United States
             Environmental Protection
             Agency
             Office Of Water
             (4203)
EPA 833-B-99-002
August 1999
vyEPA
Toxicity Reduction
Evaluation Guidance For
Municipal Wastewater
Treatment Plants

-------

-------
                                        EPA/833B-99/002
                                           August 1999
  Toxicity Reduction Evaluation Guidance
for Municipal Wastewater Treatment Plants
              Office of Wastewater Management
            U.S. Environmental Protection Agency
                 Washington, D.C. 20460

-------
                                Notice and Disclaimer

The U.S. Environmental Protection Agency, through its Office of Water, has  funded, managed, and
collaborated in the development of this guidance, which was prepared under order 7W-1235-NASX to
Aquatic Sciences Consulting; order 5W-2260-NASA to EA Engineering, Science and Technology, Inc.; and
contracts 68-03-3431,  68-C8-002, and 68-C2-0102 to Parsons Engineering Science, Inc.  It has been
subjected to the Agency's peer and administrative review and has been approved for publication.

The statements in this document are intended solely as guidance. This document is not intended, nor can it
be relied on, to create any rights enforceable by any party in litigation with the United States. EPA and State
officials may decide to follow  the guidance provided in this document, or to act at variance with the
guidance, based on an analysis of site-specific circumstances. This guidance may be revised without public
notice to reflect changes in EPA policy.

-------
                                         Foreword

This document is intended to provide guidance to permittees, permit writers, and consultants on the general
approach and procedures for conducting toxicity reduction evaluations (TREs) at municipal wastewater
treatment plants. TREs are important tools for Publicly Owned Treatment Works (POTWs) to use to identify
and reduce or eliminate toxicity in a wastewater discharge. TREs may be required by the discharger's
National Pollutant Discharge Elimination System (NPDES) permit or through state or federal enforcement
actions.  Dischargers can use the guidance to evaluate the nature and sources of effluent toxicity before a
TRE becomes a regulatory requirement. Whether the TRE  is voluntary or mandated, this guidance can be
helpful in preparing and executing a plan to address effluent toxicity.

This guidance describes the general approaches that have been successfully used in municipal TREs. Each
TRE will be different; therefore, the strategy for conducting TREs should be tailored to address site-specific
conditions. The components of a TRE may include the collection and review of pertinent data; an evaluation
of the treatment facility to identify conditions that may contribute to effluent toxicity; identification of
effluent toxicants using toxicity identification evaluation (TIE) procedures (USEPA 1991a, 1992a, 1993a,
1993b, 1996); location of the sources of toxicants and/or toxicity using chemical analysis or refractory
toxicity assessment (RTA) procedures; and the evaluation, selection, and implementation of toxicity control
measures.  Dischargers are encouraged to develop a TRE plan that describes the initial  components to
perform in the TRE. Following initial testing, the results can be used to provide direction for further testing
to identify the cause(s) and source(s) of toxicity and evaluate and select methods for toxicity control.

This document is an update of the municipal TRE protocol that was published in 1989 (USEPA, 1989a).
Much experience has been gained since 1989, including the use of a number of freshwater and estuarine/
marine species in acute and chronic TRE studies and the development of additional procedures for TIE and
RTA studies. In most cases, the approaches and methods described in the municipal TRE protocol have been
validated through TRE studies and other municipal TREs (Amato et al., 1992; Bailey et al, 1995; Botts et
al., 1990, 1992, 1993, 1994; Collins et al., 1991; Fillmore  et al., 1990; Lankford and Eckenfelder, 1990;
Morris et al., 1990, 1992). Important lessons have been learned and this information has been incorporated
in this guidance where  possible.   Additions to this guidance include considerations in evaluating the
operation and performance of current publicly owned treatment works (POTW) technology, descriptions of
current TIE procedures for acute and short-term chronic toxicity (USEPA  199la, 1992a, 1993a, 1993b,
1996), updated methods for tracking sources of acute and chronic toxicity in POTW  sewer collection
systems, and additional recent case studies on acute and chronic TREs using freshwater and estuarine/marine
species.  Information is also provided on sampling  requirements, equipment and facilities, quality
assurance/quality control, and health and safety.

The updated TIE guidance procedures  are important tools  for conducting TREs including  Toxicity
Identification Evaluation: Characterization of Chronically Toxic Effluents,  Phase I (USEPA, 1992a),
Methods for Aquatic Toxicity  Identification Evaluations: Phase I Toxicity Characterization Procedures,
Second Edition (USEPA, 199la), Marine Toxicity Identification Evaluation (TIE) Guidance Document,
Phase I  (USEPA, 1996), Methods for Aquatic Toxicity Identification Evaluations: Phase II  Toxicity
Identification Procedures for Samples Exhibiting Acute and Chronic Toxicity (USEPA, 1993 a), and Methods
for Aquatic Toxicity Identification Evaluations: Phase III Toxicity Confirmation Procedures for Samples
Exhibiting Acute and Chronic Toxicity (USEPA, 1993b).  The acute and chronic whole effluent toxicity
testing manuals should also be reviewed during the TRE process (USEPA 1993c, 1994a, 1994b, 1995). These
manuals  describe procedures for the toxicity tests that are the core of the TREs.
                                               in

-------
                                         Contents
Foreword 	iii
Acronyms, Abbreviations, and Symbols	viii
Acknowledgments	 xii

Section 1   Introduction 	  1
           Background 	  1
           TRE Goals and Objectives	  3
           Components of the Municipal TRE	  3
           Limitations of the TRE Guidance  	  5
           Organization of the TRE Guidance 	  6

Section 2   Information and Data Acquisition	  7
           Introduction	  7
           Review of Effluent Toxicity Data	  7
           Toxicants Identified in POTW Effluents	  8
           POTW Design and Operations Data  	  10
           Pretreatment Program Data  	  12

Section 3   Facility Performance Evaluation 	  15
           Introduction 	  15
           POTW Performance Evaluation	  15

Section 4   Toxicity Identification Evaluation	 29
           Introduction 	 29
           Toxicity Tests	 29
           TIE Procedures	 32
           Additional Characterization Procedures for Evaluating the Effect of Ion Composition ... 34
           TDS Toxicity	 34
           Ion Imbalance	 35

Section 5   Toxicity Source Evaluation	 38
           Introduction 	 38
           Sampling Approach	 40
           Sampling Conditions  	 40
           Chemical-Specific Investigation 	 40
           Refractory Toxicity Assessment 	 41
           POTW Wastewater Profile	 43
           Findings of the Toxicity Source Evaluation	 51

Section 6   Toxicity Control Evaluation	 53
           Introduction	 53
           Identifying Toxicity Control  Options	 53
           Pretreatment Control Evaluation	 56
           In-Plant Control Evaluation	 57
           Toxicity Control Selection	 61
                                             IV

-------
                                 Contents (continued)

Section 7   Toxicity Control Implementation	  64
           Introduction  	  64
           Implementation	  64
           Follow-Up Monitoring	  64

Section 8   Quality Assurance/Quality Control	  65
           Introduction  	  65
           Sample Collection and Preservation  	  65
           Chain-of-Custody	  66
           TRE Procedures	  66
           Equipment Maintenance	  68
           Documentation and Reporting of Data	  68
           Corrective Action	  68

Section 9   Health and Safety	  69
           Introduction  	  69
           Sample Collection and Handling	  69
           TRE Methods	  69
           General Precautions	  70

Section 10  Facilities and Equipment	  71
           Introduction  	  71
           Toxicity Identification Evaluations 	  71
           Refractory Toxicity Assessment and Treatability Tests  	  71
           General Analytical Laboratory Equipment	  72

Section 11  Sample Collection and Handling	  73
           Introduction  	  73
           Sampling Location	  73
           POTW Sampling	  73
           Sewer Discharge Sampling	'.	  74

Section 12  References 	  75

Section 13  Bibliography	  82

                                       Appendices

A  Original Case: Histories: Lesson Learned	  84
B  TRE Case Study:  Central Contra Costa Sanitary District, Martinez, California  	  93
C  TRE Case Study:  City of Reidsville, North Carolina	  98
D  TRE Case Study:  City of Durham, North Carolina  	  105
E  TRE Case Study:  Michigan City Sanitary District, Indiana	  113
F  TRE Case Study:  Central Contra Costa Sanitary District, Martinez, California, and Other
    San Francisco Bay Area POTWs	  119
G  TRE Case Study:  Linden Roselle Sewerage  Authority, New Jersey 	  133
H  Toxicity Control Options for Organophosphate Insecticides	  142
I   Pretreatment Program Chemical Review	  147
J   Refractory Toxicity Assessment Protocol: Step-by-Step Procedures	  153

-------
                                          Figures

Figure 1-1.  TRE flow diagram for municipal wastewater treatment plants  	  4
Figure 3-1.  Flow diagram for a facility performance evaluation	  16
Figure 4-1.  Flow diagram of a toxicity identification evaluation	  30
Figure 5-1.  Flow diagram for a toxicity source evaluation	  39
Figure 5-2.  Schematic of a refractory toxicity assessment test	  42
Figure 5-3.  Theoretical results of inhibition testing 	  51
Figure 6-1.  Flow diagram for a toxicity control evaluation	  54
Figure A-l.  Acute LC50 of Hollywood effluent versus diazinon concentration	  86
Figure A-2.  Correlation of diazinon TUs versus whole effluent TUs 	  87
Figure A-3.  Correlation of diazinon and CVP TUs versus whole effluent TUs  	  87
Figure C-l.  Results of RTA (rounds 1 and 2)	  101
Figure C-2.  Results of RTA (round 3)	  102
Figure D-l.  Flow diagram for wastewater treatment situations	  107
Figure E-l.  Acute and chronic effluent toxicity: 1991  through 1992	  117
Figure F-l.  Effluent TUs versus diazinon TUs in the CCCSD effluent samples 	  122
Figure F-2.  Percent mass contribution to sources to the CCCSD influent 	  125
Figure F-3.  Mean diazinon and chlorpyrifos concentrations (ฑstd) in influent and effluent from
            three Bay Area POTWs	  127
Figure F-4.  Mean chlorpyrifos and diazinon concentrations (ฑstd) in influent and effluent from
            nine Bay Area POTWs during August 1997	  128
Figure H-l.  Diazonin removal as a function of SRT, HRT, and MLSS concentration	  145
                                             VI

-------
                                         Tables

Table 2-1.  Toxicants Identified in POTW Effluents	r		  9
Table 2-2.  Example POTW Design and Operation Data	  11
Table 2-3.  Example Pretreatment Program Data	  14
Table 5-1.  POTW Wastewater Profile Analyses for a Refractory Toxicity Assessment	  44
Table 5-2.  Example of Bracketing the LC50 Concentration in the RTA Sewer Wastewater Test  ...  46
Table 5-3.  Comparison of Control Test and Industrial Wastewater Spiked Test Results	  46
Table 6-1.  An Example of the Comparison of In-Plant Ammonia Treatment Alternatives	  55
Table 6-2.  POTW In-Plant Control Technologies for Categories of Toxic Compounds	  58
Table A-l.  Comparison of Whole Effluent TUs and Methanol Fraction TUs	  88
Table A-2.  Summary of TIE Phase H Results	  89
Table B-l.  Summary of Results of Phase I TIE Conducted on Two Effluent Samples with
           D. excentricus	  94
Table B-2.  NOECs Obtained for D. excentricus and 5. purpuratus Exposed to Different Metals  ...  95
Table B-3.  Comparison of Effluent Concentration of Selected Metals with NOECs Derived from
           Laboratory Studies with D.  excentricus	:	  95
Table B-4.  Comparison of NOECs, LOECs, and Percent Fertilization Obtained with D. excentricus
           Exposed to Effluent and Seawater Spiked with Cu	  96
Table B-5.  Percent Fertilization Obtained with D. excentricus Exposed to Effluent and Effluent
           Spiked with Cu	  96
Table C-l.  Chronic Toxicity of Raw Industrial Wastewaters	  99
Table C-2.  Description of Industries Evaluated in the RTA	  99
Table C-3.  Comparison of Operating Conditions for the City of Reidsville POTW Processes
           and RTA Simulation Tests  	 100
Table D-l.  Farrington Road and Northside Simulation Operating Conditions	 108
Table D-2.  Comparison of Calibration Test Results to Permit Limitations and Design Criteria	 109
Table D-3.  Total Phosphorus Results for the Calibration Tests Conducted on April 10-11, 1990  .. 109
Table D-4.  Comparison of Simulation Test Results to Performance Criteria	 Ill
Table D-5.  Toxicity of Simulation Effluents to C. dubia	 Ill
Table E-l.  Acute Toxicity Characterization Test Results from April 1991 Through June 1991	 114
Table E-2.  Toxicity Characterization Test Results form July 1991 Through October 9, 1991	 115
Table E-3.  Acute and Chronic Toxicity of MCWTP' s Effluent (with and without added EDTA)
           from October 1991 Through January 1992	,	 116
Table F-l.  Matrix of Results of Phase I TIE Conducted on Five Effluent Samples with C. dubia .. 121
Table F-2.  Summary of TIE Phase H Results	 121
Table F-3.  Summary of Follow-Up TIE Studies  	 123
Table F-4.  Diazinon and Chlorpyrifos  concentrations in Wastewater Samples from Selected
           Residential and Commercial Sources in the CCCSD	 124
Table G-l.  TIE Phase m Results:  Non-Polar Organic Compound Confirmation (LRSA POTW) .. 135
Table G-2.  Results of Refractory Toxicity Assessment, July and October 1993	 137
Table 1-1.  PPCR Data Sheet  	 148
Table 1-2.  Data Sheet for Regression Analysis	 149
Table 1-3.  Summary of the PPCR Chemical Optimization Procedures	 150
                                             Vll

-------
                     Acronyms, Abbreviations, and Symbols

AA           Atomic absorption
AGO         Administrative consent order
Ag           Silver
Alum         Aluminum sulfate
A/Oฎ         Patented biological nutrient removal process
AQUIRE      Aquatic Information Retrieval Toxicity Data Base
APE         Alkylphenol ethoxylates
ASCE        American Society of Civil Engineers
ASM         American Society for Microbiology
ASTM        American Society for Testing and Materials
ATP         Adenosine triphosphate
BNR         Biological nutrient removal
BOD         Biochemical oxygen demand
BOD5         Five-day biochemical oxygen demand
BOD^)        Twenty-day biochemical oxygen demand
ฐC           Degrees centigrade
CaCO3        Calcium carbonate
CADPR       California Department of Pesticide Registration
Cb           Concentration of specified analyte in filtrate of return activated sludge biomass
CCCSD       Central Contra Costa Sanitary District
C. dubia       Ceriodaphnia dubia, cladoceran
Cd           Cadmium
Ce           Effluent concentration of the specified analyte
CERCLA      Comprehensive Environmental Response, Compensation, and Liability Act
CFR         Code of Federal Regulations
ChV         Chronic value, geometric mean of the no observed effect concentration (NOEC) and
              lowest observed effect concentration (LOEC)
Ci            Influent concentration of the specified analyte
CO2          Carbon dioxide
COG         Chain-of-custody
COD         Chemical oxygen demand
Cpe           Concentration of specified analyte in primary effluent
Cr            Chromium
CSO         Combined sewer overflow
CTAS         Cobalt thiocyanate active substances
Cu           Copper
CuSO4        Copper sulfate
cu ft/lb        Cubic feet per pound
CV           Coefficient of variation
CVP         Chlorfenvinphos
Cw           Concentration of specified analyte in sewer wastewater
CWA         Clean Water Act
CIS           Carbon-based resin used in solid phase extraction (SPE) columns to evaluate for the
              presence of nonpolar organic toxicants
D. excentricus  Dendraster excentricus, sand dollar
D. magnet      Daphnia magna, cladoceran
D. pulex       Daphnia pulex, cladoceran
DMR         Discharge monitoring report
                                           vni

-------
             Acronyms, Abbreviations, and Symbols (continued)
DO
EC50
ECOTOX
EDTA
ELISA
EXTOXNET
F/M
F/Mb
Fw
g/L
gpd/sq ft
gpm/sf
GAC
GC
GC/MS
H&S
Hg
HNO3
H202
HPLC
HRT
ICP
ICp
IWC
IWS
L
LC50
LRSA
LOEC
m3/m2/day
MB AS
MCRT
MCWTP
mg/L
mg O2/hr/g
mg O2/L/
             Dissolved oxygen
             Effective concentration causing a 50% effect in the test species
             Ecotoxicology Database Retrieval System
             Ethylenediaminetetraacetate
             Enzyme-linked immunosorbent assay
             Extension Toxicology Network
             Food to microorganism ratio
             Food to microorganism ratio
             Flow concentration factor for RTA testing
             Grams per liter
             Gallons per day per square foot
             Gallons per minute per square foot
             Granular activated carbon
             Gas chromatography
             Gas chromatograph/mass spectrophotometer
             Health and safety
             Mercury
             Nitric acid
             Hydrogen  peroxide
             High-performance liquid chromatography
             Hydraulic  retention time
             Inductively coupled plasma spectrometry
             Inhibition  concentration causing a percent effect (p) in the test species (e.g., IC25, IC50)
             Instream waste concentration
             Industrial waste survey
             Liter
             Lethal concentration causing a 50% mortality in exposed test organisms
             Linden Roselle Sewerage Authority
             Lowest observed effect concentration
             Cubic meters per square meter per day
             Methylene blue active substances
             Mean cell residence time
             Michigan  City Wastewater Treatment Plant
             Milligram per liter
             Milligram dissolved oxygen per hour per gram
             Milligram dissolved oxygen per liter per gram mixed liquor volatile suspended solids
 gMLVSS/min per minute
M. bahia     Mysidopsis bahia, mysid shrimp
             Milliliter
             Mixed liquor suspended solids
             Mixed liquor volatile suspended solids
             Mass spectrometry
             Material safety data sheet
             million gallons per day
              Sodium chloride
             Disodium phosphate
             Monosodium phosphate
              North Carolina Division of Environmental Management
 mL
 MLSS
 MLVSS
 MS
 MSDS
 mgd
 NaCl
 Na2HPO4
 NaH2PO4
 NCDEM
                                             IX

-------
              Acronyms, Abbreviations, and Symbols (continued)

 NETAC      USEPA' s National Effluent Toxicity Assessment Center
 NH3          Un-ionized ammonia
 NH4*         Ammonium ion
 NH3-N        Total ammonia as nitrogen
 Ni           Nickel
 NIOSH       National Institute of Occupational Safety and Health
 NJDEP       New Jersey Department of Environmental Protection
 NOEC        No observed effect concentration
 NO3-N        Nitrate as nitrogen
 NJPDES      New Jersey Pollutant Discharge Elimination System
 NPDES       National Pollutant Discharge Elimination System
 NPO         Non-polar organic
 NTIS         National Technical Information Service
 O2           Oxygen
 O. mykiss     Oncorhynchus mykiss, trout
 OSHA        Occupational Safety and Health Administration
 OUR         Oxygen uptake rate, a measure of the activity of activated sludge biomass
 P            Phosphorus
 PAC          Powdered activated carbon
 Pb           Lead
 PBO          Piperonyl butoxide, a metabolic blocking agent used to detect for the presence of
              organophosphate insecticides
 pH           Logarithm of the reciprocal of the hydrogen ion concentration
 PO4-P         Orthophosphate as phosphorus
 POTW        Publicly owned treatment works
 P. promelas    Pimephales promelas, fathead minnow
 PPCR         Pretreatment program chemical review
 Qi           POTW influent flow rate
 Qw           Sewer wastewater flow rate                           !
 QA/QC       Quality assurance/quality control
 r value        Correlation coefficient
 RAS          Return activated sludge
 RBC          Rotating biological contactor
 RCRA        Resource Conservation and Recovery Act
 RREL         Risk Reduction Engineering Laboratory
 RTA          Refractory toxicity assessment
 RWQCP      Regional Water Quality Control Plant
 SARA         Superfund Amendments and Reauthorization Act
 SBOD5        Five-day soluble biochemical oxygen demand
 SCOD         Soluble chemical oxygen demand
 sf            square foot
 SIC           Standard industrial classification
 SNAP         Sewer network analysis program
 SNH3-N       Soluble ammonia - nitrogen
SO2           Sulfur dioxide
 SOP          Standard operating procedures
 SOR          Surface overflow rate
SOUR         Specific oxygen uptake rate

-------
             Acronyms, Abbreviations, and Symbols (continued)

S. purpurtus    Strongylocentrotus purpurtus
SP            Soluble phosphorus
SPI           Simulated plant influent
SPE          Solid phase extraction
SRT          Sludge retention time
SS            Suspended solids
SSO          Sanitary sewer overflow
SSUR         Specific substrate utilization rate
SVI          Sludge volume index
TBOD5        Total five-day biochemical oxygen demand
TCIP         Toxics control implementation plan
TCLP         Toxicity characteristic leaching procedure
TDS          Total dissolved solids
TIE          Toxicity identification evaluation
TKN         Total Kjeldahl nitrogen
TMP         Toxicity management program
TOC         Total organic carbon
TP           Total phosphorus
TRC         Total residual chlorine
TRE         Toxicity reduction evaluation
TSD         Technical Support Document
TSDF         Treatment, storage, and disposal facility
TSS          Total suspended solids
TU           Toxic unit
TUa          Acute toxic unit
TUc          Chronic toxic unit
ug/L         Microgram per liter
jam          Micron
UMVUE      Uniformly Minimum Variance Unbiased Estimator
USD         Union Sanitary District
USEPA       United States Environmental Protection Agency
UV          Ultraviolet
UV-VIS       Ultraviolet - visible spectrophotometer
Vb           Volume of biomass for RTA testing
Vr           Reactor volume for RTA testing
Vnb          Volume of non-toxic biomass for RTA testing
Vpe          Volume of primary effluent for RTA testing
Vsw         Volume of synthetic wastewater for RTA testing
Vw          Volume of sewer wastewater for RTA testing
VSS         Volatile suspended solids
WEF         Water Environment Federation
WPCF        Water Pollution Control Federation (currently Water Environment Federation)
WWTP       Wastewater treatment plant
xg           Times gravity
Zn           Zinc
ZSV         Zone setting velocity
                                             XI

-------
                                   Acknowledgments

 This guidance was prepared through the cooperative efforts of USEPA; Aquatic Sciences Consulting; EA
 Engineering, Science and Technology, Inc.; Parsons Engineering-Science, Inc.; and Burlington Research,
 Inc. The principal authors of the municipal TRE protocol (USEPA, 1989a) were John A. Botts, Jonathan
 W. Braswell,  Jaya Zyman, William L. Goodfellow, Samuel B. Moore, and Dolloff F. Bishop. Although
 much of the information presented in the TRE protocol (USEPA, 1989a) is included herein, the primary
 authors of this guidance—John A. Botts, William L. Goodfellow, Mark A. Collins, Timothy L. Morris, and
 Richard A. Diehl—have updated the TRE protocol (USEPA, 1989a) using information from recent studies
 and guidance.  Contributors to specific sections of this guidance are listed below.

 Sections 1 through 7 - Main Guidance:

 John A. Botts, Aquatic Sciences Consulting, Woodbine, Maryland
 Jonathan W. Braswell, Parsons Engineering-Science, Inc., Fairfax, Virginia
 Timothy L. Morris, ENTRIX, Inc., Wilmington, Delaware
 William L. Goodfellow, EA Engineering, Science and Technology, Inc., Sparks, Maryland
 Dolloff F. B ishop, USEPA, Cincinnati, Ohio
 Elizabeth Sullivan, CH2M Hill, Reston, Virginia
 H. Jeffrey Elseroad, EA Engineering, Science and Technology, Inc., Sparks, Maryland
 Samuel B. Moore, Burlington Research, Inc., Burlington, North Carolina
 Richard A. Diehl, Burlington Research, Inc., Burlington, North Carolina
 John Cannell, former USEPA employee, Portland, Oregon

 Sections 8 through 11 - Laboratory Guidelines:

 John A. Botts, Aquatic Sciences Consulting, Woodbine, Maryland
 Elaine Wilson, Parsons Engineering-Science, Inc., Fairfax, Virginia
 Jaya Zyman, CH2M Hill, Reston, Virginia

 Appendices A  through G - TRE Case Summaries:

 John A. Botts,  Aquatic Sciences Consulting, Woodbine, Maryland
 Mark A. Collins, Parsons Engineering-Science, Inc., Fairfax, Virginia
 Jeffrey Miller, AQUA-Science,  Inc., Davis, California
 Timothy L. Morris, ENTRIX, Inc., Wilmington, Delaware
 Lauren Fillmore, Parsons Engineering-Science, Inc., Fairfax, Virginia
 Mary Welch, Science Applications International Corporation, Denver, Colorado
 James Salisbury, Parsons Engineering-Science, Inc., Fairfax, Virginia
 Christina L. Cooper, Aquatic Sciences Consulting, Woodbine, Maryland
William Clement, Great Lakes Environmental Center, Columbus, Ohio
Bart Brandenburg, Central Contra Costa Sanitary District, Martinez, California
Bhupinder Dahliwal, Central Contra Costa Sanitary District, Martinez, California
Jim Kelly, Central Contra Costa Sanitary District, Martinez, California
Jerry Rothrock, City of Reidsville, Reidsville, North Carolina
A.T. Rolan, City of Durham, Durham, North Carolina
W.W. Sun, City of Durham, Durham, North Carolina
Vicki Westbrook, City of Durham, Durham, North Carolina
Susan Turbak,  City of Durham, Durham, North Carolina
Gary Fare, Linden Roselle Sewage Authority, Linden, New Jersey
                                             xn

-------
Judy Spadone, Linden Roselle Sewage Authority, Linden, New Jersey

Appendix H - Toxicity Control Options for Organophosphate Insecticides

John A. Botts, Aquatic Sciences Consulting, Woodbine, Maryland
Carlos Victoria-Rueda, Parsons Engineering-Science, Inc., Austin, Texas
Lauren Fillmore, Parsons Engineering-Science, Inc., Fairfax, Virginia
Dennis Tierney, Novartis Crop Protection, Greensboro, North Carolina

Appendix I - Pretreatment Program Chemical Review:

Richard A. Diehl, Burlington Research, Inc., Burlington, North Carolina
Samuel B. Moore, Burlington Research, Inc., Burlington, North Carolina

Appendix J - Refractory Toxicitv Assessment Protocol: Step-bv-Step Procedures

John A. Botts, Aquatic Sciences Consulting, Woodbine, Maryland
Timothy L. Morris, ENTRDC, Inc., Wilmington, Delaware
Mark A. Collins, Parsons Engineering-Science, Inc., Fairfax, Virginia

The following individuals are gratefully acknowledged for their technical review and advice on this
guidance:

Stephen Bainter, USEPA, Region VI, Dallas, Texas
Stephen L. Bugbee, retired USEPA employee, Petersburg, Pennsylvania
Debra Denton, USEPA, Region K, San Francisco, California
Teresa J. Norberg-King, USEPA, Duluth, Minnesota
Laura Phillips, USEPA Office of Wastewater Management, Washington, D.C.
Donna Reed-Judkins, USEPA Office of Science and Technology, Washington, D.C.
Carl Potter and Richard Dobbs - USEPA Office of Research and Development, Cincinnati, Ohio
William J. Rue, EA Engineering, Science and Technology, Inc., Sparks, Maryland
G. Michael DeGraeve - Great Lakes Environmental Center, Traverse City, Michigan
Sheila Frace, USEPA Office of Science and Technology, Washington, D.C.
Jeffrey Lape, USEPA Office of Wastewater Management, Washington, D.C.
Linda Anderson-Carnahan, USEPA, Region IV, Atlanta, Georgia
Donald I. Mount, AScI Corporation, Duluth, Minnesota
Carlos Victoria-Rueda, Parsons Engineering-Science, Inc., Austin, Texas

Stephen Bainter, Debra Denton, and Teresa J. Norberg-King were technical advisors for the preparation of
this guidance (order 7W-1235-NASX to Aquatic Sciences Consulting). Ms. Laura Phillips and Mr. Stephen
Bugbee served as USEPA Technical Project Managers under order 7W-1235-NASX to Aquatic Sciences
Consulting and order 5W-2260-NASA to EA Engineering, Science and Technology, Inc., respectively.

Peer review was conducted following the USEPA's Science Policy Council Handbook  for Peer Review
(January 1998). The review comments are gratefully acknowledged and the changes were incorporated, as
appropriate.  USEPA's Office of Water, Office of Wastewater  Management provided support for  the
development of this guidance.
                                              xm

-------

-------
                                            Section 1
                                         Introduction
Background
The Clean Water Act (CWA) (United States Federal
Water Pollution Control Act Amendments, Public Law
92-500  of  1972) prohibits the discharge  of  "toxic
pollutants in toxic amounts."   In the CWA, the
mechanism for regulating discharges to the Nation's
waterways   is  the National  Pollutant  Discharge
Elimination System (NPDES). Permits issued under
NPDES  may  contain  effluent  limits and   other
requirements based on ambient water quality standards
for the protection of aquatic life and human health.
The water quality-based approach applies criteria for
both chemical specific parameters and whole effluent
toxicity to ensure that toxic pollutants are controlled
and water quality standards are maintained (Federal
Register 23868,  1989).  This integrated approach to
water quality protection is described  in  detail in
USEPA's Technical Support  Document for  Water
Quality-Based Toxics Control (hereafter referred to as
the TSD, 199 Ib).

"Whole effluent toxicity" refers to the results of acute
and  chronic aquatic toxicity tests used to monitor
discharges  to surface waters.  Acute  toxicity is a
measure of primarily lethal effects that occur  over a
short period of time (i.e., 96 hours or less). Chronic
toxicity refers to sublethal effects, such as inhibition of
fertilization, growth, and reproduction that occur over
a longer exposure period (e.g., 7 days). Acute and
chronic effects to ;aquatic species are measured using
standard procedures (40 CFR  136.3) as specified in
NPDES permits. USEPA has published manuals that
describe the toxicity test methods for freshwater and
estuarine/marine organisms (USEPA 1993c,  1994a,
1994b,   1995).   On  October 26,  1995,  USEPA
promulgated a final rule under the CWA that adds
whole effluent toxicity testing methods to  the list of
nationally applicable methods in 40 Code of Federal
Regulations (CFR) Part 136.  These methods can be
accessed electronically along with all other approved
analytical methods on CD-ROM (USEPA, 1997).

Effluents from permitted facilities are monitored, and
where a reasonable potential exists to exceed numeric
toxicity criteria,  NPDES permit limits  for  whole
effluent  toxicity   are   established  (40   CFR
122.44(d)(l)(iv)). Whole effluent toxicity limits may
also be established where there is reasonable potential
to exceed a narrative toxicity criterion in the receiving
water (40 CFR 122.44(a)(l)(v)). A toxicity reduction
evaluation (TRE) may be used to identify and reduce
or eliminate sources of effluent toxicity whether or not
there are whole effluent toxicity limits in the NPDES
permits. For example, where a permittee has no whole
effluent  toxicity  limits in its  current permit but
discovers a toxicity  problem, it may use a TRE to
reduce or eliminate effluent toxicity, ensure that there
is no reasonable potential that its discharge will exceed
toxicity criteria  and possibly obviate  the need for
whole effluent toxicity limits in a subsequent permit.
On the other hand, if a permit contains whole effluent
toxicity  monitoring  requirements  or  limits  and
unacceptable toxicity is  observed, the  permitting
authority may require the permittee to perform a TRE
through  special  conditions   in  the permit  or  an
enforcement action.

The TSD defines a TRE as "a site specific study
conducted in a stepwise process designed to identify
the causative agents of effluent toxicity,  isolate the
sources of toxicity, evaluate the  effectiveness of
toxicity control options, and then confirm the reduction
in effluent toxicity"  (USEPA, 1991b).  USEPA has
developed procedures that can be  used to conduct
TREs (USEPA 1989a, 1989b, 199 la, 1992a, 1993a,
1993b, 1996).

This document represents the first update of USEPA's
Toxicity Reduction Evaluation Protocol for Municipal

-------
 Wastewater Treatment Plants (1989a). This guidance
 provides a general framework for conducting TREs at
 publicly  owned  treatment  works  (POTWs)  and
 describes the available methods and procedures that
 experience to date has shown to be most useful.  It is
 designed for POTW staff, consultants, and regulatory
 agency staff who are implementing TREs to identify
 and reduce or eliminate sources of effluent toxicity.
 Where possible, POTW staff are encouraged to use the
 guidance  before  the  discharge  of whole effluent
 toxicity is subject to regulatory review and action.

 This guidance presents methods and procedures that
 are useful to:

  • Develop and implement a TRE plan.
  • Evaluate the results and data generated during the
    TRE.
  • Develop a sound scientific and engineering basis
    for the selection and implementation of toxicity
    control methods.

 This guidance supports the strategy desribed  in the
 TSD (USEPA, 1991b) for integrated toxics control
 using whole effluent toxicity and pollutant specific
 limits.  It is well recognized that while POTWs may
 achieve effluent limits for conventional pollutants, the
 discharge of effluent toxicity, volatilization of toxic
 materials, and contamination of sewage sludges can
 still occur. The focus of this guidance is the reduction
 of whole  effluent toxicity at municipal wastewater
 treatment plants.

 The methods and decision points that comprise a TRE
 are described in the context of an overall generalized
 approach.'Each municipality must  address regulatory
 issues and treatment operations that are unique to each
 POTW; therefore, not all components of this guidance
 will apply in every case. POTW staff may also select
 components to address specific questions about the
causes  and sources of effluent toxicity; however, the
decision to choose a particular step should be based on
technically sound information. Given the site-specific
nature of TREs, POTW staff will  need to develop a
TRE plan that  describes the overall approach  and
components of the guidance to be implemented.

In most cases, 'the approaches and methods described
in the  TRE protocol  (USEPA, 1989a)  have been
validated by USEPA TRE research studies and other
municipal TREs (Amato et al., 1992; Bailey et al.,
 1995; Botts et al., 1990, 1992, 1993, 1994; Collins et
 al.,  1991; Fillmore  et  al.,  1990;  Lankford  and
 Eckenfelder,   1990;  Morris  et  al.,  1990,  1992).
 Appendix A provides the original case studies from the
 municipal TRE protocol (USEPA, 1989a). Additional
 examples  of  successful  TREs  are  presented in
 Appendices B through H of this guidance.  The TRE
 guidance includes information learned  from  these
 studies. Major changes include:

  • Information  on toxicants commonly  found in
    POTW effluents and the conditions that influence
    their toxicity (Section 2).
  • Considerations in evaluating the operation and
    performance of POTWs with respect to conditions
    that may contribute to effluent toxicity. Additional
    information  is provided on operations review of
    biological nutrient  removal (BNR)  processes
    (Section 3).
  • A brief description of the  use of updated toxicity
    identification evaluation  (TIE) procedures for
    acute and short-term chronic toxicity (Section 4).
    The reader is referred to USEPA's guidance on
    TIE procedures for further details (USEPA 1991a,
    1992a, 1993a, 1993b, 1996).
  • Refined step-by-step guidance for tracking sources
    of acute and chronic toxicity in POTW collection
    systems (Section 5).
  • Additional recent TRE case studies that describe
    approaches for identifying the causes and sources
    of acute and chronic effluent toxicity and practical
    methods for toxicity reduction (Appendices B
    through H).

The methods and procedures described herein  will
continue to be updated and refined based on the results
of further studies.

Professional judgment is required  in  selecting the
appropriate steps for  identifying toxicants and for
evaluating  options for controlling  effluent toxicity.
USEPA has developed TIE procedures to use as tools
for TRE studies.  These TIE manuals (USEPA 1991a,
1992a, 1993a,  1993b,  1996) describe procedures for
characterization,  identification, and confirmation of
effluent  toxicants.   TIE procedures  are  a basic
component of the municipal TRE  and the USEPA
guidance manuals should be obtained and reviewed
prior  to  implementing a TRE.  USEPA  also  has
developed  a  generalized protocol   for conducting
industrial TREs (USEPA, 1989b).

-------
TRE Goals and Objectives
It is the responsibility of POTW staff to conduct a TRE
to identify and reduce or eliminate sources of effluent
toxicity and to fully comply with applicable toxicity-
based NPDES permit limits. The goal of the TRE may
be to achieve compliance with a whole effluent toxicity
limit; however, POTW staff are encouraged to use the
guidance to evaluate effluent toxicity before it becomes
a regulatory issue. The TRE goal and implementation
schedule should be clearly defined with the regulatory
authority as part  of the preparation of the TRE plan.
The regulatory authority will review the TRE plan and
carefully monitor the progress of the TRE, providing
direction as needed.

The following objectives may be defined to accomplish
the TRE goal:

  •  Evaluate the operation and performance of the
    POTW  to   identify   and  correct  treatment
    deficiencies contributing to effluent toxicity (e.g.,
    operations  problems,  chemical additives,   or
    incomplete treatment).
  •  Identify the compounds causing effluent toxicity.
  •  Trace the effluent toxicants and/or toxicity to their
    sources (e.g., industrial, commercial, or domestic).
  •  Evaluate, select, and implement toxicity reduction
    methods or  technologies  to control effluent
    toxicity  (i.e.,  in-plant or pretreatment  control
    options).

These objectives are applied to meet the TRE goal of
compliance with regulatory requirements.

Components of the Municipal  TRE
A generalized flow diagram for a TRE program is
illustrated in Figure 1-1.  A brief description of each
major TRE component is presented below along with
the section number in the guidance in which additional
information is provided.

Information and Data Acquisition (Section 2)
The first step in a TRE is the collection of information
and analytical data pertaining to effluent toxicity. This
information  includes  data  on  the operation  and
performance of the POTW,  such  as  plant design
criteria and discharge monitoring reports (DMRs), and
data from the POTW's pretreatment program, such as
industrial  waste  survey (IWS) information, permit
applications, and industrial user compliance reports.
The  POTW performance data  and  pretreatment
program information are used in the second stage of
the TRE, as described below.

Facility Performance Evaluation (Section 3)
Operations and performance data can be reviewed in a
POTW performance evaluation to indicate possible in-
plantsources of toxicity or operational deficiencies that
may be contributing to the effluent  toxicity.  If a
treatment deficiency is  causing noncompliance with
conventional pollutant permit limits, studies should be
conducted to evaluate treatment modifications before
proceeding further in the TRE.  These studies should
evaluate the toxicity reduction that can be achieved by
correcting treatment deficiencies. If plant performance
is not a  principal cause  of  toxicity or treatment
modifications do not reduce effluent toxicity, a logical
next step is to identify the cause(s) of toxicity using
TIE procedures.

Pretreatment  program data also can be gathered to
prepare a data base on the wastewaters discharged to
the POTW collection system. These data can be used
in the latter stages of the TRE to assist in tracking the
sources  of  toxicity   and/or   toxicants  that  are
contributing to POTW effluent toxicity.

Toxicity Identification Evaluation (Section 4)
This section  provides a brief overview of  the TIE
procedures. TIE procedures are available to evaluate
the causes of acute and short-term chronic toxicity.
When implementing a TIE, the reader is  advised to
consult  USEPA's TIE procedures  for  freshwater
species (1991a, 1992a, 1993a,  1993b)  or estuarine/
marine species (1996). The generic TIE protocol is
performed in three phases: toxicity characterization
(Phase  I), toxicant  identification (Phase II),  and
toxicant confirmation (Phase HI). Phase I characterizes
the types of effluent toxicants by testing the toxicity of
aliquots  of effluent samples that have  undergone
bench-top  manipulation   (e.g.,  pH  adjustment,
filtration). An evaluation of common POTW effluent
toxicants  such   as  ammonia, chlorine,  and
organophosphate  insecticides may be included in
Phase I.  Phases II and HI involve further treatments in
conjunction with chemical analyses  to identify and
confirm  the  compounds  causing effluent  toxicity.
USEPA's Phase H and m procedures  (1993a, 1993b)
for freshwater species are generally applicable  for
estuarine/marine species.

Toxicity Source Evaluation (Section 5)
A toxicity source evaluation involves the sampling and
analysis  of wastewaters discharged from sewer lines

-------
                   Voluntary or Mandatory TRE
                          TRE Test Plan
                               I
                 Information and Data Acquisition
                               I
                  Facility Performance Evaluation
     Pretreatment Program
           Review
POTW Performance
    Evaluation
                            Additional
                           Information
                            Required?
      No
                              TIE
                           Additional
                           Information
                           Required?
      No
                    Toxicity Source Evaluation
                           Additional
                           Information
                           Required?
      No
                                 Yes
                                                     Toxicity Control Evaluation
                                      Pretreatment Control
                                           Evaluation
                                   In-Plant Control
                                     Evaluation
                                                       Select Control Options
                                                  Toxicity Control Implementation
                                                     and Follow-Up Monitoring
Figure 1-1. TRE flow diagram for municipal wastewater treatment plants.

                                                4

-------
and indirect dischargers such as industrial users and
commercial facilities. Two types of source evaluation
studies  may be  performed:  chemical  tracking or
toxicity-based tracking.

Chemical-specific tracking is recommended when the
POTW  effluent toxicants have been identified and
confirmed in the TEE, and can be readily traced to the
responsible sewer dischargers.  Toxicity tracking is
used  when TIE data indicate the type  of effluent
toxicant, but the specific toxicant(s) is not identified.
Toxicity tracking involves treating the sewer samples
in a bench-scale treatment simulation prior to toxicity
measurements to account for the toxicity removal that
is provided by the POTW.

The sampling strategy for toxicity source evaluations
involves two tiers.  Tier I focuses on sampling and
analysis of the main sewer  lines  in the collection
system.  Tier  n involves  testing sewer lines and
indirect  dischargers upstream of the main lines
identified as being toxic in Tier I. This tiered approach
can be  used to identify the contributors of toxicity
and/or  toxicants by  eliminating  segments  of  the
collection  system that  do not contribute toxicity/
toxicants.

Toxicity Control Evaluation (Section 6)
Using the results of each of the above TRE elements,
alternatives for effluent toxicity reduction are evaluated
and   the most  feasible option(s)  is  selected  for
implementation.  Effluent toxicity may be controlled
either through  pretreatment  regulations  or in-plant
treatment modifications or additions. In some cases,
several control methods may be required to achieve the
desired toxicity reduction. Selection of control options
is usually based on technical and cost criteria.

If the  toxicity source  evaluation is successful in
locating the sources that are contributing the  POTW
effluent toxicants, local limits can be developed and
implemented.  If in-plant control appears to be  a
feasible approach, treatability testing may be  used to
evaluate methods for optimizing existing treatment
processes  and to assess  options  for  additional
treatment.

Toxicity Control Implementation (Section  7)
The  toxicity  control  method or  technology is
implemented and follow-up monitoring is conducted to
ensure  that the  control method achieves the  TRE
objectives  and meets permit limits.
Limitations of the TRE Guidance
This guidance describes procedures for evaluating and
implementing controls for reduction of whole effluent
toxicity.   Procedures  for the reduction  of  toxic
pollutants in residuals, biosolids, and air emissions at
POTWs are not discussed. The reader may consult the
Standards for the Use or Disposal of Sewage Sludge
(40 CFR  Part 503)  regarding the control of toxic
materials in biosolids.

The municipal TRE guidance was developed based on
the results and findings of TRE and TIE studies. The
following limitations have been  identified in these
studies:

  • Intermittent   or  ephemeral  toxicity  may  be
    challenging   to  characterize using  TEE/TRE
    procedures.  In these cases, modifications to TRE
    procedures may be needed to achieve the best
    possible  results  (see   Sections  4  and  5).
    Discussions with the regulatory authority also may
    help to identify the most appropriate approach for
    complying with effluent toxicity requirements.
  • As described   in  this  guidance,  alternative
    procedures are available if traditional methods
    such as TEE testing are not successful. Additional
    TRE  procedures, especially tools  for toxicity
    source evaluations, have not been widely used, but
    may be helpful if careful consideration is given to
    their design and application.
  • As more TRE studies  are completed,  more
    information  is  available on  the feasibility and
    effectiveness of in-plant and pretreatment toxicity
    control  options.  Examples  of TREs  in which
    toxicity   controls  have  been  successfully
    implemented are provided in Appendices B, C, D,
    E, G, and H.
  • The TRE guidance is designed to help public
    works managers select appropriate toxicity control
    approaches.   As  such,  it  does  not discuss
    regulatory procedures that  may  be useful for
    assessing the need for,  or compliance with,
    toxicity requirements, such as the determination of
    reasonable potential, dilution factors, and permit
    limits. The importance of these procedures in the
    evaluation of whole effluent toxicity is mentioned
    in Section  2 and  is discussed more fully  in
    USEPA's TSD (1991b).

-------
Organization of the TRE Guidance
This  guidance  is  organized  according  to  the
components of the TRE flow diagram shown in Figure
1-1.

Sections 1 through 7 describe the  primary  TRE
elements noted above. Changes to the municipal TRE
protocol (USEPA, 1989a) include more information on
toxicants commonly identified in POTW effluents,
suggestions for  evaluating  the effect  of  POTW
operations on effluent toxicity, an overview of updated
TIE procedures  for acute  and short-term  chronic
toxicity,  and refined  step-by-step  procedures for
tracking sources  of acute and  chronic  toxicity in
POTW collection systems.

Sections 8 through 11 provide information on the
TRE  requirements  for  quality assurance/quality
control, health and safety, facilities and equipment, and
sample collection and handling.

Sections 12 and 13 list the references and bibliography
cited in this guidance.
Appendix A presents the original case histories (given
in the municipal TRE protocol, USEPA, 1989a) along
with commentary on how the TEE/TRE procedures
have been updated to better address toxicity observed
in future studies.

Appendices B through G provide new in-depth case
examples of municipal TREs. These new examples
include  summaries of four chronic TRE studies  and
two acute TRE studies.

Appendix  H is  a  new  appendix that describes
approaches for addressing effluent toxicity caused by
organophosphate insecticides.

Appendix I describes a chemical-specific approach for
TREs that may be  applied in limited circumstances.

Appendix J is referenced in Section 5 (toxicity source
evaluation) and  presents  an updated step-by-step
procedure for tracking sources of toxicity in POTW
collection systems.

-------
                                           Section 2
                          Information and Data Acquisition
Introduction
The first step in a TRE is to gather all information and
data that may relate to effluent toxicity and that might
prove useful in planning and conducting the TRE.
This  information can be categorized  as POTW
treatment plant data and pretreatment program data.
The pertinent POTW information includes historical
effluent toxicity data  as well as information on the
treatment  plant's1  design  capabilities,  treatment
performance, and operation and maintenance practices.
Appropriate pretreatment program information consists
of IWS data,  industrial user permits, pretreatment
inspection reports, and monitoring and compliance
reports. If a pretreatment program is not in place,
POTW staff may need to collect monitoring data on the
POTW industrial  users  and,  where   necessary,
appropriate controls should be considered to ensure
good effluent quality.

Background information may provide insight into the
nature of effluent toxicity and can be used to select the
initial  steps to take in the TRE.   However,  it is
important ttot to draw conclusions about the causes and
sources of toxicity in the beginning of the TRE unless
corroborative testing is performed.  A summary of
information recommended for a TRE is provided in the
following subsections.

Review of Effluent Toxicity Data
Information and  data acquisition should include a
careful review of recent effluent toxicity data.  This
review should be used to confirm the effluent toxicity
results and  the potential  to cause adverse instream
effects. The data also can be used to evaluate general
toxicity characteristics, such as temporal  variability,
species sensitivity, and whether the toxicant(s) is fast
or slow acting.

In some states, laboratories are required to be certified
to perform toxicity tests. Toxicity test data reports also
may be reviewed by regulatory staff to confirm that the
tests meet basic  quality assurance/quality control
(QA/QC) requirements. However, this is usually the
exception; most state and regional regulatory agencies
do not have certification programs for toxicity testing
and reports  may not be formally reviewed.  As an
initial step in the TRE, POTW staff should conduct an
independent review of the toxicity test reports to verify
the quality of the reported data, especially results that
have triggered TRE requirements.

It  often is beneficial to  develop a profile on the
characteristics of effluent toxicity using the available
historical data. Information on toxicity variability, the
relative  sensitivity of  various  test species to the
effluent,  and  effluent characteristics  [e.g.,  pH,
alkalinity,  hardness,  conductivity,  total  residual
chlorine (TRC), and dissolved oxygen (DO)] can
provide important clues  about the  nature of the
toxicity. These characteristics can be compared to
POTW and pretreatment information to help determine
if  effluent toxicity may  be related  to  operational
practices or sewer discharges. This  information also
can be  used  as part  of the TIE (Section 4) to help
identify the causes of effluent toxicity.

The data review may show that some test conditions
such as pH may  artificially change during testing.
Typically, the pH of toxicity test solutions tends to drift
upward over time, which can cause pH sensitive
compounds  such as ammonia and metals to exhibit
toxicity. With the consent of the regulatory authority,
it may be possible to modify the test procedures to
control  pH  drift  (USEPA  1993c  and  T.  Davies,
USEPA,  Office  of  Water,  Memorandum  on
Clarifications Regarding Flexibility in 40 CFR Part
136 Whole Effluent Test Methods,  April 10, 1996).
Modifications may also be allowed to better reflect the
range of temperatures and hardness observed in the
receiving water. Depending on the temperatures to be

-------
 considered, it is recommended to use a different test
 species  rather   than   modify  the  recommended
 temperature range for a given test species.

 Prior to a TRE, POTW staff or the regulatory agency
 may evaluate the "reasonable potential" for exceeding
 a toxicity-based water quality standard to determine if
 a  permit limit is required.  If  there is reasonable
 potential to cause instream toxicity or contribute to an
 excursion  above a narrative  criterion,  a statistical
 approach may be used to calculate a toxicity-based
 permit limit. This approach may also be applied during
 the course of a TRE to assess compliance with a permit
 limit or a water quality standard.   For example,
 improvements in effluent quality resulting from the
 TRE could lower effluent toxicity to a  point where
 there is no  longer a reasonable potential to exceed the
 permit limit.  Or, these improvements  may reduce
 effluent  variability.  The reduced variability  could
 result in a smaller coefficient of variation (CV), which
 would lessen  the potential for excursions above the
 TRE goal.  The reader is referred to USEPA' s TSD for
 details on these procedures (USEPA, 1991b).

 The TSD  also discusses the use of dilution, and
 particularly the use of high-rate diff users, in achieving
 compliance  with  toxicity-based  water   quality
 standards.  The dilution determination, if allowed by
 applicable  regulations, is one of  the first  steps in
 characterizing the effluent for toxicity-based permitting
 (USEPA, 1991b). Public works managers, who are
 initiating  a  TRE,  may  choose  to evaluate  the
 application of appropriate mixing zone allowances to
 eliminate the potential  for  instream effects.   A
 shoreline outfall, for instance, may not qualify for any
 dilution  when   determining an  acute   toxicity
requirement. Use of a  diffuser constructed in deeper
 water may allow the effluent to achieve sufficient
dilution in  the rapid-mixing, near-field area to meet
permit requirements.  Similar results may be obtained
by moving an outfall  from a small or  intermittent
stream, where no dilution is available under low flow
conditions,  to a larger permanent stream, with greater
dilution.  It should be noted, however, that less costly
toxicity control approaches than outfall relocation may
be identified during the course  of the  TRE.   The
process of  selecting the most feasible and practical
control option(s) is described in Section 6 of this
guidance.
 Toxicants Identified in POTW Effluents
 As noted, the occurrence of toxicity and the treatment
 process  operations  are unique  to  each  POTW;
 therefore, the causes of effluent toxicity are likely to be
 different for each case.  Nonetheless, some toxicants
 have been identified at many  POTWs.  A list of
 toxicants  commonly found in POTW effluents, the
 levels of concern, and potential sources is presented in
 Table 2-1. The levels of concern are to be used as a
 general guide, not as absolute values. Due to the site-
 specific nature of effluent  toxicity, these data are
 intended only as background information to consider in
 the process of conducting a TRE. It is important to
 stress  that  a  direct   comparison  of  chemical
 concentrations to toxicity data reported in the literature
 often provides misleading information. The toxicity of
 effluent  constituents is affected  by many factors
 including the effluent  matrix  and toxicity test
 conditions. The most effective way to identify causes
 of effluent toxicity is by applying the TIE procedures,
 which are described in Section 4 of this guidance.

 Some of the information that can be collected to help
 evaluate the contribution of the toxicants to  effluent
 toxicity is provided in Table 2-1.  Toxicity information
 on specific parameters can be obtained fromUSEPA's
 Aquatic Information Retrieval  Toxicity Data  Base
 (AQUIRE,  1992),  TIE manuals  (USEPA  1991a,
 1992a, 1993a, 1993b, 1996), Extension TOXicology
 NETwork (EXTOXNET, 1998), peer-reviewedjournal
 articles, and other sources. AQUIRE information can
 be   obtained   through   the  National   Technical
 Information Service (NTIS) in Springfield, Virginia, or
 through several commercial vendors. USEPA's Mid-
 Continent Ecology Division (Duluth, Minnesota) will
 be offering Internet access to AQUIRE data  in early
 1999 through  its  Ecotoxicology Database Retrieval
 System (ECOTOX).  The Extension TOXicology
 NETwork is  currently available on  the Internet at
 http://ace.orst.edu/info/extoxnet/.   When reviewing
 toxicological data, it is  important to ensure that the
references for the data have been peer-reviewed and
the values given are considered to be accurate.

In some cases, toxicological data may be presented in
toxic units (TUs)  instead of in lethal concentrations
causing a 50% mortality in exposed test organisms
(LC50) or no observed effect concentrations (NOEC).
TUs are the inverse of the  percent  concentration

-------
Table 2-1. Toxicants Identified in POTW Effluents
Toxicant Type
Chlorine
Ammonia
Non-polar organics,'
such as
organophosphate
insecticides (e.g.,
diazinon, malathion,
chlorpyrifos, and
chlorfenvinphos)
Metals [e.g., cadmium
(Cd), copper (Cu),
chromium (Cr), lead
(Pb), nickel (Ni), zinc
(Zn)]
Other treatment
chemical additives such
as dechlorination
chemicals and polymers
Surfactants
Total dissolved solids
(TDS)
Level of Concern*
0.05 to 1 milligram per
liter (mg/L)
5 mg/L as NH3-N
Diazinon: 0.12-0.58
microgram per liter
(ug/L)
Chlorpyrifos: 0.03 ug/L
Varies
Varies
Varies
1,000-6,000 uhmos/cm
depending on endpoint,
species tested, and TDS
constituents
Potential Source
POTW disinfection
process
Domestic and industrial
sources
POTW sludge
processing sidestreams
Homeowners,
apartments,
veterinarians, pest
control, lawn care, and
commercial businesses
Treatment additives in
POTW
Industrial users
Disinfection,
dechlorination, sludge
processing, and solids
clarification in the
POTW
Industrial users
Industrial users
Sludge processing
sidestreams
Information Needed
to Assess Toxicity
TRC, temperature, and pH upon
receipt of effluent sample and
during toxicity test
Toxicity degradation tests
TIE Phase I testsf
Ammonia-nitrogen upon receipt
of effluent sample
pH, temperature, and salinity
during toxicity test
TIE Phase I testsf
High resolution analysis of
organophosphate insecticides
TIE Phase I testsf
Dissolved metals, effluent
hardness (mg/L as CaCO3), and
alkalinity upon receipt of
sample
TIE Phase I testsf
Vendor information on toxicity
of products
Dosage rates
Effluent characteristics that
affect toxicity (e.g., pH)
TIE Phase I testsf
Methylene blue active
substances (MBAS) and cobalt
thiocyanate active substances
(CTAS)
TIE Phase I testsf
TDS, ion analysis, and anion/
cation balance
TIE Phase I testsf
 * As referenced by USEPA (1992a) and D. Mount (personal communication, AScI Corp, Duluth, Minnesota, 1991) for
   chlorine; USEPA (1992a) for ammonia; TRAC Laboratories (1992), Bailey et al. (1997) for diazinon and chlorpyrifos; and
   USEPA (1992a) for TDS.
 t The contribution of effluent constituents such as chlorine, ammonia, organic compounds, metals, and TDS to effluent
   toxicity can be most effectively evaluated using the TIE Phase I procedures described in Sections 3 and 4 of this guidance
   and the USEPA manuals (1991a, 1992a, 1996).

