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)

-------