TVA
ERA
Tan «§§§©<§
Valley
Authority
United States
Environr
Agency
Energy Demonstrations and
Technology
Chattanooga, TN 37401
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-79-236
November 1979
Design of a Monitoring
Program for Ash Pond
Effluents
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-236
November 1979
Design of a Monitoring Program for
Ash Pond Effluents
by
FA Miller, III, T.V.J. Chu, and R.J. Ruane
TV A Project Director
Hollis B. Flora II
Tennessee Valley Authority
1140 Chestnut Street, Tower II
Chattanooga, Tennessee 37401
Contract No. IAG-D5-E-721
Program Element No. TNE624A
EPA Project Officer: Michael C. Osborne
Industrial Environmental Research Laboratory
Office of Environmental Engineering and Technology
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ill
DISCLAIMER
This report was prepared by the Tennessee Valley Authority and has
been reviewed by the Office of Energy, Minerals, and Industry, United
States Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the
views and policies of the Tennessee Valley Authroity or the United
States Envrionmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation for use.
-------
iv
CONTENTS
Page
Disclaimer ii
Acknowledgements xi
Abstract xii
Section
1 Introduction 1
2 Conclusions and Recommendations 6
3 Summary of TVA Data from 1970 to 1975 10
Individual Ash Pond Effluent
Characteristics 10
Relationships Between Plant Operation
Conditions and Ash Pond Effluent
Characteristics 42
Relationship Between the Character-
istics of the Intake Water and
Ash Pond Effluent 45
Indirect Monitoring Methods 58
Comparison of Weekly and Quarterly
Sampling 61
Comparison of Grab and Composite
Sampling 63
4 Procedure for Designing an Ash Pond
Monitoring Program 67
Data Requirements 68
Variation of the Data with Time 68
Distribution of the Data . 68
Estimation of the Mean as a Function
of the Precision 69
Estimation of the Precision 73
Stepwise Summary of the Design
Procedure 75
5 Ash Pond Monitoring Program for Plant E 76
Description of Plant E 76
Mechanics of the Ash Pond System at
Plant E 77
Summary of the Ash Pond Effluent
Characteristics at Plant E 88
-------
CONTENTS (Contined)
Section Page
Variation of the Ash Pond Effluent
Characteristics at Plant E with
Time 94
Statistical Distribution of the Effluent
Characteristics at Plant E 100
Estimation of the Mean as a Function
of the Precision 108
Selection of the Precision 116
Estimated Sampling Frequencies 118
Example Sampling Program for Plant E 127
Summary 131
6 Ash Pond Monitoring Program for Plant J 132
Description of Plant J 132
Mechanics of the Ash Pond System
at Plant J 132
Summary of the Ash Pond Effluent
Characteristics at Plant J 133
Variation of the Ash Pond Effluent
Characteristics at Plant J with Time .... 134
Statistical Distribution of the Effluent
Characteristics at Plant J 141
Estimation of the Mean as a Function
of the Precision 150
Selection of the Precision 160
Estimated Sampling Frequencies 166
Example Sampling Program for Plant J 168
7 Future Applications 176
References 178
Appendix A 179
Appendix B 182
Appendix C 186
Appendix D 188
Appendix E 190
-------
VI
LIST OF TABLES
Table Page
1 TVA Steam Plant NPDES Monitoring Requirments
for Ash Pond Effluents—Effective June 1976
to July 1, 1977, and July 1, 1977, to the
Present 3
2 Chemical Effluent Guidelines and Standards for
Steam-Electric Power Generating Plant Ash
Ponds 5
3 Summary of Weekly Ash Pond Effluent Data from
1970 through 1975 11
4 TVA Ash Ponds Which Showed a Yearly Cycle 32
5 Summary of Quarterly Trace Metal Data for
Ash Pond Intake and Effluent Streams 34
6 Summary of Plant Operation Conditions and
Ash Pond Effluent Characteristics of TVA
Coal-Fired Power Plants 43
7 Linear Correlation Coefficients Significant
at the 95% Level of Confidence for Plant
Operating Conditions 44
8 Summary of Weekly Ash Pond Intake Water Data
for 1974 and 1975 46
9 Correlation Coefficients for the Ash Pond
System at Plant E 50
10 Lagged Correlation Coefficients for
Plant E 57
11 Comparison of Weekly Intake and Effluent
Suspended Solids Concentrations for
1974 and 1975 at TVA Ash Ponds 59
12 Number of Ash Ponds Whose Average Effluent
Concentrations Exceed Those of the Intake
Water 60
13 Comparison of Quarterly and Weekly Sampling
Programs 62
14 Chemical Analysis of Ash Pond Effluent and
Intake Water used for Sluicing During
Preliminary Survey at Plant E 82
-------
Table
vii
LIST OF TABLES
(Continued)
15 Chemical Analysis of Ash Pond Effluent and
Intake Water Used for Sluicing During the
February Survey at Plant E 84
16 Suspended Metals Concentration for the
February Ash Pond Survey at Plant E 86
17 Average Chemical Analysis of the Ash Pond
Effluent and Intake Water Supply During
Both Ash Pond Surveys 87
18 Summary of the Ash Pond Effluent Character-
istics at Plant E for the Two Ash Pond
Surveys and the Quarterly Monitoring
Program During 1974 and 1975 89
19 Ash Pond Effluent Characteristics for
Plant E 91
20 Linear Correlation Coefficients for the
Various Ash Pond Effluent Parameters
at Plant E 92
21 Type of Distribution and Statistical
Characteristics of the Ash Pond
Effluent at Plant E 107
22 Comparison of the Ash Pond Effluent Char-
acteristics Following a Normal Distribu-
tion at Plant E with Ash Pond Effluent
Limitations or Water Quality Criteria
(Based on Data Collected Prior to
January 1978) 109
23 Comparison of the Ash Pond Effluent Char-
acteristic Following a Log Normal Dis-
tribution at Plant E with Ash Pond
Effluent Limitations or Water Criteria
(Based on Data Collected Prior to
January 1978) 110
24 Upper and Lower Limits for the Critical
Range of the Precision for the Effluent
Characteristics of Plant E 117
25 Assumed Water Quality Characteristics
for the Receiving Stream at Plant E 119
-------
Vlll
Table
LIST OF TABLES
(Continued)
26 Required Precision for the Monitoring Program
of Plant E Assuming an Average Allowable
Concentration in the Receiving Stream
Equal to the EPA Proposed Water Quality
Criteria 120
27 Required Precision for the Monitoring Program
of Plant E Assuming an Average Allowable Con-
centration in the Receiving Stream Equal to
the Maximum Value Reported for the Intake
Water 121
28 Number of Samples Required to Estimate the
Yearly Mean Within 20% of the True Yearly
Mean of Plant E 123
29 Estimate Sampling Frequencies for the Moni-
toring Program at Plant E Assuming Allowable
Average Concentrations in the Receiving Stream
Equal to the EPA Water Quality Criteria and
Maximum Value Reported for the Intake Water .... 125
30 Deviation of the Yearly Sample Mean from the
True Mean for the 99% Confidence Level at
Various Sampling Frequencies 126
31 Example Sampling Program for Plant E 128
32 Ash Pond Effluent Characteristics at
Plant J 135
33 Linear Correlation Coefficients for the
Various Ash Pond Effluent Parameters
at Plant J 136
34 Selected Sampling Period, Type of Dis-
tribution and Statistical Character-
istics of the Ash Pond Effluent at
Plant J 149
35 Comparison of the Ash Pond Effluent Char-
acteristics Following a Normal Distribu-
tion at Plant J with the Ash Pond Effluent
Limitations or Water Quality Criteria 151
-------
IX
Table
LIST OF TABLES
(Continued)
36 Comparison of the Ash Pond Effluent Character-
istics Following a Lognormal Distribution of
Plant J with the Ash Pond Effluent Limitations
or Water Quality Criteria 152
37 Upper and Lower Limits for the Critical
Range of the Precision for the Effluent
Characteristics of Plant J 161
38 Assumed Water Quality Characteristics for
the Receiving Stream at Plant J 163
39 Required Precision for the Monitoring Program
at Plant J Assuming an Average Allowable Con-
centration in the Receiving Stream Equal to
the EPA Proposed Water Quality Criteria 164
40 Required Precision for the Monitoring Program
at Plant J Assuming an Average Allowable Con-
centration in the Receiving Stream Equal to
the Maximum Value Reported for the Intake
Water 165
41 Number of Samples Required to Estimate the
Yearly Mean Within 20% of the True Yearly
Mean for Plant J 167
42 Estimated Sampling Frequencies for the Moni-
toring Program at Plant J Assuming Allowable
Average Concentrations in the Receiving Stream
Equal to the EPA Water Quality Criteria and
Maximum Value Reported for the Intake Water .... 169
43 Deviation of the Yearly Sample Mean From the
True Mean for the 99% Confidence Level at
Various Sampling Frequencies 170
44 Example Sampling Program for Plant J 171
-------
LIST OF FIGURES
Figure Page
1 Typical Hydraulic Ash Sluicing System 2
2 Variation of Plant E Ash Pond Effluent
Characteristics with Time for the
Period 1974 to 1976 28
3 Variation of Plant J Ash Pond Effluent
Characteristics with Time for the Period
1970 to 1976 30
4 Number of TVA Ash Ponds Whose Average Effluent
Concentration Equals or Exceeds Various
Given Concentration 38
5 Relationship of Ash Pond pH and Intake Water
Alkalinity for Plant J 54
6a Relationship of Suspended Solids in the Ash
Pond Effluent and the Intake Water Supply
for Plant E 55
6b Relationship of Suspended Solids in the Ash
Pond Effluent and the Intake Water Supply
for Plant J 56
7 Comparison of Grab and Composite Samples for
Four TVA Ash Pond Effluents 65
8 Example of a Cumulative Frequency Plot 70
9 Example of a Plot of the Number of Samples
Versus the Deviation from the True Mean 72
10 Vertical Profile of Ash Pond Characteristics
of Plant E During Thermal Stratification
and Isothermal Periods 78
11 Concentration of Rhodomine WT Dye in
Plant E Ash Pond Effluent with Time 81
12 Relationship Between Flow and Suspended
Solids in the Ash Pond Effluent at
Plant E 93
13 Variation of Ash Pond Effluent Character-
istics with Time at Plant E 95
14 Cumulative Frequency Plots for the Ash
Pond Effluent at Plant E 101
-------
XI
LIST OF FIGURES
(Continued)
15 Number of Samples Required for a Given
Precision for the Plant E Ash Pond
Effluent Ill
16 Variation of Ash Pond Effluent Character-
istics with Time at Plant J 137
17 Cumulative Frequency Plots for the Ash
Pond Effluent at Plant J for the Period
January 1 to December 31 142
18 Cumulative Frequency Plots for the Ash
Pond Effluent at Plant J for the Period
November 1 to April 30 143
19 Cumulative Frequency Plots for the Ash
Pond Effluent at Plant J for the Period
May 1 to October 31 146
20 Number of Samples Required for a Given
Precision for the Ash Pond Effluent
at Plant J for the Period January 1
to December 31 153
21 Number of Samples Required for a Given
Precision for the Ash Pond Effluent
at Plant J for the Period November 1
to April 30 154
22 Number of Samples Required for a Given
Precision for the Ash Pond Effluent
at Plant J for the Period May 1 to
October 31 157
-------
xii
ACKNOWLEDGEMENTS
This study was initiated by TVA as part of the project entitled
"Characterization of Effluents from Coal-Fired Utility Boilers," and is
supported under Federal Interagency Agreement Numbers EPA-IAG-D5-E-721
and TV-41967A between TVA and EPA for energy-related environmental
research. Thanks are extended to EPA project officers, Michael C.
Osborne and Dr. Ron A. Venezia, and TVA project director, Dr. Hollis B.
Flora II. Appreciation is also extended to Richard M. Bittman, David J.
Bruggink, Blake Harmon, Walter S. Liggett, Jr., Kenneth L. Ogle, Frank G.
Parker, Randall L. Snipes, Virgil E. Vandergriff and J. Michael Wyatt.
-------
xiii
ABSTRACT
The objective of this research was to develop a procedure for
designing a sound monitoring program for fossil-fueled power plant ash
pond effluents. Those factors which influence the effluent character-
istics and are of importance in designing a sound ash pond monitoring
program were determined based on a review of the plant operating char-
acteristics and ash pond effluent characteristics for the TVA fossil-
fuel power plant system. A statistical procedure for determining the
sampling frequency of chemical characteristics in ash pond effluents was
then developed based on the following equation:
n =
where n is the sample size,
t is the value of "student's" t for a given significance
level,
L is the precision, and
S is the sample standard deviation.
The precision is given by |J-X where (J is the population mean and X is
the sample mean. Two methods of determining the precision were presented.
The first involves selecting a precision value in order to estimate the
population mean within a given percentage. This method gives the number
of samples required to estimate the population mean within some degree
of certainty. The second involves calculating a precision value by
subtracting an estimate of the population mean from either the ash pond
effluent limitation established by EPA or a desirable water quality
criterion. This method gives the number of samples required to show
that the effluent is in compliance with the effluent limitation or below
the water quality criteria. The method chosen to compute the precision
depends on the purpose of the monitoring program.
The use of this procedure was demonstrated for two of TVA's ash
pond systems. Example monitoring programs utilizing this procedure
indicated that the sampling effort for trace metals in the ash pond
effluent at both plants could be substantially decreased. This
procedure should be a useful tool to managers in determining the
resources needed for monitoring. The procedure may also be used to
indicate when part of the investment in pollution control measures
may be justified to offset the cost of monitoring to show compliance.
The major limitations of the procedure are that: (1) It relies on
maintaining the same type of power plant and ash pond operating condi-
tions in the future as were used during the period when the design data
set was collected; (2) it depends heavily on the establishment of effluent
limitations; (3) the effluent must be in compliance; and (4) it cannot
be applied generically to all ash pond effluents, but must be applied
individually to each effluent.
-------
SECTION 1
INTRODUCTION
The Tennessee Valley Authority (TVA) in conjunction with the Environ-
mental Protection Agency initiated a study entitled "Characterization of
Effluents from Coal-Fired Utility Boilers," to characterize the various
effluents associated with coal-fired generating facilities. As part of
that study a procedure for designing an ash pond effluent monitoring
program to fulfill the requirements of the National Pollutant Discharge
Elimination System was initiated and demonstrated for two of TVA's power
plants. The procedure is such that it can be applied to other ash pond
effluents outside of the TVA system. The information presented in this
report represented conditions as they existed up to January 1978. These
conditions are subject to change due to plant modifications made in an
effort to achieve full compliance with NPDES permit requirements.
The Tennessee Valley Authority operates 12 coal-fired power plants
which supply approximately 65 percent of the system's total power genera-
tion (28 million kilowatts). In 1975, approximately 34 million tons of
coal were burned resulting in an estimated 5.3 million tons of ash
material. This ash material is comprised of varying portions of pyrites,
bottom ash, and fly ash depending on the method of firing, source of
coal, and fly ash collection systems used at the plant. The fly ash can
be further classified as mechanically collected (MC) or electrostatically
hot and cold collected (ESP). The majority of this ash material is
transported hydraulically from the point of production to a settling and
disposal pond. A typical ash sluicing and disposal system is shown in
Figure 1. The water used for sluicing the ash to the pond is then dis-
charged back to the original receiving stream. In 1975 this resulted
in an average effluent discharge of greater than 240 MGD for the total
TVA system, or about 13,200 gpd per MW.
In 1967 TVA initiated a periodic sampling program of the surface
water discharges from these coal-ash disposal ponds. In 1970, TVA began
collecting weekly grab samples and analyzed the samples for pH, alka-
linity, hardness, conductivity, total and dissolved solids, and turbidity.
Quarterly grab samples collected since 1968 have been analyzed for eight
additional parameters (Ca, Mg, Cl, Na, Fe, Mn, S04, and Si) and those
quarterly samples collected since 1973 have also been analyzed for trace
metals (Al, As, Ba, Be, Cd, Cr, Cu, Pb, Hg, Ni, Se, Ag, and Zn), phosphorus,
ammonia nitrogen, and cyanide.
As a result of the 1972 Amendments to the Federal Water Pollution
Control Act (Public Law 92-500), TVA began an ash pond effluent monitor-
ing program for its 12 coal-fired power plants to comply with the National
Pollutant Discharge Elimination System (NPDES). The requirements for
this program since June 1976 are shown in Table 1. The required frequencies
for some parameters have been increased at certain plants since July 1,
1977, as shown in parentheses in Table 1.
The Environmental Protection Agency (EPA) promulgated effluent
limitations guidelines in 1974 for the achievement, by 1977, of best
practicable control technology currently available (BPCTCA) and, by
-------
-2-
MECHANICAL COLLECTOR
AIR \ /
INTAKE-7 V/
\7 I
ASH HOPPER
ASH INTAKE J^SLUICE WATER INLET
WATER EJECTOR-*A [
^
AIR
INTAKE
ELECTROSTATIC
PRECIPITATOR
FLY ASH HOPPER
ASH INTAKE
WATER EJECTOR
SPRAY WATER INLET
I
BOTTOM ASH HOPPER
-DRY BOTTOM
HYDRAULIC SLUICE
GATE CYLINDER
ASH SLUICE TRENCH
TO ASH
DISPOSAL
POND
ASH PUMP
Figure 1. Typical Hydraulic Ash Sluicing System
-------
-3-
TABLE 1. TVA STEAM PLANT NPDES MONITORING REQUIREMENTS FOR
ASH POND EFFLUENTS—EFFECTIVE JUNE 1976 TO JULY 1, 1977, AND
JULY 1, 1977, TO THE PRESENT
1.
2.
3.
4.
Plant
A
B
C
D
E
F
G
H4
I
J
K
L
Flow
W
M
C (W)
W
W
W
W (D)
W
W
W
W
W
Frequency of
Oil &
pH Grease
W M (2M)
W (C) M (2M)
W (C) M (2M)
W (C) M (2M)
W M
W (C) M (2M)
W (C) M (2M)
W (C) M (2M)
W (C) M (2M)
W (C) M (2M)
W M (2M)
W M (2M)
Monitoring1
Susp.
Solids
2M
M (W)
2M (W)
M (2M)
M
M (2M)
W
M (2M)
M (2M)
M (2M)
M (2M)
M (2M)
Parenthesis indicates revised sampling frequency after
Mercury sample required 2M.
Heavy metals, also, required at plant intake, one point
and the West Knox Utility District intake.
Sampling required for two ash pond effluents.
Metals
M
M
M2
M3
Q
M
M
M
M
M
Q
Q
July 1, 1977.
in the Clinch River,
Frequency Code: C - Continuous M - Once per month
D - Once per day 2M - Twice per month
W - Once per week Q - Once per quarter
-------
-4-
1983, of best available technology economically achievable (BATEA) for
the steam-electric power generating point source category (1). A dis-
cussion of applicable control technology may be found in reference 1. A
summary of current EPA effluent guidelines for ash pond discharges from
steam-electric power generating plants is shown in Table 2. The major
goal of past monitoring programs was to provide an up-to-date data base
from which to assess the potential for adverse environmental effects
from this type of power plant discharge. An additional goal of the
monitoring program today is to show that the effluent is in compliance
with the effluent limitations shown in Table 2.
Ash pond effluent sampling frequencies utilized by TVA in past
monitoring programs and by EPA in the proposed NPDES permit have been
based on "educated guesses" without the benefit of formal study to
establish adequate, statistically sound sampling frequencies. Improperly
established frequencies could result in either unobserved, excessively
variant parameter levels or too frequent, costly, and unnecessary sampling
and laboratory analyses. The NPDES permit allows TVA with EPA concurrence
to adjust monitoring frequencies if studies indicate changes are justified.
A sound monitoring program should be based on knowledge of the
system and statistical analysis of the data gathered. The better these
two aspects are integrated, the more meaningful the monitoring program.
Therefore, the objective of this report is to develop a procedure for
designing a sound monitoring program for ash pond effluents. To reach
this objective, this report includes studies:
1. To determine those factors which influence the effluent
characteristics and are of importance in designing a sound
ash pond monitoring program.
2. To determine if some of these parameters could be omitted
from sample analyses or if some parameters could be used to
estimate other parameters.
3. To develop a statistical procedure for determining sampling
intervals to meet standards and establish water quality trends.
4. To use this procedure to determine statistically sound moni-
toring programs for two of TVA's ash pond systems.
5. To indicate how this procedure could be applied to other
ash pond systems.
Items 1 and 2 were accomplished by analysis of available ash pond
effluent characteristics from 1970 to 1975, along with an identification
of intake water quality parameters and steam plant operating charac-
teristics which may influence the strategy used in developing a monitor-
ing program for ash pond effluents. Item 3 was accomplished by modification
of statistical methods available in the literature for determining the
sample size to estimate population means at various confidence levels.
Item 4 was accomplished by conducting a detailed sampling program of the
ash pond effluents from two of TVA's steam plants and item 5 was
accomplished by outlining how the procedure developed during this study
could be applied to other power plant ash ponds.
-------
-5-
TABLE 2. CHEMICAL EFFLUENT GUIDELINES AND STANDARDS FOR STEAM-ELECTRIC
POWER GENERATING PLANT ASH PONDS1 2
BPCTCA
July 1. 1977
BATEA
July 1, 1983
New Source
Standards
pH
polychlorinated biphenyls
Bottom Ash Transport Water
total suspended solids
oil and grease
6.0-9.0
zero
Average Daily
Daily Maximum
6.0-9.0
zero
6.0-9.0
zero
Average Daily Average Daily
Daily Maximum Daily Maximum
30 100 30 -f 12.5 100 4- 12.5 30 -r 20 100 -r 20
15 20 15 -f 12.5 20 -r 12.5 15 -r 20 20 T 20
Fly Ash Transport Water
total suspended solids
oil and grease3
30
15
100
20
30
15
100
20
zero
zero
zero
zero
1. Taken from "Development Document for Effluent Limitations Guidelines and New Source
Performance Standards for the Steam-Electric Power Generating Point Source Category,"
U.S. Environmental Protection Agency, Report No. EPA-440/l-74-029-a, (October 1974).
2. All units are in mg/1. Allowable discharge is the quantity obtained by multiplying
30 by the ratio of flow for sluicing to flow discharged and dividing by 12.5 or 20.
3. For wet ash handling systems new sources must have zero discharge; however, this
limitation for any runoff from the ash storage pile for the dry ash handling
system was remanded by the Fourth Circuit Court of Appeals in July 1976.
-------
-6-
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
The following conclusions with respect to the variation of and
trends displayed by the TVA ash pond effluents were observed:
1. Seven of the ash ponds exhibited yearly cycles in pH, total
alkalinity, conductivity, dissolved solids, and total solids.
2. The remaining ash ponds exhibited no yearly pH cycle, but three
had yearly cycles for alkalinity, dissolved solids, and total
solids.
3. None of the ash ponds exhibited a yearly cycle for suspended
solids, turbidity, or flow.
4. The concentration of most metals in the ash pond effluent
appear to vary with time. The variation differs for each
element within each effluent.
5. The pH of the ash pond effluents in the TVA system vary from
acidic to alkaline.
Based on the review of the plant operating characteristics and ash
pond effluent characteristics for the TVA fossil fuel power plant system,
the following conclusions were derived:
1. The pH of the ash pond effluent was highly correlated with the
percentage of CaO in the fly ash and the sulfur content of the
coal.
2. The effluents from plants which receive coal from western
Kentucky and southern Illinois (sulfur content of coal usually
2.8 to 4 percent and calcium content of fly ash usually 2.4 to
5.0 percent) are basic while those from plants which receive
coal from eastern Tennessee, eastern Kentucky, and Virginia
(sulfur content of coal usually 2 percent or below and calcium
content of fly ash usually 2.2 percent or below) are neutral
or slightly acidic.
3. The suspended solids concentration of the effluent correlated
highly with the percentage of CaO in the fly ash, pH of the fly
ash, and pH of the ash pond effluent.
4. The relationship between intake water quality characteristics
and the ash pond effluent water quality characteristics varied
for the different ash pond systems. For example, there was a
significant correlation between the intake water pH and the pH
of the ash pond effluent and also between the intake water
dissolved solids and the pH of the ash pond effluent for Plants
G, H, and J. These same plants also had a significant adverse
correlation between the intake water conductivity and ash pond
effluent pH. There was also a significant correlation between
-------
-7-
intake water hardness and effluent pH for Plant J and a signifi-
cant negative correlation between intake water hardness and
effluent pH for Plants G and H. However, only Plants G and J
had significant correlations between intake water alkalinity
and effluent pH. For Plant G it is negative, while for
Plant J it is positive.
5. More than half of the ash ponds increased the average concen-
trations of Al, ammonia, As, Ba, Cd, Ca, Cl, Cr, Pb, Hg, Ni,
Se, silica, sulfate, and Zn over that in the intake water.
6. The detention time in the ash ponds is a function of the mixing
of the pond contents. The mixing is a function of the wind
conditions, pond geometry, and plant operating conditions such
as the number of units operating, load capacity, and number of
ESP units operating.
7. Trace metals in the ash effluents are interrelated with one
another at Plants E and J, respectively.
8. Based on statistical regression analyses using the available
data on operating conditions of TVA steam plants (such as ash
content, sulfur content, coal usage, ESP and mechanical ash
collector efficiencies, intake water characteristics and fly
ash characteristics) and ash pond systems, there were not any
useful relationships that could explain, predict, or control
the ash pond effluent water quality. However, if data on the
coal characteristics (including trace elements), intake water
characteristics, and detailed plant operating conditions (such
as coal load and time of sluicings) were available on a routine
frequency over the same time frame, such relationships could
conceivably be developed.
Conclusions related to the frequency and method of collection of
ash pond effluent samples were:
1. Quarterly sampling for suspended solids during 1973 to 1975
yielded approximately the same yearly average as did weekly
sampling for the following ponds: A fly ash, A bottom ash, D,
G, H, I, and L.
2. Quarterly sampling for pH during 1973 to 1975 was adequate to
predict the yearly average pH within 0.5 pH units for all
plants except Plant B fly ash in 1973, Plant C west in 1974,
and Plant L in 1973.
3. Grab samples were selected over composite samples for the moni-
toring programs recommended in this study because they were
easier to collect. More study is needed to determine if
composite samples would be more representative of the system.
4. Each pond is site-specific with respect to effluent character-
istics which require monitoring attention. Therefore, each
pond must be studied separately in order to establish the most
cost effective monitoring program for the entire TVA system.
-------
-8-
Conclusions regarding the development of a procedure for statistically
designing an ash pond effluent monitoring program were:
1. The procedure for determining the sampling frequency of
chemical characteristics in ash pond effluents presented in
this report is based on the following equation:
t2S2
where n is the sample size,
t is the value of "student's" t for a given
significance level,
L is the precision, and
S is the sample standard deviation.
2. The precision is given by |J-X where |J is the population mean
and X is the sample mean. Two methods of determining the
precision were discussed. The first involves selecting a
precision value in order to estimate the population mean
within a given percentage. This method gives the number of
samples required to estimate the population mean within some
degree of certainty. The second involves calculating a
precision value by subtracting an estimate of the mean from
the effluent limitation or water quality criteria. This
method gives the number of samples required to show the
effluent is in compliance with the effluent limitation or
below the water quality criteria.
3. Designing a monitoring program to estimate the mean value of
all parameters within the same percentage of the true mean for
that parameter may lead to over-sampling for some parameters
and under-sampling of others, because this approach does not
take into account the significance of the concentration in the
waste stream and it tends to reduce the precision value
(increase the number of samples) as the concentration in the
waste stream decreases.
4. Designing a monitoring program based on collecting samples of
all parameters at the same frequency may lead to over-sampling
for some parameters and under-sampling of others from a statis-
tical standpoint. For example, some parameters are estimated
more accurately than others, possibly making comparisons
between parameters misleading.
5. Care should be exercised in establishing averaging periods for
effluent limitations because the averaging period greatly
affects the sampling frequency.
6. By utilizing the procedure presented in this study, the sampling
effort for trace metals in the ash pond effluents at Plants E
and J could be substantially decreased. For example, the sampling
-------
-Q-
program at Plant E was reduced from a total of 56 analyses per
year for 12 different elements to 48 analyses per year for 9
different elements. At Plant J the reduction was from 156
analyses per year for 12 different elements to 35 analyses per
year for 12 different elements.
7. The following example sampling program was developed for
Plant E: once per year for Cr, Cu, Fe, Pb, Mn, pH, and Zn;
twice per year for As; 4 times per year for Se; and 36 times
per year for suspended solids.
8. The following example sampling program was developed for
Plant J: once per year for Cu, Fe, and Mn; twice per year for
As and Zn; 4 times per year for pH and selenium; and 24 times
per year for suspended solids.
9. The procedure for designing a monitoring program presented
here should be a useful tool to managers in determining the
resources needed for monitoring.
10. The procedure may also be used to indicate when part of the
investment in pollution control measures may be justified to
offset the cost of monitoring to show compliance.
11. The major limitations of the procedure are: (1) It relies on
maintaining the same type of operating conditions in the
future as were used during the period when the design data set
was collected; (2) it depends heavily on the establishment of
effluent limitations; (3) the effluent should be in compliance;
and (4) it cannot be applied generically to all ash pond
effluents, but must be applied individually to each effluent.
12. The procedure for determining sampling frequencies presented
here should be applied to the remaining TVA facilities once
the plant modifications to meet environmental regulations have
been completed.
13. Permission should be sought to alternate the NPDES monitoring
program to reflect the results of this study and work performed
in recommendation 1.
14. Less emphasis should probably be given to routine monitoring
programs and more emphasis given to special or intensive
studies directed at determining the effects of power plant
operations on the ash pond effluent water quality and the
effect of ash pond effluent water quality on the receiving
stream water quality and its habitant.
-------
-10-
SECTION 3
SUMMARY OF TVA DATA FROM 1970 TO 1975
This section summarizes the data available from 1970 to 1975 on
individual ash pond effluent characteristics, the relationships between
plant operating conditions and ash pond effluent characteristics, the
relationships between the intake water and ash pond effluent charac-
teristics, comparisons between weekly and quarterly sampling, and
comparisons between grab and composite sampling.
INDIVIDUAL ASH POND EFFLUENT CHARACTERISTICS
The ash pond effluent data collected on a weekly basis at each of
TVA's steam plants from 1970 through 1975 are summarized in Table 3.
The maximum, average, and minimum values are given by year for flow, pH,
phenolphthalein alkalinity, total alkalinity, hardness, conductivity,
total solids, dissolved solids, suspended solids and turbidity. Care
should be taken in comparing the values for a particular ash pond from
year-to-year because of changes in the analytical procedures, the type
data reported, and the ash pond location; the most important analytical
change being the one used to determine the solids content. From 1970 to
1973, the effluents were analyzed for total and dissolved solids and the
suspended solids concentrations were calculated by difference. Starting
in 1974, the samples were analyzed for suspended and dissolved solids
and the total solids concentration calculated by summation.
Plants A and B have separate ash ponds for bottom ash and fly ash
while the remaining plants have ash ponds which receive both bottom and
fly ash. Although the pH of the ash pond effluent varies from acidic to
alkaline from plant to plant, a survey of the data in Table 3 indicates
the pH of a particular ash pond is relatively constant from year-to-year.
For the most part, the average pH from year-to-year for a particular
plant only varies about half of a pH unit while the difference in the
maximum or minimum value from year-to-year is approximately one pH unit.
The fly ash pond at Plant A had yearly pH averages of 6.5 in 1970, 5.4
in 1972, and 4.0 in 1975. Except for 1971 the yearly average pH for the
Plant A fly ash pond decreased with time from 1970 to 1975. The pH at
Plant D was substantially lower in 1970 (average 6.5) than in the later
years (average 8.5). The average pH of the effluent at Plant G increased
from 5.7 in 1972 to 9.8 in 1973. Beginning in 1973 the sampling location
for Plant G changed from the old pond to the new one. This change in
operation along with others probably accounts for this increase. From
1973 to 1975 the pH stablized with the average pH, maximum pH, and
minimum pH values being within 0.4, 0.1, and 1.6 pH units, respectively.
The average pH at Plant I decreased from between 11.1 to 11.3 during
1970-1974 to 9.8 in 1975. The pH of a particular ash pond effluent
within any one year can vary from 1 to 6 units. However, plants A, D,
G, and I are exceptions.
The yearly average suspended solids concentration during the period
1970 to 1975 varied by more than 20 mg/1 for all the ash ponds except
those at Plants C, G, H, and I, with some varying by as much as 50 mg/1.
-------
-11-
TABLE 3. SUMMARY OF WEEKLY ASH POND EFFLUENT DATA FROM 1970 THROUGH 1975
Plant A (Bottom Ash)
Parameter
Flow (GPM)
PH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCO )
J
Hardness
(mg/1 as CaCO,,)
J
Conductivity
([Jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max
Avg.
Min.