-------
values and are calculated by dividing 100% by acute or
chronic  percent  effluent   values   or  chemical
concentration data. For example, a chronic TU (TUc)
of 2 is equivalent to an NOEC value of 50% effluent
(i.e.,  100%/50%).  Likewise,  if the  LC50  of a
compound is 20 pg/L, an effluent sample with 100
\igfL contains 5 acute TU (TUa) of the compound (i.e.,
100 ng/L/20  |ig/L).   TU values are  helpful in
understanding the relative contribution of toxicants to
effluent toxicity and are often used in interpreting TEE
data (Section 4).  For example, if one of two
compounds is contributing to effluent toxicity (e.g., 4
TUc of compound A and 1.5 TUc of compound B), it
may be possible to focus on controlling the major
toxicant if compliance with the permit limit (e.g., 3
TUc) can be achieved. However, consideration should
be given to possible antagonistic effects between the
toxicants such that removal of one toxicant may cause
the other toxicant to  exhibit greater toxicity.   An
overview of this is provided in Section 4 and the TIE
manuals (USEPA 1991a, 1992a, 1993a, 1993b, 1996)
provide thorough guidance.

POTW Design and Operations Data
POTW design and operations information can indicate
possible in-plant  sources of toxicity  or operational
problems that might  be  contributing to treatment
interferences and the pass through of toxicity.  In the
beginning of the THE, it is often helpful to briefly
review the operations  and performance of the major
unit treatment processes.  Notes can be added to
POTW data base  about  initial impressions and
potential problem areas that should be investigated
further in the POTW Performance Evaluation (see
Section 3).

The types of POTW data to be gathered include:

  •  Background information on treatment plant design
    and operation.
  •  Data routinely collected for NPDES DMRs and
    treatment process control.
  •  Existing  data on potential  effluent  toxicants,
    including  chlorine, ammonia, organophosphate
    insecticides, surfactants, metals,  and treatment
    additives (e.g., polymers, chlorine, dechlorination
    chemicals).

A list of useful POTW data is provided in Table 2-2.

The POTW data can be compared to the profile on
effluent toxicity characteristics to determine if toxicity
may be related to operation and performance. Several
questions can be posed, including:

  •  Is  toxicity  apparent during certain operational
    events,  such  as  when  treatment  upsets  are
    observed, when treatment units are taken offline
    for maintenance, or as a result of other operating
    practices (e.g., excess chlorine addition)?
  •  Does  toxicity exhibit a  weekly,  monthly, or
    seasonal pattern?  For example, if the POTW is
    operated to achieve seasonal ammonia removal, is
    toxicity  present in the period  when ammonia
    removal is not practiced?  What process control
    parameters may be related to toxicity?  Is toxicity
    apparent with changes in hydraulic and pollutant
    loadings to the primary  sedimentation process,
    changes  in biological  treatment  parameters
    [e.g.,  mixed liquor suspended solids (MLSS)
    concentration, DO concentration, sludge volume
    index (SVI), mean cell residence time (MCRT)],
    changes in filtration rates in filters, or changes in
    application  rates of chlorine and dechlorinating
    agents?
  •  Is  toxicity apparent when the type and dose of
    treatment additives change?  For example, did
    toxicity occur when a different polymer or other
    coagulant/flocculent aid was used?

In the  beginning of the TRE, emphasis  should be
placed on effluent concentrations  of  ammonia  and
chlorine, which are  common  toxicants  in  POTW
effluents (USEPA 1991a, 1992a, 1993a, and 1993b).
The toxicity  of ammonia  is dependent on effluent
characteristics such as the pH, temperature, and salinity
of the sample, as well as the sensitivity of the species
being  tested.   Therefore,  it  will be  difficult to
determine the toxicity  of ammonia based solely on a
comparison  of  literature   values   to  effluent
concentrations.  Likewise, the toxicity of chlorine will
depend on the form of chlorine, which may be in the
free form  as  chlorine,  hypochlorous  acid,  or
hypochlorite  ion,  or in  the  combined form as
chloroamines or nitrogen trichloride. The sum of the
free and combined  chlorine,  termed  total residual
chlorine or TRC, is matrix dependent.  Chloramines,
which  are  formed  by chlorine  combining  with
ammonia, can  be  more toxic than  free chlorine
(AQUIRE, 1992).

Assessments of the contribution of ammonia, chlorine,
and other compounds to effluent toxicity can be made
using Table 2-1 as a guide. As stated in Table 2-1, if
                                                10

-------
Table 2-2. Example POTW Design and Operation Data
  1.  NPDES permit requirements
     a.  Effluent limitations
     b.  Special conditions
     c.  Monitoring data and compliance history
     d.  Dilution studies or modeling results

  2.  POTW design criteria
     a.  Hydraulic loading capacities
     b.  Pollutant loading capacities
     c.  Biodegradation kinetics calculations and assumptions

  3.  Influent and effluent pollutant data
     a.  Ammonia
     b.  Residual chlorine
     b.  Other pollutants of concern such as non-polar organic compounds (e.g., organophosphate insecticides),
         metals, and TDS (see Table 2-1)
     c.  Conventional pollutant data, including five-day biochemical oxygen demand (BOD5), chemical oxygen
         demand (COD), total organic carbon (TOC), total suspended solids (TSS), volatile suspended solids (VSS),
         total Kjeldahl nitrogen (TKN), ammonia-nitrogen (NH3-N), total phosphorus (TP), orthophosphate (PO4-P),
         and nitrate-nitrogen (NO3-N),to evaluate treatment performance
     d.  Parameters, including  pH, hardness, and alkalinity, to evaluate the toxicity of suspect compounds
         (see Table 2-1)

  4.  Process control data
     a.  Chemical usage for each treatment process (e.g., coagulants for primary sedimentation, lime for biological
         treatment, polymers for tertiary clarification; see Table 2-1)
     b.  Process control data for primary sedimentation (i.e., hydraulic loading capacity and BOD5 and TSS removal)
     c.  Process control data for activated sludge [e.g., food to microorganism (F/M) ratio, MCRT, MLSS, sludge
         yield, removal efficiency of BOD5, COD, TKN, NH3-N, TP, PO4-P, NO3-N, and other pollutants specified
         in the permit].
     d.  Process control data for secondary and tertiary clarification [e.g., hydraulic and solids loading capacity, SVI,
         sludge blanket depth]
     e.  Number of process units online and number offline for maintenance

  5.  Operations Information
     a.  Reports on previous operation and maintenance evaluations, including engineering studies and USEPA and
         state compliance inspections
     b.  Operating logs
     c.  Standard operating procedures
     d.  Operation and maintenance practices (e.g., filter backwash procedures)
  6.  Process sidestream characterization data
     a.  Chemical usage for sludge processing, including thickener, digester, and dewatering processes
     b.  Pollutant data for sludge processing sidestreams, including ammonia, metals, organophosphate insecticides,
         and TDS (see Table 2-1)
     c.  Incinerator scrubber waste stream, including data on possible formation of cyanide (see discussion in
         Section 3)
     d.  Tertiary filter backwash
     e.  Cooling water
  7.  Wastewater bypass, combined sewer overflow (CSO), and sanitary sewer overflow (SSO) for bypasses or
     overflows that are discharged to the POTW effluent
     a.  Frequency
     b.  Volume
                                                    11

-------
chlorine is suspected of contributing to toxicity, TRC
should be measured when the sample arrives and when
toxicity tests are initiated because chlorine usually
dissipates rapidly. If ammonia is a concern, conditions
that affect its toxicity, including sample pH, should be
carefully monitored during  toxicity tests.   This
information is necessary to determine the concentration
of the toxic un-ionized form of ammonia (NH3) in the
toxicity test. In addition to toxicity data, USEPA's
AQUIRE data base (1992) includes information on the
conditions of the toxicity test that may have influenced
the  reported  toxicity   of  chemicals of  concern.
However, not  all conditions may be recognized or
reported, which may limit the utility for TIE tests.

Surfactants also have been identified as toxicants in
POTW effluents (Diehl and Moore, 1987; Ankley and
Burkhard, 1992; Botts et al.,  1994).   These studies
focused  on  characterizing the type  or  source of
surfactants  rather than trying  to  identify  the  toxic
surfactant compound because analytical methods are
not readily available to detect and quantify surfactant
compounds in complex effluents. Municipal effluents
contain   numerous  substances that  interfere  with
surfactant analysis.  Also, surfactants are actually
mixtures  of  many homologues  and   oligomers;
therefore, the composition and toxicity of surfactants is
complex and often  variable (USEPA, 1993a).   One
exception is  a class of surfactants, referred to as alkyl
phenol ethoxylates (APEs), that can be analyzed by a
gaschomatograph/mass spectrometer (GC/MS) (Giger
et al., 1981). Also, it may be helpful to characterize
surfactants  by  the  nature  of  their polar segment;
surfaqtants  may be classified  as nonionic, anionic,
cationic,  and amphoteric.  Many surfactants tend to
sorb to CIS resin and analyses of the methanol extracts
from solid phase extraction (SPE) columns can help to
indicate  the type  of  surfactant  causing toxicity
(USEPA, 1993a).  The American  Public Health
Association (1995) describes methods for determining
anionic surfactants as MB AS and nonionic surfactants
as CTAS.

The contribution of effluent  constituents  such as
ammonia and chlorine to effluent toxicity can be most
effectively   evaluated   using   the  TIE  procedures
described in Sections  3 and 4.  Non-polar organic
compounds  (such as organophosphate insecticides),
metals, surfactants,  and TDS also can be effectively
evaluated using these procedures.  In Phase I of the
TIE, various sample manipulations and toxicity tests
are performed to determine how toxicity is affected by
removing or isolating a particular group of toxicants.
These  procedures  establish  a  cause  and  effect
relationship between toxicants  and whole  effluent
toxicity.  If a toxicant  is indicated through Phase I
testing, additional TIE  procedures can be applied to
identify (Phase II) and confirm (Phase ID) the toxicant
(see Section 4).

Treatment  additives can be screened by obtaining
toxicity data from product vendors or by performing
toxicity tests on samples that have been treated using
typical chemical dosages.   If  toxicity tests  are
performed, it is important to simulate the conditions
occurring in the treatment process where additives are
being used because  some portion  of the additives is
usually  removed in the treatment process.   For
example, polymers are largely bound with suspended
solids in the wastewater being treated and only minor
amounts may pass through in the final effluent (Hall
andMirenda, 1991).

Once the toxicants are identified, the POTW data will
be useful in evaluating and selecting in-plant toxicity
control options (see Section 6).

Pretreatment Program Data
Pretreatment program  information  may  provide
evidence that can be  used to  identify sources of
toxicants or  toxicity in the  wastewater collection
system.  For this reason, pretreatment data should be
briefly reviewed in the beginning of the TRE. As an
initial step, pretreatment data can be compared to the
profile on POTW effluent toxicity characteristics to
determine if toxicity may be related to a particular type
of discharge.  This  review may attempt  to  answer
several questions, including:

  • What changes in POTW influent characteristics
   may be observed during toxic periods (e.g.,  pH,
   alkalinity,  suspended  solids,  hardness,
   conductivity, DO,  color)?  Also,  does  toxicity
   occur during changes in hydraulic and pollutant
   loadings to the POTW?  Can these characteristics
   be related to certain types of discharges?
  • Does toxicity occur when treatment upsets are
   observed at the POTW? Can the upsets be related
   to a particular discharge(s)?
  • Does  toxicity  exhibit  a  weekly, monthly, or
   seasonal pattern  that may be related to production
   schedules of certain industries?  For example, is
                                                 12

-------
    toxicity  observed  when  an   industry   is
    manufacturing a particular type of product? Also,
    does toxicity abate when the industry is shutdown
    for maintenance or holidays?
  •  If the POTW accepts hauled wastes, is toxicity
    apparent when a particular hauler delivers wastes?

Appropriate pretreatment  program information to
review includes the data on the industrial users of the
POTW  [e.g.,  industrial  manufacturers,  Resource
Conservation and Recovery Act  (RCRA)  waste
disposers,   and  Comprehensive   Environmental
Response,   Compensation,   and  Liability   Act
(CERCLA) dischargers] and the toxic pollutant data on
the POTW  waste streams.   A list of suggested
pretreatment data is shown in Table 2-3.

The  POTW pretreatment  program  data  can  be
reviewed as  part of a Pretreatment Program Review
(described in Section 3). The summarized data may be
useful in locating the sources of toxicants identified in
the TIE (see Section 4).  In cases in which effluent
toxicants are not identified, the pretreatment program
data can be used to develop a sampling and analysis
program to track sources of toxicity in the collection
system (see Section 5).
                                                13

-------
Table 2-3. Example Pretreatment Program Data
  1.  POTW influent and effluent characterization data
     a.  Toxicity
     b.  Priority pollutants                                                             ,
     c.  Hazardous pollutants
     d.  Pollutants listed in Superfund Amendments and Reauthorization Act (SARA) Title 313
     c.  Other chemical-specific monitoring results (e.g., industry raw materials and products)
  2,  Sewage residuals characterization data (e.g., raw, digested, thickened, and dewatered sludge, composted biosolids, and incinerator
     ash)                                                                             !
     a.  Toxicity characteristic leaching procedure (TCLP)
     b.  Chemical data
  3.  IWS
     a.  Information on industrial users with categorical standards or local limits and other significant non-categorical industrial users
         -   number of industrial users
         -   discharge flow
         -   chemical usage                                                            ',
     b.  Standard Industrial Classification (SIC) code
     c.  Wastewater flow
     d.  Types and concentrations of pollutants in the discharge                            j
     c.  Products manufactured
     f.  Description of pretreatment facilities and operating practices
  4.  Industrial User Permits
     a.  Pretreatment standards
         —   categorical standards                                                       '
         -   local limits
         —   prohibited discharge standards
     b.  Monitoring requirements                                                       |
  5.  Annual pretreatment program report                                                ;
     a.  Schematic of sewer collection system
     b.  Industrial user monitoring and inspection data collected by POTW staff
         -   discharge characterization data                                              ;
         -   spill prevention and control procedures
         -   hazardous waste generation                                                 '
     c.  Industrial user self-monitoring data                                              •
         —   discharge characterization data                                              ',
         -   flow measurements
         -   description of operations
         —   compliance schedule (if out of compliance; e.g., notice of slug loading)
  6.  Headworks analysis for local limits
  7.  Industrial user compliance reports
  8.  Waste hauler monitoring data and manifests       \                                  '
  9.  RCRA reports [if the POTW is considered a hazardous waste treatment, storage, and disposal facility (TSDF)]
     a.  Hazardous waste manifests
     b.  Operating record
     c.  Biennial report
     d.  Unmanifested waste report                                                     '
  10. CERCLA reports (if the POTW accepts wastes from a superfund site)
     a.  Preliminary site assessment
     b.  Site investigations
     c.  Remedial investigations
     d.  Feasibility studies                                                             \
     e.  CERCLA decision documents
  11. Information on POTW treatment interferences (e.g., biological process inhibition); example data include:
     a.  Evidence of slug loadings                                                      :
     b.  Decreased pollutant removal
     c.  Decreased oxygen uptake rates (OURs), SVI, and sludge yield in biological treatment process
     d.  Increased requirement for chemical usage (e.g., chlorine, coagulants, flocculents)
     c.  Decreased filtration rate and increased backwash frequency for filtration treatment
                                                            14

-------
                                            Section 3
                            Facility Performance Evaluation
Introduction
POTW  treatment  deficiencies  that  cause  poor
conventional pollutant removal can have an adverse
effect on toxicity reduction as well. As an initial step
in the TRE, effluent  toxicity data (Table 2-1) and
POTW operations and performance data (Table 2-2)
should be evaluated to indicate potential toxicants of
concern and to identify treatment deficiencies or in-
plant sources of toxicity that may be responsible for all
or part of the effluent toxicity.  POTW pretreatment
program data (Table 2-3) should also be reviewed to
indicate possible sources of toxicity and summarized
for use in later steps of the TRE such as the toxicity
source evaluation (Section 5).

POTW Performance Evaluation
A POTW performance evaluation can be conducted to
indicate conventional pollutant treatment deficiencies
or in-plant sources of toxicity that may be contributing
to effluent toxicity.  Conventional pollutant treatment
deficiencies include the inability to meet permit limits
for BOD, TSS, and nutrients.   These deficiencies
should be corrected before initiating  a full  TRE
because improved treatment also may reduce effluent
toxicity.  In-plant sources of toxicity may be present
even  if the  POTW is  meeting  permit  limits for
pollutants other than toxicity. An example of an in-
plant source of toxicity includes  inadequate solids
separation in the final clarifier, which may result in the
discharge of toxic material bound to suspended solids.
Also, incomplete biological treatment may cause the
pass-through of biodegradable  toxicants.  Other in-
plant sources of  toxicity  may include treatment
additives  used in toxic amounts or additives that
contain toxic impurities.  If deficiencies are found in
the POTW performance evaluation, improvements can
be implemented to eliminate the causes of toxicity.
Several examples of operating conditions that  have
contributed to effluent toxicity at POTWs and the steps
taken to correct  the problem  are included  in the
 following  discussion of  the  POTW  performance
 evaluation process.

 A flowchart for conducting a POTW performance
 evaluation is presented  in Figure 3-1.  The POTW
 performance evaluation involves a review of the major
 treatment unit processes  (e.g., primary sedimentation,
 activated sludge,  and secondary  clarification) using
 wastewater  characterization  data   and  process
 operations information.  A TIE Phase I analysis (as
 described below  and in  Section 4)  also  can  be
 performed to indicate the presence of effluent toxicants
 caused  by  incomplete  treatment (e.g., ammonia),
 routine  operating practices  (e.g., chlorine), or the
 discharge of organophosphate pesticides in the POTW
 collection  system.    Ammonia   and   chlorine  are
 commonly found to cause toxicity in POTW effluents
 and should be evaluated at this stage of the POTW
 performance evaluation.  As  noted in Section 2,
 ammonia  and  chlorine may be  of concern  at
 concentrations greater than 5 mg/L and 0.01 mg/L,
 respectively, depending on the effluent matrix and the
 species  being tested.  Levels of concern for other
 relatively  common  effluent  toxicants  are listed in
 Table 2-1. Special consideration also should be given
 to chemicals used in the treatment process such as used
 or reused waste materials and coagulants, which may
 contribute to toxicity due to pass-through of residual
 concentrations or impurities in the product.

 Based on   the process  review  results  and  TIE
 characterization data, options for improving operations
 and performance may be selected and evaluated in
 treatability studies. If treatability tests are successful in
 identifying  options   for  improving  conventional
pollutant treatment and toxicity reduction, the TRE
proceeds to the selection  and implementation of those
options (Section 6). If no treatment deficiencies or in-
plant sources of toxicity are observed, or the treatment
alternatives  do  not  reduce effluent  toxicity  to
                                                 15

-------
                                 Facility Performance Evaluation
                                Information and Data Acquisition
                                                                            Phase I Toxicity
                                                                            Characterization
                        Evaluation of POTW Operation and Performance

                      Evaluate Common Toxicants
                          Ammonia, Chlorine, Surfactants, Organophosphate
                          Pesticides, Metals, Treatment Additives, TDS
                      Evaluate Conventional Pollutant Treatment
                          — Preliminary Treatment
                          — Primary Sedimentation
                          - Biological Treatment
                          - Secondary/Tertiary Clarification
                          — Filtration
                          — Disinfection/Dechlorination            ;
                          — Process Sidestreams/Bypasses
                      Evaluate In-Plant Sources of Toxicity          j
                          — Disinfection Chemicals                 :
                          — Coagulants/Flocculents
                          — Toxic Impurities in Additives
                             No
          Plant
         Failure
        Observed?
                         Bench-Scale
                         Conventional
                       Treatability Tests
                         Pilot-Scale
                        Conventional
                      Treatability Tests
          TIE
    Toxicity Adequately
Reduced by Modification of
   Treatment/Operation?
Toxicity
 Control
Selection
Figure 3-1. Flow diagram for a facility performance evaluation.

                                               16

-------
acceptable  levels,  a  complete  effluent  toxicity
characterization should be performed using the TIE
procedures described in Section 4.

POTWs  are  subject  to  both  variable  influent
characteristics and changing operating conditions that
may have a significant effect of effluent toxicity. The
POTW performance evaluation should be conducted
during a period when the influent loadings and facility
.operations are representative of average conditions. If
effluent toxicity varies seasonally or as a result of a
specific operational condition, the POTW performance
evaluation should be scheduled to coincide  with the
expected toxic event. Due to the variability inherent in
POTW operations,  it may be  necessary to conduct
additional   POTW  performance  evaluation
investigations during the course of the TRE. For
example,  POTW performance evaluations  may  be
useful   when  performed   before  and  after
implementation of facility modifications, changes in
industrial user activities,  or variations  in effluent
toxicity.

Operations and Performance Review
The operations and  performance review involves the
evaluation of the maj or POTW unit processes using the
information described in Table  2-2.   This review
focuses on the secondary treatment system because
secondary treatment is responsible for removing the
majority of the conventional and toxic pollutants from
municipal wastewater. Deficiencies in this system are
more likely to result  in incomplete treatment  of
wastewater  toxicity.  For example, problems with
nitrification treatment may cause toxic concentrations
of ammonia to pass through in the effluent. Other unit
processes  to  be   evaluated  include   primary
sedimentation, disinfection, and advanced treatment
processes such as filtration.

Procedures  for evaluating and  improving  POTW
operations and performance are described in USEPA' s
Handbook on Retrofitting POTWs (USEPA, 1989c).
This handbook describes a two-step process to improve
POTW performance: comprehensive performance
evaluation  and  a  composite correction program
approach. The comprehensive performance evaluation
involves a thorough review of the POTW design and
operating conditions to identify problem areas. The
composite correction program involves the systematic
identification and implementation of improvements
with an emphasis on low-cost solutions. Other useful
sources of information include  a joint publication by
Water Environment Federation and American Society
of Civil Engineers  entitled  Design of Municipal
Wastewater Treatment Plants (WEF/ASCE,  1992a,
1992b)   and  Metcalf  and  Eddy's  Wastewater
Engineering Treatment, Disposal, and Reuse (1991).
Computer software programs, including USEPA's
POTW Expert (1990), have also been developed to
"troubleshoot" operations and performance problems.
In addition, USEPA (1993d) has a data base on
pollutant removal efficiencies (RREL Treatability Data
Base, Version 5)  for various treatment processes.
Although this guidance does not specifically address
toxicity, correcting conventional pollutant treatment
problems  and controlling  in-plant  toxicants  may
improve toxicity reduction.  In addition to the noted
guidance, public works managers are advised to use
the services  of  a professional engineer who  has
experience with the POTW treatment system.

Preliminary Treatment
Preliminary treatment processes that may be used to
enhance toxicity  control include equalization/storage
and oil and grease removal. Equalization basins can be
effective in dampening  the effect of slug loads of
toxicity or to equalize flow and organic loadings to
achieve consistent subsequent treatment of the influent
wastewater.  Oil and grease  removal can assist in
removing toxicants associated with oil and grease and
to minimize the impact of oil and grease on the POTW.
                 TRE Example

 A municipality in Texas experienced effluent toxicity
 that  was related to a volatile  organic compound
 entering the POTW. A pre-aeration system was to be
 added to the influent headworks or the grit removal
 system; however, before construction was started, a
 city employee noticed a strong odor in a sewer line
 that was related to the volatile toxicant.  The source of
 the volatile compound was identified and controlled.
 As   a  result,  effluent  toxicity  was eliminated
 (S. Bainter, personal communication, USEPA, Dallas,
 TX, 1998).
Primary Sedimentation
Primary treatment processes are designed to reduce the
loading of TSS, BOD5, and COD on the secondary
treatment system. Toxic pollutant removal also can
occur by sedimentation of insoluble or paniculate
wastewater constituents.  Optimal removal of both
toxic   and  conventional  pollutants  in   primary
                                                 17

-------
sedimentation  ultimately  reduces  the amount of
material to be  treated in the biological  treatment
process.

Primary clarifier performance can be evaluated by
comparing BOD5 removal to the surface overflow rate
(SOR), which is the average daily flow divided by the
clarifier surface area. A clarifier operating at an SOR
of less than 24 cubic meters per square meter per day
(nrYnrVday) [600 gallons pre day per square  foot
(gpd/sq ft)] should remove 35 to 45% of the influent
BODS.  A clarifier operating at an  SOR  of 24 to
40 rrrVrrrVday (600-1,000 gpd/sq ft) should remove
25 to 35% of the influent BOD5 (USEPA, 1989c). In
most cases, COD removal performance is comparable
to the BODS removal performance.   If the primary
clarifiers do not achieve the expected BOD5 or COD
removal, engineering  studies should be initiated to
determine the need for additional clarifier capacity.

Removal of toxicity  associated with TSS may be
enhanced by addition of coagulants  to the primary
clarifiers. The optimum conditions for coagulation and
flocculation of toxicants can be determined by j ar tests.
These tests are used to establish the optimum type and
dosageof coagulant, the propermixing conditions, and
the flocculent settling rates  for enhanced toxicant
removal (Adams et al., 1981).

A key operating parameter  for controlling clarifier
performance  is sludge removal.  Primary clarifiers
generally  function best  with  a  minimum  sludge
blanket.  Sludge withdrawal should be adjusted to
maintain the primary sludge concentration in the range
of 3 to 6% total solids (USEPA, 1989c).

Biological Treatment
Biological  treatment  is a  critical  process at  most
POTWs because it is the process that converts organic
matter and nutrients  to  settleable microorganisms.
Toxic pollutant removal during biological  treatment
can occur by biodegradation, oxidation, volatilization,
and adsorption  onto the biological floe. Key factors
affecting the removal of toxic pollutants are the rates of
biodegradation, tendency to volatilize, oxidize, or sorb
onto solids, and the degree to which the pollutants may
inhibit the treatment process.

Ammonia is a common cause of effluent toxicity at
POTWs that do not include nitrification treatment. As
noted by USEPA (1991a), ammonia is often present in
effluents in concentrations varying from 5 to 40 mg/L.
These concentrations can cause toxicity depending on
several factors that  affect the toxicity of ammonia,
including pH, temperature, DO, and TDS. A simple
TIE procedure for checking whether effluent toxicity
may be related to ammonia is described below (see
"TIE Phase I Tests").  Literature data on ammonia
toxicity (USEPA, 1985a) should only be used as a
general guide because ammonia toxicity is significantly
affected by slight pH changes.

The most commonly used biological treatment systems
can be defined as either suspended growth processes,
such  as  conventional  activated  sludge,  contact
stabilization,  and extended aeration; or fixed film
processes,  such  as trickling filters, denitrification
filters, and rotating biological contactors (RBC). To
simplify the discussion of biological treatment, the
following   subsections  focus  on   evaluating  the
performance   of  conventional  activated  sludge
processes  and related BNR processes, which are the
systems most widely used in POTWs.

Conventional activated sludge treatment is an aerobic
process that can be accomplished in one stage or zone.
BNR processes integrate carbon oxidation, as achieved
in conventional  activated  sludge  treatment,  with
treatment  stages   designed  for  nitrification,
denitrification, and  enhanced biological phosphorus
removal.   These stages require specific treatment
conditions, including anaerobic, anoxic, and aerobic
zones  in  the mixed liquors.  The  stages  may be
separated by physical divisions, non-discrete zones, or
by  operating cycle  (WEF/ASCE,  1992b).   The
sequence and sizing of the BNR stages depend on the
effluent nitrogen and phosphorus concentrations that
must be achieved.

Conventional  activated sludge  processes  remove
phosphorus and nitrogen in the course of converting
organic matter  to  new  biomass.    The  typical
phosphorus content of microbial cells is  1.5 to 2% on
a  dry-weight basis  (WEF/ASCE,   1992b).  BNR
processes enhance phosphorus removal by utilizing the
sequence of an anaerobic stage followed by an aerobic
stage,  which results in the  selection of a  biomass
population capable of concentrating phosphorus from
4 to 12% of the microbial cell mass.  Enhanced
nitrogen removal in BNR processes is a two stage
process: nitrification oxidizes ammonia to nitrite and
then to nitrate, and denitrification reduces the nitrate to
nitrogen gas. The nitrogen and phosphorus  removal
processes  can be used independently (e.g., oxidation
                                                 18

-------
ditches and A/Oฎ process, respectively) or can be
integrated into a combined nutrient removal process
(e.g.,  A2Oฎ and  Bardenphoฎ processes).   A  wide
variety of BNR systems are in operation,  some of
which are proprietary; therefore, specific information
on the process being studied  may  be obtained by
consultation with the system vendors.
                 TRE Example

 A United States east coast municipality implemented
 nitrification to achieve ,a seasonal NH3-N limit of
 1 mg/L (Engineering Science, Inc., 1994). ThePOTW
 typically achieved less than 1 mg/L NH3-N.  As a
 result, the POTW effluent eliminated chronic toxicity
 to fathead minnows (Pimephales promelas) from May
 1 through September  31 each year; however,  the
 effluent continued to be toxic during the remainder of
 the year. In an effort to comply with the permit limit
 for chronic toxicity, nitrification was extended for the
 full  year.  This  modification eliminated chronic
 toxicity to fathead minnows throughout the year.
The parameters that are typically used to evaluate the
operational capability of an activated sludge system
include organic loading,  oxygen requirement,  and
MCRT.  Additional important operating conditions
include the alkalinity requirement for nitrification, and
the BOD5 requirement for phosphorus  removal and
denitrification. Operating values for these parameters
can  be  compared  to  design  specifications  or
recommended criteria  to  determine how well the
processes are being operated.

Organic Loading
Organic loading affects the organic removal efficiency,
oxygen  requirement,  and  sludge production  of
activated  sludge processes.   The most common
measure  of organic loading in suspended  growth
processes is the F/M ratio, which is the organic load
removed per unit of mixed liquor volatile suspended
solids (MLVSS) in the aeration basin per unit time.
High F/M ratios (i.e., high organic loading to MLVSS)
will result in a low organic removal efficiency,  low
oxygen requirement, and high sludge production. Low
F/M ratios (i.e., low organic loading to MLVSS) will
lead to high organic removal efficiencies and  low
sludge production, but high oxygen requirements.
If the suspected toxicants are biodegradable or partition
to activated sludge,  the  MLVSS of the treatment
process should be increased to the maximum levels that
can be maintained at  the POTW.  The maximum
MLVSS  often  will  be  limited  by the  available
secondary clarifier capacity. It is important to consider
the effect of increased MLVSS on secondary solids
separation and the TSS concentrations of the clarifier
effluent.  The Patapsco Wastewater Treatment Plant
(WWTP) in Baltimore, Maryland, was operated at an
F/M ratio of 0.40  Ib BOD5/lb MLVSS-day instead of
the design F/M ratio of 0.55 Ib BOD5/lb MLVSS-day,
because  the POTW could  not  achieve consistent
wastewater treatment at the higher organic loading.
The increased MLVSS levels were thought to be
necessary because of the toxic effect that industrial
wastewaters were having on  the  activated sludge
biomass (Slattery, 1987). For optimal treatment, it may
be necessary to maintain F/M ratios that are on the low
end of the range typically observed for biological
treatment processes.  The F/M ratio in an activated
sludge system is generally maintained in the range of
0.2 to  0.4 Ib BOD5/lb MLVSS-day for  conventional
activated sludge, 0.05 to 0.15 Ib BOD5/lb MLVSS-day
for extended aeration, and 0.2 to 0.6 Ib BOD5/lb
MLVSS-day for  contact stabilization  (Metcalf  and
Eddy,  1991).   The recommended F/M  ratio  for
integrated BNR processes is generally in the range of
0.1 to 0.25 Ib BOD5/lb  MLVSS-day  (Metcalf  and
Eddy, 1991).

Influent BOD5 concentrations should be high  relative
to phosphorus levels to ensure optimal phosphorus
uptake and removal  in BNR processes.  Although
optimum conditions  vary according to the  system
design, the ratio of total influent BOD5 (TBOD5) to
influent TP should be 20:1 to 25:1 to meet an effluent
TP level of 1.0 mg/L or less.  More importantly, the
ratio of influent soluble BOD5 (SBOD5) to influent
soluble phosphorus (SP) should be 15:1 (WEF/ASCE,
1992b).

The presence of biodegradable  material   also is
necessary for denitrification in the first anoxic stage of
BNR processes.   For example, the Water Research
Commission (1984) found that the TKN to COD ratio
should be  less than 0.08 to accomplish  complete
denitrification with the Bardenphoฎ process. In most
cases, carbon must be added to the anoxic stage either
by internal recycling of BOD5 in process streams (e.g.,
nitrified  effluent) or  by chemical addition (e.g.,
methanol or acetate).
                                                 19

-------
Oxygen Requirement
Microorganisms in the activated sludge system require
oxygen to metabolize organic material and nutrients
and breakdown biodegradable toxicants. Oxygen in
diffused air or pure oxygen systems also may oxidize
toxicants. Oxygen deficient conditions can result in
lower treatment efficiency and, as a result, a greater
potential for pass-through of toxic material. To ensure
an adequate  supply  of oxygen, the DO  level for
conventional  activated  sludge  (carbon  oxidation)
should be at least 2  mg/L during average loading
conditions  and 0.5  mg/L  under peak  loadings
(WEF/ASCE, 1992a).  Typical air requirements are
1,500 cu ft/lb BOD5 load for conventional activated
sludge and contact stabilization, and 2,000 cu ft/lb
BODS  load for extended aeration (USEPA, 1989c).
Air requirements for nitrification are higher because
4.2 to 4.6 mg of oxygen are required per mg of NH3-N
oxidized as compared to 0.6 to  1.1 mg of oxygen
needed per mg of  BOD5  oxidized (WEF/ASCE,
1992b).  DO concentrations of 2 to 2.5 mg/L are
needed for nitrification in activated sludge processes
with short  retention  times.  In integrated activated
sludge/nutrient removal processes, sufficient oxygen
should be provided to achieve carbon oxidation and
complete nitrification at the maximum daily loading
rate.

Some of the oxygen consumed in the aerobic stage is
in the form of NO3-N. This oxygen source can be
recovered in  the denitrification process,  which  is
generally located in an anoxic stage at the head of the
BNR system.  Approximately 2.86 mg of oxygen is
recovered for each mg of NO3-N reduced by biological
denitrification.  Internal recycling from the aerobic
stage to the anoxic stage  can  decrease the oxygen
required for nitrification by 50 to 60%. Carryover of
molecular oxygen from the aerobic stage to the anoxic
stage should be minimized by regulating the internal
recycle rate. Generally, the recycle rate should be no
more than three to four times the influent flow rate for
these systems (WEF/ASCE, 1992b).

The transfer of oxygen from the gas phase to the liquid
phase is a function of the aeration equipment and the
basin mixing conditions.  USEPA (Handbook for
Retrofitting POTWs, 1989c) describes a procedure for
estimating the oxygen transfer capacity in aeration
basins  based  on equipment specifications.  Another
estimate of  oxygen  transfer  capacity  involves
comparing  OUR of  the biomass  to the  calculated
theoretical oxygen demand for the  aeration system
(USEPA,  1989c).  If the OUR  results indicate an
oxygen demand that is greater than the calculated
oxygen demand, the oxygen supply may be inadequate.
The opposite case (i.e., higher  theoretical oxygen
demand than actual  oxygen demand) is  preferred;
however,  a  substantial difference  may indicate
inhibition of biomass activity.
                 TRE Example

 OUR measurements were used to document the start-
 up performance of the  activated  sludge treatment
 process at the Patapsco Wastewater Treatment Plant
 (Botts et-al., 1987). During the start-up, the OUR of
 the biomass averaged 20 milligrams O2 per liter per
 hour per gram MLSS (mg O2/L/hr/g MLSS), and the
 POTW frequently exceeded its conventional pollutant
 permit limits.   As the biological system became
 acclimated to the wastewater, the effluent quality
 improved  and  the biomass  OUR increased to an
 average of 50 mg O2/L/hr/g MLSS.
Oxygen is not desirable in denitrification processes and
the initial (anaerobic) stage of biological phosphorus
removal processes. Molecular oxygen must be absent
for denitrifying organisms to reduce NO3-N to nitrogen
gas.    Phosphorus  removing  organisms  require
anaerobic  conditions  to  accomplish   the  initial
phosphorus release step  (resynthesis occurs in the
subsequent   aerobic   zone).   NO3-N  and  DO
concentrations in the anaerobic zone  should be kept
below 1 mg/L (WEF/ASCE, 1992b).  Although mixing
is usually required in anaerobic and anoxic basins, it is
minimized to prevent oxygen transfer to the mixed
liquors. Also, DO carry over from the  aerobic zone to
the anoxic (denitrification) zone should  be regulated
through independent control of aeration equipment at
the end of the aerobic zone.

Mean Cell Residence Time
In the course of biological treatment, activated sludge
microorganisms convert some of the  organic matter
and nutrients in the wastewater to new cell mass.
Toxic constituents may also be degraded or adsorbed
onto the biomass. To achieve optimal treatment, the
biomass concentration in the aeration tank is held at a
constant level by routinely wasting the excess sludge.
Sludge mass control can be practiced by maintaining a
consistent average  age  of  activated sludge  (i.e.,
MCRT) in the system.  The MCRT is calculated by
dividing the total sludge  mass in the system by the
                                                20

-------
amount of sludge that is wasted each day (i.e., Ib/lb per
day).

The optimal MCRT for toxicity reduction will depend
on the type of toxicant(s) in the wastewater.  Some
compounds may be more efficiently removed by a
younger biomass (low MCRT) and other toxicants are
treated better with an older biomass (high MCRT)
(Metcalf and Eddy, 1991). In general, biodegradable
toxicants  are  more efficiently removed  using  a
relatively long MCRT (WEF/ASCE, 1992a); therefore,
the MCRT may be set at the high end of the range of
values  typically   used  for  biological  treatment.
Hagelstein and Dauge (1984) also found improved
toxicity reduction of a petroleum waste  at MCRTs
greater than 10 days.  Typical MCRTs for aeration
processes are 6 to 12 days for conventional activated
sludge, 10 to 30 days for contact stabilization, and 20
to 40 days for extended aeration (USEPA, 1989c).
System MCRTs for nitrification and BNR processes
are generally long (20 to 40 days) because the growth
rates of nitrifying organisms are slower compared to
those  for  heterotrophic   organisms   found  in
conventional activated sludge systems (WEF/ASCE,
1992b).  Overly long MCRTs should be avoided
because subsequent denitrification treatment may be
adversely affected if essential carbon has been depleted
in the nitrification/carbon oxidation stage.
                 TRE Example

 In 1992,Novartis Crop Protection, Inc., in cooperation
 with Makhteshim-Agan of North America, Inc., the
 two  principal  manufacturers  of organophosphate
 insecticides in North America, evaluated the removal
 of diazinon  and chlorpyrifos by various treatment
 methods (Novartis, 1997). Anecdotal evidence from
 other studies (Fillmore et al., 1990) and  treatability
 studies by Novartis suggested that adsorption onto
 solids was the dominant  mechanism for removal of
 organophosphate insecticides.  The treatability tests
 performed in the 1997 study showed that  about 30%
 of the diazinon and 85 to 90% of the chlorpyrifos
 present in POTW primary influent samples  is adsorbed
 onto primary influent solids and approximately 65 to
 75% of the  diazinon added to the mixed liquor is
 adsorbed onto the biomass. Diazinon adsorption was
 greater for a 30-day MCRT biomass than for a 15-day
 biomass.  Chlorpyrifos  strongly adsorbed to the
 biomass; none remained  after biological treatment.
 These results suggest that longer MCRTs may improve
 removal of organophosphate insecticides.
Additional  Considerations  for  BNR  Process
Control
Additional considerations for BNR process control
include  maintaining  sufficient  alkalinity, proper
management of sludge processing sidestreams, and
achieving efficient solids separation in the secondary
clarifier.    Nitrification  reduces the  wastewater
alkalinity by 7.2 mg/L as CaCO3 for each mg of NH3-N
oxidized. However, about 40 to 50% of the alkalinity
destroyed  by  nitrification  can  be  restored  in
the  denitrification  process.    The  carbonate/CO2
equilibrium in the mixed liquor determines the pH. As
a general rule, alkalinity should be maintained above
50 mg/L in the nitrification process in order to keep the
pH high enough for optimal treatment.

Secondary Clarification
In order for activated sludge and BNR processes to
operate  efficiently,  the secondary  clarifier  must
effectively separate solids from the liquid phase and
concentrate  the solids for subsequent return to  the
aeration basin.  In addition to clarifier design, solids-
liquid separation is influenced to a large degree by the
aeration basin operating conditions such as DO levels,
F/M ratio, and MCRT. If the MLVSS and MCRT of
the aeration basin is  to  be maximized  for toxicity
control, it is important to consider the impact of this
change on secondary solids separation and effluent
TSS concentrations. Sludge settling characteristics are
affected by how the aeration basin is operated. Low or
high DO levels in the  aeration basin can result in the
growth of filamentous bacteria (e.g., Norcardia spp.
and Sphaerotilus natans, respectively) that can hinder
solids  settling,  whereas DO  levels of 2 to 4 mg/L
promote the growth of "zoogleal-type" bacteria, which
aggregate into fast settling floes. At very high organic
loadings (high F/M), the activated  sludge can be
dispersed and will not settle well.  This condition was
observed at the East Side Sewage Treatment Plant in
Os wego, New York, which experienced sludge bulking
due to high effluent organic loadings (USEPA, 1984a).
Sludge settleability was improved by increasing  the
MCRT and the sludge return rate.

The performance of  secondary clarifiers in solids-
liquid separation is dependent on a variety of factors
including clarifier configuration, SOR, clarifier depth
at the weirs, the type  of sludge removal mechanism,
and the return sludge flow rate. USEPA's Handbook
on Retrofitting POTWs (1989c) describes a system for
scoring secondary clarifier performance based on these
factors.   Design clarifier  SORs  for conventional
                                                 21

-------
activated sludge processes are typically in the range of
400 to 800 gpd/sq ft (Metcalf and Eddy,  1991).  In
general,  integrated BNR  processes  require  larger
clarifier capacity than conventional activated sludge
processes, particularly where effluent quality must be
high.

Clarifier capacity is important for phosphorus removal,
because phosphorus is associated with the biomass,
which can be carried over into the final effluent as
residual solids.  At peak sustained flow,  a clarifier
SOR  of 800 gpd/sq ft is recommended to achieve a
final  effluent TP concentration  of 2  mg/L.  Lower
effluent TP limits may require chemical treatment
using a metal salt of iron, aluminum, or calcium, and,
perhaps, a follow-on filtration process. For example,
effluent phosphorus levels at the Jerry Sellers POTW
in Cocoa, Florida, were reduced from an average of 2.9
mg/L to less than  0.2 mg/L with the addition of
aluminum sulfate, commonly referred to as alum
(WEF/ASCE, 1992b). In general, effluent TP levels of
0.2 to 0.5 mg/L can be met through chemical addition
with a clarifier capacity of 500 gpd/sq ft (WEF/ASCE,
1992b).

Process Sidestreams and Wastewater Bypasses
Some wastewater and sludge treatment processes can
produce sidestream wastes that may have a deleterious
effect on the wastewater treatment system or might
contribute to effluent toxicity.  In addition, raw or
partially treated wastewater that bypasses part or all of
the treatment system can add substantial toxicity to
POTW discharges (Mosure et al.,  1987).

Examples of  POTW  sidestreams include  sludge
processing wastewaters (from thickening, digestion,
and dewatering of sludges), cooling water blowdown,
incinerator scrubber blowdown,  and backwash from
tertiary filters. Sidestreams from anaerobic digestion
and sludge dewatering can contain high concentrations
of BODj, COD, nitrogen, and phosphorus that  can
represent a significant loading to the aeration  basin.
Nitrogen and phosphorus in sidestreams are a concern
for BNR processes, which can be compromised unless
the loadings are removed, equalized, or  separately
treated.  Also, some sidestreams may contain toxic
materials such as metals and cyanide that may pass
through the  POTW.  For example, cyanide maybe
formed during incineration of biosolids. Once formed,
the cyanide  may  be  captured in the incinerator
scrubbers and introduced into the treatment system via
the scrubber waste stream.  If the treatment process
does not degrade the cyanide, toxic concentrations may
be discharged in the POTW effluent. Also, sufficient
amounts of cyanide may be present to cause inhibition
of the biological treatment process, which leads to the
release of more cyanide in the POTW effluent.

In some municipalities, storm water and sewage are
still collected in the same sewer system. When a large
storm event occurs, the CSO is  often diverted away
from all  or part of the treatment system to prevent
hydraulic over loading.  In some cases, untreated
overflows or bypasses may be directed into the POTW
effluent,  which can cause the discharge of relatively
high  concentrations  of toxic  and  conventional
pollutants.

The POTW performance evaluation should include a
review of data on process  sidestreams,  wastewater
bypasses, and overflows that are discharged into all or
part of the POTW or into the final effluent. Additional
analytical and toxicity data may be  needed  to
characterize the levels of toxic pollutants and toxicity
in these waste streams. Information on the frequency,
volume,  and  toxicity  of  sidestream  discharges,
bypasses, and overflows also can be compared  to
historical effluent toxicity data to evaluate possible
trends or relationships. This information can be used
to determine if the discharges are a significant source
of pollutants or toxicity, and whether current treatment
practices  are  sufficient  to remove  toxicity.   If
necessary, consideration may be given to enhanced
treatment of process sidestreams to remove toxicants
such as metals. Enhanced treatment may  involve
changing the dosage of currently used coagulants
applied for solids separation or adding new coagulant
and flocculent  aids.   Bench-scale jar tests can be
performed to determine the optimum type and dosage
of coagulant and the appropriate treatment conditions.

Advanced Treatment Processes
Advanced treatment processes may be included  in
some  POTWs to achieve pollutant removal beyond
what  is  provided by  biological treatment.   These
processes may include filtration, adsorption, chemical
treatment, air stripping, and breakpoint chlorination.
Of these  processes, chemical treatment and filtration
are  most  commonly used in POTWs, particularly for
enhanced phosphorus removal.
                                                22

-------
Chemical Treatment
Chemical treatment can contribute to toxicity where
toxic residual concentrations or contaminants in the
product are present in the final effluent  Chemicals
used in the latter stages of wastewater treatment are of
particular concern because final treatment processes
(e.g., tertiary clarification, chlorination) are less likely
to remove residual concentrations (Note: potentially
toxic disinfection  byproducts, including  residual
chlorine, are discussed in a following subsection).
When  used  wisely,  treatment additives  such  as
coagulants,  flocculent  aids, and hydrogen peroxide
(H2O2) can also improve toxicity treatment.

Chemicals of concern in POTWs include chemicals
formerly classified as hazardous waste. Under RCRA,
a  hazardous waste sold to a  POTW is  no longer
considered a hazardous  waste.  According to 40 CFR
Part 261.21(c)(5)(ii):

    "A material  is 'used'  or 'reused' if it is...
    employed  in  a   particular   function  or
    application as  an  effective substitute  for a
    commercial  product   (for example,   spent
    pickle liquor used as phosphorus precipitant
     and  sludge  conditioner in  wastewater
    treatment)."
                  TRE Example

  An example of the potential problems that may occur
  with process chemicals was the use of a dechlorination
  agent at several City of Houston wastewater treatment
  plants (S. Bainter, personal communication, USEPA,
  Dallas, TX, 1998).  The City had routinely passed
  effluent toxicity tests until a dechlorination chemical
  was obtained from a new vendor.  When the chemical
  was applied, effluent toxicity was observed at each of
  the POTWs. At the time, the City did not know the
  chemical may be a problem and proceeded to retain
  consultants to conduct TREs at the facilities. In the
  meantime, the supply of the dechlorination chemical
  was depleted and the city turned to a new source of the
  chemical.  When the new chemical Was applied, the
  POTWs started to pass  the effluent toxicity  tests.
  POTW staff can avoid similar problems if vendors are
  queried  about potential contaminants in the waste
  chemicals  or toxicity tests are performed on  product
  samples to verify their suitability.
lime,  alum, sodium aluminate, ferric chloride, and
ferrous sulfate. These coagulants have generally not
been found to be toxic at the concentrations typically
used for phosphorus removal.   For example, alum
dosages as high as 20 mg/L did not cause chronic
effluent toxicity in treatability tests conducted at the
City of  Durham, North Carolina  (Appendix  D).
Nonetheless,  steps  should  be  taken to  prevent
excessive and inadvertent chemical use. Also, each
chemical  additive  used  for  treatment should  be
evaluated as a potential source of toxicity, not just
suspect chemicals.

If toxicity  is  associated with  suspended solids,
chemical  treatment conditions  may be modified to
enhance toxicity removal. The optimum conditions for
coagulation can be determined by conducting jar tests.
These tests can be used to establish the optimum type
and dosage of coagulant, the proper mixing conditions,
and  the  flocculent  settling  rates  for improved
phosphorus and/or toxicity removal (Adams  et al.,
1981).
 Chemical treatment is often practiced for phosphorus
 removal at POTWs.  Typical coagulant aids include
                 TRE Examples

  Studies have been performed to evaluate the reduction
  of organophosphate insecticides by chemical treatment
  (Novartis, 1997). Chemical precipitation using ferric
  chloride and polymer  was found to only slightly
  reduce diazinon levels.  No major change in diazinon
  concentrations was observed whether the coagulants
  were added  to  primary  wastewater or secondary
  treated wastewater prior to clarification. Chlorination
  treatment was effective in reducing diazinon  from
  secondary clarifier effluent; however, chronic toxicity
  was unchanged. Qualitative results suggest that the
  chlorine oxidized diazinon to diazoxon, a byproduct
  that exhibits similar toxic effects as diazinon. The
  results of additional treatments for diazinon are given
  in Appendix H.
  A study conducted for  San  Francisco Bay area
  POTWs also evaluated  the effect of chlorine  on
  organophosphate  insecticide   concentrations
  (AQUA-Science, 1995). This study evaluated the use
  of household bleach as  a measure  that residential
  customers  could use to  degrade diazinon in spray
  container rinsate and chlorpyrifos from pet flea washes
  prior to disposal into the sewer. Samples of tap water
  were  spiked  with  diazinon  (60.0  ug/L)  and
  chlorpyrifos (10.0 ug/L) and treated with either 0.005
  or 5% solutions of household bleach for 24 hours.
  Results  showed that  both bleach concentrations
                                                   23

-------
  reduced concentrations of the insecticides by 86 to
  92%. The study suggested that household bleach may
  be an effective pretreatment for waste solutions of the
  two insecticides prior to disposal. Additional studies
  are planned to further define bleach exposure times
  and concentrations under actual use conditions and to
  characterize the chemical oxidation products produced
  by the chlorine treatment. Additional information on
  this study is presented in Appendix F.

  H2O2 has been used by a North Carolina municipality
  to control toxicity associated with non-polar organic
  toxicants  (Aquatic  Sciences  Consulting,  1997).
  Although the specific  toxic compounds were  not
  identified; jar tests with H2O2 showed a substantial
  reduction in chronic toxicity to Ceriodaphnia dubia
  (C. dubia) at dosages ranging from 1 to 10 mg/L H2O2.
  Since the City began adding H2O2 to the POTW
  effluent (final concentration of 5 to 7 mg/L), results of
  a single C. dubia test show that the effluent NOEC
  was reduced from <15 to 90%.
Granular Media Filtration
Granular media  filtration is  usually  applied  after
biological treatment to remove residual suspended
solids, paniculate BOD5, or insoluble phosphorus.
Rlter influent is often chemically pretreated to enhance
removal of suspended  solids and  phosphorus.  In
addition to the metal salts noted above, polyelectrolytes
may be added to improve coagulation and flocculation
of  chemically treated  influents.  Typical  polymer
dosages are 0.5 to 1.5 mg/L for settling of flocculent
suspensions before filtration and 0.05 to 0.15 mg/L
when added directly to the filter  influent.  Some
polymers can be toxic to aquatic life (Hall and
Mirenda, 1991); therefore, polymers  used  in the
filtration process should be evaluated for the potential
to contribute to effluent toxicity.

Poor filter  performance  should  be  investigated,
especially   the pass-through  of  potentially toxic
material. Loss of suspended solids and other pollutants
may result from high hydraulic and solids loadings,
excessively long filtration  cycles,  and incomplete
backwashing and cleaning of the filter.

Disinfection
Disinfection  is generally achieved by  treating the
secondary  effluent with  chlorine and allowing a
sufficient  contact  period  prior  to  discharge.
Alternative disinfection  practices such  as ultraviolet
(UV) radiation are becoming more popular because of
 concerns about the effects of chlorine on aquatic life
 and human health.  :

 The chlorine disinfection process should be carefully
 evaluated because residual chlorine and other by-
 products of chlorination  (i.e.,  mono-  and dichloro-
 amines, nitrogen trichloride) are toxic to aquatic life
 (Brungs, 1973). Chlorine dosages are usually based on
 the level of residual chlorine to be maintained in the
 final effluent as specified  in the NPDES permit. The
 POTW performance evaluation should  focus on the
 minimum amount of chlorine that can be applied to
 achieve the required residual chlorine concentration. In
 some cases, the TRC level  specified in the NPDES
 permit may be sufficient to cause effluent toxicity. In
 general, TRC concentrations above 0.05 mg/L are a
 concern (D.  Mount personal communication, AScI
 Corp, Duluth, Minnesota,  1991), although its toxicity
 will depend on the effluent matrix and the species used
 for effluent monitoring. Residual chlorine levels can
 be compared to toxicity data reported in the literature
 (USEPA, 1984b) to determine  if chlorine may be a
 potential cause of effluent toxicity. If dechlorination
 is practiced following chlorination, information on the
 type  and amount of oxidant-reducing material also
 should be obtained.

 A chlorination process that is not continuously adjusted
 to varying flow and chlorine demand may  cause
 effluent toxicity. Fortunately,  this problem can be
 corrected easily by more frequent monitoring of the
 chlorine residual in effluent samples and more frequent
 adjustments  in  the   addition  of   chlorine  and
 dechlorination  chemical.    Flow-proportional  feed
 equipment for chlorine and dechlorinating  agents
 should be used to minimize the potential for excess
 chemical addition.

 TIE Phase I Tests
 TIE  Phase I tests  (USEPA 1991a,  1992a) can be
 conducted in parallel with the above operations and
 performance review to obtain information on the types
 of compounds causing effluent toxicity. An overview
 of the Phase I procedure is described in Section 4 of
 this guidance.

 TIE  Phase I  testing  in  the  POTW  performance
 evaluation focuses on characterizing toxicants that may
be present in the  effluent  because of inadequate
treatment performance or routine operating practices.
Phase I  results,  when taken together with the POTW
performance evaluation data, may provide  important
                                                 24

-------
clues about possible in-plant toxicants. Based on this
information, treatability tests  may  be designed to
evaluate methods for removing the suspected effluent
toxicants.

TIE Phase I testing includes several characterization
steps that can be used to indicate the presence of "in-
plant toxicants" such as ammonia and chlorine.  One
step involves pH adjustment of the effluent sample to
three distinct pHs, such as pH 6, 7, and 8 prior to
toxicity testing to indicate the effect of pH changes on
effluent toxicity.  The pH  adjustment will  shift the
equilibrium concentration of ammonia  between its
toxic form (NH3) and its essentially nontoxic form
(NH4).   As pH increases, the percentage of total
ammonia (NH3 and NH4) present as NH3 increases. If
adjusting the effluent sample pH to 8 increases the
toxicity and if lowering the effluent sample pH to 6
decreases the toxicity, the identity  of  the effluent
toxicant would be consistent with ammonia (Section
4). Another  Phase I step  is  designed to indicate
whether wastewater oxidants, such as TRC (i.e., free
chlorine and mono- and dichloroamines), are causing
toxicity.   Sodium thiosulfate, a reducing agent, is
added  to eliminate TRC and  other oxidants.  The
thiosulfate is added to serial dilutions of the effluent
sample with 1 or 2 levels added across the dilutions.
Toxicity tests on samples with and without thiosulfate
treatment are used to indicate if oxidants such as TRC
may be causing effluent toxicity.

It is important to note that each of  the TIE Phase I
characterization steps  described above  addresses a
broad class of toxicants rather than  specific effluent
constituents,  such  as  ammonia and TRC (USEPA,
1991a).  For example, the toxicant  affected by pH
adjustment may be a  pH  sensitive  compound that
behaves  in the same manner as ammonia.  Also, the
oxidants  that  are  neutralized  in   the  thiosulfate
treatment step include bromine, iodine, and manganous
ions in addition to TRC. Also, some cations, including
selected  heavy metals, are complexed by thiosulfate
and may be rendered nontoxic (Hockett and Mount,'
1996).   Therefore, the Phase I  results should be
compared with information from the POTW operations
and performance review to  substantiate the  evidence
for a particular toxicant. Using the previous example,
the assumption that TRC is causing oxidant toxicity
would be corroborated if operations data show that
toxic concentrations of chlorine are maintained in the
final effluent (see Table 2-1 for levels of concern).
Organophosphate insecticides also have been identified
as causes of effluent toxicity at POTWs (Ankley et al.,
1992; Amato et al., 1992; Bailey et al., 1997; Botts et
al.,  1990; Burkhard and Jenson, 1993).  TIE Phase I
procedures that affect Organophosphate insecticides
include CIS SPE and treatment  with  a metabolic
blocker, piperonyl butoxide (PBO). PBO can be added
to effluent samples or methanol eluates from C18 SPE
columns to block the toxicity of metabolically activated
toxicants like Organophosphate insecticides. PBO has
been shown to block the acute toxicity of diazinon,
parathion,  methyl  parathion,  and  malathion  to
cladocerans, but does not decrease the acute effects of
dichlorvos, chlorfenvinphos, and mevinphos (Ankley
etal., 1996). A reduction in toxicity by PBO treatment
together with toxicity removal by the C18 SPE column,
recovery from the CIS SPE column, and effluent
concentration data can provide strong evidence for the
presence  of Organophosphate  insecticides.    An
exception is chlorpyrifos, which is not recovered well
from C18 SPE columns (see Appendix F).

Conventional Wastewater Treatability Testing
The operations and  performance information may
identify areas in the POTW where improvements in
conventional pollutant  treatment may  reduce  the
pass-through of  toxicity.   This  information  and
the optional TIE Phase  I data also may indicate in-
plant  sources  of  toxicants  such  as  process
sidestreams or over chlorination. Using these data, a
wastewater treatability program may be devised and
implemented to assess in-plant options for improving
conventional treatment  and  eliminating  in-plant
sources of toxicity.

Treatability  studies  are  recommended  prior to
comprehensive TEE testing (Section 4) in situations
where improvements in treatment operations  and
performance  are needed  to  attain   acceptable
conventional pollutant treatment.   Otherwise,  TIE
testing of poor quality effluents could lead to erroneous
conclusions about the nature of effluent toxicity.  For
example, inadequate conventional pollutant treatment
could cause toxic materials to pass through the POTW
that would  otherwise be removed.   In  the POTW
performance evaluation, treatability studies  should
focus on conventional pollutant treatment deficiencies
that are suspected of contributing to effluent toxicity.
The scope of the treatability studies program should be
based  on clear evidence of a  consistent treatment
deficiency causing toxicity over time.  If sufficient
                                                 25

-------
information   is   not  available   to  develop   a
straightforward treatability program, additional data
must be gathered in the subsequent stages of the TRE
before in-plant toxicity control (Section  6)  can  be
evaluated.

Treatability studies can vary from a simple evaluation
such as testing the effect of TRC reduction on effluent
toxicity to an extensive effort involving long-term
bench- and pilot-scale work. Prior to beginning these
studies, the POTW operations and performance data
and the optional TEE Phase I results should be carefully
reviewed and an appropriate treatability test program
should be developed using best professional judgment.
The nature and variability of effluent toxicity must be
completely assessed (Section 4) prior to implementing
an extensive treatability effort.

A treatability program can be devised to evaluate
modifications  in  existing   treatment   processes.
Evaluating new or additional treatment units should be
attempted only after further effluent characterization
studies  (i.e., TIE)  have been  performed.   POTW
performance evaluation treatability testing may involve
physical/chemical  treatment approaches,  such  as
coagulation and precipitation, solids  sedimentation,
granular media filtration, powdered activated carbon
adsorption, or biological treatment approaches, such as
activated sludge or sludge digestion.

Toxicity control is the ultimate  goal of the TRE;
therefore, toxicity tests should be performed in addition
to the  conventional  pollutant analyses normally
conducted in treatability studies.  Toxicity tests are
used  to assess  the   capability  of  the  treatment
modifications for toxicity reduction. In some cases, the
waste streams to be tested may exert  a high oxygen
demand and  aeration may be needed to maintain a
minimum DO level of 4 mg/L in the toxicity test.
Aeration may affect toxicant characteristics; therefore,
it may be necessary to use an alternative test method,
such as Microtoxฎ, that is not  affected by low DO.
Side-by-side  testing  with alternative methods  or
species and the definitive test can be used to select a
procedure that correlates well with the definitive test.
This  initial testing  will  help to  ensure  that  the
alternative test method or species is sensitive to the
effluent toxicants of concern.

The following subsections briefly describe some of the
treatability tests that  can be  used to determine if
improvements  in  existing  conventional pollutant
treatment will reduce effluent toxicity. If bench-scale
tests suggest that toxicity can be reduced, follow-up
pilot-scale  or full-scale testing is recommended to
confirm the initial results. As shown in Figure 3-1, if
this testing is successful in identifying improvements
in conventional pollutant treatment that will  achieve
acceptable levels of effluent toxicity, the TRE proceeds
to the selection and implementation of those options
(Sections 6 and 7).  If, however, the treatability data
indicate  that improved in-plant treatment will not
reduce effluent  toxicity to acceptable levels, other
approaches must be investigated, including TIE testing
(Section 4).

Chemical  Treatment
Chemical  treatment  may  be  applied  in  primary
sedimentation, secondary clarification, filtration, and
sidestream treatment processes. As noted above, jar
tests can be used to determine the optimum type and
dosage of chemical, the proper mixing conditions, and
the flocculent settling rates for improved conventional
pollutant and toxicity removal (Adams et al., 1981).
As noted, some chemical additives, including polymers
(Hall and Mirenda, 1991), can be toxic; therefore, the
toxicity of the chemicals should be evaluated as part of
treatability testing.

Sedimentation
Sedimentation processes remove suspended solids or
flocculent  suspensions  from the  wastewater.   In
general, sedimentation in POTWs is characterized by
flocculent  settling  for  wastewater  (i.e.,  primary
clarification) and zone settling for mixed liquors (i.e.,
secondary clarification) and sewage sludges  (sludge
thickening).

Flocculent settling rates can be converted to a  clarifier
SOR by measuring the flocculent percent removal with
time in a settling column test (Adams et al., 1981). If
coagulants  are needed, the optimum conditions for
flocculation can be determined from jar tests, as noted
above. A series of settling column tests can  then be
performed  to compare particle settling profiles for
various coagulant doses and mixing conditions.

Zone settling also can  be evaluated in settling column
tests. The settling velocity of mixed liquor or sludge is
determined by measuring the subsidence of the liquid-
solids interface over time (Adams et  al., 1981).  A
series of tests are performed using the anticipated
                                                 26

-------
range of suspended solids loadings to the clarifier.
Test results are used to calculate a solids flux curve
that can be used for clarifier design.

Activated Sludge
Continuous flow and batch biological reactor tests can
be used to assess pollutant and toxicity treatability, and
to predict the process kinetics of an activated sludge
system. A series of bioreactors are generally operated
under a range of MCRT values to determine optimum
operating conditions (Adams et al., 1981).

The  operational performance of bioreactors can be
evaluated by measuring pollutant removals,  OUR,
MLVSS, and the zone settling velocity (ZSV) of the
sludge. These measurements are used to determine the
biodegradation kinetics of the wastewater, the potential
for treatment inhibition,  and the preferred  sludge
settling  conditions.    Samples  of the  influent,
intermediate  treatment stages, and  effluent  of the
bioreactors can also be tested for toxicity to evaluate
the system's toxicity reduction capability. Appendix D
provides an example of the use of batch treatability
tests to evaluate toxicity reduction in a BNR process.

If results of bench-scale treatability tests suggest that
full-scale  treatment will reduce  effluent toxicity,
follow-up  pilot-scale  or  full-scale  tests  are
recommended to confirm the results.

Granular Media Filtration
Toxicity removal by filtration can be evaluated in
bench-scale tests or in  full-scale tests  of existing
processes. The main  parameters to be evaluated in
filtration testing include hydraulic loading rate, media
type and configuration, and, if necessary, type and dose
of chemical coagulant (Adams et al., 1981). Filtration
testing results  can be used to correlate removal of
suspended solids and toxic compounds with  loss of
toxicity. These results are ultimately used to establish,
the optimum design and operational conditions for
conventional pollutant and toxicity removal, including
filter type and loading rates, media  characteristics,
backwashing, and headless development. Examples of
the use of filtration in toxicity treatability studies are
presented in Appendices C and D.

Activated Carbon Adsorption
Activated carbon may be applied in powdered form to
the activated sludge process or may be used in granular
form in a post treatment process (e.g., columns). The
capability of carbon adsorption for treatment of organic
wastewater constituents or toxicity is determined by
conducting batch isotherm tests and continuous-flow
tests (Adams et al., 1981).

The effectiveness of  carbon in removing BOD5,
selected  organic contaminants  (e.g.,  phenols),  or
toxicity is predicted by adding varying amounts  of
powdered activated carbon  (PAC) to wastewater
samples  and  measuring  removal  of the  organic
constituents or toxicity. The equilibrium relationship
between  a wastewater and carbon usually can be
described either by a Langmier or Freundlich isotherm.
A plot of equilibrium concentration  versus carbon
capacity  is  used  to  select  the required  PAC
concentration to add to activated sludge processes.

Continuous-flow tests are required to confirm the batch
isotherm results. PAC tests involve adding PAC to
bench- or pilot-scale biological reactors and monitoring
the removal of the organic wastewater constituents or
toxicity.
                 TRE Example

 Activated carbon was investigated as a toxicity control
 option in the TRE at the Linden Roselle Sewerage
 Authority's (LRSA) POTW in New Jersey (Appendix
 G). Both PAC and granular activated carbon (GAC)
 were expected to remove non-polar organic toxicity in
 the effluent; however, the costs were determined to be
 prohibitive. It was also anticipated that carbon would
 concentrate the toxicants in the mixed liquor and cause
 unacceptable sludge quality.
Pretreatment Program Review
POTW pretreatment program data (Table 2-3) may
provide information that can be used in subsequent
steps of the TRE such as the toxicity source evaluation
(Section 5). Information on the main trunk lines and
the types of indirect dischargers in the sewer collection
system can be used to devise a sampling strategy for
tracking the sources of toxicants or toxicity.  In some
cases, the pretreatment program data may be sufficient
to identify the sources of effluent toxicants identified
in the TIE. In most cases,  however, additional data,
such as wastewater flow and toxicant concentrations in
indirect discharges, will be needed to track the sources
of toxicants or toxicity.

The information needed to conduct a toxicity source
evaluation is presented in Section 5. In a USEPA TRE
                                                  27

-------
research study conducted in Linden, New  Jersey,
pretreatment  program  data   on   wastewater
characteristics of the main sewer lines and industrial
dischargers were used to develop a comprehensive
toxicity tracking program (see Appendix G). Sources
of toxicity were successfully identified by devising a
sampling schedule that accounted for periods of normal
industry activity and periods of temporary shut-down
for industry maintenance. The level of toxicity from
the industries was found to  vary  with the industry
production schedules.

It may be possible, in a few cases, to identify the toxic
sources by comparing chemical-specific data on the
POTVV effluent to information on suspected sources of
the toxic  pollutants.   This pretreatment program
chemical review (PPCR) approach is recommended
only in situations where the POTW has only a few
indirect dischargers that have relatively non-complex
wastewaters.

PPCR methods are described in Appendix I.  These
methods involve a direct comparison of industry
chemical data  to POTW effluent toxicity.   It is
important  to emphasize  that drawing preliminary
conclusions based on PPCR results can be misleading
because pretreatment monitoring information could be
incomplete, analytical techniques may not be sensitive
to low levels of effluent toxicants,  and the estimated
toxicity of individual compounds may not reflect the
whole effluent toxicity. Domestic sources of toxicants
                 TRE Example

 The PPCR approach was  applied at the Mt. Airy
 POTW in  North Carolina that receives industrial
 wastewater from only a few sources, all of which are
 textile industries (Diehl and Moore, 1987).  Detailed
 information on  the manufacturing processes and
 wastewater discharges of the industries was gathered,
 including data on the toxicity and biodegradability of
 raw and manufactured chemicals as provided in
 material safety data sheets (MSDS) and the scientific
 literature.   This information was used  to identify
 industrial chemicals with a relatively high potential to
 cause toxicity. Subsequent chemical analysis of the
 POTW effluent was  performed to  evaluate  the
 presence of  the  suspected  industrial toxicants.
 Effluent results  were  then compared to literature
 toxicity values for individual compounds. Using this
 approach, APE surfactants, largely attributed to textile
 industries, were identified as the primary causes of
 POTW effluent toxicity.
such as organophosphate insecticides  also may be
responsible for effluent toxicity (see Appendices A
and F).  In summary, comparisons of toxic pollutant
concentrations to effluent  toxicity  may  yield false
correlations. Whenever possible, results of TIE testing
should be used in lieu of PPCR results because the TIE
establishes a cause and effect relationship between
toxicants and effluent toxicity.
                                                 28

-------
                                            Section 4
                           Toxicity Identification Evaluation
Introduction
The TEE is an integral tool in the TRE process and is
applied to evaluate the acute and short-term chronic
toxicity of effluents and other samples. Toxicity is the
trigger for TREs; therefore, the toxicity test is used in
the TIE as the detector for chemicals causing effluent
toxicity.  Many types  of test  species  and  test
conditions, including lethal  (acute) and  sublethal
(chronic) measures for both freshwater and saltwater,
can be adapted for use in the TIE. The use of modified
effluent monitoring procedures, which incorporate the
permit test species or a suitable surrogate, will help to
ensure that the toxicants identified are the ones that
specifically affect the species of concern. In the TIE,
the toxicity test is used to track changes in the presence
and  magnitude  of toxicity  as  the   effluent is
manipulated to isolate, remove, or render biologically
unavailable specific types of constituents (e.g., volatile,
filterable, oxidizable). These procedures relate toxicity
to the wastewater' s physical/chemical characteristics to
determine the compound(s) causing effluent toxicity.

This section of the guidance is intended to be a general
guide for TIEs.   For specific guidance on  how to
conduct TIEs, the reader should consult USEPA' s TIE
manuals (USEPA 1991a, 1992a, 1993a, 1993b, 1996).
The TIE procedures consist of three phases:  Phase I
involves characterization  of the  toxic  wastewater
components,  Phase  n is  designed to  specifically
identify the toxicants of concern, and Phase IH is
conducted to confirm the causes of toxicity. Figure 4-1
presents the logical progression of these three phases
within the framework of a municipal TRE.  USEPA
has published guidance documents for performing each
phase of the TIE procedures.  Phase I procedures are
available to characterize acute (USEPA 1991a, 1996)
and short-term chronic toxicity (USEPA 1992a, 1996).
Phase  H procedures  (USEPA, 1993a)  and Phase m
procedures (USEPA, 1993b) are used to identify and
confirm the  causes of acute or  chronic  toxicity,
respectively. Each TIE is unique and a strategy should
be developed for each study that accounts for site-
specific conditions and allows flexibility in the study
design,  including the use of alternative tools  and
techniques noted in the TIE documents.

Several  effluent  samples  should  be  tested  to
characterize the magnitude and variability of effluent
toxicity over time. Failure to understand the variability
in whole  effluent toxicity and individual toxicants
could lead to  the selection of controls that do not
consistently reduce toxicity to  compliance levels.
Sampling requirements for TIEs are described in
Section 11.  In addition to effluent testing, the TIE
procedures  can  be  applied  in  toxicity  source
evaluations (Section 5) to obtain information about the
causes of toxicity in sewer wastewater or industrial
discharges.

Toxicity Tests
The choice of acute or short-term chronic  tests in the
TIE should be determined based on discharge permit
requirements and the toxicity exhibited by the effluent.
Modifications  to the  whole effluent  toxicity  test
procedures specified in the permit (USEPA 1993c,
1994a, 1994b,  1995) have been made to streamline the
TIE process. These modifications are described in the
respective TIE characterization,  identification,  and
confirmation   manuals  and  include  smaller  test
volumes,  shorter test duration,  smaller number of
replicates, reduced number of test concentrations, and
reduced frequency of sample renewal (USEPA 1991a,
1992a, 1996).  In addition, it is often more useful to
evaluate only  one effluent sample in chronic TIEs
instead of multiple samples (e.g., two, three, or seven)
as is typically used for chronic toxicity monitoring.
Reducing the scale of the toxicity tests improves the
efficiency  of  processing  the  large  number  of
subsamples usually generated in the TIE.  During the
                                                 29

-------
        Pretreatment
          Program
           Data
          Toxicity
          Control
         Evaluation
                                             TIE
                                 Evidence of Effluent Toxicity
                          Phase I - Toxicity Characterization Procedures
                              Initial Toxicity (Acute Phase I)
                              Baseline Toxicity
                              pH Adjustment (Tier 2 in Chronic Phase I)
                              Aeration
                              Filtration
                              Cl8 SPE Treatment                 ;
                              Sodium Thiosulfate
                              EDTA Additions
                              Graduated pH Adjustments
   Additional
TIE Information
   Required?
No
                                Phase II - Toxicant Identification
                               Phase III — Toxicant Confirmation
                             No
  Additional
  Information
  Required?
                        ::
                        :]
Toxicity Source
   Evaluation
     Tier 1
            POTW In-Plant
                Control
               Evaluation
Figure 4-1. Flow diagram of a toxicity identification evaluation.

                                              30

-------
confirmation stage  .of the TIE  (Phase ffl),  whole
effluent toxicity test methods are applied to confirm
that the toxicant(s) identified in Phases I and n is the
cause of the observed effluent toxicity.

TEE  procedures have  been designed to utilize both
freshwater and estuarine/marine species in acute and
short-term chronic tests (USEPA 1991a, 1993a, 1993b,
1996).  Most POTW discharges to freshwater  are
monitored with the cladoceran, C. dubia,  and/or
P. promelas or,  less commonly,  the cladocerans,
Daphnia  magna or Daphnia pulex, and the trout,
Oncorhynchus mykiss. C. dubia and P. promelas were
used in the  development of the TEE procedures and
many  subsequent  TIEs  have  been  performed
successfully with these species (USEPA 1991a, 1992a,
1993a, 1993b), including the case studies presented in
Appendices A, E,  and  F.   TIEs  also  have been
performed with trout (Goodfellow et al., 1994) and the
green alga,  Selenastrum capricornutum (Walsh and
Garnas, 1983).

In addition, USEPA has provided guidance for the use
of Atlantic,  Pacific, and Gulf coast estuarine/marine
species in TIEs (USEPA, 1996). A compilation of
marine TIE studies has  been  prepared (Burgess,
personal   communication,  USEPA,  Narragansett,
Rhode Island,  August,  1998).   TEEs  have been
performed using  mysid shrimp, Mysidopsis bahia
(Morris et al., 1990; Collins, et al., 1994; Burgess et
al.,  1995; Douglas  et al., 1996), the grass  shrimp,
Paleomonetes pugio (Goodfellow  and  McCulloch,
1993), the mussels, Mytilus edulis (Edile et al., 1995)
and Mytilus califomianus (Higashi et al., 1992), the
sheephead   minnow,   Cyprinodon   variegatus
(Goodfellow and  McCulloch, 1993; Burgess et al.,
1995; Douglas et al.,  1996),  the inland silverside,
Menida beryllina (Burgess et al., 1995), the purple
urchin, Strongylocentrotus purpurtus  (Bailey et al.
1995; Jiriketal., 1998), the urchin, Arbaciapunctulata
(Burgess  et al., 1995), the sand dollar, Dendraster
excentricus  (Bailey et al.  1995),  the  red abalone,
Haliotis rufenscens (Griffin et al.,  1993), the alga,
Champiaparvula (Burgess et al., 1995), the giant kelp,
Macrocystis pyrifera (Higashi et  al., 1992; Griffin et
al., 1993), and other estuarine/marine species (Higashi
et al.,  1992; Weis et  al., 1992).  Case studies that
utilized the  echinoderms, S. purpuratus and  D.
excentricus, and the mysid shrimp, M.  bahia,  are
presented in Appendices  B  and  G,  respectively.
Although many species can be used in TIEs, the use of
the species that is specified in ,the NPDES permit or
that triggered the TRE is encouraged to ensure that the
toxicants identified are the ones that affect the species
of concern.  Also, NPDES permit species are more
widely used in TEEs; therefore, extensive published
data are generally available to help characterize and
identify the toxicants affecting these species.

The TEE should incorporate modifications in toxicity
test procedures that are specified in the permit, to the
extent practicable. If pH control in the toxicity tests is
allowed  (T.  Davies,  USEPA, Office  of  Water,
Memorandum on Clarifications Regarding Flexibility
in 40 CFR Part 136 Whole Effluent Test Methods,
April 10,1996), the effects of pH should be addressed
when  evaluating  effluent  toxicants.    However,
procedural modifications should be limited to steps that
are easy and practical to implement.

Effluents with intermittent and ephemeral toxicity may
be challenging to characterize using TEE procedures.'
Intermittent toxicity may require adjustments in  the
TIE  such as performing  frequent toxicity screening
tests  over time to ensure that toxic  samples  are
collected. Some effluents also may exhibit toxicity that
dissipates after the samples are received and the initial
and baseline toxicity tests are performed. If possible,
this  ephemeral  toxicity  may be  characterized  by
conducting both the baseline test and TIE treatments
immediately upon  sample receipt.  Also, it may be
possible to shorten the time between sample collection
and testing (i.e., <36 hours) or use grab  samples in
addition to composite samples. Depending on the level
of effluent toxicity,  it also  may  be challenging to
discern differences in toxicity following the various
TIE   treatments.     Steps  that   may   improve
characterization of these samples include adding more
replicates and/or effluent concentrations  in toxicity
tests used in the TIE and testing more samples to
evaluate   trends  in  the  toxicity  characteristics.
Additional information on  this topic is given in
USEPA's TIE manuals (1991a, 1992a, 1996).

Effluent monitoring data often includes information on
the relative sensitivity of test organisms. It is generally
recommended that initial TEE testing be performed
using the test species that has been shown to be most
sensitive  to the effluent.   In cases  where equal
sensitivity is observed, the organism that is easiest to
use  in  the  TIE should be selected.   Phase  Ed
confirmation tests should utilize each of the species
required by the discharge permit to  ensure that all
toxicants of concern have been determined.
                                                 31

-------
A brief description of the TIE procedures is provided
below.  Examples of TIE applications in municipal
TREs are presented in Appendices A through G.

TIE Procedures
Acute Toxicity Characterization (Phase I)
The first step in the TEE is to characterize effluent
toxicity using the Phase I approach (USEPA, 199 la).
This procedure involves several bench-top treatment
steps to indicate the general types of compounds that
are causing effluent toxicity.  An initial toxicity test is
performed to determine if the sample is acutely toxic.
Simple manipulations for removal or alteration of
effluent toxicity are then performed and the resulting
treated samples and the original sample are tested for
toxicity.  The physical/chemical characteristics of the
toxicants are indicated by the treatment steps that
reduced toxicity relative to the baseline test.

The Phase I characterization includes the following
tests:

  •  Initial toxicity (unaltered effluent)
  •  Baseline toxicity(unaltered effluent)
  •  pH adjustment (pH 3 and 11)
  •  Filtration/pH adjustment (pH 3 and 11)
  •  Aeration/pH adjustment (pH 3 and 11)
  •  CIS SPE/pH adjustment (pH 3 and 11)
  •  Sodium thiosulfate additions
  •  Ethylenediaminetetraacetate (EDTA) additions
  •  Graduated pH adjustments.

USEPA recommends performing the full suite of Phase
I procedures on  initial effluent  samples (USEPA
199 la, 1996). As information is obtained on the nature
and variability of toxicity, additional Phase I tests may
focus on the steps that are successful in affecting
toxicity.  The aeration procedure is used to determine
if toxicity is associated with volatile or oxidizable
compounds.  The filtration procedure is designed to
evaluate  whether toxicity  is  in  the   suspended
particulate phase or in the soluble fraction. Aeration
and filtration, in conjunction with pH adjustments, are
used  to  evaluate  the  volatility  and solubility  of
toxicants such as ammonia, hydrogen  sulfide,  and
metals. The toxicity of oxidants and certain metals is
evaluated   by  adding   sodium  thiosulfate.
Cationic   metal  toxicity  is   determined   by
ethylenediaminetetraacetate  EDTA  additions  and,
possibly, by the graduated pH  procedure.   The
graduated pH step is used to evaluate for the presence
of pH sensitive compounds such  as ammonia.  An
aliquot of the effluent sample also is used to evaluate
the presence of pH sensitive compounds such  as
ammonia.  In addition, an aliquot of the effluent
sample is passed through a CIS SPE column that
selectively  removes  non-polar organic compounds
(USEPA, 1991a).

In general, the TIE procedures used for marine species
are similar to those used for freshwater species, except
that samples used in marine TIEs  must be adjusted to
the salinity appropriate to the species being tested
(USEPA, 1996). As part of the  development of the
marine TIE procedures, USEPA found that marine
species can tolerate  EDTA and  sodium thiosulfate
additions at concentrations that can affect toxicants of
concern.  Marine species can also  tolerate methanol at
concentrations that are necessary to evaluate non-polar
organic compounds  with the CIS  SPE column.
However, there are exceptions to the methods used for
freshwater species for TIE steps.  Due to the strong
carbonate buffering capacity of seawater, it is difficult
to characterize pH dependent toxicants using acids,
bases, and organic buffers. The only efficient method
for maintaining pH in the pH manipulation procedures
is to use controlled atmospheric  chambers.  Also,  a
higher range of pH values is used  in the graduated pH
procedure because of the sensitivity of some marine
species to lower pH.

When  characterizing toxicity to  marine species,
USEPA recommends adjusting the salinity of samples
before performing Phase I manipulations (USEPA,
1996; Ho et al., 1995).  However, if a Phase I TIE is
being conducted to help identify  potential treatment
options for the POTW, the salinity of the samples may
be adjusted after the TIE manipulations are performed.
This approach is necessary to ensure that toxicity
removal in the TIE reflects the conditions that would
occur in  the  POTW (i.e., mimics  treatment before
discharge to saline waters).

Subsequent   tests  are  recommended to further
characterize effluent toxicity. These tests are described
in the acute Phase I document (USEPA, 199la) in the
"Interpretation of Results/Subsequent Tests" sections
for each procedure. Some of these procedures include
elution of the CIS SPE column with methanol to retain
possible toxicants for further testing. If Phase I does
not  adequately  characterize  the  toxicants,  other
techniques can be used, such as ion exchange resins for
anions and cations; XAD (a commercially available ion
exchange  resin)  and activated carbon for various
                                                 32

-------
inorganic and  organic  compounds;  and molecular
sieves,  such  as  Sephadex  resins  that  separate
compounds by molecular weight (Walsh and Garnas,
1983; Lankford and Eckenfelder, 1990; Burgess et al.,
1997). TDS is an example of a toxicant that may not
be  well characterized  in the  TIE.   Methods  for
characterizing  and  identifying  TDS  toxicity  are
presented below.

The characterization procedures are relatively broad
and can indicate more  than one class  of toxicant.
Additional tests are needed to delineate the nature of
the  toxicity if significant  toxicity  changes  occur
following the Phase I tests. For example, the CIS SPE
column procedure, which is designed to determine if
non-polar organic compounds contribute to toxicity,
also can remove other  compounds such as metals.
USEPA (1991a, 1992a) reports that aluminum, nickel,
and zinc concentrations maybe adsorbed onto the CIS
SPE resin. Confirmation that the CIS SPE column
removed non-polar organic compounds is obtained by
eluting the column with methanol to try to recover the
toxicity.  If toxicity can be recovered in the methanol
eluate, then a non-polar organic toxicant is  likely
causing toxicity because metals do  not elute with
methanol. If toxicity adsorbed by the CIS SPE column
is not recovered by the methanol elution, the column
may have removed  toxicants other  than non-polar
organic compounds, such as  metals, or the non-polar
organic compounds may have a higher affinity for the
SPE column resin than methanol.   Appendix E
provides a case example in which toxicity due to
metals was removed by the CIS SPE procedure.

When  the primary toxicant is  present  in  high
concentrations, it may mask other potential toxicants,
making it  difficult  to  detect changes  in toxicity
following the TIE treatments.  Modified procedures
can be designed to control or account for the toxicity of
the primary toxicant.  Ammonia is a common example
of a toxicant that may need to be controlled in the TIE
(e.g., pH  control) in order to evaluate secondary
toxicants (see Appendix  G).

Pretreatment  program  data   and chemical-specific
effluent data may provide useful information to assist
in the Phase I characterization. By reviewing available
information,  compounds  that  are  known to  be
problematic can be compared to the Phase I results to
assist in indicating the effluent toxicants. This data
comparison should not, however, replace the Phase II
and HI analyses.
After successful completion of Phase I, it may not be
necessary to proceed to Phases n and IE.  If the
effluent toxicity  can be  isolated  to  a  class of
compounds, POTW staff may opt to evaluate the
treatment of effluent toxicity.  These studies  may
involve bench-scale or pilot-scale testing procedures
described in Section 6. However, if toxicity remains
following implementation of toxicity control methods,
the TIE should begin again with Phase I.  In most
cases,  a complete TIE using all three phases  will
provide results that will lead to a more cost-effective
evaluation of toxicity control approaches.

Chronic Toxicity Characterization  (Phase I)
The chronic TIE Phase I procedures (USEPA 1992a,
1996) are similar to the acute Phase I procedures and
include  aeration,  filtration,  CIS SPE  treatment,
chelation with  EDTA,  oxidant  reduction and/or
precipitation with  sodium thiosulfate, and graduated
pH testing.   The chronic test measures  sublethal
effects,  such  as  reproduction,  fertilization,  cyst
development,  and/or growth.  These measurements
may  be   affected  by  the  TIE  manipulations.
Adjustments have been made in the TIE procedures to
limit toxicity artifacts.  As  in acute TIEs, additional
steps are recommended to evaluate potential toxicity
artifacts, including use of system blanks and replicate
tests.

The same freshwater species typically used in acute
TIEs (i.e., C. dubia, P. promelas, and, less commonly,
D. magna or D. pulex) can be applied in chronic TIEs.
Species that have been used in chronic marine TIEs
include those noted  above  in the  section  titled
"Toxicity Tests."

Two  tiers   of the  Phase  I characterization  are
recommended for the chronic TIE. Tier 1 is performed
without  major   pH   adjustments.     Consistent,
representative blank tests with reconstituted water are
not readily  obtained at higher pHs;  therefore, the pH
adjustment procedures used  in the  acute  TIE are
separated into Tier 2. Tier 2 is performed only when
Tier 1  does not provide sufficient information about
the types of compounds causing toxicity, and includes
adjusting the effluent sample to pH 3 and 10 as part of
the filtration and aeration steps and pH 9 for CIS SPE
treatment.

Tier I of the chronic Phase I characterization consists
of the following:
                                                 33

-------
 •  Baseline toxicity
 •  Aeration
 •  Filtration
 •  CIS SPE treatment (including tests on post CIS
    SPE treatment and methanol eluate)
 •  Sodium thiosulfate additions
 •  EDTA additions
 •  Graduated pH adjustments.

The Tier 2 tests are to be conducted when Tier 1 does
not provide sufficient information and consists of
filtration, aeration, and the CIS treatment technique of
Tier 1 with an effluent sample adjusted to both pH 3
and 10. Tier 2 of the chronic Phase I characterization
consists of the following:

 •  pH adjustment
 •  Aeration and pH adjustment
 •  Filtration and pH adjustment
 •  CIS SPE treatment and pH adjustment (including
    tests  on post CIS SPE treatment and methanol
    eluate).

Additional Characterization Procedures for
Evaluating the Effect of Ion Composition
Although toxicity caused by ion composition is more
commonly  found in industrial effluents,  ion-based
toxicity has been reported at POTWs (Rodgers 1989a,
1989b; Douglas and Home,  1997; Dawson et al.,
1997). Ion composition  can cause toxicity in two
ways:  relatively high levels  of TDS can  inhibit
osmotic regulation  in  freshwater species,  and an
imbalance in ion composition, particularly calcium
carbonate  levels,  can   adversely  affect  marine
crustaceans (Ward, 1989; MacGregor et  al.,  1996;
Mickley et al., 1996). The later mechanism primarily
affects crustaceans such as the mysid shrimp, M. bahia,
which require minimum  concentrations of calcium
carbonate for survival, growth, and reproduction.

Procedures for  evaluating toxicity  caused by  ion
composition are available (USEPA 1991a,  1992a;
Goodfellow et al., 1998).  The following summary is
intended to provide an overview of procedures that can
be used to evaluate ion-based toxicity.

TDS Toxicity
As a general guide, TDS may contribute to acute
toxicity when conductivity exceeds 3,000 and 6,000
jihmos/cm at  the LC50  for C.  dubia and fathead
minnows, respectively (USEPA, 1991a). For chronic
toxicity, TDS may be a concern when conductivity
exceeds 1,000 and  3,000  |jhmos/cm at the  lowest
observed effect concentration (LOEC) for C. dubia and
P. promelas, respectively (USEPA,  1992a).   The
conductivity of 100%  effluent  is not  the relevant
reading,   but   rather  the  conductivity   at  the
concentrations bracketing the  effluent  LC50 and
NOEC.

C. dubia's higher sensitivity to TDS as compared to
P. promelas can provide additional evidence for TDS
toxicity. Also, the cladoceran, D. magna, exhibits less
sensitivity to TDS than the cladocerans, C. dubia and
D. pulex (API, 1998). These species generally show
similar sensitivities  to  most toxicants  (Mount and
Gulley, 1992); therefore, the difference in sensitivity to
TDS can be useful in characterizing TDS toxicity.  It
is the  toxicity of the individual ions that actually
constitutes TDS toxicity; therefore, it is important to
review the literature  for toxicity data on specific ions.
A thorough review of the toxicity of common ions to
freshwater and  marine  organisms  was  recently
published  by the American Petroleum Institute (API,
1998).

An approach for evaluating TDS toxicity may consist
of the following steps:

1.  Monitor  the effluent  for TDS  and  if the
    conductivity exceeds  the levels given  above,
    measure the major cations (calcium, magnesium,
    sodium,  potassium)  and  anions   (carbonate,
    bicarbonate, sulfate, chloride).  A  cation/anion
    balance should  be  performed  to ensure that all
    major ions have been accounted for.
2.  For freshwater  effluents,  conduct toxicity tests
    using  D. magna,  C.  dubia, and D. pulex. A
    greater sensitivity  by C. dubia  and D.  pulex
    compared  to D.  magna,   together with  high
    conductivity  readings,  provides  a  weight of
    evidence for TDS toxicity.
3.  If  TDS toxicity is suspected, review  the ion
    analysis data gathered above and prepare a stock
    solution of the ions in proportion to the amounts
    typically observed in  toxic  effluent  samples.
    Collect  an effluent  sample  and  immediately
    measure  the  constituent cations and  anions.
    Prepare a mock  effluent by adding the solution to
    deionized  water to yield the same  cation/anion
    concentrations observed in the effluent  sample.
    Measure the toxicity of the  effluent sample and
    mock  effluent. If the toxicity is similar, additional
    evidence is provided for TDS toxicity.
                                                34

-------
4.  If TDS toxicity is indicated, additional procedures
    can be used to determine the extent to which TDS
    contributes to effluent toxicity.  A sample of the
    effluent can be prepared for toxicity testing by
    setting up an appropriate dilution series and then
    adjusting the TDS levels in each dilution to the
    same  TDS level as the 100% effluent using the
    stock  solution (prepared above).  Each effluent
    dilution is then tested individually  for toxicity.
    Comparable  results  for  each effluent dilution
    provides additional evidence for TDS toxicity.
5.  Additional testing can be performed to identify the
    TDS constituent(s) that are causing toxicity. The
    toxicity of various cations and anions is  well
    known and a review of the literature (e.g., ENSR,
    1998) can be helpful in indicating potential ions of
    concern. The ions of concern can be evaluated by
    spiking the ions into dilution water and measuring
    the resulting toxicity.   It should be noted that
    toxicity may be caused by a combination of many
    ions that exert their influence together. Therefore,
    a single salt may not be solely responsible for the
    observed toxicity.
                 TRE Example

 Some of these procedures were used in a TIE at a
 POTW  in  Georgia (Dawson et  al.,  1997) where
 chronic effluent toxicity to C. dubia was observed.
 TIE characterization tests conducted on the effluent
 did not show a reduction in toxicity as a result of the
 Phase I manipulations.  Independent analyses of the
 effluent indicated elevated chloride concentrations. A
 mock effluent was prepared as described in step 3,
 above, and the ion mixture was found to be as toxic to
 C. dubia as the POTW effluent. Laboratory toxicity
 data for sodium chloride (NaCl) were used to confirm
 that the  effluent  chloride  levels would  impair
 reproduction in C. dubia at the effect concentration.

 Additional TIE studies were performed on the Georgia
 POTW  effluent using calcium addition and species
 sensitivity tests.  Calcium has been found to reduce
 chloride  toxicity  in  waters with   similar  ion
 composition as the effluent and addition of calcium to
 effluent samples reduced toxicity. Toxicity tests using
 D. magna, which has been shown to be less sensitive
 to chloride than C. dubia, also provided evidence for
 chloride toxicity.  Overall, the TIE results identified
 chloride as a major contributor to effluent toxicity.
Ion Imbalance
Calcium and carbonate, in proper balance, with other
natural ions, are essential for the formation of new
exoskeleton for mysid shrimp and other crustaceans.
At low calcium carbonate levels (i.e., 15 mg/L CaCO3),
Ward (1989)  observed  60% mortality in mysids
between the 48-hour and 72-hour exposure periods,
which corresponds well with the mysid molting cycle.
Low CaCO3 concentrations also appear to enhance
mysid sensitivity to other  toxicants.  Ward  (1989)
observed a significant increase in the toxicity of Cd to
mysids when calcium carbonate levels were reduced.

The investigator should consider the potential effect of
ion balance as part of the  TIE.  Ion imbalance can
contribute  to   apparent  toxicity in  some  marine
crustaceans when CaCO3 concentrations are 15 mg/L
or less.

Interpretation of Phase I Characterization Results
The following information on  the interpretation  of
Phase I characterization results is paraphrased from the
TIE manuals.  The Phase I characterization provides
information on the types of toxicants in the POTW
effluent. In reviewing the Phase I data, whether for
acute or chronic toxicity characterization, caution is
needed to avoid making inaccurate conclusions about
the results.  For example, as noted above, toxicity
removal by CIS SPE treatment does not necessarily
mean that non-polar organic toxicants  are present.
Toxicity must be recovered in the methanol eluate test
to provide evidence for non-polar organic toxicants.

The following guidance is given by USEPA (1992a)
for interpreting Phase I data on various  types  of
toxicants. Note that the reduction or elimination of
toxicity is determined  by comparing toxicity before
treatment, as  measured  by the baseline test,  with
toxicity after treatment.

Non-Polar Organic Toxicants
Non-polar organic toxicants may be indicated if:

  • Toxicity in the post CIS SPE column test was
    absent or reduced.
  • Toxicity is recovered in the methanol eluate test.
    However, in those instances where methanol does
    not recover toxicity from the CIS SPE column,
    other solvents may be needed to elute the toxicants
    (USEPA,  1993a).
                                                  35

-------
  •  Toxicity is reduced by adding PBO to effluent
    samples or methanol from the methanol eluate
    test.  PBO blocks the toxicity of metabolically
    activated   toxicants   like   organophosphate
    insecticides (USEPA, 1992a).

Cationic Metals
Cationic metals may be indicated if:

  •  Toxicity is removed or reduced  in the EDTA
    addition test.
  •  Toxicity is removed or reduced in the post CIS
    SPE column test.
  •  Toxicity is removed or reduced in the filtration
    test, especially when pH adjustments are coupled
    with filtration.
  •  Toxicity is removed or reduced in the sodium
    thiosulfate addition test.
  •  Erratic dose response curves are observed.

None of these characteristics is definitive, with the
possible exception of EDTA. la addition, toxicity may
be pH sensitive in the range at which the graduated pH
test is performed, but may become more or less toxic at
lower or higher pH depending on the particular metal
involved.     This  characteristic  has  not  been
demonstrated for chronic toxicity to the extent it has
for  the acute  toxicity of several metals (USEPA,
1991a).

Surfactants
Surfactants may be indicated if:

  •  Toxicity is reduced or removed in the filtration
    test.
  •  Toxicity is reduced or removed by the aeration
    test. In some cases, toxicity may be recovered
    from the  walls of the aeration vessel  using a
    dilution water or methanol rinse.
  •  Toxicity is reduced or removed in the post CIS
    SPE column test. The toxicity may or may not be
    recovered in the methanol eluate test. If a series of
    methanol concentrations (e.g., 25, 50,75, 80, 85,
    90, 95, and 100% in water) is used to elute the
    column, toxicity may be observed in multiple
    fractions.
  •  Toxicity is reduced or removed in the post C18
    SPE  column  test  using  unfiltered  effluent.
    Toxicity reduction/removal  is similar to  that
    observed in the filtration test and toxicity may or
    may not be recovered in the methanol eluate test
    or by extraction from the glass fiber filter used in
    the filtration test.
  •  Toxicity degrades over time as the effluent sample
    is held in cold storage (4ฐC).  Degradation is
    slower when the effluent sample is stored in glass
    containers instead of plastic containers.

Ammonia
Ammonia may be indicated if:

  •  Toxicity  increases  in the graduated pH test at
    higher pH.
  •  The effluent is more toxic to P. promelas than to
    C. dubia.
  •  Note:  If the concentration of total ammonia (as
    nitrogen) is 5 mg/L or more and chronic toxicity is
    a concern, the potential  for  ammonia toxicity
    should be evaluated.

Drawing conclusions about ammonia toxicity  based
solely on observed concentrations can be misleading,
especially where chronic toxicity is a concern because
of  the  uncertainty about  the chronic  effects  of
ammonia.  Ammonia is an example of a toxicant that
acts independently of  other toxicants in effluents.
Even though ammonia concentrations may appear to be
sufficient  to cause all of the effluent toxicity, other
toxicants may be present and may contribute to toxicity
when ammonia is removed.

Oxidants
Oxidants may be indicated if:

  •  Toxicity  is removed or reduced in the sodium
    thiosulfate addition test.
  •  Toxicity is removed or reduced in the aeration
    test.
  •  The sample is less  toxic over  time when held at
    4ฐC (and the type  of container does not  affect
    toxicity).
  •  C. dubia are more sensitive to the effluent than P.
    promelas.

The presence of TRC in the effluent is not enough to
conclude that toxicity is due to an oxidant. However,
TRC concentrations  of 0.05-0.1  mg/L or more in
100% effluent provides strong evidence for oxidant
                                                 36

-------
toxicity.   Further evidence would be  provided, if
dechlorination with sulfur dioxide (SO2) or another
dechlorinating agent removes or reduces  toxicity
(USEPA, 1992a).

TDS
TDS may be indicated if:

  •  pH adjustments do not remove or reduce toxicity
    and a precipitate  is not visible  in  the  pH
    adjustment test, pH adjustment and filtration test,
    orpH adjustment and aeration test.
  •  There is no loss of toxicity in the post CIS SPE
    column tests, or a partial loss of toxicity, but no
    change inconductivity measurements.
  •  There is  no change in toxicity with the EDTA
    addition test, sodium thiosulfate addition test, or
    the graduated pH test.
  •  There is a greater sensitivity by C. dubia and D.
    pulex compared to D. magna, together with high
    conductivity readings.
  •  A mock effluent prepared with the same ions as
    the effluent exhibits similar toxicity as the effluent.
  •  Toxicity is removed or reduced by ion exchange
    resin.
  •  Toxicity is not removed or reduced by passing the
    effluent over activated carbon.

Appendices A, B, E, F, and G provide example Phase I
data  and  describe how  results  are used to  select
additional TIE procedures for testing.  The Phase n
and Phase m procedures (USEPA 1993a and 1993b)
are applicable to both acutely and chronically toxic
samples.
Acute and Chronic Toxicity Identification
(Phase II)
The  Phase n guidance  manual  (USEPA,  1993a)
describes  procedures that can be used to  identify
specific  toxicants   such  as  non-polar  organic
compounds, ammonia,  cationic  metals, chlorine, or
toxicants  removed  by  filtration.   The Phase n
procedures are applicable to both  acute and chronic
toxicant identification.  Phase n uses treatment and
toxicity testing techniques similar to Phase I  and
incorporates chemical-specific analyses to identify the
toxicants.   Examples of TIE Phase n studies are
provided in Appendices A, B, E, F, and G. Appendices
A, F, and G describe the use of Phase n techniques for
non-polar   organic   compounds,  including  high-
performance liquid chromatography  (HPLC) for the
isolating toxicants. Appendices B and E describe the
application of Phase n procedures for identifying toxic
metals.    Appendix G  describes  how Phase n
procedures were used to identify ammonia toxicity.

Acute and Chronic Toxicity Confirmation
(Phase III)
The toxicants identified in Phase EL may be confirmed
by a series of Phase HI steps, including correlating
toxicity and toxicant concentration from multiple
samples, observing test organism symptoms, evaluating
species  sensitivity,  spiking  effluent samples  with
suspected toxicants, and performing a mass balance to
account for all of the effluent toxicity. In many cases,
it  will be  appropriate  for the Phase I, n, and m
evaluations  to  overlap  because  confirmation
information can be obtained during Phases I and n.
Examples  of TIE confirmation testing are provided in
Appendices A, B,  E, F,  and G.
                                                 37

-------
                                            Section 5
                                Toxicity Source Evaluation
Introduction
Once the effluent toxicants have been identified, a
follow-up evaluation can be conducted to locate the
sources of the toxicants. This evaluation may involve
a review of existing pretreatment program data or data
from the collection and analysis of additional samples
from industrial users. In some cases, the TIE may not
identify  the  specific compounds causing effluent
toxicity and,  in the absence of data on toxicants, the
sources of toxicity must be tracked.   Examples of
compounds that are not easily identified  in the TIE
include  surfactants  and  some  non-polar  organic
compounds (other than organophosphate insecticides).
Although the class of compounds may be indicated in
the TIE, it may not be possible to locate the sources
without information on the specific toxic compounds.
In these cases, a guidance is available to track the
sources of toxicity.

A toxicity source evaluation is conducted to locate the
sources of influent  toxicity or toxicants that  are
contributing  to the POTW  effluent toxicity.  This
evaluation is performed in two tiers whether chemical-
specific or toxicity tracking is to be performed:

  •  Tier I—generally involves sampling and analysis
    of wastewater samples collected from the main
    POTW sewer lines.
  •  Tier II—is performed using samples collected
    from tributary sewer lines or point sources on the
    main sewer lines found to be toxic in Tier I.

This tiered tracking approach can be used to identify
the sources  of toxicity and/or toxicants  through a
process of eliminating segments of the collection
system that prove to be non-toxic.

The flow diagram for the toxicity source evaluation is
presented in  Figure 5-1.  The choice  of chemical-
specific analyses or toxicity tests for source tracking
will depend on the TIE data on the POTW effluent
toxicants.   A  chemical-specific   investigation  is
recommended in cases where the effluent toxicants
have been  confirmed and  can be traced to the
responsible  sewer dischargers.  If the sources of
toxicants are located, the TRE can then proceed to the
evaluation of local pretreatment limits as described in
Section 6.   Toxicity tracking, using the refractory
toxicity assessment (RTA) approach described herein,
is required in situations where the TIE does not provide
conclusive  data on the effluent toxicants.  Prior to
toxicity analysis, sewer samples are subjected to the
same type of treatment as is provided by the POTW for
its influent wastewaters.  This treatment step allows a
measurement  of  "refractory"  wastewater  toxicity,
which is the toxicity that passes through the POTW
and causes effluent toxicity.  If toxicity tracking is
successful   in  locating  the  sources  of  toxicity,
pretreatment requirements can be set to reduce the
refractory toxicity contributed to the POTW.

In some cases, industrial users may modify or cease the
discharge of  toxicity before  specific  sources are
identified. The abatement of effluent toxicity during
the course of TREs is not uncommon; however, efforts
to ensure ongoing compliance can be difficult when the
original sources of toxicity are not  located.   These
situations  dramatize  the importance of collecting
information  on industrial pretreatment activities and
POTW operations in the early stages of the TRE. As
part of the toxicity source evaluation, POTW staff can
request industrial  users  to submit  weekly or daily
reports of production and waste discharge activities
that can  be used  to  indicate  potential sources of
toxicity. This information also is helpful in subsequent
pretreatment control studies, if an industrial user is
identified as a source of toxicity (Botts et al., 1994).
                                                 38

-------
                                   Toxicity Source Evaluation
                                   Review of TRE Information

                            Results of Toxicity Identification Evaluation
                            Results of Pretreatment Program Review
                                             1
                                   Select Sampling Locations
                   Tier I
       Chemical-Specific Investigation
              of Sewer Lines or
             Indirect Dischargers
                                            Tier II
                             Tier I
                  Track Toxicity in Sewer Lines
                          Using RTA

                 Bench-Scale POTW Simulation
                 Using:
                 •  Sewer Wastewater Spiked
                   into POTW Influent
                 •  POTW Influent
                  TierH
       Chemical-Specific Investigation
           of Indirect Dischargers
                                                                      1
                            Tier II
                 RTA of Sewer Line Tributaries
                   and Indirect Dischargers

                 • POTW Simulation Tests
                 • Inhibition Tests (Optional)
                 • TIE Phase I Tests (Optional)
      Review Tier I/II
     Results and Repeat
    Testing if Necessary
 Additional
Information
 Required?
TIE
                                        Toxicity Control
                                          Evaluation
Figure 5-1. Flow diagram for a toxicity source evaluation.

                                              39

-------
Sampling Approach
The "Tier I" and "Tier H" designations refer to the
sampling approach to be taken in tracking the sources
of toxicity and/or toxicants.

Tier I- Toxicity Source Evaluation
Sampling locations for Tier I testing are established by
reviewing the pretreatment program data (Section 3)
and selecting appropriate sampling points on the main
sewer lines. In some cases, industrial users or tributary
sewer lines may be selected when substantial evidence
is  available  on  potential  sources of  toxicity  or
toxicants.

If the toxicants have been identified and chemical-
specific tracking is to be performed in Tier I, sampling
locations  can include industrial users who have
manufacturing processes or use raw materials that are
known or suspected of containing the toxicants (e.g.,
metals from metal finishers).   If the  toxicant is
contributed by a large number of dischargers, sewer
line sampling is recommended in Tier I. For example,
sewer line sampling was conducted to determine the
sources of organophosphate insecticides in the City of
Fayetteville, North Carolina, sewer system and several
sewer  systems in the San  Francisco Bay Area in
California (see Appendix F). These studies indicated
that two  insecticides, diazinon and chlorpyrifos, are
widely distributed in POTW collection systems.

If toxicity tracking is to be performed in Tier I, each
major sewer line should be sampled to ensure that all
possible  sources  in the  collection  system  are
considered.   Indirect discharger  sampling  is  not
recommended in Tier I because of the large number of
sources that may need to be evaluated.  Sewer line
testing may ultimately reduce the number of sampling
points by eliminating segments of the collection system
where toxicity is not observed (USEPA, 1983a).

In  the RTA study conducted at Fayetteville, North
Carolina (Fillmore et al., 1990), sewer wastewater
samples  were  initially  collected from  manholes
throughout the collection system because of the large
number of potential sources of toxicity.  Sources of
toxicity were subsequently  identified by testing the
indirect dischargers located on the toxic sewer lines.

Tier II- Toxicity Source Evaluation
Results of Tier I are used to  establish the sampling
locations for Tier II.  The toxic sewer lines identified
by toxicity or toxicant tracking in  Tier I can be further
evaluated in Tier n by sampling indirect dischargers or
tributary sewers on the toxic sewer lines.

Information on classes of toxicants such as surfactants
can be obtained by coupling the RTA protocol with
selected TEE procedures. For example, in the TRE at
the LRSA, New Jersey, sources of non-polar organic
toxicants were identified by passing RTA test samples
through CIS SPE columns (see Appendix G). Sources
of toxicity were indicated if toxicity was observed in
methanol eluates from the columns.

Sampling Conditions
Whether  sampling of  sewer  lines  or  indirect
dischargers is conducted, 24-hour flow proportional
composite samples are recommended to characterize
daily variations in toxicant concentrations or toxicity.
In some cases, samples may be collected over less than
24 hours  to observe  the  contribution  of potential
intermittent sources of toxicants or toxicity.

Flow data must be gathered in order to determine the
relative contributions of toxicants or toxicity from the
sewer lines or indirect dischargers. Flow data can be
used to calculate the toxicant loadings, which will be
needed to develop local  pretreatment  limitations
(Section 6).  Row data also will be needed to conduct
RTA testing, as described later in this section.

The sampling period for both sewer lines and indirect
dischargers should account for:

  • Discharge  schedules  for  indirect  dischargers
    (i.e., intermittent versus continuous).
  • Temporary shut-down schedules for industry
    maintenance.
  • Coordination with routine pretreatment program
    monitoring, if possible.

For example, in the LRSA TRE (Appendix G), sources
of refractory toxicity  were identified by  sampling
during periods of normal industry activity and during
a period of temporary industry shut-down.

Other  considerations for  sampling are described in
Section 11.   QA/QC sampling  requirements  are
discussed in Section 8.

Chemical-Specific Investigation
A chemical-specific approach can be used to trace the
influent sources of toxicants if definitive TIE data on
the causes of POTW effluent toxicity are available.
                                                 40

-------
This approach is not recommended in cases in which
the TIE data only indicate a broad class of compounds
(e.g., polar organic compounds), because the toxicants
may be contributed by a variety of sources that will be
difficult  to  pinpoint  by   chemical  tracking.
Chemical-specific tracking should be conducted after
the  effluent  toxicants  have  been  identified  and
confirmed in TIE Phase n and in tests (USEPA 1993a
and 1993b).

The  chemical-specific  approach  involves  testing
indirect dischargers or sewer line samples for toxicants
using chemical analysis  techniques. In some cases,
existing pretreatment program data maybe adequate to
identify the indirect dischargers that are contributing
the toxicants.   It is  likely,  however, that further
sampling  and analysis  will be necessary,  because
pretreatment  program data generally  do not include
information on toxicants typically identified in  TIE
tests (e.g., compounds other than regulated pollutants).
Existing pretreatment program data may be used to
reduce  the  amount  of  sampling  and analysis by
indicating which sources contribute toxic pollutants
that are similar to the effluent toxicants.

Chemical analysis methods for potential toxicants such
as ammonia, metals, and  organic compounds are
described in several  USEPA documents (USEPA
1979b, 1983b, 1985b) and Standard Methods for the
Examination of Water and Waste-water (APHA, 1995).
USEPA (1997) provides all of USEPA's methods for
analysis of water on a CD-ROM. USEPA's Phase H
TIE manual (1993a) also provides guidance for the
analysis of organophosphate insecticides, surfactants,
and metals.  Analytical methods for organophosphate
insecticides have been improved to achieve the lower
detection limits necessary to assess insecticide toxicity
(USEPA, I993a; Durban etal., 1990).  Enzyme-linked
immunosorbent  assay (ELISA) procedures  also are
available for selected organophosphate insecticides,
metals, and other compounds.  ELISA methods offer
the advantage of low cost, rapid sample processing,
and field portability; however, these methods may not
be  specifically  approved  by USEPA.  Additional
analytical techniques can  be  found in American
Society for Testing and Materials (ASTM)  manuals
and  peer-reviewed journals such as  the Analytical
Chemistry Journal. A qualified chemist should verify
the selected analytical method in the laboratory prior to
sampling and analysis.
A literature search also can be made to determine if the
toxicant could be a biodegradation product resulting
from  POTW treatment.  Where clear evidence is
available to show that the toxicant is a treatment by-
product, the sewer sample should be  analyzed for the
precursor form(s) of the toxicant as well as the toxicant
itself.

In cases where  chemical tracking  is  successful in
locating the sources of the POTW effluent toxicants,
the TRE can proceed to the selection and development
of toxicity control options such as local pretreatment
regulations (Section 6).   Information on  toxicant
distribution can be used in developing pretreatment
control options.   For example, although  a primary
contributor of ammonia was identified in  the LRSA
TRE  (Appendix  G),   system-wide  pretreatment
limitations  were  adopted to address all non-domestic
sources  of ammonia.  In  other situations, control
methods  other than pretreatment limitations, such as
public education,  may be  needed to control the
discharge of a widely used toxicant.  Public education
has been successfully used at a number of POTWs
(Appendix  H) to control the use of organophosphate
insecticides,  which can be contributed from many
domestic and commercial  areas of the  collection
system.

If the responsible indirect dischargers are not located,
the TIE results should be reviewed to confirm previous
conclusions. The chemical analysis results also should
be  carefully  reviewed  to  determine if errors or
waste water matrix effects may have caused inaccurate
results. In cases where the chemical-specific approach
is ultimately  not successful, the source  evaluation
testing should be repeated using toxicity tests in lieu of
chemical analyses, as described below.

Refractory Toxicity Assessment
Toxicity tracking  may be  required when the  TIE
characterizes  the toxicity as broad classes of toxicants
or identified toxicants cannot be confirmed. Toxicity
tracking also may be useful in situations in which there
are multiple effluent toxicants and the occurrence of
these toxicants  in the POTW effluent  is highly
variable. In such cases, toxicity testing may be more
cost-effective than chemical tracking.

The toxicity  found in influent wastewaters is not
necessarily the same toxicity that is observed in the
                                                 41

-------
 POTW effluent because  the POTW is  capable  of
 removing some toxic wastewater constituents.  The
 amount  of sewer  wastewater toxicity  that could
 potentially pass through the POTW must be estimated
 by treating sewer samples in a simulation of the POTW
 prior to toxicity analysis. This treatment step accounts
 for the toxicity removal provided by the POTW.

 A protocol has been  developed  for predicting the
 potential for a sewer discharge to contribute to acute or
 chronic toxicity in POTW effluents.  This protocol,
 referred   to as  the  RTA  procedure,  has  been
 successfully used to track sources of acute and chronic
 toxicity using both freshwater and estuarine/marine
 species (Morris et al., 1990; Botts et al., 1992, 1993,
 1994).  Examples of RTA studies are presented  in
 Appendices C, D, and G.

 The RTA protocol has been designed to  simulate
 conventional activated sludge processes,  although it
 has  also  been adapted to other POTW treatment
 processes including single and two-stage nitrification
 systems  (Collins,  et  al.  1991), BNR processes
 (Appendix  D), and  filtration  treatment  systems
 (Appendices C and D). The RTA procedure described
 herein  involves   treating  sewer  samples  in  a
 bench-scale, batch  simulation of  a  conventional
 activated sludge process and measuring the resulting
 toxicity.    Batch  simulations are  appropriate for
 plug-flow biological systems because batch processes
 behave over time as plug-flow processes do with flow
 time.  Batch biological reactors have been used by
 several researchers to screen wastewaters for activated
 sludge inhibition (Grady, 1985; Adams et al., 1981;
 Philbrook and Grady, 1985; andKangetal., 1983) and
 non-biodegradable aquatic toxicity (Hagelstein and
 Dauge, 1984; Lankford et al., 1987; and Sullivan et al.,
 1987). Hagelstein and Dauge (1984) and Lankford et
 al. (1987) have found that  toxicity  measurements
 coupled with bioreactor tests can be a pragmatic way
 to evaluate refractory wastewater toxicity.

 The RTA protocol was developed in the USEPA TRE
 research study at the City of Baltimore's Patapsco
 POTW (Botts et al., 1987) to evaluate the potential for
 indirect  dischargers to contribute refractory toxicity.
 Additional USEPA TRE research studies in Linden,
 New Jersey; High Point, North Carolina; Fayetteville,
 North Carolina; and Bergen County, New Jersey were
 conducted to improve the RTA approach (Morris et al.,
 1990;  DiGiano, 1988; Fillmore et al., 1990; Collins et
 al., 1991). The RTA procedure described herein is a
 refined version of the method given in the municipal
 TRE protocol (USEPA, 1989a).

 The batch reactor used in RTA testing is designed to
 simulate,  as   close  as  possible,  the  operating
 characteristics of the POTW's activated sludge process
 (e.g., MLSS concentration, DO level, and F/M ratio).
 Two types of batch reactors  are used, as shown in
 Figure 5-2.  One reactor serves as the control and
 consists  of the POTW influent and return activated
                    Wastewater and
                   Return Activated
                      Sludge
            Plastic or Glass
              Container
                                                    Air Supply
                                                    (Oil-Free)
Air Line
Tubing
                                                   Air Stone
                             Control Reactor
                              POTW Influent
                                (Control)
                 Spiked Reactor
                 Sewer/Industrial
                Wastewater Spiked
                into POTW Influent
Figure 5-2.  Schematic of a refractory toxicity assessment test.

                                                 42

-------
sludge (RAS).  The other reactor consists of sewer
waste water spiked into the mixture of POTW influent
and RAS.  If the effluent toxicity of the reactor spiked
with sewer wastewater is increased  relative to the
unspiked reactor, the  sewer  wastewater would  be
considered a  source  of refractory toxicity.  In the
spiked reactor,  sewer wastewater is  tested together
with the POTW influent in order to observe possible
interactive  effects (e.g., additivity, antagonism) that
can occur when the wastewater and the total POTW
influent are combined and treated in the POTW.

A general description of the RTA procedure is
presented below. A step-by-step protocol for  RTA
testing is provided in Appendix J. The basic steps in
the RTA approach are:

  •  Conduct  conventional  pollutant  analyses  to
    develop a profile for each wastewater to be tested
    in the RTA.
  •  Perform  toxicity tests on the POTW's  RAS
    (filtrate)  to determine its  potential to cause an
    interference in RTA testing.
  •  Collect,  characterize, and prepare  wastewater
    samples for RTA tests.
  •  Calibrate the RTA batch  reactors to  achieve a
    treatment level comparable to that of the POTW's
    activated sludge process.
  •  Calculate wastewater volumes to be used in RTA
    tests.
  •  Set-up and operate the RTA batch reactors.
  •  Analyze batch effluent toxicity.
  •  Evaluate the potential for the sewer waste waters to
    inhibit activated sludge treatment (optional).
  •  Conduct TIE Phase I tests to indicate the types of
    refractory toxicants  in  the  sewer wastewater
    (optional).
  •  Interpret  the results.

It is important to emphasize that the RTA protocol
should be modified to address site-specific conditions.
For example, Appendix C describes an RTA study that
simulated a filtration treatment process in addition to a
nitrification  treatment process.    The  following
summary of  the RTA protocol is intended to be a
general guide to evaluating sources of toxicity  using
simulations of suspended biological growth processes.
Best professional judgment is important in adapting the
procedures to treatment processes and conditions that
are unique to each facility.
POTW Wastewater Profile
The first step in the RTA is to characterize the POTW
influent (primary effluent), sewer wastewaters, and
RAS to be used in RTA testing. The POTW influent
wastewater should be collected from the effluent of the
POTW primary  treatment process because primary
effluent is treated by the activated  sludge process,
which is the main process to be simulated in the RTA.
RAS is recommended for use in batch testing because
it is in a concentrated form that can be easily diluted to
the target MLSS concentration.  Mixed liquor from the
POTW's aeration basins can be used in lieu of RAS;
however, the activated sludge will need to be thickened
to the same suspended solids concentration as the RAS
before use.

Table 5-1 presents the analyses and information that
are needed to characterize the wastewaters to be used
in RTA testing.  This information will  be useful for
determining the following operating conditions for the
RTA batch reactors, including:

  •  Determining the volume of sewer wastewater to
    use in testing based on sewer line and indirect
    discharger flow-rate data.
  •  Determining whether nutrient addition is necessary
    using information on the ratio of organic content
    (BOD5 or COD) to nutrient concentrations (TKN
    and TP).
  •  Selecting a test period for the RTA reactors that is
    based on the  organic content of the  sewer
    wastewater.  Some sewer wastewaters  may have
    substantially higher COD concentrations,  which
    will increase the initial COD level in the RTA
    reactor.  A longer treatment time may be needed to
    ensure that the wastewater is treated to the same
    level as the POTW influent.

Biotnass Toxicity Measurement
Sometimes the  RAS  used in testing can cause an
interference in the measurement of refractory toxicity.
In the  Patapsco  TRE, filtered samples of RAS were
found  to be acutely toxic to C. dubia  (Botts  et al.,
1987).  This toxicity was related to residual biosolids
that passed through the filter [10-micron (|im) pore
size].  The toxic biosolids caused the batch reactor
effluents in RTA tests to be acutely toxic and masked
the refractory toxicity of the wastewaters being tested.
This biomass interference reduced the effectiveness of
the RTA test for determining the sources of refractory
                                                 43

-------
Table 5-1. POTW Wastewater Profile Analyses for a Refractory Toxicity Assessment
                  Waste Stream
               Information Required
  RAS
TSS
VSS
NH3-N
pH
  Primary effluent
BOD5
COD
TSS
TP
TKN
NH3-N
pH
  Sewer line or indirect discharger wastewater
Location in collection system
Number/type of indirect dischargers
Flow, million gallons per day (mgd)
BOD5
COD
TSS
TKN
NH3-N
TP
pH
  Other indirect discharger data
Type of discharger
Wastewater pretreatment system
Operations/production schedule
toxicity at the Patapsco WWTP.  Additional tests
demonstrated that RAS toxicity could be removed by
filtration of the coarse filtrate through a 0.2 |im pore-
size filter or by centrifugation of the coarse filtrate at
10,000 times gravity (xg) for 20 minutes (Botts et al.,
1987).

Additional information obtained in the Linden Roselle,
New Jersey; Fayetteville, North Carolina; and Bergen
County, New  Jersey USEPA TRE research studies
indicated that the POTW RAS filtrate was not acutely
toxic, and therefore did not cause an interference in
RTA testing (Morris et al., 1990; Fillmore et al., 1990;
Collins et al., 1991). The existing data on the toxicity
of sewage sludges  are not  sufficient, however, to
evaluate the occurrence of biomass toxicity at POTWs.
The following discussion provides information on how
to proceed, if POTW biomass toxicity is observed.
Prior  to conducting the RTA, toxicity tests of the
POTW  activated sludge should  be performed to
determine if the biomass is toxic. This testing involves
    toxicity measurement of two aliquots of RAS: coarse
    RAS filtrate and  coarse RAS filtrate subjected  to
    centrifugation to remove colloidal particles. The RAS
    should first be filtered through a coarse glass fiber
    filter (e.g., 10 (im pore size), which is the same type of
    filter used  for  suspended  solids analysis  (APHA,
    1995).  Following coarse filtration,  an aliquot of the
    RAS filtrate should be further treated by centrifugation
    at 10,000 xg for 10 to 15 minutes. Alternatively, the
    coarse filtrate could be  filtered through a 0.2 jam
    membrane filter. However, tests should be conducted
    to confirm that  soluble toxicity is  not removed by
    sorption onto the filter.

    Both the RAS filtrate and centrate should be tested for
    acute or  chronic  toxicity  using limited-scale tests
    (USEPA  1993c, 1994a,  1994b).  If results show that
    centrifugation does not reduce biomass toxicity relative
    to the coarse filtrate, then  the RAS is not  likely  to
    cause an interference in RTA testing. In this case, the
    POTW biomass  can be used directly in RTA testing
                                                 44

-------
and centrifugation of the RTA batch effluents will not
be required. Samples of RAS should be periodically
analyzed for toxicity during RTA testing to monitor for
possible biomass toxicity.

If the biomass coarse filtrate is observed to be more
toxic than the RAS centrate, the biomass toxicity may
interfere with RTA testing. Two options are available
in this case: removal of the toxic biosolids by fine
particle filtration (or centrifugation) of batch test
effluents, and use of an alternate biomass.

Biomass toxicity may be removed by applying the fine
particle filtration or centrifugation treatment steps to
the batch test effluents.  In this case, the resulting RTA
effluent toxicity will only indicate soluble refractory
toxicity, not the total refractory toxicity (i.e., soluble
and paniculate).

Another approach to remove biomass toxicity is to use
a non-toxic biomass such as another POTW biomass or
a commercially available (freeze-dried) preparation. A
surrogate biomass will not be acclimated to the influent
wastewaters of the POTW being evaluated; therefore,
it may not treat the wastewaters as well as the POTW's
biomass.   Nonetheless, an alternate biomass can
provide a level of treatment that will approximate the
refractory toxicity of the sewer wastewater. It may be
helpful to conduct a parallel series of RTA tests using
the toxic POTW biomass.  The use of toxic POTW
biomass is suggested because  it is acclimated to the
POTW influent wastewaters and will therefore provide
a level of batch treatment that is more similar to the
POTW  treatment   than  that  provided  by the
unacclimated alternate biomass.   In this  case, fine
particle filtration  or centrifugation  is  required to
remove the  interfering biomass  particles prior to
toxicity analysis.   By  performing RTA tests with
POTW biomass in parallel  with RTA tests with
alternate biomass, both the soluble and total refractory
toxicity of the wastewater may be estimated.

An alternate biomass may be useful in cases where it is
necessary to simulate future modifications or additions
to the POTW activated sludge treatment process (e.g.,
conversion  from conventional activated  sludge to
nitrification).   In these cases,  a biomass that  is
indicative of the future activated sludge, may not be
directly available at the POTW. An alternate biomass
can  be obtained from another  POTW  that has a
biological treatment process similar to the treatment
process planned for the POTW. A TRE conducted by
the City  of Durham, North Carolina, used  this
approach to evaluate the toxicity reduction capability
of  planned  nutrient removal  treatment  systems
(Appendix D).

RTA Reactor Calibration Testing
Generally, ideal plug-flow conditions do not occur in
activated  sludge processes;  therefore, it  will  be
necessary to adjust the RTA batch treatment conditions
to account for the actual level of treatment achieved in
the POTW.  One method  of controlling the treatment
efficiency of activated sludge processes is to adjust the
biomass   concentration,  measured   as   MLVSS
concentration.    Batch  calibration  tests  can  be
performed using a series of MLVSS concentrations
and the  MLVSS  concentration that most closely
simulates the POTW treatment efficiency can be
selected for RTA testing.

Prior  to calibration  testing,  a   target  MLVSS
concentration can be estimated using mathematic
models.   In  the Fayetteville TRE,  a steady state,
completely mixed, multi-stage model (Grady  and Lim,
1980)  was used to determine biokinetic coefficients
that best modeled the POTW treatment performance
(Fillmore et al., 1990).   The biokinetic coefficients
were  then used in a steady  state plug-flow model
(Kornegay,   1970)  to  calculate  a   batch  MLVSS
concentration that would theoretically simulate the
POTW treatment efficiency. The model results were
confirmed in bench-scale, batch reactor tests using a
range  of  MLVSS  concentrations,  including the
theoretical MLVSS concentration and several MLVSS
concentrations that bracketed the theoretical value. In
this case, the MLVSS concentration determined from
the calibration tests matched the theoretical MLVSS
value (Fillmore et al., 1990).

POTW primary effluent is  typically used  in  RTA
calibration testing.  The  treatment efficiency of the
batch  reactors  can be  evaluated  by  periodically
collecting and  analyzing  samples  for COD  and
toxicity.  TKN, NH3-N, and TP may also be monitored
if the  batch  reactors are simulating  BNR treatment
systems.  Results of the  batch reactor tests are then
compared to COD, nutrient, and toxicity data for the
POTW final effluent to indicate  which batch reactor
achieved treatment comparable to the POTW. If there
are large differences between batch effluent results and
POTW effluent results, it may be necessary to evaluate
                                                 45

-------
different  MLVSS   concentrations  in   additional
calibration tests. In the Durham TRE, calibration tests
were used to define batch operating conditions for a
five-stage BNR process (Appendix D).

The RTA  calibration  study  can also be used to
establish an appropriate test  dilution series for the
toxicity tests of batch effluents.  Where possible, the
dilution series for toxicity  tests should bracket the
acute or chronic toxicity value (i.e., LC50 or ICp) as
closely as possible in order  to reduce the span of the
95% confidence limits. Increased confidence in the
data is important because sources of refractory toxicity
are indicated based on a comparison between effluent
toxicity results for the sewer wastewater-spiked reactor
and the POTW influent (control) reactor (Figure 5-2).
The following example illustrates this point.

A wastewater from an industrial user is spiked into
POTW influent sample and tested using the RTA
procedure.  The acute toxicity of the RTA effluent is
measured using two different dilution series: one test
series  encompasses  a  wide  range  of  samp?e
concentrations and  the other series more closely
brackets the expected LC50. The dilution series and
resulting survival and LC50 values  are  shown in
Table 5-2.

Using dilution series #1, the LC50 for the industrial
wastewater test would be 35% sample with confidence
limits (95%) of 25% to 50%  (based on binomial
model). Using series #2, the LC50 would also be 35%
sample, but the confidence limits (95%) would  be
much tighter at 31 to 39% (based on probit method).
The results for the RTA control test together with the
industrial  wastewater  spiked  test  are  shown in
Table 5-3.

In this example, if the control reactor LC50 had been
50% with confidence limits of 42 to 62%, the industrial
wastewater would have been indicated as a possible
source of toxicity based on the results of series #2,
because the 95% confidence limits do not overlap (i.e.,
31 to  39%  versus 42 to 62%).  However, if dilution
series #1 had been used, the industrial wastewater may
not have been judged to be a toxic source because the
confidence limits overlap (25 to 50% versus 42 to
62%). The partial mortality in the 35% concentration
in series #2 (Table 5-3) helps to more precisely define
the LC50. The narrow confidence limits in series #2
support the conclusion that the refractory toxicity of
the wastewater is significantly greater than the POTW
influent control (i.e., confidence limits do not overlap).

In the Reidsville and LRSA TREs (Appendices C and
G), the results of preliminary toxicity analyses were
used to adjust the dilution series to closely bracket the
expected IC25 and LC50 value of the batch effluent
samples.  This  approach allowed the identification of
sources of refractory toxicity that would not have been
indicated using a standard toxicity test dilution series.

Sample Collection
Wastewater and activated sludge samples  should be
collected according to the procedures described in
Section 11.  Sample volumes will be based on the
subsample  volumes  needed  for periodic  reactor
measurements and batch effluent toxicity testing.
Table 5-2. Example of Bracketing the LC50 Concentration in the RTA Sewer Wastewater Test
Test
Series #1
Series #2
Percent Survival in Sample Concentration
100
0
100
0
50
0
50
0
25
100
35
50
12.5
100
25
100
6.25
100
12.5
100
0
100
0
100
Source: Fillraore et al., 1990.
Table 5-3.  Comparison of Control Test and Industrial Wastewater Spiked Test Results
Test
Series #1
Series #2
Control Test
LC50(CI)*
50% (42-62)
50% (42-62)
Industry Spiked Test 1
LCSO(CI) L
35% (25-50)
35% (31-39)
Potential Source
of Toxicity?
No
Yes
* Confidence intervals (95%) shown in parentheses.
                                                46

-------
The volumes of wastewater and RAS to be used in
RTA testing will depend, in part, on whether acute or
chronic toxicity will be measured. Generally, a batch
reactor volume of 3 liters  (L) is  sufficient when
standard freshwater and marine/estuarine species (i.e.,
C. dubia, Daphnia sp., P. promelas, M. bahia) are to
be used for testing the acute toxicity of batch effluents.
A batch reactor volume of 10 L is adequate to measure
chronic batch effluent toxicity using the 7-day C. dubia
test.

Sample Characterization
Average characteristics of the sewer wastewater and
POTW primary effluent can be determined using the
wastewater profile data (Table 5-1). These data should
include historical results of BOD5, COD, TKN, TP,
TSS, NH3-N, and pH analyses. Analyses should also
be performed on the samples collected for RTA tests to
ensure  that  the  wastewater   characteristics  are
consistent with historical data.

Preparation of RTA Test Mixtures
Two batch influent solutions are prepared for each test
of a sewer wastewater sample: sewer sample spiked
into POTW primary effluent, and primary effluent
alone.  The sewer  sample may be collected from a
sewer line or an industrial discharge.  The amount of
sewer sample to be used in testing should reflect the
percent volume of sewer wastewater  in the POTW
influent.  In some cases, the wastewater toxicity from
small contributors may not be readily observed when
the wastewater is  mixed by percent  volume with
POTW influent. In these cases, it may be necessary to
use a  greater volume of sewer wastewater than is
typically contributed to the POTW.

The volume of sewer wastewater (Vw) sample to be
added to the batch reactor is calculated as follows:
          Vw (L) = -=- x (Vr - Vb) x Fw,
                   Qi

where:  Qw is the sewer wastewater flow rate (mgd).
        Qi is the average POTW influent flow rate
        (mgd).
        Vr is the total reactor volume (gal or L).
        Vb is the volume of RAS biomass (gal or L).
        Fw is the sewer wastewater flow concentra-
        tion factor (e.g., 1, 2, 10 times the sewer
        wastewater flow).
The selection of a flow concentration factor (Fw) will
depend on the percent flow of the sewer wastewater in
the POTW influent.  A conservative, yet realistic,
approach would be to use a Fw that is based on the
maximum daily wastewater flow from the sewer
discharge in the past year. The Fw should not cause
the sewer wastewater to be  100%  of  the reactor
wastewater  volume.   For  example,  if the sewer
wastewater flow is greater than 20% of the POTW
influent flow, a Fw of less than 5 should be used. It is
necessary to test the mixture of sewer wastewater and
POTW primary effluent in order to evaluate  the
interactive effects  (e.g., additive or antagonistic) that
can realistically occur when these wastewaters are
combined at the POTW. All sewer samples should be
tested using the same Fw to allow a comparison of
batch effluent toxicity between the various sewer
wastewaters.

After determining the Vw, the volume of primary
effluent (Vpe) to be added to the batch reactors can be
calculated as:

              Vpe = (Vr-Vb-Vw).

The batch reactor influents are prepared by mixing the
Vw and Vpe for the sewer wastewater spiked reactor
and measuring Vpe for the control reactor. In some
cases, it may be necessary to adjust the nutrient levels
or pH of the batch influents prior  to testing  as
described below.

The BOD5/TKN/TP ratio of the batch reactor influents
should be compared to the average BOD5/TKN/TP
ratio  of the POTW  influent, as  determined from
historical or profile data. The sewer wastewater added
in the batch reactor influent may be deficient in
nutrients, especially if industrial wastewaters are used.
If necessary, nitrogen  and/or phosphorus should be
added so that the  BOD/TKN/TP ratios of the batch
reactor influent and POTW influent are similar.
Unless the BOD5 to nutrient ratios for the batch reactor
influent and POTW influent are  clearly dissimilar,
nutrient addition is not recommended be'cause of the
potential  for added  nutrient salts to change  the
sample's toxicity.

Using the profile data, BOD5 and nutrient (TKN, TP)
concentrations (C) in the batch reactor  influent
(spiked) are calculated as follows:
                                                47

-------
       C (mg/L) = (vPexCPg)+(VwxCw)
                       (Vpe+Vw)

where:  Vpe  is the volume of primary effluent  in
        reactor (L).
        Cpe is the BOD5 or nutrient concentration in
        primary effluent (mg/L).
        Vw is the volume of sewer wastewater  in
        reactor (L).
        Cw is the BOD5 or nutrient concentration in
        sewer wastewater (mg/L).

The typical BOD5/TKN/TP ratio for municipal sewage
is 100:5:1 (WEF/ASCE, 1992a). This ratio will ensure
that sufficient nutrients are available for consistent
batch treatment of the sewer wastewaters. If necessary,
phosphorus should be added in the form of three parts
monosodium phosphate (NaH2PO4)  to four  parts
disodium phosphate (NajHPOJ. Nitrogen should be
added as urea nitrogen, except in cases where ammonia
is suspected as a cause of effluent toxicity, because
urea nitrogen can be converted to ammonia during
biological treatment.

Following nutrient  addition, the pH  of the batch
influents may need to be adjusted to within the average
range of pH for the POTW influent.  Typically, the
range of POTW influent pH values will be pH 6 to 9.
Hydrochloric acid and sodium hydroxide can be used
for pH adjustment.

Following nutrient addition and pH adjustment, the
batch  influent  toxicity  should be   measured  to
determine if the added nutrients or pH adjustment
cause  a change  in  sample toxicity.    Substantial
differences between the initial toxicity and the adjusted
sample toxicity may indicate the presence of specific
types of toxicants. Use of pH adjustment for toxicity
characterization is discussed in the USEPA TIE Phase
I manuals (USEPA 1991a and 1992a).

The volume of RAS biomass (Vb) to be used in batch
testing should yield a batch MLVSS concentration that
is equal to the target MLVSS concentration determined
in calibration testing (see above). The amount of RAS
to be  added to the total reactor  volume (Vr)  is
calculated as follows:

    Vb(L) .Target MLVSS (mg/L)
               RAS VSS (mg/L)
This equation also is used to determine the alternate
(non-toxic) biomass volume (Vnb), if required.
                    j
Synthetic Wastewater Testing (Optional)
In some cases it may be important to determine the
amount of refractory toxicity of the sewer wastewater
excluding the effects of other influent wastewaters. A
batch influent solution pontaining sewer sample spiked
into a synthetic wastewater can be used to  determine
the individual refractory toxicity of the sewer sample.
The synthetic wastewater will provide a standard
substrate that will allow consistent treatment of the
sewer wastewaters.

A synthetic wastewater should be  prepared that has a
COD concentration that is equal to the average COD
concentration of the  POTW  primary  effluent. The
volume of synthetic wastewater (Vsw) to be added to
the batch reactor is calculated using the same equation
that is used to calculate the volume of POTW primary
effluent.    A  synthetic  wastewater has  not  been
developed that is consistently non-toxic (DiGiano,
1988).  Prior to use in  RTA testing, the synthetic
wastewater should be tested for toxicity to ensure that
it will not interfere with the measurement of refractory
toxicity.

Performance of RTA Tests
RTA  testing is initiated when  the batch influent
solutions are mixed  with RAS and diffused air is
applied to the mixture. The  aeration rate  should be
adjusted to maintain a DO concentration equal to the
DO level observed  in the POTW activated sludge
treatment  process.   Mechanical mixing  using  a
magnetic stirrer and  teflon-coated stir bars may be
required to ensure complete mixing in the reactor. The
RTA tests must be performed in appropriate laboratory
fume hoods to prevent exposure of laboratory staff to
any toxic vapors stripped from the wastewater samples
(Section 9).

The organic loading to the batch reactors can vary
substantially  depending  on  the  type  of  sewer
wastewater being tested.  For example, a wastewater
with a high COD concentration (e.g., >5,000 mg/L) is
likely to increase the COD loading  to the RTA reactor.
The effect of this variation on batch treatment can be
minimized by adjusting the reactor treatment time to
achieve a constant "food-to-microorganism ratio" in
the batch reactor (F/Mb).  F/Mb should be  similar to
                                                48

-------
the F/M of the POTW biological treatment process.
This adjustment will allow the biodegradable material
in the batch influent to be reduced to approximately
equal levels in all RTA tests.  The required batch test
period (d) can be calculated as follows:

          , _ Batch Influent COD(mg/L)
               MLVSS (mg/L)xF/Mb

where:  F/Mb is equal to the calculated F/M of the
        primary effluent reactor (i.e., COD/MLVSS x
        treatment period in days).

Both acute  and chronic refractory toxicity can be
measured in  RTA  testing.   In order  to  obtain
comparable toxicity results, RTA testing should utilize
the same species that was used for TIE tests or routine
compliance monitoring. Use of toxicity screening tests
such as  bacterial  bioluminescence   tests  (e.g.,
Microtoxฎ)  in  conjunction  with the  preferred test
species may provide additional information.  These
screening tests are recommended when  the waste
streams to be tested exert a high oxygen demand (i.e.,
high BOD  concentration)  which would  otherwise
require aeration during testing and a possible loss of
toxicants.     Standard  procedures   for   toxicity
measurement are not practical due to the large number
of samples that will need to be processed in the RTA.
Instead, simplified acute toxicity  test procedures, like
those presented in the USEPA TIE Phase I manuals
(USEPA  199 la,  1992a, 1996)   are recommended.
Likewise,   simplified  procedures  for  short-term
measurement  of  chronic toxicity (USEPA  1992a,
1996) are recommended for chronic refractory toxicity
assessments.  Oris et al. (1991) and Masters et  al.
(1991) describe the use of an abbreviated version of
the 7-day chronic C. dubia test, referred to as the 4-day
test. However, the 7-day test has been the method of
choice for  most  RTA  studies because the  use  of
younger test animals provides more consistent results.
Therefore, 7-day test data  are better for discerning
differences between toxic and nontoxic  sources.

The batch  test mixtures are  prepared for  toxicity
analysis  by  allowing  the  mixed liquors to settle,
decanting the clarified supernatant, and filtering the
supernatant through a coarse glass fiber filter.  The
coarse filtration step is used to more closely simulate
the POTW clarification process because solids settling
in bench-scale containers is not as efficient as the
POTW settling process.  Note that this step may not be
required if the RTA includes a simulation of effluent
filtration processes at the POTW (see Appendix C). If
toxic biomass  is  used in the RTA tests, further
particulate removal is required to measure the soluble
refractory toxicity in the sewer wastewater. In this
case, the  coarse  filtrate can be filtered through a
0.2 nm pore-size glass filter or centrifuged at 10,000
xg  for  10 to  15 minutes (American  Society  for
Microbiology, 1981) to remove colloidal size particles
from the  wastewater.  Membrane filters such  as
cellulose nitrate filters may not be appropriate because
some soluble organic constituents may absorb onto the
filter. Prior to sample filtration, all filters should be
washed and filter blanks should be prepared using the
steps described in Section 8 and Appendix J.

Data Evaluation
Results of RTA testing are used to locate the sources
that are contributing refractory toxicity to the POTW.
A discussion  of  the  evaluation of RTA results is
provided as follows.

Results of RTA Tests if POTW Biomass is Non-
toxic
Results for each sample analysis will consist of data on
two types of batch tests: tests of sewer sample spiked
into primary effluent, and a control test using primary
effluent alone. The batch test of the sewer sample/
primary effluent will indicate the toxicity that would
realistically occur upon  mixture  of  the  sewer
wastewater with POTW influent. Results of this test
are compared to results of the primary effluent control
test to determine if the addition of sewer wastewater
decreases the refractory toxicity (e.g.,  dilution  or
antagonistic effect) or increases the refractory toxicity
(e.g., additive effect) of the primary effluent.

If the effluent toxicity of the sewer sample/primary
effluent test is greater than the effluent toxicity of the
primary effluent control test, the sewer wastewater
source may be a  contributor of refractory toxicity.
POTW influent  and sewer  wastewater toxicity is
known to vary significantly over time; therefore, each
wastewater source should be tested several times over
an extended period (e.g., three times during both cold
and warm weather months) to determine the overall
potential for the discharge to cause POTW effluent
toxicity.  Results of Tier I sewer line tracking can be
used to prepare a list of the toxic sewer lines. This list
can be compared to a sewer collection system map to
indicate tributary sewer lines or indirect dischargers to
be tested in Tier EL
                                                 49

-------
The TRE case study summaries in Appendices C and
G describe how RTA results  were used to indicate
sources of refractory toxicity  in the Reidsville and
LRSA TREs, respectively. These studies illustrate the
need  to test  several samples from each wastewater
source in order  to account for  the variability in
refractory toxicity over time.

Results of RTA Tests if POTW Biomass is
Toxic
In situations where the RAS coarse filtrate is found to
be more toxic than the RAS  centrate, RTA tests may
use alternate (non-toxic) biomass in addition to tests
with the POTW biomass. The data for each sewer
sample analysis will consist of results  of two  batch
tests using alternate biomass (i.e., one test of sample/
primary effluent, and one test of primary effluent) and
results of two batch tests using toxic POTW biomass.
The results  of tests that use alternate  biomass will
provide an estimate of the total refractory wastewater
toxicity.  The disadvantage of these tests is that the
alternate biomass  is  not acclimated to the  POTW
influent wastewaters; therefore, it may not provide the
same level of treatment as  the POTW  acclimated
biomass.

Batch tests using toxic POTW biomass better reflect
the treatment  efficiency  of  the  activated sludge
process; however, manipulation of the batch effluent
(i.e., centrifugation or small particle filtering) removes
particles  that normally are  present in the  POTW
effluent.   Batch effluent treatment is  necessary to
remove the interfering toxic biomass, but this treatment
may artificially change batch effluent toxicity. The
advantage of toxic  biomass  tests is that the soluble
refractory toxicity  of source  wastewaters can  be
determined.    The  non-toxic biomass  tests cannot
provide as  good an estimate of soluble  toxicity,
because alternate biomass is not  acclimated to the
POTVV influent wastewaters.  If both toxic biomass
and alternate nontoxic biomass are used in testing,
results are obtained on both  the soluble and total
refractory toxicity of the sewer wastewater.

Inhibition Testing (Optional)
Inhibitory wastewater may upset the normal operation
of the POTW biological treatment process to the extent
that  it causes toxicity pass-through.    Biological
treatment  inhibition may occur  by three primary
mechanisms: competitive inhibition, non-competitive
inhibition, and substrate  inhibition.  The effect of
competitive  inhibitors is most pronounced  at low
substrate   concentrations.    Inhibition  by  non-
competitive inhibitors such as chromate or other heavy
metals  is  observed  over  a  range of  substrate
concentrations.  The third mechanism of biological
inhibition, substrate inhibition, occurs at high substrate
concentrations.

Only substrate inhibition can be practically evaluated
in batch treatment tests. An example of the effects of
substrate inhibition on biological activity is shown in
Figure 5-3. This figure shows that substrate utilization
normally achieves a constant maximum rate as the
wastewater concentration is increased.  If inhibitory
substances are present in the wastewater, the substrate
uptake  rate  would decrease  as the  wastewater
concentration is increased further.

Substrate inhibition can be assessed by monitoring
removal of substrate (e.g., BOD5, COD, TKN, andTP)
and oxygen uptake rates in the RTA batch reactors. A
series  of dilutions of the  sewer line  or indirect
discharger wastewater is tested with POTW biomass:
one with 100% indirect discharger wastewater and at
least three consisting of serial dilutions (e.g., 50%,
25%, and 12.5%) of sewer wastewater.  A range of
wastewater dilutions is necessary to compare organic,
nutrient, and oxygen removal rates over a range of
substrate concentrations. At high wastewater strengths
[e.g., 1 mg/L  soluble COD  (SCOD) to 4 mg/L
MLVSS], biomass activity  will generally reach  a
maximum  rate  (Figure  5-3).   When  wastewater
concentrations are  increased, a decrease in COD,
nutrient, and oxygen removal rates would indicate the
presence of inhibitoiy materials.

SCOD,  ammonia  (SNH3-N), and phosphorus (SP)
removal can be used to calculate the specific substrate
utilization rate (SSUR).  The SSUR is reported in units
of mg/L of soluble substrate  per gram MLVSS per
minute (g MLVSS/min), and is calculated using the
equation:
    SSUP, -
            MLVSS (g/L) x Test Period (min)

where:  Ci is the influent substrate concentration as
        SCOD, SNH3-N, or SP.
        Ce is the substrate concentration in periodic
        samples collected from the batch reactor.

The POTW biomass used in batch testing contains
residual SCOD, SNH3-N  and SP remaining from
                                                 50

-------
                                                                              Non-inhibitory
                                                                              wastewater
                                                            Inhibitory
                                                            wastewater
Figure 5-3. Theoretical results of inhibition testing.

biological treatment that must be accounted for when
calculating  batch  effluent concentrations.    The
correction for biomass SCOD, SNH3-N, and  SP is
calculated by the following equation:

 SCOD = [(Vr) x (Ce, mg/L)] - [(Vb) x (Cb, mg/L)]
                          Vr
where:  Cb is the concentration  of SCOD, SNH3-N,
        and SP in the RAS filtrate.
        Vr is the total volume in the batch reactor (L).
        Vb is the volume of RAS added to the reactor
        (L).

Oxygen utilization can be measured as a  specific
oxygen uptake rate (SOUR).  SOUR is reported in
units of mg O2/L/g MLVSS/min and is calculated as
follows:
 SOUR = -
                Oxygen Consumed (mg/L)
        MLVSS (g/L) x DO MeasurementPeriod (min)

The SSUR and SOUR data for the four wastewater
concentrations can be plotted as shown in Figure 5-3.
A reduction in the SSUR and SOUR rates for the full
strength sample test relative to the SSUR and SOUR
rates for the sample dilution tests would indicate the
presence of inhibitory material in the sewer wastewater
sample. The degree of inhibition can be inferred by the
amount of deviation in biomass activity rates between
the full strength sample test and the sample dilution
tests.

Phase I Toxicity Characterization (Optional)
TIE Phase I tests can be applied to the batch effluent of
the indirect  discharger/primary  effluent test to
determine the types of toxicants causing refractory
toxicity in the sewer wastewater.  Results of the TIE
Phase I testing can be compared to TIE results for the
POTW effluent to determine if the sewer wastewater
contains  the same  types of refractory toxicants that
were observed in the POTW effluent. Sources that
discharge the same types of toxicants as those found in
the POTW  effluent  would  be candidates  for a
pretreatment control evaluation (Section 6). The TIE
Phase I procedure is described in Section 4.

Findings of the Toxicity Source Evaluation
The results of Tier I and Tier n testing should be
sufficient to confirm the sources of POTW effluent
toxicants or refractory toxicity. This information can
be used to evaluate and select pretreatment control
options (Section 6).

It is possible that the toxicity source evaluation results
will suggest  that no  single sewer line  or indirect
discharger is a source of refractory toxicity. This case
may occur if the sources of toxicants or  toxicity are
                                                 51

-------
widely dispersed throughout the  collection system.
Examples  of dispersed toxicants include  organo-
phosphate insecticides (e.g., diazinon) and ammonia.
The inability to locate the toxicant or toxicity sources
may also indicate that the sewer sampling points did
not include all possible sources of the toxicants or
toxicity.  In this case, it may be necessary to evaluate
additional sewer lines in the collection system.
In situations where the toxicity  source evaluation
proves to be a prodigious task, the permittee may elect
to evaluate alternatives for in-plant toxicity control
(Section 6).  The choice of pretreatment or in-plant
controls may be determined by assessing the best use
of the resources that are available for the TRE.  In this
regard, POTW staff have the option to recover costs
associated with toxicity source evaluation through the
process of local limits development.
                                                 52

-------
                                            Section 6
                               Toxicity Control Evaluation
Introduction
The goal of the TRE is to select and implement toxicity
control methods and technologies that will achieve
compliance with the permit limits for effluent toxicity.
Toxicity control evaluation involves  assessing the
potential  control  options  and selecting the best
option(s) for toxicity reduction based on technical and
cost considerations. Figure 6-1 illustrates the process
of evaluation and selection of toxicity control options.
Toxicity control may be accomplished either through
the implementation of pretreatment requirements or
POTW  modifications.  Examples of pretreatment
controls include local limits development and waste
minimization/pollution  prevention   requirements.
POTW   modifications  may   include changes  in
treatment chemical  usage,  enhanced  operational
strategies, or addition of treatment processes.

Criteria for the selection of the preferred toxicity
control option(s) should be defined at the beginning of
the toxicity control evaluation. Recommended criteria
include:

  •  Compliance with effluent toxicity limits
  •  Compliance with other permits
  •  Capital, operational, and maintenance costs
  •  Ease of implementation
  •  Reliability
  •  Environmental impact.

Cost will be a primary selection criterion; however, the
selected control option must offer the best potential for
consistent, reliable toxicity reduction with the least
impact on other permit requirements.  The selection
criteria should be used initially to screen all candidate
control  options to  determine which alternatives merit
further study. The preferred options can then undergo
an in-depth review in a pretreatment control evaluation
(e.g., local  limits  development) or in-plant control
evaluation (e.g., treatability studies). Information from
these evaluations will be used  to select the  most
feasible  option(s)  based  on   a  more  thorough
comparison of the criteria listed above.   The final
selection  process   may  require  a  quantitative
examination of the options using a scoring and ranking
system. Table 6-1 presents a matrix of in-plant toxicity
control options for the TRE case example provided in
Appendix G.  Further discussion of the final selection
process is provided at the end of this section.

Identifying Toxicity Control Options
The TRE guidance is designed to identify possible
methods for toxicity reduction at the earliest possible
stage in the TRE process. As shown in the overall
schematic of the TRE process (Figure 1-1), sufficient
information may be  available  for toxicity  control
evaluation  at   the   completion  of  the  POTW
performance   evaluation  conventional  pollutant
treatability tests, TIE  tests,  and tbxicity/toxicant
tracking. Control options must be identified based on
ample data that clearly demonstrates  the option's
technical feasibility.

POTW Performance Evaluation Treatability Tests
Treatability  testing   in  the  POTW  performance
evaluation  may identify  options  for   improved
conventional pollutant  treatment that also reduce
effluent toxicity to acceptable levels (Section 3). In
addition, the  optional TIE Phase I tests may provide
information on the presence of in-plant toxicants such
as suspended solids or chlorine that is corroborated in
the operations and performance review. The treatment
steps in TIE Phase I also may provide information on
treatment options for control of the in-plant toxicants.

Potential  control  options may  involve  treatment
modifications or  additions  that are necessary to
improve conventional pollutant treatment and to reduce
or eliminate in-plant sources of identified toxicants.
Examples of these control options include dechlorin-
                                                 53

-------
                                    Toxicity Control Evaluation
                                       Develop Selection Criteria

                               Compliance with Effuent Toxicity Limits
                               Compliance with Other Permits
                               Capital, Operational, and Maintenance Costs
                               Ease of Implementation
                               Reliability
                               Environmental Impact
                                     Select Control Alternatives
                                       Based on TRE Results
                                               i
                  Evaluation of
           In-Plant Control Alternatives

         Process Enhancement
         POTW Modifications and Additions
         Treatability Testing
                                                                        1
         Evaluation of
Pretreatment Control Alternatives

• Allowable Headworks Loading
  Analysis
• Public Education
• Local Limits
• Industrial User Management
• Case-by-Case Permitting
                                         Information and
                                           Data Review
                                       Control Alternatives
                                           Assessment
                                         Select Toxicity
                                       Control Alternatives
                                       Implement Toxicity
                                       Control Alternatives
Figure 6-1. Flow diagram for a toxicity control evaluation.

                                               54

-------
Table 6-1. An Example of the Comparison of In-Plant Ammonia Treatment Alternatives (Ammonia
Concentrations of 90 mg/L NH3-N or Higher)
Treatment Technology
1 . Single-stage biological
nitrification
2. Two-stage biological
nitrification
3. Biological nutrient
removal with nitrification
4. Ammonia air stripping
5. Selective ion exchange
(including resin
regeneration)
6. Breakpoint chlorination
Capital Costs
Millions*
9.1
, 11.5
18.7
11.2
28.0
7.5
O&M
Costs
Millionst
1.5
2.4
2.8
1.3
6.2
6.8
Equivalent
Annual Cost
Millions*
3.4
5.0
6.4
3.3
12.7
11.5
Relative
Practicality!
Low
Impractical
Low
Very Low
Very Low
Very Low
Relative
Reliability*
Low
Low
Low
Low
Low
Low
 * Approximate capital costs based in part on WPCF Nutrient Control Manual cost curves (WPCF, 1983). Values reflect conditions
   of 17 mgd and 90 mg/L NH3-N. The values presented here have been modified from the cost curves to reflect engineering and
   contingency costs at 25% and contractor's overhead and profit at 15%.
 t Approximate overhead costs based on WPCF Nutrient Control Manual cost curves. Values reflect conditions of 17 mgd and
   90mg/LNH3-N.
 t Approximate equivalent costs amortized over 20 years, assuming an annual 5.00% increase in operation and maintenance costs
   and an estimated annual interest rate of 8.86%.
 ง Relative practicality based on typical technology applications, available land space, overall costs, and/or chemical usage
   requirements.
 # Relative reliability based on potential inhibition, temperature and pH sensitivity, and evidence that the technology is proven
   reliable at 17 mgd and 90 mg/L NH3-N. Scores of "low" to "high" were used.
 Source:  LRSA (1991). Additional information on this TRE is presented in Appendix G. All costs shown are in 1991 dollars.
ation treatment to eliminate toxic levels of chlorine and
biological treatment  optimization  (e.g.,  increased
MCRT) to remove toxic ammonia concentrations.

TIE Tests
Results of TIE Phase I testing (Section 4) may indicate
the types of treatment that can be used to remove broad
classes of effluent toxicants (e.g., filterable material,
metals, organic compounds). For example, filterable
toxicants may be removed by granular media filtration.
The feasibility of options for removing classes  of
toxicants can  be evaluated in the  POTW in-plant
control evaluation.

Alternatively, results of TIE Phases n and in may help
to identify and confirm the specific effluent toxicants
(Section 4).  If the pretreatment  program data are
adequate to  determine the sources of the toxicants,
local limits  can be developed and evaluated in the
pretreatment control  evaluation.    In  this  case,
pretreatment control would be preferred over in-plant
control because the costs of implementation are usually
lower. If pretreatment program data on the toxicants
are not available, chemical-specific  testing will be
necessary to track the sources of the toxicants before
toxicity control selection can proceed.

Chemical-Specific Investigation
Chemical-specific tracking in the Tier I - toxicity
source evaluation may locate the sources of the POTW
effluent toxicants (Section 5). Once the sources have
been identified, pretreatment control  options such as
local limits or waste minimization requirements can be
developed and evaluated.

Refractory Toxicity Assessment
Results of the Tier H RTA testing may identify the
indirect dischargers contributing refractory toxicity to
the POTW. Based on these results, POTW staff can
require the indirect discharger to limit the discharge of
wastewater toxicity even though the toxic wastewater
constituents have not been identified.  In some cases,
                                                    55

-------
 POTW staff may elect to perform optional TIE Phase I
 analyses  to  provide  information   on  the  toxic
 constituents  in  the  indirect discharger wastewater.
 This additional testing may be conducted  so that
 numerical pretreatment limits can be set.

 Toxicity Control Screening Process
 Using  appropriate selection  criteria,  the  preferred
 toxicity control options are identified.  Available
 options can be compared using a ranking system (e.g.,
 on a scale of 1 to 10). This screening process may be
 relatively simple, although some estimate of costs (i.e.,
 order of magnitude) will be useful in selecting the most
 practical options. The selected options are then studied
 in the pretreatment control  evaluation and in-plant
 control evaluation, as described below.

 The example  matrix in Table 6-1 compares in-plant
 control options  for ammonia toxicity.  In this  case,
 costs and qualitative measures were used to rank the
 various options.  All of the in-plant control options
 were found to be impractical or costly; therefore, the
 sewerage authority investigated pretreatment controls.
 The source of a majority of the ammonia loading was
 an industry, which was considered to be controllable.
 As  a  result,  the  sewerage  authority required the
 industry to implement ammonia control methods.  The
 cost to the authority was relatively low and involved a
 headworks analysis for ammonia and reissuance of
 discharge permits. Additional information on this TRE
 is provided in Appendix G.

 Pretreatment Control Evaluation
 Pretreatment control  options can be  developed by
 public works managers to prevent the pass-through of
 toxicants, toxicity, and inhibitory material that  have
 been traced EO  indirect dischargers.   The primary
 advantages of pretreatment control of toxicity are that
 a  smaller  volume  of waste can be managed by
 addressing individual sources and the costs are usually
 the responsibility of the industrial users. Pretreatment
 requirements may involve a public education effort or
 the  implementation  of  narrative   or numerical
 limitations for POTW users.

 The  toxicants  to be controlled may not be  the same
 parameters that are currently regulated under the
 pretreatment program.  Examples of these types of
 toxicants include organophosphate insecticides, TDS,
 biocides, and specialty chemicals used by industries.
In cases where current pretreatment regulations are
 inadequate to address sources of toxicants or toxicity,
 POTW  staff should revise or adopt  new  permit
 regulations  or ordinances, as appropriate.  In these
 cases, it may be necessary to initiate the following
 steps to control toxicants or toxicity:

  • Investigate public education  approaches, if the
    toxicant is  widely  used  in  the  service  area
    (e.g., organophosphate insecticides).
  • Perform an allowable headworks loading analysis.
  • Decide   whether to  establish local  limits  or
    implement a more directed approach, such as
    industrial user  management or  case-by-case
    requirements.
  • Develop  a  monitoring program  to  evaluate
    compliance with the requirements.

 These steps  are described below.

 Public education has been successfully used to control
 toxicity  at POTWs.   Organophosphate insecticides
 such as diazinon and malathion have been identified as
 effluent toxicants at many POTWs, especially in the
 southeast and southwest United States (Norberg-King
 et al., 1989). Insecticides can be discharged by many
 users in the POTW service area, including pest control
 businesses,   veterinarians,  lawn   care  businesses,
 apartment complexes, restaurants, hotels/motels, office
 buildings, and homeowners. These users are usually
 not included under pretreatment programs and it may
 be impractical to control these sources by regulating
 each  discharge.   Studies at POTWs in California
 (Singhasemanon  et  ! al.,   1997),  Texas  (City  of
 Greenville,  1991), Oklahoma (Engineering-Science,
 Inc., 1992), and North Carolina (Fillmore et al., 1990)
 have  determined that public education is  a  viable
 option for control of organophosphate insecticide
 toxicity attributed to multiple sources. Recommended
 steps in a successful public awareness program include
 identifying  the  significant users  of  insecticides,
 developing education materials targeted to users, and
 distributing the materials on an ongoing basis during
 periods  of expected insecticide use.   The City of
 Greenville also enacted an ordinance to encourage the
 environmentally sound use of insecticides and require
 merchants to display public education materials where
 insecticides  are  soH (City  of Greenville, 1991).
Additional information on the identification  and
control of organophosphate insecticides is presented in
Appendices F and H.  ^Public education efforts may be
applied to control other effluent  toxicants that  are
widely used in  POTW service areas  and are not
                                                 56

-------
practical to  regulate  through  local  pretreatment
limitations.

POTW  staff have successfully used revised or new
pretreatment regulations  to reduce POTW effluent
toxicity (Appendices C, E, and G). Local pretreatment
limits can be developed to control sources of toxicants
or toxicity identified in the toxicity source evaluation.
USEPA's Guidance Manual on the Development and
Implementation of Local Discharge Limitations Under
the Pretreatment Program (USEPA, 1987b) describes
several approaches for developing local limits. These
approaches include:

  •  Allowable   Headworks    Loading  Method:
    Numerical  limits  are  defined  based  on  the
    maximum pollutant  loadings that will allow
    compliance with receiving water  quality criteria,
    sludge quality  criteria, or protection  against
    treatment interferences.
  *  Industrial User Management Method:  Based on
    an in-depth review of indirect discharger practices,
    POTW staff can  set narrative limits for chemical
    management practices (e.g., chemical substitution,
    spill prevention, and slug loading control).
  •  Case-by-Case  Permitting:    Technology-based
    limits are established based on levels that can be
    feasiblely  and   economically  achieved   by
    industries.

Some of the local limits approaches can be adapted to
address effluent toxicants or toxicity. For example, the
allowable headworks  loading method is well-suited for
developing limits  to prevent the pass-through of
toxicants identified in POTW effluent TIE tests and
located by chemical-specific analyses in the toxicity
source  evaluation.   This method can  be used to
establish the maximum level of the toxicant that can be
safely received by the POTW without exceeding the
effluent toxieity limit.   The  LRSA,  New Jersey,
conducted an allowable headworks loading analysis to
address industrial sources of ammonia (see Appendix
G). The results of the analysis were used to develop
local limits for controllable sources in order to reduce
effluent toxicity caused by ammonia.

The industrial user management method provides a
framework for implementing chemical management
practices including slug discharge control. In cases in
which slug loadings contribute to POTW effluent
toxicity, spill prevention  or load equalization can be
implemented at the industrial facility to moderate the
slug loadings.   USEPA's  Guidance  Manual for
Control of Slug Loadings to POTWs (1988b  and
1991c) describes methods for the development of slug
loading control programs.

The case-by-case permitting method can be used when
the POTW effluent toxicants cannot be identified, but
TIE information on the general classes of toxicants is
available or sources of toxicity have been located in the
toxicity source evaluation.    Using TIE data,  an
engineer  may  be  able  to  select  a  pretreatment
technology that can remove general types of toxicants
(i.e., non-polar organic compounds).  In cases where
the sources of toxicity have  been identified, POTW
staff  have the authority to  require  the  indirect
discharger to take  steps to  limit the discharge of
refractory or inhibitory toxicity (USEPA, 1987b).

Although  USEPA  and the  States  with  approved
pretreatment programs have  overview authority, the
choice of which  approach  to use  for local limits
development is the municipal government's decision.
The goal in developing  local limits is to  implement
pretreatment regulations that are technically and legally
defensible. Local limits can include provisions for
equitable recovery of costs associated with the toxicity
source evaluation and limits development.

In-PIant Control Evaluation
The objective of the in-plant control evaluation is to
select and evaluate feasible treatment options for the
reduction   of   effluent   toxicity  at  the  POTW.
Treatability testing may be conducted to determine the
toxicity   removal  effectiveness   and  operating
characteristics of the candidate treatment options.
These tests should use acute or chronic toxicity tests
and chemical  analyses to evaluate the removal of
specific toxicants and/or toxicity. The resulting data
provide a  basis for the final selection and conceptual
design of feasible POTW process  modifications or
additions.

It is  important to  consider  that major changes in
treatment  plant facilities or operations may not be
practical due to the  cost of  new facilities or the
complexity of additional process operations. In these
situations, pretreatment  control  of toxicity may be
preferred over in-plant control. Wherever possible, the
in-plant control evaluation should  be performed in
conjunction with the pretreatment control evaluation to
identify the most technically feasible and cost-effective
control option.
                                                 57

-------
Review Existing Information
The first step in the in-plant control evaluation is to
review the POTW performance evaluation data on the
POTW design (Section 3) to establish the physical
space available  for new  process additions and to
determine the idle facilities and equipment that could
be  used  for  toxicity  control.    Operations  and
maintenance information also should be reviewed to
determine if the POTW is capable of handling  the
increased operational control that may be required with
process modifications or additions. In addition, POTW
performance  evaluation   information  should  be
reviewed to determine how the control options might
be integrated into the overall treatment system design.

TIE results on identified effluent toxicants can be used
to determine in-plant  control  options.   Although
information on specific toxicants is well suited for the
application of pretreatment control limitations, POTW
staff may choose to evaluate in-plant control of these
toxicants. An example is the treatment of ammonia by
optimizing the POTW activated sludge process (e.g.,
increase MCRT) to achieve nitrification.   In some
cases, TIE Phase I data on the classes of effluent
toxicant can be used to select options to be examined.
For example, if filterable  material  is  the  principal
effluent toxicant,  possible options would include
improved solids  clarification  or  granular  media
filtration.

In-plant  toxicity  control  may  be  achieved  by
enhancement of the existing treatment system or by the
implementation of additional treatment processes.  In-
plant control alternatives for different categories of
toxicants are summarized in Table 6-2. A description
of these control alternatives is provided as follows.

Process Enhancement
Biological Process Control
Biological process control is most easily applied to
suspended growth systems (e.g., conventional activated
sludge  and  BNR  processes),  although  some
modifications to fixed film processes (e.g., trickling
filters and RBCs) may be feasible. The performance of
activated  sludge  and  BNR systems  is generally
controlled by adjusting several process parameters,
including MCRT, MLSS, DO levels, recycle ratio, and
F/M ratio.  The treatment efficiency of the activated
sludge  system  is   optimized   by varying  these
interrelated process parameters.  A description of the
use of operational parameters for toxicant control is
provided as follows: "Removal of biodegradable toxic
compounds  in suspended growth systems  may be
improved  by increasing the MCRT" (Adams  et al.,
1981). MCRT can be increased by lowering the excess
sludge wasting rate. Longer MCRTs are necessary for
nitrification  and   can  be  beneficial  for  the
biodegradation of some types of organic compounds.
An example of this approach was practiced at a POTW
on  the United  States'  east  coast (Judkins  and
Anderson, 1992).   The  facility  was  retrofitted to
achieve nitrification  to reduce  ammonia.  Existing
treatment  capacity,  including aeration  basins  and
secondary clarification, was available to accommodate
the longer MCRTs and detention times needed to
accomplish  nitrification  and denitrification.   The
retrofits involved increasing the air supply, changing
the air diffuser pattern,  adding an anoxic zone in the
aeration basins, increasing the MCRT, and modifying
the return sludge flow. Usually, mixed liquor from the
aerobic zone of the  biological  treatment process is
recycled  to  the   anoxic  zone  to   accomplish
denitrification. However, it was possible in this case to
use the existing return sludge pumps to recycle the
secondary clarifier underflow to the anoxic zone. The
cost  of  the retrofit consisted  of approximately
$100,000 in capital costs and an increase in annual
operating costs of about 25%.

High MLSS concentrations have been shown to
minimize  the  effects  of  inhibitory pollutants  on
activated  sludge  treatment systems  (WEF/ASCE,
Table 6-2. POTW In-PIant Control Technologies for Categories of Toxic Compounds
Biodegradable
Organic Compounds
and Ammonia*
Biological process control
Nutrient addition
Coagulation/precipitation
Non-Biodegradable
Organic Compounds
Filtration
Activated carbon
Coagulation/precipitation
Volatile Organic
Compounds |
Biological process control
Aeration

Heavy Metals and
Cationic Compounds
Filtration
Coagulation/precipitation
pH adjustment
* Air stripping, breakpoint chlorination, and ion exchange also may be considered for ammonia removal; however, the cost of these
technologies and the use of toxic additives such as chlorine often preclude their use.
                                                 58

-------
1992a).   High MLSS concentrations increase  the
potential  for biodegradation  and sorption  of toxic
wastewater constituents and can help to protect the
treatment process from shock loadings. The maximum
MLVSS  will often  be  limited  by the  available
secondary clarifier capacity. It is important to consider
the effect of increased MLVSS on secondary solids
separation and the TSS concentrations of the clarifier
effluent.

A decrease in F/M  (based  on BOD5) effectively
decreases the organic waste loading per unit of
biomass, which may improve the biodegradation of
some toxic compounds (Adams et al., 1981).  The F/M
ratio is inversely related to MCRT.

Biological  process  control   is  not   as  easily
accomplished for fixed film  processes,  such as
trickling filters or RBCs.  Some adjustments can be
made, however, such as varying the amount and point
of wastewater recirculation in  a trickling  filter to
potentially increase  the removal of toxicants or
toxicity. In addition, secondary clarifier effluent can
be recirculated to  dilute high-strength wastes prior to
treatment in a trickling filter or RBC.  In some cases,
inhibitory pollutants may cause excessive sloughing of
the fixed film biomass. This problem may be rectified
by returning thickened secondary clarifier solids to the
fixed film process to help maintain a  proper biomass
population.

Chemical Addition
The addition of chemicals or additives to waste streams
in existing POTW treatment processes may improve
toxicant or toxicity removal. Nutrients can be added to
influent  wastewaters  that  have low  nutrient  levels
(relative  to  their  organic strength) to  improve
biological treatment. Lime or caustic chemicals can be
used to adjust wastewater pH for optimal biological
treatment or for  coagulation  and  precipitation
treatment. Other chemical coagulants are used to aid
in removal  of insoluble toxicants and to  improve
sludge settling. Powdered  activated  carbon may be
applied in activated sludge systems to remove toxic
organic compounds.  A description of each of these
treatment additives is provided as follows.

Addition of phosphorus, nitrogen, or sulfur may in
some cases improve biological treatment of industrial
wastewaters with low nutrient concentrations.  The
optimal BOD/TKN/TP ratio for municipal activated
sludge  treatment  is  100:5:1.  Lime and caustic
chemical addition may be used to increase influent
wastewater pH prior to primary sedimentation in order
toenhance the precipitation of heavy metals. Chemical
addition may also be appropriate for removal of metals
in sidestreams from sludge processing. Some metals,
however,  such as iron and  chromium will go into
solution rather than precipitate at alkaline pH.  The
optimum pH range for metals precipitation varies for
each type of  metal  and the solubility/precipitation
equilibrium can be affected by other factors such as
dissolved solids  concentrations in  the  wastewater.
Lime and caustic  chemicals also provide additional
alkalinity, which is essential  for biological treatment,
especially nitrification treatment, processes.

Polymers and inorganic coagulants such  as alum and
ferric chloride can be introduced to POTW waste
streams  to  help remove  insoluble   pollutants.
Coagulants may be added to influent wastewater to
increase the sedimentation of toxic constituents in
primary treatment and thereby minimize the loading of
toxicants  on  the biological  treatment  process.
Coagulants also can be added after the activated sludge
aeration basins to control sludge bulking or reduce
effluent suspended solids, which may be associated
with effluent toxicity.  The  optimum conditions for
coagulation can be determined by conducting bench-
scale jar tests. These tests are used to establish the
optimum coagulant type and dose, the proper mixing
requirements,  and the flocculent settling rates  for
treatment (Adams et al., 1981).

Coagulants can adversely affect the characteristics of
sewage sludges, which could affect the sludge disposal
method.  Coagulants may increase the toxicity of the
sludge (as measured by a TCLP) as a result of the
removal of toxic wastewater constituents or as a result
of the toxicity of the coagulant itself (e.g.,  metal salts).
Therefore, coagulants should be evaluated carefully
prior to use.

The addition of PAC to an activated sludge unit may
increase the removal of  toxic organic chemicals.
Organic pollutants that are not biodegraded can be
removed by adsorption onto the surfaces  of activated
carbon particles.   Activated carbon also  improves
sludge settleability by providing a substrate onto which
sludge  floes  can agglomerate.     Although  PAC
processes  have been used in municipal  wastewater
treatment, studies (Deeny et al., 1988) have shown that
PAC regeneration  by wet-air oxidation breaks down
the activated carbon particles to carbon fines, which
                                                 59

-------
cany over the secondary clarifier weirs. In some cases,
periodic additions of PAC to an aeration basin can be
used to minimize the effects of toxic slug loadings,
thereby improving the stability of the activated sludge
system.

POTW Modifications and Additions
Where process enhancement is not feasible or will not
provide adequate toxicant removal, physical addition
to or modification of the POTW can be undertaken.
Additional  treatment  processes  could   include
equalization prior to treatment, instrumentation control,
BNR, and advanced wastewater treatment processes
such  as  coagulation/flocculation,  granular  media
filtration, and GAG treatment. Public works managers
also may consider enhancing effluent dilution through
the addition of an outfall diffuser or relocation of the
outfall to a larger water body.

Equalization
Equalization can  be used  prior to  the biological
treatment  process  to  dampen the effect of slug or
diurnal loadings of high-strength industrial wastes.
Equalization facilities  can be provided to either
equalize   wastewater  flows   or   wastewater
concentrations. Flow equalization is partially provided
by existing primary sedimentation tanks and can be
enhanced by increasing the size of the primary tankage.
Concentration equalization  requires mixing of the
wastewater to moderate intermittent pollutant loadings;
therefore, separate facilities must be provided.

Instrumentation Control
Instrumentation/monitoring can be used to help control
slug loadings of  toxic constituents in the POTW
influent wastewater.  For example, transient metals
loadings may be monitored by continuously measuring
the pH and conductivity of the influent wastewater. A
significant change in pH or an increase in conductivity
may indicate a slug loading of toxic  material, which
can be manually or automatically diverted to a holding
basin.  After equalization, the diverted wastewater can
be slowly added back to the influent waste stream to
dilute the material prior to treatment.

Outfall Diffuser/Relocation
Public works managers may choose to evaluate the
alternative of installing  a diffuser or relocating the
outfall to  achieve better dilution.  For example, the
extension of a shoreline outfall to a submerged high-
rate diffuser in deeper water may promote rapid mixing
and achieve an acute dilution factor of 10 or more. If
allowed by applicable state water quality standards, the
effectiveness  of  outfall  relocation  or  diffuser
installation can be evaluated along with other control
options.  The reader is referred to USEPA's TSD
(1991b) for a discussion of the role of dilution in
permitting  for whole effluent toxicity control and
details on mixing zone analyses and high-rate diffusers.

Advanced Wastewater Treatment
POTWs  that  only utilize  primary  and  secondary
wastewater treatmetit may achieve toxicity reduction
by the addition of advanced  wastewater  treatment
processes   such   as   coagulation/flocculation,
sedimentation, grariular media filtration, and granular
activated carbon. Each of these processes can provide
enhanced removal' of some toxicants and toxicity.
Treatability tests used to evaluate treatment process
additions are described below.

Treatability Testing
Bench-scale  and  pilot-scale  treatability  tests  are
commonly used to evaluate treatment options that have
been selected for testing. Bench-scale or pilot-scale
tests offer several advantages compared to full-scale
testing, including a more manageable test unit size and
the ability to vary the operating conditions to evaluate
toxicity reduction.  Treatability  methods can range
from   simple  jar  tests for  testing coagulation/
flocculation options  to flow-through bioreactors for
investigating thebipdegradation kinetics of wastewater
treatment.
                  j
During treatability testing, influent, effluent, and
sidestream waste waters of the treatment simulation are
tested for acute or chronic toxicity. Toxicity testing is
used to assess the effectiveness of the treatment option
in reducing wastewater toxicity and to determine the
fate of toxicity in the treatment process. Initial testing
should  use the  simplified  toxicity test  methods
described in the TIE manuals (USEPA 1991a, 1992a,
1996) because of the large number of samples that may
need  to be tested.   Toxicity screening tests such as
Microtoxฎ also may be used in conjunction with the
required test species to provide additional information.
These tests are recommended for waste streams with a
high oxygen demand (i.e., high BOD5 concentration),
which would otherwise require aeration when testing
with  permit species.  Aeration  should be avoided
because it may remove volatile or oxidizable toxicants.

Definitive  acute or  chronic toxicity tests  (USEPA
1993c,  1994a, 1994b, 1995) should be used at the
                                                  60

-------
completion of treatability testing to verify the option's
capability  in  meeting  the NPDES  permit  limit.
Optional TIE Phase I analyses also may be performed
on treatability test samples to confirm toxicant removal
by the treatment option.

Activated Sludge/BNR Treatment
The basic  parameters  of interest  in  the design  of
activated sludge/BNR systems include organic loading,
oxygen requirements, nutrient requirements, sludge
production, sludge settleability, and internal recycle
rates.  Continuous flow systems are most useful for
evaluating  activated  sludge/BNR systems; however,
batch  systems  may provide  sufficient treatability
information in some cases.  An example of the use of
batch  treatment  tests  in  a TRE is provided  in
Appendix D. This study determined that an upgrade of
a conventional activated sludge process to a five-stage
BNR process would achieve compliance with chronic
toxicity limits. Follow-up monitoring upon completion
of the upgrades confirmed the toxicity reduction.

Consideration should be given to evaluating design
specifications   and operating  conditions   that  are
expected to optimize the treatment of toxicants and
toxicity. These parameters may include relatively long
MCRTs and high  MLVSS levels,  which have been
shown to improve toxic pollutant removal and protect
the  process from inhibitory wastes (Hagelstein and
Dauge, 1984; WEF/ASCE,  1992a).

Coagulation/FIocculation
The  evaluation  of  coagulation  and flocculation
treatment involves the use of bench-scale jar tests or
zeta potential tests to provide  information on  the
optimum coagulant type and dosage, mixing rates, and
flocculent  settling rates for removal  of solids and
flocculent suspensions (Adams et al.,  1981).  Results
of these tests are used to devise treatability tests to
evaluate the sedimentation of flocculent suspensions.

Sedimentation
Sedimentation  involves the removal  of suspended
solids or flocculent suspensions by gravity settling.
Sedimentation is evaluated  by conducting a series of
settling column tests that measure the settling rates of
solids or flocculent suspensions (Adams et al., 1981).
Test results are  used to calculate a settling profile that
can be used for  clarifier design.
Granular Media Filtration
Filtration testing involves scaled-down models (usually
pilot-scale) of full-sized filters. The choice of filter
media and test-flow rates should correspond to the
intended design and operating criteria.  Although the
process scale  is reduced, the bed gradation and
thickness should be equivalent to anticipated full-scale
processes   in   order  to  predict  actual  treatment
performance (Adams et al., 1981).

Granular Activated Carbon
The carbon  adsorption  isotherm test is used  to
determine the optimum type and dosage of activated
carbon for wastewater treatment (Adams et al., 1981).
Results of this test are used to prepare bench-scale or
pilot-scale carbon columns that can be used to evaluate
carbon exhaustion  rates  and  the  effect  of carbon
regeneration on toxicity removal performance.

Toxicity Control Selection
The final process of toxicity control selection involves
an assessment of potential control options and selection
of the best option(s) for toxicity reduction based on
several criteria. In most cases, both a pretreatment
control evaluation and an  in-plant control evaluation
will have been  performed;  therefore, the review
information should include the data developed in both
evaluations.

The choice of in-plant toxicity control or pretreatment
toxicity control will depend largely on the technical
and  economic  feasibility of  POTW  treatment
modifications and pretreatment controls. Pretreatment
control will be feasible in situations where the TIE data
and the toxicity source evaluation data are sufficient to
definitively identify the sources of toxicity. These data
should provide an  indication of the  variability  of
toxicity and toxicants in the indirect discharge. If these
conditions are  satisfied, POTW staff  can set local
limits using the methods  outlined above.  In-plant
control  will  be  preferred  in cases where  the
implementation of feasible treatment modifications or
additions is more practical than pretreatment control.
Data obtained  in treatability studies should include
information on the variability of toxicity treatment
performance and the design criteria for implementing
the treatment option. In-plant options provide POTW
staff a direct method of controlling effluent toxicity;
however,  in-plant modifications  or additions  may
substantially increase process operation requirements
and maintenance costs.
                                                 61

-------
Selection ofToxicity Control Options
Final selection  of the  preferred toxicity  control
option(s) involves a comparison of the options using
appropriate criteria (see example in Table 6-1). It may
be necessary to select and implement more than one
control  option  to  ensure  compliance with effluent
toxicity requirements.   The  preceding evaluations
should provide sufficient information to document the
technical and cost considerations for each option.

Compliance with Effluent Toxicity Limits or
Requirements
Data gathered through the TRE should indicate that the
selected option will consistently achieve compliance
with the  toxicity  permit requirement.   Sufficient
information should be provided to show that the option
will reduce effluent toxicity even during periods of
maximum occurrence of toxicity. If this information
includes bench and/or pilot-scale  treatability  data,
scale-up factors must be incorporated into estimates of
toxicity reduction to adjust for differences in treatment
efficiency between laboratory and full-scale treatment
systems. Likewise, safety factors should be included
in the calculation of local limits to allow for variation
in toxicant loadings to the POTW.  It  also may be
necessary to conduct a sensitivity analysis to evaluate
the  effectiveness  of the  options  under  variable
conditions (e.g., variable toxicant loadings or treatment
performance).

A relative  scoring system can  be  used to  rate the
overall potential for the options to achieve permit
compliance. The scores can be entered into a matrix
table like that shown in Table 6-1.

Compliance with Other Permits
Steps taken to  reduce effluent  toxicity may have a
detrimental effect on other permitted activities such as
sludge disposal or air emissions.  If toxicants are
expected to be transferred to sludge or air, the potential
effects on limitations specified in residuals  and air
permits  should be estimated.  Each option should be
rated for its potential to comply with related permits.

Capital, Operation, and Maintenance Costs
Sufficient detail on costs should be presented to allow
a straight-forward comparison of the control options.
Cost estimates should include the effort and materials
required for planning, implementation, operation, and
maintenance of the options. Cost information may be
obtained  from  equipment  vendors,  engineering
consultants, and existing data for comparable systems.
Costs for requisite environmental and construction
permits should be included.

In some cases, it may be possible to recover some of
the  costs  of  implementation   from  responsible
dischargers. For example, municipalities may apply
surcharges to local limits or request in-kind funding for
POTW modifications or additions to recover the costs
of toxicity control.  Anticipated cash returns should be
included in the final cost estimate.

Costs for all options can be ranked and a score can be
assigned and entered into a matrix table. Weighting
factors may be incorporated into the scoring if funding
of some options is uncertain.

Ease of Implementation
Factors such as land availability, permits, operability,
and maintenance will have a major influence on the
selection of options involving POTW modifications or
additions. Likewise, the economic impact and level of
community  cooperation   anticipated  from  new
pretreatment regulations will  affect the selection of
pretreatment control options. Public works managers
should develop a list, of all potential constraints as well
as benefits of the candidate control methods. Benefits
should address items  other  than effluent toxicity
reduction such as improved treatment conditions or
better  cooperation  among POTW  users.   Each
constraint and benefit  can  be assigned  a weighted
score, the individual values can be summed for each
option, and the total value entered into the matrix table.

Reliability
The   selected  option(s)   must   be  dependable.
Pretreatment approaches or treatment processes that
tend to malfunction or  fail because of difficulties in
executing complicated  operational plans should  be
avoided. Experiencein  implementing similar projects
will be useful in defining the reliability of the options.
Public works managers  should consider each option's
operational history, maintenance  requirements, and
longevity.

Environmental Impact
Some  options   may   require  the  evaluation  of
environmental issues  related to  the  protection  of
wetlands, rare and endangered species, and cultural
resources. Although the costs of these evaluations are
included under the above cost criteria, other factors
will  affect the  decision-making  process,  including
public perception, time  period for permit approval (if
                                                 62

-------
needed), and potential remediation issues. A score can
be developed based oh these factors and entered into
the matrix table.

Comparison of the Toxicity Control Options
Scores developed in the criteria evaluation are summed
for each option.  These scores will  incorporate all
necessary weighting factors; therefore, the total scores
for each option can be compared directly. The options
can be ranked according to their scores and the highest
ranked option(s) can be selected for implementation.
In some cases, it may be necessary to select more than
one  toxicity  control option  to  ensure that permit
compliance will be achieved. This approach is highly
recommended when the control options are relatively
inexpensive to implement, operate, and maintain.
                                                 63

-------
                                           Section 7
                           Toxicity Control Implementation
 Introduction
 Once the evaluation and selection of toxicity control
 options has been completed, the final steps in the TRE
 are the implementation of the selected pretreatment
 and in-plant control options and follow-up monitoring
 to ensure permit compliance. The degree of effort in
 the implementation step will depend on the severity of
 the effluent toxicity and the complexity of the selected
 control approaches. Depending on the  findings of the
 TRE, implementation may involve relatively minor
 changes  such  as   modifying  POTW  operating
 procedures or more complex modifications such as
 expanding  the  POTW's pretreatment  program or
 designing and constructing new treatment facilities.

 Implementation
 Using the results of the previous steps in the TRE, a
 Toxics Control Implementation Plan (TCIP) should be
 developed. The TCIP should summarize the results of
 the TRE, results of the screening and  selection of
 toxicity control options, and justification for selecting
 the preferred toxicity control option(s).  For in-plant
 control options, the TCIP should provide the basis of
 design for the  selected control options,  including
 capital and operating costs, and a schedule for design
 and construction. For pretreatment control options, the
 TCIP should specify the basis of  selection  and
 technical justification for local limits and discharger
 monitoring methods.  In addition, the procedures for
 implementing revised pretreatment regulations also
 should be defined.

Follow-Up Monitoring
 Once a control technology has been implemented, a
follow-up monitoring program should be developed
and implemented to ensure the effectiveness of the
selected control option(s).   In most  cases,  the
conditions and frequency of monitoring will be set by
the  regulatory agency.   If source controls  are
implemented, POTW staff should specify additional
monitoring requirements for indirect dischargers under
the pretreatment program.  These requirements may
include verification of statements from industries that
the required reduction of toxicity has been made.

The POTW effluent monitoring program should be
designed to provide data to ensure that toxicity has
been reduced to acceptable levels and that the TRE
objectives have been met. This program may involve
more frequent monitoring than is required by the
NPDES  permit, including monitoring to evaluate
daily,  weekly,  monthly, or  seasonal variations  in
effluent toxicity that were observed during the TRE.
Follow-up monitoring should utilize the test species
and methods  specified in the discharge permit.
Additional tests, including surrogate methods applied
in  the TRE,  may  be  included to  re-evaluate
correlations between test  species that may have
changed as a result of the effluent toxicity reduction.

Any effluent  toxicants  that were determined to be
present prior  to  implementation of the control
technology  should  be  monitored to ensure that
concentrations are below toxic levels.  Approved
analytical methods! will generally be applied; however,
screening methods such as ELISA tests or other field
kits, which may not be specifically approved by
USEPA,  can be used to evaluate trends and identify
potential problems for follow-up testing.  As with
toxicity monitoring, the analytical program should re-
evaluate trends in toxicant concentrations  observed
during the TRE. A discussion of an ongoing POTW
monitoring program for organophosphate pesticides is
described in Appendix F.
                                                64

-------
                                          Section 8
                         Quality Assurance/Quality Control
Introduction
A QA/QC program for the TRE should be developed
and implemented to  ensure  the  reliability of the
collected data.  The QA/QC program should include
addressing the monitoring of field sampling  and
measurement  activities,  the  review of laboratory
analysis  procedures,  and the documentation  and
reporting of the analytical data. A QA/QC program
should be designed so that corrective action can be
quickly implemented to detect and eliminate erroneous
or questionable data without  undue expense to the
project or major delays in the schedule.

The POTW laboratory manager should ensure that the
specific QA/QC requirements for TRE activities are
addressed by the facility's QA/QC plan. If a private
consultant is to be used for all or part of  the TRE
testing, the POTW laboratory manager should request
a QA/QC plan from the consultant and review the
consultant's proposed QA/QC activities.  Whether the
TRE is to be performed by the POTW laboratory or by
a consultant, it is essential that the project organization
include  competent  chemists,  lexicologists,  and
engineers who have  adequate knowledge of TRE
methods.

The QA/QC plan should be  prepared  prior to the
initiation of the TRE and should contain the following
elements:

  •  QA/QC objectives
  •  Sample collection and preservation techniques
  •  Chain of custody procedures
  •  Analytical QA/QC
  •  Laboratory equipment maintenance
  •  QA/QC training requirements
  •  Documentation and reporting procedures
  •  Corrective action protocols.
Sample Collection and Preservation
To  ensure quality  control in  sample collection
activities, the TRE sampling plan (Section 11) should
be strictly followed.  In addition, the QA/QC plan
should state the minimum sample volumes, maximum
sampling holding times,  and  sample preservation
techniques for each analytical method. The sampling
requirements for conventional and priority pollutant
analyses  are described in USEPA's Methods for
Chemical Analysis of Water and Wastes (USEPA,
1983b) and Standard Methods for the Examination of
Water and Wastewater (APHA, 1995).   Sampling
requirements for acute toxicity tests are provided in
USEPA's Methods for Measuring the Acute Toxicity
of Effluents to Freshwater and Marine  Organisms
(USEPA, 1993c) and Methods for Aquatic Toxicity
Identification  Evaluations:  Phase  I,   Toxicity
Characterization  Procedures  (USEPA,  1991a).
Sampling requirements for chronic toxicity tests  are
provided in USEPA's  Short-Term Methods for
Estimating  the  Chronic Toxicity of Effluents and
Receiving Waters to Freshwater Organisms (USEPA,
1994a),  Short-Term  Methods for  Estimating  the
Chronic Toxicity of Effluents and Receiving Waters to
Marine and Estuarine Organisms (USEPA, 1994b),
Short-Term Methods for Estimating the  Chronic
Toxicity of Effluents and Receiving Waters to West
Coast Marine  and Estuarine Organisms (USEPA,
1995),   Toxicity   Identification   Evaluation:
Characterization of  Chronically Toxic  Effluents,
Phase  I (USEPA,  1992a), and Marine  Toxicity
Identification Evaluation (TIE) (USEPA,  1996).

It is important to routinely assess the effects of sample
holding times on wastewater toxicity to predict how
long samples can be kept before changes in toxicity
occur.  For example,  the acute TIE Phase I manual
(USEPA, 1991a) describes how  testing the sample
toxicity on the  day of collection and comparing this
initial toxicity  to its baseline toxicity (tested 1 day
                                               65

-------
 later) can provide information on appropriate sampling
 holding times for toxicity analysis. In chronic TIEs,
 effluent manipulations are performed on the day the
 sample is received so that the possible effects of any
 toxicity degradation are minimized (USEPA, 1992a).

 Other  QA/QC  considerations  for  TRE  sample
 collection include routine cleaning and inspection of
 automatic  sampling  equipment,  cleaning  sample
 containers  according to the requirements for each
 analytical method, and collecting  duplicate samples
 and field blanks.   When preserving  samples  for
 chemical analysis, only analytical grade preservatives
 should  be  used  to  avoid contamination  and
 overestimation  of   analyte   concentrations.
 Unpreserved samples that are to be used for toxicity
 and chemical analyses require sample containers that
 are  both  lexicologically  and analytically  clean.
 Equipment  and containers  used for  toxicity test
 samples require special cleaning procedures outlined
 in USEPA manuals (1993c, 1994a, 1994b).

 Chain-of-Custody
 A chain-of-custody (COC) form should accompany all
 samples to document the collection, preservation, and
 handling of samples. The COC form should indicate
 the sample identification number, sample type (i.e.,
 composite or grab), date and time of collection, a brief
 description of the sample, number of samples taken,
 name of the person taking the sample and performing
 field measurements, and sample characteristics such as
 temperature, pH, total and free residual chlorine, and
 conductivity. A field book also should be used to
 record any field observations or conditions noted
 during  sampling  along   with  other  pertinent
 information. Each laboratory should identify a sample
 custodian to log in and store samples collected during
 the TRE. The sample custodian should acknowledge
 receipt of samples by signing the COC form and
 noting the date and time of sample receipt, the sample
 identification number, the laboratory assession code,
 and sampling information such as temperature, pH,
 and TRC.  Upon receipt of the sample, a  sample
tracking form should be used to record the date, time,
 and volume of aliquots  of the sample removed for
analysis, the analyst, and any changes in the nature of
 the sample, including its toxicity, over time. All COC
and sample tracking forms should be maintained in a
permanent file so that information on specific samples
can be traced easily.
 TRE Procedures
 Analytical tests should provide data of an acceptable
 quality for characterizing wastewater toxicity and for
 evaluating methods  and technologies  for toxicity
 reduction.  Several test methods described in this
 document are not standard procedures  and require
 careful  attention to  unique  QA/QC  procedures.
 Special QA/QC procedures for each major TRE test
 are  discussed  below.   Whenever  possible,  these
 procedures should be followed to ensure precise and
 accurate results.

 Toxicity Identification Evaluation
 Special precautions for TIE tests are discussed in the
 Phase I, E, and Hi! manuals (USEPA 1991a, 1992a,
 1993a, 1993b, 1996).  In general, strict adherence to
 standard quality control practices is not required for
 conducting Phase I analyses due to the large number
 of toxicity tests to be performed and the tentative
 nature of the toxicant characterization. Nonetheless,
 system blanks and controls should be used whenever
 possible to indicate toxicity artifacts caused by the
 characterization procedures.   In Phase n  more
 attention should be i paid to quality control in order to
 identify interferences in toxicant characterization and
 identification. Evert greater attention to quality control
 is needed in Phase ;in.  Sample manipulation should
 be  minimized  in Phase El  to  prevent  analytical
 interferences and toxicity artifacts. Field replicates,
 system blanks, controls,  and calibration  standards
 should be used extensively to allow a  precise and
 accurate determination of the sample toxicants and
 toxicity.

 Specific precautions for characterization (Phase I) and
 toxicity testing in TIE analyses are provided below.

 Aeration
 For air stripping or aeration tests, only a high quality
 compressed air source should be used. Oil, water, and
 dirt are undesirable contaminants in compressed air;
 therefore, it is important to use equipment and filters
 that  generate   dry,  oil-free  air.   Oil-sealed  air
compressors should not be used.  Simple aeration
devices, such as those sold for use with aquariums, are
 acceptable provided that the  ambient laboratory air is
uncontaminated (USEPA, 1991a). Recommendations
for  in-line filters, for  air  exchange  systems in
laboratories are provided by USEPA (1993c).
                                                66

-------
Filtration
High  purity water,  which has been  adjusted to a
specified pH, should be used to rinse filters between
filtration steps (USEPA 1991a, 1992a).  Filtration
equipment should be rinsed with  10%  nitric acid
(HNO3), acetone, and high purity water between
sample aliquots. Filtration equipment should be made
of plastic to avoid leaching of metals or other toxicants
during acid washes.  Toxicity can be  checked by
testing filtered dilution water.

pH Adjustments
Concerns in the pH adjustment steps involve artificial
toxicity caused by excessive ion concentrations from
the addition of acid and base solutions, contamination
from  impure acid  and  base solutions,  and silver
contamination from some pH probes (USEPA 1991a,
1992a). The baseline toxicity test acts as a control for
indicating whether addition of acid and base solutions
increases effluent toxicity. Ultra-pure acids and bases
should be used to minimize artificial toxicity. During
pH measurement, toxic concentrations of silver can
leach from refillable calomel electrodes; therefore,
only solid state pH probes should be used.

Methanol/C18 SPE Column
HPLC  grade methanol is  required for CIS SPE
column preparation and extraction steps.  A blank
toxicity test should  be conducted for each methanol
reagent lot. In addition, a toxicity blank should be
performed on each  CIS SPE column to check for
resin-related toxicity (USEPA 199 la, 1992a).

Sodium Thiosulfate Addition
The TIE manuals  (USEPA  1991a,  1992a, 1996)
provide  information  on  the  toxicity  of sodium
thiosulfate to several freshwater and marine species.
These manuals  prescribe the  amount  of sodium
thiosulfate to use in  testing. If alternative species are
to be used, the species tolerance should be evaluated
by adding increasing quantities of sodium thiosulfate
to aliquots of the sample, testing the resulting toxicity,
and comparing the toxicity to the sample's baseline
toxicity.

EDTA Addition
The TIE manuals (USEPA 1991a, 1992a, 1996) also
prescribe the concentration  of EDTA ligand to be
added to samples. If alternative species are to be used
in the TIE, the same test approach noted above for
sodium thiosulfate can be applied.
Toxicity Tests
The organisms used to test the sample toxicity prior to
and following each characterization step should not be
subject  to  undue stresses  such as contamination
(USEPA, 1991a). The test organisms should have had
no prior exposure to pollutants and their sensitivity
should be constant over time. To assess changes in the
sensitivity of the test organisms, a standard reference
toxicant test should  be performed on a regular basis
and accompanying quality control charts should be
developed (USEPA 1993c, 1994a, 1994b). Reference
toxicant tests should be performed monthly.  If test
organism cultures are not maintained in the laboratory,
reference toxicant tests should be performed with each
group  of  test organisms  received, unless  such
information is available from the vendor.  Information
on obtaining and culturing species for toxicity testing
is provided in  the  acute and chronic toxicity test
manuals (USEPA 1993c, 1994a,  1994b).

The quality of the dilution water used in toxicity tests
will depend on the purpose of the TIE test and whether
the  test  is  being  performed for  toxicant
characterization (Phase I), identification (Phase H), or
confirmation (Phase ID). Much of Phase I and parts of
Phase  II  rely  on  relative  toxicity  measurement;
therefore, water that is of consistent quality and will
support the growth and reproduction of the test species
is suitable for these phases of the TIE (USEPA 199 la,
1992a, 1996). The objective of Phase m, however, is
to confirm the  true cause  of  toxicity;  therefore,
artifacts are to be excluded and the choice of dilution
water should follow Phase HI  guidance (USEPA,
1993b). Guidance for preparing the dilution waters is
described by USEPA (199 la, 1992a, 1996).

USEPA (199la)  recommends feeding  cladocerans
(i.e., C.  dubia and Daphnia sp.) in the TIE test
solutions at the beginning of acute TIE toxicity tests.
Daily feeding is  required  in the  chronic TIE tests
(USEPA, 1992a). Feeding requirements for selected
species are described in the acute and chronic toxicity
test manuals (USEPA 1993c, 1994a, 1994b, 1995).

Sample pH  should be recorded  at each  sample
renewal. Additional pH measurements may be needed
during the test, especially if ammonia toxicity is a
concern.

DO measurements may be made at sample renewal or
at the end of the exposure period in the TIE. In cases
where low DO is a problem, DO adjustment should be
                                                 67

-------
 performed at a rate that will not intentionally change
 the sample toxicity.

 Refractory Toxicity Assessment and Treatability
 Tests
 RTA and treatability tests are subject to a variety of
 potential interferences due to the large number of
 variables that must be accounted for and controlled
 during testing. When performing RTA and treatability
 analyses, it is  important to hold  all parameters
 potentially affecting toxicity constant so that sample
 toxicity is the sole variable. Important parameters to
 be controlled in RTA testing include the test solution
 temperature, DO level, and pH.

 The QA/QC concerns for toxicity analysis in RTA and
 treatability tests are the same as those noted above for
 TIE tests. Selection and use of test species and dilution
 water should follow  procedures given in the USEPA
 Phase I manual (USEPA 1991a, 1992a).

 Potential sources of toxicity contamination should be
 identified through the use of system blanks. As in TIE
 testing, the filters used in RTA testing should be tested
 to  determine if toxicity is added during filtration.
 Each of the solutions used in RTA testing, including
 activated sludge, should be checked for toxicity.  In
 the Patapsco TRE, the RAS used in the RTA batch
 tests was found to be acutely toxic to C. dubia (Botts
 et al., 1987). Steps for addressing RAS toxicity are
 described in Section 5. Similarly, the reagents used in
 treatability testing such as chemical coagulants should
 be screened for toxicity.

 Held replicates,  calibration standards, and analytical
 replicates should be routinely performed during RTA
 and treatability  testing.   Results  of these  quality
 control analyses can be used to calculate the precision,
 accuracy, and the sensitivity of each physical/chemical
 analysis method used in these studies.

 Chemical Analyses
 Quality control for chemical analyses includes the use
 of  calibration standards, replicate analyses, spiked
 sample analyses, and performance standards.  The
 detection limits  and  the recommended reagents for
 method  calibration  and spiking are discussed  in
 USEPA's Methods for Chemical Analysis of Water
 and Wastes (USEPA, 1983b) and Standard Methods
for Examination of Water and Waste-water (APHA,
 1995).  General information on laboratory  quality
 control for chemical analyses is provided in USEPA's
 HandbookjbrAnalytical Quality Control in Water and
 Wastewater Laboratories (USEPA, 1979a).
Equipment Maintenance
All facilities and equipment such as pH, DO, and
conductivity meters, spectrophotometers, GC/MS, and
HPLC instruments should be inspected and maintained
according to manufacturers' specifications. Standard
operating procedures (SOP) should be followed for
routine maintenance land calibration of each analytical
instrument.  A maintenance log book also should be
kept for each major laboratory instrument.

The measurement of toxicity or trace compounds in
wastewater  samples  requires  the use of carefully
cleaned instruments and glassware.  Instruments that
involve flow-through analysis such as  automated
spectrophotometers should be inspected to ensure that
flow-through parts ' (i.e.,  tubing) are periodically
replaced. New glassware may be contaminated with
trace amounts of metals;  therefore, any glassware
being used in toxicity tests for the first time should be
soaked for three days' in 10% HNO3 (USEPA, 1991a).
For  subsequent use in TIB and toxicity  tests,  the
glassware should  be  washed  with phosphate-free
detergent, and sequentially rinsed with 10% HNO3,
acetone, and finally high-purity water (USEPA 1993c,
1994a, 1994b).

Documentation and Reporting of Data
Basic steps in a successful QA/QC program are the
documentation of the analytical data in meaningful,
exact terms, and reporting the analytical  data in a
proper form for future interpretation and use.  To
ensure the reliability of the data, its handling must be
periodically monitored and reviewed.  This review
generally consists of three elements: an assessment of
laboratory record-keeping procedures, a review of the
data calculations, and a review of the final reported
data.  On the basis: of these  review steps and  the
QA/QC analyses for precision and accuracy, the data
are accepted or rejected.  This  review process is
essential because some or all records may have to be
submitted for review by State or Federal regulatory
agencies.

Corrective Action
Procedures should be established  to ensure  that
QA/QC  problems  such  as  improper  sampling
techniques,  inadequate  COC  records,  and  poor
precision  and   accuracy   results   are   promptly
investigated and  corrected.    When  a  QA/QC
deficiency is noted, the cause of the condition should
be determined and corrective action should be taken to
preclude repetition.
                                                 68

-------
                                           Section 9
                                     Health and Safety
Introduction
A health and safety (H&S) plan may be necessary to
establish policies and procedures to protect workers
from hazards posed by TRE sampling and analytical
activities.   The general guidelines outlined in this
section  should be  integrated into  existing  H&S
programs even if a specific H&S plan is not required.
Whether a specific H&S plan is necessary or not will
depend  on  the conditions under which  the TRE is
being conducted. For example, if the POTW operates
under an RCRA permit by rule, then H&S must be
addressed when collecting and analyzing hazardous
wastes.

Important considerations for H&S for TRE studies
include:

  •  Identification of personnel responsible for H&S
    matters
  •  H&S information and training activities
  •  Protective equipment required for TRE activities
  •  Materials cleanup and disposal procedures
  •  Emergency response contingencies.

Detailed information on the preparation and scope of
H&S plans is provided in the Occupational Safety and
Health Administration's (OSHA's) Safety and Health
Standards,  General Industry (OSHA, 1976).   The
following   subsections  discuss   specific   H&S
considerations for selected TRE activities.

Sample Collection and Handling
Working with waste streams, of unknown composition
is inherent to  TREs.  Samples  of industrial  sewer
discharges, municipal wastewater, and sewage sludge
can contain a variety of toxic and hazardous materials
(e.g., pathogens, carcinogens) at concentrations that
can be harmful to human health.
It  is the  responsibility of the laboratory  sample
custodian  to ensure that TRE samples are properly
stored,  handled,  and  discarded  after  use  (see
Section 8).    Upon sample  storage, the . sample
custodian should indicate the H&S considerations for
sample handling and disposal.

Exposure to toxic and hazardous sample constituents
should be minimized during sampling handling. The
principal routes of human exposure to toxics is via
inhalation, dermal  absorption, and/or  accidental
ingestion.  Exposure can be minimized through the use
of proper laboratory safety equipment such as gloves,
laboratory aprons or coats, safety glasses, respirators,
and laboratory hoods. Laboratory hoods are especially
important  when testing wastewaters containing toxic
volatile substances such as volatile priority pollutant
compounds, hydrogen sulfide, or hydrogen cyanide.
Proper dermal  protection such as using neoprene
gloves for solvent-containing wastes also is important.
Laboratory  managers   should  consult   the
manufacturers' specifications in selecting appropriate
clothing materials for  protection  against specific
chemicals.

Residual wastewater samples  and  wastes  generated
during TRE studies should be disposed of properly.
Residual  municipal  wastewater  and other  non-
hazardous wastes can be disposed directly into the sink
drain if the TRE is being conducted at the POTW.
Residual industrial samples and other wastes that may
contain hazardous materials should be decontaminated
and/or disposed of in accordance with hazardous waste
regulations (NIOSH, 1977).

TRE Methods
Specific precautions to be, followed for selected TRE
techniques are described below.
                                                69

-------
Toxicity Identification Procedures (TIE)
USEPA's TIE Phase  I manuals (USEPA  1991a,
1992a,  1996) address  the  general  H&S concerns
involved in performing TIE testing.  Ventilation is a
specific concern when performing the Phase I aeration
procedure. The aeration test should be performed in
laboratory hoods to prevent exposure to toxic volatile
compounds or pathogens resulting from aeration.

H&S considerations  for aquatic toxicity testing are
addressed in USEPA's toxicity test manuals (USEPA
1993c, 1994a, 1994b, 1995). Special precautions need
to be taken  for on-site mobile laboratories in the
handling and transportation of chemicals, supply  of
adequate ventilation  and safe electrical power, and
disposal of waste materials.

Refractory Toxicity Assessment and Treatdbility
Tests
Proper ventilation also is important when conducting
RTAs and treatability tests in the laboratory.  Hoods
should be used to capture and vent potential volatile
compounds  that are stripped from  the  wastewater
during biological treatment tests.

Physical/chemical treatability testing may involve the
use of hazardous reagents such as acids or caustics.
Caution should be taken in the handling and disposal
of these chemicals.

Chemical Analyses
Several reagents used for chemical-specific analyses
(e.g.,  priority  pollutants, COD,  etc.) are  toxic or
hazardous substances.  Analysts  should be familiar
with safe handling procedures for all reagents used in
testing, including the practice of proper chemical
storage to avoid  storing incompatible chemicals
together (NIOSH, 19J77;OSHA, 1976). After use, the
waste  chemicals should be  converted  into a  less
hazardous form in  the laboratory before disposal or
disposed of by a commercial disposal specialist.

General Precautions
USEPA (1977) and the American Chemistry Society
(1979)   describe  additional  laboratory  safety
procedures that can be used in TRE studies, including:

  •  Use of safety arid protective  equipment such as
    eye protection (safety goggles, eye wash), fire
    hazard protection (smoke and fire detectors, fire
    extinguishers), and electrical shock protection
    (ground-fault interrupters for wet laboratories).
  •  Protocols for emergency response and materials
    cleanup.
  •  Personnel training in H&S procedures.
                                                70

-------
                                          Section 10
                                Facilities and Equipment
Introduction
Laboratories should be equipped with all the basic and
specialized laboratory equipment required to conduct
the TRE, and laboratory personnel should be skilled
and experienced in operating this equipment.  The
facilities and equipment needed to perform a TRE will
be different for each POTW and will depend on the
type of testing to be performed in the TRE. In general,
the minimum facilities and equipment for initiating a
TRE will include the equipment needed for toxicity
and Tffi testing (USEPA 1991a, 1992a, 1993c, 1994a,
1994b,  1996).  As additional information becomes
necessary, facility and equipment needs will depend
on  the  physical/chemical  characteristics of the
causative toxicants aiid the toxicity control approaches
to be evaluated. For example, the selection of bench-
scale  equipment and/or  pilot  plant  facilities for
treatability studies will be  dictated by the control
options to be tested (e.g., physical/chemical processes
such as filtration or biological processes).

The choice of whether to work on-site or off-site will
depend on the stage of the TRE, the  approach for
tracking sources of  toxicants or toxicity, and the
requirements for treatability testing. In general, the
equipment and time required for conducting TEE tests
makes on-site testing less  feasible. If the loss of
sample toxicity over time is minimal, TIE samples can
be shipped and tested off-site, usually at much less
cost than on-site testing.  If toxicity tracking  using
RTA tests is required, on-site testing is recommended
for the treatment phase of  the RTA, because fresh
samples of the POTW RAS biomass must be used.
Treatability tests that require continuous supplies of
POTW  influent or  process  wastewaters and/or
activated sludge (i.e., flow-through bioreactor tests)
also may be more efficiently  conducted in on-site
facilities. Some treatability evaluations require unique
or  sophisticated equipment  (e.g.,  ultra-filtration
apparatus) that is not readily available for on-site
work. In these situations, the equipment vendor may
be able to conduct the required tests at their facility.

The general equipment requirements for each of the
main TRE methods are summarized below.   H&S
equipment is discussed in Section 9.

Toxicity Identification Evaluations
Laboratories should be equipped with the equipment
and materials needed to conduct the TEE,  including
filtration and air-stripping equipment, pH and DO
meters,  CIS SPE  columns,  fluid  metering pumps,
required reagents  for the TEE  manipulations, and
facilities for organism handling, water preparation,
sample holding, and glassware cleaning. Equipment
requirements for culturing standard test species are
described in USEPA's acute and chronic toxicity test
manuals (USEPA 1993c, 1994a, 1994b, 1995).

More sophisticated analytical equipment is required
for the TEE Phase n toxicant identification and TEE
Phase  in toxicant confirmation procedures.  The
choice of analytical instruments for these procedures
will depend  on  the  compound  to be measured.
Equipment may include an analytical balance, a
GC/MS, an  HPLC,  an atomic  absorption  (AA)
spectrometer, an inductively coupled plasma  (ICP)
spectrometer, an ultraviolet-visible spectrophotometer
(UV-VIS), an ion  chromatograph, ion  selective
electrodes, a pH meter,  a conductivity meter,  and a
refractometer.   Use  of inert  materials  such   as
perfluorocarbon plastics for TEE Phases n and HI are
recommended to protect  against toxicity artifacts
(USEPA, 1991a).


Refractory Toxicity Assessment and
Treatability Tests
Laboratories  should  be equipped  with  the  basic
equipment for setting up and operating the RTA batch
                                                 71

-------
reactors, including an air supply, electrical supply, and
a laboratory hood.  Instruments for monitoring the
batch reactors include  respirometer and/or oxygen
meter, pH meter, ion selective electrode meter and
probes, total  TOC analyzer, spectrophotometer for
COD  and  nutrient  (e.g.,  ammonia  and nitrate)
analyses.   A drying  oven,  muffle  furnace,  and
analytical balance will be needed for TSS and VSS
measurements.

The equipment for toxicity testing will depend on the
choice of toxicity screening tests. Depending on the
species to be used, it may be more economical to
culture the test organisms than to purchase them.  In
some cases, rapid screening tests such as a bacterial
bioluminescence test (e.g., Microtoxฎ) may be used as
a surrogate  method for toxicity  testing (see the
Billerica, Massachusetts, case history in Appendix A).

General Analytical Laboratory Equipment
General laboratory equipment such as refrigerators, a
water  purification'  system,  and commonly  used
reagents  are neede'd  to support the  TIE and RTA
analyses.  The type of water purification  system
needed for testing !is described by USEPA (1993c,
1994a, 1994b, 1995).
                                               72

-------
                                          Section 11
                           Sample Collection and Handling
Introduction
The most important criterion in sampling is to obtain
a sample  that is representative of the discharge.
Several samples will need to be collected to ensure
that the samples represent the typical lexicological and
chemical quality of the wastewater. Guidelines for
sample collection and handling are presented in the
acute and chronic toxicity test manuals  (USEPA
1993c, 1994a, 1994b, 1995) and the Phase I TEE
documents (USEPA 1991a, 1992a, 1996). The WEF
also  has  published  a useful guide  to sampling at
POTWs (WEF, 1996).

A sampling plan should be prepared to document the
procedures to be followed in TRE sampling.  This plan
should include:

  •  A description of sampling locations
  •  Sampling equipment and methodology
  ซ  Sample deli very requirements.

These elements are discussed in  the following
subsections.  QA/QC procedures for sampling are
addressed in Section 8. Procedures include preparing
COC forms, maintaining sampling  equipment,  and
identifying  the  minimum  volume  requirements,
holding  times,  and preservation  techniques  for
samples.

Sampling Location
Sampling locations  should  be established where
representative samples can be readily obtained. When
sampling waste streams within the POTW, care should
be taken to exclude unwanted waste streams and select
a sampling point that is most representative of the
discharge (e.g., the common discharge channel for
secondary clarifiers). The sampling location for the
POTW  effluent   should  correspond  with  the
biomonitoring sampling  point stated in the NPDES
permit. If the permit does not specify whether the
effluent sample is to be collected prior to or following
the chlorination/dechlorination treatment process, the
choice of a  sampling location  will depend on the
toxicants of concern. Generally, sampling at the point
of final  discharge  is  the  best  option; however,
sampling  both  before  and  after  chlorination/
dechlorination may help to  determine if toxicity is
caused by chlorination (i.e., TRC) or dechlorination.
If samples are collected following the chlorination
process, free chlorine and TRC should be measured
when sampling is completed and upon initiation  of
toxicity tests. These results will provide information
on the potential for chlorine toxicity.

Wastewater sampling for toxicity source evaluations
requires knowledge of sewer discharge locations.
Sampling may be conducted at the point of sewer
discharge or within the sewers in the municipal sewer
collection system. The choice of sampling points for
sewer line  tracking may  be based  on existing
pretreatment program data.  If these data  are not
available, a sampling scheme can be devised to locate
sources of toxicity by testing and eliminating segments
of the collection system that prove to be non-toxic.  In
some cases,  indirect dischargers may have multiple
sewer discharges that  need to be included when
sampling.

RTA testing requires samples of the POTW influent
(primary effluent) and activated sludge.  Primary
effluent samples should be collected at the overflow
weirs of the primary sedimentation tanks. Activated
sludge samples can be collected from the aeration
basin effluent weirs or the RAS pipelines.

POTW Sampling
The choice of grab or composite samples of POTW
waste streams (i.e., effluent and influent wastewater
and process  waste streams)  will  depend on  the
physical/chemical characteristics and variability of the
                                                73

-------
toxicants.  Initial effluent  toxicity characterization
(TEE Phase  I)  should utilize 24-hour composite
samples in order to ascertain  the daily, weekly, or
seasonal variability of the causative agents. If effluent
toxicity  is not  easily  observed  in  24-hour  time
composites, flow proportional composite or  grab
samples may be used to observe possible flow-related
peaks of toxicity. In the latter phases of the TEE, grab
samples may be used to determine the variability in the
type and concentration of effluent toxicants (USEPA
1991a, 1992a). A discussion of the use of grab versus
composite sampling for toxicity tests  is provided by
USEPA (USEPA 1993c, 1994a, 1994b, 1995). The
choice of sampling techniques for chemical-specific
analyses is dependent on the type of compounds to be
measured (e.g.,  grab sampling for volatile organic
compounds).

When  evaluating the treatment efficiency of the
POTW or its unit processes, collection of the influent
and effluent  wastewaters should be  lagged by the
hydraulic retention  time (HRT)  of  the treatment
process in order to obtain comparable samples. For
example, if the HRT of the treatment plant is 20 hours,
the effluent sampler should be timed to start 20 hours
after influent  sampling is  initiated.    Likewise,
sampling of  wastewater from industries or sewers
should account for the travel time in the collection
system (i.e., POTW influent sample collection should
lag industry sample collection).

Samples also should be collected during representative
discharge periods.   An evaluation  of the POTW
operations and performance at the time of sampling
can  be made by comparing  the effluent sample
concentrations of BOD5, TSS, and other pollutants to
long-term historical averages and/or permitted values
for those parameters.

Effluent samples are often collected, shipped, and
stored in plastic containers. However, some toxicants
such as surfactants may adsorb to plastic.  A simple
way to check for this characteristic is to collect and
ship samples in both glass and plastic containers, then
test the samples for toxicity  (USEPA, 1991a).   A
greater  loss  of  toxicity in  plastic  containers  as
compared to  glass containers may  indicate  the
presence of toxic surfactants.

The sample volume requirements for TIE Phase I tests
are provided  by  USEPA  (1991a,  1992a, 1996).
Volume requirements for POTW samples that are used
in RTA tests are given in Section 5.

If TEE or physical/chemical treatability testing is being
conducted off-site, samples should be shipped on ice
to maintain the sample temperature at 4ฐC. RTA and
some biological treatability tests require fresh  or
continuous samples of POTW waste streams, which
requires testing to  be conducted on-site. Samples of
RAS and activated sludge should be delivered to the
on-site laboratory and used immediately in testing to
prevent changes in the biomass that can occur during
long-term  storage. Biomass  samples should  be
vigorously aerated  for a minimum of 15 minutes
before use in the RTA or treatability tests.  POTW
influent and process wastewater samples required for
on-site RTA or treatability studies should be used on
the day of sample collection.

Sewer Discharge Sampling
The choice of grab or composite samples of indirect
discharges will  depend on the physical/chemical
characteristics and variability of the toxicants. The
sample type also will be dictated by the stage of the
toxicity source evaluation.  In Tier I testing, 24-hour
flow  proportional  composite   samples  are
recommended to characterize daily variability while
accounting for variations in flow. Flow proportional
sampling  should be scheduled  to coincide with
production schedules for industrial  discharges,  the
frequency of intermittent inputs for RCRA discharges,
and the schedule of remedial activities for CERCLA
discharges.  This information is usually available in
the POTW's pretreatment program reports.

Sampling techniques'for flow proportional composites
should account  for the potential loss of volatile
compounds.    For  samples collected for chemical
analysis or refractory toxicity testing, zero headspace
sampling methods can be used to minimize volatile
losses. In some cases, grab sampling may be used in
lieu of zero  headspace methods to reduce sampling
costs; however, care should be exercised in  collecting
samples that are representative of the discharge.

In Tier n, grab sampling can be used in addition to
composite sampling; to  assess the variability  of the
toxicants.  This type of sampling requires in-depth
knowledge  of the  production schedules and the
pretreatment operations of the discharger.
                                                74

-------
                                          Section 12
                                         References
Adams, C.E., D.L. Ford, and W.W. Eckenfelder.
    1981. Development of Design and Operational
    Criteria for Wastewater Treatment. Enviro Press,
    Nashville, Tennessee.

Amato,  J.R.,  D.I.  Mount,  EJ.  Durban,  M.L.
    Lukasewycz, G.T.  Ankley, and  E.D.  Robert.
    1992.  An Example of  the Identification of
    Diazinon as a .Primary Toxicant in an Effluent.
    Environ. Toxicol. Chem. 11: 209-16.

American Chemistry Society.   1979.   Safety in
    Academic   Chemistry  Laboratories.    ACS
    Publication Committee on Chemical Safety, Third
    Edition.     American   Chemistry  Society,
    Washington, D.C.

American Petroleum Institute (API).  API. 1998. The
    Toxicity of Common Ions to Freshwater and
    Marine Organisms.  Washington, DC, November
    1998, Doc. No. 0300-029.

American Public Health Association (APHA).  1995.
    Standard Methods for the  Examination of Water
    and Wastewater. 19th ed. American Public Health
    Association, American Waterworks Association,
    Water EnvironmentFederation. Washington, D.C.

American Society for Microbiology (ASM).  1981.
    Manual for Methods for General Bacteriology.
    Washington, D.C.

Ankley, G.T., J.R. Dierkes, D.A. Jensen, and G.S.
    Peterson. 1991. Piperonyl Butoxide as a Tool in
    Aquatic  Toxicological   Research  with
    Organophosphate   Insecticides.    Ecotoxicol.
    Environ. Saf.  21: 266-74.

Ankley, G.T., andL.B. Burkhard. 1992. Identification
    of Surfactants as Toxicants in a Primary Effluent.
    Environ. Toxicol. Chem. 11: 1235-48.

AQUA-Science. 1995. Diazinon and Chlorpyrifos in
    Wastewater from Residential and  Commercial
    Sources. Report for Central Contra Costa Sanitary
    District, Martinez, California.  AQUA-Science,
    Davis, California.

Aquatic Sciences Consulting.  1997. Evaluation of
    Sources of Toxicity and  Toxicity  Treatment
    Options at the City ofHenderson 's Nutbush Creek
    Wastewater Treatment Plant.  Prepared for the
    City ofHenderson, North Carolina.

AQUIRE.   1992.   Aquatic Information  Retrieval
    Toxicity Data  Base.   Office  of Research  and
    Development, National Health and Environmental
    Effects  Research  Laboratory, Mid-Continent
    Ecology Division, Duluth, Minnesota.

Bailey,  H.C.,  J.L.  Miller,  M.J.  Miller,  and  B.S.
    Dhaliwal.   1995.    Application   of Toxicity
    Identification   Procedures  to  the  Echinoderm
    Fertilization Assay to  Identify Toxicity  in  a
    Municipal Effluent.  Environ.  Toxicol.  Chem.
    14(12): 2181-86.

Bailey, H.C., J.L. Miller, M.J. Miller, L.C. Wiborg, L.
    Deanovic, and T. Shed. 1997. Joint Toxicity of
    Diazinon and Chlorpyrifos Under Conditions of
    Acute Exposure to Ceriodaphnia dubia. Environ.
    Toxicol. Chem. 16: 2304-08.

Botts, J.A., J.W.  Braswell, E.G. Sullivan, W.C.
    Goodfellow, B.D. Sklar, and A.G. McDearmon.
    1987.   Toxicity Reduction Evaluation at the
    Patapsco  Wastewater Treatment Plant.  Water
    Engineering Research  Laboratory, Cincinnati,
    Ohio.  Cooperative Agreement No. CR812790-
    01-1. NTIS # PB 88-220 488/AS.
                                                75

-------
 Botts, J.A.,  L.B.  Fillmore, EJ. Durban,  W.A.
    Goodfellow, T. Pereira, and D.R Bishop. 1990.
    Evaluation of the Role of Diazinon in the Toxicity
    of a Municipal Wastewater Treatment Plant
    Effluent.  Proceedings  of the Third National
    Pesticide  Conference,  November,   1990,
    Richmond, Virginia.

 Botts, J.A.,  T.L.  Morris, J.E.  Rumbo, and C.H.
    Victoria-Rueda.     1992.     Case  Histories-
    Municipalities.     In   Toxicity  Reduction:
    Evaluation and  Control.    D.L.  Ford,  ed.
    Technomic   Publishing  Co.,  Lancaster,
    Pennsylvania.

 Botts, J.A., T.L. Morris, M.A. Collins, T.M. Schmitt,
    andE.K. Wilson. 1993. Evaluating the Impact of
    Industrial Discharges   to   Publicly  Owned
    Treatment  Works:  The Refractory Toxicity
    Assessment Protocol. Annual Conference and
    Exposition of the Water Environment Federation,
    Anaheimv California.

 Botts, J.A., T. Schmitt,  E. Wilson, M. Collins, D.
    Waddell,  R.  Diehl,  and L.  Ehrlich.    1994.
    Refractory Toxicity Assessment: An Alternative
    Approach for Chronic Toxicity  Reduction
    Evaluations. Annual Conference and  Exposition
    of the Water Environment Federation, Chicago,
    Illinois. Paper # AC944404.

 Brungs.W.A.  1973. Effects of Residual Chlorine on
    Aquatic Life. Journal Water Pollut. Control Fed.
    45(10): 2180-93.

 Burgess,  R.M, K.T. Ho, M.D. Tagliabue, A. Kuhn, R.
    Comeleo,  P. Comeleo,  G. Modica, and G.E.
    Morrison. 1995. Toxicity Characterization of an
    Industrial and a Municipal Effluent Discharging to
    the Marine Environment. Marine Pollut. Bull. 30:
    524-35.

Burgess,  R.M., J.B. Charles, A. Kuhn, K.T. Ho, L.E.
    Patton,andD.G.McGovern. 1997. Development
    of a  Cation-Exchange Methodology for Marine
    Toxicity Identification Evaluation Applications.
    Environ. Toxicol. Chem.  16: 1203-11.

Burkhard, L.P., and J.J. Jenson.  1993.  Identification
    of Ammonia, Chlorine, and Diazinon as Toxicants
    in a Municipal Effluent. Arch. Environ. Contam.
    Toxicol. 25: 506-15.
 City of Greenville. 1991. TRE Phase B Final Report.
    City of Greenville, Texas.

 Code  of Federal Regulations, Title 40, Part 503.
    1993.  Standards for the  Use of Disposal of
    Sewage Sludge, 58 FR 9387. Office of Federal
    Register,  National  Archives  and  Records
    Administration, Washington, D.C.

 Collins, M.A., T.L. Morris, J.A. Botts, T. Norberg-
    King,  J.  Thompson, and D.I.  Mount.   1991.
    Chronic  Toxicity Reduction Evaluation at the
    Bergen County Utilities Authority  Wastewater
    Treatment Plant, Draft Report.  USEPA, Risk
    Reduction Engineering Laboratory, Cincinnati,
    Ohio.  USEPA Contract No. 68-03-3431.

 Collins, M.A., E.K. Wilson, C.A. LaRocca, and W.L.
    Goodfellow.   , 1994.     Chronic  Toxicity
    Identification  Evaluation   Analyses  Using
    Mysidopsis  bahia.  Annual Conference  and
    Exposition of the Water Environment Federation,
    Chicago,  Illinois.

 Dawson, T.,  K.  Lott, and  M.  Deadrick.   1997.
    Toxicity Identification Evaluation of Washington,
    Georgia,  Wastewater Treatment Plant Effluent.
    USEPA  Contract  No.  68-W3-0034.   Mid-
    continent Ecology Division, Duluth, Minnesota.

 Deeny, K., J. Heiclrnan, and J. Condren.   1988.
    Evaluation  of aj Full-Scale  Activated Sludge
    System Utilizing  Powdered  Activated Carbon
    Addition  with  Wet Air  Oxidation.  Annual
    Conference  of the Water  Pollution  Control
    Federation, Dallas, Texas.

 Diehl,  R., and S. Moore.  1987.  Case History:  A
    North  Carolina:  Municipal  TRE.  Toxicity
    Identification/Reduction  Evaluation  Workshop,
    Annual Conference of the Water Pollution
    Control Federation, Philadelphia, Pennsylvania.

DiGiano,F. 1988. Case History:  Toxicity Reduction
    Evaluation  at High Point, North Carolina.
    Virginia  Water Pollution Control Association
    Wastewater   Toxics  Management  Seminar,
    November 3, 1988, Richmond, Virginia.

Douglas, W.S., S.S. Grasso, D.G. Hutton, and K.R.
    Schroeder.  1996;  Ion Imbalance as a Source of
                                               76

-------
       Toxicity in Estuarine Effluent. Arch. Environ.
       Contam. Toxicol.  31:426-32.

Douglas,  W.S.,  and M.T. Home.   1997.   The
    Interactive Effects of Essential Ions and Salinity
    on the Survival of Mysid shrimp, Mysidopsis
    bahia in 96-hour Acute Toxicity Tests of Marine
    and Estuarine Effluents.  Environ. Toxicol. and
    Chem.  16(10): 1996-2001

Durban, E.J., M.T. Lukasewycz, and J.R. Amato.
    1990. Extraction and Concentration of Nonpolar
    Organic Toxicants from Effluents  Using Solid
    Phase Extraction. Environ. Toxicol. Chem. 9:
    463-66.

Edile, D., S. Newton, N.  Lewnes,  G.N. Cherr, L.
    Schipper, and J. Anderson. 1995. A Marine TIE:
    When Phase I Becomes a Research Project.  16th
    Annual Meeting of the Society of Environmental
    Toxicology and  Chemistry,  Vancouver, B.C.,
    Canada, November 1995.

Engineering-Science, Inc.  1992. Toxicity Reduction
    Evaluation:   Toxicant  Identification  and
    Confirmation. Submitted to the  City of Lawton,
    Oklahoma.

Engineering-Science, Inc. 1994. Quarterly Progress
    Report (July 1 to September 30, 1994) for the
    Rivanna Water  &  Sewer  Authority  Toxicity
    Reduction Evaluation. Prepared for the Rivanna
    Water  &  Sewer  Authority,  Charlottesville,
    Virginia.

EXTOXNET   (Extension Toxicology  Network).
    1998.   Developed by University of California,
    Davis,  Oregon  State  University,  Michigan
    State University, Cornell University,  and the
    University of  Idaho.   On  the  Internet at
    http://ace.orst.edu/info/extoxnet/.

Federal Register.  1989.  USEPA: National Pollutant
    Discharge Elimination System; Surface Water
    Toxics Control Program. USEPA, Volume 54,
    No. 105, Friday, June 2, 1989.

Fillmore, L.B., T.L. Morris,  T.L.  Champlin,  M.C.
    Welch, and J.A. Botts. 1990. Toxicity Reduction
    Evaluation at the City ofFayetteville Cross Creek
    Wastewater Treatment Plant, Draft Report,  1990.
    USEPA, Risk Reduction Engineering Laboratory,
    Cincinnati, Ohio. USEPA Contract No. 68-03-
    3431.

Giger, W., E. Stephanou, and C. Schaffner.  1981.
    Persistent   Organic  Chemicals  in  Sewage
    Effluents: Identification  of  Nonylphenols  and
    Nonylphenolethoxylates by Gas  Capillary  Gas
    Chromatography/Mass   Spectrometry.
    Chemosphere 10(11/12):  1253-63.

Goodfellow, W.L.,  and W.L. McCulloch.  1993.
    Toxicity or Physiological Imbalance?  Inorganic
    Salts TIE Meeting (work group), Houston, Texas,
    November 14, 1993.

Goodfellow, W.L., V.A. Sohn, and Kotulak, M.A.
    1994.  Performance of  a Second  Generation
    Toxicity  Reduction Evaluation.  15th  Annual
    Meeting  of  the   Society   of Environmental
    Toxicology  and Chemistry,  Denver,  Colorado,
    October 29 - November 3, 1994.

Goodfellow, W.L., P.B. Dorn, L.W.  Ausley,  D.T.
    Burton, D.L. Denton, D.R. Grothe, M.A. Heber,
    TJ. Norberg-King, and J.H. Rodgers, Jr. 1998.
    Submitted:   Environmental  Toxicology  and
    Chemistry.  The Role of Inorganic Ion Imbalance
    in Aquatic Toxicity   Testing.    Society  of
    Environmental  Toxicology  and  Chemistry,
    Foundation of Environmental Education, Whole
    Effluent Steering Committee, Pensacola, Florida.

Grady,  C.P.L.     1985.    Biodegradation:     Its
    Measurement   and  Microbiological  Basis.
    Biotechnology and Bio engineering 27: 660—74.

Grady, C.P.L.,  and H.P. Lim.   1980.   Biological
    Wastewater Treatment: Theory and Applications.
    Marcel Dekker, Inc., New York, New York.

Griffin, D.M., K.F. Kline, and L.S. Targgart.  1993.
    Development of Phase I TIE Methods for the
    Giant Kelp, Macrocystis  pyrifera, and  Red
    Abalone,  Haliotis  rufenscens.  13th  Annual
    Meeting of the  Society  of  Environmental
    Toxicology and Chemistry, Cincinnati, Ohio.

Hagelstein K., and R.R. Dauge.  1984. Use of Long
    Solids Retention Times to Improve the Activated
    Sludge  Biodegradation  of  a Complex Waste.
    Proceedings of the  Purdue  Industrial Waste
    Conference, Ann Arbor, Michigan.
                                                77

-------
 Hall, W.S., and R.J. Mirenda. 1991. Acute Toxicity
    of Wastewater Treatment Polymers to Daphnia
    pulex and the Fathead Minnow (Pimephales
    promelas) and  the Effects of  Humic Acid on
    Polymer Toxicity. Res. Journal of Water Pollut.
    Control Fed.  63(6): 895-99.

 Higashi, R.M.,  G.N. Cherr, J.M. Shenker,  J.M.
    MacDonald, and D.G. Crosby.   1992.  A Polar
    High Molecular Mass Constituent of Bleached
    Kraft Mill Effluent is Toxic to Marine Organisms.
    Environ. Sci. and Tech. 26: 2413-20.

 Ho, K.T., K. Mitchell, M. Zappala, and R.M. Burgess.
    1995.  Effects  of Brine Addition  on  Effluent
    Toxicity and  Marine Toxicity Identification
    Evaluation  (TIE)  Manipulations.  Environ.
    Toxicol Chem.  14: 245-49.

 Hockett, J.R., and D.R. Mount. 1996. Use of Metal
    Chelating Agent to Differentiate Among Sources
    of Aquatic Toxicity. Environ. Toxicol. Chem. 15:
    1687-94.

 Jink, A.W., S.M. Bay, D. J. Greenstein, A. Zellers, and
    S-L. Lau.  1998. Applications of TIEs in Studies
    of  Urban  Stormwater Impacts   on  Marine
    Organisms.  In Environmental  Toxicology and
    Risk Assessment: 7th Volume, ASTM STP 1333,
    E.E. Little, A.J. SeLonay, and B.M. Greenberg,
    eds. American Society for Testing and Materials,
    Philadelphia, Pennsylvania.

 Judkins, J.F., and G.A. Anderson.   1992.  Using
    Denitrification for pH Control. Water Environ. &
    Tech. February: 62-65.

 Kang, S.J., L.B. Pugh, and J.A. Borchardt. 1983.
    ATP  as  a  Measure  of  Active  Biomass
    Concentration  and  Inhibition  in  Biological
    Wastewater Treatment Processes. Proceedings of
    the 38fh Purdue Industrial Waste Conference, Ann
    Arbor, Michigan.

Komegay, B.H.   1970.  The Use of  Continuous
    Culture  Theory in the  Selection of Biological
    Reactor Systems. 43rd Annual Conference of the
    Water Pollution Control Federation,  Boston,
    Massachusetts, October.
Lankford, P.W.,  and W.W.  Eckenfelder.   1990.
    Toxicity Reduction in Industrial Effluents. New
    York, New York:  McGraw-Hill.

Lankford, P.W., W.W. Eckenfelder Jr., and K.D.
    Torrens.  1987.;   Technological Approaches to
    Toxicity Reduction in Municipal and Industrial
    Wastewaters.  Annual Meeting  of the Virginia
    Water Pollution  Control Association, Norfolk,
    Virginia.

Linden Roselle Sewerage Authority. 1991. Linden
    Rosette  Sewerage Authority -  1991 Toxicity
    Reduction Evaluation Project Report.  Prepared
    by Engineering-Science, Inc., Fairfax, Virginia, in
    association with EA Engineering, Science,  and
    Technology, Inc., Sparks, Maryland.

MacGregor, R., M.  Tucker,  and J.S. Qi.   1996.
    Application of Ion-balancing Methods for Marine
    Whole Effluent  Toxicity  Tests. IT"1 Annual
    Meeting  of  the  Society of Environmental
    Toxicology and Chemistry, Washington, D.C.

Masters, J.A., M.A. Lewis, D.H. Davidson, and R.D.
    Bruce.  1991.     Validation  of  a  Four-Day
    Ceriodaphnia Toxicity  Test  and  Statistical
    Considerations;  in Data  Analysis.  Environ.
    Toxicol. Chem. 10: 47-55.

Metcalf  and  Eddy,   Inc.    1991.    Wastewater
    Engineering  Treatment, Disposal,  and Reuse.
    Third Edition. New York, New York: McGraw-
    Hill.

Mickley,  M., J. Briceno, B. Borkman, K.  Fucik, J.
    Case, and T. Nickel.   1996.   Common  Ion
    Toxicity in R.O.  Membrane  Concentrate. 17th
    Annual Meeting of the Society of Environmental
    Toxicology and Chemistry, Washington, D.C.

Morris, T.L., J.A. Botts, J.W. Braswell, M.C. Welch,
    and W.L. Goodfellow. 1990. Toxicity Reduction
    Evaluation  at ;the Linden Roselle Sewerage
    Authority Wastewater Treatment Plant,  Draft
    Report, 1990.  tJSEPA, Office of Research and
    Development,   Risk  Reduction  Engineering
    Laboratory, Cincinnati, Ohio.  USEPA Contract
    No. 68-03-3431.
                                               78

-------
Morris, T.L.,G. Fare, and J.Spadone. 1992. Toxicity
    Reduction Evaluation at the Linden  Roselle
    Sewerage Authority Wastewater Treatment Plant.
    Water Environ. Tech. June: 8-16.

Mosure,  T.E.,  J.P.  Pierko,  J.M.  Schmidt, R.A.
    Monteith,  and J.S. Mosser.   1987.   Toxicity
    Reduction  Evaluation  at the  Akron Water
    Pollution Control Station, Draft Report. USEPA,
    Office of  Research  and  Development, Risk
    Reduction Engineering Laboratory, Cincinnati,
    Ohio.     Cooperative  Agreement  No.
    #CR813681010.

Mount, D.R., and D.D. Gulley. 1992.  Development
    of a  Salinity/Toxicity Relationship to  Predict
    Acute Toxicity of Saline Waters to Freshwater
    Organisms. Gas Research Institute, Environment
    and Safety Research  Department, Chicago, IL,
    October, 1992.

National Institute of Occupational Safety and Health
    (NIOSH).  1977. Working With Carcinogens.
    Public  Health  Service,  Centers  for  Disease
    Control. Publication No. 77-206.

Norberg-King, T.J., M. Lukasewcyz, and J. Jenson.
    1989.  Results of Diazinon  Levels in POTW
    Effluents in the United States.  National Effluent
    Toxicity Assessment Center, Duluth, Minnesota.
    Technical Report 14-89.

Novartis Crop Protection, Inc., andMakhteshim-Agan
    of North America, Inc.  1997.  Investigation of
    Diazinon Occurrence, Toxicity, and Treatability
    in  Southern  United States  Publicly  Owned
    Treatment  Works.    Technical  Report  3-7,
    Environmental Affairs Department, Greensboro,
    North Carolina.

 Occupational  Safety and Health  Administration
    (OSHA).    1976. OSHA Safety  and Health
    Standards,  General  Industry. 29  CFR 1910.
    OSHA 2206 (Revised).

 Oris, J.T., R.W. Winner, and M.V. Moore.  1991. A
    Four-day Survival and Reproduction Toxicity Test
    forCeriodaphniadubia. Environ. Toxicol. Chem.
     10: 217-24.
Philbrook,D.M.,andC.P.L.Grady. 1985. Evaluation
    of Biodegradation Kinetics for Priority Pollutants.
    Proceedings of the Purdue Industrial  Waste
    Conference, Ann Arbor, Michigan.

Rodgers, J.H., Jr.   1989a.  Species tolerances for
    NPDES  bioassays:   Volume   I,   freshwater
    organisms.  American Petroleum Institute.  No.
    4483. Washington, D.C.

Rodgers, J.H., Jr.   1989b.  Species tolerances for
    NPDES bioassays: Volume II, marine organisms.
    American  Petroleum  Institute.   No.  4482.
    Washington, D.C.

Singhasemanon, N., C. Nordmark, andT. Barry. 1997.
    Diazinon and Chlorpyrifos in  Central Contra
    Costa Sanitary District Sewer System, Summer
    1996.  State  of  California   Environmental
    Protection  Agency, Department of  Pesticide
    Regulation.   Sacramento,  California.   Draft
    Report.

Slattery, G.H.   1987.  Effects of Toxic Influent on
    Patapsco Wastewater Treatment Plant Operations.
    Annual Conference  of the  Water Pollution
    Control Federation, New Orleans, Louisiana.

Sullivan, E.G., D.F. Bishop, J.A. Botts, J.W. Braswell,
    G.H.  Slattery, and W.L. Goodfellow.   1987.
    Effluent  Toxicity  Monitoring  Methodology
    Evaluated  for   Five   Industrial  Dischargers.
    Proceedings  of the  Purdue Industrial Waste
    Conference, Ann Arbor, Michigan.

TRAC Laboratories, Inc.   1992.  Unpublished data.
    Prepared by TRAC Laboratories, Inc., Denton,
    Texas.

USEPA.   1977.  Occupational Health and Safety
    Manual. Office of Planning and Management,
    Washington, D.C.

USEPA.  1979a. Handbook for Analytical Quality
    Control in Water and Wastewater Laboratories.
    EPA/600/4-79/019. Analytical Quality Control
    Laboratory, Cincinnati, Ohio.

USEPA.   1979b.  Methods 624 and 625: GC/MS
    Methods for Priority Pollutants. Federal Register
    44(223):69532-58.
                                                79

-------
 USEPA. 1982. Handbook for Sampling and Sample
    Preservation  of  Water   and  Wastewater.
    EPA/600/4-82-019. Cincinnati, Ohio.

 USEPA. 1983a. Guidance Manual for Pretreatment
    Program   Development.  Office   of  Water
    Enforcement and Permits, Washington, D.C.

 USEPA. 1983b. Methods for Chemical Analysis of
    Water  and  Wastes.  EPA/600/4-79-020,
    Cincinnati, Ohio.

 USEPA.  1984a.  Handbook on Improving POTW
    Performance Using the  Composite Correction
    Program Approach. EPA/625/6-84/008. Center
    for  Environmental   Research  Information,
    Cincinnati, Ohio.

 USEPA. 1984b. Ambient Water Quality Criteria for
    Chlorine -1984. Office of Water Regulations and
    Standards,  Criteria and Standards  Division,
    Washington, D.C.

 USEPA. 1985a. Ambient Water Quality Criteria for
    Ammonia - 1985.  Office of Water Regulations
    and Standards, Criteria and Standards Division,
    Washington, D.C.

 USEPA.  1985b.  Master Analytical  Scheme for
    Organic Compounds  in  Water.  EPA/600/4-85-
    008.  Office of Research  and Development,
    Cincinnati, Ohio.

 USEPA. 1987a.  Permit Writer's Guide to  Water
    Quality-Based Permitting for Toxic Pollutants.
    Office  of  Water Enforcement  and  Permits,
    Washington, D.C.

 USEPA.   1987b.    Guidance  Manual  on  the
    Development  and  Implementation of  Local
    Discharge Limitations Under the Pretreatment
    Program.   Office of Water Enforcement and
    Permits, Washington, D.C.

 USEPA.  1988a.  Methods  for Aquatic  Toxicity
    Identification Evaluations:   Phase I  Toxicity
    Characterization Procedures. EPA/600/3-88/034.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.  1988b.   Development of Slug  Loading
    Control Programs for Publicly Owned Treatment
    Works.   Office; of  Water Enforcement  and
    Permits, Washington, D.C.

 USEPA.   1989a.   Toxicity  Reduction Evaluation
    Protocol for Municipal Wastewater Treatment
    Plants. EPA/600/2-88/062.  Office of Research
    and Development, Risk Reduction Engineering
    Laboratory, Cincinnati, Ohio.

 USEPA.   1989b.   Generalized  Methodology for
    Conducting  Industrial  Toxicity   Reduction
    Evaluations.   EPA/600/2-88/070.     Water
    Engineering Research Laboratory, Cincinnati,
    Ohio.

 USEPA.  1989c.  Handbook: Retrofitting POTWs.
    EPA/625/6-89/020.  Center for Environmental
    Research Information, Cincinnati, Ohio.

 USEPA. 1990.  User Documentation: POTW Expert,
    Version 1.0.  EPA/625/11-90/001.  Office  of
    Research and Development,  Cincinnati, Ohio.

 USEPA.   1991a.   Methods for  Aquatic Toxicity
    Identification  Evaluations:   Phase I, Toxicity
    Characterization Procedures.   Second Edition.
    EPA/600/6-91-003. National Effluent Toxicity
    Assessment Center, Duluth,  Minnesota.

 USEPA.  1991b.  Technical Support Document for
    Water Quality-Based Toxics Control. EPA/505-2-
    90-001.  Office  ;of  Water Enforcement and
    Permits, Washington, D.C.

 USEPA.   1991c.  Guidance Manual for Control of
    Slug Loadings to 'POTWs. 21W-4001.  Office of
    Water Enforcement and  Permits, Washington,
    D.C.

 USEPA.   1991d.   Methods for Determination  of
    Metals in Environmental Samples.  EPA 600/4-
    91/010.  Office  of Research and Development,
    Washington, D.C.;

 USEPA.  1992a. Toxicity Identification Evaluations:
    Characterization of Chronically Toxic Effluents,
    Phase I. EPA/600/6-91-005F. National Effluent
    Toxicity Assessment Center, Duluth, Minnesota.

USEPA.  1993a.  Methods for Aquatic Toxicity
    Identification Evaluations:  Phase II Toxicity
    Identification Procedures for Samples Exhibiting
                                               80

-------
       Acute and Chronic Toxicity. EPA/600/R-92-
       080. National Effluent Toxicity Assessment
       Center, Duluth, Minnesota.

USEPA.   1993b.  Methods for Aquatic Toxicity
    Identification Evaluations.   Phase HI Toxicity
    Confirmation Procedures for Samples Exhibiting
    Acute and Chronic Toxicity. EPA/600/R-92-081.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.  1993c.  Methods for Measuring the Acute
    Toxicity of Effluents and Receiving Waters to
    Freshwater and  Marine  Organisms.   Fourth
    Edition.  EPA/600/4-90-027F.  Environmental
    Monitoring Systems Laboratory, Cincinnati, Ohio.

USEPA.  1993d.  Treatability Data Base, Version 5.
    EPA/600/C-93/003a.     Risk  Reduction
    Engineering Laboratory, Cincinnati, Ohio.

USEPA.  1994a. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to Freshwater Organisms. Third Edition.
    EPA/600/4-91/002. Environmental Monitoring
    and Support Laboratory, Cincinnati, Ohio.

USEPA.  1994b. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to Marine  and Estuarine Organisms.
    Second  Edition.  EPA/600/4-91/003.
    Environmental  Monitoring  and   Support
    Laboratory, Cincinnati, Ohio.

USEPA.   1994c.  Methods for Determination of
    Metals in Environment Samples — Supplement I.
    EPA 600/R-94/111.  Office of Research  and
    Development, Washington, D.C.

USEPA.  1995. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to West  Coast Marine and  Estuarine
    Organisms.   EPA/600/R-95-136.   National
    Exposure Research Laboratory, Cincinnati, Ohio.

USEPA.    1996.   Marine Toxicity  Identification
    Evaluation (TIE) Guidance Document, Phase I.
    EPA/600/R-96/054.   Office of Research  and
    Development, Washington, D.C.

USEPA.  1997. Methods and Guidance for Analysis
    of Water CD ROM Version 1.0. Office of Water,
    EPA-821-97-001. Washington, D.C.
Walsh, G.E., and R.L. Garnas. 1983. Determination
    of Bioactivity of Chemical Fractions of Liquid
    Wastes Using Freshwater and Saltwater Algae and
    Crustaceans. Environ. Sci. Tech. 17: 180-82.

Ward, S.H. 1989. The Requirements for a Balanced
    Medium in  Toxicological Experiments  using
    Mysidopsis bahia with Special  Reference  to
    Calcium Carbonate. Environmental Toxiocology
    and Hazard Assessment: 12th Volume, ASTMSTP
    No. 1023:  402-12.   U.M. Cowgill and L.R.
    Williams, eds. American Society for Testing and
    Materials, Philadelphia, Pennsylvania. 402-12.

Water Environmental Federation.  1996. Wastewater
    Sampling for Process and Quality Control - MOP
    OM-1.    Water   Environment  Federation,
    Alexandria, Virginia.

Water Environment Federation and American Society
    of Civil Engineers (WEF/ASCE). 1992a. Design
    of Municipal  Wastewater  Treatment Plants,
    Volume  I,   Chapters  1-13.   WEF Manual of
    Practice No. 8 andASCE Manual and Report on
    Engineering Practice No. 76. Water Environment
    Federation, Alexandria, Virginia, and American
    Society of Civil Engineers, New York, New York.

Water Environment Federation and American Society
    of Civil Engineers (WEF/ASCE). 1992b. Design
    of Municipal  Wastewater  Treatment Plants,
    Volume  II,  Chapters 13-20.  WEF Manual of
    Practice No. 8 and ASCE Manual and Report on
    Engineering Practice No. 76. Water Environment
    Federation, Alexandria, Virginia and American
    Society of Civil Engineers, New York, New York.

Water Pollution Control Federation (WPCF).  1983.
    Nutrient Control.   Manual of Practice  FD-7,
    Facilities Design, Alexandria, Virginia.

Water Research Commission (WRC). 1984. Theory,
    Design  and Operation of Nutrient Removal
    Activated Sludge Processes. Wiechers HNS, ed.
    Water Research Commission, Pretoria,  South
    Africa.

Weis, P., J.S. Weis, C-M Chen, and A. Greenberg.
    1992. Treated Municipal Wastewaters: Effects of
    Organic Fractions on Development and Growth of
    Fishes. Environ. Toxicol. Chem. 11: 1451-59.
                                               81

-------
                                          Section 13
                                        Bibliography
Ahel,M.,W. Giger, and M. Koch.  1985. Behavior of
    Nonionic Surfactants in Biological Wastewater
    Treatment.  In Organic Micropollutants in  the
    Aquatic Environment. Proceedings  of the  4th
    European Symposium, Vienna, Austria.

Ahel, M., and W. Giger.  1985.  Determination of
    Alkylphenols and  Alkylphenol  Mono-  and
    Diethoxylate in Environmental Samples by High-
    Performance Liquid Chromatography. Analytical
    Chem. 57(8): 1577-83.

Ahel, M., and W. Giger 1985.  Determination of
    Nonionic   Surfactants  of   the  Alkylphenol
    Polyethoxylate Type by High-Perf ormance Liquid
    Chromatography.   Analytical  Chem.  57(13):
    2584-90.

Ayres, G.H., 1970. Quantitative Chemical Analysis.
    New York: Harper & Bow.

Code of Federal Regulations,  Title 40, Part 503.
    1993.  Standards for the  Use of Disposal of
    Sewage Sludge, 58 FR 9387.  Office of Federal
    Register,  National   Archives  and  Records
    Administration, Washington, D.C.

DeRenzo, DJ. 1981.  Pollution Control Technology
    for  Industrial  Wastewater.     Noyes  Data
    Corporation, New Jersey.

Eckenfelder, W.W. 1980.  Principles of Water Quality
    Management.  Boston,  MA:   CBI  Publishing
    Company.

Federal Register.  1984.  USEPA.  Development .of
    Water Quality-B ased Permit Limitations for Toxic
    Pollutants; National  Policy, Vol. 49, No. 48,
    March 9,1984.
Giger, W., H. Brunn;er, and C. Schaffner. 1984. 4-
    Nonxylphenol in Sewage Sludge: Accumulation
    of Toxic Metabolites from Nonionic Surfactants.
    Science 225: 623-25.

Goodfellow, W.L., and W.L. McCulloch, 1987. A
    Technique for the Rapid Evaluation of Effluent
    Acute Toxicity. In Proceedings of the 8th Annual
    SETAC Meeting, ;Pensacola, Florida.

Jirka, A.M., and M.J. Carter.  1975. HACH COD
    Procedure  for  the Bausch & Lomb  Spec 20.
    Analytical Chem: 47(8).

Marcomini, A., and W. Giger, 1987. Simultaneous
    Determination of Linear Alkylbenzenesulfonates,
    Alkylphenol Polyethoxylates and Nonylphenol by
    High-Performance   Liquid  Chromatography.
    Analytical Chem. 59(13):  1709-15.

McEvoy, J., and W. Giger, 1986.  Determination of
    Linear Alkyl-benzenesulfonates in Sewage Sludge
    by High-Resolution Gas Chromatography/Mass
    Spectrometry.  Environ. Sci. Tech. 20(4): 376-83.

Patterson, J.W.,  1985.   Industrial  Wastewater
    Treatment Technology. 2nd Edition. Boston, MA:
    Butterworth.

Peters, R.W., and Y.| Ku, 1984.  Removal of Heavy
    Metals from Industrial Plating Wastewaters by
    Sulfide  Precipitation.  Proceedings  of  the
    Industrial  Wastes  Symposium  (57th Annual
    Conference).

Techline   Instruments,  Inc.    1984.    Techline
    Laboratory  Respirometer Operating Manual,
    Fond du Lac, Wisconsin.
                                               82

-------
USEPA.  1983. Treatability ManwaZ. EPA/600/2-82-
    00 la. Office  of Research and  Development,
    Washington, D.C.

USEPA.   1985b.  Protocol Development for the
    Prediction of the  Fate of  Organic Priority
    Pollutants in Biological Wastewater Treatment
    Systems. Risk Reduction Engineering Laboratory,
    Cincinnati, Ohio.

USEPA.  1988. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to Marine and Estuarine Organisms.
    EPA/600/4-87-028. Environmental Monitoring
    and Support Laboratory, Cincinnati, Ohio.
USEPA.  1989.  Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to  Freshwater Organisms.    Second
    Edition.  EPA/600/4-89/001.   Environmental
    Monitoring Systems Laboratory, Cincinnati, Ohio.

Water  Pollution   Control   Federation,   1982.
    Wastewater Treatment Plant Design, Manual of
    Practice No. 8. Alexandria, Virginia.

Water Pollution Control Federation (WPCF). 1983.
    Nutrient Control.   Manual  of Practice FD-7,
    Facilities Design.  Alexandria, Virginia.

Water Environment Federation, 1990. Operation of
    Wastewater Treatment Plants.   A  Manual of
    Practice.  Alexandria, Virginia.
                                                 83

-------
                                          Appendix A
                      Original Case Histories:  Lessons Learned
 Since the USEPA research studies in the 1980s and the
 first TREs performed to meet permit requirements,
 there  have  been  significant  advances  in  the
 development and  refinement  of TRE procedures.
 These advancements become apparent upon review of
 the original case histories published in the first edition
 of the TRE manual.  The case histories have been
 revisited in this manual to note the lessons learned and
 new approaches that can be taken to conduct TREs.

 Many lessons have been learned in applying Tffi/TRE
 procedures to different types  of effluents using  a
 variety  of freshwater and  saltwater test species.
 Perhaps  the most  significant improvements  in  the
 methods  since  the original  case  histories were
 performed have been the development and application
 of methods to:

  • Identify causes of short-term chronic toxicity to
    both freshwater and estuarine/marine species.
  • Track sources of chronic toxicity that can not be
    readily characterized in the TIE.
  • Characterize, identify, and confirm organophos-
    phate insecticide toxicity.
  • Characterize toxic metals using improved  EDTA
    and sodium thiosulfate tests.
  • Characterize surfactant toxicity using multiple TIE
    manipulations.
  • Confirm toxicants by the correlation approach.

The use of some of these updated methods is described
below using the original case histories as examples.
The following summaries are intended to show how
similar TREs can be performed more quickly, cost-
effectively,  and  accurately  using  the  current
procedures. These summaries also portray the steps
taken over the  last 10 years to  improve the TRE
procedures.
 Baltimore, Maryland
 In January 1986, USEPA, in cooperation with the City
 of Baltimore, began the first research study to develop
 a pragmatic  approach  and methods for conducting
 TREs at WWTPs {Bolts et al., 1987).  The City's
 Patapsco WWTP was selected for this study because of
 evidence of acute and  chronic effluent toxicity.  In
 addition, USEPA wfas interested in conducting a TRE
 at an urban WWTP, like the Patapsco WWTP, which
 receives its influent from a wide range of industrial
 discharges.   The  objectives of  the  TRE were  to
 characterize the WWTP's capability for treatment of
 toxicity, evaluate techniques to identify the specific
 components of toxicity, and assess methods to trace
 toxicity to its source(s).

 The study results showed that the WWTP influent had
 significant acute and chronic toxicity as measured by
 C. dubia [(mean 48-hour LC50=2.6% and mean 7-day
 chronic value (ChV)=1.2%], M. bahia (mean 96-hour
 LC50=23%), and Microtoxฎ (EC50=8%).  Although
 significant   toxicity  reduction   occurred  through
 treatment, substantial toxicity remained. The 48-hour
 LC50 for C.  dubia  averaged 6.3%  effluent.  An
 evaluation of the WWTP  operations  indicated that
 treatment performance  was not the major cause of
 effluent toxicity.

 Results of the TIE showed that acute effluent toxicity
 was removed by passing effluent samples through a
 C18 SPE column.  Recovery of toxicity in the 75 to
 95%  methanol/water eluates from the C18 column
 suggested that the toxicants were non-polar organic
 compounds  with relatively  high octanol-to-water
partition coefficients. However, GC/MS analysis of
the toxic non-polar  organic  fractions  was  not
successful in identifying the specific nonpolar organic
                                                84

-------
toxicants. Additional testing showed that the toxicants
sorbed onto suspended solids in the effluent.  Solids
greater than 0.2 |im were found to be the major toxic
fraction.
             TIE Procedure Update

 Since this study, USEPA developed procedures for
 identifying non-polar organic toxicants (1993a).  If
 non-polar organic toxicity is indicated in the Phase I of
 the  TIE,  the  toxicant(s)  can  be isolated   and
 concentrated to improve the chances of identification
 using GC/MS analysis.   This approach has been
 helpful in identifying organophosphate insecticides as
 causes of  effluent  toxicity at some POTWs  (see
 examples below and Appendix F).
 An evaluation of wastewa.ter samples from selected
 candidate industries was performed to determine the
 major contributors of refractory toxicity to the WWTP.
 An important goal  of this study was to develop and
 evaluate methods for tracking sources of toxicity in
 POTWs.   A protocol  was designed to measure the
 toxicity remaining after treatment at the WWTP, which
 is the toxicity that passes through in the final effluent.
 The  residual  or "refractory" toxicity of five major
 industrial  users of the WWTP was evaluated by
 treating wastewater samples in a bench-scale batch
 simulation of the WWTP activated sludge process.
 Microtoxฎ results  indicated that two of  the  five
 industries were contributing refractory toxicity to the
 WWTP. Results of C. dubia tests were inconclusive
 due to an interference in the treatment simulation. This
 interference appeared to be caused by residual toxicity
 in the RAS used in testing.
             RTA Procedure Update

  Biomass toxicity may be reduced by washing the
  RAS  with  buffer  solutions or laboratory  water.
  Alternatively, a surrogate biomass from a POTW with
  a similar type  of biological treatment process may
  be obtained for  testing.  Details are presented  in
  Section 5.
 Hollywood, California
 In the  late  1980s ;and  early  1990s, the USEPA
 laboratory in Duluth, Minnesota, tested several POTW
 effluents in the process of developing TIE procedures.
 One of these effluents was the City  of Hollywood
POTW, which exhibited acute toxicity to C. dubia
(Amato et al., 1992).

TIE Phase I tests showed that treatment with a CIS
SPE column was the only step that reduced effluent
toxicity.  Acute toxicity was recovered from the CIS
column  by eluting the column  with methanol.
Additional CIS SPE column tests performed on  16
effluent samples showed that toxicity was consistently
eluted in the 80 and 85% methanol fractions, which
suggested that the  cause of toxicity  was the same
among the various  samples.  These results  provided
evidence that the toxicant(s) was a non-polar organic
compound(s).  Further concentration and separation of
the toxic fractions was done, followed by confirmation
GC/MS analyses of the fractions. Analysis of selected
80 and 85%  methanol fractions by  GC/MS found
sufficient concentrations of the insecticide diazinon to
account for the observed acute toxicity to C. dubia.
             TIE Procedure Update

  In recent TIE guidance, USEPA (1991 and  1993a)
  recommends  adding a metabolic blocker, PBO, to
  toxic effluent samples  or  methanol eluates as  a
  subsequent test  for the presence of metabolically
  activated toxicants like organophosphate insecticides.
  PBO has been shown to block the acute  toxicity of
  diazinon, parathion, methyl parathion, and malathion
  to cladocerans, but does not affect acute sensitivity to
  dichlorvos, chlorfenvinphos, and mevinphos (Ankley
  et al., 1991).  A reduction in acute or chronic toxicity
  by the PBO addition together with toxicity removal by
  the  CIS SPE column  and  concentration data  can
  provide strong evidence for the presence of selected
  organophosphate insecticides.
 In the confirmation step (USEPA, 1989b), three Phase
 in confirmation steps were used to confirm diazinon as
 a cause of effluent toxicity: toxicant correlation, mass
 balance, and additional species testing.

 Toxicant correlation was evaluated by plotting effluent
 diazinon concentrations and effluent LC50 values as
 shown in Figure A-l. The correlation coefficient (r
 value) was significant and confirmed that, from sample
 to sample, diazinon was consistently the cause of acute
 effluent toxicity. Also, the intercept of the regression
 line at 100% effluent (0.325) was near the diazinon
 LC50 of 0.35 ng/L, which indicated that diazinon
 accounted for nearly all of the observed acute effluent
 toxicity.
                                                   85

-------
     10
   I  i:
    10-'
       10
                       LC50 (%)
                                              102
 Figure A-l.  Acute LC50  of Hollywood effluent versus
 diazinon concentration (actual correlation shown by solid
 line; predicted 1:1  correlation by dashed line)  (Source:
 USEPA, 1988).
              TEE Procedure Update

  USEPA  (1993b)  recommends  a straight-forward
  correlation  approach to determine if a  consistent
  relationship exists between the concentration of the
  toxicant(s)  and effluent toxicity.   This approach
  involves comparing the toxic units of the toxicant to
  whole effluent toxic units. Toxicant concentrations are
  converted to toxic units (i.e., measured concentration
  divided by the toxicant's acute or chronic endpoint)
  and the resulting values are plotted versus whole
  effluent toxic units.  Since this study, additional acute
  toxicity data for diazinon and other organophosphate
  insecticides have become available for calculating
  toxic units for  these toxicants (Ankley et al., 1991;
  Amato  et al.,  1992; and Bailey et al., 1997).  The
  correlation approach is useful for determining the
  extent to which the identified toxicants contribute to
  effluent toxicity. Using the above example, diazinon
  would be confirmed as the primary toxicant if the slope
  is 1 and the intercept is 0 for a plot of diazinon toxic
  units versus effluent toxic units.  In some cases,
  additional  toxicants may be indicated  using  this
  technique (see  the City of Largo, Florida, example
  below).
The mass balance  confirmation approach involved
passing samples through a CIS SPE column, eluting
the  column  with   a  series  of  eight  methanol
 concentrations, and testing the toxicity of the methanol
 fractions.  The combined toxic, combined nontoxic,
 and all fractions were combined and tested at whole
 effluent concentrations.  The results showed that the
 toxicity of the combined toxic fractions was similar to
 the toxicity of all fractions together and the toxicity of
 the original effluent samples. These results provided
 further  confirmation  that  effluent  toxicity  was
 associated with non-polar organic toxicants.

 The final confirmation step involved testing effluent
 samples with P. promelas, which are at least 100 times
 less  acutely  sensitive to diazinon than  C. dubia
 (USEPA 1987,1988). Test results showed only slight
 acute toxicity to the  minnows as compared to  the
 average acute LC50 of about 60% for C. dubia. Acute
 toxicity to P. promelas was interpreted as evidence that
 a toxicant other than diazinon was  present in  the
 samples. However, this additional toxicant(s) was not
 a significant contributor to toxicity and its identity was
 not evaluated.  In! summary,  the Phase  HI testing
 confirmed that diazinon was the principal effluent
 toxicant.

 Largo, Florida
 The USEPA Duluth Laboratory also evaluated effluent
 samples from the City of Largo POTW.  A TIE was
 performed to identify the causes of acute effluent
 toxicity (USEPA, 1987).

 TIE Phase  I  tests showed  that CIS SPE column
 treatment removed acute effluent toxicity. Toxicity
 was not reduced by the other Phase I treatments,
 including  filtration,  EDTA  addition, or  sodium
 thiosulfate addition.

 An additional 18 effluent samples were passed through
 CIS SPE columns in Phase H Elution of the columns
 with  methanol showed  that  acute  toxicity was
 consistently isolated in the 75 and 80% methanol
 concentrations, although occasional toxicity was also
 observed in the 70 and 85% methanol concentrations.
 GC/MS  analysis of  the toxic fractions  identified
 diazinon as a cause of acute effluent toxicity.

In Phase HI, five confirmation steps were used to verify
that diazinon was: the cause  of  effluent toxicity:
toxicant correlation;  toxicant spiking, mass balance,
additional species testing, and test species symptoms.

Acute effluent toxicity and diazinon concentrations
were converted to TUs and were plotted to determine
                                                  86

-------
the toxicant correlation to whole effluent  toxicity
(USEPA, 1989b). As shown in Figure A-2, more acute
toxicity  was present than would be explained by
diazinon alone; the slope of the linear regression was
less than 1 and all of the plotted data points are below
the expected 1:1 relationship for diazinon and effluent
toxicity.   Spiking experiments also  showed that
doubling the  concentration of diazinon  in  effluent
samples did not result in a doubling of effluent toxicity.
These results suggested that diazinon was not the sole
cause of acute effluent toxicity.
    5.00
     4.00 -
  I
          — — — Expected Regression
                Observed Regression
     3.00 -
     2.00 -
     1.00-
     0.00
        0.00     1.00     2.00    3.00     4.00     5.00
                   Whole Effluent TUs
 Figure  A-2.  Correlation of diazinon TUs versus whole
 effluent TUs (Source: USEPA, 1988).
              TIE Procedure Update

  The current approach  (USEPA, 1993b) is to plot
  effluent TUs on the Y-axis (dependent variable) and
  toxicant  TUs on the x-axis (independent variable).
  See Figure A-3.
 Follow-up GC/MS analyses identified chlorfenvinphos
 (CVP) and malathion in effluent samples. Malathion
 did not appear in concentrations high enough to cause
 acute  toxicity  to  C.   dubia,  although   CVP
 concentrations were sufficient to contribute to effluent
 toxicity  (48-hour LCSOs  of  1.4  and  0.35 ug/L,
 respectively,  according  to  D.   Mount,  personal
 communication, USEPA, Duluth, Minnesota, 1989).

 The correlation  analysis  was  repeated using  the
 summed toxic units for both diazinon and CVP versus
whole effluent toxic units (USEPA, 1993b). As shown
in Figure A-3, the slope of the regression line was
close to 1, the y-intercept was nearly zero, and the
r-value indicated a good correlation (r = 0.73). These
results show that  diazinon and CVP accounted for
nearly all of the acute effluent toxicity.
                                                          5-
           t    Observed
        — — —  Theoretical
                                 = 0.73
                                slope = 0.82
                                y-intercept = 0.46
       0123456
                TUs of Suspect Toxicant(s)
 Figure A-3.  Correlation of diazinon and CVP TUs versus
 whole effluent TUs (Source: USEPA, 1993b).


 Additional confirmation testing involved analyzing 13
 effluent samples using  the CIS  SPE  column mass
 balance approach. As shown in Table A-l, in 12 of the
 13  tests,  the  toxicity  of all  methanol  fractions
 combined was slightly greater than the toxic fractions
 combined.  Various  mixtures of the three  identified
 insecticides were tested to determine if interactive
 effects (i.e., antagonistic or synergistic) could account
 for the difference in toxicity.  These tests  showed that
 the toxicity of the insecticides was strictly additive.
 These results indicated that the higher toxicity of "all
 fractions"  compared to  the toxicity  of the  "toxic
 fractions" may be due to another unidentified toxicant,
 rather than interaction among the identified toxicants.

 The additional toxicity observed in the "all fraction"
 test  was attributed to 70% methanol/water fraction,
 which exhibited slight and intermittent toxicity. This
 fraction  was  initially  included  in  the  "nontoxic
 fraction" test;  however,  the mass balance approach
                                                    87

-------
      Table A-l. Comparison of Whole Effluent TUs
      and Methanol Fraction TUs
Sample
A
B
C
D*
E
F*
G
H
I
J
K
L
Main
Acute Toxic Units (TUa)
Whole
Effluent
1.18
2.00
1.93
<1.00
2.00
1.15
1.33
3.70
2.86
2.27
2.27
2.27
2.13
All-
Fractions
1.64
2.94
2.86
1.15
1.75
1.06
1.52
3.03
2.86
1.72
2.04
1.67
2.18
Toxic-
Fractions
1.43
3.13
2.53
<1.00
1.64
<1.00
1.13
2.86
2.44
1.64
2.00
1.59
2.00
 * Values excluded from mean calculations due to less-than values.

 indicated it to be a slightly toxic fraction.  When the
 toxic units of the 70% fraction are added to the "toxic
 fraction" result, nearly all of the toxicity is accounted
 for. Due to the intermittent toxicity of this fraction,
 additional  testing to  identify the toxicant was  not
 performed.

 Additional species testing with P. promelas provided
 further   evidence   that   the   toxicants   werp
 organophosphate insecticides. No acute toxicity was
 observed with P. promelas, which are known to be
 orders of magnitude less sensitive to diazinon than
 C. dubia (USEPA 1987,1988).

 As a final confirmation step, the  same symptoms to
 C.  dubia were observed after exposure to effluent
 samples, toxic methanol fractions, and laboratory water
 spiked with Bear lethal levels of diazinon, CVP, and
 malathion. Similar symptoms were observed for all
 test solutions, which suggested that the same toxicant
 was responsible in each case.

 Lawton, Oklahoma
The City of Lawton was required by USEPA Region 6
to initiate a TRE study in 1991, based on evidence of
chronic effluent toxicity at its POTW (Engineering
Science, Inc., 1991). The permit limit of no chronic
lethality  at  the  critical  instream dilution  of 96%
(i.e., NOEC >96% effluent) was exceeded. Toxicity
 test results showed that the effluent was toxic to C.
 dubia, but not P. promelas.

 TIE  Phase  I  tests  were  conducted in  1991  to
 characterize  the  chronic   effluent   toxicants
 (Engineering Science, Inc.,  1991).  The permit limit
 was based on lethality to C. dubia and P. promelas in
 chronic toxicity tests; therefore, the TEE tests focused
 on lethality instead of reproduction or growth effects.
 The Phase I tests evaluated percent survival of C.
 dubia, the most sensitive organism, over 5 to 7 days in
 100% effluent. In addition, acute lethality results (48-
 to 72-hour exposure) also were collected to assist in the
 evaluation.

 The results indicated a consistent reduction in effluent
 toxicity  by passing \ samples through the C18 SPE
 column.  Chronic lethality data showed that no other
 treatment consistently removed toxicity. Toxicity was
 recovered by eluting the C18 SPE column with
 methanol, which indicated the presence of nonpolar
 organic toxicants. Sample adjustment to pH 3 and pH
 11 also reduced toxicity in all but two samples, which
 suggested that the toxicants could be denatured under
 acidic or basic conditions.
             TIE Procedure Update

 As noted above, PBO, a metabolic blocker, can be
 added to toxic effluent samples, CIS SPE fractions, or
 HPLC fractions  to test for the presence of meta-
 bolically activated toxicants such as organophosphate
 insecticides.       j

 Reproduction data for C. dubia, although not required
 as part of compliance testing for the Lawton POTW,
 may have been useful in characterizing the effluent
 toxicants.  These data may provide a more sensitive
 endpoint than  survival  in  100% effluent  when
 comparing the effects of the various TIE treatments.
TIE Phase n tests were performed on three samples
evaluated in the Phase I characterization and involved
the following steps as described by USEPA (1989a):

  •  C18  SPE columns  were eluted with a series of
    increasing methanol concentrations (25,  50, 75,
    80, 85, 90, 95, and 100%) to isolate the toxicants.
  •  The  acute toxicity  of each eluted fraction was
    determined and ithe fractions  found  to be toxic
    were combined.  The combined  toxic fractions
    were then reconcentrated using a second C18 SPE
                                                 88

-------
    column. Acute toxicity tests were used instead of
    chronic toxicity tests because the methanol elution
    concentrated the toxicants to acutely toxic levels.
  •  The concentrated sample was separated into 30
    fractions using HPLC and the. toxicity of each
    fraction was measured. Again, the toxic fractions
    were combined and reconcentrated on another CIS
    SPE column.
  •  The combined toxic sample was then analyzed by
    GC/MS.

As  shown in  Table A-2, toxicity was consistently
isolated in the 75  and  80% methanol fractions,
although  toxicity  was also recovered  in  the  50%
methanol fraction of one sample. Further separation of
the  toxicants  by  HPLC  recovered toxicity  in  a
relatively narrow band of fractions (fractions 22 to 28).

Table A-2.  Summary of TIE Phase II Results
Sample
Sample Collection Data (1992)
4/28
6/11
7/16
C. dubia percent survival in 100% sample
Original effluent
Post CIS SPE
SPE eluate
(Ix effluent)
50
100
0
0
100
0
0
80
0
Toxic methanol fractions (>20% mortality)
Methanol/water
(Ix effluent cone.)
HPLC fraction no.
(Ix effluent cone.)
50%
75%
80%
15
22-25
30
75%
80%
25
28 .
75%
80%
22
24
Organophosphate insecticides in effluent (ng/L)
Diazinon
Diazinon oxon
0.22
0.1
0.42
<0.1
0.71
1.45
GC/MS analysis of the toxic HPLC fractions identified
several potentially toxic compounds,  including the
organophosphate  insecticide,   diazinon,  and  its
metabolite, diazinon oxon (Table A-2). The 48-hour
LC50 of diazinon to C. dubia is reported to range from
0.35 to 0.61 ug/L (Amato et al., 1992; Ankley et al.
1991).  Based on the low  end of this range, the
diazinon concentrations in the Lawton effluent were
high enough to cause acute toxicity to C. dubia in two
of the three samples tested (0.42 and 0.71 |ig/L for the
June and July samples, respectively).
C.  dubia acute toxicity  tests were conducted to
evaluate the potential contribution of diazinon oxon to
effluent toxicity. The 48-hour LC50 for diazinon oxon
was determined to be 1 |J.g/L. These data indicate that
the diazinon oxon concentration in the July effluent
sample (1.45 |ig/L) was high enough to contribute to
the observed acute toxicity.
             TIE Procedure Update

 Data  on the chronic  toxicity of  organophosphate
 insecticides  is limited.  Unpublished data (TRAC
 Laboratories, 1992) suggest  that C.  dubia may be
 chronically sensitive to 0.12 to 0.38 jag/L diazinon (see
 also Section 2).  Chronic data would have been useful
 in defining the potential for diazinon to contribute to
 chronic toxicity at the Lawton POTW.
Further testing focused on confirming the contribution
of diazinon and diazinon oxon to effluent toxicity. A
partial Phase HI confirmation was performed using the
following steps (USEPA, 1989b):

  • Assessing diazinon's physical/chemical properties
    in relation to the TIE results.
  • Determining the contribution of diazinon  and
    diazinon oxon to whole effluent toxicity based on
    measured effluent concentrations.
  • Reviewing effluent toxicity  data  for  a 3-year
    period to determine if the occurrence of effluent
    toxicity  matched  seasonal  insecticide   use
    (Engineering Science, Inc.,  1992).

Diazinon matches  the  general  toxicant  profile
developed as part of the TIB. Removal of diazinon on
the CIS SPE column and its elution at high methanol
concentrations   is  consistent  with   diazinon's
characteristic as an organic chemical of low polarity.
The observed reduction in toxicity by pH adjustment
also is indicative of diazinon's tendency to break down
under acidic and alkaline conditions.

Concentrations of diazinon and diazinon oxon were
measured in 13 effluent samples collected from April
1 through August 21, 1992. Chronic toxicity data for
the insecticides  were  not  available  at  the time;
therefore, it was not possible to apply the correlation
approach. However, in seven cases, diazinon exceeded
the 0.35 [ig/L acute toxicity value for C. dubia. In two
of  these  cases, diazinon  oxon  concentrations  also
exceeded the acute toxicity value of 1.45 (ig/L- These
                                                  89

-------
data suggested that diazinon and diazinon oxon were
likely to cause mortality equal to or greater than that
found in the effluent samples.

A review of effluent toxicity data from 1989 to 1992
indicated a greater incidence of toxicity in the spring
and summer of each year when insecticides are most
often used. Effluent toxicity decreased in late summer
and fall  and generally disappeared in  the winter
months. These data support the evidence that toxicity
is associated with insecticides.
             TEE Procedure Update

 Confirmation of the role  of diazinon  and other
 toxicants would  have been more definitive if the
 current Phase HI procedures (USEPA, 1993b) for
 chronic toxicants had been applied. Useful procedures
 for confirming organophosphate insecticide toxicity
 include the correlation, mass balance, and  species
 symptoms approaches. An example of the use of these
 procedures is presented in Appendix F.
Based on previous studies (City of Greenville, 1991; C.
Kubula, personal communication, City of Greenville,
Texas, Public Works Department, 1992), the City of
Lawton decided to implement a public awareness
program  in  1993  to  control  the  discharge  of
insecticides to the POTW.  Information on the proper
use  and  disposal of insecticides  was printed in
newspaper articles  and  on  monthly  water  bills
(Engineering-Science, Inc.,  1993).   An electronic
message sign with insecticide information was also
located at major intersections. Since August 1993, the
POTW effluent has met  the toxicity permit  limit
(NOEC >96% effluent) with the exception of 2 months
in 1994 and several months in 1995 (as of September
1997).  Although diazinon was not confirmed as ;an
effluent toxicant, the City's ongoing insecticide control
effort appears to have been  successful in achieving
compliance with the chronic toxicity limit.

Akron, Ohio
A survey of six Ohio municipal wastewater treatment
plants was conducted to determine the level of toxicity
reduction  that can occur in POTWs (Neiheisel et al.,
1988).  Of the six WWTPs, the  City of Akron's
Botzum WWTP  received the most toxic influent
wastewater.   Significant  toxicity reduction  was
achieved through treatment; however, the effluent had
an impact on the Cuyahoga River.  A bioassessment
study of the river in 1984 revealed a severe impairment
to aquatic communities downstream of the WWTP
discharge. A review of the WWTP's operating records
showed a history ,of intermittent bypasses of raw
wastewater during storm events.

Based on the survey results, the Botzum WWTP was
selected by USEPA as a site for a TRE research study.
The research study focused on conducting toxicity tests
of the effluent and the bypassed wastewater and
characterization of • the variability and sources of the
impairment to the receiving  water (Mosure et al.,
1987).  In addition, TIE tests were performed to try to
identify the effluenjt toxicants.

Toxicity test results indicated that although CSOs may
contribute intermittently to poor  river quality,  the
continuous effluent discharge was probably the major
cause of the observed impact (Mount  and Norberg-
King, 1985).

The  TIE testing isolated toxicity on  the C18  SPE
column and the  toxicity  was eluted in  the 85%
methanol/water fraction (Mosure et al.,  1989). These
results suggested that non-polar organic compounds
were a principal cause of effluent toxicity. Metals also
were implicated as effluent toxicants. However, before
toxicant identification  and confirmation could be
performed, effluent toxicity abated.

The cause of this abatement is not known, although the
following events may have contributed to the improved
effluent quality. These events include:

  •  Increasing MLSS concentrations in the WWTP
    aeration basins.
  •  The shutdown of a large chemical manufacturing
    plant that discharged to the WWTP.
  •  Overall  improvements in WWTP  operation and
    the pretreatment program (Mosure  et al., 1987).

Biological surveys of the  Cuyahoga River in 1986
continued to show  poor water quality despite the
decrease in effluent toxicity (Mosure et al., 1987). It is
possible  that other dischargers to  the river were
contributing to the impairment or the recovery rate of
the river was slower than anticipated.

Billerica, Massachusetts
A study  was conducted at the City  of Billerica's
WWTP to evaluate sources of toxicity in the facility's
                                                 90

-------
       Toxicity Control Evaluation Update

 Abatement of effluent toxicity during the course of
 TREs is not uncommon. However, efforts to ensure
 ongoing compliance can be difficult when the original
 causes and sources of toxicity are not known. These
 situations dramatize the importance of documenting
 industrial pretreatment activities and POTW operations
 in the early stages of the TRE. Weekly or daily reports
 of production and  waste  discharge activities  by
 industrial users can provide a useful history of events
 that can be used to indicate potential sources.  This
 information is also helpful in subsequent pretreatment
 control studies if an industrial user is identified as a
 source of toxicity (Botts et al., 1994).
collection system (Durkin et al., 1987). A purpose of
the study was to evaluate the usefulness of Microtoxฎ
as a tool for tracing sources of toxicity.

The Billerica study was conducted in five stages:

  • Screening for WWTP influent toxicity.
  • Testing samples  from  pump  stations in  the
    collection system.
  • In-depth testing to determine the time of day when
    toxicity was observed at the pump stations.
  • Testing of the main sewer lines  above the pump
    stations where toxicity was indicated.
  • Final testing of tributary sewers.

Of the 11 pump stations tested, 2 were found to have
highly toxic wastewaters.   In one of these pump
stations, high levels of toxicity occurred only during
the 8 a.m: to 2 p.m. time period. Further investigation
of the  intermittently  toxic pump  station  provided
evidence  that the principal source of toxicity was an
industrial park.

References
Amato,   J.R.,   D.I.  Mount, EJ.  Durban,  M.L.
    Lukasewycz,  G.T. Ankley,  and E.D.  Robert.
     1992.  An  Example of the  Identification  of
    Diazinon  as a  Primary Toxicant in an  Effluent.
    Environ. Toxicol. Chem. 11: 209-16.

Ankley, G.T.,  J.R. Dierkes, D.A. Jensen, and G.S.
    Peterson.  1991.  Piperonyl Butoxide as a Tool
    in  Aquatic   Toxicological  Research  with
    Organophosphate  Insecticides.     Ecotoxicol.
    Environ. Saf.  21: 266-74.
            RTA Procedure Update

 Toxicity screening tools such as Microtoxฎ have been
 used to identify sources of toxicity in POTW collection
 systems.   It is necessary to first  determine if  a
 correlation exists between the compliance test and the
 screening test to ensure that the toxicity measured by
 the surrogate tool is the same toxicity indicated by the
 species used for compliance testing.  This correlation
 can be performed using POTW effluent; however, it is
 important  to note that correlation  results  may be
 different for individual industrial discharges.  As  a
 result, the screening test may yield false positive or
 false negative results.

 The advantage of screening tests is that a large number
 of samples can be processed at relatively low cost. As
 an alternative to these tools, POTW staff may consider
 using  the permit test species in  an  abbreviated test
 procedure such as that used in the TIE (USEPA1991).
 The  cost  of these  tests can  be  comparable  to
 commercially available screening tests if the number of
 replicates or sample concentrations is reduced or the
 exposure time is decreased.

 Although this study indicated a  potential source of
 toxicity, a  final determination of the source(s) of
 toxicity would require first treating the sewer samples
 in a simulation of the POTW to provide an accurate
 estimate of the refractory toxicity of the waste stream.
 Otherwise, as  discussed  in Section 5, the toxicity
 results may overestimate the toxicity of the discharge
 because some toxicity removal generally occurs in the
 POTW. A description of the updated RTA protocol is
 given in Section 5.
Bailey, H.C., J.L. Miller, M. J. Miller, L.C. Wiborg, L.
    Deanovic, and T. Shed. 1997.  Joint Toxicity of
    Diazinon and Chlorpyrifos Under Conditions of
    Acute Exposure to Ceriodaphnia dubia. Environ.
    Toxicol Chem. 16: 2304-08.

Botts,  J.A.,  J.W.  Braswell,  E.G.  Sullivan, W.C.
    Goodfellow, B.D. Sklar, and A.G. McDearmon.
    1987.   Toxicity  Reduction  Evaluation at  the
    Patapsco  Wastewater Treatment Plant.   Water
    Engineering Laboratory, Cincinnati, Ohio.

Botts, J.A., T. Schmitt, E. Wilson, M. Collins, D.
    Waddell,  R.  Diehl,  and L.  Ehrlich.   1994.
    Refractory Toxicity Assessment: An Alternative
    Approach   for  Chronic  Toxicity  Reduction
    Evaluations. Technical paper  presented  at  the
    Annual Conference and Exposition of the Water
                                                   91

-------
    Environment Federation, Chicago, Illinois. Paper
    #AC944404.

City of Greenville.  1991. TRE Phase B Final Report.
    Prepared for the City of Greenville, Texas.

Durkin, PP., C.R. Ott, D.S. Pottle, and G.M. Szal.
    1987.   The Use of the Beckman  Microtoxฎ
    Bioassay System to Trace Toxic Pollutants Back
    to Their Source in Municipal Sewerage Systems.
    Proceedings of  the  Water  Pollution Control
    Federation Association Annual  Meeting,  New
    Hampshire.

Engineering-Science, Inc. 1991.  Toxicity Reduction
    Evaluation Phase I Report. Submitted to the City
    of Lawton, Oklahoma.

Engineering-Science, Inc. 1992.  Toxicity Reduction
    Evaluation:   Toxicant  Identification  and
    Confirmation.  Submitted to the City of Lawton,
    Oklahoma.

Engineering-Science, Inc. 1993.  Review ofDiazinon
    Control Options.  Submitted to the City  of
    Lawton, Oklahoma.

EXTOXNET.  1990. Pesticide Information Profile for
    Diazinon,   Extension  Toxicology   Network.
    Oregon State University, Corvalles, Oregon.

Microbics Corporation.  1982.   Microtoxฎ System
    Operating Manual. Carlsbad, California.

Mosure,  T.E., J.P.  Pierko,  J.M.  Schmidt,  R.A.
    Monteith,  and J.S.  Mosser.  1987.   Toxicity
    Reduction   Evaluation  at  the  Akron  Water
    Pollution  Control  Station,   Draft  Report.
    Cooperative agreement #CR813681010. USEPA
    Office  of Research and Development   Risk
    Reduction  Engineering  Laboratory, Cincinnati,
    Ohio.

Mount, D.I., and T.J. Norberg-King. 1985.  Toxicity
    Reduction  Evaluation:  Akron STP - Part I.
    USEPA Environmental  Research Laboratory,
    Duluth, Minnesota.

Neiheisel, T.W., W.B. Horning,  B.M.  Austern, D.F.
    Bishop,  T.L. Reed, and J. F. Estemik.  1988.
    Toxicity Evaluation at  Municipal Wastewater
    TreatmentPlants.7oซrwa/o/W2/erPo//H/. Control
    Fed.  60(1): 57-67.
TRAC Laboratories, Inc., 1992.  Unpublished data.
    Prepared by TRAC Laboratories, Inc.  Denton,
    Texas.       '

USEPA. 1987. Toxicity Identification Evaluation of
    the Largo, Florida POTW Effluent. J.R. Amato,
    D.I. Mount, M. Lukasewycz, E. Durham, and E.
    Robert.  National  Effluent Toxicity Assessment
    Center, Duluth, Minnesota.

USEPA. 1988. Memorandum from Donald Mount to
    USEPA Region IX. June 25, 1988.

USEPA.   1989a.}  Methods  for  Aquatic  Toxicity
    Identification Evaluations: Phase II:  Toxicity
    Identification Procedures. EPA/600/3-88/035.
    Office of Research & Development,  Duluth,
    Minnesota.   :

USEPA.   1989b.•  Methods  for  Aquatic  Toxicity
    Identification Evaluations, Phase HI.   Toxicity
    Confirmation ] Procedures. EPA/600/3-88/036.
    Office of Research & Development,  Duluth,
    Minnesota.   ;

USEPA.   1991.   Methods for  Aquatic  Toxicity
    Identification Evaluations:   Phase  I  Toxicity
    Characterization Procedures.  Second Edition.
    EPA/600/6-91-003.  National  Effluent  Toxicity
    Assessment Center, Duluth, Minnesota.

USEPA. 1992.  Toxicity Identification Evaluations:
    Characterization of Chronically Toxic Effluents,
    Phase I. EPA/600/6-91-005F.  National  Effluent
    Toxicity Assessment Center, Duluth, Minnesota.

USEPA.  1993a. ;  Methods for  Aquatic  Toxicity
    Identification Evaluations: Phase II  Toxicity
    Identification Procedures for Samples Exhibiting
    Acute and Chronic Toxicity. EPA/600/R-92-080.
    National Effluent  Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.   1993b.   Methods for  Aquatic  Toxicity
    Identification Evaluations. Phase III  Toxicity
    Confirmation Procedures for Samples Exhibiting
    Acute and Chronic Toxicity. EPA/600/R-92-081.
    National Effluent  Toxicity Assessment Center,
    Duluth, Minnesota.
                                               92

-------
                                        Appendix B

                                     TRE Case Study:
        Central Contra Costa Sanitary District, Martinez, California
Abstract
TRE Goal:
Test Organisms:

TRE Elements:
Toxicant Identified:
Toxicity Controls:
NOEC;>10%
Echinoderms (S. pufpuratus
and D. excentricus)
TIE
Copper
Pretreatment requirements
Summary
Chronic toxicity was detected in a municipal effluent
with the echinoderm fertilization assay. D. excentricus
(sand dollar) appeared more sensitive to the effluent
than did S. purpuratus (purple urchin). A Phase I TEE
was conducted using procedures described by USEPA
(1988a)  that  were  adapted to  the  echinoderm
fertilization toxicity test. The Phase I TIE implicated
cationic metals as the cause of chronic toxicity, and
follow-up investigations suggested that Cu was the
primary cation responsible. As part of the TIE, toxicity
tests were conducted on ammonia and several cations.
No observable effect concentrations for D. excentricus
were >13.4 jag/L silver (Ag), >9.4 |ag/L Cd, 3.8 to 13.1
(ig/L  Cu, >0.7 |ag/L mercury (Hg), and 10 mg/L
nitrogen as total  ammonia.  The data also suggested
that inter-specific differences in sensitivity to Cu and
ammonia  exist  between  D. excentricus  and  S.
purpuratus.

Key Elements
1.  TIE procedures for freshwater organisms can be
    successfully  modified  to   apply  with  the
    echinoderm fertilization toxicity test.
2.  This study demonstrated that  Cu could have
    accounted for the  intermittent effluent toxicity
    observed.
3.  Echinoderms  exhibited  comparatively   high
    sensitivity to Cu with EC50s for both species of
    approximately 25 ug/L.
4.  Source control  measures  were  successful  in
    reducing  Cu  concentrations  by approximately
    25%.

Introduction
Permit Requirements
The Central Contra Costa Sanitary District (CCCSD,
Martinez, California) was required by the State Water
Quality Control Board, San Francisco Bay Region, to
conduct a TRE to identify the chemical constituents in
their final effluent that were responsible for observed
chronic toxicity in the echinoderm fertilization toxicity
tests. Results of monthly compliance tests showed
frequent exceedance  of the discharge permit  limit
(NOEC ;> 10% effluent).

Description of the Treatment Plant
The CCCSD WWTP  provides  secondary  level
treatment for combined domestic, commercial, and
industrial wastewater from a 126-square mile area with
a population of approximately 400,000. The treatment
plant has an average dry weather design capacity of 45
mgd and currently discharges an annual average flow
of 38.7 mgd into upper San Francisco Bay. Treatment
facilities consist of screening, primary sedimentation,
activated sludge, and secondary clarification followed
by  chlorination in contact basins.   In the treatment
process,  waste-activated sludge  is thickened via
flotation thickeners, and lime is  added to  assist in
dewatering with centrifuges. The combined primary
and waste-activated  sludge   is   dewatered  and
incinerated in multiple-hearth furnaces. The effluent
TSS and BOD concentrations average <10 mg/L. Total
ammonia concentrations range from 10-35 mg/L with
an average of 25 mg/L.
                                                93

-------
Toxicity Identification Evaluation
General Procedures
The echinoderm  fertilization toxicity  tests weire
conducted on the final effluent according to published
procedures (Dinnel, et al., 1982, as modified by S.
Anderson,  1989) using the  West Coast species S.
purpuratus and D. excentricus. The purpose of the test
is to determine the concentration of a test substance
that reduces egg fertilization by exposed sperm relative
to fertilization in a control solution. Two species were
used in this test because the echinoderms are obtained
from feral populations which are gravid at different
times during the year. Effluent samples were 24-hour
flow-proportional composites. Samples were screened
for toxicity within 36 hours of collection. The effluent
salinity was adjusted to 30% using hypersaline brine
(90%), and the pH was adjusted to 8.0 ฑ 0.05.

Phase I TIE Studies
The results of this TIE have been published elsewhere
(Bailey, et al., 1995). The Phase I TIE included the
procedures described by USEPA  (1988a).   After
completing the TIE manipulations, the effluent was
salinity and pH adjusted as previously noted.

Table B-l.  Summary of Results of Phase I TIE Conducted
on Two Effluent Samples with D. excentricus
Treatment
PH3
pHll
Filtration
Aeration
EDTA
Sodium thiosulfate
Post CIS SPE
column
Mcthanol cluate
add-back
Sample 1
No effect on toxicity
Eliminated toxicity
No effect on toxicity
No effect on toxicity
Eliminated toxicity
Eliminated toxicity
No effect on toxicity
No toxicity
Sample 2
Increased toxicity
No effect on toxicity
No effect on toxicity
No effect on toxicity
Eliminated toxicity
Eliminated toxicity
No effect on toxicity
No toxicity
Phase I TIEs were conducted on two effluent samples.
The data for both samples (Table B-l) suggested that
EDTA and sodium thiosulfate were consistently the
most   effective  treatments  in  reducing toxicity.
Extraction of the sample with SPE columns did not
reduce toxicity, suggesting that non-polar organics and
weak organic acids  and bases  were not causes  of
toxicity.  This conclusion is supported by the fact that
clution of the columns with methanol did not yield
toxicity.  The effectiveness of EDTA in eliminating
toxicity suggested  that  a  divalent  cation(s)  was
responsible for toxicity in the samples tested.  The
concurrent  effectiveness  of sodium thiosulfate  in
reducing toxicity suggested that the potential suite of
cations was limited to Cd, Cu, and Hg (USEPA, 1991).
In one case, toxicity also appeared to be increased by
temporarily reducing the sample pH to 3; greater
toxicity at lower pHs has been associated with Cu
(Schubauer-Berigan et al., 1993).

Because the  effluent samples contained moderate
levels of ammonia (20—25 mg/L total ammonia), the
potential contribution of ammonia to effluent toxicity
was determined by comparison with ammonia toxicity
tests.   This approach was  taken because the TIE
guidelines evaluate ammonia toxicity by adjusting the
pH of the test solution and preliminary data indicated
that   these  pH  adjustments   adversely  affected
fertilization success.

Contribution of Ammonia to Toxicity
Ammonia  toxicity tests were conducted in natural
seawater spiked with ammonia chloride; fertilization
success was evaluated using logarithmically spaced
concentrations across a range of 1.0 to 100.0 mg/L N
as total ammonia.  Test solutions were adjusted to pH
8.0 ฑ 0.05 prior to exposure.

The NOECs for D. excentricus and S. purpuratus were
both 10 mg/L N  as total ammonia.  Based on the
unionized   fraction,  the  NOECs were  Q.21  and
0.17 mg/L N for D. excentricus  and S. purpuratus,
respectively (calculated per USEPA, 1988a). However,
large differences existed between the response of the
two species at concentrations higher than the NOEC.
For S. purpuratus, the IC25 was greater than 100 mg/L
N as  total ammonia (1.69  mg/L N as  unionized
ammonia)  compared with an IC25 estimate of 16.5
mg/L N (0.34 mg/L N as unionized ammonia) for D.
excentricus.  Because the upper limit of ammonia
concentrations in the effluent was 25 mg/L N as total
ammonia, these results suggested that ammonia alone
could not account for NOECs that were ^33% effluent,
a concentration that would correspond to a maximum
of 8.25 mg/L N as total ammonia.

Identification and Confirmation of the Role of
Cationic Metals
Sensitivity of echinoderms to cationic metals
Once it appeared that a divalent cation was responsible
for the effluent's toxicity, candidate metal ions (Cd, Cu,
and Hg) and Ag were evaluated for toxicity  with
                                                 94

-------
D. excentricus  and S. purpuratus.  Metal solutions
were prepared in moderately hard freshwater (USEPA,
1991) using reagent grade salts of Cu, Cd, and Hg.
The CCCSD also was concerned about the potential
for Ag to contribute to effluent toxicity; therefore, tests
were performed with silver salts. Stock concentrations
of metals were confirmed by either graphite furnace
(Ag, Cu, and Cd) or cold vapor (Hg) AA spectroscopy
(APHA, 1989). Hypersaline brine was then added (1/3
brine:2/3 metal solution) to bring the salinity to 30%,
and the pHs of the solutions were adjusted to 8.0 ฑ
0.05 prior to exposure. This procedure was analogous
to the preparation of the effluent samples  prior to
testing.  Serial dilutions that incorporated a 50%
dilution factor were made from the stock solutions to
achieve exposure concentrations that bracketed those
found  in the effluent.   The NOECs from multiple
toxicity tests  on  Ag, Cd,  Cu,  and  Hg  with  D.
excentricus and S. purpuratus are summarized in Table
B-2.   Side-by-side comparisons between the two
species are shown by the paired values in the table.

Table B-2.  NOECs Obtained for/), excentricus and
S. purpuratus Exposed to Different Metals*
Metal
Ag
Cd
Cu
Hg
'. NOECS (ug/L)
D. excentricus
>13.4
>9.4
>67.0
10.0
13.1
5.4
3.8
8.0
>0.7
>2.2
S. purpuratus
>13.4
Not tested
>67.0
20.0
19.7
Not tested
Not tested
Not tested
>0.7
Not tested
 * When seasonally available, concurrent tests were conducted
 with both species. Values given as jig metal/L (Bailey et al.,
 1995).
In  some  cases,  seasonal  spawning  constraints
precluded  conducting  concurrent  tests  with  S.
purpuratus. One comparison was conducted with Ag;
the NOECs for both species were >13.4 ng/L. Two
tests were conducted with Cd; in both cases the highest
concentrations tested (9.4 and 67.0  (ig/L) failed to
produce  any measurable  effects  on  fertilization
success.  Five tests were performed on Cu with D.
excentricus. The NOECs ranged between 3.8 and 13.1
ug/L with  an average of 8.1  pg/L. In two  of three
concurrent tests with S. purpuratus, the NOECs were
1.5 to 2 times greater than those obtained with D.
excentricus. In two tests  with  Hg, no effects on
fertilization success were found at concentrations up to
0.7 and 2.2 ug/L, respectively.

Comparison of  toxic concentrations of metals
with concentrations found in the effluent
The NOECs for each of the metals were compared
with the discharger's analytical records to determine
which metals were present individually in the effluent
at concentrations high enough to inhibit fertilization
success. Toxicity ratios were calculated for each metal
[metal concentration  in effluent (|ig/L)  -=- NOEC
(ug/L)].  A ratio greater than 1 suggested that the
metal(s) was present in the effluent at concentrations
high enough to produce toxicity. Conversely, a ratio of
1, or less, suggested that the concentration of metal
was 13.4
>9.4; >67
3.8-13.1
>0.7; >2.2
Ratio
<;0.3
sO.2
0.4-5.3
<;0.6
* Values given as 7.5 (80.0 ฑ 2.0)ug metal/L.

Confirmation of the role of Cu in effluent toxicity
The next confirmation step compared fertilization
success in an effluent sample against that in seawater
spiked  with copper sulfate  (CuSO4)  to  the same
concentration found in the effluent.  These exposures
were  conducted simultaneously  using  the  same
gametes fromD. excentricus. Fertilization success also
                                                  95

-------
was evaluated in an effluent  sample spiked with
different concentrations of Cu, such that subsamples of
the effluent contained 1, 2, and 3 times the amount of
Cu (measured concentrations) as the original sample.
Serial dilutions, which incorporated a 50% dilution
factor, were then prepared from the unspiked and 2x
spiked samples and fertilization success evaluated with
ฃ>. excentricus. Depending on the results, it could be
determined whether Cu was responsible for toxicity in
the effluent. The reasoning was if Cu was the primary
factor  controlling toxicity,  then  the  LOECs  and
NOECs obtained for the spiked and unspiked samples
should  be the same, based on  Cu concentration.
Similarly, based on percent effluent, the NOEC and
LOEC associated with the spiked sample should be one
dose level lower than in the unspiked sample.

The results of parallel toxicity tests with D. excentricus
on  effluent   and seawater spiked  with  Cu  at
concentrations found in the effluent are summarized in
Table B-4. Based on Cu concentration, the NOECs
and LOECs were the same between the effluent sample
and the concurrent toxicity test with seawater spiked
with Cu.  Furthermore, the percent fertilization was
similar at corresponding Cu concentrations in both
toxicity tests. These data suggested that Cu accounted
for the reduction in fertilization success associated with
this effluent sample.   Fertilization success  in  an
effluent sample and the same sample spiked with Cu is
shown in Table B-5.

Discussion
The data demonstrated that procedures for conducting
TIEs with freshwater organisms can be successfully
applied to the echinoderm fertilization toxicity test.
Table B-4.  Comparison of NOECs, LOECs, and Percent
Fertilization Obtained withD. excentricus Exposed to Effluent
and Seawater Spiked with Cu
Treatment
Effluent
Seawater Cu
spike
: NOEC* .
3.8 (89.3 ฑ 3.0)
: 7.5 (80.0 ฑ2.0)
•:'LOEC*;:''
7.5 (73.3 ฑ6.1)
7.5 (80.0 ฑ 2.0)
* Percent fertilization given in parentheses (mean ฑ SD).

 The results of this study suggest that Cu could have
 accounted for the intermittent toxicity demonstrated by
 the echinoderm fertilization test. Of the four metals
 identified in the Phase TIE, Cu was the only one that
 occurred in the  effluent at  concentrations  that
 overlapped the toxic range.   Confirmatory studies
 conducted with two different effluent samples  also
 showed that Cu could account for the adverse effects
 observed with the whole effluent.  Paired tests  also
 suggested that  Cu  exhibited  greater  toxicity to
 D. excentricus than to S. purpuratus. This is important
 because  S.  purpuratus   generally  exhibited  less
 sensitivity to the effluent.

 Source control measures implemented by the CCCSD
 successfully reduced Cu concentrations in the effluent
 by 25%. This reduction made it difficult to obtain
 samples with sufficient toxicity to fully complete the
 confirmatory phase of the TEE.  In fact, nearly all the
 samples tested at the end of the TIE failed to produce
 a measurable response with S. purpuratus.

 Acknowledgments
 This work was supported wholly  or in  part by the
 CCCSD, Martinez,  California.  Bart Brandenburg,
    Table B-5. Percent Fertilization Obtained with D. excentricus Exposed to Effluent and Effluent Spiked with Cu*
Unspiked effluent
Effluent (%)
0.0
8.4
16.8
33.5
67.0 (IxCu)
ug/LCu
0.0
0.8
1.6
3.3
6.6
Fertility (%)
96.0 ฑ 2.5 ,
96.7 ฑ3.1,
97.3 ฑ 1.2
91.3 ฑ1.2
82.0 ฑ 4.7f
Effluent spiked with Cu
Effluent (%)
0.0
8.4
16.8
33.5 (IxCu)
67.0 (2xCu)
67.0 (3xCu)
ug/LCu
0.0
1.6
3.3
''. 6.6
13.2
19.8
Fertility (%).'
96.0 ฑ 2.5
90.7 ฑ 2.3
90.3 ฑ 2.3
83.3 ฑ 2.7|
74.8 ฑ 2.2f
71.7 ฑ 12.9f
* Fertilization data are the means and standard deviations of three replicates.
t Significantly less than controls; p < 0.05.
                                                 96

-------
Bhupinder Dhaliwal, and Jim Kelly of the CCCSD
managed the various aspects of this project. The TRE
studies were conducted at AQUA-Science, Davis,
California, under the direction of Jeffrey L. Miller,
Ph.D., and Michael J. Miller.

References
American  Public   Health Association  (APHA),
    American Water Works Association, and Water
    Pollution Control Federation.   1989.  Standard
    Methods  for the Examination  of Water  and
    Wastewater.  17th ed.  American Public Health
    Association,  Washington, D.C.

Anderson,   S.L.   1989.      Effluent   Toxicity
    Characterization Program.  Report to the  San
    Francisco Bay Regional Water Quality Control
    Board. Oakland, California.

Bailey, H.C., J.L.  Miller, M.J. Miller,  and B.S.
    Dhaliwal.   1995.    Application  of Toxicity
    Identification Procedures  to  the  Echinoderm
    Fertilization  Assay  to Identify Toxicity  in a
    Municipal Effluent. Environ. Toxicol. Chem.  14:
    2181-86.

Dinnel, P.A., Q.J.  Stober, S.C. Crumley, and R.E.
    Nakatani.  1982. Development Of A Sperm Cell
    Toxicity Test For Marine Waters. In J.G. Pearson,
    R.B.  Foster, and W.E.  Bishop, eds.  Aquatic
    Toxicology  and  Hazard   Assessment:  Fifth
    Conference.  STP 766.  American Society  for
    Testing and Materials, Philadelphia, Pennsylvania.

Schubauer-Berigan, M.K., J.R. Dierkes, P.D. Monson,
    and G.T. Ankley.  1993.  pH Dependent Toxicity
    of Cd, Cu, Ni, Pb, and Zn to Ceriodaphnia dubia,
    Pimephales  promelas,  Hyalella  azteca,  and
    Lumbriculus variegatus. Environ. Toxicol. Chem.
    11: 1261-66.

USEPA.   1988a.   Methods For Aquatic Toxicity
    Identification Evaluations:  Phase I Toxicity
    Characterization Procedures. EPA/600/3-88/034.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.   1988b.   Methods For Aquatic Toxicity
    Identification Evaluations:   Phase II Toxicity
    Identification Procedures.  EPA/600/3-88/035.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.   1988c   Methods For Aquatic Toxicity
    Identification Evaluations:  Phase III Toxicity
    Confirmation Procedures.  EPA/600/3-88/036.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.    1991.    Methods for  Aquatic Toxicity
    Identification  Evaluations: Phase I,  Toxicity
    Characterization Procedures.  Second Edition.
    EPA/600/6-91-003.  National Effluent Toxicity
    Assessment Center, Duluth, Minnesota.

USEPA.  1995. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to West Coast Marine  and Estuarine
    Organisms.  EPA/600/R-95/136.   National
    Exposure Research Laboratory, Cincinnati, Ohio.
                                               97

-------
                                         Appendix C
                                            !                       i
                                     TRE Case Study:
                          City of Reidsville, North Carolina
Abstract
TRE Goal:
Test Organism:
TRE Elements:

Toxicant Identified:
Toxicity Controls:
NOEC >90%
C. dubia
TIE and Toxicity Tracking
Assessment (RTA)
Surfactants
Pretreatment requirements
Summary
The TRE study used a novel approach to identify the
sources of POTW  effluent toxicity.   Subsequent
modifications  in  chemical  usage by  industrial
contributors successfully reduced effluent toxicity to
the NOEC limit in  1994.  Further studies are in
progress  to ensure consistent compliance with the
toxicity limit.

Key Elements
1.  Other TRE procedures  can  be used if the  TIE
    cannot identify the effluent toxicants.  One such
    procedure uses a toxicity-based tracking approach
    to locate the sources of  toxicity in municipal
    collection systems.
2.  The toxicity-based tracking approach, referred to
    as the RTA procedure, can be adapted to fit the
    site-specific conditions at each POTW.
3.  Once identified,  the toxic contributors  can be
    required  through the  industrial  pretreatment
    program  to  reduce  the discharge  of toxicity.
    Practical  control techniques  are  available to
    industries,  including   substitution   of toxic
    chemicals, waste minimization, and  pollution
    prevention.

Introduction
The City of  Reidsville was required by the North
Carolina Division  of Environmental Management
(NCDEM) to conduct a TRE based on evidence of
chronic effluent toxicity  at  its  POTW.   Monthly
NOECs for C. dubia have averaged about 35% effluent
since 1992. These values  show that chronic effluent
toxicity has consistently exceeded the discharge permit
NOEC limit of 90% effluent.

Background
In 1992, the City submitted a TRE plan and initiated
TIE studies to  determine the cause(s) of the effluent
toxicity. Chronic TIE Phase I (Tier I) tests indicated
that surfactants were the principal toxicant group. This
evidence was based on toxicity reduction by filtration,
aeration, and CIS SPE in the Phase I tests.  TIE Phase
n tests were performed to try to identify the toxic
surfactant  compounds; however, the results  were
inconclusive because of the difficulty in isolating the
toxicants and  the lack of  conventional  analytical
techniques for  surfactant compounds.  The toxicants
removed by the CIS SPE  column were recovered by
eluting  the column with  methanol,   but  toxic
compounds could not be identified in the column
extract (Burlington Research Inc., 1993).

In cases where  the TIE is not successful in identifying
the effluent toxicants, other TRE steps can be used to
gather  information  on the  nature  and  sources  of
effluent toxicity. USEPA and several  municipalities
have worked together in  USEPA funded studies to
develop the RTA method, which can be used to assess
the potential  toxicity  contribution  from indirect
dischargers in sewerage collection systems (USEPA,
1989a; Botts et al., 1987; Morris et al, 1991; Fillmore
et al., 1990; Collins et al., 1991). The RTA procedure
involves treating industrial wastewater samples in a
bench-scale, batch simulation of the POTW, and
measuring  the resulting  toxicity.    The toxicity
                                                98

-------
remaining  after  batch treatment,  referred  to  as
refractory toxicity, represents the toxicity that passes
through the POTW and is discharged in the effluent.
Several municipalities have successfully used the RTA
procedure to identify industrial sources of toxicity
(Botts  et  al.,   1992;  Morris  et  al.,   1991;  and
Engineering-Science, Inc., 1992).

Description of Treatment Plant
The major treatment processes at the Reidsville POTW
are extended activated sludge treatment and filtration.
Influent  wastewater,  which  averages  2.8 mgd,  is
initially screened and then treated in two activated
sludge aeration basins equipped  with  mechanical
surface aerators.   Both  carbonaceous BOD  and
ammonia are removed in this  single-stage aeration
system. After 48 hours contact time, the basin effluent
flows to the final clarifiers for solids clarification. The
clarified effluent is then passed through sand filters to
remove  remaining  suspended  solids   that  may
contribute  to effluent BOD.  The  filter effluent is
disinfected with chlorine gas and dechlorinated and
aerated prior to discharge. Waste activated sludge is
thickened and aerobically digested for land application.

Refractory Toxicity Assessment Procedure
Selection of Industries for Testing
Acute  and  chronic toxicity tests were performed on
raw wastewater from the seven permitted significant
industrial users  in the Reidsville collection system.
The industrial  wastewater samples  were tested  at
concentrations   that  reflected  the  average flow
contribution of the industries to the POTW (dilutions
were made with reconstituted lab water).

The results showed that five of the seven industries
were contributing chronic toxicity to the POTW (Table
C-l).  It is possible that at  least some of the raw
wastewater toxicity would be removed by treatment at
the POTW; therefore, the five toxic industrial users
were selected for further evaluation by RTA testing. A
description of the industries evaluated in the RTA is
provided in Table C-2.

Test Procedure
A step-by-step  description of the RTA  procedure is
given  in Section 5 and Appendix  J.   The generic
procedure must be adapted to simulate the treatment
processes and operating conditions at each POTW.
Several types of treatment processes can be simulated,
including conventional activated sludge systems (Botts
et al.,  1987; Morris et al., 1991; and Fillmore et al.,
Table C-l. Chronic Toxicity of Raw Industrial
Wastewaters
•Industry
A
B
C
D
E
F
G
C. dubia Chronic Pass/Fail Result*
May
1992
Fail
Fail
Fail
Fail
Pass
Pass
Pass
June
1992
Fail
NTf
Fail
NT
Pass
Pass
Pass
July
1992
Fail
Fail
Fail
Fail
Fail
Pass
Pass
April
1993
Fail
Fail
Fail
Fail
Fail
NT
NT
 * Tests were conducted using industrial wastewater diluted
   according to its percent contribution to the total POTW
   influent.
 t NT = Not tested.

Table C-2. Description of Industries Evaluated in the
RTA
Industry
A
B
C
D
E
Domestic
Type
Textile
Tobacco
Products
Can
Making
Food
Processing
Metal
Finishing

Flow
(mgd)
1.072
0.308
0.085
0.189
0.031

%Flow* to
POTW
65
28
10
12
2
38
 * Based on maximum industrial flow and minimum
   POTW influent flow, except for domestic, which is
   based on average flow and minimum POTW influent
   flow.
1990), single  and two-stage nitrification  processes
(Collins et al., 1991), and BNR systems (Botts et al.,
1992).

The RTA simulated the two main treatment processes
at the Reidsville POTW: the activated sludge and sand
filtration processes.  Wastewater samples  were first
treated in biological reactors and then the clarified
effluents were passed through a bench-scale sand filter
column.
                                                 99

-------
Two types of simulations were tested as shown in
Figure 5-2 (see Section 5).  A control simulated the
existing treatment conditions and treated  only the
POTW influent. The second simulation evaluated the
addition of die industrial discharge to the POTW and
treated the industrial wastewater spiked into the POTW
influent.

The amount of industrial wastewater spike represented
the conservative condition of maximum industrial flow
and minimum total influent flow at the POTW.  The
operating conditions for the simulations are described
in Table C-3.

 Table C-3. Comparison of Operating Conditions for the
 City of Reidsville POTW Processes and RTA Simulation
 Tests
POTW Process
Specifications
Treatment
Plant
RTA
Simulation
Activated Sludge Process
Mixed liquor solids
(mg/L)
DO (mg/L)
Treatment period
(hours)
2,200-2,500
>2
48
2,240-2,740
2.4-9.2
48
Sand Filter Process
Filtration rate
(gpm/sf)
Total filter area (sf)
Sand particle size
(mm)
Sand depth (inches)
Water depth on
filter (ft)
Backwash rate
(gpm/sf)
0.8
2,520
0.45
10
0-7
12
0.8
0.09
0.45
10
0.1-2.5
5 (estimated)
The results of the control and spiked simulations are
compared to determine  whether addition of  the
industrial wastewater increases effluent toxicity.  Ah
industry would be considered a source of toxicity if the
effluent of the spiked simulation is more toxic than the
control effluent.

Sampling
Three rounds of RTA tests were performed over a
4-month period. Twenty-four hour composite samples
of the  industrial   wastewaters, POTW  influent,
domestic  wastewater,  and  POTW effluent  were
collected for testing. In addition, a grab sample of the
POTW  RAS was  collected  on the day of testing.
Domestic wastewater was tested because TRE studies
at other municipalities have shown that domestic
sources can contribute to effluent toxicity (Botts et al.,
1990). The POTW effluent served as a baseline for
comparison with the: RTA control to determine if the
treatment performance of the simulations and the
POTW were similar)

Toxicity Monitoring
Following biological treatment, the clarified reactor
effluents were passed through the sand filter column
and  the  resulting filtrates were tested  for  chronic
toxicity using C. dubia, the test species specified in the
NPDES permit. Each RTA effluent sample was used
for both test initiation and renewals on days 3 and 5 of
the toxicity test (USEPA, 1989b).

Results          ;
Source  Characterization
Two rounds of RTA tests were used to characterize the
sources  of toxicity.  As shown in Figure C-l, the
effluent TUc for the two control simulation tests in
Round 1 were 3.8 and 3.1. These values compare well
with the POTW effluent (TUc =3.6).  The control
simulation effluents in Round 2 also exhibited similar
toxicity  (TUc =3.0 and 2.9)  as the POTW effluent
(TUc =3.4).  These results indicate that the RTA test
accurately simulated the POTW with respect to toxicity
treatment.         !

As shown in Figure C-l,  the  effluent from the
simulation spiked with Industry A wastewater was
about twice as toxic (TUc=6.7) as the control effluents
in both rounds of tests. Effluent TUc values for the
simulations spiked  with other industrial wastewaters
were similar to or less than the effluent TUc for the
controls.

The  results  of both rounds  of testing  indicate  a
potential for Industry A to contribute toxicity to the
POTW.  The results for the simulations spiked with the
other industrial wastewaters suggest that Industries B,
C, D, and E do not contribute measurable toxicity to
the POTW.

Toxicity Confirmation
A recent study for a New Jersey municipality found
that an industry was contributing toxicity in amounts
                                                100

-------
Round 1 Toxicity Results
6_

u 5
O A
I i
H 3 -
o
1 2
a















v
ซ *
_g.








%flil
5SO.l1
2^i**s : ^i^i
l^ ^ ง
^ ง: ' •
Control Control POTW A B C D
Simulation Tests
Note: Industry E was not tested in round 1.
Round 2 Toxicity Results
6_

u
i
O A -
^ 4
I ,
H 3
'I

r\














^
%
^ f ?,
^&.v v
1 C*




E












i
Domestic





*"j p

^ s 5
Control Control POTW A B C D
Simulation Tests

1
E





^ %
;c5
i




i
Domestic
1010P-10
Figure C-l. Results of RTA (rounds 1 and 2).
                                                   101

-------
high enough to mask other smaller sources of toxicity
(Morris et al., 1991). It was necessary to remove the
larger source of toxicity from the RTA test regime
before other significant sources could be identified.
The City of Reids ville decided to conduct a third round
of  tests to  determine  if  a  similar  situation  was
occurring at their POTW.

Round 3 involved using a mock influent that did not
contain Industry A wastewater. The mock influent was
used in lieu of the POTW influent for the controls and
the spiked simulations. The mock influent consisted of
samples collected from each major sewer line with the
exception  of  the  sewer  receiving  Industry  A
wastewater.

Toxicity results for the RTA simulation effluents are
presented in  Figure  C-2.   A comparison of results
shows that the  effluent of the  Industry A spiked
simulation was several times more toxic (TUc=6.8)
than the control effluent (TUc =1.2).  These results
provide further evidence that Industry A is a source of
toxicity. The simulations spiked with Industry C and
D wastewater had  similar effluent toxicity (TUc=1.3
for both) compared to the control. These data indicate
         that Industries C and D are not contributing significant
         toxicity to the POTW.

         The simulation spiked with domestic wastewater had
         greater effluent toxicity (TUc=2.3) than the control
         (TUc=1.2).   These  results  suggest that this  waste
         stream may be a source of toxicity; however, results of
         Round 1 and  2 indicate  that domestic wastewater
         collected from other areas of the collection system is
         not a problem.  Further studies are planned to evaluate
         the potential toxicity contribution from domestic
         sources throughout the collection system.

         Discussion       ,
         The results of this study indicate that Industry A is a
         major contributor to chronic effluent toxicity  at the
         Reidsville POTW. None of the other industries  (B, C,
         D, and E) were found to discharge measurable toxicity
         even after the potential  toxicity  interference from
         Industry A was removed.

         In January 1994, the City of Reidsville implemented a
         program to  minimize or eliminate the discharge of
         industrial chemicals that may contribute to the POTW
         effluent toxicity.  Although the RTA results indicated
                                     Round 3 Toxicity Results
            Control   Control   POTW
A        B        C
 Simulation Tests
Domestic
   Note: A replicate control and POTW effluent were not tested in round 3. Industries B and E were not indicated to be sources
         of toxicity in rounds 1 and 2; therefore, these industries were not tested in round 3.
                                                                                             1010P-11
Figure C-2. Results of RTA (round 3).
                                                 102

-------
that Industry A is the major contributor of chronic
toxicity, all of the City's eight permitted industrial
users  were requested to participate.  The program
involved:

  •  An evaluation of current chemical usage and the
    selection of alternative materials of low toxicity,
    low inhibition potential, and high biodegradability.
  •  An  on-site  evaluation of waste-minimization
    practices by the North Carolina Office of Waste
    Reduction.

Particular attention was given to surfactant products or
chemicals with surfactant constituents because the TIE
had indicated surfactants to be the principal toxicant in
the POTW effluent.  Industries were requested to
maintain chronological records of changes in chemical
usage, production, and housekeeping practices. These
records were used to compare the timeline of industry
modifications to results of chronic toxicity monitoring
at the POTW.

Follow-up monitoring results  showed a substantial
reduction in effluent toxicity.   Beginning in March
1994, the IC25 values (an endpoint that approximates
the NOEC) for 7 of 10 monthly C. dubia toxicity tests
were 2:90%. A review of the industries' chronological
records established a correlation between toxicity
reduction  and   chemical  optimization  practices,
especially those implemented at Industry A.

However, in 1995 occasional chronic effluent toxicity
was again observed., Since early 1997, the effluent has
exhibited   consistent   chronic    toxicity
(NOEC=30-45%).  Current studies are  focusing on
treatment with polymer, which has shown to reduce
toxicity in bench-scale tests.  The City is also working
with the industries to implement additional chemical
optimization and waste minimization practices.  In
addition, construction is underway to extend the outfall
from a small creek to a river, which will afford greater
dilution. In 1998, the City will need to meet a revised
chronic toxicity limit of an NOEC of approximately
61%.

Summary
The RTA protocol was initially developed as part of
TRE research studies funded by the USEPA  Risk
Reduction Engineering Laboratory in Cincinnati, Ohio.
The   procedure  was  intended to  be  used by
municipalities as a tool for tracking sources  of toxicity
in sewer  collection systems;  however,  the  RTA
approach has  evolved to  suit other purposes.   In
addition to toxicity tracking (Collins et al., 1991), the
RTA protocol has  been used  to determine  the
compatibility  of  planned  discharges  to POTWs
(Engineering-Science,  Inc.,  1992,  1993)  and  to
establish compliance with toxicity-based pretreatment
limits (Morris et al., 1991).

Acknowledgments
Burlington Research, Inc. (Burlington, North Carolina)
and  Engineering-Science,  Inc.  (Fairfax,  Virginia),
conducted the TRE study. Burlington Research, Inc.,
performed the  TIE  tests, with  assistance  by  EA
Engineering, Science and Technology, Inc. (Sparks,
Maryland), and the evaluation of industrial chemical
usage. Engineering-Science, Inc., performed the RTA
study.  Burlington Research, Inc., and Engineering-
Science,  Inc.,  acknowledges Mr. Jerry Rothrock
(Director of Public Works, City  of Reidsville), and
Donald Waddell, LisaHaynes, Mitzy Webb, and James
Fain  (Hydro   Management  Services)   for  their
assistance.

The  material presented in this  appendix includes
copyrighted data presented in a technical paper for the
67th Annual Water Environment Federation Conference
(Botts et al., 1994).  WEF has granted permission to
include the data in this document.

References
Botts,  J.A., J.W.  Braswell,  E.G.  Sullivan,  W.L.
    Goodfellow, B.D. Sklar, and A.G. McDearmon.
    1987.    Toxicity  Reduction  Evaluation  at  the
    Patapsco  Wastewater  Treatment Plant. Risk
    Reduction Engineering Laboratory,  Cincinnati,
    Ohio. Cooperative Agreement No. CR812790-01-
    1. NTIS # PB 88-220 488/AS.

Botts,  J.A.,  L.B. Fillmore,  E.J.  Durban,  W.A.
    Goodfellow, T. Pereira, and D.F. Bishop.  1990.
    Evaluation of the Role of Diazinon in the Toxicity
    of a Municipal  Wastewater Treatment  Plant
    Effluent   Proceedings of the  Third National
    Pesticide   Conference,  November,   1990,
    Richmond, Virginia.

Botts, J.A.,  T.L. Morris, J.E.  Rumbo,  and C.H.
    Victoria-Rueda.   1992.  "Case Studies - Munici-
    palities." In Toxicity Reduction: Evaluation and
    Control. D.L. Ford, ed.  Technomic Publishing
    Co., Lancaster, Pennsylvania.
                                                103

-------
Bolts, J.A., T. Schmitt, E. Wilson, M. Collins, D.
    Waddell, R.  Diehl,  and L.  Ehrlich.   1994.
    Refractory Toxicity Assessment: An Alternative
    Approach  for  Chronic  Toxicity  Reduction
    Evaluations.  Technical paper  presented at the
    Annual Conference and Exposition of the Water
    Environment Federation, Chicago, Illinois. Paper
    #AC944404.

Burlington Research, Inc.  1993. Toxicity Reduction
    Evaluation: CityofReidsville, Phase I/II Report.
    Burlington, North Carolina.

Collins, M.A., T.L. Morris, J.A. Botts, T. Norberg-
    King, J.  Thompson,  and D.I. Mount.   1991.
    Chronic Toxicity Reduction Evaluation Study at
    the Bergen County Utilities Authority Wastewater
    Treatment Plant. Draft Report, 1991. USEPA,
    Risk  Reduction  Engineering   Laboratory,
    Cincinnati,  Ohio.   USEPA  Contract  No.
    68-03-3431.

Engineering-Science, Inc.   1992.   Results of the
    Refractory Toxicity Assessment for an Auto Parts
    Manufacturer. Prepared by Engineering-Science,
    Fairfax, Virginia.

Engineering-Science, Inc.   1993.   Results of the
    Refractory  Toxicity Assessment  for a  Store
    Fixtures Manufacturer. Prepared by Engineering-
    Science, Fairfax, Virginia.
    Welch, and J.A
Fillmore, L.B.,  T.L. Morris, T.L. Champlin, M.C.
.Botts. 1990. Toxicity Reduction
    Evaluation at the City of Fayetteville Cross Creek
    Wastewater Treatment Plant. Draft Report, 1990.
    USEPA, Risk Reduction Engineering Laboratory,
    Cincinnati, Ohio.  USEPA Contract No. 68-03-
    3431.        :

Morris, T.L., G. Fare, and J. Spadone. 1991.  Toxicity
    Reduction Evaluation at the  Linden  Rosette
    Sewerage Authority Wastewater Treatment Plant.
    Presented at the 64th  Water Pollution Control
    Federation Conference, Toronto, Ontario, October
    7-10, 1991. #AC91-055-001.

USEPA.   1989a. ; Toxicity  Reduction Evaluation
    Protocol for Municipal Wastewater  Treatment
    Plants.   EPA/600/2-88/062.  Risk  Reduction
    Engineering Laboratory, Cincinnati, Ohio.

USEPA.  1989b. Short-term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to Freshwater  Organisms.  EPA/600/4-
    89/001.  Environmental Monitoring and Support
    Laboratory, Cincinnati, Ohio.

USEPA.   1991b.  technical Support Document for
    Water  Quality-Based   Toxics  Control.
    EPA/505/2-90/001. Office of Water, Washington,
    D.c.         ;
                                               104

-------
                                        Appendix D

                                    TRE Case Study:
                           City of Durham, North Carolina
Abstract
TRE Goal:
Test Organism:
TRE Elements:

Toxicant Identified:
Toxicity Controls:
NOEC = 100%
C. dubia
Toxicity treatability
evaluation
TEE not performed
Proceeded  with planned
POTW upgrades
Summary
The City of Durham evaluated the expected toxicity
reduction to be achieved by planned upgrades of their
POTWs. Chronic toxicity reduction was evaluated
through the use of bench-scale simulations of  the
upgraded POTWs.  Results  indicated that the new
POTWs would reduce chronic toxicity to compliance
levels. Based on this evidence, the TRE was waived
until the new POTWs were online and effluent toxicity
reduction could be confirmed. The upgraded POTWs
became  operational  in  late   1994  and  effluent
monitoring results have shown no chronic toxicity after
consistent treatment performance was achieved.

Key Elements
The TRE study used a unique approach to evaluate
chronic toxicity reduction.  This approach may be
useful  to  other .municipalities  that have  TRE
requirements,  yet are planning upgrades  of their
POTWs. The key elements of interest in the City of
Durham study include the following:

1.  In cases where POTW staff  are planning to
    upgrade their POTWs, it may be more practical to
    evaluate the toxicity reduction to be achieved by
    the upgrade than to conduct TIE tests on  the
    existing POTW effluent.  The treatability approach
    is recommended when the upgrade is expected to
    improve toxicity reduction, such as nitrification
    treatment  for  ammonia   removal;  however,
    additional  evidence is needed to confirm the
    expectation.
2.   A  bench-scale  simulation  of the  upgraded
    treatment system can be  used to generate an
    effluent that is similar to the effluent expected for
    the new POTW.  Calibration tests should be
    performed to ensure  that the quality of the
    simulation effluent is similar to that of the planned
    POTW effluent.
3.   The treatability approach should  be  thoroughly
    described in the  TRE plan and  the regulatory
    authority should accept the plan prior to testing.

Introduction
Permit Requirements
Since 1987, NCDEM has required the City of Durham
to monitor the effluents of its four POTWs for chronic
toxicity using the North Carolina pass/fail test.  The
pass/fail test consists of 10 replicates of the effluent at
the critical instream waste concentration (IWC) and a
control. The effluent test concentrations corresponding
to  the IWC were 63.8% for the Eno River POTW,
100% for Lick Creek POTW,  98.7%  for Farrington
Road POTW, and 100% for Northside POTW.  The
test results indicated unacceptable levels of chronic
effluent toxicity for each of the four POTWs. In each
case, a statistically lower number of C. dubia young
were  observed in the  effluent  concentration as
compared to the control.

Based on  the effluent toxicity monitoring results,
NCDEM required the City of Durham to initiate a TRE
in January 1990. The goal of the TRE was to identify
methods for reducing chronic effluent  toxicity  to
acceptable levels at each of the treatment plants by
                                               105

-------
 January 1991. The City of Durham submitted a plan
 within 60 days that described a unique approach for
 implementing the TRE program.

 Instead of the traditional TRE approach of testing the
 existing effluents, the City proposed to evaluate the
 expected chronic toxicity reduction to be achieved by
 planned upgrades to the POTWs.  Toxicity reduction
 would be evaluated through the use of bench-scale
 simulations of the upgraded POTWs. This approach
 was favored over conventional TRE methods, such as
 TIE tests, because it was anticipated that the degree
 and nature of the effluent toxicity would change upon
 startup of the new treatment plants.

 Description  of the Treatment Plants
 In 1990, the City of Durham, North Carolina, had four
 POTWs: Eno River (2.5 mgd), Farrington Road (13
 mgd), Lick Creek (1.5 mgd), and Northside (10 mgd).
 In anticipation of the need for additional treatment
 capacity, the  City decided to close the Eno River and
 Lick Creek treatment plants and divert the flow to an
 expanded Northside plant. At the same time, NCDEM
 established draft permit limits for several parameters,
 including phosphorus. The new permit limits would
 require  advanced wastewater treatment; therefore, in
 addition to the Northside plant expansion, the City of
 Durham decided  to  upgrade  the  Northside   and
 Farrington  Road POTWs plants to include BNR
 treatment.

 During  the TRE, the Northside POTW comprised
 primary treatment followed by  trickling filters,  a
 single-stage   nitrification   process,  secondary
 clarification, and chlorine disinfection. The Northside
 POTW  upgrade involved building a new treatment
 system  in parallel with the existing system, which
 would treat the flow diverted from the former Eno
 River and Lick  Creek plants.   The new treatment
 system was planned to consist of primary clarifiers and
 a five-stage BNR process designed to remove nitrogen
 and phosphorus.  Effluents from the new and existing
 treatment systems will be combined, treated with
 aluminum sulfate (alum), passed through a filtration
 process, and disinfected by UV light prior to discharge
 to Ellerbe Creek.

The  Farrington  Road POTW was  planned to be
 converted from a two-stage nitrification process to a
 five-stage BNR  process similar in design  to  that
planned for  the  Northside  plant.   Final effluent
 treatment, like the Northside plant, will involve alum
 treatment, filtration, and UV disinfection.

 Wastewater Treatment Plant Simulations
 The new treatment processes for the Northside and
 Farrington Road POTWs were planned to be similar;
 therefore, the simulation designs were nearly identical.
 A batch mode  of operation instead of a continuous
 flow mode was selected to reduce study costs. Both
 simulations, as shown in Figure D-l, comprised a BNR
 process, followed by alum flocculation, settling, and
 effluent filtration.  Phosphorus and nitrogen removal
 was achieved in  the; BNR  process,  which involved
 treating the influent wastewater with activated sludge
 in five consecutive stages (anaerobic, anoxic, aerobic,
 anoxic, and aerobic).; The BNR process effluent was
 then treated with aliim and passed  through  a dual
 media filter column to remove additional phosphorus.
 Chronic toxicity tests using  C. dubia (USEPA, 1989)
 were performed on the final simulation effluents to
 evaluate the expected effluent quality of the full-scale
 treatment systems.  '

 Simulation of the Northside POTW involved treating
 the combined influents of the three POTWs scheduled
 for consolidation:  the  Eno River, Lick  Creek, and
 Northside plants.  The influents were combined in
 proportion  to  their respective  flow  rates.    The
 Farrington Road POTW influent was used directly in
 the simulation tests of  the Farrington Road facility.
 Each simulation influent was settled for approximately
 2 hours to simulate primary sedimentation.

 The activated sludge used in the simulations was
 collected from a municipal treatment plant that had a
 BNR process similar to the system planned for the City
 of Durham POTWs.; RAS was  collected from  the
 plant's clarifier return line  and mixed liquor solids
 were collected from the aeration basins.  RAS was
 mixed with the simulated primary effluent in the first
 BNR simulation  stage (anaerobic).   Phosphorus
 removal was enhanced in the subsequent BNR stages
 by replacing a portion of the RAS with nitrate rich,
 aeration basin sludge.  The nitrate was an essential
 source of oxygen for phosphorus removing bacteria in
the BNR anoxic stage.

Following biological treatment, the activated sludge
was settled and  the clarified effluent  was withdrawn
and treated with alum. Alum treatment involved flash
mixing and settling. The clarified supernatant was then
                                                106

-------
   RAS
   Primary Effluent
    15-Gallon
    Vessel
  Replace RAS
  with Nitrate
  Rich Sludge
     Air
               Anaerobic
                 Stage
           1st Anoxic
             Stage
                  Aerobic
                   Stage
2nd Anoxic
  Stage
    Air
             Alum
                                                    Supernatant
        2nd Aerobic
           Stage
Clarification
   Stage
Alum Treatment
     Stage
                                            4-Inch Diameter
                                            Column

                                            18 Inches
                                            Anthracite
                                                                                     12 Inches Sand
                                                                                         Final
                                                                                         Effluent
                                                                          Filtration Stage
Figure D-l. Flow diagram for waste-water treatment simulations.
passed through an anthracite/sand filter column, which
was operated in a constant headless mode.  Prior to
testing, the anthracite and sand in the filter columns
was distributed by backwashing the columns in  the
upflow direction using tap water. The filter columns
were  then  rinsed  with  deionized  water  in  the
operational (downflow) mode.

The general operating  conditions for the treatment
simulations are shown in Table D-l.  Some of  the
operating procedures for the simulations were modified
during calibration testing to achieve  the  desired
treatment performance.

Calibration of the Treatment Simulations
Prior to the toxicity evaluation, calibration tests were
performed to match  the  simulation performance to
expected performance for the upgraded POTWs. Also,
several  toxicity tests were performed during  the
calibration testing  to  verify that  the  simulation
materials and additives (i.e., activated sludge, alum)
would not introduce unexpected toxicity. The toxicity
tests followed USEPA procedures (1989) for C. dubia,
the test organism specified  in the City's discharge
permits.
                           The calibration testing involved varying the operation
                           of the simulations and monitoring the resulting effluent
                           quality.  The objective was to achieve a reduction in
                           influent concentrations of BOD5, COD, TKN, NH3-N,
                           NO3-N, TP, PO4-P, and TSS to levels approximating
                           those expected in the effluents of the planned treatment
                           plants.   Treatment performance was  evaluated by
                           varying the treatment times for each step.

                           The treatment times evaluated during the calibration
                           testing were 90,100, and 110% of the design HRT. A
                           summary of the conventional pollutant results for the
                           calibration study is shown in Tables D-2 and D-3.
                           Also shown are the monthly average permit limitations
                           and the design effluent characteristics for the planned
                           facilities.

                           Biological Treatment
                           All. BNR process simulations successfully achieved
                           carbonaceous BOD5 removal  and nitrification.  As
                           shown  in Table D-2, the batch biological process
                           removed BOD5, COD, and ammonia concentrations to
                           well below design effluent levels. TKN concentrations
                           in the simulation effluents also met the design levels.
                                                 107

-------
 Table P-l. Farrington Road and Northside Simulation Operating Conditions
Parameter
Farrihgton Road
POTW
Design*
Simulation
Northside
POTW
Design*
Simulation
Biological Treatment Step
Primary effluent volume
Eno River
Lick Creek
Northside
Average MLSS
Minimum DO
Anaerobic
1st Anoxic
1st Aerobic
2nd Anoxic
2nd Aerobic
Temperature (ฐC)
24.5 mgd



3,000 mg/L

Omg/L
Omg/L
2 mg/L
Omg/L
4 mg/L
10-26
100%



3,508 mg/L

<0.2 mg/L
<0.2 mg/L
2mg/L
<0.2 mg/L
4 mg/L
20-25

3.00 mgd
6.94 mgd
1:1.53 mgd
3,000 mg/L

!Omg/L
0 mg/L
2 mg/L
Omg/L
4 mg/L
12-29

14.0%f
32.3%f
53.7%t
3,481 mg/L

<0.2 mg/L
<0.2 mg/L
2 mg/L
<0.2 mg/L
4 mg/L
20-25
Alum/Filtration Treatment Steps
Alum dose after biological
treatment
Depth of anthracite/sand in filter
Constant headless level in filter
Average filtration rate
10 mg/L
878"
4ft
2.4 gpm/ft2
20 mg/L
878"
4ft
2.4 gpm/ft2 $
.5 mg/L
18712"
2-8 ft
4 gpm/ft2
10 mg/L ง
18712"
4ft
4.1 gpm/ft2 #
 * Source: Hazen and Sawyer; R.L. Taylor, personal communication to J.A. Botts, Design Information for the Treatment
   Plant Expansions. December 10,1990, Raleigh, North Carolina.                   ;
 t Percent of total simulation influent volume.                                    '
 $ Filtration rate was 4.2 gpm/ft2 for April 4-5 simulation.                          ;
 ง Alum dosage increased to 20 mg/L for April 10-11 simulation.
 # Filtration rate was 7.1 gpm/ft2 for April 4—5 simulation.
The BNR simulations did not consistently achieve the
effluent permit levels for phosphorus (Table D-2). No
phosphorus removal was observed in the April 4-5 test.
For subsequent tests, the percentage of aeration basin
sludge added to the anoxic  stage was  increased to
stimulate phosphorus  removal.  This  modification!
resulted  in a decrease in phosphorus to near design
levels in the April 10-11 test.  As shown in Table D-3,
phosphorus was initially released by the bacteria in the
anaerobic stage, which is common in BNR systems.
However, unlike the April 4-5 test, the phosphorus
was re-assimilated in the anoxic and aerobic stages as
would be expected.  These results demonstrated that
phosphorus removal can  be achieved in the batch
simulation  tests. The lack of phosphorus removal in
the April 18-19 test appeared to be related to the poor
quality of the activated sludge on the day of testing.

The BNR simulations also did not achieve consistent
denitrification (Table D-2). The Northside simulation
test  on  April  10-11  reduced  nitrate  to  a level
(1.7 mg/L)  close to the design effluent concentration
(1.0 mg/L). All other simulation tests achieved only
slight nitrate removal. The lack of nitrate removal in
                                                 108

-------
Table D-2. Comparison of Calibration Test Results to Permit Limitations and Design Criteria (mg/L)
Parameter
Monthly Average*
Effluent Permit
Limits
Designf Effluent
Characteristics
Calibration Results
Apr 4-5
Apr 10-11
Apr 18-19
Northside POTW
BODS
COD
TSS
TKN
NH3-N
NO3-N
TP
24.0/12.0 $
NA*
30
NA
16.0/8.0 ง
NA
2
5
51
10
1.5
0.5
4.75
0.5
1
21
0
1.5
0.2
5.9
6
1
17
5
1.5
0.1
1.7
0.8
1
26
0
0.9
0.05
12.4
6
Farrington Road POTW
BOD5
COD
TSS
TKN
NH3-N
NO3-N
TP
10.0/7.0 $
NA
30
NA
4.0/2.0 $
NA
2
5
45
10
1.5
0.5
1
0.5
1
23
1
1.9
0.1
7.1
7.4
1
26
5
1
0.1
6.5
0.6
1
23
2
0.8
0.1
14.7
7.1
  * Values are interim limits for the period beginning January 1, 1991, and lasting until 3 months after construction completion.
  t Source:  Hazen and Sawyer, R.L. Taylor, personal communication, to J.A. Bolts, Design Information for the Treatment
   Plant Expansions December 10,1990, Raleigh, North Carolina.
  $ Winter and Summer limits, respectively.
  ง No limit established in permit.
 the Farrington Road simulation may have been due to
 the  short  anoxic treatment time  (approximately 3
 hours) as compared to the Northside simulation (more
 than 4 hours). The simulation procedure was modified
 to increase the anoxic treatment time for the Farrington
 Road simulation to attempt to achieve denitrification
 during the effluent toxicity evaluation.

 The  toxicity test  results indicated that the  RAS
 supernatant used in simulation testing was not acutely
 toxic (LC50 z 100%). Therefore, the activated sludge
 was  not  expected  to  cause  an  acute  toxicity
 interference in the simulation tests.

 Alum Treatment
 As  shown  in Table D-3, only a  slight removal of
 phosphorus was observed in the alum treatment step.
 Solids flocculation did not occur at the designed alum
dosages (10 mg/L for Farrington Road POTW and
5 mg/L for Northside POTW).  Alum dosages were
Table D-3.  Total Phosphorus Results  (mg/L)  for the
Calibration Tests Conducted on April 10-11,1990
Wastewater/Sludge
Influent
RAS
Basin sludge
Biological treatment
Anaerobic effluent
1st aerobic effluent
2nd aerobic effluent
(Clarifier effluent)
Alum treatment supernatant
Farrington
Road
Simulation
5.49
13.5
4.13
32.2
2.33
1.48
1.06
Northside
Simulation
3.95
13.5
4.13
20.7
3.05
1.78
1.55
                                                   109

-------
increased two-fold; however, no additional phosphorus
removal was achieved.

The effect of alum on effluent toxicity was evaluated
by comparing the toxicity of the wastewater before and
after alum treatment. The results show that the alum
did not add acute toxicity to the wastewater (i.e., LC50
>100% before and after alum addition).

Filtration Treatment
The filter columns were very efficient in removing
suspended solids (Table D-2).  As a result, nutrients
and  COD associated with the solids were further
reduced. Total phosphorus concentrations decreased
by nearly half after filtration (Table D-3).

The deionized water rinsates from the filter columns
were analyzed for toxicity prior to testing. The results
indicated that the filter media would not add acute
toxicity to the simulation effluent (rinsate LC50
Discussion of Calibration Results
The calibration results indicated that bench-scale tests
could effectively simulate the effluent quality expected
for the new POTWs. Pollutant removal was similar
whether the simulations were tested at 90,  100,  or
110%  of  the  design HRT.  BOD5,  COD, TKN,
ammonia, and TSS were consistently reduced to levels
expected to be achieved by the planned facilities.
Although nitrate and phosphorus were  not treated to
design effluent levels, no adverse effects on toxicity
treatment in the simulations were anticipated.  The
calibration results also indicated that the simulation
materials would not contribute artifactual toxicity.

Toxicity Treatment Evaluation
Tests of the calibrated simulations were performed to
determine if the new POTWs would eliminate chronic
toxicity. The operating parameters for the simulations
were based on the design HRT treatment condition
(100%). An exception was the treatment time for the
second anoxic treatment stage of the Farrington Road
simulation,  which   was  increased   to  stimulate
denitrification. In addition, the alum dosages for both
simulations were increased to enhance the flocculation
necessary for phosphorus removal.

The treatment plant simulations were implemented on
two occasions. Performance criteria were applied  to
ensure that the effluent  quality was  sufficient for
toxicity evaluation.    These  criteria,  shown   in
Table D-4, were based on the treatment performance
that was consistently achieved in the calibration tests.

Treatment Performance Results
A summary of the conventional pollutant results for the
simulation effluents is shown  in Table  D-4.   The
results show that the simulations consistently achieved
the design effluent concentrations for BOD5, COD,
TSS,  and ammonia.  Effluent  TKN concentrations
were within the simulation performance criterion of
5  mg/L.   The  effluent concentrations  of  total
phosphorus and nitrate also were within the simulation
performance criteria levels.  Overall, the simulation
effluents were judged to be  suitable for  toxicity
analysis based on the simulation performance criteria.

Toxicity Evaluation Results
Results of toxicity tests, presented in Table D-5, show
that the simulation effluents were  not acutely toxic to
C. dubia (48-hour LCSOs 100% effluent).  Chronic
toxicity results show that the simulation effluents did
not inhibit C. dubia reproduction (NOEC of 100%
effluent).  Only the effluent of the Farrington  Road
simulation on May 29-30,1990, adversely affected C.
dubia survival (NOEC = 75% effluent). The chronic
toxicity of this effluent was due to significant mortality
in the 100% effluent concentration.

Sulfide  was detected in the May 29-30 Farrington
Road simulation effluent at a concentration that may be
chronically toxic to C. dubia (1.6 mg/L). The sulfide
NOEC for D. magna at pH 7.6-7.8 is reported to be
1.0 mg/L (USEPA, 1990). Although the toxicity of
sulfide to C. dubia is unknown,  the sensitivities of D.
magna and C. dubia to many classes of toxicants are
similar (Mount and Norberg, 1984). The pH values of
the Farrington Road simulation effluent and the value
used for the reported test also were similar (i.e., 7.85
versus 7.6 to 7.8); therefore, the potential toxicity of
sulfide in the simulation sample should be comparable
to that of the reported test (Note: the concentration of
hydrogen  sulfide, the most toxic form  of  sulfide,
increases when pH decreases). Based on this evidence,
the chronic toxicity  observed  in the May 29-30
Farrington Road simulation effluent may be related to
sulfide.            i

Discussion
The TRE study was completed within the  1-year time
frame specified by NCDEM. The  results of this study
indicated that the addition of new  BNR and filtration
treatment processes at the City of Durham POTWs
                                                110

-------
- Table D-4. Comparison of Simulation Test Results to Performance Criteria
Parameter
• Simulation
Performance Criteria
'(mg/L)*
Simulation Effluent Results (mg/L)
May 29-30
June 6-7
Northside POTW
BOD5
COD
TSS
TKN
NH3-N
NO3-N
TP
5
51
10
5
0.5
15
8
1
22
3
2
0.1
5.5
1.2
1
21
2
- NAf
0.1
11.3
3.2
Farrington Road POTW
BOD5
COD
TSS
TKN
NH3-N
NO3-N
TP
5
45
10
5
0.5
15
8
1
22
4
2.3
0.1
5.2
1.5
1
22
1
NA
0.1
9.3
3.8
   *  Simulation performance criteria based on calibration results and design effluent levels (Hazen and Sawyer; R.L. Taylor,
     personal communication, to J.A. Botts, Design Information for the Treatment Plant Expansions. December 10,1990,
     Raleigh, North Carolina).
   f  NA = not available.
 Table D-5. Toxicity of Simulation Effluents to C. dubia*
Date
May 29-30, 1990
June 6-7, 1990
Simulation
Farrington Road
Northside
Farrington Road
Northside
48-hour LC50
(%Effluent)
100
>100
>100 .
>100
NOECf ' '
(%Effluent)
75 ง
100
100
100
LOECI
(%Effluent)
100 ง
>100
>100
>100
   * 7-day chronic toxicity test (USEPA Method 1002.0) according to USEPA (1989).
   t NOEC for Northside is based on survival and reproduction. Results for Farrington R'oad are based on survival.
   t LOEC for Northside is based on survival and reproduction. Results for Farrington Road are based on survival.
   ง Denotes statistically significant inhibition of survival.
  would reduce chronic  effluent toxicity to  levels
  required under the North Carolina discharge permit.
  Sulfide, a potential effluent toxicant, was not expected
  to be a problem because the final effluents of the new
  treatment plants are aerated to meet instream DO
  standards.    The  sulfide  should be  volatilized or
  oxidized in this aeration step.
The POTW upgrades were implemented beginning in
November  1994.   Results of effluent  monitoring
through  the second quarter of 1997  show  that the
POTWs  are in compliance with the chronic toxicity
limits. The limits were revised to NOECs 2:90% for
both plants.  One test failure was observed in January
1995; however, this result may have been related to the
                                                      111

-------
start-up of the new treatment processes. Since then,
the City has passed all quarterly tests at both POTWs.

Bench-scale batch tests were  successfully used  to
simulate the treatment processes planned for the new
POTWs, including the BNR treatment process.  In
addition  to carbon removal  and  nitrification, the;
simulations  achieved  phosphorus removal to  near
permit levels.  Although nitrate was not reduced  to
permit levels,  the observed concentrations did not
cause chronic toxicity.

The study  findings suggest  an  alternative  TRE
approach is appropriate in cases where POTW staff is
planning upgrades or improvements to their WWTPs.
Toxicity reduction can be evaluated by conducting
bench-scale batch simulations of the planned upgrades.
This testing can be used to determine the potential for
compliance with discharge limits for toxicity. If non-
compliance  is  anticipated,  further testing can be
performed to evaluate the  additional  improvements
necessary for toxicity reduction.  In cases where the
conclusions of a bench-scale toxicity  evaluation are
uncertain, pilot-scale tests may be warranted.
                                               i
Acknowledgments
The City of Durham TRE study was conducted by
Engineering-Science, Inc., under contract to Hazen and
Sawyer, P.C., and the City of Durham.  The assistance
of A.T. Rolan, W.W. Sun, Vicki  Westbrook, and
Susan Turbak (City of Durham); Linda Forehand,
Tim Morris, Jim  McReynolds, and Tory Champlin
(Engineering-Science,   Inc.);  and  William  L.
Goodfellow   (EA   Engineering,  Science   and
Technology, Inc.) are acknowledged.

The  material presented  in  this  appendix  includes
copyrighted  information  published  in  text by
Technomic  Publishing  Company,  Lancaster,
Pennsylvania  (Botts  et  al,  1992).   Technomic
Publishing Company has granted permission to include
the information in this document.

References
Botts,  J.A., T.L. Morris, J.E.  Rumbo, and C.H.
    Victoria-Rueda.  1992. Case Histories-Munci-
    palities. In Toxicity Reduction:  Evaluation and
    Control.  D.L. Ford, ed. Technomic Publishing
    Co., Lancaster, Pennsylvania.

Engineering-Science, Inc.  1989.  Test Plan for the
    City  of Durham. Toxicity Reduction Evaluation
    Project. Prepared for the City of Durham, North
    Carolina.

Engineering-Science, Inc. 1990. Results of the Devel-
    opment and Implementation  of  Waste-water
    Treatment Plant Simulations  for the City of
    Durham TRE. Prepared for the City of Durham,
    North Carolina.  .

Mount, D.I., and TJ. Norberg.  1984. A Seven-Day
    Life Cycle Cladoceran Test.  Environ.  Toxicol.
    Chem. 3:425-34.

USEPA.  1989. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Water to Freshwater Organisms. Second Edition.
    EPA/600/4-89/001.   Office of Research  and
    Development, Cincinnati, Ohio.

USEPA.  1990.  Aquatic Information Retrieval Toxi-
    city  Data  Base,    Office  of  Research  and
    Development, National Health and Environmental
    Effects  Research  Laboratory,  Mid-Continent
    Ecology Division, Duluth, Minnesota.
                                               112

-------
                                        Appendix E

                                     TRE Case Study:
                     Michigan City Sanitary District, Indiana
Abstract
TRE Goal:         LC50 ;> 100%, NOEC ;>62%
Test Organisms:    C. dubia and P. promelas
TRE Elements:     TIE
Toxicant Identified: Metals
Toxicity Controls:   Pretreatment requirements

Summary
Acute and chronic TIE studies indicated that metals
were the primary cause  of effluent toxicity.  An
industrial user was identified as a major source of
metals  loadings to  the  POTW. The POTW staff
required the industrial user to discontinue a cadmium
plating operation and, as a result, the POTW effluent
has achieved compliance with the acute and chronic
toxicity limits (MCSD, 1993).

Key Elements
1.  Less expensive acute TIE procedures can be used
    in  lieu of  chronic  TIE  procedures to help
    characterize the causes of chronic effluent toxicity.
    However, chronic TIE testing is needed to confirm
    the acute TIE results.
2.  CIS   SPE  can  remove  toxicity  caused  by
    compounds  other  than  non-polar  organic
    compounds. In this study, CIS SPE treatment
    removed toxicity caused by metals. These results
    demonstrate the importance of needing to recover
    toxicity from  the CIS  SPE column   before
    concluding that non-polar organic compounds are
    causing effluent toxicity.
3.  TIE  Phase I  data may  provide  sufficient
    information to proceed  to  the selection  of
    pretreatment  controls  for  toxicity  reduction.
    Although specific toxic metals were not identified
    in this study, evidence of metals toxicity  was
    successfully used to set pretreatment requirements.
 Introduction
 Permit Requirements
 The NPDES permit for the Michigan City Wastewater
 Treatment Plant (MCWTP) requires acute and chronic
 toxicity monitoring using C. dubia and P. promelas.
 The  permit  specifies  that the  effluent  must not
 demonstrate  chronic  effluent toxicity at effluent
 concentrations of 62% or less (< 1.6 TUc) and that the
 effluent must not be acutely toxic (e.g., LC50 ^ 100%,
 < 1.0 TUa). Based on evidence of unacceptable acute
 and chronic toxicity, Michigan City was  required to
 perform a TRE.  The Michigan City Sanitary District
 submitted a TRE plan and initiated TIE testing. The
 objective of the TIE was to characterize, identify, and
 confirm the  causes  of acute  and chronic effluent
 toxicity so that an appropriate  toxicity reduction
 strategy could be developed and implemented.

 Description of Treatment Plant
 The MCWTP comprises an activated sludge process
 with single-stage  nitrification and advanced  waste
 treatment of the secondary effluent.  The facility is
 designed for an average wastewater flow of 12-million
 gallons per day  (mgd) and 96.7% removal of BOD5
 and  96% removal of suspended solids.   Monthly
 average effluent limits  for ammonia are 2 mg/L in
 summer and 6 mg/L in the winter. Influent phosphorus
 is reduced with an iron salt added at the aeration tanks.
 Additional phosphorus and suspended solids removal
 is accomplished by sand filtration of the secondary
 effluent.  Total phosphorus is reduced by 80%, which
 results in effluent concentrations of less than 1 mg/L.
 Post aeration equipment is provided to increase the
 effluent DO concentration prior to discharge to Trail
 Creek. During the months of June through September
 (which coincides with the seeding of Trail Creek with
..•smoltSia'ndilater.fishiDnignation up;Trail Creek), a:pure
                                               113

-------
 oxygen system supersaturates the plant effluent to a
 DO concentration in excess of 13.0 mg/L.

 Toxicity Identification Evaluation
 Initial Toxicity Characterization
 When both acute and chronic toxicity requirements
 must be met, POTW staff must decide whether to use
 acute or  chronic TEE  procedures to determine the
 effluent toxicants. Acute TIE procedures can be used
 to provide information about the causes of chronic
 toxicity and   may  be   preferred because  they are
 simpler and less costly than chronic TIE tests. Follow-
 up confirmation tests can be performed using chronic
 TIE procedures to determine if additional toxicants are
 contributing to chronic toxicity. If an effluent exhibits
 marginal and intermittent acute toxicity, it may not be
 possible to identify the causes of effluent toxicity using
 acute  TIE procedures.   In  this  case, chronic TIE
 procedures should be used.

 The initial TIE  work  at the MCWTP  focused on
 characterizing  the causes of acute  effluent toxicity
 because previous testing indicated  that the effluent
 exhibited consistent acute toxicity. C. dubia were used
 as the test organism based on previous tests showing it
 to be more sensitive to  the MCWTP effluent than P.
promelas.
The toxicity characterization tests conducted during the
first quarter of the TIE program included the following
effluent manipulations:

  • Pressure filtration (1.0 um filter).
  • Submicron  filtration (0.22  um filter) following
    pressure filtration (performed on one sample)
  • Aeration.
  • C18 SPE following filtration.
  • Cation  resin treatment following filtration/CIS
    SPE treatment.
  • Anion resin treatment following filtration/CIS
    SPE treatment.  :

As shown in Table E-l, the four effluent  samples
characterized from April through  June  1991 were
consistently toxic and the magnitude of  toxicity was
similar in each  sample (1.5 to 2.5 TUa).   Slight
reductions in toxicity occurred following filtration and
aeration and acute toxicity was completely removed by
the CIS SPE column. Toxicity removal by the cation
and anion resins could not be determined because the
sample was first passed through the CIS SPE  column,
which removed  all of the toxicity.  In retrospect, it
would have been preferable to treat the samples with
the ion exchange resins following filtration rather than
after CIS SPE treatment. Relatively nonpolar organic
compounds are preferentially adsorbed onto  the CIS
SPE column; therefore, toxicity removal by the CIS
Table E-l. Acute Toxicity Characterization Test Results from April 1991 Through June 1991
Characterization Test
Baseline (whole effluent)
Filtration
Aeration co
Post CIS SPE t
Cation exchange T
Anion exchange ฃ
C dubia LC50 (TUa)*
4/18/91
42 (2.4)
51 (2.0) f
40 (2.5)
>100 (0.0)
>100 (0.0)
>100 (0.0)
5/16/91
40 (2.5)
79 (1.3) t
62(1.6)
>100 (0.0)
>100 (0.4)
>100 (0.0)
6/5/91
46 (2.2)
54(1.9)f
51 (2.0)
>100 (0.0)
>100 (0.0)
>100 (0.2)
6/19/91
67 (1.5)
ง
ง
>100 (0.0) #
ง
ง
 * C. dubia 48-hour LC50 values expressed as percent effluent with acute TUs (100/LC50) in parentheses.
 t Effluent first pressure filtered through a Gelman A/E glass fiber filter (1.0 um).
 t Effluent first pressure filtered through a Gelman A/E glass fiber filter (1.0 um), followed by filtration through a Micro
   Separation, Inc., 0.22 um nylon filter.
 ง Characterization manipulation not conducted.
 # Fine stream of air bubbles passed through an effluent sample placed in a graduated cylinder.
 o> Effluent sequentially pressure filtered (1.0 um) and passed over a CIS SPE column.
 i Effluent passed directly over a CIS SPE column.                                '
 t Effluent passed over a Bio-Rex MSZ 50 cation resin after pressure filtration and CIS SPE treatment.
 € Effluent passed over a Bio-Rex MSZ 1 anion resin after pressure filtration and CIS SPE treatment.
                                                  114

-------
SPE treatment during the initial characterization tests
suggested  that  non-polar  or  semi-polar  organic
compounds were causes of effluent toxicity.

Evaluation of Toxicity Removed by CIS SPE
The CIS SPE column can remove toxicants other
than  non-polar  organic  compounds,  including
organometallic   complexes,  certain  metal   ions,
surfactants, and some high molecular weight organic
compounds.   Accordingly, additional  tests  were
performed from July through October 1991 to obtain
information about the types of compounds removed by
the CIS SPE treatment.  In an attempt to recover
toxicity from the CIS SPE column, sequential elutions
were performed with methanol, methylene chloride, 3N
hydrochloric acid, and 9N sodium hydroxide. Metals
were  evaluated  as possible  causes  of  toxicity
concurrently with the CIS SPE tests. Metals toxicity
was investigated by adding EDTA to whole effluent
samples and testing for acute toxicity. EDTA forms
complexes with many toxic metals and, when added at
appropriate concentrations, can render metals non-
toxic.

Results of the CIS SPE column and EDTA tests are
summarized in Table E-2. In contrast to previous tests,
the acute toxicity of the whole effluent from August
through October 1991 was variable and intermittent
(Table E-2). Four of the seven effluent samples were
not acutely toxic. The three acutely toxic samples were
rendered  non-toxic by  the CIS  SPE treatment;
however, toxicity was not recovered by eluting the CIS
SPE columns with methanol, methylene chloride, 3N'
hydrochloric acid, or 9N sodium hydroxide. Toxicity
could not  be successfully eluted from CIS  SPE
columns  using   conventional  organic  extraction
techniques; therefore, it was concluded that the toxicity
removed by the column was not caused by typical non-
polar or semi-polar organic compounds.

Addition of EDTA to the three acutely toxic samples
eliminated acute toxicity, suggesting that  toxicity was
caused by metals.  The EDTA results provide evidence
that the toxicity removed by the CIS SPE column was
not  caused  by  non-polar  or semi-polar organic
compounds.  Instead,  it  indicated that metals  or
organometallic complexes were removed in the CIS
SPE column tests.  These results demonstrate the
importance of needing to recover toxicity from the C18
SPE column before concluding that non-polar organic
compounds are a cause of effluent toxicity.

Evaluation of Metal Toxicity
Additional testing was performed to evaluate metals as
a cause of chronic effluent toxicity to C. dubia.
Chronic tests were used to  help avoid problems
associated  with   the  intermittent acute  toxicity;
however, acute toxicity endpoints (e.g., 48-hour LC50)
were also  obtained from the chronic  tests.  During
October 1991 through  January 1992,  7-day  static
renewal C. dubia survival and reproduction tests were
performed on whole effluent samples and  whole
 Table E-2. Toxicity Characterization Test Results from July 1991 Through October 9,1991
1 <
Sample Date
7/10/91
7/24/91
8/07/91
8/22/91
9/11/91
9/25/91
10/09/91
C,dซWaLC50(TUa)*
Baseline
(Final Effluent)
>100 (0.2)
>100 (0.0)
61 (1.6)
52 (1.9)
>100 (0.4)
>100 (0.2)
<100 (>1) #
Post CIS SPE t
ง
>100 (0.2)
>100 (0.0)
>100 (0.4)
ง
>100 (0.0)
ง
EDTA*
ง
>100 (0.0)
>100 (0.0)
>100 (0.0)
ง
>100 (0.0)
>100 (0.0) co
  * C. dubia 48-hour LC50 values expressed as percent with TUs (100/LC50) in parentheses.
  t Effluent passed over a C18 SPE column.
  $ EDTA was added to the final effluent at a concentration of 186 mg/L.
  ง Characterization manipulation was not conducted.
  # Test conducted only in 100% effluent; as a result, LC50 and TUa values could not be calculated.
  co EDTA concentration in the 10/09/91 sample was 18.6 mg/L.
                                                 115

-------
 effluent samples with EDTA added.  As shown in
 Table E-3, three of the five samples exhibited acute
 toxicity and  four of the five were chronically toxic.
 The 48-hour LC50 values for all of the EDTA treated
 samples were greater than 100% effluent.  EDTA
 addition  also  eliminated chronic  toxicity in  two
 samples and reduced chronic toxicity in a third sample.
 These results provided additional evidence that metals
 cause acute effluent toxicity, and also suggested that
 metals were  a primary  cause  of  chronic effluent
 toxicity.

 The  correlation approach and spiking  approach
 described by USEPA (1989a) were used  to confirm
 that metals  were  causing effluent toxicity.    The
 correlation  approach is  intended  to evaluate  the
 relationship between the  concentration of suspected
 toxicants and effluent toxicity.  Toxicity and metals
 data (aluminum, Cd, Cu, Ni, and Zn) for six effluent
 samples were compared by correlation analysis.  All
 metals were measured as total metals.

 Linear regression analysis indicated a good correlation
 (regression  coefficient  of  0.72) between  effluent
 toxicity and  effluent Cd concentrations.  However,
 when data from May 1991 through December 1992
 were pooled with the data set, the correlation between
 effluent toxicity and effluent Cd concentrations  was
 not statistically significant. A comparison of the mean
 Cd concentrations from samples collected during a
 toxic period (May 1991 to December 1991), and those
 taken  during a  non-toxic  period  (May 1992 to
 December 1992) indicated a trend.   The mean Cd
 concentration was 4.1 |jg/L during the toxic period and
 0.47  ug/L during the non-toxic period.  These data
 provide evidence that Cd was contributing to effluent
 toxicity.  No significant correlation  was  observed
 between effluent toxicity and the concentration of the
 other metals or the sum of all the metals.

 The objective of the spiking approach was to determine
 whether an increase in the concentration of a suspected
 toxicant would cause  a proportional  increase in
 toxicity.   Chronic  C.  dubia toxicity  tests were
 performed on three chronically toxic effluent samples
 both with and without added Cd, Cu, Ni, and Zn. The
 metals  were added  in  nominal   concentrations
 approximating those typically found in the MCWTP
 effluent. The results indicated that effluent toxicity did
 not consistently increase when the metals were spiked
 individually or in combination.  Therefore, the results
 of the spiking tests  did not confirm that Cd or other
 metals were contributing to effluent toxicity.

 Toxicity Control Evaluation and
 Implementation
 Although the TIE did  not conclusively  identify the
 specific causes  of effluent  toxicity,  the weight of
 evidence indicated that effluent toxicity was caused by
 metals.   As a  result, Michigan  City  investigated
 possible sources of metals in the collection system.
 Pretreatment program data indicated that a cadmium
plating  facility in the MCWTP service area was
consistently out of  compliance with pretreatment
limitations  for metals.    Based on  the persistent
pretreatment permit violations,  the cadmium plating
Table E-3. Acute and Chronic Toxicity of MCWTP's Effluent (with and without added EDTA) from October 1991 Through
January 1992
Sample Date
10/30/91
11/14/91
12/04/91
12/18/91
01/08/92
Final Effluent
Acute
LC50 (TUa) t
73 (1.4)
>100 (0.0)
>100 (0.0)
84 (1.2)
60 (1.7)
Chronic
NOECt
50
62
100
<50
<50
Final Effluent with EDTA Added *
Acute
LC50 (TUa) t
>100 (0.0)
>100 (0.0)
>100 (0.0)
>100 (0.0)
>100 (0.0)
Chronic
NOEC $ . . ; v:
100
62
100
100
50
 * EDTA concentration in the 10/30/91 and 11/14/91 tests was 5 mg/L. EDTA concentration in the 12/04/91,12/18/91 and
   01/08/92 tests was 10 mg/L.                                              :
 t C. dubia 48-hour LC50 values expressed as percent effluent with acute TUas (100/LC50) in parentheses.
 t Reproduction NOEC values expressed as percent effluent calculated from 7-day static-renewal chronic tests with C. dubia.
                                                116

-------
company was issued a consent decree to terminate their
cadmium plating operation.  The  cadmium plating
operation was shut down in April 1992.

The impact of the shutdown on effluent toxicity was
evaluated by performing 4-day modified chronic C.
dubia tests on whole effluent samples at approximately
2-week intervals from May through September 1992
(total of nine tests). The 4-day modified chronic tests
consisted of four concentrations and a control, five
replicate test chambers per concentration, and the tests
were initiated with 3-day old C. dubia. This modified
approach has been demonstrated to produce results that
are comparable to the 7-day test (Masters et al., 1991).
The results of these tests showed that acute and chronic
effluent toxicity to C. dubia had been eliminated.

Discussion
Subsequent chronic  testing with  C.  dubia and P.
promelas using compliance monitoring procedures
(USEPA, 1989b) confirmed the reduction in effluent
toxicity following shutdown of the cadmium plating
operation.   The  acute  and chronic  toxicity of the
MCWTP effluent from inception of the TRE through
December 1992 is summarized in Figure E-l.  The
correlation between  the cadmium plating operation
shutdown and improved effluent toxicity is clearly
evident. Based on the improved effluent toxicity, the
TRE was terminated and semiannual acute and chronic
toxicity compliance monitoring was initiated.

However,  starting  in  August  1996  significant
reproductive effects were observed in 100% effluent as
compared to the test control. Subsequent TIE testing
was inconclusive because  effluent samples  were
nontoxic.  Michigan City has submitted a letter to the
Indiana Department  of Environmental Management
(IDEM) requesting changes in the effluent monitoring
program.  The requested changes include the use of
reconstituted laboratory water as dilution water in lieu
of receiving water to minimize potential contamination
and reducing the frequency of monitoring if no toxicity
is observed in three consecutive tests.  As of October
1997, a decision from IDEM was still pending.

Acknowledgments
This  TIE  effort was conducted by  Great Lakes
Environmental Center, Traverse City, Michigan, under
contract to the Michigan City Sanitary District.

References
Masters, J.A., M.A. Lewis, D.H. Davidson, and R.D.
      3.0 T
                                        Chronic Toxicity   —•—Acute Toxicity
                                                         Cadmium plating operation discontinued
                                           Toxicity Test Date


          Note 1. Recorded toxicity is an artifact of the statistical analysis.
          Note 2. TUa's and chronic TUs are based on 48-hour LC50 and NOEC values for C.dubia, respectively.
 Figure E-l. Acute and chronic effluent toxicity: 1991 through 1992.

                                                  117

-------
    Bruce. 1991. "Validation of a Four-Day C. dubia
    Toxicity  Test  and  Statistical  Considerations
    in Data Analysis" Environ. Toxicol. Chem.  10:
    47-55.

Michigan City Sanitary District (MCSD). 1993. Draft
    Final Report on Toxicity Identification/Reduction
    Evaluation  for  Michigan  City  Waste-water
    Treatment Plant.   Prepared  by  Great  Lakes
    Environmental Center, Traverse City, Michigan.

USEPA.  1989a.   Methods for Aquatic  Toxicity
    Identification  Evaluations:  Phase III Toxicity
    Confirmation  Procedures.   EPA/600/3-88/036.
    Office  of Research  and Development, Duluth,
    Minnesota.    ;

USEPA.  1989b. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters  to  Freshwater  Organisms.   Second
    Edition. EPA/600/4-89/001. Office of Research
    and Development, Cincinnati, Ohio.
                                               118

-------
                                        Appendix F

                                     TRE Case Study:
         Central Contra Cost Sanitary District, Martinez, California,
                   and Other San Francisco Bay Area POTWs
Abstract
TRE Goal:

Test Organism:
TRE Elements:

Toxicants Identified:
Toxicity Controls:
No significant acute toxicity
at 100% effluent
C. dubia
TIE  and  source identifi-
cation
Diazinon and chlorpyrifos
Multi-faceted public aware-
ness  program;   ongoing
program to identify  and
control  sources;  ongoing
effort to  identify  POTW
processes  and  operations
that   effectively  remove
organophosphate
insecticides.
Summary
Acute toxicity to C. dubia was consistently detected in
a POTW effluent. Application of Phase I, H, and m
TIE procedures showed that the toxicity was caused by
diazinon and one or more additional organophosphate
insecticides.    Follow-up  studies,  which  required
development of more sensitive analytical methods,
showed that chlorpyrifos was present at levels that
exceeded the NOEC in all effluent samples that were
toxic to C. dubia. Influent and effluent monitoring
studies of San Francisco Bay Area POTWs identified
large differences in both  influent loading and removal
of the two insecticides between the POTWs.  All the
POTWs sampled achieved  substantial  removal of
diazinon and chlorpyrifos from influent wastewater.
Higher removal of both insecticides were generally
associated with POTWs that had filtration treatment,
extended mean cell residence times, chlorine contact
times,  and/or  long  retention  in ponds.   Source
identification studies showed that the majority of the
influent mass loading of the two insecticides was from
residential sources. A multi-faceted outreach program
was  initiated  within the  POTW  service  area.
Monitoring  of effluent  toxicity  and  insecticide
concentrations to assess the effectiveness of the public
outreach program is on-going.

Key Elements
1.  The organophosphate insecticides, diazinon, and
    more recently, chlorpyrifos, have been implicated
    as  causes  of  toxicity to  C.  dubia  in  POTW
    effluents.
2.  Published TIE procedures are available to identify
    organophosphate  insecticide  toxicity (USEPA
    1991,1992,1993a, 1993b, 1996). Application of
    new methods and procedures assisted in providing
    a more  quantitative assessment of the  role of
    diazinon and chlorpyrifos in effluent toxicity.
3.  Source   identification  studies  at  the  CCCSD
    demonstrated that the majority of the diazinon and
    chlorpyrifos influent loading was from residential
    sources.
4.  Regional influent and effluent monitoring studies
    demonstrated patterns  in influent diazinon and
    chlorpyrifos loadings  at the  CCCSD, which
    suggest there were demographic differences in use
    and disposal  practices  for  organophosphate
    insecticides.
5.  A multi-faceted public outreach  program was
    implemented in the POTW service area.  The
    effectiveness of the program is being assessed by
    frequent  measurements of  influent and effluent
    levels of diazinon and chlorpyrifos and effluent
    toxicity tests.
                                              119

-------
6.  Monitoring studies showed that San Francisco Bay
    Area POTWs achieve substantial removal of both
    diazinon and chlorpyrifos.  The highest levels of
    removal are associated with systems that have
    filtration systems, extended MCRTs, and/or longer
    chlorine contact times.

Introduction
Permit Requirements
During 1990-1991, the CCCSD conducted an effluent
toxicity characterization program in which 18 acute
toxicity tests were performed. The effluent produced
detectable acute toxicity to C. dubia in 12 of the 18 test
events. The  CCCSD's  NPDES permit requires no
significant acute toxicity at 100% effluent; therefore, a
TRE study was required by the California State Water
Quality Control Board, San Francisco Bay Region, to
determine the causes and sources of the acute toxicity.

This study was performed in addition to the TRE study
that addressed effluent  toxicity caused by Cu (see
Appendix B).  The CCCSD was required to meet
permit limits  based on toxicity testing using both C.
dubia and echinoderms.

Description of the Treatment Plant
A description of the treatment plant is presented in
Appendix B.

Facility Performance Evaluation
As part of the TRE study, the CCCSD conducted an
internal facility performance evaluation to determine if
the  treatment system   was  operating   at  design
performance specifications.  A review of all relevant
operating parameters  indicated that  there were  no
obvious performance deficiencies. During this period,
monthly effluent  tests   showed intermittent  acute
toxicity to C. dubia, but no toxicity was detected to
juvenile P. promelas (15- to 60-day-old).

Toxicity Identification  Evaluation
USEPA TIE methods  were used as  guidance in
conducting the Phase I (1988a), Phase H (1988b) and
Phase IE TEE studies (1988c).

Phase I TIE - Toxicity Characterization
A total of five Phase I TIE studies were conducted with
the CCCSD final effluent to characterize the class of
the toxicant(s) responsible for the acute toxicity to C.
dubia. Tests were 48-72 hours in duration and TIE
treatments were not renewed during the tests.  TIE
treatments were conducted on  100%  effluent.  The
results, shown in Table F-l, indicated that the toxicity
was consistently reduced by treatment with CIS SPE
columns at pHz (initial pH of the sample) and PBO
addition. Treatments that produced a partial decrease
in toxicity in two or more samples included adjustment
to pH 3 and aeration. Treatments that consistently did
not decrease toxicity included pH adjustments, sodium
thiosulfate, EDTA, or graduated pH treatment.

The results of the Phase I TIE studies  showed that
acute toxicity was consistently reduced by the C18
SPE  column treatment, which removes  non-polar
organic chemicals.  The methanol  eluates from the
C18 SPE column  were toxic when added to dilution
water at a concentration equivalent to 1.5 times (1.5X)
the concentration in the effluent sample. It is important
to note that the 1.5X calculation assumes that the
toxicity was  completely removed  from the effluent
sample by the C18 SPE column and further, that the
toxicity was completely recovered from the column in
the methanol eluate.

PBO  was effective in preventing acute toxicity to  C.
dubia in all five samples. PBO blocks the metabolic
activation and subsequent toxicity of organophosphate
insecticides,  which  require metabolic activation  to
exhibit  toxicity  (Ankley  et   al.,   1991).    The
ineffectiveness  of sodium thiosulfate  and  EDTA
suggest that oxidants and/or cationic metals were not
implicated in the toxicity. The results of the graduated
pH test also suggested that ammonia did not contribute
to toxicity. Overall, the Phase I TIE results indicated
that the effluent toxicity was due to non-polar organic
toxicant(s), specifically one or more organophosphate
insecticides,  which require  metabolic activation  to
produce toxicity.  Diazinon, a metabolically activated
organophosphate  insecticide, has  been reported  to
cause toxicity in municipal effluents (Norberg-King et
al., 1989; Amato  et al., 1992);  therefore, subsequent
Phase EC  studies focused  on  identifying organo-
phosphate insecticides. Effluent and diazinon-spiked
laboratory  water were used to  determine if the TIE
treatments produced similar effects.

Phase II- Toxicity Identification
A total of four effluent samples were  processed in
Phase n.  PBO completely prevented toxicity in  all
four  effluent samples, suggesting that metabolically
activated  organophosphate   insecticides  were
responsible for the  acute toxicity.  The Phase I TIE
showed that the toxicity could be both removed by and
                                                120

-------
 Table F-l. Matrix of Results of Phase I TIE Conducted on Five Effluent Samples with C. dubia
         Treatment
   Reduces Toxicity Due To
                                                           Samples with Substantially Reduced Toxicity
  CIS SPE column (pffi)*

  CIS eluate toxic'

  PBO addition

  Filtration

  Aeration

  Adjustment to pH 3

  Adjustment to pH 11

  Thiosulfate addition

  EDTA addition

  Graduated pH test
Non-polar organics, metals

Confirms non-polar organics

Organophosphate insecticides

Filterable toxicants

Volatile/oxidizable toxicants

Acid hydrolyzable toxicants

Base hydrolyzable toxicants

Oxidants, some metals

Cationic metals

Ammonia, metals
 * pHz = initial pH.

 recovered fromCIS SPE columns; therefore, the Phase
 n TIE procedures focused on the use of the columns to
 fractionate  the sample for further characterization.
 Aliquots of the samples were concentrated on CIS SPE
 columns and the columns were eluted with a series of
 methanol:water mixtures  (USEPA,  1993a).   Acute
 toxicity tests were then conducted on each fraction at
 1.5X the original effluent concentration.

 The 75% fraction from all the effluent samples was
 acutely toxic.  In some samples,  adjacent fractions
 (e.g., 70,  80, and 85%) also exhibited acute toxicity.
 The toxic fractions were combined, concentrated, and
 sequentially  fractionated  using  HPLC.    For
 comparison, an analytical standard of diazinon was run
 immediately prior  to each effluent sample HPLC run.
 A total of 30 fractions were collected during the HPLC
 linear gradient  (30-100%  methanol:water  for  25
 minutes with 5 minutes at  100%  methanol).  Each
 fraction was assayed at 1.5X the original effluent
 concentration with C. dubia, and toxic fractions were
 treated with  PBO  to  ascertain  the presence  of
 organophosphate  insecticides.  This procedure was
 similar to that described by USEPA (1993a).  The
results are summarized in Table F-2.

The diazinon  standard  consistently produced acute
toxicity in one fraction  (19), and in one HPLC run,
                         toxicity also was observed in another fraction (18). All
                         four effluent samples also produced acute toxicity in
                         fraction 19 and occasionally in adjacent fractions (18
                         and 20).

                         As shown in Table F-2, in all cases, PBO provided
                         protection against acute toxicity in the HPLC fractions
                         in which toxicity occurred (18-20). However, PBO
                         did not protect against the toxicity of fractions 12 and
                         13.  The  results of the PBO treatment of the toxic
                         fractions suggested that one  or more  metabolically
                         activated  organophosphate   insecticides,  such  as
                         diazinon, had a role in the toxicity of all four effluent

                         Table F-2.  Summary of TIE Phase II Results
Sample
Diazinon (Runs 1-4)
Effluent 1
Effluent 2
Effluent 3
Effluent 4
Toxic Fractions
18*+, 19*+
18*+, 19*+
12-f, 19*+
18*+, 19*+, 20$
13f-, 19*+,
                          * PBO provided full protection against toxicity.
                          t PBO provided no substantial protection against toxicity.
                          t PBO provided partial protection against toxicity.
                                                 121

-------
samples.  Diazinon consistently eluted in the  same
fractions that were identified in the effluent samples;
therefore, further studies focused on confirming the
presence of  diazinon  in  the  HPLC fractions and
refining procedures for the accurate determination of
diazinon in effluent samples.  This latter aspect was
challenging because diazinon is toxic to C. dubia at
low concentrations (LC50=0.26-0.58 ug/L) (USEPA,
1991; Ankley et al., 1991; Bailey et al., 1997), and the
CIS SPE column extracts of  the effluent samples
contained numerous interferences which made analysis
by gas chromatography (GC) problematic. Diazinon
analysis  generally followed procedures described by
USEPA (1993a). Diazinon was quantitatedby GC/MS
using selected ion monitoring. The detection limit for
this procedure  in the  CCCSD effluent matrix was
0.010 ug/L of diazinon.

Phase IH- Toxicity Confirmation
The role of diazinon in the CCCSD's effluent toxicity
was assessed using the correlation approach (USEPA,
 1988c).  Hie purpose of the correlation approach is to
determine whether there is a consistent relationship
between the  concentration of the suspected toxicant
and the degree of effluent toxicity. If the correlation is
not robust,  the role of the suspect toxicant in the
effluent toxicity should be re-examined.

A total of seven CCCSD effluent samples collected
during July  and August 1992 were  evaluated  by
comparing the expected toxicity based on diazinon
 (48-hour LC50=0.38 ug/L) with the measured effluent
 toxicity. The 48-hour toxicity of the effluent samples
 ranged from 1.25-2.17 TUa. Diazinon concentrations
 in these samples ranged fromO. 120-0.280 ug/L, which
 corresponds  to 0.32-0.74 TUa based on the 48-hour
 LC50 for diazinon (i.e., 0.12  ug/L + 0.38 ug/L and
 0.28 ug/L -T- 0.38 ug/L).  The oxygen analog of
 diazinon  (diazinon   oxon)   was  not  detected
 (<0.010 Ug/L) in any of the effluent samples analyzed.
 Treatment of the toxic samples with PBO resulted in
 full reduction of toxicity in  five  samples,  partial
 reduction in one sample, and no  reduction in one
 sample. The effluent TUa and diazinon TUa values for
 the seven toxic samples are plotted in Figure F-l along
 with the theoretical regression line, which depicts the
 case where all of the toxicity measured in the sample is
 due to diazinon (diazinon TUa = effluent TUa).   •

 The linear regression of effluent TUa versus diazinon
 TUa had an R2 value of 0.75 (p<;0.01), which indicates
 that diazinon concentrations can account for 75% of
    2.5
     2-
    1.5 H
      1 -
    0.5
Effluent Regression Line
r2 = 0.87
slope =1.69         /
y-intercept = 0.74    S
                        Theoretical Regression Line
             —r™
              0.5:
                       1
       —i—
       1.5
2.5
                      Diazinon TUa
Figure F-l. Effluent TUs versus diazinon TUs in the CCCSD
effluent samples.

the variability in the toxicity of the effluent samples.
However, the regression  is  above the theoretical
regression  line,  which  suggests  that either  the
analytical procedure for diazinon  was  consistently
detecting less than the actual effluent concentration,
and/or there  were one or  more additional toxicants
present in the effluent samples.  Further studies were
undertaken to assess both possibilities.

Analytical procedures were reviewed by the CCCSD
and were found to have acceptable levels of precision
and accuracy.  In an effort to identify the missing
toxicant(s), more rigorous extraction procedures were
applied to additional samples of effluent  that were
toxic to  C.  dubia.    The  effluent samples  were
exhaustively   extracted with methylene  chloride,
evaporated to dryness, and resolubilized in hexane.
Analysis  of  the extracts  by  GC/MS  revealed the
presence  of  chlorpyrifos,  a metabolically activated
organophosphate insecticide, in all the toxic effluent
samples at concentrations greater than the NOEC  of
0.030  ug/L  (AQUA-Science,  1992;  Bailey  et al.,
 1997).

Follow-Up TIE Studies
Before further Phase El studies were initiated, a series
of studies were conducted to validate the Phase I and
n TIE findings for diazinon and to determine why the
Phase n TIE process failed to identify chlorpyrifos as
 a toxicant in the CCCSD effluent. The results of these
 studies are summarized in Table F-3.
                                                  122

-------
 Table F-3. