuster, swampbuster) that
G-7
-------
concurrently address questions about agricultural overproduction, soil
erosion, and wetlands destruction. The Water Quality Act of 1987
provides States with new and more broadly based incentives to define and
address the nonpoint pollution issues that chronically impede attainment
of the goals of the Clean Water Act. Substantial funding is authorized
for both these initiatives. Neither can be expected to yield instan-
taneous cures for the longstanding national and regional issues that they
address. Both, however, offer hope for major improvements over the long
term if resources are allocated to implement these programs.
EPA maintains it has sufficient authority and responsibility to oversee
the preparation and implementation of measures to improve water quality
nationally. TVA supports EPA and the States' efforts to use their
authority and to make available the resources necessary to address water
quality problems. Our experience with water quality in TVA's reservoir
system indicates that no major improvement will occur until point and
nonpoint source controls are considered and implemented as part of the
overall water quality improvement strategy. To do this in the most cost
effective way, innovative approaches may be necessary, such as allowing
States to use more of their allocation of the existing $2.4 billion
construction grants or revolving loan program to fund cost-offective
nonpoint pollution control projects. In recent studies with EPA and the
State of Tennessee, TVA has found that innovative solutions can be more
cost-effective than building traditional advanced wastewater treatment
plants. If the Tennessee Valley and the Nation are to achieve Clean
Water Act goals, the States should be provided with the flexibility to
make tradeoffs in point versus nonpoint abatement, once secondary
treatment requirements are met, and in targeting clean water funding to
where it is most cost-effective.
As questions arise with regards to eutrophication, toxics, and sediment
buildup in streams and reservoirs, it becomes more apparent that water
resource interests need to take a more active role in the triennial
review of State water quality standards. Areas of particular interest
include:
1. Stream Use Classification—TVA believes that lake and reservoir
water quality can be managed better if a distinction is made between
free-flowing streams, lakes, impoundments, and tailwaters. TVA has
offered to work with the States and EPA in developing criteria and
standards for impounded waters and tailwaters that recognize
the hydrologic and related physical, chemical, and biological
differences between these water bodies.
2. Impaired Waters Designation(s)—Additional measures are needed to
control pollutants that are causing cultural eutrophication, toxics
impacts, and sediment buildup in streams and reservoirs. Water
bodies with particular use impairments should be identified
categorically as impaired waters, e.g., "Nutrient Sensitive Waters",
"Erosion Sensitive Watershed." Following this designation, water-
shed or areawide pollutant control programs should be developed that
emphasize the control or treatment of point and nonpoint source
contributions.
G-8
-------
Taken together these policy issues strongly suggest the need to reexamine
resource management and pollution control strategies. Better integration
of these programs has not only become an important policy issue in the
Tennessee Valley, but nationally as originally recognized by the Water
Resource Council (WRC) in its "Second National Water Assessment,"
10 years ago.
Water Quality Research Recommendations
Additional activities beyond those now carried out by the Federal
government are needed to adequately address water quality concerns. The
following are critical research needs that have been identified by TVA's
environmental engineers and scientists.
Water Quality Criteria—Existing criteria and standards are essentially
based on "fixed" concentrations for a single constituent. For example,
EPA's dissolved oxygen (DO) criteria document has progressed from
recommended minimum concentrations for selected levels of protection to
the point where they consider the type of resource to be protected, i.e.,
whether it is a warm or cold water fishery, life stage of the fishery
resource and substratum or habitat condition. This is a significant
improvement, but the criterion remains a single stream value and thus
fails to fully address the needs of resource managers. In highly
regulated rivers like those managed by Federal water resource agencies,
water quality criteria and standards are needed that take into consider-
ation (1) the interaction between multiple constituents, (2) the rate of
change that could result in unacceptable environmental conditions, and
(3) the frequency and duration of exposure of target organisms.
Total Dissolved Gases (TDG)—TDG is another issue deserving attention
because they have been observed at relatively high concentrations during
flood control operations at dams on the Tennessee River and elsewhere.
However, the effects on downstream fisheries have not been adequately
assessed to determine if these elevated concentrations are significant.
Hydrogen Sulfide—Hydrogen sulfide is being detected in the releases from
water resource projects. This constituent is extremely toxic at very low
concentrations (i.e., 2 ug/1). Limited information is available on the
occurrence of hydrogen sulfide in reservoir releases because present
analytical methods are not sufficient to detect toxic levels. Infor-
mation about the rate of oxidation of hydrogen sulfide is also limited,
but it can vary considerably because of naturally occurring catalysts
that result in rapid oxidation; however, in the absence of such
catalysts, hydrogen sulfide can persist downstream for several miles
before it is oxidized to a nontoxic state. Hence, additional field data
are needed on the occurrence and persistence of hydrogen sulfide.
Temperature—The releases from some reservoir projects can contain water
that is too cold for desirable fish growth in the downstream tailwater,
and in some cases spawning activities can be significantly affected by
temperatures in the reservoir releases. Work at the Bureau of
Reclamation's Flaming Gorge project has demonstrated one method for
improving temperatures. However, less expensive means may be available.
G-9
-------
For example, surface water pumps may be feasible to mix the upper layers
that have warmer temperatures with the colder layers in such a way that
these two water sources are blended before being released through
low-level outlets.
Minimum Flows—There are two key issues related to the provision of
minimum streamflows below hydropower projects: (1) the quantity of flow
required to achieve fishery and aquatic life objectives and (2) the means
of providing such minimum flows, i.e., the use of "continuous" flows
versus the use of "pulsing" flows that are more desirable from a
hydropower viewpoint. Concerning the first issue, there is a need to
better predict the effects of various streamflow levels on aquatic life.
There are several available methods for predicting the effects of flow
quantity on aquatic life, but their accuracy is suspect. In addition,
because recreationists are the most immediate beneficiaries of tailwater
improvements, better methods are needed to determine from their viewpoint
which flows are desirable.
Reservoir Research Recommendations
Reservoir Models—Even though significant progress has been made in
developing methods for analyzing reservoir water quality, new
mathematical modelling methods in two dimensions have been applied to
only a few reservoirs. Additional application of such models is
critically needed to more fully understand the water quality changes that
occur within a reservoir. Stream models have been in existence since the
1920s, and estuary models have been available since about 1970, whereas
2-D reservoir water quality models have only become available during the
1980s. These models have significantly improved the ability to assess
and predict such things as DO and the mechanisms that consume oxygen
within the reservoir; however, even with recent developments with the 2-D
model, the effects of inflow water quality on reservoirs has still not
been clearly defined. There is a need to identify the reservoir water
quality effects of various sources of contamination within watersheds,
particularly nutrients from point and nonpoint sources. The direct
linkages between these sources of contaminants and effects on overall
water quality in receiving impoundments have not yet been defined
quantiatively . Sound assessments of the mechanisms that affect water
quality within reservoirs are also needed to determine the effects of
changes resulting from reservoir operations. For example, in the TVA
system, consideration is being given to maintaining summer pool levels
through the end of October each year, and sufficient modeling has not
been conducted to predict any resultant changes in water quality.
Modeling is also needed to predict water quality in large embayments to
reservoirs. From a biological standpoint, embayments are extremely
important, yet these bodies of water are not usually considered in
reservoir models.
Hypolimnion Water Quality—If any resultant flow through a project is
substantially reduced for a portion of the year, it is not possible at
this time to predict the effects on existing uses. Water quality issues
such as low DO in the hypolimnion can result and, in turn, contribute to
the potential release of hydrogen sulfide as well as manganese from the
sediments.
6-10
-------
Technology Development Recommendations
Aerating Turbines—Considerable research and development activities have
been initiated on various methods to increase water quality in hydropower
releases. Several of these methods have been evaluated over the past
five years on full-scale operations. However, most of them are retrofit
technologies. There is a national need to develop new designs for
turbine replacement wheels in the next decade. The objective would be to
incorporate aeration capabilities in the turbine system in such a manner
that minimum impacts would occur to power generation as well as
operations. A large number of existing turbines in the United States
will be replaced over the next two decades; therefore, this proposal
offers a significant opportunity to enhance the dissolved oxygen
concentrations in turbine releases nationally.
Habitat Modifications—Habitat modifications within tailwaters could be a
less costly alternative to present minimum flow strategies. Such
modifications would be one-time expense items as opposed to annual
operational expense items. This concept can also be extended to certain
areas within reservoirs. It is believed that some water uses can be
significantly enhanced by improving localized areas of the reservoirs.
One example is the embayment enhancement option. Another example is the
provision of refuges for fish that may require the higher dissolved
oxygen levels found in cool waters, e.g., a submerged reservoir can be
constructed within a reservoir to contain high-density cold water and
provide habitat for striped bass.
Operational Monitoring Recommendations
Reservoirs with long retention times and low nutrient levels can develop
low dissolved oxygen concentrations in the hypolimnion, triggering the
release of hydrogen sulfide, iron, and manganese. In some cases, such
occurrences may be avoided by changes in outlet levels or release
patterns. In other cases, they may be aggravated by changes in release
patterns that result from reservoir operations for other purposes, e.g.,
raising pool levels for longer durations during the recreation season can
cause some significant changes in reservoir release patterns. Water
quality within reservoirs may be affected by changes in watersheds and
year-to-year variation in hydrology in the form of annual
rainfall/runoff. All of these interactions are complex and changes in
the water quality of the reservoir cannot simply be attributed to one
factor. Therefore, it is necessary to conduct appropriate monitoring and
to determine the cause-and-effect relationships for processes within
reservoirs. Mathematical models calibrated to field data offer one of
the best tools for exploring reservoir operations and evaluating
management alternatives. Reservoirs are much more complex than
free-flowing streams, resulting in three-dimensional variations in water
quality as opposed to single-dimensional.
G-ll
-------
G-12
-------
Iron
Manganese
Low D.O.
Sill f iCiP
Turbidity
Low Temperature
High Temperature
Dissolved Solids
Metal Contaminant
Organic Contaminant
High Nutrients
Algae
Sediment Accum.
Pool Elev. Fluct.
Shore Erosioi
Taste 5 Odor
pH/Acidity
Bacteria
Parasites
Organics-Fish Flesh
Metals-Fish Flesh
Continuous
Seasonal
Infrequent
No Problem
No Data
40 60
Percent
80
100
Iron
Manganese
Low O.Q.
Hydrogen Sulfide
Turbidit
Low Temperature
High Temperature
Dissolved Solids
Metal Contaminant
Organic Contaminant
High Nutrients
Algae
Macrophytes
Sediment Accum.
Pool Elev. Fluct.
Shore Erosi
Taste S Odor
pH/Acidity
Bacteria
Parasites
Organics-Fish Flesh
Metals-Fish Flesh
Severe
Significant
Minor
No Impact
No Data
\-\<-VSZ-V.-'SSSSSSSSSSSS^^^^
Figure 2 • Rate of recurrence (upper graph) and user impact assessment for TVA
reservoir pools for the indicated parameters.
G-13
-------
Iron
Manganese
Low 0.0.
Hydrogen Sulfide
Turbidity
Low Flow
High Flow
Fluctuating Flow
Low Temperature
High Temperature
Fluctuating Temp.
Dissolved Solids
Metal Contaminant
Organic Contaminant
Gas Supersaturation
High Nutrients
Algae
Macrophytes
Shore Erosion
Taste S Odor
pH/Acidity
Bacteria
Parasites
Organics-Fish Flesh
Metals-Fish Flesh
Iron
Manganese
Low 0.0.
Hydrogen Sulfide
Turbidity
Low Flow
High Flow)
Fluctuating Flow
Low Temperature
High Temperature
Fluctuating Temp.
Dissolved Solids
Metal Contaminant
Organic Contaminant
Gas Supersaturation
High Nutrients
Algae
Macrophytes
Shore Erosion
Taste 8 Odor
pH/Acidity
Bacteria
Paras ites
Organics-Fish Flesh
Metals-Fish Flesh
Continuous
Seasonal
Infrequent
No Problem
No Data
Percent
Severe
Significant
Minor
No Impact
No Data
B5QH
Percent
Figure 3 . Rate of recurrence (upper graph) and user impact assessment for TVA
reservoir tailwaters for the indicated parameters.
G-14
-------
APPENDIX H
U.S. BUREAU OF RECLAMATION"S ASSESSMENT
SUPPLEMENT
EPA solicited comments from the States and Federal agencies for
this report to Congress. Federal and non-Federal organizations with
responsibility and activity involving impoundments were included. At the
January 5, 1988, meeting the Tennessee Valley Authority (TVA), Bureau of
Reclamation (USBR), and Corps of Engineers (COE) were invited to provide
an agency assessment of water quality at their dams. Dams from these
three agencies represent a wide range of geography, climate, and
operational situations.
This chapter presents the results of these agency assessments. Tt
utilizes an alternate approach to that used in the main body of this
report. The three agencies felt there were important limitations in
trying to perform a national overall assessment of the status of water
quality at dams without detailed appreciation of regional effects. These
limitations center on three main points:
1. The basic assumptions of what constitutes a dam-related water
quality problem are over-simplified. The prime example of this
is the assumption that thermal stratification ir any reservoir
with low level outlets automatically causes downstream problems
of low dissolved oxygen, dissolved iron and managanese, and
hydrogen sulfide. This is only the case where stratification
lasts long enough to exhaust the supply of dissolved oxygen (DO)
in the hypolimnion, and it is a function of the initial DO
levels, the biochemical oxygen demand in the hypolimnion, and the
hydrodynamics of flow through the bottom of the reservoir to the
outlets. An examination of the USBR assessment presented below,
for example, will show that problems of low DO, dissolved iron
and manganese, and hydrogen sulfide are relatively rare at Bureau
dams, although summer thermal stratification is nearly universal
and low level outlets are not uncommon.
2. The analytical tools used are too general and, in some cases,
flawed. The use of a Froude number calculated on the basis of a
flow equal to the spillway capacity to estimate stratification
potential is considered to have serious limitations. The Oak
Ridge study cited in the evaluation of low DO levels in power dam
releases is specific to the southeastern United States, and it is
definitely not representative of USBR and COE power dams in the
western arid region. An evaluation of downstream effects on the
basis of a comparison of mean annual upstream and downstream data
is not meaningful, in that it AS not the annual average, but the
seasonal variation that is important in an aquatic environmental
parameter like xlow or temperature. The r-enrichment analysis
uses statistical models that have been shown to be too general
for specific field situations.
3. The data base (STORET) used in the analyses is not sufficiently
comprehensive in its representation, has no control on the
H-l
-------
quality or comparability of data on a case by case basis, and is
arguably biased toward problem cases.
Given these basic limitations, the agencies felt that conclusions reached
on the basis of this overall evaluation are likely to be unreliable and
unrepresentative .
As a result of these concerns, the three agencies were each asked to
prepare a short assessment of the status of water quality at their dams.
This gives an analysis of rather specific situations, with TVA dams
located in a mature, we 11 -developed and rather industrialized extended
river basin; the COE dams are concentrated in the industrialized areas of
the Southeast and tue Ohio River Basin, coastal areas, the Pacific
Northwest, and along main stem navigable rivers; and USBR dams are all in
the western states.
In making their assessment, it was felt that a better approach to this
subject is to, first of all, realize that dam-related water quality
problems are much more s ite- specif ic than general. The case studies
included in this report as Chapter IV make this point abundantly clear,
while the agency assessments presented below further emphasize the
variety of conditions.
The three agencies agreed that in t^is situation, the best way to assess
the nature, extent, and severity of impact of dam-related water quality
problems is to survey «s many as possible 01 the existing impoundments
and ask those directly responsible for their daily operation what the
problems are and what impacts are being felt. They feel this subjective
approach is more likely to yield an accurate assessment of actual
conditions than is an indirect approach that relies on over-simplified
assumptions, a "data-rich but information-poor"*- data base, and the
"illusion of technique"^ of generalized models.
To this end, the three agencies adopted a questionnaire originally
developed by Kennedy, Gunkel, and Gaugush ot trie U.S. Army Corps of
Engineers Waterways Experiment Station (WES) at Vicksburg, Mississippi,
as a uniform instrument for collecting dam-related water quality
information on their respective projects. A copy of the questionnaire,
with its instruction sheet, is attached at the end of this chapter
(Attachment A) .
Basically, the questionnaire presents field personnel with a
comprehensive list or water quality attributes, and asks them to
subjectively rate each attribute in the tributary, pool, and tailwater of
1 Ward, R.C., J.C. Loftis, and G.B. McBride. 1986. The "data-rich
but information-poor" syndrome in water quality monitoring. Environmental
Management, V.10, N.3, pp.292-297.
2 Behnke, R.J. 1987. The illusion 01 technique and fisheries
management. Proceedings of the 22nd Annual Meeting of the Colorado-
Wyoming Chapter of the American Fisheries Society, March 11-12, 1987,
Laramie, WY, pp.48-51.
H-2
-------
each reservoir on the basis of the extent to which this attribute is a
problem, the level of impact of the attribute on user benefits, and the
reliability of the data upon which the rating is being made. A computer
program, also developed at WES, compiles the data from the questionnaires
into a SAS data file. Statistical analyses can then be performed on the
various attributes and their ratings using the SAS software. In the
short time allowed for this study, the three agencies limited their
efforts to frequency analyses of each attribute's extent and impact in
the pools and tailwaters of their reservoirs. Given more time, these
analyses could be extended to reservoir tributaries and to the reported
quality of available data. Finally, available quantitative data from
agency files and STORET could also be added to the analyses, if there
were a desire to go beyond an assessment of the situation to an
investigation of parameter interrelationships.
The agencies feel the present limited analysis does, however, give an
accurate picture of the known extent of given water quality conditions
across a broad range of geography, climate, and project operating
criteria, along with an assessment of the perceiveu impacts of these
conditions on user benefits. All of this information is presented, not
from the perspective of a simplified general overview of the way
reservoirs should theoretically behave, but from the point of view of
field personnel who are directly responsible for daily dam operations and
the delivery of promised project benefits.
The remainder of this chapter contains the results of each agency's
assessment of water quality related to its dams, beginning with the
Bureau of Reclamation, and continuing through the Corps of Engineers and
TV A. Each agency's section is organized as follows:
1. statement of policies and procedures followed by the agency in
the development and management of water resources.
2. assessment of water quality with respect to the agency's dams.
3. recommendations for policies and practices for water quality
aspects of dams and important related scientific and research
needs.
H-3
-------
THE BUREAU OF RECLAMATION
The Bureau of Reclamation of the U.S. Department of the Interior Is
responsible for the development and conservation of the Nation's water
resources In the Western United States.
Reclamation was formally begun 86 years ago with the passage of The
Reclamation Act of 1902. In most areas of the 17 Western States,
which constitute the area served by the Reclamation program, less than
20 inches of moisture falls each year. However, several Important
rivers, fed mainly by the melting snow packs in the mountains, flows
through these States. A basic function of Reclamation Is to harness
these streams and to store their surplus waters in times of heavy
runoff for later use when the natural flow is low.
Reclamation's original purpose "to provide for the reclamation of arid
and semi arid lands in the Vest" today covers a wide range of
Interrelated functions. These Include providing municipal and
Industrial water supplies; hydroelectric power generation; Irrigation
water for agriculture; water quality Improvident; flood control; river
regulation and control; fish and wildlife enhancement; outdoor
recreation; and research 1n atmospheric water management and
alternative energy sources, such as wind and solar power. Reclamation
1s also a primary source of research In design, construction, and
development of materials used in water management structures.
Reclamation programs most frequently are the result of close
cooperation with the U.S. Congress, other Federal agencies, States,
local governments, academic Institutions, water-user organizations,
and other concerned groups.
H-4
-------
As of September 30, 1986, the Federal Investment In completed
Reclamation project facilities totaled $8.7 billion. This Investment
Includes $1.7 billion In specific Irrigation facilities. $1.7 billion
1n electric power facilities, $0.3 billion in nunlclpal and Industrial
facilities, and $5.0 billion In multipurpose and other facilities.
About 80 percent of this total Investment is reimbursable to the
Federal Treasury. Reclamation project facilities In operation during
1986 included: 355 storage reservoirs; 254 diversion dams; 15,804
miles of canals; 1,382 wiles of pipelines; 276 miles of tunnels;
37,263 miles of laterals; 17,002 miles of project'drains; 240 pumping
plants; 50 hydroelectric powerplants; and 288 circuit miles of
transmission lines. Nearly 10 million acres of Western farm land
receive full or supplemental irrigation, over 20.5 million people
receive municipal and Industrial water, 53.2 million visitor days of
recreation are recorded annually, and 13.8 million kilowatts of
Installed hydroelectric power capacity exists.
Completed water service facilities are transferred to local water user
organizations for operation and maintenance as soon as the
organizations become capable of assuming these functions. Reclamation
operates and maintains hydroelectric powerplants and some water
storage and supply works on multi-purpose projects. Of the 235
operating projects or units providing service in 1986, 172 were
operated entirely by water user organizations, 39 were operated
jointly by water user organizations and Reclamation, and 24 were
operated entirely by Reclamation.
H-5
-------
Unique Features of the Bureau of Reclamation
Among Federal programs, the Reclamation program 1s unique 1n several
respects. First. It 1s regional In nature, 1t 1s United by
Reclamation Law to the 17 contiguous States lying wholly or partly
west of the 100th meridian. Second, the Reclamation program has from
the beginning, in contract to other Federal public works programs,
been based on the principle of repayment by direct beneficiaries
(water users, conservancy districts, power customers, etc.) As law
and policy were revised and broadened over the years to accommodate
multiple purpose projects, some program costs became nonreimbursable.
The total amount repaid through fiscal year 1986 is over $2 billion,
representing 23 percent of completed plant-in-service investment of
(8.7 billion. Repayment of costs is greater in Reclanation than in
any other Federal resource development program.
Another significant feature of the Reclamation program is the economic
analysis that accompanies any proposed project. This evaluation
consists of (1) plan formulations studies to determine optimum size
and mix of features; (2) benefit-cost analysis aimed at the question
of economic justification for the project; (3) cost allocation studies
to assign costs to the various reimbursable and nonreimbursable
functions; and (4) repayment analysis to determine if the project
reimbursable costs can be repaid in accordance with requirements of
law and policy.
H-6
-------
Reclamation's Salinity Control Program
A large body of law surrounds the Reclamation salinity control program
and specific projects. The individual projects are subjected to
congressional hearings, authorization and funding. In 1972, an
amendment to the Federal Water Pollution Control Act,
Public Law 92-500 (known commonly as the Clean Water Act) sets forth a
public policy for restoration and maintenance of water quality
'-,'
standards, including benefical use designations and numeric salinity
criteria. In June 1974, Congress enacted the Colorado River Basin
Salinity Control Act, Public Law 93-320, which directed the Secretary
of the Interior to proceed with a program to enhance and protect the
quality of water available in the Colorado River for use in the
United States and the Republic of Mexico. Reclamation is leading a
strong and aggressive salinity control program launched by these acts
of Congress.
The overall approach in meeting the salinity standards for the
Colorado River 1s to prevent salt from entering and nixing with the
river's flow. A number of agricultural, point, and diffuse sources of
salinity have been identified in the Colorado River Basin.
Reclamation's salinity control program will Implement controls at
those sites which contain salt sources that can be intercepted and
prevented from entering the river at least cost.
The estimated salinity control program potential, through existing and
planned Reclamation projects coordinated with State and other Federal
agencies, h^s a projected total salt reduction in the Colorado River of
2.06 billion tons per year.
H-7
-------
The Changing Nature of the Reclamation Program
The shift in emphasis away from agriculture and toward municipal and
industrial water, recreation, and fish and wildlife enhancement has
been the trend for the past two decades. And as fewer large federally
financed water projects are constructed, Reclamation is responding by
developing alternative means of supplying water through Improved
system management, joint use of surface and ground-water supplies, and
re-evaluating priority of use.
Management of water resources remains a critical necessity for the
arid west. Reclamation is examing opportunities to Increase water and
power operating efficiencies and to Identify opportunities for
non-Federal partnerships for water resource development. The goals
and the objectives of Reclamation are:
* Continue to provide the world's best engineering and
construction expertise for water resource projects.
Projects that are justified by local area need, support
and funding.
- Improve efficiency by properly maintaining; upgrading
and enhancing existing water and power resource facilities.
- Emphasize non-structural water efficiency and conservation
alternatives.
- Maintain the technical water development capability to
assist State, and local governments, as well as other
non-Federal entities.
H-8
-------
Inprove joint tnd nultlple Und tnd water resource use,
Including perfecting recharge and conjunctive use programs
for ground Mater.
Continue leadership 1n establishing effective and
acceptable water quality programs and standards.
Maintain the safety of the Nation's dams, reservoirs
and waterways.
Develop remedies for water-related hazardous waste
problems.
H-9
-------
ASSESSMENT OF WATER QUALITY CONDITIONS IN BUREAU OF RECLAMATION
RESERVOIRS AND TAILWATKBS
Information on water quality conditions and user impacts for all Bureau
of Reclamation (USSR) storage reservoirs and tailwaters was solicited by
distributing a questionnaire (attachment A) developed at the Corps of
Engineers Waterways Experiment Station (WES) to all six Bureau regional
offices. All regions responded, and a total of 250 questionnaires were
returned. The geographical distribution of this response by state is
shown in figure 1. Since there are approximately 349 USER storage
reservoirs in eighteen western states (USBR 1986), this response
represents nearly 72% of the total.
Information obtained on the frequency of occurrence of various water
quality conditions in USBR reservoirs and their impact upon user benefits
are summarized in figures 2 and 3, respectively. Tailwater conditions
and their impacts are shown in figures 4 and 5, respectively.
What is most immediately apparent in these four figures is that in an
average of 54% of the reservoir cases and 59% of the taiiwater cases,
there are no data upon which to make an evaluation of the conditions or
their impact on user benefits. Further, if the assumption is made that
the reason that no questionnaires were received on the other
approximately 99 USBR reservoirs listed in the 1986 statistical
compilation is that no data were available, then the total percentages
of no data would rise to 68% for reservoirs and 71% for tailwaters.
Water quality data are usually only collected on a particular project
when some problem is noted or suspected, or when some change in the
structure or operation is contemplated. Consequently, the picture of
water quality conditions in Bureau reservoirs and tailwaters given by the
available information is probably somewhat skewed toward those situations
where some problem is perceived or an impact is felt. The following
assessment is, therefore, probably rather conservative.
Data sources for figures 6 through 9 are the same as figures 2 through 5,
except that the percentages are based only on those cases for which data
are available. The total number (N) of reservoirs or tailwaters with
information on a particular condition is iit»i.ed on the right side of each
graph. These numbers range from 56 to 149, or from 16% to 43% of USBR
storage reservoirs.
Figures 6 and 7 suggest that tne iuain conditions affecting user benefits
in Bureau reservoirs are drawdown, pool fluctuation, turbidity, sediment,
and shore erosion. None of these are water quality conditions per se,
but arise from the way water storage reservoirs are operated in an arid
climate, where spring snowmelt or winter rains are the major source of
runoff, and drawdown is continuous throughout the long dry season.
Drawdown was rated as having a severe impact on user benefits in six USBR
impooundments, and a significant impact in 33 others, out of a total
sample of 107 reservoirs with information available. Thus, the
cumulative percentage of reservoirs with data in which drawdown was rated
H-10
-------
as having at least a significant impact on user benefits is 36.4%.
Corresponding cumulative percentages for the other four conditions are:
pool fluctuation 35.3%, turbidity 13.2%, sediment 12.8%, and shore
erosion 10.2%. The last condition, shore erosion, had no severe impact
ratings out of a total of 108 impoundments ratea.
The second ""~st important set of reservoir water quality conditions
affecting user benefits are related to eutrophication: algae, high
nutrients, low dissolved oxygen, and taste and odor problems. Cumulative
percentages of reservoirs with data where these conditions were rated as
having at least significant impacts on user benefits were: algae 16.5%,
high nutrients 15.2%, low dissolved oxygen 13.5%, and taste and odor
problems 11.8%. There were no severe impact ratings for high nutrients
or taste and odor problems, however.
Although iron is often present in Bureau reservoirs (figure 6), it was
rated as having at least a significant impact on user benefits in only
4.4% of the rated reservoirs (figure 7). In fact, it should be noted
that only drawdown and pool fluctuation were perceived as having really
significant impacts on reservoir user benefits (figure 7).
Tailvater conditions and user impacts are depicted in figures 8 and 9,
respectively. Here again, the major impact-producing conditions seem to
cluster around the mode of operation of water supply reservoirs in an
arid region: high flow, low flow, turbidity, and high temperature. Of
these, high flow was rated as having significant user impacts in 21.1% of
the tailwaters with data, but in no case was it rated as having a severe
impact. The other three conditions were rated as having at least a
significant impact in 13.2%, 11.6%, and 10.9% of the rated tailwaters,
respectively. Taste and odor problems were rated as significant in 13.3%
of the tailwaters with available data, uut were not considered severe in
any case.
CONCLUSIONS
Three main conclusions may be drawn from this assessment of water quality
in USER reservoirs and tailwaters.
1. Data on water quality conditions and user impacts for about 30%
of the Bureau's approximately 349 storage reservoirs are
presented here. Since this group is somewhat biased toward
projects where problems have been noted or suspected, the
resulting assessmenr may be a bit pessimistic.
The main impact-producing conditions identified for Bureau
reservoirs are drawdown and pool fluctuation, which are
associated with a mode of operation that combines rapid spring
filling of reservoirs with a steady withdrawal of water to
satisfy irrigation, municipal, and industrial demands during the
long dry season. These two conditions were rated as having at
least significant impacts on user benefits in 36.4% and 35.3%,
respectively, of the reservoirs with available data.
H-ll
-------
High flow was the main impact-producing condition noted in USER
tailwaters, probably reflecting the high spring inflows and
spillway discharges of the mid-19801 s. This condition was rated
as having a significant impact on user benefits in 21.1Z of the
rated tailwaters, but in no case was the impact rated as severe.
The next two impact-producing conditions cited were low flow and
taste and odor problems, each witn a cumulative rating of at
least significant in about 13% of tv>«» tailwaters with available
data. There were no severe ratings for tailrace taste and odor
problems, however.
REFERENCES
U.S. Bureau of Reclamation. 1986. Statistical Compilation of
Engineering Features on Bureau of Reclamation Projects. USER,
Denver, CO, 167 pp.
H-12
-------
0?
C3 rv
O
H-13
-------
o
it
O t/l
oc ui
a: to
OCD
Odd
tu 2
«>io
UJ
a:
z>
o
d
-------
cnm
z
-------
o
in
CM
II
2^
vf
UJ
c/>
UJ
o:
a:
m
en
_
o
to
a:
l
o
E
o
•z.
o
o
Or
a:
UJ
I
CL
13
O
*tf
0
i
€>
e
V»
5
8
cu
o
l—>
o
oc
o
X
UJ
H-16
-------
o
in
CO
CC
i
to
UJ
2
UJ
CD
to
ID
2-
O
O
DC
u.
o
o
I
If)
UJ
a:
Z5
O
O
O
_ in
LJ
«I
a.
O
m
on
o
01
iO
CM
Bill's
C\£Q o
0 v> « O I Q_
o o *>
T J- •*-*
2. X. OT
UJ
cc
O
-------
oicjico o «cj- n en o
o O «—t
3
Ixl ~
II
UJ
UIC/)
oo
uicr
Eg
Q-
ru
o w o TJ zi
~ c * ~ -^
o
o>
8
0)
"o
CO
to
Q
JOWOTWOT-^CCCt^.DW
<£ o«c5cx: E"5 S O5*oo
•«O»*- Q-'D*^uj"o^-O*»
r-»3 P*>Po?2>mp
X
LU
o»
X
H-18
-------
"§
OLU
Q_ LU
LJ
O
iZ
CU
«J,
*•*
LJ vt
o:
o
-------
to
UJ
I
to
oc
UJ
to
UJ
o:
cc
m
to
to
a:
UJ
I
to
o
5
a
o
o
« a. to «> «> c
ft- ft. ^K ^P>% *••» « * ^M.
ca
o
on
CL.
€»
™ p
n s
Ji
o
o:
UJ
I-
X
UJ
H-20
-------
UJ
(T
§
o:
£
2
I
.
LJ
5
m
o:
UJ
(O
ID
_
O
o:
i
O f^ o vo CM «—« O r-- o —• o *=r LT> 001 CTI o CT> G"> in u~> UD
oo •—< co •—< cr> 01 en o o o ^r »—« cr> »-i .—i o 01 00 *£> r-- ro cr> in
I
II
I «*
I *••
_J «o
O)
ui
o:
^
o
C t> O *> >» > 3t ^4>«>CLOT«COCW«>CO-JiC)r>sOW
o 2 O ^2»— o o o ^ ^ P ^ "5 ^ o ^~ ~ o v 5; o .2 •— *c ^
3 O 0>C - -
*— _J -4: ^S CL Q.
/
o
%
5 E E --g >
^J Q^ ^J _L O
"o *~ *~ 3 co
§•2
CO
« ox
-c"§
oo 5
H-21
-------
RECOHMEHDATIONS
The following recommendations represent ideas that should help
Congress better evaluate water quality (WQ) issues relating to
dams and impoundments. All of the ideas are more or less related
and will have direct implications for all of the water resource
management agencies (USER, COE, TV A, referred to as the Agen-
cies). The principal policy implications of these recommenda-
tions are that the Agencies will need to shift more priority
towards WQ issues anu cooperate to accomplish technical goals.
Thus coordination between the Agencies will be a primary require-
ment of these initiatives. The section following the recommenda-
tions provides a suggested organizational structure for these
activities.
1. Avoid establishing uniform standards for WQ of impoundments
and dan discharges that ignore regional differences between res-
ervoirs and the often conflicting demands of water users.
A cursory examination of the WQ summaries seen in Figures 2
through 9 reveals that the problems identified for reservoirs in
the arid West are very different from those in the East. Nutri-
ent loading and eutrophication are the principal concern for TVA
reservoirs that are located in a much more industrialized and
densely populated area. In the West, pool fluctuations and flow
are the major concern for USER reservoirs.
Most projects must plan the O&M of their reservoirs to balance
the WQ (or water supply) demands of many user groups, each with
often contradictory requirements. Each reservoir has a unique
set of water uses and range of WQ conditions, and compromises are
often necessary since WQ cannot alway» be maximized for every
user. Rigid WQ standards would ignore the complexity of reser-
voir O&M vs. user WQ demands, and remove the flexibility neces-
sary to prioritize water uses.
For example, selective withdrawal to maximize downstream WQ for
fisheries may conflict with protection of WQ in the reservoir.
Reservoir fisheries and recreation may suffer to benefit down-
stream fisheries or vice versa. Resolving such issues is
difficult and requires that the Agencies maintain management
flexibility to prioritize water uses. While uniform WQ standards
do represent a simple and expedient approach, they would prevent
the Agencies from considering the complex mix of user demands
necessary to properly manage reservoirs.
2. Establish a cnmmnn. geographic information system/data base
(GIS/DB) for all national water resource agency impoundments.
H-22
-------
One of the major difficulties in studying this subject on a national
scale is the lack of a unified data set with comprehensive and
unbiased information. Given the current situation where each
agency has incomplete and unvalidated data sets stored in incom-
patible computer formats (or in printed summaries), it is diffi-
cult to make an accurate or reasonably precise evaluation of dam
water quality problems at the national level.
Water quality problems do not occur regularly in impoundments or in
downstream releases. However, in certain situations, depending upon many
factors discussed in this report, water quality problems can and do
occur. Preventative measures must then be instituted in new projects or
corrective actions taken at existing facilities. In order to identify
the locations, causes, and nature of water quality problems, an inventory
of water quality at major Federal impoundments is needed. Such a data
base would be a valuable resource for improved management of national
water resources. Inventories by Reclamation, COE, and TVA, prepared on a
compatible basis, would contain information typical of most situations in
the U.S. With proper coordination with non-Federal dam owners, the
usefulness and scope of these three data bases could be further expanded.
The Agencies should establish a common, validated GIS/DB that
would contain layers of data currently available from other
sources (USGS, EPA, etc,) along with data specifically related to
water projects.
A common format, distributed GIS would allow the retrieval of
other agency data when needed and would provide Congress, and
State and Federal environmental enforcement agencies rapid access
to both quantitative data and subjective evaluations for water
quality in reservoirs. This would facilitate improved decision
processes for identifying water quality problems and provide a
means for researchers, planners, and policy makers to evaluate
the relationships between water quality and other variables such
as land use, population, reservoir morphology, pollution sources,
and O&M strategies.
Technical Aspects
Examples of currently available digital data sets that could be
of use in evaluating the water quality problems of impoundments
include:
** USGS digital elevation data
** USGS streamflow and Benchmark Sample WQ data
** USGS bedrock geology maps
** SCS soil classification (mineralogy) maps
** USDA crop cover maps
** USFS timber cover maps
** Specific land classification and vegetative
cover studies based on digital analysis of
remote sensing data and satellite images
** EPA STORET water quality data
** EPA National Surface Water Survey alkalinity
H-23
-------
maps
** Subjective evaluations of WQ problems,
severity, and frequency from COE data set
** USER acid precipitation sensitivity data
** Engineering blueprints for TVA, COE, USER dams
** Population and economic data
** NOAA meteorological data
** Historical O&M records
** Hydrology data for specific projects
** Dam safety data for specific projects
The Agencies should agree to adopt hardware, software, and data
format protocols currently being used and developed by the USGS
in order to take advantage of available digital data sets and to
ensure data compatability. ARC/INTO is a GIS/DB that is widely
used and is available on VAX minicomputers. USGS is also devel-
oping microcomputer-based GIS systems that rely on compact disk
technology to store large data sets.
By adopting standard configurations, a lower-cost, distributed
approach to the GIS/DB can be used that eliminates the need for
expensive mainframe computers and management of a centralized
"super" data system. Each individual Agency would be responsi-
ble only for data directly related to their projects, and would
import other data sets for specific data evaluation tasks. In a
similar manner, national evaluations could be performed by tem-
porarily combining the Agency records for the most recent infor-
mation.
The final, and possibly most important aspect of this effort
involves implementation of a thorough quality assurance/quality
control (QA/QC) program for data entered into the GIS/DB. This
would address the issues of reliability of existing data,
establishment of appropriate techniques for WQ sampling and
chemical analyses, and procedures for documentation, error
detection and correction. The major portion of these protocols
have already been established by EPA and USGS, so minimal
development time will be required.
Policy Implications
The most significant policy implications of the GIS/DB will be
the necessity of securing development and long-term maintenence
funding from Congress, and adjusting Agency priorities to reflect
a stronger and more consistent committment to water quality is-
sues. To this end it is recommended that Congress be provided
with information that would encourage the development of this
GIS/DB. Also, the ongoing USGS GIS activities should be encour-
aged and promoted by the water resources agencies.
The development of a GIS/DB represents a complex task that will
require careful planning and coordination between the water
resource agencies, EPA, USGS and several other Federal and State
agencies. While organizational structures are in place for
H-24
-------
coordination between the water resource agencies, development of
a GIS/DB will require a wider and more comprehensive coordination
effort.
We would recommend that an Interagency Management Oversight Group
be formed to guide the development of the GIS/DB (and other WQ-
related initiatives). A permanent GIS/DB Working Task Group,
comprised of technical personnel who would be responsible for
implementing and maintaining the system, and providing the inter-
agency peer-communication necessary to maximize available resour-
ces and prevent costly duplication of effort, should be
appointed.
Members of the Management Oversight Group would provide policy
review, help to champion the GIS/DB concept inside their respec-
tive agencies, and encourage the proper funding priority to en-
sure a successful effort.
3. Pursue Congressional appropriations to adequately fund
reservoir monitoring programs.
One of the major problems with the STORET WQ data used in the
contractor report is that the available reservoir data is usually
collected during problem episodes, thus introducing significant
bias in the data. Also, WQ monitoring programs are inconsis-
tently funded, often resulting in non-representative data sets,
poor planning, lack of sampling design or quality assurance. In
order to address these problems, the following is recommended:
** Establish a standard funding period for post-
impoundment and post-mitigation action monitoring.
** Develop a stratified, randomized monitoring
program, much like the EPA Surface Water Survey, that
would provide an unbiased background data set to use
for national WQ evaluations in reservoirs.
** Develop Congressional committment to WQ by
encouraging a consistent, long-term approach to funding
of reservoir WQ monitoring.
** Ensure that monitoring is performed in accordance
with accepted QA/QC protocols to guarantee quality of
data.
** Include validated data in the reservoir GIS.
Technical Aspects
Monitoring should be performed with an adequate QA/QC program in
place, and resulting data should be included in the GIS/DB.
Development of a statistically randomized "background" sampling
program will be based on currently accepted methods for water
H-25
-------
quality sample network design.
Policy Implications
This recommendation is closely related to the GIS/DB initiative
and shares many of the policy issues identified above. While
Congress would have to appropriate funding for monitoring activ-
ities, it is also important to remember that the Agencies must
also develop an internal committment to water quality and moni-
toring that will ensure consistent long-term priority.
Establishment of a common QA/QC program for reservoir monitoring,
a task that will also be important to the GIS/DB, will require
coordination between the Agencies. It is recommended that the
QA/QC and the water quality monitoring network design for the
randomized program be developed by an interagency technical group
with review by EPA and statisticians familiar with sampling
design.
4. Develop better methods for Measuring the relationship between
WQ problems and effects on user benefits.
A discussion of water quality problems implies an adverse effect
on some recognized use of the water; e.g., drinking water, body
contact recreation, fish production, or agriculture. In recent
years, the economic impacts of salinity have been measured and
are currently used to select cost-effective alternatives for
meeting treaty WQ obligations for the Colorado River. Unfortun-
ately, there is no such standardized approach or methodology
available to quantify relationships between concentrations of
WQ constituents and the effects on user benefits or costs. Thus,
it is difficult to assign costs, to WQ problems or the mitigation
strategies chosen for a given problem.
By developing these relationships, agencies could prioritize
corrective action plans based on the severity of impacts to eco-
nomic benefits, recreational use, or intangibles such as aesthe-
tic quality. This priorxtization would also assist in the plan-
ning and design of cost-effective mitigation measures.
Technical Aspects
This process would involve applied research to develop and/or
evaluate existing methods for quantifying the costs (or loss of
benefits) associated with specific WQ constituents. Such an
effort would require expertise in risk assessment and related
statistical specialties.
Policy Implications
Investigation and implementation of this idea will once again
require coordination between the Agencies, and an appropriate
technical task group should be formed that will report to the
H-26
-------
Interagency Management Oversight Group. Policy guidance will
also be needed to establish appropriate risk factors inherent for
different WQ constituents and intended reservoir use.
5. Encourage public involvement in reservoir WQ issues by
establishing Citizen Advisory Boards or Councils.
Citizens1 Advisory Councils would help the Agencies identify
priority WQ issues in reservoirs and provide an organized
mechanism for public input or advise. Such Councils or Boards
would also help Agencies in several other important ways:
** Advisory Council(s) would foster a more cooperative
relationship between agencies, environmental groups,
user interest groups, and the general public. The net
result would be a more positive public image for the
agencies and a more obvious committment to water
quality issues.
** Good communication regarding WQ problems from the
public would prevent embarrassing public relations
problems due to lack of responsiveness through the
usual bureaucratic channels.
** The Advisory Council(s) would help provide lobbying
influence for funding of research and monitoring of WQ
problems in reservoirs.
Technical Aspects
None, although informed technical personnel should be involved in
Advisory Council activities.
Policy Implications
Once again, coordination between agencies should be important.
Advisory Councils may be nationally-oriented, however, it is more
likely that they will be of local interest. One agency's
experience with a particular problem could be very valuable in
helping another deal with the public on sensitive environmental
issues. An interagency task group that meets occasionally should
provide the means for effectively sharing such experiences.
****************************
The recoonendations detailed above should be coordinated accord-
ing to the organizational chart detailed on the following page.
H-27
-------
Organizational Structure for implementation of WQ recommendations
******************************************
* INTERAGENCY MANAGEMENT OVERSIGHT GROUP *
******************************************
Personnel: Agency administrators with WQ experience
Function: Policy oversight, program funding and
advocacy, Congressional liason
*
*
*
*
*
**************************
* GIS/DB Technical Group *
**************************
Personnel: Technical with CIS and computer experience
Function: Develop and coordinate approach, procurement and lAGs,
and actual implementation of system
*
*
*
*
*
******************************
* Risk/Cost Assessment Group *
******************************
Personnel: Technical with EIS, risk analysis and statistics
experience.
Function: Develop and implement assessment methodology
*
*
*
*
*
***************************************
* Advisory Council Coordination Group *
***************************************
Personnel: Administrative with WQ and public relations experience
Function: Communication between Agencies on WQ Issues
H-28
-------
APPENDIX I
AGENCY QUESTIONNAIRE USED IN
DEVELOPING ASSESSMENTS
SUPPLEMENT
Information is needed to prepare a summary statement on the present
status of water quality at Bureau dams. This will be used in a section
prepared by the Bureau for a report by EPA to be submitted to Congress
on water quality and dams.
INSTRUCTIONS
For each dam, complete the questionnaire using available information
and knowledge based upon day-to-day familiarity with the project.
Water Quality Evaluation
For dams or items for which information is not available, enter: NIA.
Provide the name of the dam, project, and category using the following:
Category
A. Constructed and operated by Bureau of Reclamation
B. Rehabilitated and operated by Bureau of Reclamation
C. Constructed by others, operated by Bureau of Reclamation
D. Under construction by Bureau of Reclamation
E. Constructed by Bureau of Reclamation, operated by others
F. Rehabilitated by Bureau of Reclamation, operated by others
G. Constructed and operated by others
H. Constructed under loan program
The evaluation table lists several commonly cited water quality considerations
for reservoir projects. For each water quality consideration and each
project location type (i.e., tributary, pool, and tailwater), evaluate
the extent of the problem (Column a), the relative impact on user benefits
(Column b), and the reliability of data upon which the evaluation is
based (Column c). Brief remarks which would aid in the interpretation
of this information may be entered where indicated.
1-1
-------
Problem Evaluation (Column a)
-_
0 :No problem evaluation has been Bade.
1 s Chronic or continuous problea.
2 :Intermittent problea occurring on • seasonal or event basis.
3 :Occasional problea occurring infrequently on an annual basis.
4 :No problea.
User Xepact (Column b)
Code
0 :Information concerning user impacts is not available.
1 :Severe impact resulting in the longtent loss of one or store
user benefits.
2 :Significant Impact Mhich restricts but does not eliminate
user benefits.
3 :Minor impact which does not restrict user benefits.
4 :No impact on user benefits.
Data Reliability (Column c)
Code
0 :Deta or infor&ation are not available.
1 :Based on reliable data covering appropriate tine fra&e.
2 :Based on scattered or incomplete data.
3 :Based on infomal information.
Water crudity considerations for which evaluations are requested are
briefly described below.
Iron - Elevated concentrations of dissolved or particulate iron.
tenganese - Elevated concentrations of dissolved or particulate
•engines*.
3. Low Dissolved Oxygen - Concentrations beloK saturation.
4. Hydrogen sulfide * Elevated concentrations or obvious odors.
5. Turbidity - Reduced water clarity due to suspended inorganic solids.
6. Low Flow - Insufficient flow in the tailwter.
7. Eigh How - Excessive discharge flows in the tailtater.
6. Fluctuating Flc« - Excessive or unnatural changes ir floa.
S. low Temperature - Temperature below expected or desirable level.
10. Elgh Temperature - Temperature above expected or desirable level.
11. nuctUBting Temperature - Undesirable changes in temperature.
12. Dissolved Solids - Elevated concentrations of total dissolved solids.
12. betel Contaminants - Elevated concentrations of ftetals other than iron
end langanese.
14. Organic Contaminants - Presence of man-Bade organic coapounds.
IS. Gts Supersaturatioi. - Dissolved nitrogen gas concentration above
saturation.
1€. Eigh Nutrients - Excessive nitrogen and phosphorus concentrations.
17. £Lgae - Excessive algal biotas s or chlorophyll concentration.
1-2
-------
18. K&crophytes - Excessive or undesirable growths of rooted or floating
aquatic plants.
19. Sediaent Accu&ulation - Excessive or undesirable sediment accumulation.
20. DraHdown - Prolonged periods of IOM pool elevation Kith undesirable
iapacts.
21. Pool Elevation Fluctuation - Undesirable i&pacts due changing pool.
22. Shoreline Erosion - Loss of banks or shoreline due to erosion.
23.'Taste and Odor - Taste or odor in raw and/or finished potable tater.
24. pH or Acidity - pR significantly below neutrality or nigh level of
acidity.
25. Bacteria - Excessive levels of any sdcrobe.
26. Parasites - Presence of any aniaal or hua&n parasite.
27. Other - Specify other stater quality considerations as needed.
28. Other - Specify other Mater quality considerations as needed.
29. Other - Specify other Mater quality considerations as needed.
30. Other - Specify other Mater quality considerations as needed.
1-3
-------
1. Dam:
3. Category;
-
I !
1 Kater Quality |
1 Consideration I
!, T " '
I Tributary
I •
1
IX. Iron ||
I ii
{2. Kanganese ||
1 it
|3. LOH D.O. ||
I II
14. Bydrogen Sulfide ||
15. Turbidity ||
1 M
Ife. Low Flow ||
1 II
17. High Flow ||
1 II
IE. Fluctuating Flow |
IS- Low TcMerature
I
|1C. High Teaoeraturc
1
Ill.Fluctuating lesp.
|12.Dis3olvc4 Solids
I
|13.Ketel Contaminant
1
IK.Orotnlc Conterir*.
!
|15. Ges Sur>ersaturat.
I
I It. High Kutrientr
1
jL7.Alcee
!
IlE.Kacroshvtcj
I
I IS. Sediment Accut.
i
«
|2C .DraMOKs
I
1 21. Pool Eiev nuct.
I
|22. Shore Eros.
I
I
—
WHHW
t
-
«^™^«^
w—_
c
^— —
^_^^^_
mm ^_^
&•
rrojeci:
Pool || ItUwter
M
a
_
_
— m^^m
b
VH^^V
c || a
M
II
||
II
II
II
II
II
||
II
II
II
l|
II
_ 11
II
(I
II
|l
1
_
1
b
c
•••^^H
fie&arks |
-
-
1-4
-------
Water Quality Description (Continued):
| || Tributary II Pool || Ttilmter
f Rater Ou.lity II II II
| Consideration || a
I II
| 23. Taste and Odor ||
I II
|24.pH/Acidity ||
1 II
1 25. Bacteria ||
1 II
(26. Parasites ||
1 II
I27.0ther £ )|l
1 II
I2B. Other i )ll
1 II
129. Other ( )ll
1 II
130. Other I )ll
1 II
b
c II a
II
II
II
II
II
II
II
II
II
II
II
II
11
II
II
II
i;
b
c II a
II
1
1
b
c
Eenarks
•fr U.S GOVERNMENT PRINTING OFFICE 1989— 617-003 ' 0 "» 8 6 1
1-5
------- |