Max
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
39500
21233
700
8.1
7.0
5.3
<1
120
60
10
800
205
87
1700
376
210
87A
295
106
798
203
50
590
96
5
70
32
15
1971
21000
13586
8000
7.8
7.3
5.7
<1
155
87
38
260
166
100
510
314
220
743
312
149
688
211
111
359
100
3
91
34
6
1972
25000
15654
7800
8.0
7.4
6.5
<1
124
86
42
324
180
110
415
315
205
624
236
109
285
172
54
351
64
5
94
28
0
1973
22800
17792
4500
7.9
7.1
4.1
<1
120
80
20
260
153
90
730
331
215
1030
242
103
404
176
69
657
66
7
410
40
11
1974
23000
15415
5000
7.9
7.2
4.1
<1
160
88
49
394
130
76
910
313
210
394
212
99
366
158
77
274
54
5
96
32
10
1975
23000
20189
13320
7.8
7.1
6.1
<1
110
47
20
NA
NA
754
224
113
342
172
70
412
51
8
NA1
1. NA = data not available
-------
-12-
TABLE 3 (continued)
Plant A (Fly Ash)
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaCCL)
Total Alkalinity
(mg/1 as CaCO.,)
j
Hardness
(rag/1 as CaC00)
J
Conductivity
(|jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
5000
1250
200
7.6
6.5
3.5
<1
120
68
25
900
258
15
2100
815
245
3941
753
150
2000
488
25
1199
265
<1
150
50
12
1971
9740
5460
300
7.1
4.8
3.7
<1
60
29
7
690
384
136
1500
846
400
1224
694
71
1165
546
239
210
148
2
91
17
2
1972
13000
8175
5880
7.4
5.4
4.0
<1
65
28
5
590
445
170
970
800
300
957
640
293
730
550
200
426
89
5
59
14
5
1973
10190
6617
4500
6.3
4.5
3.7
<1
70
23
2
520
350
185
1010
809
640
893
606
370
820
513
141
256
93
15
36
14
4
1974
8700
6219
3100
5.8
4.2
3.6
<1
20
11
3
455
280
196
1125
813
615
737
545
253
734
508
241
220
37
3
26
12
1
1975
17910
7166
3000
7.1
4.0
3.4
<1
35
22
11
NA
NA
799
545
284
781
530
276
51
15
3
NA
-------
-13-
TABLE 3 (continued)
Plant B (Bottom Ash)
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaCOj
Total Alkalinity
(mg/1 as CaCCL)
J
Hardness
(mg/1 as CaCOj
j
Conductivity
((Jmhos/ctn)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1973
NA
9.1
8.2
7.6
4
<1
<1
73
59
44
108
79
56
260
200
160
914
224
76
500
145
33
706
79
7
50
43
<25
1974
NA
9.5
8.0
6.5
19
14
2
72
54
6
225
106
60
490
247
160
501
219
29
474
152
5
202
67
7
35
25
20
1975
NA
8.5
8.0
7.4
2
<1
<1
65
54
45
NA
NA
351
137
55
186
112
37
196
25
6
NA
-------
-14-
TABLE 3 (continued)
Plant B (Fly Ash)
Parameter
Flow (GPM)
PH
Phenolphthalein
Alkalinity
(mg/1 as CaCCL)
Total Alkalinity
(rag/1 as CaC00)
j
Hardness
(mg/1 as CaCCL)
o
Conductivity
(|Jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1973
NA
11.4
9.2
6.5
258
53
<1
300
119
35
512
354
195
1800
888
395
1690
724
319
1224
624
199
1491
100
6
30
25
<25
1974
NA
11.3
9.4
5.0
176
37
5
226
76
17
505
304
125
1600
688
310
996
582
122
812
488
84
470
94
7
25
24
10
1975
NA
11.0
9.2
5.3
65
24
3
100
58
11
NA
NA
719
479
110
711
461
101
157
18
5
NA
-------
TABLE 3 (continued)
-15-
Plant C - East
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCO )
J
Hardness
(mg/1 as CaCO )
J
Conductivity
(|jrahos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
11100
6410
3525
8.2
7.5
6.6
<1
169
92
60
246
194
144
576
441
310
535
360
261
412
312
216
252
48
4
85
38
5
1971
8975
5710
3400
7.9
7.3
6.6
<1
132
90
40
270
188
126
675
467
312
572
350
247
434
304
202
198
46
3
60
31
5
1972
9152
6050
3400
8.2
7.2
6.4
<1
118
64
10
390
210
90
670
476
195
539
358
213
488
318
140
108
39
5
58
28
10
1973
9475
7090
4171
7.6
7.2
6.4
<1
120
72
24
280
203
68
662
469
195
527
367
200
510
324
124
202
43
1
76
30
20
1974
9850
7690
3683
7.8
7.1
6.4
<1
140
69
24
350
222
25
713
521
250
881
409
207
524
360
180
614
48
5
90
33
10
1975
11100
8410
5463
7.7
7.1
5.1
<1
114
71
8
NA
NA
628
370
200
489
343
165
317
27
3
NA
-------
-16-
TABLE 3 (continued)
Plant C - West
Parameter
Flow (GPM)
?H
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCCL)
j
Hardness
(mg/1 as CaCOj
j
Conductivity
(pmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
3200
1600
800
9.2
7.8
4.8
20
17
12
120
79
12
160
123
84
448
316
212
311
231
160
309
200
124
103
30
2
84
34
25
1971
1706
1636
900
9.6
7.8
5.4
22
14
8
120
81
14
200
133
82
558
351
230
393
240
166
374
218
136
87
21
<1
65
30
25
1972
2400
1685
1600
8.9
7.5
4.1
<1
94
58
5
180
124
80
458
326
216
338
226
176
290
200
134
156
26
2
68
29
15
1973
2467
1704
1139
8.5
7.6
5.6
<1
114
68
4
222
126
68
550
319
184
422
245
166
402
202
103
197
42
6
95
39
25
1974
1867
1620
800
8.7
7.2
3.9
<1
118
70
4
272
132
80
775
340
208
641
259
120
629
215
100
129
43
7
90
36
<15
1975
1867
1631
800
8.6
7.5
4.1
<1
108
69
30
NA
NA
454
269
212
447
234
149
141
35
3
NA
-------
-17-
TABLE 3 (continued)
Plant D
Parameter
Flow (GPM)
pH
Pheno Iphtha le in
Alkalinity
(mg/1 as CaCCL)
Total Alkalinity
(mg/1 as CaCCL)
J
Hardness
(mg/1 as CaCC- )
O
Conductivity
(pmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
7200
6760
2800
8.0
6.5
4.3
<1
95
29
3
200
149
102
605
349
215
1189
337
158
457
244
106
931
93
4
65
32
<25
1971
9470
6923
2800
9.7
8.4
6.6
12
8
1
137
50
28
220
146
100
460
298
195
299
192
66
272
171
56
107
21
1
60
26
<25
1972
15840
5935
600
9.7
8.5
6.6
14
6
1
70
54
37
195
142
90
385
290
200
306
185
75
300
164
45
235
21
2
40
25
<25
1973
14000
7983
1050
9.3
8.6
7.7
19
6
1
97
60
37
183
130
68
380
259
190
455
168
54
375
143
40
193
25
<1
30
25
<25
1974
14590
8103
820
9.1
8.4
7.4
12
5
2
78
56
35
145
123
107
304
271
233
250
178
121
223
161
82
61
17
3
40
26
<25
1975
16470
8586
1050
9.3
8.4
7.5
13
4
1
104
69
48
NA
NA
279
183
130
247
168
122
50
15
3
NA
-------
-18-
TABLE 3 (continued)
Plant E (New Ash Pond Started 7-1-74)
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaCO.)
Total Alkalinity
(mg/1 as CaCO.J
J
Hardness
(mg/1 as CaCO.)
.3
Conductivity
((jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/D
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
OLD
1974
6850
3400
0
12
10.8
7.5
400
142
5
490
176
22
670
266
80
2100
796
195
866
369
122
804
340
110
123
29
3
NA
New
1974
6650
6057
5650
11.5
11.2
10.2
200
114
26
500
154
42
400
288
76
1150
819
285
535
393
138
522
381
135
43
12
2
NA
1975
6850
5658
4380
11.5
11.1
10.4
180
97
35
240
128
53
NA
NA
600
404
223
598
398
220
12
4
<1
NA
-------
-19-
TABLE 3 (continued)
Plant F
Parameter
Flow (GPM)
PH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCO )
Hardness
(mg/1 as CaCO_)
J
Conductivity
(|Jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(jcin
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1974
40000
32940
23000
11.4
11.1
10.5
150
98
21
173
115
33
400
304
88
1250
915
450
795
472
14
648
431
12
182
40
1
25
14
1
1975
35000
28293
15000
11.2
10.7
9.1
126
67
8
140
78
28
NA
NA
874
392
111
871
386
105
43
6
<1
NA
-------
TABLE 3 (continued)
-20-
Plant G
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCO_)
o
Hardness
(mg/1 as CaCO )
j
Conductivity
(pmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
(New Pond Started
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
NA
10.6
6.9
4.0
66
4
<1
85
18
0
375
203
100
630
452
128
492
318
142
437
295
135
137
22
<1
NA
in 1970 Began Sampling in
OLD
1971
NA
9.5
4.7
3.2
12
<1
<1
28
2
0
300
248
161
870
586
355
636
445
228
618
425
219
198
20
1
NA
1972
NA
8.2
5.7
3.3
<1
60
17
0
630
226
38
1100
568
180
1004
455
112
994
429
79
93
26
3
NA
1973
10000
10000
10000
10.4
9.8
9.1
38
25
12
72
51
36
360
199
150
480
387
295
345
280
227
324
261
213
59
19
1
<25
<25
<25
1973)
NEW
1974
10000
7347
2500
10.5
9.5
8.2
44
19
2
66
44
20
244
197
160
500
345
41
400
316
225
381
296
210
64
20
3
<25
<25
<25
1975
7500
4826
2500
10.4
9.4
7.5
38
16
6
72
44
22
NA
NA
365
251
182
323
232
164
74
19
5
NA
-------
-21-
TABLE 3 (continued)
Plant H
Parameter
Flow (GPM)
PH
Phenolphthalein
Alkalinity
(mg/1 as CaCO )
Total Alkalinity
(mg/1 as CaCO-)
J
Hardness
(rag/1 as CaCO_)
-J
Conductivity
(pmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
3400
2926
100
9.4
8.4
7.6
20
10
2
100
70
50
260
133
74
1300
704
380
742
422
267
728
400
7
295
21
2
40
26
<25
1971
3400
2639
324
8.9
8.1
7.0
18
7
1
104
75
8
300
138
80
830
517
330
572
354
235
564
337
214
56
17
5
45
26
<25
1972
3662
2631
1584
8.7
8.0
7.5
5
4
3
107
73
45
150
103
60
490
353
280
331
255
188
307
236
169
71
19
2
60
27
<25
1973
3362
3261
3175
9.6
8.8
7.6
25
14
3
100
78
55
155
117
80
500
395
270
372
284
210
364
268
200
90
16
3
80
27
<25
1974
3362
2583
25
9.4
8.3
7.3
20
9
1
95
61
20
140
114
80
480
389
220
379
265
118
365
250
100
103
15
3
35
25
2
1975
3362
2233
1715
9.4
8.7
7.1
68
11
1
120
64
34
NA
NA
411
305
190
376
292
176
35
12
3
NA
-------
-22-
TABLE 3 (continued)
Plant I (North Outfall)
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaC00)
J
Hardness
(mg/1 as CaC00)
j
Conductivity
(|Jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
NA
12.1
11.3
10.4
284
128
45
290
151
66
302
193
112
1400
735
320
816
396
237
800
371
233
97
24
<1
<25
<25
<25
1971
5565
4614
3363
12
11.3
10
190
116
21
230
140
70
293
181
115
960
635
250
420
293
188
414
269
171
123
23
<1
<25
<25
<25
1972
9021
4971
3353
12.3
11.3
10.7
190
119
51
215
140
74
550
232
117
1140
653
375
376
278
202
349
255
149
128
24
2
<25
<25
<25
1973
18132
5740
656
12
11.4
10.5
290
144
50
317
165
70
420
231
30
1400
728
280
491
307
181
470
285
166
75
22
2
<25
<25
<25
1974
23964
11124
3160
10.7
11.1
10.6
204
108
55
225
134
78
318
191
112
960
570
365
429
285
163
409
264
142
102
21
1
25
13
1
1975
No
Discharge
At
Present
-------
-23-
TABLE 3 (continued)
Plant I (South Outfall)
Parameter
Flow (GPM)
PH
Pheno Iphtha le in
Alkalinity
(mg/1 as CaCO )
Total Alkalinity
(mg/1 as CaCO )
J
Hardness
(mg/1 as CaC03)
Conductivity
((jmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
NA
12.1
11.3
10.3
280
118
34
284
138
55
285
189
131
1197
718
340
836
400
229
823
380
225
83
20
<1
<25
<25
<25
1971
8572
7409
6149
12
11.3
10
196
114
20
215
136
55
295
177
97
990
615
265
432
284
179
412
266
148
116
18
<1
<25
<25
<25
1972
9021
7870
5969
12.1
11.3
10.8
200
120
62
230
139
80
570
235
128
1100
633
330
403
282
182
348
249
154
114
33
<1
<25
<25
<25
1973
17144
9830
1212
11.9
11.3
10.7
240
140
96
260
162
105
320
224
44
1120
730
400
506
301
192
443
277
182
63
24
1
<25
<25
<25
1974
23518
13786
2190
11.6
11.1
10
193
106
35
233
132
60
320
199
120
915
575
250
547
292
162
441
217
160
275
15
<1
25
12
1
1975
32178
27000
19791
11.1
9.8
8.9
65
27
7
94
71
32
NA
NA
364
240
75
345
223
65
78
18
2
NA
-------
-24-
TABLE 3 (continued)
Plant J
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCCL)
o
Hardness
(mg/1 as CaCCO
,3
Conductivity
(pmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
14810
14159
8886
8.3
6.0
3.3
1
1
1
52
25
3
128
106
73
415
323
250
375
210
84
362
197
78
64
14
2
25
22
<2
1971
19840
13840
10960
8.8
6.5
3.5
4
2
1
87
42
3
135
96
50
345
268
170
262
193
101
247
177
100
72
16
<1
31
12
2
1972
18320
14841
12200
8.7
6.1
3.3
5
2
1
84
41
3
134
101
62
440
284
200
501
193
113
235
168
102
360
26
<1
28
8
2
1973
19840
15457
10880
8.2
6.0
3.6
<1
82
37
3
151
104
70
550
325
230
341
233
159
284
201
137
128
32
2
95
44
7
1974
24140
11860
3460
8.8
6.5
3.3
9
5
2
96
47
3
152
102
2
465
298
215
617
240
92
294
193
66
431
47
1
74
11
2
1975
28000
14870
9700
9.1
6.2
3.4
13
4
1
81
36
2
NA
NA
719
206
93
282
168
80
542
38
3
NA
-------
TABLE 3 (continued)
-25-
Plant K
Parameter
Flow (GPM)
pH
Phenolphthalein
Alkalinity
(mg/1 as CaCO )
Total Alkalinity
(mg/1 as CaCCL)
3
Hardness
(mg/1 as CaC03)
Conductivity
(pmhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
NA
11.8
11.2
10.0
304
139
37
362
188
68
380
236
88
1650
969
390
653
405
257
510
351
179
374
53
2
<25
<25
<25
1971
16000
13184
8000
11.7
11.4
11.1
330
160
102
372
191
130
422
264
212
1900
967
650
566
373
256
564
370
253
20
3
1
<25
<25
<25
1972
35000
16323
1500
12.5
11.4
11.0
358
153
50
400
181
72
460
238
112
2000
1046
390
508
320
117
442
314
116
66
6
1
<25
<25
<25
1973
40500
18172
4500
11.4
11.0
10.5
151
81
35
187
112
58
242
175
132
720
507
27
318
215
30
310
203
23
37
13
2
<25
<25
<25
1974
37500
25859
18000
11.4
11.0
9.4
115
69
15
146
103
70
231
173
118
680
438
280
427
288
131
416
272
106
59
16
2
<25
<25
<25
1975
37000
23311
18000
11.2
10.3
8.9
102
40
6
133
84
54
NA
NA
966
319
187
404
268
172
273
29
6
NA
-------
-26-
TABLE 3 (continued)
Plant L
Parameter
Flow (GPM)
PH
Phenolphthalein
Alkalinity
(mg/1 as CaC03)
Total Alkalinity
(mg/1 as CaCCL)
j
Hardness
(mg/1 as CaC00)
o
Conductivity
((Jhos/cm)
Total Solids
(mg/1)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
Max.
Avg.
Min.
1970
17222
14189
10830
11.3
10.2
9.1
100
39
14
135
78
40
238
161
110
530
351
220
539
246
124
296
210
103
329
36
3
85
29
<25
1971
17546
14370
9597
10.7
9.4
7.2
60
21
2
110
74
36
210
143
97
380
327
250
496
244
180
286
210
154
210
34
<1
45
27
<25
1972
17000
13223
10000
10.5
9.2
8.0
40
16
2
95
76
27
210
132
88
440
312
220
308
198
80
290
185
72
34
13
2
42
26
<25
1973
18500
15188
11000
11.1
10.0
8.0
95
36
1
140
88
53
320
171
105
710
348
170
320
222
122
318
206
91
90
17
<1
41
26
<25
1974
17000
13698
8000
11.5
10.3
9.1
204
50
15
214
96
52
310
185
105
870
371
200
352
242
130
338
223
118
92
20
4
25
21
5
1975
19000
14596
8000
11.2
10.4
9.4
102
46
15
150
74
40
NA
NA
288
217
108
284
209
100
52
8
2
NA
-------
-27-
The yearly average suspended solids concentration decreased with time
over the period for which data is given in Table 3 for over half of the
ash ponds. Exceptions to this are the ponds at Plants G, H, I, J, and K
and the west pond at Plant C where they remained constant or increased.
The yearly dissolved solids averages in a particular ash pond
effluent varied from year to year by as high as 167 mg/1 at Plant K to
as low as 34 mg/1 at the west pond at Plant C. Likewise, the yearly
total solids average varied from as high as 245 mg/1 for the fly ash
pond at Plant B to as low as 35 mg/1 at Plant E.
The amount of suspended solids variation from year to year due to
natural background variation is hard to determine because of changes in
the ash pond structures in an effort to lower suspended solids loadings
from the ponds and changes in analytical procedures. However, the trend
of decreasing suspended solids was probably a result of efforts to
reduce the suspended solids concentrations to levels below 30 mg/1.
Therefore, the decrease was not observed at Plants G, H, and I because
they were already below 30 mg/1. The effluent concentrations at Plant J
and K actually increased from yearly averages in the teens to yearly
averages near 30 mg/1. These increases were probably associated with a
decrease in the pond hydraulic detention times and an increase in the
ash and dirt content of the coal.
The yearly average alkalinity for a particular ash pond did not
vary by more than 40 mg/1 as CaC03, except for the fly ash ponds at
Plants A (57) and B (61) and the combined pond at Plant K (107). The
yearly average hardness did not differ by more than 60 mg/1 from year to
year except for the ponds at Plants A bottom ash (75) and fly ash (187)
and K (91). The average yearly conductivity varied by more than 100
|jmhos/cm for the ash ponds at Plants H (351), I (165), and K (539) and
the fly ash pond at Plant B (200). There is nothing significant about
the values of 40, 60, and 100. They were presented only as a reference
and for the sake of comparison.
The ash pond at Plant I had two distinct discharges from 1970 to
1974 (north and south). The water quality characteristics as reported
in Table 3 are similar for each outfall. There is a difference in the
flow rate and suspended solids concentration of the two effluents which
may be affected by the detention time or flow pattern of the ash pond,
but with the limited data this is hard to verify. However, the data
appear to indicate that location of the outfall does not effect the
water quality characteristics of the effluent provided adequate ash
settling times are provided and the water within the ash pond is well
mixed.
The weekly effluent data for each pond were plotted with respect to
time. Examples of these plots are shown in Figures 2 and 3. The type
of trends exhibited by each ash pond effluent characteristic was deter-
mined by observation of these plots. Figure 2 is representative of the
trends displayed by the ponds in which there is no yearly pH cycle,
while Figure 3 is representative of those with a yearly pH cycle. A
summary of the type of trends exhibited by each pond is given in Table 4.
The type of trend has been defined as cyclic (Yes) or noncyclic (No)
within a one year period.
-------
12
11
/"\
•
.**
§18
-o
§
*• 9
8 -
7
1874
1388
g888
o
o
I486
!288
1975
1976
1974
1975
1976
8808
6808
4888
2808
1974
1975
YEAR
197C
1088
888
s
*680
V.
a
E
288
1974
1975
YEAR
1976
Figure 2. Variation of Plant E Ash Pond Effluent Characteristics with Time
for the Period 1974 to 1976
to
00
i
-------
680
in
a
a
268
2888
1588
1888
see
1971 1972 1973 1974 1975 1976
8
1978
1971 1972 1973 197-4 1975 1976
•wu
^388
>
\s
Q
HI
0288
a
Lu
o
v>
2 188
8
19
. •
•
••
• • • 1
*.\
. .' * " .*• *• '
•f .JL : ':\ fi^y^ V>V:
•'• *'I*I; 1*" ."*'" x' •*"• •
•- '
, • •
«
78 1971 1972 1973 1974 1975 19
YEAR
OHJM
/•^
^ 488
V^
( —
M
o 290
o
o
8
76 19
•
•
• •
• ' •
' » •
• « •
•» • v •
• • *
• • *. • • • m
»•»*•' "• . . fc " • •' * * §
.*%•* • *£? *• ."r* • *. -:f •
• •• • " •• / • ^ "•• *jr
•£ :*• "• 'O
• •
•
^ ,.. i .... i .... i .... i .... i • - ••
78 1971 1972 1973 1874 1975 19
YEAR
76
Figure 2 (Continued)
-------
10
a
^t
§
T
r
X
a
>
7-
"V •
••
.
..-
'
o
W
0
X
a
E
158 f-
188 h
*-
1978 1971 1972 1973 1974 1975 1976
1978 1971 1972 1973 1974 1975 1976
20888
3
O
18009
• * •
t •
158
O
o
0
o
a. 108
OT
d
a:
58
1978 1971 1972
1973
YEAR
1974 1975 1976
0
1970 1971 1972
o
I
1973
YEAR
1974 1975 1976
Figure 3. Variation of Plant J Ash Pond Effluent Characteristics with Time
for the Period 1970 to 1976
-------
600
Ct
LJ
288
2800
1500
en
E
1000
500
1975 1976
0
1970
1971
1972 1973 1974 1875 1976
•too
^300
X.
E
\s
M
0200
DISSOLVED
i
8
19
. •
m
* * • •
•" "\
• 9 * •. • •
• • 9 «^ •
•*^ " % • " " **m * \*« i •*•••
/, -'"*'• *'• "••" . •' '•:'.
•* • •«. /
••«• . i •
-
, • *
•
70 1971 1972 1973 1974 1975 19
YEAR
oo»
TY Cumho«/cm3
i
CONDUCTIVE
0
76 19
•
•
. "
• * •
• • •
. • *
,. " '. •-
f\ ..%•. ... •.:
• •
•
70 1971 1972 1973 1974 1975 19
YEAR
76
u>
I
Figure 3 (Continued)
-------
TABLE 4. TVA ASH PONDS WHICH SHOktD A YEARLY CYCLE
Parameter Plant A Plant B Plant C Plant D
Bottoa Fly Bottom Fly
Ash Ash Asa Ash East West
Flow No No No No No No No
Pheaolphthaliea No No No No No No Vo
Alkalinity
Turbidity No No No No No No No
Suspended Solids No No No No No No No
Hardness Ye* No Yes Yes Yes
Yes - Indicates cycle
Plant E Plant f
No No
Yes Yes
No
Ho No
Yes Yes
Yes
Plant G Plant H Plant [ Plant J Plant K
Old
No
No
No
No
No
No
North South
New Outfall Outfall
No No No No No No
No Yes Yes Yes No No
No No No Ho No No
No No No No No No
No No Yes Yes Yes No
Plant L
No
No
No
No
1
CO
ro
i
-------
-33-
From this sunanary suspended solids, flow, and turbidity showed no
yearly cycle at any of the ash ponds. Seven of the ash ponds exhibited
a yearly pH cycle. These same seven ash ponds exhibited yearly cycles
for total alkalinity, conductivity, dissolved solids, and total solids.
Five out of seven ponds also exhibited a yearly cycle for hardness. For
those ash ponds which had no yearly pH cycle, only three showed yearly
cycles in total alkalinity, dissolved solids, and total solids.
These plots revealed an interesting trend in the alkalinity of most
of the ash pond effluents. The total alkalinity is usually approximately
50 mg/1 as CaC03 between December and April for all pond effluents. From
this time on, except for the effluents from the ponds for Plant C, D, and
J and the fly ash pond for Plant A, the total alkalinity either increases
or remains approximately the same, rarely dropping below 50 mg/1 as CaC03.
The pH normally followed the same pattern as the total alkalinity with
the pH being the lowest during the later part of the year or first part
of the next year. This relationship between effluent pH and effluent
alkalinity is not surprising since alkalinity increases with pH.
In 1973, TVA began collecting ash pond effluent and water intake
samples quarterly for trace metal and calcium, chloride and silica
analyses. A summary of this data for 1973 through 1975 is given in
Table 5. Discussion of these data collected prior to 1973 was excluded
because it was collected at infrequent intervals. The summary consists
of the average, maximum, and minimum concentrations for each element.
The average was calculated by substituting a value equal to the minimum
detectable amount (MDA) when the reported value was less than the MDA.
Thus, the average may be biased upward if there are a significant number
of reported values less than the MDA. Those parameters most likely
affected are As, Ba, Be, Cd, Cr, Pb, Hg, Ni, and Se.
The average values for the ash pond effluents given in Table 5 are
plotted in Figure 4 against the number of ash ponds equal to or exceed-
ing that average concentration. For example, 7 of the 15 ash ponds have
an aluminum concentration greater than or equal to 2 mg/1. Figure 4
also allows a known average concentration of a particular element in the
effluent of one ash pond to be compared with the concentrations of that
element in the other TVA ash pond effluents.
The average concentrations of calcium, chloride, iron, magnesium
and manganese varied considerably from effluent to effluent while the
average concentrations of aluminum, arsenic, silica, and sulfate varied
slightly from effluent to effluent. The average concentrations of
barium, cadmium, chromium, copper, lead, mercury, nickel, selenium, and
zinc were approximately the same in all the ash pond effluents. However,
the fly ash pond at Plant A had considerably higher concentrations of
aluminum, cadmium, chromium, copper, lead, nickel, silica, sulfate, and
zinc than any of the other effluents. The combined ash pond effluent at
Plant D had a considerably higher concentration of selenium than the
rest of the effluents, while the ash pond effluent from Plant H had a
considerably higher concentration of arsenic than the others. Except
for the fly ash pond at Plant A (0.75 mg/1) and the combine ponds for
Plant H (0.34 mg/1) and L (0.52 mg/1), the average effluent ammonia
concentration was less than 0.2 mg/1. These ammonia concentrations come
primarily from the intake water; however, peak concentrations may result
during metal cleaning operations within the plant.
-------
-34-
TABLE 5. SUMMARY OF QUARTERLY TRACE METAL DATA FOR ASH POND INTAKE AND EFFLUENT STREAMS
Aluminium
Ammonia aa N
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chloride
Chromium
Copper
Cyanide
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
Selenium
Silicii
Silver
Diuolved Solid!
Suspended Solids
Sulf.te
Zinc
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFT
RW
EFF
RW
EfT
RW
EFF
RW
EFT
RW
EFT
RW
EFT
RW
EFF
RW
EFT
RW
EFF
RW
EFT
RW
EFF
RW
EFF
RW
EFF
RW
EIT
RW
EFF
RW
EFF
RV
EFF
RW
EFF
RW
1
Minimum
0.5
0.5
0 04
0.02
<0 005
<0 005
<0.1
<0.1
<0.01
CO. 01
CO. 001
<0.001
23
21
4
4
<0.005
CO. 005
0.01
0.04
<0.01
1.7
1.1
CO. 010
<0.010
0.3
4.1
0.07
O.OB
<0.0002
<0.0002
<0.05
<0.05
h
Maximum
8.6
1.6
0.31
0.08
0.055
CO. 005
0.3
co. 1
co. 01
CO. 01
5.01
0.01
200
20
11
7
0,026
CO.OOS
0.20
0.02
CO. 01
30
0.90
0.048
CO. 01
21
4.7
3.6
0.08
0.0042
CO. 0002
0.14
CO. 05
0.056
0.002
22
7.2
CO. 01
0.05
710
100
78
14
470
18
0.55
0.04
Minimum
0.6
0.4
CO. 01
0.04
<0.005
CO.OOS
CO. I
cO.l
<0.0!
CO. 01
CO. 001
CO. 001
27
17
4
4
<0 . 005
CO. 005
CO. 01
CO. 01
CO. 01
0.14
0.32
CO. 01
CO. 01
0.2
3.6
0.02
0.04
CO. 0002
CO. 0002
CO. 05
<0.05
0.001
C0>002
3.1
3.2
CO. 01
0.01
40
90
2
8
17
9
0.01
0.01
Plant 8
Fly Ash
Average
1.6
0.8
0.07
0.08
0.029
CO.OOS
O.I
<0.1
<0.01
CO. 01
0.001
0.004
152
19
6
5
0.013
CO. 005
0.03
0.02
CO. 01
1.4
0.57
0.015
<0.01
3.6
4.3
0.12
0.06
0.0008
CO. 0002
0.05
CO. 05
0.015
CO. 002
7.1
5.4
CO. 01
0.02
458
93
13
11
214
12
0.05
0.02
Maximum
4.8
1.6
0.20
0.08
0.070
<0.005
0.2
co.l
<0.01
<0.01
0.002
0.01
430
20
8
7
0.036
CO. 005
0.10
0.02
CO. 01
7.1
0.90
0.030
<0.01
6.8
4.7
0.63
0.08
0.0056
CO. 0002
0.08
<0.05
O.C64
CO. 002
22
7.2
CO. 01
0.05
1100
100
39
H
480
18
0.13
0.04
Effluent d»ta bated on yean 1973-1975
Raw water Intake data baaed on yean 1974 and 1975
-------
-35-
TABLE 5 (Continued)
Aluminum
Ammonia as N
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chloride
Chromium
Copper
Cyanide
I ron
Lead
Magnrs mm
Manganese
Mercury
Nickel
Seieni um
Silicia
Si Iver
Dissolved Solids
Suspended Solids
SuHate
Zinc
EFF
RW
EFF
RW
Err
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RU
EFF
RU
EFF
RU
EFF
RW
EFF
RU
EFF
RW
EFF
RW
F.FF
RW
EFF
RW
EFF
RW
EFF
RW
CFK
RU
F.tF
RU
Minimum
0.3
0.6
0.02
0.03
CO. 005
<0.005
<0. 1
<0.1
<0.01
<0.01
0.002
<0.001
45
15
7
7
CO. 005
<0. 005
<0.01
0.03
<0.01
0.33
1 .0
<0.010
<0.010
1.4
6.5
0.13
0.12
CO. 0002
<0. 0002
<0.05
<0.05
CO. 001
<0.001
4.7
55
CO. 01
<0.01
260
160
3
11
110
0.07
0.02
O.OJ
Plant C
Average
1.5
4.7
0.11
0. 14
0.013
o.ooe
0.2
0.1
<0.01
<0.01
0.006
0.001
78
29
11
11
0.006
0.012
O.Ob
0,11
0.01
1.7
6.5
0.021
0.022
10
9.5
0.20
0.31
0.0034
0.0004
0.05
<0.05
0.010
0.002
7.4
6. 1
0.01
0.01
345
205
18
46
15fl
23
0. 13
0.08
Maximum
3.8
15
0.34
0.33
0.05
0.026
0.4
0.2
<0.01
<0.01
0.013
0.002
100
45
16
16
0.008
0.041
0.10
0.22
<0.01
4.1
14
0.069
0.047
16
14
0.34
0.53
0.0074
0.0016
0.07
<0.05
0.080
0.004
11
7.9
0.01
<0.01
460
240
37
150
200
52
0.27
0. 13
Minimum
0.5
1.3
<0.02
0.03
<0.005
<0.005
<0.1
<0.1
<0.01
<0,01
<0. 001
<0.001
19
15
8
7
<0.005
<0.005
<0.01
0.03
<0.01
0.72
1.4
<0.010
<0.010
6.3
6.5
0.05
0. 12
CO. 0002
<0.0002
<0.05
<0.05
CO. 001
-------
TABU 5 (Continued)
-36-
--
Aluminum
Ammonia as N
Arsenic
Ba rium
Beryllium
Cadmium
Calcium
Chloride
Chromium
Copper
Cyanide
I ron
Lead
Magnes i urn
Manganese
Mercury
Nickel
Selenium
Si licia
Silver
Dissolved Solids
Suspended Solids
SuHate
Zmi
EFF
RW
EFF
RW
EFF
RW
EFF
RU
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
KFF
KW
Minimum
0.8
<0.1
0.03
0.02
-------
-37-
TABLE 5 (Continued)
A 1 urn i mum
Ammonia as N
Arsenic
Barium
Beryllium
Cadmi UA
Calcium
Chloride
Chromi um
Copper
Cyanide
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
Selenium
Silicia
Silver
Dissolved Solids
Suspended Solids
Sulfate
Zinc
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
RW
EFF
EFF
EFF
RW
EFK
RW
Minimun
0.4
0.3
O.CI
0.01
0.005
0.005
<0.1
<0.1
CO. 01
<0.01
<0.00]
<0.001
20
4
2
2
<0.005
<0.005
0.02
<0.01
<0.01
0.1
0.26
<0.010
<0.010
3.9
1.2
0.05
0.03
<0.0002
< 0.0002
<0.05
<0.05
<0.001
'0.001
3.5
1.0
<0.01
<0.01
140
30
1
5
56
9
0.02
0.03
Plant J
Average
2.6
0.7
0.05
0.04
0.041
0.018
0.2
0.2
CO. 01
<0.01
0.001
0.001
34
15
5
2
0.005
0.005
0.11
0.08
<0.01
2.4
0.7
0.015
0.010
6.7
4.5
0.38
0.07
0.0003
0.0003
0.05
<0.05
0.004
0.003
6.4
3.9
<0.01
<0.01
202
89
15
13
119
22
0.07
0.06
Maximum
7.6
1.4
0.08
0.23
0. 130
0.110
0.3
0.4
-------
-38-
c
o
2
4J
e
0)
u
B
O
15 13 II 9 7 5 3
Number of ash ponds equal to or exceeding
a Riven concentration
Figure 4. Number of TVA Ash Ponds Whose Average Effluent
Concentration Equals or Exceeds Various Given
Concentrations
-------
-39-
0.14
0.12
0.10
0.08
e
5
4J
2
4J
s
o
u
I 1 I 1 1 1 1 1 1 1 1 1 1 1
0.03 -
0.02 -
0.01 -
QOO u—
15
13
Number of ash ponds equal to or exceeding
a given concentration
Figure A (Continued)
-------
-40-
-, , , ,
a
o
1.4
12
LO
0.8
0.6
0.4 -
0.2 -
0.0
O.7 -
0.6 -
0.5
0.4
0.3 -
0.2-
0.1
0.0
D Bo
O Zn
O NH
D Cu
A Mn
15
Number of ash ponds equal to or exceeding
a given concentration
Figure 4 (Continued)
-------
-41-
600
500
400
300
200
100
4J
£
3
140
120
100
80
60
40
20
O Susp. Solids
A Co
15
13
II 9 7 5 3
Number of ash ponds equal to or exceeding
a given concentration
Figure 4 (Continued)
-------
-42-
The concentrations of most metals in the ash pond effluent appear
to vary with time. The amount of variation differs for each element
within each effluent. This is indicated by the difference between the
minimum and maximum values for each element. A more detailed statistical
analysis to determine whether this variation was significant was not
performed because there were insufficient data.
RELATIONSHIPS BETWEEN PLANT OPERATION CONDITIONS AND ASH
POND EFFLUENT CHARACTERISTICS
Relationships between the ash pond effluent and the plant operating
conditions were developed on a plant-to-plant basis in order to increase
knowledge of the system and aid in developing an ash pond effluent
monitoring strategy. Plant operating conditions are defined as those
parameters or processes which can vary with time either with or without
man's control. They include such things as coal characteristics, raw
water quality characteristics, ash collector efficiencies and quantities
of coal burned. A summary of the plant operating conditions, ash
characteristics, and ash pond effluent characteristics at each plant are
given in Table 6. These data are rough estimates and represent average
values for varying time periods with a span of five years. They were
obtained from various sources and most often the data were collected for
other purposes. For example, the coal data were obtained from TVA
Division of Power Production. The ash characteristics were obtained
from routine analyis conducted by TVA's Singleton Materials Engineering
Laboratory in Knoxville, Tennessee. The intake and effluent data were
obtained from analysis of routine samples collected by personnel of
TVA's Division of Power Production and analyzed by TVA's Laboratory
Branch of the Division of Environmental Planning. The average values
are also for varying periods of time within a span of five years.
Linear correlation coefficients were developed for the data shown
in Table 6. The significant coefficients at the 95 percent confidence
level are given in Table 7. They are based only on the data for Plants
C-L. The data for Plants A and B were excluded from the correlation
analysis because these two plants operate separate ash ponds for fly ash
and bottom ash. The data for Plant C was included although it has
cyclone boilers. Therefore the ash produced at Plant C is about 60
percent bottom ash, whereas at the other plants, which have pulverized
coal boilers, the ash produced is only about 30 percent bottom ash.
Unfortunately no data were available on the bottom ash characteristics.
Correlation is a measure of the degree of association between
parameters and may give valuable insight into the relationships between
plant operating conditions and the ash pond effluent. Even though the
data in Table 6 are only estimates, the correlations can be used to group
parameters that behave similarly and identify pairs of parameters that
should be plotted and studied more carefully as predictors. The corre-
lation coefficient depends primarily on the amount of variation for the
parameters as well as on their measured error and the actual relation
between them. In general, a large correlation coefficient between two
parameters may be due not so much to a direct relationship between them
as much as to their common dependence on other parameters ("lurking
variables"). For example, two seasonal parameters may show a strong
-------
TABLE 6. SUMMARY OF PLAltT OPERATION CONDITIONS AND ASH POND EFFLUENT CHARACTERISTICS OF TVA COAL-FIRED POWER PLANTS
Parameters Plant A
Method of Firing Cyclone
Coal Source W. Kentucky
Afifa Content io Coal, « 18.8
Fly Ash of Total Ash, I 30
Sulfur Content in Coal, I 4.1
Coal Usage at Full Load 22901
(tons/day)
Niartier of Units 3
ESP Efficiency, 1
Mechanical Ash Collector 98
Efficiency, X
Overall Efficiency, X 98
(gal/ ton) 98106
pH of Intake Water 7.7
Suspended Solids Concentration 60
(•g/l as CaCOj)
X SiOj in Fly Ash NA
X CaO in Fly Ash NA
X Fe20j in Fly Ash NA
X Al20j in Fly Ash NA
X HgO in Fly Ash NA
X SO in Fly Ash NA
X Moisture in Fly Ash NA
pH of Fly Ash NA
Ash Pond Effluent pH 4.4*
7._2°
Ash Pond Effluent Suspended 2s5
Solids («jg/l) 55
Fly ash pond only
Bottoa ash pond only
NOTE: Intake water characteristics based
Ash pond effluent characteristics
Plant B
Circular
Wall Burners
W. Kentucky
14.8
50
50
3314
4
-
-
-
7.5
41
56
NA
NA
NA
NA
NA
NA
NA
NA
9 «^
8V
«b
64b
Plant C
Cyclone
W. Kentucky
11
30
70
3.0
7848
3
-
90-99
23065
7.4
81
83
47.6
1.72
11.3
22.7
0.93
22
1.04
2.9
7.1°
30C
Plant D
Tangential
E. Kentucky
15.5
75
25
1.2
8240
1
99
-
99
10770
7.5
15
95
NA
NA
NA
NA
NA
NA
NA
NA
8.4°
19C
Plant E
Circular
Wall Burners
W. Kentucky
15.3
67
33
4.1
12897
5
74
80
95
9585
7.0
17
53
46.9
4.66
14.9
18.6
1.33
1.5
0 32
11.8
11. 1C
<10C
Plant F
Opposed
W. Kentucky
S. Illinois
16.3
80
20
3.7
24525
2
99
-
19490
7.4
24
69
NA
NA
NA
NA
NA
NA
NA
NA
11. 1C
10<
Plant G
Tangential
W. Kentucky
15.7
80
20
3.5
10503
4
60
-
98-99
12345
7.3
12
63
53.7
2.36
9.6
26.4
1.12
1.09
0.37
4.5
9.5C
20C
Plant H
Tangential
Virginia
E. Kentucky
E. Tennessee
15
67
33
1.8
8057
4
-
-
99
11425
7.0
21
73
52.5
2.19
10.2
25.5
1.42
1.9
0.63
3.6
8.7C
19C
Plant I
Wall Burner
W. Kentucky
14
70
30
3.7
14460
10
75
-
75.5
42430
7.4
IS
58
58.7
3.17
10.7
23.9
1.24
1.2
0.22
4.6
11.0C
19C
Plant J
E. Kentucky
E. Tennessee
19.1
75
25
2.1
16193
9
70
95
98
9520
7.6
IS
55
50.4
1.92
11.6
25.2
1.29
0.54
0.21
4.0
7.5°
25C
Plant K
Wall Burner
S. Illinois
W. Kentucky
15.6
75
25
2.8
15304
10
60
95
98
17265
7.6
38
66
NA
NA
NA
NA
NA
NA
NA
NA
10. 8C
1?C
Plant L
Wall Burner
U. Kentucky
X. Alabama
16
75
25
2.8
17691
8
60
99
70
15370
7.5
6
63
45.3
4.91
17.0
27.0
1.22
1.16
0.87
65
10. lc
15C
on 1974 and 1975 weekly savples.
based on 1970-1975 weekly sanples.
-------
-44-
TABLE 7. LINEAR CORRELATION COEFFICIENTS SIGNIFICANT AT THE 95%
LEVEL OF CONFIDENCE FOR PLANT OPERATING CONDITIONS
Parameter
Parameter
Correlation
Coefficient
Ash Pond Effluent pH
Ash Pond Effluent pH
Ash Pond Effluent pH
Suspended Solids Ash
Pond Effluent
Suspended Solids Ash
Pond Effluent
Mechanical Ash Collector
Efficiency %
CaO Content of Ash %
CaO Content of Ash %
Ash Content of Coal %
Suspended Solids in the
Ash Pond Effluent
Sulfur Content of Coal
CaO Content of Ash %
CaO Content of Ash %
Fly Ash pH
AUO- Content of Ash
Fe203 Content of Ash
Fly Ash pH
SO. Content of Ash %
J
-0.856
0.649
0.792
-0.859
-0.840
0.951
0.863
0.812
-0.863
-------
-45-
association as they fluctuate together over time. Lurking variables in
addition to time are to be found when looking at relationships from
plant to plant. A significant correlation at the 95 percent level of
confidence is one greater than 0.632 for all correlations except those
involving the fly ash characteristics. Significant correlations with
the fly ash characteristics are represented by an R value greater than
0.754. These values were obtained from Freund 1967 (2) and are based
on the number of data points used to determine R. As the number of data
points increases, the R value for the 95 percent level of confidence
decreases.
Not all correlations with R values greater than 0.632 or 0.754
represent meaningful relationships. For example, the R value for the
comparison of the pH of the intake water used for sluicing with the
efficiency of the mechanical ash collector was 0.903, but there is no
logical reason these two parameters should correlate. Therefore, this
coefficient represents a meaningless relationship. One reason that a
high coefficient was obtained was that coincidentally the lowest pH
value and the lowest mechanical ash collector efficiency occurred at
the same plant representing one sixth of the data.
Table 7 indicates that the pH of the ash pond effluent is mainly
influenced by the calcium content of the fly ash (R = 0.792) and the sulfur
content of the coal (R = 0.649). Since the sulfur content of coal varies
with its source the following generalization can be made. The effluents
from plants which receive coal from western Kentucky and southern Illinois
(sulfur content usually 2.8 to 4 percent) are basic while those from plants
which receive coal from eastern Tennessee, eastern Kentucky, and Virginia
(sulfur content usually 2 percent or below) are neutral or slightly acidic.
An exception to this is the effluents from the separate ponds at Plant A
and the combined pond at Plant C.
Suspended solids in the effluent exhibited significant (with a
95 percent confidence coefficient) negative correlations with the
percent of CaO in the fly ash and pH of the fly ash. There was also a
significant negative correlation between the effluent pH and the
effluent suspended solids which can not be explained. In addition, the
pH of the fly ash correlated significantly with the percent of CaO in the
ash.
RELATIONSHIP BETWEEN THE CHARACTERISTICS OF THE INTAKE WATER AND ASH
POND EFFLUENT
In 1974 and 1975 weekly samples of the intake water used for
sluicing at each plant were collected at approximately the same time as
the ash pond effluent samples and analyzed for pH, alkalinity, dissolved
solids, and suspended solids. A summary of this data is given in Table 8.
These weekly data were combined with the corresponding weekly ash pond
effluent data for 1974 and 1975 and linear correlation coefficients
developed for the four plants: E, G, H, and J. The R values for these
correlations are shown in Table 9 for these four plants. An R value
greater than 0.205 indicates a significant correlation of at least the
95 percent confidence level for Plants G, H, and J, while an R value of
0.273 indicated a significant correlation of at least the 95 percent
confidence level for Plant E.
-------
TABLE 8. SUMMARY OF WEEKLY ASH POND INTAKE WATER DATA FOR 1974 AND 1975
Parameter
pH
Phenolphthalein
Alkalinity
(mg/1 as Ca(X>3)
Total Alkalinity
(mg/1 as CaCO )
j
Hardness
(mg/1 as CaCO )
j
Conductivity
((jmhos/cm)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Plant
1974
8.1
7.7
7.4
<1
130
97
74
150
110
75
340
252
150
204
108
25
334
60
7
100
34
12
A
1975
8.2
7.6
7.1
<1
130
97
68
NA
NA
259
137
60
187
48
11
NA
Plant
1974
8.6
7.5
7.0
<1
71
56
28
80
57
55
200
142
70
164
96
11
230
41
1
25
24
15
B
1975
8.5
7.6
7.0
<1
64
54
35
NA
NA
137
94
34
145
23
8
NA
Plant
1974
7.8
7.4
6.8
<1
150
83
50
196
111
70
438
266
158
357
177
94
172
81
9
220
67
<25
C
1975
7.9
7.4
6.9
<1
144
92
56
NA
NA
339
229
177
155
45
15
NA
Plant
1974
8.5
7.8
7.5
9.0
1.2
1.0
120
95
63
142
114
91
298
220
180
223
125
60
44
15
2
27
25
<25
D
1975
8.6
7.8
7.3
6.0
0.7
<1
120
98
64
NA
NA
215
137
90
34
11
2
NA
-------
TABLE 8 (continued)
Parameter
PH
Phenolphthalein
Alkalinity
(mg/1 as CaCO )
Total Alkalinity
(mg/1 as CaCO-)
j
Hardness
(mg/1 as CaCO )
o
Conductivity
(pmhos/cm)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/D
Turbidity
(JCU)
Max
Avg
Win
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Plant
1974
8.3
7.4
7.0
<1
90
53
34
160
80
52
200
157
115
190
106
64
50
17
4
60
27
<22
E
1975
7.8
7.3
7.0
<1
58
47
34
NA
NA
187
102
34
184
22
3
NA
Plant
1974
8.2
7.8
7.4
<1
94
69
38
243
108
84
280
205
165
587
128
33
68
24
9
150
56
22
F
1975
8.2
7.7
6.8
<1
90
66
42
NA
NA
224
107
12
87
19
2
NA
Plant
1974
7.8
7.5
7.3
90
2
<1
90
63
52
140
72
62
230
184
115
156
116
79
81
24
5
59
31
25
G
1975
8.0
7.6
7.4
<1
90
56
40
NA
NA
136
91
32
175
18
2
NA
Plant
1974
8.4
7.0
7.6
5
<1
<1
90
73
8
95
80
60
396
256
195
250
190
122
64
21
5
90
30
<25
H
1975
8.6
7.9
7.5
<1
100
82
60
NA
NA
360
173
122
72
19
3
NA
-------
TABLE 8 (continued)
Parameter
pH
Phenolphthalein
Alkalinity
(mg/1 as CaCO )
Total Alkalinity
(mg/1 as CaCO )
J
Hardness
(mg/1 as CaCO )
j
Conductivity
((Jmhos/cm)
Dissolved Solids
(mg/1)
Suspended Solids
(mg/1)
Turbidity
(JCU)
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Max
Avg
Min
Plant
1974
7.9
7.8
7-4
<1
74
58
53
79
68
60
200
144
120
287
98
23
25
15
4
78
24
10
I
1975
8.1
7.5
7.1
<1
62
54
15
NA
NA
223
105
25
117
22
3
NA
Plant
1974
8.6
7.6
6.5
1
<1
<1
108
52
7
98
54
8
225
133
43
218
91
10
43
15
2
25
9
3
J
1975
8-4
7.4
6.0
2
<1
<1
96
44
3
NA
NA
183
79
12
42
13
1
NA
Plant
1974
8.3
7.9
7.6
<1
96
66
50
135
86
64
320
185
90
237
125
15
176
38
9
80
54
45
K
1975
9.9
7.9
7.6
<1
68
56
42
NA
NA
227
136
68
127
32
9
NA
Plant
1974
8.3
7.8
7.5
<1
80
68
38
83
70
53
170
148
125
162
93
14
36
6
2
70
22
1
L
1975
8.0
7.6
7.4
<1
67
57
37
NA
NA
272
89
10
58
11
2
NA
oo
-------
-49-
The R values in Table 9 indicate that there are several significant
relationships between several intake water quality characteristics and
the pH of the ash pond effluent for Plants G, H, and J. However, these
relationships varied for the different plants. For example, there was a
significant correlation between the intake water pH and the ash pond
effluent pH and also between the intake water dissolved solids and the
ash pond effluent pH for Plants G, H, and J. These same plants also had
a significant negative correlation between the intake water conductivity
and ash pond effluent pH. There was also a significant correlation
between intake water hardness and effluent pH for Plant J and a
significant negative correlation between intake water hardness and
effluent pH for Plants G and H. However, only Plants G and J had
significant correlations between intake water alkalinity and effluent
pH. For Plant G it is negative while for Plant J it is positive.
Of these correlations with the pH of the effluent, the most meaning-
ful with respect to prediction of the pH are the ones with intake alka-
linity, since alkalinity is a measure of the resistance of the system to
changes in pH. Figure 5 shows this relationship for Plant J more clearly.
During periods when the alkalinity is near zero, the pH drops below
four, whereas, with a normal alkalinity in the range of 30-90 mg/1, the
pH is approximately eight. In order to maintain a pH between six and
nine at Plant J, an intake alkalinity of around 10 mg/1 is needed.
There were few significant correlations between the suspended
solids in the ash pond effluent and any of the intake water quality
characteristics. The ash pond effluent suspended solids were negatively
correlated with the hardness in the intake water at Plants J and G. The
effluent suspended solids were also negatively correlated with the
conductivity of the intake water at Plant J. The effluent suspended
solids were negatively correlated with the alkalinity of the intake
water at Plant H.
The effluent suspended solids were not significantly correlated
with the intake suspended solids at any one of the four plants. However,
Figure 6 indicates that suspended solids peaks in the effluent may
correspond to suspended solids peaks in the intake water when lag times
of one to two weeks are considered. To test this hypothesis correlation
coefficients were developed for Plants E and J by incorporating a lag
time of one sample period (approximately seven days) between intake and
effluent samples. By lagging the data sets in this manner the detention
time of the ash pond is somewhat accounted for. The results of this
comparison are given in Table 10. Only the correlations for intake
versus effluent characteristics are given since intake versus intake and
effluent versus effluent remained the same as before. Significant
correlations between intake suspended solids and effluent suspended
solids were not obtained by lagging the two data sets. Furthermore, the
correlation coefficients for the data in Table 9 were higher than those
for the lagged data set. An exception to this at Plant E is the
correlation between the ash pond effluent conductivity and the intake
dissolved solids. Exceptions to this at Plant J are the correlations
for intake pH with effluent dissolved solids, intake total alkalinity
with effluent dissolved solids, intake hardness with effluent flowrate,
intake hardness with effluent conductivity and intake conductivity with
-------
TABLE 9. CORRELATION COEFFICIENTS FOR THE ASH POND SYSTEM AT PLAHT E
IntakeIntakeIntakeEffluentEffluentEffluentEffluent
Intake Total Intake Intake Dissolved Suspended Intake Effluent Effluent Phenolphthalein Total Effluent Effluent Dissolved Suspended Effluent
pH Alkalinity Hardness Conductivity Solids Solids Turbidity Flo»rate pH Alkalinity Alkalinity Hardness Lonductivity Solids Solids Turbidity
Intake pH
Intake Total
Alkalinity
Intake
Hardness
Intake
Conductivity
Intake
Dissolved
Solids
Intake
Suspended
Solids
Intake
Turbidity
Effluent
Flourate
Effluent pH
Effluent
Phenolphthalein
Alkalinity
Effluent
Total
Alkalinity
Effluent
Hardness
Effluent
Conductivity
Effluent
Dissolved
Solids
Effluent
Suspended
Solids
Effluent
Turbidity
1.000
0.460
0.684
0.159
0.178
-0.257
-0.253
0.160
0.098
0.112
0.163
0.089
-0.156
0.153
-0.069
1.000
-0.034 1.000
-0.133 0.307 1.000
0.295 0.215 -0.135
-0.149 0.109
-0.071 -0.081
0.111
0.155
0.087
0.090
0.115
0.118
0.038
-0.458
1.000
-0.345
0.029
-0.026
0.210
-0.203
0.002
0.237
0.075
1.000
0.078
0.381 -0.356 0.538 0.200 0.032 0.026 0.056 0.701
0.393 0.212 -0.498 0.225 0.014 -0.035 0.045 0.664
0.024 0.005 -0.421 0.395 -0.099 0.028 0.068 0.778
0.102 -0.255 -0.391 0.274 -0.150 0.035 0.106 0.875
0.328 -0.3OT -0.641 0.335 0.101 0.026 -0.205 0.678
-0.051 -0.250 -0.129 -0.188 0.223 -0.125 0.115 0.139
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
1.000
0.953
0.655
0.850
0.781
1.000
0.704 1.000
0.836 0.714 1.000
0.743 0.743 0.904
0.078 0.293 0.377
0.000 0.000 0.000
l.OOC
0.026 1.000
0.000 0.000 1.000
o
i
-------
TABLE 9 (Continued). CORRELATION COEFFICIENTS FOR ME ASH POND SYSTEM AT PLANT C
Intake
pH Alkalinity
Intake f» 1.000
Intake Total
Alkalinity -0.045 1-000
Intake
Hardness -0.187 0.812
Intake
Conductivity -0.229 0.890
iDtake
Dissolved
Solids -0.093 0.504
Intake
Suspended
Solids -0.146 -0.309
Intake
Turbidity -0.361 -0.369
Effluent
Flowratc -0.&92 -0.387
Effluent pH 0.391 -0.381
Effluent
Phenolphthalein
Alkalinity 0.297 -0 328
Effluent
Total
Alkalinity 0.088 0.001
Effluent
Hardness 0.202 -O.li]
Ef f lueot
Conductivity 0.239 -0.193
Effluent
Dissolved
Solids -0.082 0.174
Effluent
Suspended
Solids -0.084 -0.166
Effluent
Turbidity 0.000 0.000
Intake Intake Effluent Effluent Effluent Effluent
Hardness Conductivity Solids Solids Turbidity Flowrate pH Alkalinity Alkalinity Hardness Conductivity Solids Solids Turbidity
1.000
0.692 1.000
O.S74 0.579 1.000
-0.537 -0.581 -0.016 1.000
-0.382 -0.309 -0.398 0.775 1.000
-0.623 -0.623 -0.055 0.382 0.459 1.000
-0.595 -0.452 -0.313 -0.056 0.042 0.347 1.000
-0.504 -0.429 -0.176 0.038 O.lOi 0.580 0.740 1.000
-0.227 -0.166 0.069 -0.036 0.258 0.328 0.046 0.468 1.000
0.068 -0.108 -0.254 -0.054 -0.175 0.2«i5 0.027 0.154 0.282 1.000
-0.083 -0.066 -0.225 -0.147 -0.318 0.2S6 0 103 0.216 0.279 0.670 1.000
0.062 -0.196 0.442 -0.088 -0.384 0.333 0.025 0.196 0.207 0.784 0.513 1.000
-0.209 -0.196 -0.056 0.067 0.074 -0.033 0.010 -0.073 -0.092 -0.197 0.219 -0.038 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 -0.319 0.035 -0.243 0.836 1.000
-------
TABLE 9 (Continued). CORRELATION COEFFICIENTS FOR THE ASH POND SYSTEH AT PLANT H
Intake
Intake pH 1.000
Intake Total
Alkalinity O.U9
Intake
Hardness -0.227
Intake
Conductivity -0. U9
Intake
Dissolved
Solids 0.173
Intake
Suspended
Solids -0.105
Intake
Turbidity -0.052
Effluent
Flovrate 0.079
Effluent pH 0.363
Effluent
Pbenolphthalein
AlLaliDity 0.191
Effluent
Total
Alkalinity 0.105
Effluent
Hardness 0.415
Effluent
Conductivity 0.411
Effluent
Dissolved
Solids 0.372
Effluent
Suspended
Solids -0 137
Effluent
Turbidity -0.024
Intake Intake Intake Effluent Effluent Effluent Effluent
1 .000
0.495 1.000
0.391 0.684 1.000
0.258 0.441 0.756 1.000
-0.196 -0.160 -0.104 -0.167 1.000
0.032 O.OS4 0.014 -0.077 0.688 1.000
-0.161 0.023 -0.034 0.111 -0.051 0.139 1.000
-0.051 -0.328 -0.011 0.345 0.007 0.017 0.008 1.000
0.057 -0.215 0.063 0.143 -0.084 -0.167 -0.042 0.380 1.000
0.340 0.252 0.127 0.006 -0.061 0.025 -0.032 -0.306 0.195 1.000
-0.063 -0.047 0X190 0.215 0.214 0.276 O.S39 0.052 -O.MS 0.177 1.000
-0 224 -0.315 0.002 0.197 0.193 0.119 0-365 0.286 0.085 0.000 0.777 1.000
0.012 -0.408 -0.017 0.491 -0.075 0.077 0 278 0 506 0.235 -0 039 0.682 0.837 1.000
-0.497 0.026 -0.072 -0.084 0.201 -0.096 -0.010 -0.312 -0.102 -0.132 -0.051 0.054 -0 117 LOW
-0.187 -0.116 -0.341 -0.432 -0.0*6 -0017 -0 952 -0 253 -0.249 -0.069 -O.H3 -0.172 -0.156 0.419 1.000
to
-------
TABLE 9 (Continued). CORRELATION COEFFICIENTS FOR THE ASH POND SYSTErt AT PLANT J
Intake pH
Intake Total
Alkalinity
Intake
Hardness
Intake
Conductivity
Intake
Dissolved
Solids
Intake
Suspended
Solids
Intake
Turbidity
Effluent
Tlowrate
Effluent pK
Effluent
Phenolpbtha le in
Alkalinity
Effluent
Total
Alkalinity
Effluent
Hardness
Effluent
Effluent
Di ssolved
Solids
Effluent
Suspended
Solids
Effluent
Turbidity
PH
1.000
0 840
0 791
0.800
0.705
-0.012
-0.089
-0 199
0.865
0 246
0 835
0.630
0.048
0.471
-0.182
-0.434
Intake
Al ka 1 1 n i ty
1.000
0.977
0.960
0.753
-0.024
-0.057
-0.224
0.841
0 243
0 830
0.625
0. 143
0.464
-0.113
0.217
Intake Intake Effluen
Hardness Conductivity Solids Solids Turbidity Flowrate pH Alkalini
1.000
0.967 1.000
0.720 0.729 1.000
0.057 0.104 0.119 1.000
-0.078 -0.091 -0.131 -0.144 1.000
-0.401 -0.319 -0.072 0.071 -0.084 1.000
0.825 -0.805 0.689 -0.061 -0.06? -0.266 1.000
0 221 0.256 0.084 -0.082 -0.066 -0.010 0.316 1.000
0.842 0.847 0.671 -0.035 -0.161 -0.365 0875 0.340
0.707 0.684 0.408 -0.094 -0.009 -0.223 0.620 0.387
0.209 0.188 0.160 -0.180 0.034 -0.021 -0.046 0242
0.565 0.561 0.392 0.026 -0.033 -0.055 0.405 0.169
-0.219 -0.225 -0.088 -0.072 -0.031 -0.076 -0.121 -0.033
-0.248 -0.242 -0.115 -0.076 -0.050 0.069 -0.374 -0.124
Effluent Effluent Effluent
Total Effluent Effluent Dissolved Suspended Effluent
Alka1 mity Alkalinity Hardness Conductivity Sol ids Solids Turbidity
1.000
0.781 1.000
0.262 0.741 1.000
0.508 0.550 0.667
OJ
I
1.000
-0.101 -0.143 -0.172 -0.047 1.000
-0.275 0.000 0.000 0.000 0.000 1.000
-------
CO
<
I
140
130
110
90
70
50
30
10
0
9
7
5
1 II
- TOTAL ALK, OF RAW WATER SUPPLY
pH OF ASH POND EFFLUENT
nvt
i
Figure 5. Relationship of Ash Pond pH and Intake Water Alkalinity for Plant J
-------
50
30
in
a
o_
CO
20
10
RAW WATER SUPPLY
ASH POND EFFLUENT
L ' i i i i i I i
0
TIME., WKS,
Figure 6a. Relationship of Suspended Solids in the Ash Pond Effluent and the Intake Water Supply for Plant E
-------
100
90
80
70
50
CO
§ 50
a
Q
LJ
Q_
oo
30
20
10
0
ASH POND EFFLUENT
RAW WATER SUPPLY
11
21
31
51
61
71
31
91
101
"irlE, 1/KS,
Figure 6b. Relationship of Suspended Solids In the Ash Pond Effluent and the Intake Water Supply for Plant J
-------
TABLE 10. LAGGED CORRELATION COEFFICIENTS FOR PLANT E
Intake pH
Intake Total
Alkalinity
Intake
Hardness
Intake
Conductivity
Intake
Dissolved
Solids
Intake
Suspended
Solids
Intake
Turbidity
Intake pH
Intake Total
Alkalinity
Intake
Hardness
Intake
Conductivity
Intake
Dissolved
Solids
Intake
Suspended
Solids
Intake
Turbidity
Effluent
Flowrate
0.171
0.032
0.123
0.024
-0.075
0.017
0.237
-0.165
-0.181
-0.524
-0.493
-0.161
-0.064
-0.136
Effluent
PH
0.061
0.204
-0.123
-0.596
0.249
-0.038
0.028
0.805
0.759
0.769
0.766
0.680
-0.039
-0.063
Effluent
Phenolphthalein
Alkalinity
0.129
0.350
-0.282
-0.579
0.152
-0.027
0.006
LAGGED
0.231
0.200
0.217
0.241
0.126
-0.101
-0.141
Effluent
Total
Alkalinity
0.137
0.384
-0.272
-0.636
0.149
0.015
-0.001
CORRELATION
0.765
0.744
0.783
0.784
0.636
-0.052
-0.110
Effluent
Hardness
0.015
0.050
-0.243
-0.498
0.208
-0.115
0.004
COEFFICIENTS
0.613
0.590
0.640
0.646
0.485
-0.017
-0.008
Effluent
Conductivity
-0.200
0.044
-0.323
-0.650
0.289
-0.067
0.035
FOR PLAHT J
0.104
0.190
0.243
0.249
0.137
-0.134
0.023
Effluent
Dissolved
Solids
0.099
0.175
-0.242
-0.726
0.137
0.001
0.015
0.534
0.520
0.479
0.487
0.432
-0.009
-0.034
Effluent
Suspended
Solids
-0.100
-0.194
0.024
-0.243
0.036
-0.091
0.203
-0.116
-0.018
-0.171
-0.199
-0.065
0.018
0.012
Effluent
Turbidity
0
0
0
0
0
0
0
-0.273
-0.116
-0.183
-0.201
-0.225
0.085
-0.017
I
Ln
-------
-58-
effluent flowrate. In addition, at Plant J there was a significant
correlation between effluent conductivity and intake conductivity for
the lagged data set which was not observed for the original comparison.
Therefore, the results in Table 10 appear to have no real meaning.
The TVA ash ponds for which the weekly ash pond effluent data
showed a yearly cycle were given in Table 4. The ash ponds for which
the weekly intake data showed a yearly cycle were not determined because
the type of trends exhibited by the parameters in the ash pond effluent
are more important in determining how the data are treated statistically
than the trends in the intake water. In addition, the effect of intake
water quality on the ash pond effluent quality was already discussed via
the linear correlation comparisons.
In general for the year 1974 and 1975 the range over which the pH
in the ash pond effluent varies was larger than the range for the pH in
the intake water. However, some ash ponds increased the pH while others
decreased it. The dissolved solids varied over approximately the same
range for both the intake and the effluent; however, the yearly average
concentrations were normally higher for the ash pond effluents. The
range over which the suspended solids vary is greater for the intake
water than the ash pond effluent. However, the yearly average con-
centration for suspended solids increased from intake to effluent both
in 1974 and 1975 for four effluents, while it increased either in 1974
or 1975 for an additional five effluents. The suspended solids decreased
from intake to effluent for six plants in both 1974 and 1975. Table 11
gives a summary of this comparison by plant.
Also included in Table 5 was a summary of the quarterly intake data
for the years 1974 and 1975. Table 12 shows the number of ash ponds
which increase the various average concentrations of the intake water.
Although Table 12 does not give valuable insight into the effect of
variations in the concentrations of trace metals in the intake water on
the variations of trace metals in the ash pond effluent it does provide
information on the effect of using the intake water for ash sluicing.
The extent of this effect on intake water quality is very important in
designing a monitoring program for ash pond effluents. Table 12 indicates
that more than half of the ash ponds increase the concentrations of Al
ammonia, As, Ba, Cd, Ca, Cl, Cr, Pb, Hg, Ni, Se, silica, sulfate, and Zn
over that in the intake water. The range over whiqh the trace metals
vary in the ash pond effluent appears to be as great or greater than
that in the intake water.
INDIRECT MONITORING METHODS
The previous results indicate that monitoring the ash pond effluent
cannot be replaced by measurements within the operation of the power
plant itself. If such an approach is to be pursued, more detailed data
on variables such as the amount of coal burned, its characteristics, the
quality of water used for sluicing and others will have to be collected
and their relationships with the ash pond effluent characteristics
determined.
-------
-59-
TABLE 11. COMPARISON OF WEEKLY INTAKE AND EFFLUENT
SUSPENDED SOLIDS CONCENTRATIONS FOR 1974 AND 1975
AT TVA ASH PONDS
Plant
Plant A BA
Plant A FA
Plant B BA
Plant B FA
Plant C East
Plant C West
Plant D
Plant E
Plant F
Plant G
Plant H
Plant I South
Plant J
Plant K
Plant L
Intake SS
1974 1975
60
60
41
41
81
81
15
17
9
12
21
15
15
38
6
48
48
23
23
47
47
11
18
23
15
20
23
13
36
12
Effluent SS
1974 1975
54
37
67
94
48
43
17
12
40
20
15
15
47
16
20
51
15
25
18
27
35
15
4
6
19
12
18
38
51
8
Difference
1974 1975
-6
-23
+26
+53
-33
-38
+2
-5
+31
+8
-6
0
+32
-22
+14
+3
-33
+2
-5
-20
-12
+4
-14
-17
+4
-8
-5
+25
+ 15
-4
-------
-60-
TABLE 12. NUMBER OF ASH PONDS WHOSE AVERAGE EFFLUENT
CONCENTRATIONS EXCEED THOSE OF THE INTAKE WATER
Element
Aluminum
Ammonia
Arsenic
Barium
Beryllium
Cadmium
Calcium
Chloride
Chromium
Copper
Cyanide
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
Selenium
Silica
Silver
Sulfate
Zinc
No. Exceeding
10
9
15
7
1
7
15
8
10
5
3
4
8
6
5
12
10
14
12
2
15
7
-------
-61-
COMPARISON OF WEEKLY AND QUARTERLY SAMPLING
There appears to be a discrepancy between the yearly average
suspended solids concentration calculated based on the quarterly data
and the average calculated based on the weekly data. Therefore, a
comparison of these two data sets was made. The data for this
comparison is given in Table 13. Yearly average concentrations for the
years 1973, 1974, and 1975, based on the quarterly and weekly sampling
programs are given for suspended solids. The yearly average for pH
based on weekly samples is also given. The quarterly samples (one
sample collected every three months) for all TVA ash ponds were analyzed
at the Laboratory Branch in Chattanooga. The weekly samples were analyzed
at the respective steam plant laboratories. Both sets of samples were
collected by steam plant personnel. In most cases a weekly sample was
collected at the same time as a quarterly sample.
There are two possible explanations for the differences in the
averages: the first is that for each steam plant, the same lab did not
analyze the weekly and quarterly samples; and the second is that the
average based on the quarterly samples was calculated using only four
samples while the average of weekly samples is based on approximately
52 samples. From a characterization standpoint, the more samples
analyzed, the more representative the calculated average. Therefore,
the difference could be due to having a more representative average from
the weekly samples than from the quarterly. In addition to these two
explanations, the difference in suspended solids averages in 1973 could
be the result of the different procedures used for determining the
suspended solids. The quarterly samples were analyzed directly by
weighing the quantity of suspended material removed following a filtra-
tion procedure while the weekly samples were calculated by subtracting
the dissolved solids from the total solids.
In order to determine the major reason for the discrepancy, a third
average was calculated by selecting those weekly samples which were col-
lected at the same time as the quarterly samples. This average for sus-
pended solids and pH is also given in Table 13. If the yearly average
calculated from the quarterly data and the yearly average calculated
from only four of the weekly samples are in close agreement then the
difference between the two yearly averages could be assumed to be the
result of the difference in sampling frequencies. However, if these two
averages are different by more than 10 mg/1, then the difference could
be attributed to different laboratories. The value of 10 mg/1 was
obtained from a discussion in Standard Methods (3) on the precision and
accuracy of the nonfilterable residue (suspended solids) procedure. The
discussion gives a standard deviation of ±2 mg/1 at the 15 mg/1 concentra-
tion and ±24 mg/1 at the 242 mg/1 concentration. Although a measure of
the accuracy would be more appropriate, Standard Methods (3) indicates
that there is no satisfactory procedure for determining the accuracy of
the method on wastewater samples because the true value of suspended
matter is unknown. Since most of the averages were greater than 15 mg/1
and less than 242 mg/1, a value of 10 mg/1 was chosen.
Based on the previous discussion, the differences in the yearly
average suspended solids in at least 2 out of 3 years at Plants A fly
ash, A bottom ash, B fly ash, B bottom ash, I, and J were attributed to
-------
-62-
TABLE 13. COMPARISON OF QUARTERLY AND WEEKLY SAMPLING PROGRAMS
— — Suspended Solids
Plant A - Fly Ash
Plant A - Bottom Ash
Plant B - Fly Ash
Plant B - Bottom Ash
Plant C East
Plant C West
Plant D
Plant E
(New Ash Pond)
Plant F
Plant G
Plant H
Plant I
Plant J
Plant K
Plant L
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
Q
W
QW
1973
7
93
94
30
66
27
24
100
43
23
79
10
20
43
27
31
42
31
15
25
37
.
-
6
-
19
19
22
10
16
15
4
24
26
11
32
44
5
13
7
7
17
14
1974
5
37
38
88
54
40
9
94
56
33
67
33
20
48
48
20
43
50
19
17
13
3
12
22
2
40
83
14
20
17
12
15
14
2
15
16
26
47
31
7
16
14
23
20
16
1975
6
15
16
38
51
53
6
18
16
14
25
19
12
27
17
45
35
41
12
15
16
4
4
4
6
6
7
29
19
18
8
12
9
10
18
10
7
38
20
13
29
13
6
8
8
1973
.
4.5
4.6
-
7.1
7.1
9.2
9.1
-
8.2
8.2
7.2
7.2
7.6
8.3
8.6
8.4
-
-
-
—
9.8
9.9
8.8
8.8
-
11.3
11.2
-
6.0
5.7
-
11.0
11.1
-
10.0
9.3
PH
1974
-
4.2
4.2
-
7.2
7.2
9.4
8.6
-
8.0
7.9
7.1
6.9
7.2
7.4
8.4
8.5
-
11.2
11.3
11.1
11.2
9.5
9.3
8.3
8.1
-
11.1
11.2
-
6.5
6.2
-
11.0
11.2
.
10.3
10.4
1975
.
4.0
4.2
-
7.1
6.8
9.2
9.6
-
8.0
7.9
7.1
7.2
7.5
6.4
8.4
8.5
-
11.1
11.1
10.7
10.6
9.4
9.6
8.7
8.4
-
9.8
9.9
-
6.2
6.1
.
10.3
10.1
_
10.4
10.3
Note: Q - Averages based on quarterly samples analyzed by TVA Laboratory Branch.
W - Averages based on weekly samples analyzed by the respective steam
plant labs.
QW - Averages based on selected weekly samples collected at or approximately
the same time as the quarterly samples.
-------
-63-
different laboratories. The difference for one year at Plants C, D, E,
F, and G, was attributed to different laboratories. There did not appear
to be a difference in yearly average suspended solids due to different
laboratories during any of the three years for Plants H, K, and L.
If the difference in the yearly average suspended solids between
laboratories is excluded the difference between the yearly average for
quarterly and weekly sampling at Plants A fly ash, A bottom ash, D,
G, H, I, and L was less than 4 mg/1 during at least 2 of the 3 years.
The difference between quarterly and weekly yearly averages was less
than 4 mg/1 during at least 1 year for Plants A bottom ash, B fly ash,
C east, E, F, and K. For most monitoring programs a difference of
4 mg/1 in suspended solids would probably be acceptable. For other
monitoring programs a greater difference may be acceptable. The dif-
ference that can be tolerated depends on the goal of the monitoring
program. Overall the above discussion indicates that both the sampling
frequency and laboratory preforming the analysis can influence the
yearly average suspended solids concentration reported for a particular
plant during any given year.
A comparison of the yearly average pH of the ash pond effluent
based on quarterly sampling and weekly sampling showed that except for
Plant C west in 1973, Plant B fly ash in 1974 and Plant L in 1973,
quarterly sampling was adequate to determine the yearly average pH
within at least 0.5 pH unit.
Other than for pH and suspended solids, there were not enough data
available on those parameters required by the NPDES permit to determine
the influence of the sampling frequency on the calculation of the yearly
average. Flow was not included in these comparisons because all ash
pond effluents are equipped with continuous flow measuring and recording
devices.
COMPARISON OF GRAB AND COMPOSITE SAMPLING
The two most common types of samples are grab samples and composite
samples, and either may be obtained manually or automatically. Grab
samples represent the waste characteristics at the time the sample is
taken, while composite samples represent the waste characteristics of a
mixture of several individual samples whose collection frequency or
relative volume is determined based on the flow at the time of sampling.
As long as the ratio of flow to individual sample volume remains the
same, the compositing should be valid. A grab sample is preferred over
a composite sample when the waste characteristics are relatively constant
because of relative cost of collecting these two types of samples. For
such wastes an occasional grab sample may be entirely adequate to establish
waste characteristics.
Twenty-four hour composite samples comprised of grab samples taken
every hour were collected for four consecutive days at four TVA steam
plants. During at least one, and in most cases two, 24-hour sample
period, three or four grab samples were collected for comparison with
the composite samples. These results are shown in Figure 7. First, the
-------
-64-
concentration of metals in the composite samples did not vary signifi-
cantly over the four day period. For example, copper varied the most on
a percentage basis for the four plants. The range of composite samples
on the four-day periods was from 0.07 to 0.11 mg/1 at Plant C, <0.01 to
0.05 mg/1 at Plant J, <0.01 to 0.1 mg/1 at Plant H, and <0.01 to 0.03 mg/1
at Plant D. The range of grab samples is indicated by the symbol "I" in
Figure 7. The grab samples appear to be as representative of the system
as composite samples; however, more work is needed to statistically
determine the best method of sample collection. However, only grab
sampling was considered in the later sections of this report.
-------
O 24 hr COMPOSITE 5-20-73 A 24 hr COMPOSITE 9-22-75
D 24 hr COMPOSITE 9-21-79 O 24 IV COMPOSITE 9-23-79
J RANGE OF GRA8 SAMPLES
1000
IOO
10
s°
O D
p. souos
DA_
0 V
0. SEJCA
00
A
COND
O D
O 24 hr COMPOSITE 6-17-79 A 24 hr COMPOSITE 6-19-75
D 24 hr COMPOSITE 6-W-7S O 24 hr COMPOSITE 6-2O-7S
XRANeC OF GRAB SAMPLES
IOOO
100
**
o
o. SIUCA
CONO
0 SOLIOS 00^0
o
O.I
O.OI
0 0
D At>
0 ±
O D
WATER QUALITY PARAMETERS
PLANT C ASH POM)
lOp
O.I
0.01
O D
AO
0D
WATER QUALITY PARAMETERS
PLANT J ASH POND
ON
en
Figure 7. Comparison of Grab and CDmposite Samples for Four TVA Ash Pond Effluents
All parameters are in mg/l except pH (standard units) and conductivity (pmho/cm) ,
-------
O Z4 hr COMPOSITE 6-3-7S A Z4 hr COMPOSITE 6-S-7S
D Z4 hr COMPOSITE 6-4-7S O 24 hr COMPOSITE 6-6-7S
I RAHOC OF GRAB SAMPLES
1000 r
100
10
O.OI
O Z4 hr COMPOSITE 6-24-75 A 24 hr COMPOSITE 6-26-75
D Z4 hr COMPOSITE «-Z9-7S O 24 hr COMPOSITE 6-27-79
~J_ RANGE OF GRAB SAMPLES
so4 o D & o
c« D T
A- 0 4?
: T -1-! 1
; o D J-^7 :
: o D At? s. souos :
'• D. SILICA -r- -
o
O a AC D
u
o D£o
: a i
~T~ ' 1
| -'-
A -r
» i T •
TT A
D D O D |
o So -L 6
I
IOOO
100
10
i
OOI
COND
D SOUDS O D ^ &
: SO, :
0 D A.5
S. SOUPS
: ~ onA.S —I ]
; O a A O
a A o
: JH oo AO f
Al
T^ F.
: oc^ ° :
Cg |
T T "
: T z" oAl :
T-1- o
Mn n A n ^
' ° II °I
WATER QUALITY PARAMETERS
PLANT H ASH PONO
o\
WATER QUALITY PARAMETERS
PLANT D ASH POND
Figure 7 (Continued)
-------
-67-
SECTION 4
PROCEDURE FOR DESIGNING AN ASH POND MONITORING PROGRAM
Since the physical understanding of the ash pond system is limited,
more emphasis must be placed on statistical analysis of past effluent
characteristics to design the future monitoring program. There are
several statistical procedures which could be applied to determine the
proper sampling frequency to ensure an accurate estimate of the ash pond
effluent characteristics. The procedure discussed here requires the
following: (1) a baseline data set which characterizes the effluent
with a greater detail, precision, and accuracy than the data to be
obtained from the monitoring program under design; (2) the variation of
the baseline data with time; (3) an estimate of the statistical distri-
bution of the baseline data; (4) the number of samples to estimate the
mean as a function of the precision; and (5) an estimate of the desired
precision of the monitoring program under design. The procedure assumes
that the individual water quality parameters either exhibit a seasonal
trend or are randomly distributed. For those parameters that are randomly
distributed, that data are assumed to follow either a normal or lognormal
distribution. For those exhibiting a seasonal trend, the data are
divided into different sample periods which can be treated the same as a
data set that is randomly distributed. The monitoring program to be
designed ensures, within a specified degree of confidence, that the
least number of samples is collected which shows the effluent is in
compliance with a particular effluent limitation within a specified time
period. Where effluent limitations have not been established, the
monitoring program is designed for the collection of the minimum number
of samples required to estimate the yearly mean with predetermined
precision, accuracy, and confidence.
The assumption that the ash pond effluent parameters are random is
not completely valid because the gross ash pond characteristics were
shown to be affected by the type of coal burned. However, as long as
the coal characteristics and methods of operation are not changed dras-
tically from those used to design the monitoring program, the assumption
can be considered valid. The monitoring program can be evaluated after
each sampling period (the sampling period would be one year if the
objective is to estimate the yearly mean) using the same procedure used
to design the original monitoring program. This would be done by apply-
ing the method to either a combined data set consisting of the data set
for the sample period under evaluation and the previous data sets or
just the data set for the sample period under evaluation. Limiting the
evaluation to the new data set would be best in the case where the data
exhibits either a continuous increasing or decreasing trend from sample
period to sample period while the variation within a sample period
remains constant. This procedure is discussed in the following section.
Application of the method to two TVA ash pond effluents will be discussed
in subsequent sections.
-------
-68-
DATA REQUIREMENTS
Before a statistically sound monitoring program can be designed,
background information on the characteristics of the waste stream and
the entire production process which generates this waste stream is
desirable. If this information is not already available, a wastewater
survey is conducted to provide this information. The balance between
the use of statistical methods and evaluation based upon physical under-
standing is extremely important as pointed out by EPA in 1974 (4). As
physical understanding increases, the use and need for statistics
decreases.
Data such as that presented in Section 3 can be used to explain the
system and estimate the number of samples required to provide a proper
data base from which a monitoring program can be designed, or it can be
used as the data set from which to design the monitoring program. The
parameters which should be included in the monitoring program can also
be determined from the data given in Section 3.
VARIATION OF THE DATA WITH TIME
Before a monitoring program can be properly designed, the variation
of the effluent parameters with time and any periodic cycles which occur
in the system must be defined. The variation of the parameters in the
effluent with time can best be determined from plots of concentration
versus time. The concentrations are determined from past monitoring
programs or extensive waste surveys. If cycles exist within this data
set, the proper sampling frequency can be predicted based on the time
span of the cycle. For example, to define a weekly cycle, a sampling
frequency of at least twice a week would be required. However, the
cycle can be better defined by more frequent sampling. If no cycles are
indicated by the data set, it can be treated as one set of random events.
A statistical method can then be applied to this data set to determine
the number of samples which ensures estimation of the true mean within
some accuracy and precision. To estimate the true mean within some
accuracy and precision for a data set which exhibits a cycle, the data
set can be divided into the different phases of the cycle, thus creating
individual data sets of random events. The same statistical procedure
can then be applied to these individual data sets to estimate the mean
within a subset, or a procedure which will be termed "stratified sampling"
can be applied to estimate the mean for the entire data set (see Daniel
and Terrell, 1975 (5) for a complete discussion on stratified sampling).
DISTRIBUTION OF THE DATA
Once a period of random events has been established, the probability
distribution of the data within that period must be defined. The
assumption as to the underlying probability distribution of a parameter
is critical in designing a monitoring program. Sherwani and Moreau,
1975 (6) summarized the distribution of many water quality parameters as
follows:
-------
-69-
(1) The parameters have a finite range. They have a fixed
lower physical limit, in most cases equal to zero and
a variable but finite upper limit, in most cases
saturation concentration;
(2) The distribution is typically positively skewed;
(3) The parameters exhibit a periodic behavior. The
periodicity may be due either to the annual cycle in
the meteorological and hydrological environment of
the stream, or to the weekly and seasonal cycles in
the waste inputs to the stream.
They found that the majority of the water quality parameters do not
follow a normal distribution. However, they did find that several
parameters such as flow, suspended solids, conductivity, and phosphorous
followed a lognormal distribution. Berthouex and Meinert, 1977 (7)
reported that surface water concentrations in the Tennessee Valley of
Hg, Zn, Cu, Cd, and Pb followed lognormal distributions. For the pur-
poses of this project, the data sets were therefore assumed to follow
either a normal or lognormal distribution.
A method discussed by Miller and Freund, 1965 (8) will be used to
determine whether the data are best described by a normal or lognormal
distribution. The method requires that the cumulative frequency of the
data be plotted on a special probability scale against the actual
concentration. Data from a normal distribution will graph roughly as a
straight line when such a probability scale is used. When the data
graph as a straight line when the concentrations are plotted on a log-
rithmic scale, the data are more nearly lognormal. Logrithms to the
base ten will be used for this study. These plots are called cumulative
frequency plots and Figure 8 gives an example.
Additional information can be obtained from these cumulative fre-
quency plots. For example, extrapolation below the minimum detectable
amount (MDA) is reasonable, thus making it possible to estimate a geo-
metric mean for a particular element when it is below the MDA. The
concentration corresponding to 50 percent estimates the geometric mean
for plots using the logarithmic scale and the arithmetic mean for plots
using an untransformed scale. For lognormal distributions the logarithm
of the geometric mean is equal to the mean of the logarithms of the con-
centrations. These plots can also be used to estimate the probability
that a certain concentration, for instance an effluent limitation, will
be exceeded.
ESTIMATION OF THE MEAN AS A FUNCTION OF THE PRECISION
The population mean value for a given parameter is the main
interest of most monitoring programs. Therefore a method is required
for ascertaining the chance that a sample statistic such as the mean
deviates from the population parameters by a prescribed amount. The
three components to be considered are:
-------
-70-
.0
"5
"c
o
o
o
Estimated
Arithmetic
Mean
I I I
10 20
40 60
80 90 95 98
Cumulative Frequency (%)
8. Example of a Cummulatlve Frequency Plot
-------
-71-
1. the sample size,
2. the precision of the estimate, and
3. the significance level.
Should any two be available, the third can be calculated. For a normal
distribution, the population mean, |J, is estimated by X and a confidence
interval for }J is given by
X ± t ,- (1)
Vn
where, X, S are the sample mean and standard deviation, t is the value of
"student's" t for a given significance level and depends on the number of
degrees of freedom of S, and n is the sample size. For a lognormal dis-
tribution X and S are the mean and standard deviation respectively of the
logarithms of the concentrations.
The sample data are summarized in X and S. Both X and S are deter-
mined from past data or an extensive survey. The choice of a confidence
coefficient and corresponding "student's" t value depends on the con-
sequences of the estimate being incorrect. The 80, 95, and 99 percent
significance confidence levels will be considered in this study.
The precision L may be defined as limits on either side of the true
mean within which the sample mean will fall with specified probability.
That is, it is desired to have a specified degree of confidence that:
M - X < L (2)
For the lognormal distribution |j is the logarithm of the true geometric
mean and X is the mean of the logarithms of the concentrations. Given L
and S, and given the value of t corresponding to the desired confidence
coefficient, the required number, n, of samples is:
t2S2
n =
The number of samples, n, in equation 3 can be presented graphically
as a function of L for any given data set as shown in Figure 9. Figures
similar to Figure 9 can be used to determine the number of samples, with
a given confidence coefficient, required to estimate the mean within a
given precision. The sampling frequency can then be determined by
dividing the number of samples by the time period over which an estimate
of the mean is desired. Sampling periodically as suggested here is
known as systematic sampling rather than random sampling. However, in
this case both are treated as being equivalent.
To properly determine the number of samples required to estimate
the mean for a data set which exhibits seasonal behavior, the following
equation, according to Daniel and Terrell (5), may be used:
-------
QL
E
o
o
6
—i
NJ
I
DEVIATION FROM THE TRUE MEAN (L)
Figure 9. Example of a Plot of the Number of Samples Versus the Deviation
from the True Mean
-------
-73-
H
t2 I [(N /N) S ]2
1. — 1 ^ *^
n = 6=4, (4)
where
N, = size of h stratum,
N = population size, and
S, = standard deviation of h stratum.
n
H = number of strata
The allocation of the samples over the sample period can then be
determined by the following equation:
N, S,
n.= -TJLJi n C5)
IH N, S,
h=l h h
where n, represents the number of samples required for the h stratum.
This method minimizes the variance of the estimate of the yearly
mean. Most likely the number of samples within a stratum are not portional
to the length of that stratum and, therefore, an extra step is needed
to calculate the mean over the entire period. This is done by weighing
the averagesjby the proportion of the year occupied by each stratum.
That is, if X, is the average for stratum h, the weighted average is:
H N
2 \ X (6)
h=l N n
Computing the standard error of the estimated mean is only slightly more
complicated than for a simple random sample. See, for example, Snedecor
and Cochran, 1967 (9).
ESTIMATION OF THE PRECISION
Figure 9 shows that the deviation of the sample mean from the popu-
lation mean (precision) decreases with an increased number of samples,
and that the incremental change in this deviation also decreases with
an increased number of samples. This incremental change in the deviation
is therefore the key factor in identifying the most desirable sampling
frequency. This indicates that there is a critical range of values of
the deviation which should be considered in the design of a monitoring
program. This critical range is defined by the values between £x and
A2 in Figure 9. The value £lt is defined here as the lower limit for
the range of the critical deviation but actually represents a more
precise estimate than £2> while £2 is defined here as the upper limit
for the range of the critical deviation. The upper limit may be defined
-------
-74-
by the deviation produced by only one sample and the lower limit may be
defined by the deviation produced by some given maximum number of samples.
The given maximum number of samples may be a function of the resources
available such as manpower and dollars for the monitoring program.
The strategy for selecting the precision, differs depending on the
objective of the monitoring program. If the objective of the monitoring
program is to determine water quality trends or means for a given period,
then the deviation from the true mean may need to be very small. However,
there is a point where L becomes so small that the cost for collection
and analysis of the corresponding samples becomes prohibitive or
impractical. In most cases a deviation of ± 10 to 20 percent of the
sample mean would be acceptable. If the objective of the monitoring
program is to show that the effluent is in compliance with some effluent
limitation or standard, then the desired precision depends on how close
the mean is to the limitation or standard. The question of such a moni-
toring program then become: (1) "What is the probability that an estimate
of the mean would be greater than the standard when the true mean was
actually less than the standard?" (The effluent is in compliance, but
from an estimate of the mean, the effluent appears not to be in compliance.)
and (2) "What is the probability that an estimate of the mean would be
less than the standard when the true mean was actually greater than the
standard?" (The effluent is not in compliance, but from an estimate of
the mean, the effluent appears to be in compliance.) Both probabilities
depend on the number of samples used to estimate the mean. The true
mean can never be known with complete certainty, but the precision of
the estimated mean can be improved by increasing the number of samples.
Therefore the most efficient monitoring program to show compliance is
the one with the minimum number of samples so that if the effluent is in
compliance, the average of the samples shows the effluent to be in
compliance.
The minimum sampling frequency necessary to determine water quality
trends for a particular parameter can be determined directly from a
figure similar to Figure 9 or by the use of equations 2 and 3 for a
desired precision. This is done by selecting the number of samples cor-
responding to the desired deviation from the true mean. However, the
minimum sampling frequency necessary to show compliance is obtained in a
slightly different manner. The desired deviation from the mean L, is
calculated as the difference between the effluent limitation and an esti-
mate of the mean (or the logarithm of the effluent limitation and the
logarithm of the estimate of the geometric mean for lognormal distribu-
tions) from a previous sampling program. The average used to determine L
should be for the period over which the effluent limitation applies. The
average for a given period can be used to determine L for an effluent
limitation which applies over a short period as long as the data were
random over the entire period. This assumes the same average would be
obtained for an equal number of samples, no matter whether they were
collected over a month or a year. The number of samples is then deter-
mined from a figure similar to Figure 9 or equation 3 for the L value
calculated using equation 2. Division of the number of samples by the
period for which the effluent limitation applies, yields the minimum
sampling frequency which indicates compliance with the effluent
limitation.
-------
-75-
The procedure for selecting the best sampling frequency presented
in this section is limited to determining the minimum sampling frequency
for a given precision and significance level. Therefore, the procedure
should be a useful tool to managers in determining the monetary resources
needed for monitoring. However, the amount of money spent for monitoring
is a policy decision based on available resources and priorities. In
addition, the procedure showed that as the sample mean approaches the
effluent limitation, the number of samples (and, therefore, the cost)
required to show compliance increases. Therefore, this procedure may
also be used to indicate when part of the investment in pollution control
measures may be justified to offset the cost of monitoring to show
compliance.
STEPWISE SUMMARY OF THE DESIGN PROCEDURE
Methods for determining the various inputs into a procedure for
designing a monitoring program were described in the previous subsections.
A stepwise summary of the procedure is presented below.
Step 1. Develop a physical understanding of the system.
Step 2. Develop a data set which estimates the effluent characteristics.
Step 3. Determine the variation of the data with time.
Step 4. Stratify the data set by season.
Step 5. Determine the distribution of the data in these data subsets.
Step 6. Estimate the mean as a function of the precision.
Step 7. Determine the critical range of the precision and select the
desired precision for the future monitoring program.
Step 8. Determine the number of samples required to estimate the mean
within the desired precision.
Step 9. Determine the maximum resources which can be allocated to the
monitoring program.
Step 10. Select the monitoring frequency which best satisfies the
requirements in Step 8 without exceeding the maximum resources
established in Step 9.
Step 11. Repeat Steps 3 through 10 at the end of each monitoring
period in order to update the program.
Step 10 cannot be accomplished without a management decision which
establishes the maximum resources (Step 9) which can be allocated to the
monitoring program. This decision involves a host of considerations beyond
the scope of this study. Therefore, limits on the available resources will
be assumed in later sections in order to demonstrate Steps 1 through 10 of
the procedure. Once management has established the available resources,
the results of the work presented here for Steps 1 through 8 should be
easily adapted to the task of completing Steps 10 and 11.
-------
-76-
SECTION 5
ASH POND MONITORING PROGRAM FOR PLANT E
The following section demonstrates how the procedure outlined in
Section 4 was used to design a monitoring program for the ash pond
effluent at TVA's Plant E.
DESCRIPTION OF PLANT E
Plant E consists of five pulverized coal-fired units with a combined
full load capacity of 1.3 million kilowatts. Units 1 through 4 were
placed in commercial operation in 1955 and unit 5 in 1965. Full load
capacity for each of units 1-4 and unit 5 is 200,000 and 500,000 kilowatts,
respectively. At full capacity the plant consumes about 13,000 tons of
coal per day. The majority of the coal comes from western Kentucky and
has an average sulfur content of 4.1 percent and an average ash content
of 15.3 percent.
The plant also consists of eight standby gas turbine units which
have a total generator nameplate rating of 475,000 kilowatts. These
units are used primarily to meet system peak power loads and are used
between 500 and 1,000 hours per year. These units are designed to use
either natural gas or distillate fuel oil and were placed into operation
in 1972. At normal full load, each unit will consume about 4,900 gallons
of oil per hour; and when burning natural gas, each unit uses about
670,000 cubic feet per hour.
The coal-fired units 1 though 4 are equipped with mechanical fly
ash collectors (74 percent efficiency) and electrostatic precipitators
(97 percent design efficiency) while unit 5 is equipped with an electro-
static precipitator (90 percent design efficiency). The overall fly ash
collection efficiency is approximately 95 percent.
Assuming operation at full load capacity, approximately 1,900 tons
of ash per day would be produced by Plant E. This ash is sluiced to a
63 acre ash pond with a storage capacity of about 3.1 million cubic
yards which provides settling and disposal of the ash. The ash pond
effluent is discharged into the condenser cooling water discharge canal.
The ash pond effluent characteristics previously discussed for Plant E
and those to follow are for the ash pond effluent stream prior to
discharge into the condenser cooling water discharge canal.
In addition to the ash, the ash pond also receives neutralized
chemical cleaning wastes. These wastes are discharged intermittently (4
times every 3 years) and during their discharge they represent approximately
3.4 percent of the total flow from the ash pond.
-------
-77-
MECHANICS OF THE ASH POND SYSTEM AT PLANT E
A summary of the ash pond effluent characteristics for Plant E
during 1974 and 1975 was given in Section 3. There were insufficient
data on the operating conditions of Plant E during 1974 and 1975 to
determine the relationship between the ash pond effluent and plant
operation. There were also no significant correlations between the
intake water quality and the effluent water quality except when the
detention time of the ash pond was taken into consideration by lagging
the two data sets. Therefore, two ash pond surveys were conducted in
which the physical and chemical characteristics of the ash pond and its
effluent were studied and their relationship determined. The first, a
preliminary survey, was conducted during the first week of October 1975.
The second, a more detailed survey, was conducted during the week of
February 23, 1976.
Cross-sectional profiles of the ash pond on the first two days of
the preliminary survey showed the pond to be stratified with respect to
temperature. A third pond profile on the last day of the survey showed
the pond to be completely isothermal. The reason for this destratifica-
tion was uncertain. During the second survey, two more pond profiles
were performed on February 23 and 25. The results are shown in Figure
10. During the second survey tag lines were stretched across the pond
at 5 locations and each line was marked with tape at 25 percent, 50
percent, and 75 percent distances from the left bank to indicate sampling
stations. Depth, temperature, pH, conductivity, and dissolved oxygen
were measured in situ at each station with a portable water quality
analyzer. Total alkalinity and turbidity were measured at all stations
along the tag line closest to the skimmers (skimmers are outlet devices
which prevent materials floating on the ash pond surface from being
carried out in the effluent) and at each 50 percent station. The ash
pond was again stratified with respect to temperature the first day and
completely isothermal the third day of the survey. The first day was
extremely calm while the third was windy. The stratification is believed
to be a result of the heated discharge to the pond. The temperature of
the water entering the pond is elevated by addition of the hot ash.
During calm conditions, the water entering the pond appears to spread
out over the surface of the pond and flow across it in a thin stratified
layer, whereas during windy conditions the water entering the pond mixes
with the water already in the pond. Therefore, wind conditions appear
to determine the mixing of the pond.
During the period of stratification, the difference in temperature
from top to bottom decreased as the flow approached the outlet of the
pond. This decrease was a result of both surface cooling and mixing.
Although thermally stratified the first day, the pond showed no mea-
surable difference in pH, DO, and alkalinity from top to bottom. The
conductivity was higher at the surface than at the bottom during periods
of thermal stratification. However during isothermal periods the con-
ductivity was lower at the surface than at the bottom. The turbidity
was normally constant top to bottom during isothermal conditions, while
during stratified conditions it was slightly higher at the top than at
the bottom.
-------
NFUW
, M.5 W.S
EFFUUEMTfc (U.5)
Ql/
oo
I
o— •
«— 400 •
•-4OO — P
•-400 *
^ -r
tFLOW
TOTAL ALKALNTY
TURBWTY {JTU)
TEMPERATURE CO
Figure 10. Vertical Profile of Ash Pond Characteristics of Plant E During Thermal Stratification and Isothermal Periods
(the numbers in parenthesis represent readings approximately 1 foot from the bottom,
while the other numbers represent readings just below the surface)
-------
11.1
(U.I)
11.1
(11.1)
1 u.o
I (11.03)
./
20 *-4OO
11.05
(11.05)
11.1
(11.05)
11.0
(U.O)
— » • — 4(
11.05 11
(11) (11)
11.05 11
(11) (11)
U.O U
(U) (11)
» — •
p»-400— •
U
(Ul
U
(Ul
(iou«<;
CIO
(9.9)
10
„""'
EFFUOtT
C
-w
o 9.8
o (10)
•*
to
/
•--40O-
9.9 9.9 9.9 9.9
(9.91 (101 <10) (10)
9.9 9.9 10 9.9
(10) (10) (10) (10)
9.9 9.9 9.9 9.9
(11) (10) (10) (10) C^
— •
• — 400 — •
•— 400 — •
h«— 400 — •
V£>
I
PH
Dissolved Oxygen (mg/0
CONDUCTIVITY (tt r*os)
Figure 10 (Continued)
-------
-80-
During the preliminary survey, an unsuccessful attempt was made to
determine the detention time of the ash pond using Rhodamine WT dye.
The dye was injected into the sluice lines within the plant. However,
the dye was adsorbed onto the fly ash and settled out with the ash
making detection of the dye impossible. Therefore, during the second
survey the Rhodamine WT dye was poured into the headwaters of the pond
at 8:00 a.m., on February 23, 1976. Effluent samples were collected
every 30 minutes and analyzed for dye concentration by a fluorometer.
Dye was first detected in the effluent 7 hours after the dye was
injected into the pond. The peak dye concentration occurred 12 hours
after injection (see Figure 11), but dye concentrations up to 4 ppb were
still detected after 43 hours. This indicates that the flow through the
ash pond was plugflow with some mixing taking place.
The dye was injected into the pond during thermal stratification
and thus moved across the surface of the pond reducing the detention
time to approximately 12 hours. However, the morning of the 24th was
fairly windy and the pond had become mixed. Therefore, the remaining
dye was redistributed throughout the pond resulting in a slow decline of
the dye concentration after the morning of the 24th. Had the dye been
injected when the pond was completely mixed, the peak would have occurred
later and been less intense than the one shown in Figure 11. Based on
this dye study and the pond profiles, the detention time of the Plant E
ash pond is believed to vary from approximately 12 hours to 7 days
depending on the state of mixing in the pond.
During the ash pond survey in October 1975, samples of the sluice
water before additions of the ash were collected by allowing a valve
located in the intake pumping system to drip continuously for 8 hours
during each of the three sampling days. This provided three 8-hour
composite samples which represented the characteristics of the water
used for sluicing. In addition, grab samples of the ash pond effluent
were collected by a Circo automatic sampler at 30 minute intervals and
composited every 12 hours for a 72 hour period. Both influent and
effluent composite samples were analyzed for total and suspended solids
and total and dissolved Al, Ca, Cr, Cu, Fe, Mg, Pb, Zn, S04, and Si.
References for the methods used to analyze these samples are given in
Appendix A. A summary of the quality control data for TVA's Division of
Environmental Planning's Laboratory Branch is given in Appendix B. The
results are shown in Table 14. For all elements except solids, the
total and dissolved concentrations were determined analytically and the
suspended concentration determined by subtraction of the dissolved
concentration from the total concentration. The majority of the Cu and
Fe in the effluent appears to be associated with the suspended solids.
The dissolved form is the predominate form for the remaining elements.
The concentration of suspended calcium for the data given in Table 14 is
greater than the suspended solids concentration indicating an error or
lack of precision in the analytical procedures used. The analytical
procedure used to determine the suspended metal concentration is the
reason for this inconsistency. The suspended Ca concentration was
calculated by first analyzing for the total and dissolved concentrations
and then subtracting the dissolved from the total. The suspended solids
concentration was determined directly by analysis and thus is a more
accurate estimate of its value than the value given for the suspended Ca
concentration.
-------
c
o
0>
u
oo
4 -
3/23/76
3/24/76
3/25/76
3/26/76
Figure 11. Concentration of Rhodomine WT Dye In Plant E Ash Pond Effluent with Time
-------
Table 14. CHEMICAL ANALYSIS OF ASH POND EFFLUENT AND INTAKE WATER USED FOR SLUICING
DURING PRELIMINARY SURVEY AT PLANT E
Elements
Aluminum
Calcium
Chromium
Copper
Iron
Magnesium
Lead
Zinc
Solids
Sulfate
Silica
(mg/D
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Total
Dissolved
Raw Water Supply
9/30
0.9
<0.2
34
26
<0.005
<0.005
0.1
0.05
0.36
0.06
4.1
4.1
0.02
<0.01
0.04
0.02
110
6
10
10
3.1
2.1
10/1
0.9
0.4
32
28
<0.005
<0.005
0.14
0.07
0.27
0.07
5.0
4.9
0.028
0.012
0.04
0.04
100
2
10
10
2.9
2.8
10/2
0.6
<0.2
32
28
<0.005
<0.005
0.11
0.06
0.34
0.05
5.1
5.1
0.017
<0.01
0.05
0.03
110
2
11
9
2.5
2.0
9/29
Night
3.0
2.4
180
150
0.023
<0.005
0.04
0.01
0.43
0.06
0.3
0.3
<0.01
<0.01
0.04
0.02
490
22
150
150
3.2
3.0
Pond Outfall
9/30
Day
3.0
2.4
180
160
0.031
0.031
0.03
<0.01
0.35
0.05
0.3
0.3
<0.01
<0.01
0.05
0.04
400
22
150
150
3.5
3.5
Night
2.4
2.3
180
170
0.03
0.03
0.06
0.02
0.15
<0.05
0.2
0.2
0.03
<0.01
0.03
a
470
4
150
150
4.1
3.0
10/1
Day
2.8
2.4
180
160
0.026
<0.005
0.03
0.01
0.11
0.05
0.2
0.2
<0.01
<0.01
0.03
0.03
420
2
140
140
2.7
2.6
Night
2.7
2.7
170
150
0.034
0.012
0.06
<0.01
0.14
<0.05
0.3
0.3
<0.01
<0.01
0.03
0.03
360
11
150
150
2.9
2.9
10/2
Day
2.7
2.7
180
160
0.04
0.006
0.02
<0.01
0.10
<0.05
°-2 s
0.2 '
<0.01
<0.01
0.03
0.02
400
2
140
140
3.6
3.4
Not reported
-------
-83-
During the second ash pond survey, 500 milliliters of the intake
and effluent samples were filtered and the residue on the filter pad
analyzed for suspended metal concentrations. The intake samples were
collected in the same manner as during the preliminary survey. However,
the effluent samples were composited every six hours rather than twelve
hours as before. In an effort to minimize laboratory costs, only four
effluent samples and two intake samples were chosen for suspended metal
analysis. The results of this February survey are shown in Tables 15
and 16.
The average concentrations of each element in the intake water
supply and the ash pond effluent, and the difference between the two
were calculated for each survey using all of the data in Table 14 and
selected data (sample numbers 2, 3, 5, 7, 9, and 18) from Tables 15 and
16. The results of these calculations are shown in Table 17. The sum
of suspended and dissolved values given in Table 17 may not add up to
the total shown for several parameters due to round off errors and the
treatment of less than values during the averaging of the original data
sets. A negative sign indicates a decrease in concentration from the
intake water supply to the ash pond effluent while a positive number
indicates an increase in concentration.
The following conclusions were derived from Table 17.
1. Both surveys indicated that the total and dissolved aluminum
concentrations increased from intake to effluent while the
suspended concentration decreased.
2. The total calcium and chromium concentrations increased from
the intake to the effluent with the majority of the increase
being in the dissolved phase.
3. The total, dissolved, and suspended concentrations of copper
and magnesium decreased from the intake to the effluent.
4. The total iron concentration decreased from the intake to the
effluent. The decrease occurred both in the suspended and
dissolved phases. The majority of the iron remaining in the
effluent was in the suspended form.
5. Both the suspended and dissolved lead concentrations decreased
from the intake to the effluent. The lead concentration in the
effluent was usually near or below the minimum detectable limit.
6. The total zinc concentration decreased only slightly and the
decrease was primarily in the dissolved phase.
7. The sulfate and silica concentrations increased from intake to
effluent. The increases were in the dissolved form.
8. The total arsenic concentration increased from the intake to
the effluent. The increase was in the dissolved form. The data
for the February survey indicates that the dissolved concentra-
tion was higher than the total concentration. This is in error
and is attributed either to laboratory or sampling errors.
-------
TABLE 15. CHEMICAL ANALYSIS OF ASH POND EFFLUENT AND INTAKE WATER USED
FOR SLUICING DURING THE FEBRUARY SURVEY AT PLANT Ea
Solids
Sample
Number
l
*l
3b
4
5b
6b
7b
8
9b
10
11
12
13
14
15
16
17,
18b
19
20
Date
2/18/76
2/19/76
2/20/76
2/22/76
2/23/76
2/23/76
2/23-24/76
2/24/76
2/24/76
2/24/76
2/24-25/76
2/25/76
2/25/76
2/25/76
2/25-26/76
2/26/76
2/26/76
2/26/76
2/26-27/76
2/27/76
9
8
8
2
8
2
8
2
8
2
8
2
8
2
8
2
8
2
Time
a.m. -2
a.m. -2
a.m. -2
p.m. -8
p.m. -2
a.m. -8
a.m. -2
p.m. -8
p.m. -2
a.m. -8
a.m. -2
p.m. -8
p.m. -2
a.m. -8
a.m. -2
p.m. -8
p.m. -2
a.m. -8
p.m.
p.m.
p.m.
p.m.
a.m.
a.m.
p.m.
p.m.
a.m.
a.m.
p.m.
p.m.
a.m.
a.m.
p.m.
p.m.
a.m.
a.m.
Sample
Location
Intake
Intake
Intake
Intake
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Effluent
Susp
mg/1
54
15
16
8
8
4
3
4
5
7
4
3
3
2
2
I
-------
TABLE 15 (continued)
Sample
Number
!
2b
3b
4
5b
6V
7b
8
9b
10
11
12
13
14
15
16
17
I8b
19
20
Calcium Copper
mg/1 mg/1
22
21
21
23
93
97
110
110
100
94
91
95
87
92
93
98
88
87
90
91
0.02
0.02
0.24
0.05
<0.01
<0.01
<0.01
<0.01
0.09
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Iron
mg/1
1.7
1.3
2.2
0.81
0.30
0.15
0.05
0.13
0.14
0.13
0.09
0.21
0.16
0.26
0.10
0.08
0.06
0.10
0.09
0.09
Manganese
mg/1
0.07
0.05
0.33
0.05
0.04
0.03
0.02
0.05
0.02
0.01
0.01
0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Magnesium Zinc Aluminum
mg/1 mg/1 mg/1
3.1
3.2
3.2
3.4
0.5
0.5
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.4
0.5
0.5
0.03
0.03
0.14
<0.01
<0.01
0.03
0.11
0.16
<0.01
0.05
0.08
0.04
<0.01
0.02
0.02
<0.01
<0.01
<0.01
<0.01
<0.01
2.4
1.6
1.6
0.9
1.7
1.7
1.6
1.6
1.7
1.7
1.9
1.9
2.0
2.0
2.0
1.7
1.6
2.0
1.9
1.8
Arsenic0 pH
M8/1
<5
<5
<5
<5
5
5
5
10
5
10
5
5
5
5
5
5
5
10
20
20
8.0
7.9
7.7
8.5
10.95
11.0
11.23
11.2
11.3
11.3
11.1
11.2
11.3
11.3
11.3
11.3
11.3
11.35
11.3
11.36
TDSd
mg/1
91
85
85
90
330
350
420
400
385
400
385
400
400
390
410
395
390
370
350
385
Tempd
°C
14
14.5
13.5
14.5
15.5
14.5
14.5
14.5
15.0
15.0
15.0
15.5
15.5
15.5
15.5
All metal concentrations are total.
Chosen for suspended metal analysis.
cBased on past data all intake samples were assumed to be less than the minimum detectable
limit of 5 M8/L-
jField measurement.
i
00
Ln
t
-------
-86-
TABLE 16. SUSPENDED METALS CONCENTRATION
FOR THE FEBRUARY ASH POND SURVEY AT PLANT E
Sample Number
Location
Suspended Solids (mg/1)
Chromium ((Jg/1)
Lead (pg/l)b
Calcium (mg/1)
Copper (mg/1)
Iron (mg/1)
Magnesium (mg/1)
Zinc (mg/1)
Aluminum (mg/1)
f+
Arsenic (pg/1)
2
Intake
15
-------
-87-
TAIH.I-: 17.
AVERAGE CHEMICAL ANALYSIS OF THE ASH POND EFFLUENT AND INTAKE WATER SUPPLY
DURING BOTH ASH POND SURVEYS
Element
Aluminum Total
Diss
Susp
Manganese Total
n-ss
Arseni<-'
Total
Diss
Susp
Averngc Intake
Water Supply
mg/1
0.8
0.27
0.67
Ci lei lit"
Chromium
Copper
Iron
Magnesium
Lead
Zinc
Solids
SulfaU-
Silica
T-tal
0 i s s
Susp
Total
Diss
Susp
Total
IJiss
Susp
Total
Diss
Susp
Total
Diss
Susp
Total
Diss
Susp
Total
Hiss
Susp
Total
Diss
Susp
Total
Uiss
fiusp
Total
Uiss
'•usp
33
27
6
<0.05
<0.05
<0.05
0.12
0.06
0.06
0.32
0.06
0 . 26
4.73
4.7
0.03
0.02
0.011
0.017
0.043
0.03
0.01.3
107
104
3
10.3
-------
-88-
9. The total and dissolved solids concentrations increased from
the intake to the effluent but the suspended solids concentra-
tion remained approximately the same, sometimes increasing and
sometimes decreasing.
Other aspects of these surveys as they relate to developing an ash
pond monitoring program for Plant E will be discussed in the following
subsections. The results of the February survey confirmed the findings
of the September survey. Because of improved sampling procedures for
trace metals (collection of suspended and dissolved samples for metal
analysis) a better estimate of the form in which the various metals
occurred was obtained. A dye study was attempted during the first
survey but the dye was injected before the majority of the ash had had
time to settle out and the dye was absorbed into the ash and disappeared.
During the second survey a successful dye study to determine the detention
time of the ash pond was carried out by injecting the dye into the pond
after the ash had settled out. The thermal stratification of the ash
pond observed during the first survey was also confirmed.
SUMMARY OF THE ASH POND EFFLUENT CHARACTERISTICS AT PLANT E
A summary of the ash pond effluent characteristics based on the two
ash pond surveys and the quarterly monitoring program for 1974 and 1975,
is given in Table 18. The summary also includes the maximum and minimum
values reported during that sample period, indicating the range over
which the effluent characteristics vary. The October 1975 survey spanned
3 days while the February 1976 survey spanned 4 days. The quarterly
monitoring program covered six quarters or a year and a half starting in
mid-1974. The data indicate that except for Fe, Mg, and Zn, the averages
for each sample period differed for each element. However, the range
(the difference between the lowest and highest values) is greatest for
the quarterly sampling program indicating that the effluent characteris-
tics vary more over a period of a year rather than within a day. The
weekly effluent data from 1970 to 1975 for Plant E, showed that there
was no yearly cycle for flow, pH, or suspended solids but that there was
a yearly cycle for alkalinity and dissolved solids. The data from the
two ash pond surveys did not indicate a daily cycle for any of the
elements except possibly Cu and Fe during the preliminary survey. Both
Cu and Fe were consistently higher in the samples collected at night
than those collected during the day. However, this was not observed
during the February survey. Since the two surveys only span either 3 or
4 days, enough data is not available to ensure a weekly cycle does not
exist. However, based on the data, there is no reason to believe a
weekly cycle does exist.
As concluded at the end of Section 3, except for pH there is insuffi-
cient data on those parameters required by the NPDES permit for Plant E
to adequately estimate the true yearly mean. Therefore, a more inten-
sive sampling program of the ash pond effluent at Plant E was conducted
from May 1976 to February 1977 to better estimate the effluent characteris-
tics. Samples were collected on a varying workday of each week. For
example, a sample may have been collected on a Thursday one week and
Tuesday the following week. This was done to avoid sampling at exactly
one week intervals in hopes of detecting a weekly cycle, if one exists.
-------
TABLE 18. SUMMARY OF THE ASH POND EFFLUENT CHARACTERISTICS AT PLANT E FOR THE
TWO ASH POND SURVEYS AND THE QUARTERLY MONITORING PROGRAM DURING 1974 AND 1975
Element
Aluminum (mg/1)
Calcium (mg/1)
Chromium (mg/1)
Copper (mg/1)
Iron (mg/1)
Magnesium (mg/1)
Lead (mg/1)
Zinc (mg/1)
Dissolved
Solids (mg/1)
Suspended
Solids (mg/1)
Dissolved
Sulfate (mg/1)
Dissolved
Silica (mg/1)
Arsenic (mg/1)
Manganese (mg/1)
Quarterly Monitoring Program
Minimum Avg . Maximum
1.1
68
<0.005
0.02
0.05
0.1
<0.01
<0.03
240
2
100
5.9
0.03
<0.01
2.5
126
0.017
0.08
0.16
0.3
0.017
0.05
368
4
147
7.0
0.06
0.01
3.0
170
0.022
0.19
0.39
0.3
0.036
0.07
420
6
210
8.4
0.09
0.02
October 1975 Survey
Minimum
2.4
170
0.023
0.02
0.10
0.2
<0.01
0.03
349
2
140
2.6
NA
NA
Avg.
2.77-
178
0.031
0.04
0.21
0.25
0.01
0.035
413
11
147
3.1
NA
NA
Maximum
3.0
180
0.04
0.06
0.43
0.3
0.03
0.05
468
22
150
3.5
NA
NA
February 1976 Survey
Minimum
1.6
87
<0.005
<0.01
0.05
0.4
<0.01
<0.01
200
«
85
6.1
0.005
<0.01
Avg.
1.8
95
0.014
0.015
0.13
0.43
<0.01
0.04
210
4
90
6.5
0.008
0.02
Maximum
2.0
110
0.036
0.09
0.30
0.5
<0.01
0.16
230
8
94
7.2
0.02
0.05
I
00
-------
-90-
Grab samples were collected by representatives from TVA's Division of
Environmental Planning and shipped to the Laboratory Branch in Chattanooga
Tennessee. The samples were analyzed for the following parameters which
are required by the NPDES permit for Plant E: ph, flow, suspended
solids, total arsenic, chromium, copper, iron, lead, manganese, selenium,
and zinc. The NPDES permit also calls for cadmium, mercury, and nickel
to be monitored, however, these elements were not included in this study
because previous data (see Table 5) indicated the concentrations were at
or below the miminum detectable limit. In addition, the samples were
analyzed for aluminum, calcium, magnesium, dissolved silica, sulfate and
dissolved solids. These elements were included because previous data
indicated their presence. The samples were collected in the ash pond
discharge prior to mixing with any other waste stream as required by the
NPDES permit. During the sampling period the plant was operated as
normal, including the discharge of routine chemical-cleaning wastes and
air-preheater washdown to the ash pond.
The results of this intensive sampling program are given in Table
19. Beginning October 14, 1976, only every other sample collected was
analyzed because the data did not vary substantially from week to week.
A summary at the bottom of the table gives the minimum, mean, and maximum
values for each element. Linear correlation coefficients were developed
between elements. A significant correlation at the 95 percent signifi-
cance level is represented by an R value greater than 0.325 in Table 20
(2). The following parameters were correlated significantly with pH:
conductivity, calcium, dissolved solids, and dissolved silica. Copper
was negatively correlated with pH. The following parameters were signifi-
cantly correlated with dissolved solids: chromium, conductivity, dissolved
silica, and sulfate. Turbidity was negatively correlated with dissolved
solids, calcium, chromium, dissolved silica, and sulfate. Suspended
solids were significantly correlated with flow, turbidity, and aluminum.
Suspended solids were also negatively correlated with chromium, selenium,
dissolved solids, and dissolved silica. Aluminum was significantly
correlated with iron, manganese, and zinc in addition to turbidity and
suspended solids. Calcium was significantly correlated with iron,
manganese, selenium, and dissolved silica. Copper was significantly
correlated with magnesium. In addition to aluminum and calcium, iron
was significantly correlated with manganese. These correlations
indicate that the heavy metals in the ash pond effluent are definitely
interrelated with one another.
Several of the R values indicated relationships between parameters
which may be beneficial to a monitoring program. For example, turbidity
could be used as an indication of the suspended solids concentration or
conductivity could be used to indicate the dissolved solids concentration.
These two relationships have been used extensively by industry for auto-
mation of monitoring programs. However, the relationship between flow
and suspended solids may be more beneficial to controlling suspended
solids, especially if the flow could be controlled to ensure a given
suspended solids concentration. Figure 12 shows the linear regression
for flow versus suspended solids. According to the data in Figure 12,
if the flow is maintained between 17,500 and 24,320 gallons per minute
(gpm) the average suspended solids concentration should be 30 mg/1
assuming a linear regression and 95 percent confidence level. It also
-------
TABLE 19. ASH POST) EFFLITVT CKAJUCTERISTICS FOR PLA.YT E
O.t,
5/13/76
5/17/74
e/01/76
6/11/74
6/16/74
6/24/76
6/29/74
7/08/*4
7/20/74
7/26/76
8/03/76
8/12/75
8/17/76
8/26/76
9/10/76
9/16/76
9/20/74
9/30/;6
0/07/74
0/14/76
0/28/74
l/U/76
1/22/-6
2/10/76
2/20/74
l/OJ/77
1/20/77
2/r,3/77
2/15/77
2/25/7-
l.,u.u.
*,v*r*R»
l*.ii.u»
4170
4470
4170
632')
7380
7100
6120
7100
SA
7920
9050
10720
0720
1330
0970
1130
4850
5370
4420
5260
10840
11530
10900
15«80
200JO
25400
10?40
14390
;4440
-170
9570
25300
pH
11.3
11.2
10.8
11.0
11.2
11.0
11.3
11.0
SA
11.5
11.2
11.3
11.3
11.3
11.1
11.3
11.2
11.5
!I. 3
11.3
10.8
11 3
11.4
112
11.0
10.9
11.5
11.2
11.1
i: 2
:o 7
1 1 :
11.5
fjJho*/rm)
832
081
510
547
667
630
810
045
SA
1 120
805
1070
1030
1020
830
1060
890
1 l&j
835
810
6SO
850
757
580
427
960
8«2
697
745
42
79
116
(JTU)
SA
NA
SA
1-5
I.I
-------
TABLE 20. LINEAR CORRELATION COEFFICIENTS FOR THE VARIOUS ASH POND EFFLUENT PARAMETERS AT PLANT E
Flow
pH
Conductivity
Turbidity
Aluminum
Calcium
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Zinc
Selenium
Arsenic
Dissolved Silica
Sulfate
Dissolved Solids
Suspended Solids
Flow
1.000
0.127
0.080
0.652
0.254
0.137
-0.405
-0.015
-0.128
-0.047
-0.126
0.001
-0.022
-0.247
-0.065
-0.450
-0.049
-0.217
0.812
pH Conductivity
1.000
0 . K26
-0.193
-0.036
0.450
0.113
-0.392
0.192
-O.US
-0.125
0.154
-0.210
0.240
-0.033
0.457
-0.051
0.375
-0.090
1.000
-0.222
0.021
0.515
0.182
-0.271
0.287
-0. 101
-0.234
0.305
-0.213
0.128
-0.231
0.528
0.235
0.578
-0.213
Turbiilitv
1.000
0.368
-0.373
-0.403
-0.028
0.134
-J.098
-0.055
0.025
0.023
-0.293
-0.078
-0.403
-0.581
-0.332
0. 707
Aluminum
1.000
0.075
-0. 122
0.268
0.369
-0.030
-0.076
0.426
0.428
-0.060
-0. 187
-0.159
-0.054
-0.100
0.430
Calcium
1.000
0.201
-0.002
0.311
0.032
0.033
0.374
0.225
0.474
-0.165
0.418
0.200
0.26J
-0.294
Chromium
1.000
-0.2/6
-0.218
0.274
-0.283
-0.184
-0.210
0. 102
-0.042
0.229
0.41,2
0.514
-0.418
Copper
1.000
0.003
-0 125
0.580
0.126
0.244
-0.068
0.006
-0.211
0.089
-0.158
0.009
I ron
1.000
-0.03i
0.033
0.959
0.01"!
0.08?
-0.037
0.064
-0.051
-0.018
0. !01
Lead
0.
-0.
-0
0
-0
-0
0.
-0
-0.
OCu
.065
053
.086
.105
.008
080
180
.159
043
Magnes mm
I
0
-0
-0
0
-0
0.
0
-0.
.000
.039
.083
.036
. 147
.117
.061
. 199
036
Dissolved Dissolved Suspfinlfl
Manganese Zinc Selenium Arsenic Silica Sulfate Solids Soli'li
1
VO
1 . 000
0.079 1.000
0. 108 -0.063 1 -000
-0.064 -0.125 0.089 1 .000
0.172 0.005 0.327 -0.025 1.000
-0.011 0.228 0.041 -0.250 -0.146 1.000
0.027 -0.190 -0.073 -0.171 0.408 0 . 3<>6 1.000
-0.016 0.043 -0.317 -i).087 -0.047 -0.114 -0.399 1.000
-------
(0
o
CO
O)
•o
i
en
CO
—'indicates 95% confidence limit for mean
predicted value
A
A
10,000 20,000
Flow (gal/min)
30,000
Figure 12. Relationship Between Flow and Suspended Solids in the Ash Pond Effluent at Plant E
-------
-94-
shows that for a flow of 9,570 gpm the average suspended solids should
be between 8.5 and 12.5 mg/1 95 percent of the time. This agrees with
the data in Table 19. The average flow was 9,570 gpm and the average
suspended solids was 11 mg/1. At flows above 12,500 gpm there is more
lack of fit to a straight line in the data.
There are three possible explanations for the relationship between
suspended solids and flow. The first is that the large changes in the
ash pond flow correspond to the operational status of unit 5. The flow
increases when unit 5 is on-line and decreases when it is off-line. The
new electrostatic precipitator on unit 5 may produce an ash which differs
chemically and physically from the ashes from the other units. This ash
may not settle as well as the ash from the other units and thus the
apparent relationship with increased flow. The second explanation is
that the increased flow causes an increase in the velocity of the water
spilling over the skimmers which may cause an increase in the quantity
of cenospheres discharged with the effluent. The third explanation may
be a reduced detention time in the pond when unit 5 is operating.
However, more data at the higher flows are needed along with the settling
characteristics of the different ashes from the various units, the
detention time, and the effect of weir overflow rate on the discharge of
cenospheres before the apparent dependency of suspended solids on flow
can be properly explained.
The summary of the ash pond effluent characteristics for Plant E
given in Table 19 will now be used to complete steps 3 through 8 of the
design procedure summarized in Section 4.
VARIATION OF THE ASH POND EFFLUENT CHARACTERISTICS AT PLANT E WITH TIME
The variation with time of the ash pond effluent characteristics
given in Table 19 for Plant E is shown in Figure 13. Except for flow
and possibly suspended solids, there does not appear to be any trend or
cycle over the sample period for any of the effluent constituents. Flow
appears to increase in August and then again in December. This increase
in ash pond flow is due to:
1. The operation of the electrostatic precipitator on unit 5 which
was placed into service on June 1, 1976, resulting in an addi-
tional flow of 5.2 mgd. The time required for startup before
reaching full operation may account for the 2-month lay between
June and the flow increase in August.
2. Increased plant capacity factor due to increased hours of opera-
tion during extreme hot and cold weather periods resulting in
more frequent and longer duration of ash sluicing.
Suspended solids also appear to exhibit an increasing trend or
possibly a yearly cycle. The reason for this may be due to the pre-
viously discussed relationship of suspended solids with flow. If this
is the case, then, theoretically, suspended solids is not random over a
year's period. However, because previous data did not indicate a yearly
cycle, the relationship with flow has not been confirmed and for ease of
-------
JV,W^J
20POO
1
1
10,000
0
11.6
114
11.2
-------
ouu
^ 2OO
Concentrotion (m<
8
12
JT 9
'o
8
3
^
Co
A
_
AA A
: * V~ A A\
AA AA A A
^ A A
A
A
AMJJASONDJ FM
1976 1977
A Mg
-
_
-
A A
A AA A
. 4L A*CAA.A A,, ^& A A AA ^A
AMJJASONOJFM
2U
15
•^
Concentration (mg
5
5
0
15
I
|,o
a
5
Al
A
A
A
. A^
/^A^AA^^ A A A A A
AMJ JASO NO J FM
1976 1977
Fe
A
-
-
A
-
A
A AA
A A
AMJJASONOJFM
(976
1977
1976
1977
Figure 13 (Continued)
-------
VAW+
0.03
^^
^C
J"
1 Q02
•g
S
J
0.01
A
Mn
_
A
~
-
_
A
A
..
" A^wv *A A
, jf\ ^Vyg\ JL AA A A AA A ^\ A A A A A A A
yyv i ^x^yx /y\ ^v^^ "*\ /\ . /\/v s\ xy\ *\jf\
O.O25
0.020
^^
~
o>
•| 0.015
o
1
§ 0.010
J
O.O05
f\
^ Se
A
A A
A AA A
A A A A
A A A
A A
A A
A AAA
A
A
^
A
1 1 1 1 1 1 1 1 i 1 i
Q
AMJJASONDJFM "AMJJASONDJFM
1976 1977 1976 1977
0.04
v!
J 0.03
g
§ 002
J
0.01
0
Cr
A ^^
A
A A A
A A
- A AA A
A A
" * A * ^ A
A
— ^V A A. ^AA
•"• A. A *\ ^\ ^ti
A A
,,111111111
AMJJASONDJFM
O
6
^
9
C
Q
|
I
2
0
A A A A Si
A ^^ A
^^ A A A
A A A A A
A A A
A A
'
A
-
i i i i i i i i i i i
AMJ JASONDJFM
IQ7K IQ77 '976 1977
Figure 13 (Continued)
-------
U£3
020
o
§. QI5
c.
O
o
I oio
C.
S
0.05
Q
A CU
-
"
-
A A
A
~
,-
A A A
A
AA A
A A
A A ^3C\A^ ^SA A,
/^k. AA AA A AA A
iiiiiiiiiii
0.20
o>
^ 0.15
c
.0
"o
| OJO
o
o
0.05
f\
Pb
_
/& « A ^
^^^"V^C^M ^^*^vYs 4A/i^ ^ A/v MA ^ ^ A A
IIIIIIIIIII
AMJJASONDJFM "AMJJASONDJFM
1976 1977 1976 1977
i
O.lO
O.O8
0
9
.§ Q06
c
o
1
"c
g O.O4
002
f\
As
- A
_
-
-
—
A
AAA A
A A A A A
A AA A A A AA A A
A A ^^ ^- A
A A A A AA
\M*.VV*
005
I. 004
E
g
1 003
"c
§ 0.02
001
r\
/L\
Zn
"
A
-
A
_
A
A
A A.
A ^ y^^ >^ A
co
AMJJASONDJFM "AMJJASONDJFM
1976
1977
Figure 13 (Continued)
1976
1977
-------
•JV
40
o>
-§ 30
o
o
w
3 20
§
10
Q
Suspended A
Solids A A A
-
A
A
A A
A A
A
A 4& A A
A -^k A A
, A^ , A^ A A , ^£ , ,
BOO
600
I
C
o
'g 4OO
§
C
zoo
r\
A Dissolved
Solids
AA A A A
/\AQ^ A
.A .^ ...
A A^ ^^ A
A A A A
- A A
_
AMJJASONOJFM ~AMJJASONDJFM
1976 1977 1976 1977
300
|
c
•| 200
S
c
o
100
0
h A S(^
-
^| ^^
AAAA AA A
A * A A
4^ A A A A
AT/T^ ^^ A
A
A
A
i i i i i i i i i i i
AMJJASONDJFM
vO
I
Figure 13 (Continued)
-------
-100-
statistical analysis, the variation in suspended solids will be assumed
to be random. None of the other ash pond effluent constituents appeared
to be cyclic; therefore, they will also be assumed to be random.
By assuming the data to be random, the data set does not need to be
divided into smaller data sets because the data can be assumed to esti-
mate the effluent characteristics for almost any time period even though
the data were collected over a 10-month period. In other words, the
same statistical values, such as the mean and standard deviation, should
be expected for 30 samples collected randomly over 30 days as those
obtained for 30 samples collected over 365 days. This will be useful
later when the objective of the monitoring program is to show compliance
over a time period other than 10 months. Such is the case with suspended
solids. There are, however, practical limits over the time periods for
which the data set given in Table 19 should be used to estimate the
effluent characteristics. For example, to use this data to estimate the
daily average or 10-year average may not be wise. Other statistical
procedures such as those suggested by Box and Jenkins 1970 (10) should
be consulted for extrapolations of this magnitude.
STATISTICAL DISTRIBUTION OF THE EFFLUENT CHARACTERISTICS AT PLANT E
Cumulative frequency plots were prepared for the ash pond effluent
data given in Table 19 according to the method outlined by Miller and
Freund (8). These plots are shown in Figure 14. The best fit straight
line was determined by visual placement. In the top row of Figure 14,
the data are plotted to the log base ten scale while the data in the
bottom row are plotted on an arithmetic scale. The two sets of plots
were compared for each parameter to determine which plot of the data
yielded the straightest line. This determination was made by visual
inspection. If the data in the bottom row were closest to a straight
line, then the parameter was assumed to follow a normal distribution.
If the data in the top row was closest to a straight line, then the
parameter was assumed to follow a lognormal distribution. If there did
not appear to be a difference, then a normal distribution was assumed
to simplify calculations. Also, for ease of calculation these were the
only two distributions considered.
Table 21 lists the various parameters and the assumed distribution
based on this comparison. The following parameters were assumed normal:
arsenic, chromium, lead, pH, selenium, dissolved silica, and sulfate.
The following were assumed lognormal: aluminum, calcium, conductivity,
copper, dissolved solids, iron, magnesium, manganese, suspended solids,
turbidity, and zinc. These distributions are in agreement with those
reported by Berthouex and Meinert (7) for surface waters in the Tennessee
Valley. An exception is lead which followed a normal distribution in
the ash pond effluent while that in the surface waters in the Tennessee
Valley was reported to follow a lognormal distribution. However, a con-
siderable portion of the samples had Pb concentrations below the minimum
detectable limit. Table 21 contains additional information which will
be discussed in the following subsection.
-------
1000 i
Conductivity
10
o_
100
2 5 10 20 40 60 80 90 95 98
2 5 10 20 40 60 80 90 95 98
I2O
100
I
80 -
6O
Conductivity
40
2 5 IO 20 40 60 80 9O 95 98
Cumulative Frequency (%)
2 5 10 20 40 60 80 90 95 98
Cumulative Frequency (%)
11.9 -
o
h-*
I
10.5
2 5 10 20 40 60 80 9O 95 98
Cumulative Frequency (%)
Figure 14. Cumulative Frequency Plots for the Ash Pond Effluent at Plant E
-------
0.1 r r 1 1 1 1—i 1—i 1 1 ,
Min. Detectable Limit
O.OOI ' ' ' 1 1 i—i—i—i J
Z 5 IO ZO 40 6O 8O 9O 95 98
1.00
0.10
' I 1 1—i—I—1 1 r
Pb
-Mia Detectable Limit
0.100
0.010
2 5 10 2O 4O 60 80 90 95 98
O.OOI
As
2 5 10 20 40 60 80 9O 95 98
0.04
= 0.031-
o>
0.02'-
0.01 h
0.02O h
0.015 I-
O.OIO
Q005I-
'2 5 10 20 40 60 BO 90 95 98
Cumulative Frequency (%)
o.ooo
0.04 I-
0.03
0.08 I-
0.01 (-
2 5 10 20 40 60 8O 90 95 98
Cumulative Frequency (%)
o
N?
I
2 5 IO 20 40 6O 8O 9O 95 98
Cumulative Frequency (%)
Figure 14 (Continued)
-------
1000 I r
100 I
Si
10
O
o
IOO
10
Al
2 5 10 20 40 60 8O 9O 98 98
2 5 10 20 4O 60 8O SO 96 96
2 5 10 20 4O 60 SO 9O 95 98
4OO
= 3OO
o>
.2
5
IOO -
Co
I II I 1_
5 IO 2O 4O 6O SO 9O 95 98
Cumulative Frequency (%)
2 5 10 20 4O 60 SO 90 95 98
Cumulative Frequency (%)
14 (Continued)
Al
o_
2 5 10 2O 40 60 SO 90 95 98
Cumulative Frequency (%)
o
w
I
-------
10
1
§
O.I
0.01
-i—i 1—i—i—i—i—i—i 1—r
Mn
2 5 10 20 40 60 80 90 95 9B
2 5 10 20 40 60 80 90 95 98
2 5 10 20 40 60 SO 90 95 98
Z 5 10 20 40 6O 80 90 95 98
Cumulative Frequency (%)
4 -
o
Mg
Q
2 5 10 20 40 60 80 90 95 96
Cumulative Frequency (%)
Figure 14 (Continued)
4 -
Fe
o
Z 5 IO 20 40 60 80 90 95 98
Cumulative Frequency (%)
-------
o.too
2 o.oio
§
S
0.001
Se
o
o
o .
I I I I 1 1 I I I
0.01
1.0
O.I —
2 5 IO 20 4O 60 SO 9O 95 98
2 5 10 20 40 60 80 90 95 98
O.OI
Zn
2 5 10 20 40 60 80 9O 95 98
QOOO
2 5 10 2O 4O 6O 80 90 95 98
Cumulative Frequency (%)
0.20 -
0.15 I-
0.10
0.05 -
2 5 10 20 40 60 80 90 95 98
Cumulative Frequency (%)
I'lpure 14 (Continued)
0.4 -
0.3 -
0.2 -
2 5 10 20 40 60 80 90 95 98
Cumulative Frequency (%)
-------
lOOr
Suspended
Solids
I -
I I
lOOOOr
IOOO
2 5 IO 20 40 60 8O 90 95 98
40 \-
~ 301-
>>
o
§
Suspended
Solids
2 5 10 20 40 60 80 90 99 98
Cumulative Frequency (%)
Dissolved
Solids
IOOO -
IOO
I i i i i
2 5 10 20 4O 60 80 9O 95 98
550
450
350
2.50
ISO
Dissolved
Solids
2 5 10 20 40 60 SO 90 95 96
Cumulotive Frequency (%)
IOO —
2 5 10 20 40 60 80 90 95 98
450 -
350
2501-
150
2 5 10 2O 40 6O 8O 90 95 98
Cumulative Frequency (%)
I
I—i
o
I
tCant.-La.tMd)
-------
TABLE 21. TYPE OF DISTRIBUTION AND STATISTICAL CHARACTERISTICS OF THE ASH POND EFFLUENT AT PLANT E
Parameters
Aluminum
Arsenic
Calcium
Conductivity
Chromium
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Lead
PH
Selenium
Dissolved Silica
Sulfate
Suspended Solids
Turbidity
Zinc
Type of
Distribution
Lognormal
Normal
Lognormal
Lognormal
Normal
Lognormal
Lognormal
Lognormal
Lognormal
Log Normal
Normal
Normal
Normal
Normal
Normal
Lognormal
Lognormal
Lognormal
Mean
0.414
0.017
2.061
1.883
0.023
-1.605
2.548
-0.334
-0.134
-1.689
0.02
11.2
0.014
5.7
167
0.778
0.374
-1.552
Variance
0.0397
0.0017
0.0321
0.0133
0.0015
0.1463
0.0137
0.2960
0.2093
0.2042
0.0014
0.0403
0.00003
0.9573
3030
0.2298
0.1207
0.2306
Number of
Samples
33
34
33
33
34
34
33
34
34
34
34
33
33
34
32
34
25
34
(S2)(t2) for various confidence levels
99%
0.298
0.013
0.241
0.100
0.011
1.095
0.103
2.216
1.567
1.529
0.010
0.303
0.0002
7.167
22831
1.720
0.944
1.726
95%
0.165
0.007
0.133
0.055
0.006
0.606
0.057
1.227
0.868
0.846
0.006
0.167
0.0001
3.968
12610
0.953
0.514
0.956
80%
0.068
0.003
0.055
0.023
0.003
0.250
0.023
0.506
0.358 g
0 . 349 ?
0.002
0.069
0 . 00005
1.638
5192
0.393
0.210
0.395
aThe values given for lognormal distributions are for the logarithms of the concentrations while those for normal
distributions are for the untransformed concentrations.
br
JSee equation 3 for definition of (S2)(t2).
-------
-108-
Tables 22 gives the mean, appropriate ash pond effluent limitation
or proposed water quality criteria and the probability that these limita-
tions or criteria are exceeded for the effluent parameters assuming a
normal distribution. Table 23 gives the mean of the logarithms of the
concentrations, the logarithm of the geometric mean, appropriate ash pond
effluent limitation or proposed water quality criteria and the probability
that these limitations or criteria are exceeded for the effluent parameters
assuming a lognormal distribution. The mean of the logarithms of the
concentrations and the logarithm of the geometric mean different slightly
because the geometric mean was determined from the appropriate cumulative
frequency plot in Figure 14. All calculations for lognormal distributions
will be based on the values given in Table 23 for the logarithm of the
geometric mean.
For most of the elements, less than 5 percent of the samples were
below their minimum detectable limit. Lead was an exception, however,
with 80 percent of the samples being below the minimum detectable limit
of 0.01 mg/1.
The effluent limitations given in Tables 22 and 23 for pH and sus-
pended solids are those outlined for the steam-electric power generating
industry by EPA in 1974 (1) for the achievement, by 1977, of best
practical control technology currently available (BPCTCA). The pH is to
be maintained between 6 and 9 and the average daily suspended solids for
a 30-day period is to be below 30 mg/1 with a daily maximum less than
100 mg/1. Since limitations for the ash pond effluents at Plant E for
the remaining elements have not yet been promulgated, the criteria
specified in EPA's "Water Quality Criteria" (10) for dometic water
supply intakes are used. A list of the criteria are given in Appendix C.
This does not suggest that the ash pond effluent should meet these
criteria because the effluent is diluted between 20 and 80 times with
the condenser cooling water before final discharge. They are only given
for comparison purposes and as an aid in establishing the desired preci-
sion for the future monitoring program. The data in Tables 22 and 23
shows that greater than 98 percent of the time the pH is greater than 9,
whereas only 9 percent of the time the suspended solids are above 30
mg/1. Less than 2 percent of the samples had concentrations of arsenic,
chromium, lead, sulfate, copper, and zinc above the domestic water
supply criteria proposed by EPA. However, for selenium, iron, and
manganese, 70, 60, and 19 percent of the samples, respectively, were
above the domestic water supply criteria.
ESTIMATION OF THE MEAN AS A FUNCTION OF THE PRECISION
The number of samples, n, required to estimate the mean as a func-
tion of L was plotted for each parameter based on the data given in
Table 21 and equation 3. The results are shown in Figure 15. They were
constructed by dividing the values shown under the column labeled "(S2)(t2)"
in Table 21 by various values of (L)2 to yield various sample sizes, n.
The values for (S)2(t)2 given in Table 21 were obtained by multiplying
the various, S2, times the appropriate t value squared. The values used
for t are a function of the confidence level and number of data points
used to generate the variance. The t values necessary for calculating
the (S)2(t)2 values in Table 21 are given in Appendix D.
-------
-109-
TABLE 22. COMPARISON OF THE ASH POND EFFLUENT CHARACTERISTICS
FOLLOWING A NORMAL DISTRIBUTION AT PLANT E WITH ASH POND EFFLUENT
LIMITATIONS OR WATER QUALITY CRITERIA (BASED ON DATA COLLECTED
PRIOR TO JANUARY 1978)
Parameter
Mean of the
Concentrations
(mg/D
Standard or
Water Quality
Criteria
(mg/D
Frequency that
Standard is Exceeded
Arsenic
Chromium
Lead
pH
Selenium
Dissolved Silica
Sulfate
0.017
0.023
0.02
11.2
0.014
5.7
167
0.05a
0.05a
0.053
6 to 9b
0.013
c
250a
<2
<2
<2
>98
70
-
<2
Proposed EPA intake standards for domestic drinking water supplies (EPA 1976).
Effluent limitation specified in the NPDES permit. Units are standard units.
CNo criteria proposed for drinking water supplies.
-------
TABLE 23. COMPARISON OF THE ASH POND EFFLUENT CHARACTERISTIC FOLLOWING
A LOG NORMAL DISTRIBUTION AT PLANT E WITH ASH POND EFFLUENT LIMITATIONS
OR WATER QUALITY CRITERIA (BASED ON DATA COLLECTED PRIOR TO JANUARY 1978)
Mean of the
Logarithms of the
Logarithms of the
Standard or
Water Quality
Criteria
Probability that
Parameter
Aluminum
Calcium
Conductivity
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Suspended Solids
Turbidity
Zinc
Concentrations
0.414
2.061
1.883
-1.605
2.548
-0.334
-0.134
-1.689
0.778
0.374
-1.552
Geometric Mean
0.407
2.050
1.870
-1.778
2.531
-0.370
-0.148
-2.111
0.704
0.370
-1.926
(mg/1) Standard is Exceeded
b
c
c
1.0 <2
c
0.3d 60
c
0.05d 19
30e 9
c
5d <2
Values given are logarithms to the base 10 of the concentrations in mg/1.
Values given are the logarithms to the base 10 of the estimated geometric mean in mg/1.
No criteria proposed for drinking water supplies.
Proposed EPA intake standards for domestic drinking water supplies (EPA 1976).
g
Effluent limitations specified in the NPDES permit.
o
i
-------
0.02 0.04 0.06 0.08 0.10
0.02
0.04
0.06
0.08
O.IO
0.02 0.04 0.06 0.08 0.10 0.12 "0 0002 0004 0006 0008 0(
Precision (mg/l) Precision (mg/l)
Figure 15. Number of Samples Required for a Given Precision for the Plant E Ash Pond Effluent
-------
0.4 OS
20 4O 60 80 100 120 I4O
Precision (mg/l)
20
10 •
Precision (mg/l)
N3
I
Figure 15 (Continued)
-------
60
50
S 40-
Q
O
in
_ 30
20
10
Al
O.05 O.I
0.15
0.2 0.25 0.3
0.2
04 0.6
Precision (log mg/l)
0.8
10
02
04 06
Precision (log mg/l)
08
10
Figure 15 (Continued)
-------
Dissolved
Solids
99%
95%
80%
O.05 0.1
0.15 O2 0.25 O.3
0 OJ 02 0.3 0.4 0.5 0.6 O.7 OB 0.9 I.O
Suspended
Solids
"0 OJ 02 03 04 0.5 O.6 0.7 0.8 0.9
Precision (log mg/l)
1.0
0.05
0.1 0.15 0.2 0.25 03
Precision (log mg/l)
0.35
Figure 15 (Continued)
-------
0 0.1 02 03 04 0.5 0.6 0.7 0.8 O.9 1.0
•6
I
0 0.1 02 03 04 0.5 O.6 07 0.8 0.9 IO
Precision (log mo/I)
0.4 0.6 08
Precision (log mg/l)
10
Figure 15 (Continued)
-------
-116-
SELECTION OF THE PRECISION
The upper and lower limits (&i and #2) for the critical range of
the precision at the 99 percent confidence level for the ash pond
effluent characteristics at Plant E are given in Table 24. The upper
limit given is for the precision produced by one sample. However, for
all elements except for selenium and aluminum, the curve had become
asymptotic to the x-axis at a precision less than the upper limit given
in Table 24. Therefore, for some elements, the upper limit given in
Table 24 is not shown in Figure 15. They were calculated using equation
3. In some cases, the difference in the precision between one and two
samples may be significant. For example, the precision for As at one
sample is 0.114 rag/1, whereas at two samples it is 0.081 mg/1. However,
the upper limit of the critical range was given based on one sample
because that precision may be adequate for the monitoring program. The
lower limit given is for the precision produced by 52 samples or where
the curve becomes asymptotic to the y-axis, whichever gives the larger
value of L. Determining the lower limit in this manner, assumes resources
are not available for the collection or analysis of more than 52 samples
in any one sample period. The curve in Figure 15 had not become asymptotic
to the y-axis at 52 samples for any of the ash pond effluent charac-
teristics, and therefore the lower limits given in Table 24 were
determined based on the assumed availability of resources. If the value
of the precision required for the monitoring program is greater than the
upper limit, then only one sample per period needs to be collected.
However, if the precision value is less than the lower limit, then 52
samples per period would be collected. If for some reason the precision
for 52 samples is not adequate for an element of a monitoring program,
then a decision would have to be made as to whether or not to increase
the level of resources allocated to the monitoring program. If the
required precision is between the limits, then the data in Figure 15
would be consulted to determine the number of samples. Therefore, the
information in Table 24 gives valuable insight into the importance of
the required precision on the design of an ash pond effluent monitoring
program for Plant E.
Suspended solids and pH are the only parameters included in this
study for which ash pond effluent limitations have been set for Plant E.
Use of the design procedure discussed in Section 3 to show compliance
can, therefore, only be applied to suspended solids. The procedure
requires that the effluent be in compliance, and Figure 14 shows that
greater than 98 percent of the time the pH is greater than the effluent
limitation of 9. However, suspended solids is only above the effluent
limitation 9 percent of the time. Therefore, the precision required
for suspended solids can be determined by subtracting the logarithm of
the geometric mean of the concentrations given in Table 19 for suspended
solids (0.704, see Table 23) from the logarithm of the effluent limita-
tion (log 30 = 1.477). This yields a value of 0.773 for the required
precision or deviation from the true mean.
Defining the precision which should be used to design the moni-
toring program at Plant E is difficult where effluent limitations have
not yet been promulgated. One method is to assume some precision based
on a given percentage of the sample mean. For comparison purposes,
-------
-117-
TABLE 24. UPPER AND LOWER LIMITS FOR THE CRITICAL RANGE OF
THE PRECISION FOR THE EFFLUENT CHARACTERISTICS OF PLANT E
Element
Elements
Arsenic
Chromium
Lead
pH
Selenium
Dissolved Silica
Sulfate
Elements
Aluminum
Calcium
Conductivity
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Suspended Solids
Turbidity
Zinc
Lower Limit Upper Limit
of L of L
(mg/1) (mg/L)
Following A Normal Distribution
0.016
0.015
0.014
0.076
0.002
0.37
21
Following A Lognormal Distribution
0.076
0.068
0.043
0.145
0.045
0.206
0.174
0.171
0.182
0.135
0.182
0.114
0.105
0.100
0.550
0.014
2.68
151
0.546
0.490
0.316
1.046
0.321
1.489
1.25
1.24
1.31
0.97
1.31
-------
-118-
sampling frequencies based on estimating the yearly mean within 10 and
20 percent of the true mean at the 99 and 80 percent significance levels
will be discussed in the next subsection. Another method for estab-
lishing the precision is to allow for a certain level of pollutant
loading to the receiving stream. This cannot be done without some
estimate of the receiving stream water quality before addition of
pollutants. Therefore, the water quality characteristics shown in Table
25 will be assumed for the stream receiving the ash pond effluent from
Plant E. These values are based on the 1976 data for the intake water
to Plant E. They differ somewhat from the data given in Table 5 for the
intake water during 1974 and 1975. The major reason for this may be in
the difference in the number of samples used in generating the two data
sets. Note that the significance level and precision for the data in
Table 25 are not specified. For design purposes, these values will be
assumed to be absolute. In addition, some dilution factor and maximum
allowable average concentration in the receiving stream must be specified.
The dilution factor assumed for Plant E's ash pond effluent is approxi-
mately 8.6 x 103. It is based on a seven day miminum flow of 7880 cubic
feet per second (cfs) for the receiving stream and a maximum ash pond
flow of 67 cfs (~30,000 gpm). The value of 67 cfs was obtained by
rounding off the highest reported value for the flow in Table 19. Table
26 gives the allowable ash pond input to the receiving stream and precision
required by the monitoring program assuming the maximum allowable average
concentration in the receiving stream is based on maintaining the con-
centration in the receiving stream equal to or below the EPA proposed
water quality criteria for domestic water supply intakes (see Appendix C
for a summary of these criteria). Table 27 gives the same information
for a monitoring program assuming the maximum allowable average con-
centration in the receiving stream is below or equal to the maximum
value given in Table 25. Remember the value given in Table 25 represents
the maximum value reported in 1976 for the intake water to the plant. A
precision was not given for Se in Table 27 because the reported Se con-
centration in the effluent was above the maximum average allowable con-
centration calculated by this method, therefore, the procedure developed
in Section 4 for determining the number of samples to show compliance
with a selected water quality criteria could not be used. An example
calculation for the element As and an input based on the EPA water
quality criteria of the assumed allowable input to the stream and the
associated precision is shown in Appendix E. The sampling frequencies
associated with these precisions will be discussed in the following
subsection.
ESTIMATED SAMPLING FREQUENCIES
The precision required to determine the minimum number of samples
needed to show that the ash pond effluent for Plant E is in compliance
with the effluent limitation for suspended solids, was calculated to be
0.773 in the previous section. This value falls within the critical
range of the deviation for suspended solids indicating the number of
samples can be determined from Figure 15. For the 99 percent confidence
level, this means 3 samples per sample period are required. Since the
effluent limitation specifies that the concentration must not exceed an
average of 30 mg/1 for 30 consecutive days, the number of samples derived
-------
-119-
TABLE 25. ASSUMED WATER QUALITY CHARACTERISTICS FOR
THE RECEIVING STREAM AT PLANT E
Element
Aluminum (rag/1)
Arsenic (mg/1)
Calcium (mg/1)
Conductivity (|jmhos)
Chromium (mg/1)
Copper (mg/1)
Dissolved Solids
Iron (mg/1)
Magnesium (mg/1)
Manganese (mg/1)
Lead (mg/1)
pH (Standard units)
Selenium (mg/1)
Dissolved Silica (mg/1)
Sulfate (mg/1)
Suspended Solids (mg/1)
Turbidity (JTU)
Zinc (mg/1)
Average
Concentration
1.7
0.004
19
158
0.008
0.018
100
0.4
3.4
.046
0.012
6.9
0.002
4.1
22
12
7
0.015
Maximum
Concentration
2.1
0.005
26
180
0.016
0.020
120
0.54
4.7
0.1
0.016
7.3
0.002
5.0
41
18
14
0.030
-------
-120-
TABLE 26. REQUIRED PRECISION FOR THE MONITORING PROGRAM OF PLANT E
ASSUMING AN AVERAGE ALLOWABLE CONCENTRATION IN THE RECEIVING STREAM
EQUAL TO THE EPA PROPOSED WATER QUALITY CRITERIA
Maximum Average Allowable Required
Element Concentration in the Effluent Precision
Arsenic (mg/1)
Chromium (mg/1)
Copper (mg/1)
Iron
Manganese (mg/1)
Lead (mg/1)
Selenium (mg/1)
Dissolved Silica (mg/1)
Sulfate (mg/1)
Zinc (mg/1)
5.4
4.9
115.5
a
0.516
4.48
0.943
26838
586
5.383
4.877
3.841
-
1.824
4.460
0.929
26671
4.694
alntake water already exceeds the criteria.
-------
-121-
TABLE 27. REQUIRED PRECISION FOR THE MONITORING PROGRAM OF PLANT E
ASSUMING AN AVERAGE ALLOWABLE CONCENTRATION IN THE RECEIVING STREAM
EQUAL TO THE MAXIMUM VALUE REPORTED FOR THE INTAKE WATER
Maximum Average Allowable Required
Element Concentration in the Effluent Precision
Aluminum (mg/1)
Arsenic (mg/1)
Calcium (mg/1)
Conductivity (pmhos)
Chromium (mg/1)
Copper (mg/1)
Dissolved Solids (mg/1)
Iron (mg/1)
Magnesium (mg/1)
Manganese (mg/1)
Lead (mg/1)
Selenium (mg/1)
Dissolved Silica (mg/1)
Sulfate (mg/1)
Suspended Solids (mg/1)
Turbidity (JTU)
Zinc (mg/1)
48
0.122
842
2745
0.95
0.253
2452
16.9
156
6.40
0.48
a
110
2257
718
830
1.78
1.274
0.105
0.875
1.569
0.927
i.179
0.859
1.598
2.341
2.917
0.46
-
104.3
2090
2.152
2.549
2.176
aThe reported ash pond effluent concentration exceeds the maximum average
allowable concentration calculated by this method; therefore, the procedure
developed in Section 4 for determining the number of samples to show com-
pliance with a selected water quality criteria cannot be used.
-------
-122-
from Figure 15 represents a sampling frequency of 3 samples per 30 days
or 36 samples per year, assuming 30 days per month. This represents a
sampling frequency of one sample every 10 days. This assumes, of course
that the variance obtained for the data over the period of the extensive
sampling program and used to construct Figure 15, would be the same had
the period of the survey been any one month and the same number of
samples been collected. As discussed earlier, this assumption is valid
when the data are randomly distributed. Corresponding sampling fre-
quencies for the 95 and 80 percent confidence levels would be 2 and 1
times per month, respectively.
The sampling frequency of one sample per 10 days (3 per month) for
the 99 percent significance level is slightly more often than the sampling
frequency of two per month currently being required by the NPDES permit.
The current requirement results in 29 samples per year whereas 36 are
required according to the study. These additional 12 samples result in
a decrease at the 99 percent confidence level of 22 percent (from 0.219
for 36 samples to 0.268 for 24 samples) in the deviation of the logarithm
of the estimated yearly geometric mean from the logarithm of the true
yearly geometric mean and also a decrease of 22 percent in the deviation
of the daily mean for a 30-day period. The effect of establishing an
averaging period within an effluent limitation specification (i.e., 30 mg/1
in any 30-day period) is readily apparent from the above discussion. Had
the average period been shortened to 15 days or extended to 60 days, the
number of samples required per year would have been 72 and 18, respectively.
Therefore, care should be exercised in establishing these averaging periods
for effluent limitations.
The NPDES sampling frequency of 2 samples per month provides for a
95 percent confidence level and even 1 sample per month would provide
for an 80 percent confidence level. Considering the relative significance
of suspended solids to the environment, the high dilution factor by the
relatively high minimum flow in the receiving stream, and the insignifi-
cance of the potential incremental increase in suspended solids above
the effluent limitation, the 80 percent confidence level appears to be
sufficient to ensure adverse environmental impacts will not occur. By
collecting one sample per month, the 30-day average suspended solids
concentration can be shown to be below 22 mg/1 with 80-percent confidence.
The above estimates are appropriate if the average of 30 mg/1 is
interpreted to mean the geometric mean of 30 mg/1 when dealing with log-
normal data. The geometric mean is always smaller than the arthimetic
mean, thus, in effect, creating a slightly higher standard when trans-
forming the standard to a logarithmic value and comparing it with the
geometric mean. However, in this case the error introduced due to this
assumption is insignificant because the mean for suspended solids is
well below the effluent limitation.
Table 28 shows the number of samples required per year to estimate
the yearly mean (geometric mean for lognormal data) within 20 percent of
the true mean for the 99, 95, and 80 percent confidence levels. A sub-
stantial sampling effort (greater than 52 samples per year) would be
required to estimate the yearly mean within 20 percent for As, Cr, Fe,
Mg, Pb, and turbidity at all three confidence levels, whereas a minimal
-------
-123-
TABLE 28. NUMBER OF SAMPLES REQUIRED TO ESTIMATE THE YEARLY
MEAN WITHIN 20% OF THE TRUE YEARLY MEAN OF PLANT E
Element
Aluminum
Arsenic
Calcium
Conductivity
Chromium
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Lead
PH
Selenium
Dissolved Silica
Sulfate
Suspended Solids
Turbidity
Zinc
Number of
99% SL
29
719
1
1
332
6
1
259
1144
6
400
1
16
4
13
56
111
8
Samples Required Per Year
95% SL
16
387
1
1
182
4
1
144
634
4
240
1
9
2
8
31
61
5
80% SL
7
166
1
1
91
2
1
60
262
2
80
1
5
1
3
13
25
2
SL = Significance level.
-------
-124-
effort (only 1 sample per year) would be required for Ca, conductivity,
dissolved solids, and pH. The remaining parameters would require between
2 and 56 samples per year.
Selecting the sampling frequency based on a precision which estimates
the yearly mean within a given percentage of the population mean has two
major weaknesses. First, the method does not take into account the
significance of the concentration in the waste stream, and second, it
tends to reduce the precision value (increase the number of samples) as
the concentration in the waste stream decreases. As a result of these
weaknesses, a monitoring program based on this method above would result
in 1,973 turbidity samples and only 26 selenium samples per year.
Selenium in the effluent probably deserves more attention than turbidity
because of its toxicity relative to the concentrations found in the
effluent.
To overcome these weaknesses and add additional meaning to the
monitoring program, Table 29 gives the estimated sampling frequencies
for the precisions given in Tables 26 and 27. Table 29 indicates that a
sampling frequency of once per year for all parameters except for selenium
and arsenic, ensures within 99 percent confidence that the yearly average
concentration in the receiving stream will not be increased (by the ash
pond effluent) above the maximum value reported in the intake water for
1976. Arsenic would require two samples per year for the same assurance.
Likewise, only one sample per year for all parameters ensures within
99 percent confidence that the receiving stream's yearly average concen-
tration will not be increased (by the ash pond effluent) above the EPA
proposed water quality criteria. Therefore, establishing monitoring
frequencies based on maintaining the average concentration in the
receiving stream equal to or below the maximum value reported in the
intake water in 1976 automatically ensures monitoring frequencies as
great or greater than those based on maintaining the average con-
centration in the receiving stream equal to or below the EPA proposed
water quality criteria for domestic water supply intakes.
The sampling frequencies listed in Tables 28 and 29 differ con-
siderably. The frequencies based on the assumed allowable level of
increase in the receiving steam are substantially lower than those
required to estimate the mean within 20 percent. The sampling frequency
used in the final monitoring program should, therefore, be a compromise
between the frequencies given in Tables 28 and 29. As an aid in estimat-
ing the point of compromise, the deviation of the yearly sample mean
from the true mean for the 99 percent confidence level is given for the
following frequencies: yearly, quarterly, bimonthly (once every two
months), monthly, biweekly (once every two weeks), and weekly. These
frequencies were selected because they are the most widely used frequen-
cies. The data given in Table 30 indicates that the deviation of the
sample mean from the true mean varies from parameter to parameter at
each sampling frequency. This indicates the fallacy in establishing the
same monitoring frequency for every parameter. By doing so, some
parameters are estimated more accurately than others, possibly making
comparisons between parameters misleading.
-------
-125-
TABLE 29. ESTIMATE SAMPLING FREQUENCIES FOR THE MONITORING PROGRAM
AT PLANT E ASSUMING ALLOWABLE AVERAGE CONCENTRATIONS IN THE RECEIVING
STREAM EQUAL TO THE EPA WATER QUALITY CRITERIA AND MAXIMUM VALUE
REPORTED FOR THE INTAKE WATER
Number of Samples per Year'
Element
Precision based on
Water Quality Criteria
Precision based on
Maximum value reported
for the intake water
Aluminum
Arsenic
Calcium
Conductivity
Chromium
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Lead
Selenium
Dissolved Silica
Sulfate
Suspended Solids
Turbidity
Zinc
d
1
d
d
1
1
d
e
d
1
1
1
d
1
d
d
d
1
2
1
1
1
1
1
1
f
1
1
g
1
1
1
1
1
a. Values are for the 99% significance level.
b. See Table 25 for the precision values.
c. See Table 26 for the precision values.
d. Criteria not proposed for drinking water intake supplies.
e. Intake water exceeds criteria.
f. Intake concentration exceeds effluent concentration.
g. The average ash pond effluent concentration exceeds the maximum concentration
reported for the intake water.
-------
-126-
TABLE 30. DEVIATION OF THE YEARLY SAMPLE MEAN FROM THE TRUE MEAN FOR THE
99% CONFIDENCE LEVEL AT VARIOUS SAMPLING FREQUENCIES
Deviation from
Parameter
Aluminum
Arsenic
Calcium
Conductivity
Chromium
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Lead
PH
Selenium
Dissolved Silica
Sulfate
Suspended Solids
Turbidity
Zinc
Yearly
57(0.546)
87(0.114)
19(0.490)
14(0.316)
82(0.105)
37(1.046)
11(0.320)
80(1.489)
89(1.252)
37(1.237)
83(0.100)
4.7(0.550)
50(0.014)
32(2.677)
48(151)
65(1.311)
72(0.972)
41(1.314)
Quarterly
40(0.273)
77(0.057)
11(0.245)
8(0.158)
69(0.052)
23(0.523)
5.9(0.160)
67(0.744)
81(0.626)
23(0.618)
71(0.050)
2.4(0.275)
33(0.007)
19(1.339)
31(76)
48(0.656)
57(0.486)
25(0.657)
Bimonthly
35(0.223)
73(0.047)
9(0.200)
6(0.129)
65(0.043)
19(0.427)
4.9(0.31)
62(0.608)
78(0.511)
19(0.505)
67(0.041)
2.0(0.225)
30(0.006)
16(1.093)
27(62)
43(0.535)
52(0.397)
22(0.536)
the True Mean3
Monthly
28(0.156)
66(0.033)
6(0.142)
5(0.091)
57(0.030)
15(0.302)
3.5(0.093)
54(0.430)
71(0.361)
14(0.357)
59(0.029)
1.4(0.159)
22(0.004)
12(0.773)
21(44)
35(0.379)
43(0.280)
16(0.379)
Biweekly
21(0.107)
56(0.022)
4(0.096)
3(0.062)
48(0.021)
10(0.205)
2.4(0.063)
44(0.292)
62(0.245)
10(0.243)
50(0.020)
1.0(0.108)
18(0.003)
8(0.525)
15(30)
27(0.257)
34(0.191)
12(0.258)
Weekly
16(0.076)
48(0.016)
3(0.068)
2(0.044)
39(0.015)
8(0.145)
1-7(0.045)
36(0.206)
54(0.174)
8(0.171)
41(0.014)
0.6(0.076)
13(0.002)
6(0.371)
11(21)
21(0.182)
27(0.135)
9(0.182)
a. Values are given as percent of deviation from the true mean.
indicate the deviation in mg/1 or log mg/1.
Numbers in parenthesis
-------
-127-
EXAMPLE SAMPLING PROGRAM FOR PLANT E
An example sampling program for Plant E to meet NPDES require-
ments is shown in Table 31. It is based on the previous discussion and
the following criteria:
1. The element must be required by the NPDES permit.
2. The 99 percent confidence level was assumed.
3. The precision used to estimate the sampling frequency was based
on maintaining the average concentration in the receiving stream
below or equal to the maximum concentration reported for the
receiving in 1976 under the 7-day 10-year minimum flow. This
justification for trace metals was assumed because biological
studies performed for P.L. 92-500, Section 316 demonstrations,
indicated no adverse biological effects of the discharges from
Plant E.
4. If the average concentration in the effluent exceeded the maxi-
mum value reported for the intake water and the EPA proposed
water quality criteria, then the frequency was established based
on estimating the average within at least 33 percent of the true
mean.
5. For those elements for which an effluent limitation has been
set, the recommended frequency ensures an average which shows
compliance if the effluent is in compliance.
6.
v. \stttp JU-l. t*J,I.V-<^ -L i. h*ll.(^ C J_ J- A U.d.11- -LO J-J.1 LUIII^S X. iOJ-i^ti •
Unless specified, the sample(s) can be collected any time
during the averaging period.
7. If the data shows that the effluent concentration is below the
detection limit, the element will not be included in the moni-
toring program even if required by the current NPDES permit.
The remaining discussion gives the justification for this program by
element.
Aluminum
Aluminum is not recommended as part of the monitoring program
because it is not required by the NPDES permit.
Arsenic
Two samples per year are recommended. Monitoring of arsenic is
required by the NPDES permit. Two samples show with 99 percent con-
fidence that the ash pond effluent does not increase the yearly average
receiving stream concentration above 0.005 mg/1. It also allows
estimation of the yearly average concentration in the effluent within
83 percent.
-------
-128-
TABLE 31. EXAMPLE SAMPLING PROGRAM FOR PLANT E
•a
Sampling Frequency Precision
Element (No. per year) (% of true mean)
Aluminum
Arsenic
Cadmium
Calcium
Chromium
Conductivity
Copper
Dissolved Silica
Dissolved Solids
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
pH
Selenium
Sulfate
Suspended Solids
Turbidity
Zinc
0
2
0
0
1
0
1
0
0
1
1
0
1
0
0
1
4
OK
36b
0
1
.
83
-
-
82
-
37
-
-
80
83
-
37
-
-
4.7
33
_
24
-
41
a. At the 99% confidence level.
b. The frequency should be 1 sample every 10 days.
-------
-129-
Cadmimn
Cadmium is not recommended as part of the monitoring program
because the data during 1974 and 1975 indicated that the concentration
was below the detection limit.
Calcium
Calcium is not recommended as part of the monitoring program
because it is not required by the NPDES permit.
Chromium
One sample per year is recommended. Monitoring of chromium is
required by the NPDES permit. This one sample shows within 99 percent
confidence that the ash pond effluent does not increase the yearly
average receiving stream concentration above 0.009 mg/1. It also allows
estimation of the yearly average concentration in the effluent within
82 percent of the true mean.
Conductivity
Conductivity is not recommended as part of the monitoring program
because it does not provide any useful information.
Copper
One sample per year is recommended. Copper is required by the
NPDES permit and one sample shows that the ash pond effluent does not
increase the yearly average receiving stream concentration above 0.02
mg/1. It also allows the yearly mean in the effluent to be estimated
within 37 percent.
Dissolved Silica
Dissolved silica is not recommended as part of the monitoring
program because it is not required by the NPDES permit.
Dissolved Solids
Dissolved solids is not recommended as part of the monitoring
program because it is not required by the NPDES permit.
Iron
One sample per year is recommended. Iron is required by the NPDES
permit. The concentration of iron in the intake water exceeds the EPA
proposed water quality criteria for drinking water intake supplies,
however, one sample shows that the effluent does not increase the yearly
average intake concentration above 0.54 mg/1. It also allows estimation
of the yearly average concentration in the effluent within 80 percent.
-------
-130-
Lead
One sample per year is recommended. Lead is required by the NPDES
permit. This one sample shows with 99 percent confidence that the ash
pond effluent does not increase the yearly average receiving stream
concentration above 0.013 mg/1. It also allows estimation of the yearly
average concentration in the effluent within 83 percent of the true
mean.
Magnesium
Magnesium is not recommended as part of the monitoring program
because it is not required by the NPDES permit and the concentration in
the effluent was consistently less than that in the intake water.
Manganese
One sample per year is recommended. Manganese is required by the
NPDES permit and one sample shows that the ash pond effluent does not
increase the yearly average receiving stream concentration above 0.05
It also allows the yearly mean in the effluent to be estimated within
37 percent.
Mercury
Mercury is not recommended as part of the monitoring program
because the data during 1974 and 1975 indicated that the concentration
was below the detection limit.
Nickel
Nickel is not recommended as part of the monitoring program because
the data during 1974 and 1975 indicated that the concentration was below
the detection limit.
pH
One sample per year is recommended. The pH of the effluent exceeds
the limitation established in the NPDES permit greater than 98 percent
of the time and one sample estimates the yearly average within 4.7
percent of the true mean at the 99 percent confidence level.
Selenium
Four samples per year are recommended. Selenium is required by the
NPDES permit. The concentration in the intake water is consistently at
or below the minimum detectable limit of 0.002 mg/1, while the ash pond
effluent concentration is consistently above this concentration (the
average during the extensive survey was 0.014 mg/1). Therefore, four
samples were recommended because they estimate the yearly average within
33 percent.
Sulfate
Sulfate is not recommended as part of the monitoring program
because it is not required by the NPDES permit.
-------
-131-
Suspended Solids
Thirty-six samples per year at 10-day intervals are recommended.
This frequency shows with 99 percent confidence that the effluent is in
compliance with the effluent limitation of 30 mg/1 specified in the
NPDES permit.
Turbidity
Turbidity is not recommended as part of the monitoring program at
this time. However, under certain influent water quality conditions a
relationship could possibly be developed between turbidity and suspended
solids. This could allow installation of an automatic turbidity meter
with an alarm system set to activate an automatic sampler when the
effluent suspended solids exceed 30 mg/1. Such an arrangement may
reduce the number of samples required for suspended solids and reduce
sampling and analysis costs.
Zinc
One sample per year is recommended. Zinc is required by the NPDES
permit. This sample shows that the ash pond effluent does not increase
the yearly average receiving stream concentration above 0.017 mg/1. It
also allows the yearly mean in the effluent to be estimated within 41
percent.
SUMMARY
The example sampling program given in Table 31 requires a total of
48 analyses per year for 9 different elements whereas the NPDES permit
requires a total of 56 analyses per year for 12 different elements.
Under the recommended program, estimates of the yearly average were
obtained for the following elements: As, Cr, Cu, Fe, Pb, Mn, Se,
suspended solids, and Zn. The example program excludes sampling for Cd,
Ni, and Hg, which are required by the NPDES permit, because past data
showed them to be below the minimum detectable amount. The above totals
exclude pH, flow, and oil and grease.
At the time of this writing the ash pond effluent at Plant E was
considered to be in compliance with existing effluent limitations as
defined in the NPDES permit for that plant. Special provisions or plant
modifications specific for Plant E may be required in the future to
ensure continued compliance. One example of a recent provision in the
NPDES permit pertains to the pH limitation. Although the effluent
limitation for pH cited in this report was not met, the pH of the ash
pond effluent is not considered out of compliance. Provisions have been
made such that the effluent is allowed to mix with another waste stream
(condenser cooling water) before meeting the effluent limitations. This
provision was started after the extensive sampling program was begun and
therefore not included in this report.
-------
-132-
SECTION 6
ASH POND MONITORING PROGRAM FOR PLANT J
The following section demonstrates how the procedure outlined in
Section 4 was used to design a monitoring program for the ash pond
effluent at TVA's Plant J.
DESCRIPTION OF PLANT J
Plant J consists of nine pulverized coal-fired units with a
combined full load capacity of 1.7 million kilowatts. Each of units 1-4
has a maximum generator nameplate rating of 175 megawatts and each of
units 5-9 has a rating of 200 megawatts. At full capacity, the plant
consumes about 16,200 tons of coal per day. The majority of the coal
comes from eastern Kentucky and eastern Tennessee and has an average
sulfur content of 2.1 percent and an average ash content of 19.1 percent.
Fly ash control is accomplished by the use of mechanical collectors
and electrostatic precipitators installed in series on each unit. The
overall collection efficiency of the collection system is estimated at
98 percent, 70 percent efficiency for mechanical and 95 percent effi-
ciency for the electrostatic precipitators.
Assuming operation at full load capacity, approximately 3050 tons
of ash per day would be produced by Plant E. This ash is sluiced to a
275-acre ash pond with a storage capacity of about 3.25 million cubic
yards. The effluent is discharged to the condenser cooling water intake.
In addition to the ash, the ash pond also receives chemical clean-
ing wastes, coal pile drainage, and treated sanitary sewage. The coal
pile drainage and sanitary sewage flows represent approximately 0.6 and
0.04 percent of the total flow from the ash pond. The chemical cleaning
wastes are discharged intermittently (3 times per year) and during their
discharge they represent approximately 1.0 percent of the total flow
from the ash pond. These flows are assumed insignificant in determining
the overall ash pond effluent characteristics.
MECHANICS OF THE ASH POND SYSTEM AT PLANT J
A summary of the ash pond effluent characteristics for Plant J from
1970 to 1975 were given in Section 3. There were insufficient data on
the operating conditions of Plant J during this period to determine the
relationship between the ash pond effluent and plant operation. However,
there were some significant correlations at Plant J indicated in Section 3,
Of these, the most interesting was the one between the alkalinity of the
intake water and the ash pond effluent. This relationship is believed
to exist because of the large changes that occur in the alkalinity of
the intake water. The intake water used for sluicing can consist of
either Emory River water or Clinch River water or a combination of both
since the Clinch River has been known to progress as much as 14 miles up
-------
-133-
the Emory under different hydrologic conditions. According to a 1961-1962
survey, the alkalinity ranged from 64 to 108 mg/1 in the Clinch River
and from 3 to 85 mg/1 in the Emory River. This change in intake water
quality, therefore, probably accounts for the seasonal cycle indicated
in Table 4 for pH, alkalinity, conductivity, dissolved solids, and
hardness in the ash pond effluent at Plant J.
The mixing characteristics of the ash pond contents for Plant J
were not investigated as fully as those for Plant E. Cenospheres on the
surface of the ash pond at Plant J were observed to readily move about
the pond depending on wind conditions, and for this reason the mixing
characteristics were assumed to be similar to those at Plant E. The
detention time within the ash pond at Plant J was not determined by a
dye study. However, samples of the effluent during the sluicing of
chemical cleaning wastes showed the detention time of the ash pond to be
about 2 to 4 hours. However, during more windy conditions, which pro-
vide pond mixing and destratification, the detention time is probably
closer to 60 hours. At first this seems to be contradictory to the
detention times given for the ash pond at Plant E since the pond at
Plant J is larger. These shorter times are probably due to more
short-circuiting and the higher flow at Plant J than Plant E. Because
of this shorter detention time, the variation of the effluent charac-
teristics are probably more dependent on the variation in plant operating
conditions.
SUMMARY OF THE ASH POND EFFLUENT CHARACTERISTICS AT PLANT J
The weekly effluent data from 1970 to 1975 for Plant J showed that
there was a yearly cycle for pH, total alkalinity, conductivity, dissolved
solids, total solids, and hardness. The data did not indicate a yearly
cycle for flow, phenolphthalin alkalinity, turbidity or suspended solids.
The data given in Figure 7 for Plant J showed that the variation of
daily composite samples for several elements, including dissolved and
suspended solids, trace metals, and pH, over a four day period was no
greater than that exhibited by the weekly data. Therefore, a weekly
cycle is assumed nonexistent.
As concluded at the end of Section 3, except for pH and suspended
solids, there is insufficient data on those parameters required by the
NPDES permit for Plant J to adequately estimate the true yearly mean.
Therefore, a more intensive sampling program of the ash pond effluent at
Plant J was conducted from January 1977 through September 1977 to better
estimate the effluent characteristics. Grab samples were collected by
power plant personnel on a varying work day of each week the same as at
Plant E. These samples were then shipped to the Laboratory Branch in
Chattanooga, Tennessee, for analysis. They were analyzed for the
following parameters which are required by the NPDES permit for Plant J:
pH, suspended solids, flow, total arsenic, chromium, copper, iron, lead,
manganese, nickel, selenium, and zinc. The NPDES permit also calls for
cadmium and mercury to be monitored, however, these elements were not
included in this study because previous data (see Table 5) indicated the
concentrations were near the minimum detectable amount. In addition,
the samples were analyzed for aluminum, calcium, magnesium, sulfate, and
-------
-134-
dissolved solids. These elements were included because previous data
indicated their presence. The samples were collected in the ash pond
discharge prior to mixing with any other waste stream as required by the
NPDES permit. The sampling period was such that samples were collected
during all phases of the yearly cycle.
The results of this extensive sampling program are given in Table
32. A summary at the bottom of the table gives the minimum, mean, and
maximum values for each element. Linear correlation coefficients were
developed between each element. A significant correlation at the 95
percent significance level is represented by an R value greater than
0.325 in Table 33 (2). Chromium is not shown because most samples were
below the mimimum detection limit. Every element except chromium,
nickel, suspsended solids, and sulfate were significantly correlated
with pH. This is not surprising, for the elements Al, Cu, Fe, Pb, Mn,
and Zn are normally more soluble at lower pH and this trend is indicated
by the negative R values. The positive R values for Ca and Mg are also
not surprising since alkalinity increases with increasing pH. The
positive R values for As and Se indicate that their concentration
decreases with decreasing pH in the pH range 8.5 to 4.5. The pH was
significantly correlated with flow and therefore, some of the elements
correlated with pH were significantly correlated with flow. They are
Al, Ca, Fe, Mg, Se, As, and dissolved solids. Aluminum and iron were
the only elements significantly correlated with suspended solids.
Several of the trace metals were correlated with each other. These
correlations indicate that the heavy metals in the ash pond effluent are
interrelated with one another. However, development of these relationships
are beyond the scope of this study.
The only R values which indicated a relationship between parameters
which may be beneficial to a monitoring program were those with pH.
Since the pH varies considerably (4.5 to 8.5) on a seasonal basis and
some of the trace elements are dependent on the pH, pH may be useful as
an indicator of certain trace elements when the concentration of these
trace elements are at that concentration which has the most potential
for causing environmental harm.
The summary of the ash effluent characteristics for Plant J given
in Table 32 will now be used to complete steps 3 through 8 of the design
procedure summarized in Section 4.
VARIATION OF THE ASH POND EFFLUENT CHARACTERISTICS AT PLANT J WITH TIME
The variation with time of the ash pond effluent characteristics
given in Table 33 for Plant J is shown in Figure 16. The majority of
the concentrations of chromium, lead, and nickel were below the minimum
detectable amounts, and therefore the occurrence of "a cycle could not be
determined. A yearly cycle was indicated for pH as expected. In addi-
tion, a yearly cycle was indicated for Al, As, Ca, dissolved solids, Fe,
Mg, Se, and sulfate. These cycles probably exist as a result of the
cycle in pH. The data in Figure 16 also indicates a yearly cycle for
flow, however, this is not the case. At about the time the flow is
shown to increase in Figure 16, additional electrostatic precipitators
-------
TABLE 32. ASH POND EFFLUENT CHARACTERISTICS AT PLANT J
Date
1-19
1-27
2-11
2-17
2-23
J-3
J-9
3-16
3-23
3-31
4-15
4-19
4-28
5-4
5-9
5-17
5-25
6-2
6-8
6-16
6-21
6-30
7-6
7-20
7-12
7-29
8-1
8-16
8-10
8-31
9-8
9-15
9-19
9-28
Min.
Avg.
Max.
pH
4.9
6.7
7.6
7.7
7.3
5.9
4.5
4.5
4.7
7.5
5.6
NA
6.1
6.1
7.S
7.5
NA
7.8
8.1
7.8
8.1
7.6
8.5
7.9
8.2
7.4
7.4
7.3
7.4
8.2
7.6
7.4
7.4
7.0
4.5
7.0
8.5
Flow
(MGD)
20.2
21
20
21
21
19
21
19
35
37
S'A
38
39.6
41
38.8
MA
NA
39
35
49
40
39
39
40
39
39
37
36
40
32
40
39
40
19.0
33.8
49
Aluminum
(mg/1)
1.8
0.5
0.5
0.2
0.8
0.9
1.3
1.7
1. 1
.5
1.1
0.2
0.9
0.9
0.5
0.61
0.51
0.34
0.74
0.29
0.29
0.88
0.70
0.50
<0.2
0.4
0.4
0.41
0.3
0.8
0.5
0.6
0.2
0.9
0.2
0.66
1.8
Calcium Chromium
(mg/1) (mg/1)
26
32
39
31
35
19
23
26
22
19
19
26
30
26
29
36
44
28
43
42
40
48
46
48
50
48
44
50
50
51
53
54
50
70
19
38
70
<0.005
•C0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0.005
<0 . 005
<0.005
<0.005
<0.005
<0.005
Copper
(mg/1)
0.11
0.02
<0.01
0.02
0.04
0.01
0.35
0.11
0.04
0.03
0.06
0.01
0.04
0.25
0.06
<0.01
<0.01
0.06
<0.01
0.20
o.oi
0.03
0.04
16.0
0.01
0.07
0.02
<0.01
0.04
0.02
0.02
0.02
0.02
0.05
<0.01
0.06
0.35
Iron
(mg/1)
3.9
4.1
0.48
0.46
0.66
1.9
2.5
2.2
0.57
0.48
0.37
0. 12
0.67
0.86
0.13
0.25
0.3
1.0
0.30
0.42
0.37
0.34
0.38
0.49
0.36
0.47
0.50
0.33
0.44
0.24
0.55
0.38
0.23
0.56
0. 12
0.80
4.10
Lead Magnesium
(tig/ 1) (mg/1)
<0. 01
<0.01
<0.01
<0.01
<0.01
<0.01
'.0.01
0.016
0.014
<0.01
<0.01
<0.01
0.15
<0.0l
'0.01
<0.01
<0.01
<0.01
(0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
^0.01
<0.01
<0.01
^0.01
<0.01
<0.01
<0.01
<0.010
0.010
0.016
4.3
6.4
8.0
6.7
7. 1
3.5
3.0
3.9
3.3
3.8
4.0
7. !
4.9
5.7
7.2
8.3
9.4
7.8
8. 1
8.2
8.4
7.8
8.4
8.0
7.4
8.3
8. 7
7.7
8.0
10
9.6
9.6
8.4
8.0
3.0
7.0
10.0
Manganese Zinc
(mg/1) (mg/1)
0.28
0. 18
0.08
0. 10
0. 14
0.21
0.24
0.33
0.24
0.1
0.22
0. 13
0.26
0.33
0.23
0. 10
0.23
0.07
0.01
0. 11
0.09
0.08
0.09
0.07
0.04
0.29
0.30
0.22
0. 18
0.08
0.33
0.21
0.20
0.33
0.04
0. 18
0.31
0.06
0.03
<0.01
0.01
<0.01
0.01
0.13
0.07
0.05
0.01
0.20
0.07
<0.01
0. 10
0.03
0.04
<0.01
<0.01
0.05
0.01
0.09
0.03
0.01
<0.0l
0.07
0.05
0.05
0.01
0.09
0.04
0.06
0.20
0.05
0.03
<0.01
0.05
0.20
Selenium
(mg/1)
0.006
0.006
0.005
0.006
0.008
0.005
0.005
0.002
0.003
0.008
0.005
0.006
0.004
0.018
0.005
0.10
0.016
<0.001
0.006
0.006
0.007
0.020
0.018
0.016
0.018
0.007
0.006
0.010
0.011
0.016
0.006
0.016
0.012
0.005
0.001
0.009
0.020
Arsenic
(mg/1)
0.026
0.001
0.04
0.011
0.042
0.014
0.04
0.012
0.018
0.041
0.011
0.031
0.031
0.027
0.062
0. 17
0.048
<0.004
0.09
0.069
0.041
0.056
0.078
0.084
0.095
0.039
0.046
0.042
0.036
0.068
0.008
0.020
0.04
0.01
0.001
0.043
0.170
Dissolved
Nickel Solids
(mg/1) (mg/1)
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.06
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.06
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.05
0.051
0.060
150
190
180
140
NA
90
160
120
100
90
120
170
140
160
170
180
190
120
190
200
200
180
190
200
210
220
230
220
210
210
260
230
230
190
90
177
260
Suspended
Solids
(mg/1)
14
18
2
1
NA
3
6
12
2
12
5
^1
1
2
2
2
3
7
11
4
1
5
5
1
3
<1
1
2
2
6
2
7
11
24
-.1
5
24
Sulfate
(mg/1)
85
72
82
50
NA
83
64
47
68
26
100
72
93
97
9-
73 ±
120 U)
20 ^
92
88
89
100
100
100
120
130
120
66
120
95
130
120
120
130
20
90
130
-------
TABLE 33. LINEAR CORRELATION COEFFICIENTS FOR THE VARIOUS ASH POND EFFLUENT PARAMETERS AT PLANT J
Flow
pH
Aluminum
Calcium
Copper
1 ron
Lead
Magnesium
Manganese
Zinc
Selenium
Arsenic
S'ickel
Dissolved Solids
Suspended Solids
Sulfjte
Flou
1-000
0.571
-0.376
0.509
-0. 194
-0.614
-0.294
0.556
-0. 118
0.163
0.490
0.409
0. 170
0.491
-0.099
0.537
pH
1.000
-0.807
0.596
-0.507
-0.609
-0.553
0.833
-0.619
-0.335
0.428
0.439
-0.068
0.554
-0. 198
0.250
Aluminum
1.
-0.
0.
0.
0.
-0.
0.
0.
-0
-0
-0
-0
0
-0
000
374
420
562
479
.623
.454
.239
.194
.193
.007
.465
.365
.159
fair
1.
-0.
-0.
-0.
0.
-0.
-0.
0.
0.
-0
0
0
0
ium
000
306
390
294
.805
.039
.076
.457
.220
.064
.826
.135
.683
Copper
1.
0.
0.
-0.
0.
0.
-0
-0.
0
-0
0
-0
000
349
055
.417
.321
.281
.080
.097
.274
. 164
.048
.149
I ron
1.
0.
-0.
0.
0
-0
-0
-0
-0
0
-0
ooo
144
.514
.294
.043
.305
.363
.032
.278
.493
.285
Le
1.
-0.
0
-0.
-0
-0
-0
-0
0
-0
Dissolved Suspended
ad Magnesium Manganese Zinc Selenium Arsenic Nickel Solids Solids
000
.457
.343
.036
.344
.209
.077
.408
.014
.243
1.
-0
-0
0
0.
0
0
-0
0
.000
.269
.139
.438
.336
.022
.825
.150
.516
1.
0.
-0.
-0.
0.
0
0.
0.
000
.269 1.000
.294 0.073 1.000
.499 -0.186 0.423 1.000
.369 0.128 0.154 -0.045 1.000
.016 0.089 0.423 0.267 0.108 1.000
.018 0.001 0.167 -0.257 -0.183 0.105 1.000
299 0.247 0.433 0.116 0.169 0.700 -0.079
Sulfate
1
p— '
W
Os
1
1 .000
-------
X
o
•o
§.
•JV
40
30
20
10
f\
A
: FLOW
T /A A A^y^ A/5«A
AA A ^
A
-
-
- A Av/Ax
•
;
-
.
• i i i i i i i i i i
^a
20
f.
E
"s l5
.3
1
g 10
-------
^—
J
I
1
o
o
_^
^
g
s
•5
8
u^u
0.15
O.IO
O.05
000
QO8
O06
OX>4
O02
ono
: As
A
-
•—
-
~ A
; A AA
A
- A A
A^A A A .A A
jMw A»^
A «\
A A
A ^A A A A
Aj i " i A i i i i 11
J FMAMJ JASOMO
1977
Ni
,
.
A A
/&*. ^^A^^^^V /v^^. JBk/\/tftf\Abftftf\S\. M\. AAv9CkA
-
•
~*
i i i i i i i i i i i
O.OSJO
OOI5
J
8
f"~ aoio
u
§
0
O.O05
n nnn
\j.\j\j\j
~ O.OI5
^
.g
5
fnmn
W-WIW
u
O.O05
r\r\f\r\
^6 ^ A*
A A A A
A
A
A A
A A
A A
- A\ A A A^ A A
- A <(& A A A
A
A
A
A
J F M A M J J ASOND
1977
: Pb
A
A
A
-
/VS AAA/W A AAAd&AANMtJVy\A
-------
8O
— 6O
f
S
I 40
§
20
0
Co
-
A^A^A
A^A A
A A
-/\. A**
A ^AA
, i i i i i i i i i i
J FMAMJJASONO
2.0
= 1.5
J
'•s
o
0.5
0.0
!A Al
A
—
A
A A
A dk A A
A A
A A
A A
- AA A AA A A
- AAA A
A A A A
i i • i i i i i i i i
JFMAMJJASOND
1977 1977
10.0
80
1.
3
.i
B 6O
i
°
4.0
on
A_
A Mg ^
A
A AiAA A A
— ^ JT^"^ a A ^L ^V
4\ /^ ^^
A AA A
A
A
—
A
A
A
A^A
A A
A
1 1 1 1 | | 1 1 1 1 1
5.0
4-0
li 3.0
&
*
1 2.0
§
o
1.0
O
: Fe
. .
i— ^
"
-
—
-
A
- A
- A
~
_
r A A
A^A 4^A /AAA^^ AAA
- , . , A A^, , . A A .
1977
JFMAMJJASOND
1977
Figure 16 (Continued)
-------
0.4
? 0.3
01
,E
5
1
I °2
o
O.I
0-0
Cu
A
; A
A
LA A
AAA A .
A At A . V\ A
A A ^^ A *^ A A At
— ^\ -fA . . - , - - . f\ /\^^^
i A^ A ^ A, AA, A A, A , A , L , L
JFMAMJJASOND
= 03
I
S
B
I °'2
u
c
o
o
O.I
0.0
Mn
A A A A
- /A
* A
A
- A A
A A
r aa a \-VA .
r A
i i i i i i t i i i i
JFMAMJJASOND
1977 1977
1
O
U.cU
_ 015
e
1 °'°
(j
§
o
005
/t rt/\
: Zn
_
-
A
A
A A
-
A A A
- A A
A A AA A
A A
- A A A A
A\AA A A AAA AA A
i i i i i i i i i i i
\*J\J
125
100
I 75
s
1 °°
{J
25
0
^ Sulfate A A A
A A A
A A At\ X!^
_
^ A A AAA
^\ ^\ .^^ »
^v^
- A . A A
. AA A
7 A A
'.
L A
A
-
I i r i i i i i i i . i. _
J FMAMJ J ASOND
1977
JFMAMJJ ASO NO
1977
(Figure 16 (Continued)
-------
-141-
(ESP) were put into operation causing an additional estimated flow of 12
million gallons per day (MGD). Therefore, the changes in the trace
metals could be associated with the possible change in ash characteris-
tics as a result of the new ESP units. However, since the change in
flow occurs approximately one month before the pH change and the change
in trace metals occurred at approximately the same time as the pH, the
new ash probably does not account for the changes shown for trace metals
in Figure 16. A cycle was not indicated for suspended solids, lead,
zinc, manganese or copper.
By assuming there to be a yearly cycle for some parameters, the
data set must be divided into smaller data sets of random events in
order to estimate the effluent characteristics for the various time
periods. Since most of the elements exhibiting a cycle were signifi-
cantly correlated with pH and the pH appears to change in April, the
data for the following elements was divided into two data sets: pH, Al,
As, Ca, dissolved solids, Fe, Mg, Se, and sulfate. The two data sets
consist of the data from November 1 to April 30 and from May 1 to
October 31. The change in pH does not always occur at the same time
each year as indicated by the data in Figure 3, therefore, the year was
divided into a 6-month period although the low pH period may only last 3
months out of every year. For the elements defined as having a yearly
cycle the variation at any point in time is the same and only the rela-
tive concentration has changed, whereas for others the variation has
changed considerably. Calcium is a good example of the first case,
while As is a good example of the second case. For those parameters
defined as not having a yearly cycle, the data was not divided into two
distinct sample periods. Therefore, the remainder of this section deals
with the three data sets previously discussed.
STATISTICAL DISTRIBUTION OF THE EFFLUENT CHARACTERISTICS AT PLANT J
Cumulative frequency plots were prepared for the data given in
Table 32 for the time periods discussed in the previous subsection.
These plots are shown in Figures 17, 18, and 19. The best fit straight
line was determined by visual placement. Figure 17 is for the parame-
ters for which no yearly cycle was indicated. Figures 18 and 19 are
for the periods November 1 to April 30 and May 1 to October 31, respec-
tively, for those parameters for which a yearly cycle was defined.
Plots are given for the linear and logarithmic scales the same as for
Plant E. These plots were compared visually to determine the best
estimate of the type distribution of the data. For those elements
exhibiting a cycle, the distribution of the data was assumed the same
for both periods in order to simplify the calculations.
Table 34 lists the various parameters and the assumed distribution
based on this comparison. The following elements were assumed to exhibit
no cycle and be lognormal: Cu, Mn, and suspended solids. Zinc was
assumed to exhibit no cycle and be normal. For those elements which
were assumed to exhibit a yearly cycle, the following were assumed
lognormal: Ca, dissolved solids, Fe, Mg, pH, Cu, Mn, and suspended
solids, and the following were assumed normal: Al, As, Se, and sulfate.
-------
~i—i—r—i—i—i 1—r~
Suspended
Solids
Z90W ««0 <0«0
i i i i i i i
o.,L
Cu
an 1 1 i >a i i i i i ill anil i—i.—j. ' L. ' J- '—W>—A jL A
°"^9 0 ZO 40 a» B «0 W M ™"Tl 5 » ZO «0 «O 10 M 90 M
SDK) 40 eo 80 90 95 9«
-1 1 1 1—i—I—I—I r-
Suspended
tsnto 40«o IOM>
Cu
-i—i 1—i—i—i—i—i—i 1—r
Mn
Zn
QmMm Fraqmney DM
ID «0 55 16 ID 9S 9>
Cmutatlv* FraqumeyM)
to
i
Figure 17. Cumulative Frequency Plots for the Ash Pond Effluent at Plant J for the Period January 1 to December 31
-------
aoi
u*-
Al
°4s;oa!0|4oi.ol.0.0»k..
WOr
Co
»H
— lOOr 1 1 1 1—i—i—r
j [ Mg
Z 9 « 20 40 «0 80909598
2 5 10 20 40 80 80 9O 95 98
i*H
oah
"JO 2^ ' V80 ' 80 90»9 M
4O
~TIIIII\1IIT"
Co
1 1 1 1
2 5 W 20 4080 80909996
Cumutotiv* Frtqucncy (K)
OjO
2 5 10 20 4060 80909998
CumuWiw F™qu«nqf (%)
Figure 18. Cumulative Frequency Plots for the Ash Pond Effluent at Plant J
for the Period November 1 to April 30.
-------
)l
" oo
5
5
I
i
)
»
200
: Dissolved i \ SCk q :
Solids ^ ^ \ -
\ 4 '-
0^°°^ ' ,
~ ^-^5^° ~ ""^ nxf^O ^ v '°
-
-
1 i i 1 1 1 1 1 I i i IQ
o
I 1 1 1
— I — 1 1 — 1 — 1 — 1 — 1 — 1 — 1 1 1 — J
: PH :
-
- _ O O __
^.o o v
Z 5 K> M> 40 60 80 30 93 9« 29O2O 4060 8O9O9996 "291020 4060 80 90 95 9(
Dissolved
. Solids J «,
i r /" -
- 7
- BO- /
-/b
1 1 1 1 1 1 1 1 1 1 1 c
II 1 1 1 1 1 1 [ f 1
PH / _
7°° -
/ J
f /- l
>— —
-
25»ZO 4000 8090999*
Cumulotiv* Fr*qiwncy (%)
291020 4060 N909S9*
Cumutatrve FreqiMney VU
Figure 18 (Continued)
CurnAitiv* Frequency OU
-------
1 1 1 1 1
As
2 9 K> 20 4O60 60 90 99 98
~i 1 i—i—i—i—i—i—i 1 r
Se
2 5 O ZO 40 CO 809O95
231020 40 60 90 90 95 96
= O.O6
I
As
o o
25^0 tO 40 60 80 90 89 96
Cumulotiva Frequency (%)
291020 4O 60 60909996
Cumulative Fraquency (%)
~~\ 1 1 1 1 1—I 1 1 1 T~
Se
O.OB-
oatot-
0406
0.0001 L
291020 4060 809099
Cwnukjtiw Frequency (%)
Fipure 18 (Continued)
-------
lOOr
Al
CP
CO 80 90 95
lOOOr
~i—i 1—i—i—i—r
Co
100-
iCP
cP,
,0oo o _
no
Mg
00 0_
5 10 20 4060 90 90 98 98 2 5 O 20 4O6O 80909596
r—T—r I
Al
cp
O2- O ,
Q I ' I I I I I I I I
T 5 «5 3540 «O 85 90
Cumutottvt Fnqu6ncy (%)
60-
Co
Cumukrtive Frwjuency (%)
Figure 19. Cumulative Frequency Plots for the Ash Pond Effluent at Plant J
for the Period May 1 to October 31
-------
~T I 1 1 1 1 1 1 1-
Dissolved
Solids
2 5 O 20 4O60 80 90 95 98
_T 1 I
sex
00
- o
5(020406080909598
PH
_ooo oo
o—o—_
ii i i i i i i i ii
5 - O2O4O6OeO9O954«
BO
Dissolved
Solids
231020 4060 SO 9O 95 98
CunutatiM Frequency (%)
100-
90-
SO/
°2 5 K3 2O 4O60 8O 90 95 96
Cumilotivt Frequency (%)
7.0-
63-
6.0.
l I i i
10 20 4O SO SO 9O 95 98
Cumutotive Frequency (%)
Figure 19 (Continued)
-------
- o
2 5 10 20 4060 609099
10.01
1.0-
Fe
00^
II I I I I I I I II
? 9 10 20 40 SO tO 90 96 96
Se
25O2O «O 6O 90 9O 95
OM
O06
Q04
1 1 1—I—I—1—I 1 V T
—I—1—I 1 f-
As /
cP
0.00
W
/
25O2O 4O 60 90 9O 9S
CumuMim FnqiMnqr «0
0,8-
OJ6-
04
Fe
,00
2O 40 CO 90 9O 96 98
Cumutotiv* Frequency (%)
OOB-
o.on
O005-
0000
291020 4060 8O 90 95 96
Cumutotrv* Frequency (%)
00
I
Figure 19 (Continued)
-------
TABLE 34. SELECTED SAMPLING PERIOD, TYPE 01 DISTRIBUTION AND STATISTICAL CHARACTERISTICS
OF THE ASH POND EFFLUENT AT PLANT J
Parameter
Aluminum
Arsenic
Calcium
Dissolved Solids
Iron
Magnesium
pH
Selenium
Sulfate
Copper
Manganese
Suspended Solids
Zinc
Sample
Period3
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
1
2
3
3
3
3
Type of
Distribution
Normal
Normal
Normal
Normal
Lognormal
Lognormal
Lognormal
Lognormal
Lognormal
Lognormal
Lognormal
Lognormal
Lognormal
Lognormal
Normal
Normal
Normal
Normal
Lognormal
Lognormal
Lognormal
Normal
Mean
0.88
0.52
0.024
0.054
1.415
1.645
2.125
2.295
-0.054
-0.415
0.683
0.913
0.775
0.881
0.005
0.011
70
101
-1.507
-0.804
0.540
0.050
Variance
0.2631
0.0513
0.0002
0.0014
0.0108
0.0105
0.0127
0.0049
0.2095
0.0381
0.0215
0.0027
0.0084
0.0010
0.000003
0.00003
444
689
0.1904
0.628
0.1737
0.0024
Number of
Samples
13
21
13
21
13
21
12
21
13
21
13
21
12
20
13
21
12
21
32
33
32
33
(S2)(t2) for various confidence levels0
99%
2.456
0.415
0.0019
0.011
0.101
0.085
0.123
0.040
1.956
0.308
0.201
0.022
0.081
0.008
0.00003
0.0002
4283
5577
1.435
0.472
1.309
0.018
95%
1.274
0.223
0.0010
0.006
0.052
0.046
0.062
0.021
1.015
0.166
0.104
0.012
0.041
0.004
0.00001
0.0001
2151
2998
0.792
0.261
0.723
0.010
80%
0.489
0.090
0.0004
0.002
0.020
0.018
0.024
0.009
0.389
0.067
0.040
0.005
0.016
0.002
0.000006
0.00005
825
1210
0.326
0.108
0.298
0.004
aSample period number 1 is from November 1 to April 30. Sample period number 2 is from May 1 to October 31.
Sample period number 3 is for the entire year.
The values given for lognonnal distribution are for the logarithms of the concentrations while those for normal
distributions are for the untransformed concentrations.
CSee equation 3 for definition of (S2)(t2).
£
-------
-150-
These assumption are in agreement with those for the effluent at Plant E.
Exceptions, however, are Al, pH, and Zn. Table 34 contains additional
information which will be discussed in the following subsection.
Table 35 gives the mean, appropriate ash pond effluent limitation
or proposed water quality criteria and the probability that these limi-
tations or criteria are exceeded for the effluent parameters assuming
a normal distribution. Table 36 gives the mean of the logarithms of
the concentrations, the logarithm of the geometric mean, appropriate
ash pond effluent limitation or proposed water quality criteria and the
probability that these limitations or criteria are exceeded for the
effluent parameters assuming a lognormal distribution. All calcula-
tions for lognormal distributions will be based on the values given
in Table 36 for the logarithm of the geometric mean.
The effluent limitations given in Tables 35 and 36 for pH and
suspended solids are those outlined for the steam-electric power
generating industry by EPA in 1974 (1) for the achievement, by 1977, of
best practicable control technology currently available (BPCTCA). The
pH is to be maintained between 6 and 9 and the average daily suspended
solids for a 30-day period is to be at or below 30 mg/1 with a daily
maximum equal to or less than 100 mg/1. Since limitations for the ash
pond effluent at Plant J for the remaining elements have not yet been
promulgated, the criteria specified in EPA's "Water Quality Criteria"
(9) for domestic water supply intakes are used. A list of the criteria
are given in Appendix C. This does not suggest that the ash pond effluent
should meet these criteria. They are only given for comparison purposes
and as an aid in establishing the desired precision for the future
monitoring program. The data in Table 36 show that the pH during the
sample period November 1 to April 30 is not in the pH range of 6 to 9
approximately 62 percent of the time. However, during the period May 1
to October 31 the pH is not in the 6 to 9 range less than 2 percent of
the time. The suspended solids concentration is above 30 mg/1 less than
2 percent of the time. Arsenic and selenium are shown to be above the
EPA proposed water quality criteria approximately 50 percent of the time
during the period May 1 to October 31, but during the remainder of the
year, only exceed the criteria 4 percent and less than 2 percent of the
time, respectively. Iron and manganese were above the criteria greater
than 70 percent and 95 percent of the time, respectively. Copper,
sulfate, and zinc exceeded the criteria less than 2 percent of the time.
ESTIMATION OF THE MEAN AS A FUNCTION OF THE PRECISION
The number of samples, n, required to estimate the mean as a func-
tion of L was plotted for each parameter based on the data given in
Table 34 and equation 3. The results are shown in Figures 20, 21, and
22. Figure 20 is for those parameters to be sampled during the period
January 1 to December 31. Figures 21 and 22 are for those parameters
for which the data set was divided into two periods. They were con-
structed by dividing the values shown under the column labeled "(S2)(t2)"
in Table 34 by various values of (L)2 to yield various sample sizes, n.
The values for (S)2(t)2 given in Table 34 were obtained by multiplying
the variance, S2, times the appropriate t value squared. The values
-------
-151-
TABLE 35. COMPARISON OF THE ASH POND EFFLUENT CHARACTERISTICS FOLLOWING
A NORMAL DISTRIBUTION AT PLANT J WITH THE ASH POND EFFLUENT LIMITATIONS
OR WATER QUALITY CRITERIA
Parameter
Aluminum
Arsenic
Selenium
Sulfate
Zinc
Sample
Period
1
2
1
2
1
2
1
2
3
Mean of the
Concentrations
(mg/1)
0.88
0.52
0.024
0.054
0.005
0.011
70
101
0.05
Standard or
Water Quality
Criteria (mg/1)
b
b
0.05C
0.05C
0.01C
0.01C
250°
250C
5.0C
Probability that
Standard is Exceeded
-
4
50
<2
50
<2
<2
<2
al = November 1 to April 30; 2 = May 1 to October 31;
3 = January 1 to December 31.
bNo criteria proposed for drinking water supplies.
°Proposed EPA intake standards for domestic drinking water supplies
(EPA 1976).
-------
TABLE 36. COMPARISON OF THE ASH POND EFFLUENT CHARACTERISTICS FOLLOWING A
LOGNORMAL DISTRIBUTION OF PLANT J WITH THE ASH POND EFFLUENT LIMITATIONS
OR WATER QUALITY CRITERIA
Mean of the
Sample Logarithms of the
Parameter Period Concentrations
Calcium
Dissolved Solids
Iron
Magnesium
pH
Copper
Manganese
Suspended Solids
1
2
1
2
1
2
1
2
1
2
3
3
3
1.415
1.645
2.125
2.295
-0.054
-0.415
0.683
0.913
0.775
0.881
-1.507
-0.804
0.540
Logarithms of the
Geometric Mean
1.362
1.602
2.114
2.280
-0.060
-0.420
0.771
0.903
0.756
0.863
-1.658
-0.854
0.415
Standard or
Water Quality Probability that
Criteria (log mg/1) Standard is Exceeded
d
d
d
d
0.3e
0.3e
b
b
6 to 9^
6 to 9r
1.0e
0.05e
30f
_
-
_
-
80
70
_
62
<2
<2
95
<2
NJ
I
1 = November 1 to April 30; 2 = May 1 to October 31; 3 = January 1 to December 31.
Values given are logarithms to the base 10 of the concentrations in mg/1.
°Values given are the logarithms to the base 10 of the estimated mean in mg/1.
Tfo criteria proposed for drinking water supplies.
eProposed EPA intake standards for domestic drinking water supplies (EPA 1976).
Effluent limitation specified in the NPDES permit.
-------
Suspended
Solids
002
0.04 006 OJ06
Precision (log mg/Q
O.IO O.IZ
0.2
O.4 Ofi 0.8
Precision (tog mg/l)
1.0
O.I
0.2 0.3 0.4
Precision (log mg/l)
0.5 O.6
u>
i
0.02
0.04 0.06
Precision (mg/l)
0.06
0.10
Figure 20. Number of Samples Required for a Given Precision for the Ash Pond Effluent
at Plant J for the Period January 1 to December 31
-------
O.I 02 0.3 0.4 0.5
Precision (mq/l)
0.10 0.20
Precision (tog mg/l)
O30
02 0.3
Precision (log mg/l)
04
Q5
I
H-t
Cn
I
Figure 21. Number of Samples Required for a Given Precision for the Ash Pond Effluent
at Plant J for the Period November 1 to April 30
-------
O.O2
O.O4 0.06
Precision (mg/l)
0.06
0.10
O.OO4 oooe
Precision (mg/l)
0012
OO4 O.08 0.12
Precision (log mg/l)
0.16
Figure 21 (Continued)
-------
Dissolved Solids
O04
OO8 OJ2
Precisian (tog mg/0
O-I6
O-2O
OjOZ 0.04 O.O6
Precision (tog mg/0
0.08
0.10
K>
20 30 40 SO 60
Precision (mg/0
70 80
tn
-------
O2 04
O6 OS U>
Precision (mg/l)
12 14 L6
OK) 020
Precision (log mg/l)
0-30
0.2
0.4 0.6 08 LO
Precision (log mg/l)
12 14
Figure 22. Number of Samples Required for a Given Precision for the Ash Pond Effluent
at Plant J for the Period May 1 to October 31
-------
OCX 0.02 O.03 O.O4
Precision (mg/l)
0X502 OO04
Precision (mg/l)
oooe
0.1
02 03
Precision (log mg/l)
04
0-5
00
I
Figure 22 (Continued)
-------
60
OJO 020
Precision (tog mg/l)
0-3O
Dissolved Solids
0.08 0.16 024
Precision (tog mg/0
O32
K> 20
30 40
Precision (mg/l)
50 60
70
VO
I
Figure 22 (Continued)
-------
-160-
used for t are a function of the confidence level and number of data
points used to generate the variance. The t values necessary for
calculating the (S)2(t)2 values in Table 34 are given in Appendix D.
SELECTION OF THE PRECISION
The upper and lower limits (#1 and #2) for the critical range of
the precision at the 99 percent confidence level for the ash pond efflu-
ent characteristics at Plant J are given in Table 37. The upper limit
given is for the precision produced by one sample. However, for all
elements, the curve had become asymptotic to the x-axis at a precision
less than the upper limit given in Table 37 indicating the precision
could be increased significantly by the addition of one more sample.
The upper limit of the critical range was given based on one sample
because that precision may be adequate for the monitoring program. The
lower limit given is for the precision produced by 52 samples because
the curve had not become asymptotic to the y-axis for any of the elements.
This assumes resources are not available for the collection or analysis
of more than 52 samples in any one sample period. If the value for the
precision required for the monitoring program is greater than the upper
limit, only one sample per period needs to be collected. However, if
the value is less than the lower limit, then 52 samples per period would
be required. If for some reason the precision for 52 samples is not
adequate for an element of a monitoring program, then a decision would
have to be made whether or not to increase the level of resources
allocated to the monitoring program. If the required precision is
between the limits, then the data in Figures 20, 21, or 22 would be
consulted to determine the sampling frequency. Table 37, therefore,
gives valuable insight into the importance of the required precision on
the design of an ash pond effluent monitoring program for Plant J.
Suspended solids and pH are the only parameters included in this
study for which ash pond effluent limitations have been set for Plant J.
The design procedure discussed in Section 4, using the relationship,
L = (j - X, can only be applied to suspended solids and pH during the
period from May 1 to October 31. *fte precision required for suspended
solids is calculated by subtracting the estimate of the geometric mean
of the logarithms of the data given in Table 32 for suspended solids
(0.415, see Table 36) from 1.477 (log of the effluent limitation of
30 mg/1). This yields a precision of 1.062. The precision required for
the pH during the period May 1 to October_31 is calculated by substi-
tuting 0.778 (log 6) for (j and 0.903 for X. This yields a value of
0.125 for the precision.
Defining the precision which should be used to design the
monitoring program at Plant J is difficult where effluent limitation
have not yet been promulgated. One method is to assume some precision
based on a given percentage of the sample mean. For comparison purposes
sampling frequencies based on estimating the yearly mean within 10 and
20 percent of the true mean at the 99 and 80 percent significance levels
will be discussed in the next subsection. Another method for establishing
the precision L is to allow for a certain level of environmental harm
(or pollutant loading to the receiving stream). This cannot be done
-------
-161-
TABLE 37. UPPER AND LOWER LIMITS FOR THE CRITICAL RANGE OF
THE PRECISION FOR THE EFFLUENT CHARACTERISTICS OF PLANT J
Element
Sample
Period
Lower Limit
of L
(mg/D
Upper Limit
of L
(mg/1)
Elements following a normal distribution
Aluminum
Arsenic
Selenium
Sulfate
Zinc
1
2
1
2
1
2
1
2
0.217
0.089
0.006
0.015
0.0008
0.002
9
10
1.57
0.644
0.044
0.105
0.0055
0.014
65
75
0.018
0.134
Elements following a lognormal distribution
Calcium
Dissolved Solids
Iron
Magnesium
pH
Copper
Manganese
Suspended Solids
1
2
1
2
1
2
1
2
1
2
3
3
3
0.044
0.040
0.049
0.028
0.194
0.077
0.062
0.021
0.039
0.012
0.166
0.095
0.159
0.318
0.292
0.351
0.200
1.399
0.555
0.448
0.148
0.028
0.089
1.198
0.687
1.144
-------
-162-
without some estimate of the receiving stream water quality. The effluent
was discharged directly to the river (although upstream of the water
intake) during the period when the data in Table 32 was collected. The
effluent discharge location was changed after the data in Table 32 was
collected. The ash pond effluent at Plant J now discharges to the main
water intake canal serving the entire power plant at a location upstream
of the main intake pumps. Only a small fraction of this intake water
(~4%) is used for ash sluicing. The remainder is used for condenser
cooling water and other miscellaneous processes. This new discharge
location creates some degree of water reuse for ash sluicing. The
degree of reuse depends on the mixing of the streams. For the purposes
of this study, complete mixing will be assumed creating a very low
degree of reuse. The ash pond effluent characteristics can then be
assumed equal to those for the once-through system given in Table 32.
The water quality characteristics shown in Table 38 will be assumed for
the stream receiving the ash pond effluent from Plant J. These values
are based on the 1976 data for the intake water to Plant J. They differ
somewhat from the data given in Table 5 for the intake water during 1974
and 1975 mainly because of the precision involved in developing the
average.
The significance level and precision for the data in Table 38 are
not specified. For design purposes the values will be assumed to be
absolute. In addition, some dilution factors and maximum allowable
concentrations in the receiving stream must be specified. The dilution
factor assumed for Plant J's ash pond effluent into the intake water
stream is approximately 0.0435. It is based on an intake water flow of
1200 MOD and a maximum ash pond effluent flow of 50 MGD (~34,800 gpm).
The maximum ash pond flow was assumed based on the highest reported
value in Table 32. The calculations for the precision are the same as
for Plant E. However, the value substituted for the sample mean, X, in
equation 2 for those elements which were divided into two data sets was
calculated by averaging the means for the different sample periods.
This had to be done as opposed to taking the mean of the entire data
because there were more samples taken during the second period, therefore
baising the mean. But by assuming the events to be random within a
subset of the data, the mean of the entire data set can be estimated by
averaging the weighted subset means. In this case, the periods of the
subsets were equal, therefore, the mean of the entire data set could be
estimated by simply adding the subset means and dividing by 2. The
geometric means were used for those parameters following a lognormal
distribution.
Table 39 gives the allowable ash pond input concentration to the
receiving stream and precision required by the monitoring program,
assuming the maximum allowable average concentration in the receiving
stream is based on maintaining the concentration in the receiving stream
equal to or below the EPA proposed water quality criteria for domestic
water supply intakes. Table 40 gives the same information for a monitor-
ing program assuming the maximum allowable average concentration in the
receiving stream is below or equal to the maximum value given in Table
38. The sampling frequencies associated with these precisions will be
discussed in the following section.
-------
-163-
TABLE 38. ASSUMED WATER QUALITY CHARACTERISTICS FOR
THE RECEIVING STREAM AT PLANT J
Element
Aluminum
Arsenic
Calcium
Copper
Dissolved Solids
Iron
Magnesium
Manganese
Selenium
Sulfate
Suspended Solids
Zinc
Average
Concentration
(mg/1)
0.45
0.006
17.7
0.035
81
0.5
4.7
0.086
0.002
17
6
0.013
Maximum
Concentration
(mg/1)
0.60
0.010
33
0.050
140
0.84
8.4
0.120
0.002
26
12
0.020
-------
-164-
TABLE 39. REQUIRED PRECISION FOR THE MONITORING PROGRAM AT PLANT J
ASSUMING AN AVERAGE ALLOWABLE CONCENTRATION IN THE RECEIVING STREAM
EQUAL TO THE EPA PROPOSED WATER QUALITY CRITERIA
Element
Arsenic
Iron
Selenium
Sulfate
Copper
Manganese
Zinc
Maximum Average
Allowable Concentration
In the Effluent
(mg/1)
1.062
a
0.194
5600
23
a
120
Required
Precision
1.023
-
0.186
5515
3.02
-
119.95
a. Intake water exceeds the criteria.
-------
-165-
TABLE 40. REQUIRED PRECISION FOR THE MONITORING PROGRAM AT PLANT J
ASSUMING AN AVERAGE ALLOWABLE CONCENTRATION IN THE RECEIVING STREAM
EQUAL TO THE MAXIMUM VALUE REPORTED FOR THE INTAKE WATER
Element
Aluminum
Arsenic
Calcium
Dissolved Solids
Iron
Magnesium
Selenium
Sulfate
Copper
Manganese
Suspended Solids
Zinc
Maximum Average
Allowable Concentration
in the Effluent
(mg/1)
4.05
0.102
385
1497
8.66
93.5
0.002
233
0.40
0.90
150
0.18
Required
Precision
3.35
0.063
1.103
0.978
1.178
1.134
a
147
1.26
0.808
1.76
0.13
aThe reported ash pond effluent concentration exceeds the maximum average
allowable concentration calculated by this method; therefore, the procedure
developed in Section 4 for determining the number of samples to show com-
pliance with a selected water quality criteria cannot be used.
-------
-166-
ESTIMATED SAMPLING FREQUENCIES
The precision required to determine the minimum number of samples
needed to show that the ash pond effluent for Plant J is in compliance
with the effluent limitation for suspended solids was calculated to be
1.062. This value falls within the critical range of the deviation for
suspended solids indicating the number of samples can be determined from
Figure 20. For the 99 percent confidence level, this means two samples
per sample period are required. Since the effluent limitation specifies
that the concentration must not exceed an average of 30 mg/1 for 30
consecutive days, the number of samples derived from Figure 20 represents
a sampling frequency of two samples per 30 days or 24 samples per year
assuming 30 days per month. This yields a sampling frequency of one
sample every 15 days. This assumes, of course, that the variance obtained
for the data over the period of the extensive sampling program and used
to construct Figure 20, is equal to the variance had the period of the
survey been any one month and the same number of samples been collected.
This assumption is valid when the data are randomly distributed.
Corresponding sampling frequencies for both the 95 and 80 percent
confidence levels would be one sample per month.
The sampling frequency of one sample per 15 days (2 per month) for
the 99 percent significance level is the same sampling frequency currently
being required by the NPDES permit. The 24 suspended solids samples per
year allows estimation of the yearly geometric mean within 36 percent of
the true yearly geometric mean and estimation of the monthly geometric
mean within 66 percent of the true monthly geometric mean.
The average value reported for pH for the period May 1 to October
31 can be shown to be within the range of six to nine within 99 percent
confidence by collection of one sample during the period. However, the
mean for the period from November 1 to April 30 was less than 6 and,
therefore, the sampling frequency to show compliance, could not be
determined. Sampling the pH one time during the period May 1 to October
31 estimates the average pH to within ± 1.2 pH units.
The above estimates are appropriate if the average is interpreted
to mean the geometric mean when dealing with lognormal data. The geo-
metric mean is always smaller than the arthimetic mean, thus, in effect,
creating a slightly higher standard when transforming the standard to a
logarthim value and comparing it with the geometric.
Table 41 shows the number of samples required per year to estimate
the yearly mean (geometric mean for lognormal data) within 20 percent of
the true mean for the 99, 95, and 80 percent confidence levels. For
those elements which were divided into two sampling periods, the number
of samples per period are given in parenthesis. A substantial sampling
effort (greater than 52 samples per year) would be required to estimate
the yearly mean within 20 percent for As, Fe, Se, suspended soilds, and
zinc at the 99 percent confidence level, whereas a minimal effort (only
1 sample per year) would be required for calcium and dissolved solids.
The remaining parameters would require between 3 and 42 samples per year
at the 99 percent confidence level. More samples for Al and Fe should
be collected during the period from November 1 to April 30 than during
the rest of the year, while just the opposite is true for As and Se.
-------
-167-
TABLE 41. NUMBER OF SAMPLES REQUIRED TO ESTIMATE THE YEARLY
MEAN WITHIN 20% OF THE TRUE YEARLY MEAN FOR PLANT J
Number of Samples Required
Element
Aluminum
Arsenic
Calcium
Dissolved Solids
Iron
Magnesium
PH
Selenium
Sulfate
Copper
Manganese
Suspended Solids
Zinc
99% SL
42(29,13)
66(18,48)
1(1,0)
1(1,0)
290(203,87)
3(2,1)
4(3,1)
30(7,23)
12(5,7)
9
11
122
116
95% SL
22(15,7)
34(9,25)
1(1,0)
1(1,0)
148(104,44)
2(1,1)
2(1,1)
16(4,12)
7(3,4)
5
6
68
64
Per Year3
80% SL
9(6,3)
13(4,9)
1(1,0)
1(1,0)
105(74,31)
1(1,0)
1(1,0)
6(1,5)
3(1,2)
2
3
28
26
a. Numbers in parenthesis indicate the number of samples required
during the period November 1 to April 30 and May 1 to October 31,
respectively.
SL = Significance level.
-------
-168-
Table 42 gives the estimated sampling frequencies for the preci-
sions given in Tables 39 and 40. Table 42 indicates that one sample per
year for all parameters except Al, As, and Zn ensures within 99 percent
confidence that the yearly average concentration in the receiving stream
will not be increased (by the ash pond effluent) above the maximum value
reported in the intake water for 1976. Arsenic, aluminum, and zinc
would require two samples per year for the same assurance. Likewise,
only one sample per year for all parameters ensures within 99 percent
confidence that the receiving stream's yearly average concentration will
not be increased (by the ash pond effluent) above the EPA proposed water
quality criteria for domestic drinking water intakes. Therefore, estab-
lishing monitoring frequencies based on maintaining the average concentra-
tion in the receiving stream equal to or below the maximum value reported
in the intake water in 1976 automatically ensures monitoring frequencies
as great or greater than those based on maintaining the average concen-
tration in the receiving stream equal to or below the EPA proposed water
quality criteria for domestic water supply intakes.
The sampling frequencies listed in Tables 41 and 42 differ con-
siderably. The frequencies based on the assumed allowable level of
increase in the receiving stream are substantially lower than those
required to estimate the mean within 20 percent. The sampling frequency
used in the final monitoring program should, therefore, be a compromise
between the frequencies given in Tables 41 and 42. As an aid in esti-
mating the point of compromise, the deviation of the yearly sample mean
from the true mean for the 99 percent confidence level is given for the
following frequencies: yearly, quarterly, bimonthly (once every two
months), monthly, biweekly (once every two weeks) and weekly (see
Table 43). For those parameters which were assumed cyclic, the samples
are not collected uniformly over the year, but the same number of samples
are collected for the corresponding frequency. These frequencies were
selected because they are the most widely used. The data given in Table
43 also indicates that the deviation of the sample mean from the true
mean varies from parameter to parameter at each sampling frequency for
the ash pond effluent parameters at Plant J.
EXAMPLE SAMPLING PROGRAM FOR PLANT J
An example sampling program for Plant J to meet NPDES requirements
is shown in Table 44. It is based on the previous discussion and the
following criteria:
1. The element must be required by the NPDES permit.
2. The 99 percent confidence level was assumed.
3. The precision used to estimate the sampling frequency was based
on maintaining the average concentration in the receiving stream
below or equal to the maximum concentration reported for the
receiving stream in 1976 under the 7-day 10-year minimum flow.
This justification for trace metals was assumed because biological
studies performed for P.L. 92-500, Section 316, demonstrations
indicated no adverse biological effects of the discharges from
Plant E.
-------
-169-
TABLE 42. ESTIMATED SAMPLING FREQUENCIES FOR THE MONITORING PROGRAM AT
PLANT J ASSUMING ALLOWABLE AVERAGE CONCENTRATIONS IN THE RECEIVING STREAM
EQUAL TO THE EPA WATER QUALITY CRITERIA AND MAXIMUM VALUE REPORTED FOR
THE INTAKE WATER
Q
Number of Samples Per Year
Precision Based on
Precision Based on , Maximum Value Reported
Element
Aluminum
Arsenic
Calcium
Dissolved Solids
Iron
Magnesium
Selenium
Sulfate
Copper
Manganese
Suspended Solids
Zinc
Water Quality Criteria
d
1(0,1) .
d
d
e
d
1(0,1)
1(0,1)
1
e
d
1
for the Intake Water
1(1,0)
2(0,2)
1(1,0)
1(1,0)
1(1,0)
1(1,0)
f
1(0,1)
1
1
1
2
a. Values are for the 99% significance level. Numbers in parenthesis indicate
the number of samples required during the period November 1 to April 30
and May 1 to October 31, respectively.
b. See Table 39 for the precision values.
c. See Table 40 for the precision values.
d. Criteria not proposed for drinking water intake supplies.
e. Intake water concentration exceeds criteria.
f. Intake water concentration exceeds effluent concentration.
-------
-170-
TABLE 43. DEVIATION OF THE YEARLY SAMPLE MEAN FROM THE TRUE MEAN FOR THE
99% CONFIDENCE LEVEL AT VARIOUS SAMPLING FREQUENCIES
Deviation from the True Mean*
Parameter Yearly Quarterly Bimonthly Monthly Biweekly Weekly
18(0.157)
22(0.011)
2.9(0.044)
0.6(0.013)
16(0.045)
4.8(0.042)
3.0(0.026)
16(0.0015)
10(10)
9(0.166)
10(0.095)
28(0.159)
28(0.019)
Aluminum
Arsenic
Calcium
Dissolved Solids
Iron
Magnesium
PH
Selenium
Sulfate
Copper
Manganese
Suspended Solids
Zinc
62(1.129)
67(0.078)
18(0.315)
4.0(0.091)
58(0.326)
27(0.303)
19(0.191)
58(0.011)
46(73)
42(1.20)
45(0.687)
73(1.144)
73(0.134)
45(0.565) 40(0.461)
50(0.039) 45(0.032)
9.6(0.158) 8.0(0.129)
2.0(0.046) 1.7(0.037)
40(0.163) 36(0.133)
15(0.152) 13(0.124)
10(0.096) 8.6(0.078)
43(0.006) 33(0.004)
30(36) 26(30)
27(0.600) 23(0.489)
29(0.344) 25(0.280)
58(0.572) 53(0.467)
57(0.067) 52(0.055)
32(0.326) 24(0.222)
37(0.023) 28(0.015)
6.1(0.091) 4.0(0.062)
1.2(0.026) 0.8(0.018)
28(0.094) 21(0.064)
9.5(0.088) 6.5(0.059)
6.2(0.055) 4.4(0.038)
27(0.003) 22(0.0022)
20(21) 14(14)
17(0.345) 12(0.235)
19(0.198) 14(0.135)
44(0.330) 35(0.224)
44(0.039) 34(0.026)
a. Values are given as precent of deviation from true mean. Numbers in parenthesis.
indicate the deviation in mg/1 or log mg/1.
-------
-171-
TABLE 44. EXAMPLE SAMPLING PROGRAM FOR PLANT J
Sampling Frequency (No. per Period)'
November 1
Element to April 30
Aluminum
Arsenic
Cadmium
Calcium
Chromium
Copper
Dissolved Solids
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
PH
Selenium
Sulfate
Suspended Solids
Zinc
0
0
0
0
0
c
0
1
0
0
c
0
0
3
1
0
c
c
May 1 to
October 31
0
2
0
0
0
c
0
0
0
0
c
0
0
1
3
0
c
c
January 1 to
December 31
0
2
0
0
0
1
0
1
0
0
1
0
0
4
4
0,
24d
2
Precision
(% of true
yearly mean)
—
59
-
-
-
42
-
58
-
-
45
-
-
10
43
_
36
66
a. The number given for the period January 1 to December 31 equals the sum of
the other two periods.
b. At the 99% significance level.
c. Does not matter which period sample is collected.
d. The frequency should be one sample every 15 days.
-------
-172-
4. If the average concentration in the effluent exceeded the maxi-
mum value reported for the intake water and the EPA proposed
water quality criteria, then the frequency was established based
on estimating the average within at least 50 percent of the true
mean.
5. For those elements for which an effluent limitation has been set,
the recommended frequency ensures an average which indicates if
the effluent is in compliance.
6. Unless specified, the sample(s) can be collected any time during
the averaging period.
7. If the data shows that the effluent concentration is below the
detection limit, the element will not be included in the moni-
toring program even if required by the current NPDES permit.
The remaining discussion gives the justification for this program by
element.
Aluminum
Aluminum is not recommented as part of the monitoring program
because it is not required by the NPDES permit.
Arsenic
Two samples per year are recommended. Both samples should be taken
during the period from May 1 to October 31. Monitoring of arsenic is
required by the NPDES permit and two samples show with 99 percent confi-
dence that the ash pond effluent does not increase the yearly average
receiving stream concentration above 0.001 mg/1. They also allow
estimation of the yearly average concentration in the effluent within 59
percent.
Cadmium
Cadmium is not recommended as part of the monitoring program
because the data during 1974 and 1975 indicated that the concentration
was below the detection limit.
Calcium
Calcium is not recommended as part of the monitoring program
because it is not required by the NPDES permit.
Chromium
Chromium is not recommended as part of the monitoring program
because the data during 1974 and 1975 and the data in this study
indicated the concentration was below the detection limit.
-------
-173-
Copper
One sample per year is recommended. Copper is required by the
NPDES permit and one sample shows that the ash pond effluent does not
increase the yearly average receiving stream concentration above
0.046 mg/1. It also allows the yearly mean in the effluent to be esti-
mated within 42 percent.
Dissolved Solids
Dissolved solids is not recommended as part of the monitoring
program because it is not required by the NPDES permit.
Iron
One sample per year is recommended. Iron is required by the NPDES
permit. The concentration of iron in the intake water exceeds the EPA
proposed water quality criteria for drinking water intake supplies;
however, one sample shows that the effluent does not increase the yearly
average receiving water concentration above 0.53 mg/1. It also allows
estimation of the yearly average concentration in the effluent within 58
percent.
Lead
Lead is not recommended as part of the monitoring program because
the data during 1974 and 1975 and the data in this study indicated the
concentration was below the detection limit.
Magnesium
Magnesium is not recommended as part of the monitoring program
because it is not required by the NPDES permit.
Manganese
One sample per year is recommended. Manganese is required by the
NPDES permit and one sample shows that the ash pond effluent does not
increase the yearly average receiving stream concentration above 0.11 mg/1.
It also allows the yearly mean in the effluent to be estimated within 45
percent.
Mercury
Mercury is not recommended as part of the monitoring program
because the data during 1974 and 1975 indicated that the concentration
was below the detection limit.
Nickel
Nickel is not recommended as part of the monitoring program because
the data during 1974 and 1975 and this study indicated that the concentra-
tion was below the detection limit.
-------
-174-
Four samples per year are recommended. Three of the samples should
be collected during the period from November 1 to April 30 and one
during the period from May 1 to October 31. The one sample during the
period May 1 to October 31 is sufficient to show that the effluent is in
compliance with the effluent limitation during the period. The remaining
three samples, if spaced out evenly over the period, are sufficient to
show that the effluent exceeds the limitation some time during the
period. These four samples will allow estimation of the yearly mean to
within 10 percent of the true mean.
Selenium
Four samples per year are recommended. Selenium is required by the
NPDES permit. The concentration in the intake water is consistently at
or below the minimum detectable limit of 0.002 mg/1, while the ash pond
effluent was higher than that in the intake water. Therefore, this one
sample allows estimation of the yearly average concentration in the
effluent within 46 percent. It also shows with 99 percent confidence
that the ash pond effluent does not increase the yearly average intake
concentration above 23 mg/1.
Sulfate
Sulfate is not recommended as part of the monitoring program
because it is not required by the NPDES permit.
Suspended Solids
Twenty-four samples per year at intervals of 15 days are recommended.
This frequency shows with 99 percent confidence that the effluent is in
compliance with the effluent limitation of 30 mg/1 specified in the
NPDES permit.
Zinc
Two samples per year are recommended. These two samples show that
the ash pond effluent does not increase the yearly average receiving
stream concentration above 0.019 mg/1. They also allow the yearly
average effluent concentration to be estimated within 66 percent of the
true mean.
SUMMARY
The example sampling program given in Table 44 requires a total of
35 analyses per year for 7 different elements whereas the NPDES permit
requires a total of 156 analyses for 12 different elements. Under the
recommended program, estimates of the yearly average were obtained for
the following elements: As, Cu, Fe, Mn, Se, suspended solids, and Zn.
The example program also excludes sampling for Cd, Cr, Pb, Hg, and Ni
which are required by the NPDES permit, because past data showed them to
be below the minimum detectable amount. The above totals exclude pH,
flow, and oil and grease.
-------
-175-
At the time of this writing the ash pond effluent at Plant J was
considered to be in compliance with existing effluent limitations as
defined in the NPDES permit for that plant. Special provisions or plant
modifications specific for Plant J may be required in the future to
ensure continued compliance.
-------
-176-
SECTION 7
FUTURE APPLICATIONS
The procedure for designing a monitoring program outlined in
Section 4 and demonstrated in Sections 4 and 5 for ash pond effluents
has several limitations. First, the procedure relies on maintaining the
same type of operating conditions in the future as were used during the
period when the design data set was collected. If at some time after
design of the monitoring program, operating conditions change which
result in changes in the effluent characteristics, the monitoring pro-
gram may no longer be valid. Therefore, the monitoring program should
be closely evaluated if changes in the operating conditions occur.
Second, the procedure depends heavily on the establishment of effluent
limitations. Therefore, limitations should be established with full
understanding of the consequences of not complying with them. Third,
the procedure was primarily designed to indicate compliance with an
effluent limitation. In those cases where the effluent was not in
compliance, application of the procedure was difficult. Fourth, the
procedure cannot be generically applied to all ash pond effluents, but
must be applied individually to each effluent.
The monitoring program which results from the use of this design
procedure results in a program which is quite dynamic, requiring frequent
reexamination and reevaluation of data and assumptions and redevelopment
of the effluent sampling program. This plus the limitations listed
above significantly limit the attractiveness of the procedure. However,
in spite of these limitations, application of the procedure to the
effluents at Plants E and J indicated that the sampling effort for trace
metals could be substantially decreased (from 70 to 90 percent). There-
fore, it may prove beneficial to apply the procedure to the remaining
TVA ash pond effluents. In addition, the procedure should also be applied
to oil and grease samples to see if their frequency cannot be reduced.
Since each ash pond effluent is equipped with a continuous flow measure-
ment device which supplys a permanent record of the flow, there is no
advantage to applying this procedure to flow measurements. There may be
enough data collected since June 1976, as part of the NPDES program to
supply a data base for these designs. However, as pointed out in
Section 3, there are several factors within the operation of a power
plant which affect the ash pond effluent water quality characteristics.
Since TVA is making or will be making modifications through April 1979
to its coal-fired power plants and ash ponds in an effort to meet
environmental regulations, the ash pond effluent characteristics may
change. Therefore, the NPDES data collected during this period,
June 1976 to April 1979, may not be representative of the effluent
characteristics after these modifications are complete. It may be
necessary to wait until after the modifications are complete before
implementing the procedure discussed here.
-------
-177-
If the reduction in the NPDES monitoring program, once the procedure
is implemented, is as significant for the entire TVA system as for
Plants E and J, the cost savings in routine monitoring should be directed
towards more short-term, extensive surveys or studies to better quantify
the effects of power plant operations on the ash pond effluent water
quality and development of better methods of treating or using water in
power plants.
-------
-178-
REFERENCES
1. "Development Document for Effluent Limitations Guidelines and New
Source Performance Standards for the Steam-Electric Power Generating
Point Source Category," U.S. Environmental Protection Agency, Report
No. EPA-440/l-74-029-a, (October 1974).
2. Freund, John E. Modern Elementary Statistics, 3rd Edition, Prentice-
Hall, Inc., Englewood Cliffs, NJ, 1967.
3. Standard Methods for the Examination of Water and Wastewater, 14th ed.,
American Public Health Association, American Water Works Association
and Water Pollution Control Federation, 1975.
4. U.S. Environmental Protection Agency, Handbook for Monitoring Industrial
Wastewater, August 1973.
5. Daniel, Wayne W. and James C. Terrell, Business Statistics Basic Concepts
and Methodology, Houghton Mifflin Company, Boston, MA, 1975.
6. Sherwani, Jabbar K. and David H. Moreau, Strategies for Water Quality
Monitoring, Water Resources Research Institute of the University of
North Carolina, June 1975. Report No. 107.
7. Berthouex, Paul M. and Dennis L. Meinert, Water Quality at Selected
Locations in the Tennessee Valley with Recommendations for Monitoring,
Tennessee Valley Authority, Chattanooga, Tennessee, 1977.
8. Miller, Irwin and John E. Freund, Probability and Statistics for
Engineers, Prentice-Hall, Inc., Englewood Cliffs, NJ, 1965.
9. Snedecor, George W. and William G. Cochran, Statistical Methods,
Sixth Edition, Iowa State University Press, Ames, Iowa, 1967.
10. Box, George E. P. and Gwilym M. Jenkins, Time Series Analysis
Forecasting and Control, Holden-Day, Inc., 500 Sansome Street,
San Francisco, CA, 1976.
11. Quality Criteria For Water, U.S. Environmental Protection Agency,
Washington, DC 20460, September 1976.
-------
-179-
APPENDIX A
-------
Parameter
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Aluminum
Arsenic
Cadmium
Chromium
Copper
Iron
Lead
Manganese
Magnesium
Calcium
Mercury
Nickel
Selenium
APPENDIX A
ANALYTICAL PROCEDURES USED BY REFERENCE
Procedure
Atomic absorption - Direct
Digestion and Colorimetric
Atomic absorption - Gaseous Hydride*
Atomic absorption - Extracted
Atomic absorption - Extracted
Atomic absorption - Direct
Atomic absorption - Direct
Atomic absorption - Extracted
Atomic absorption - Direct
Atomic absorption - Direct
Atomic absorption - Direct
Digestion and Flameless
Atomic absorption
Atomic absorption - Direct
Atomic absorption - Gaseous Hydride
Reference MDA(pg/l)
EPA, pp. 81, 92 200
SDDC SM, pp. 62, 65 5
EPA, pp. 81, 95 2
EPA, pp. 81, 89, 101 1
EPA, pp. 81, 89, 105 5
EPA, pp. 81, 108 10
EPA, pp. 81, 110 50
EPA, pp. 81, 89, 112 10
EPA, pp. 81, 116 10
EPA, pp. 81, 114 1000
EPA, pp. 81, 103 1000
EPA, p. 134 0.2
EPA, pp. 81, 141 50
EPA, p. 95 <1
00
o
I
-------
APPENDIX A (Continued)
Parameter Procedure Reference MDA(ng/l)
14- Silica, Dissolved Colorimetric-automated EPA, p. 274 100
Molybdosilicate Automated by TVA
(Technicon Auto Analyzer)
15. Zinc Atomic absorption - Direct EPA, pp. 81, 155 10
16. Residue, Total Gravimetric - Glass Fiber EPA, p. 266 10
Filterable Filtration
17. Residue, Total Gravimetric - Glass Fiber EPA, p. 268 1
Nonfilterable Filtration ,
00
EPA - Methods for Chemical Analysis of Water and Wastes, 1974, Environmental Protection Agency, Water V
Quality Office, Cincinnati, Ohio.
SM - Standard Methods for the Examination of Water and Wastewater, 13th Edition, 1971, American Public Health
Association, New York, New York.
*This procedure used for the analysis of all samples collected after October 12, 1976.
-------
-182-
APPENDIX B
QUALITY CONTROL DATA FOR TVA
WATER QUALITY LABORATORY
-------
-183-
TABLE B-l
SHORT TERM SINGLE OPERATOR DATA
BASED ON SEVERAL REPLICATES
ANALYZED AT LEAST THREE DIFFERENT CONCENTRATION LEVELS
Parameter
Cu
Zn
Cr
Ni
Pb
Hg
As*
As**
Cd
Se
Be
Sb***
Al
Ca
Fe
Mg
Ma
Si02
Residue, Total
Filterable
Residue, Total
Nonfilterable
Sulfate
Equation for
Standard Deviation
(So=Mx+b)
Concentration
Range & Units
0.00945 x +4.50
0.00652 x +2.93
0.0454 x +2.71
0.0133 x +8.82
0.00843 x +2.47
0.0163 x +0.079
•0.0211 x 1.68
0.0429 x +0.357
0.0106 x +0.395
0.0571 x +0.100
0.00184 x +3.92
0.002 x 70
10 -
11 -
20 -
226 -
15 -
1.13 -
10 -
2 -
0.9 -
5 -
47 -
5,000 -
536 MgA
519 Mg/1
110 |Jg/l
1150 Mg/1
149 Mg/1
5.71 Mg/1
48.5 Mg/1
10 Mg/1
21.7 Mg/1
20 Mg/1
515 Mg/1
15,000 Mg/1
0.0577 x +47.4
-0.00106 x +0.635
0.00985 x +6.34
0.0387 x -0.134
0.0155 x +3.96
0.0453 x -0.268
0.000 x +3.5
0.0334 x +0.864
0.0250 x +1.12
657 to 5,200 Mg/1
22.5 to 38.5 mg/L
220 to 2,150 M8/L
6.8 to 8.6 mg/L
29 to 547 Mg/L
7.4 to 11.2 mg/L
39 to 189 mg/L
4 to 84 mg/L
26 to 34 mg/L
Range
of Bias
0 to
-2 to
-3 to
+10 to
-26 to
+5 to
-3 to
-20 to
-10 to
-1 to
-6 to
-4 to
14%
10%
0%
+14%
+3%
+38%
0%
-3.6%
+14%
+1%
+3%
-3%
0 to 18%
-11 to -10%
6%
10%
12%
-1 to 2%
-6%
-3 to
-3 to
0 to
-22 to
Not Obtainable
-15 to 3%
*From 3/76 to 10/12/76 arsenic was analyzed by the silver diethyl
dithiocarbamate method.
**From 10/12/76 to present arsenic was analyzed by the gaseous hydride
method.
***Data from EPA manual.
-------
-184-
TABLE B-2
LONG-TERM QUALITY CONTROL CHART DATA
BASED ON OBSERVATIONS FROM MARCH 1976 TO JUNE 1977*
Parameter
Cu
Zn
Cr
Ni
Pb
Hg
As**
As***
Cd
Se
Be
Sb
Al
Ca
Fe
Mg
Mn
Silica
Residue, Total
Filterable
Residue, Total
Nonfilterable
Sulfate
Observations //
120
140
180
120
200
110
55
60
169
100
69
16
61
147
151
145
150
123
245
496
232
% Relative
Standard Deviation
Mean
Concentration
(M8/D
280
310
51
570
53
1.9
25
7.4
7.9
9.0
250
1,900
1,390
12.26
670
2.65
99
4.75
441
449
10.3
Mean
%RSD
0.96
0.98
98
26
22
28
4.77
,38
.63
.95
0.93
0.81
1.77
0.67
1.53
0.83
0.84
0.68
4.98
7.76
3.00
Average
% Bias
0.93
0.75
0.39
1.22
2.36
2.01
1.21
1.98
0.75
2.75
0.65
1.52
0.715
0.52
1.13
0.49
0.18
3.94
0.979
-1.10
*For the parameters below Sb, the data are based on observations
from 8/76 to 9/77.
**From 3/76 to 10/12/76 arsenic was analyzed by the silver diethyl
dithiocarbamate method.
***From 10/12/76 to present arsenic was analyzed by the gaseous
hydride method.
-------
-185-
TABLE B-3
COMPARISON OF SHORT-TERM SINGLE OPERATOR DATA WITH THAT
PREDICTED FROM LONG-TERM QUALITY CONTROL CHART DATA
Standard Deviation
Mean Value (Mg/D (Mg/D (M8/D
Parameter from Control Charts So Predicted**** So Found*****
Cu 280 7.14 2.69
Zn 310 4.95 3.04
Cr 51 5.02 2.03
Ni 570 16.4 12.9
Pb 53 2.92 2.77
Hg 1.9 0.110 0.118
As* 25 1.15 2.98
As** 7.4 0.674 0.398
Cd 7.9 0.479 0.208
Se 9.0 0.414 0.446
Be 250 4.39 2.33
Sb 1,900 73.4*** 15.4
Al 1,390 Mg/L 128 24.6
Ca 12.3 mg/L 0.622 0.082
Fe 670 Mg/L 12.9 10.3
Mg 2.65 mg/L 0.000 0.022
Mn 99 Mg/L 5.50 0.83
Si02 4.75 mg/L 0.000 0.032
Residue, Total
Filterable 441 mg/L 3.5 23.0
Residue, Total
Nonfilterable 449 mg/L 15.9 34.8
S04 10.3 mg/L 1.38 0.31
*From 3/76 to 10/12/76 arsenic was analyzed by the silver diethyl
dithiocarbamate method.
**From 10/12/76 to present arsenic was analyzed by the gaseous
hydride method.
***Data from EPA manual.
****So predicted is found by using mean value from control charts to
solve equation for standard deviation for short-term single operator
data in Table I.
**#**So found is product of long-term RSD and mean value from control
charts.
-------
-186-
APPENDIX C
-------
-187-
APPENDIX C
WATER QUALITY CRITERIA FOR DOMESTIC WATER
SUPPLY INTAKES PROPOSED BY EPA
Element
Reference
Aluminum
Arsenic
Barium
Beryllium
Cadmium
Chloride
Chromium
Copper
Iron
Lead
Manganese
Mercury
Nickel
Selenium
Silver
Sulfate
Zinc
Domestic Water
Supply (EPA)
me/1
No criteria
0.05
1.0
No criteria
0.01
250
0.05
1.0
0.3
0.05
0.05
0.002
No criteria
0.01
0.05
250
5
-------
-188-
APPENDIX D
-------
-189-
APPENDIX D
STUDENT t VALUES1
degrees
n of freedom 99% CI 95% CI 80% CI
12 11 3.106 2.201 1.363
13 12 3.055 2.179 1.356
20 19 2.861 2.093 1.328
21 20 2.845 2.086 1.325
25 24 2.797 2.064 1.318
32 31 2.745 2.040 1.309
33 32 2.741 2.038 1.309
34 33 2.736 2.036 1.308
1. Taken from CRC Standard Mathematical Tables, 19th ed., edited by
Samuel M. Selby, The Chemical Rubber Co., 18901 Cranwood Parkway,
Cleveland, Ohio 44128, page 610.
-------
-190-
APPENDIX E
-------
-191-
APPENDIX E
EXAMPLE CALCULATION OF THE ALLOWABLE INPUT TO THE STREAM AND THE
ASSOCIATED PRECISION ASSUMING THE MAXIMUM AVERAGE ALLOWABLE CON-
CENTRATION IN THE STREAM IS EQUAL TO OR LESS THAN THE EPA WATER
QUALITY CRITERIA
Element: As
Sample Mean in the Effluent: 0.017 mg/1
Average Concentration in the
Receiving Stream: 0.004 mg/1
Water Quality Criteria: 0.05 mg/1
Maximum Ash Pond Flow: 67 cfs
7-day Minimum Flow in the
Receiving Stream: 7880 cfs
Maximum average allowable _ 7880(0.05) - 7813 (0.004)
concentration in the ~~67
effluent
= 5.4
Required Precision = 5.4 - 0.017
= 5.383
-------
-192-
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-236
t. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Design of a Monitoring Program for Ash Pond
Effluents
5. REPORT DATE
November 1979
i. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
F.A. Miller, HI, T.V.J. Chu, andR.J. Ruane
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Tennessee Valley Authority
1140 Chestnut Street, Tower H
Chattanooga, Tennessee 37401
10. PROGRAM ELEMENT NO.
INE624A
II. CONTRACT/GRANT NO.
IAG-D5-E-721
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PE
Final; 5/75 - 3/79
PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/13
15.SUPPLEMENTARY NOTES IERL-RTP project officer is Michae C. Osborne, Mail Drop 61
919/541-2915. '
16. ABSTRACT
The report describes a procedure for designing an effective monitoring
program for fossil-fueled power plant ash pond effluents. Factors that influence
effluent characteristics and are important in designing such a monitoring program
were determined following a review of plant operating characteristics and ash pond
effluent characteristics of TVA's fossil-fueled power plant system. A statistical
procedure for determining the sampling frequency of chemical characteristics in
ash pond effluents was then developed. Two ways to determine precision are descri-
bed: Method 1 involves selecting a precision value to estimate the population mean
within a given percentage; Method 2 involves calculating a precision value by subtrac-
ting an estimate of the population mean from either the ash pond effluent limitation
established by EPA or a desirable water quality criterion. Method 2 gives the num-
ber of samples required to show that the effluent is in compliance with the effluent
limitation or below the water quality criteria. The method chosen to compute the
precision depends on the purpose of the monitoring program. The procedure was
demonstrated for two TVA ash pond systems.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Pollution
Monitors
Design
Ponds
Fly Ash
Effluents
Waste Disposal
Electric Power
Plants
Fossil Fuels
Chemical Analysis
Water Quality
b.lDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Ash Ponds
c. COSATI Field/Group
13B
14B
08H
2 IB
10B
2 ID
07D
Release to Public
Unclassified
(This Report)
21. NO. OF PAGES
205
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |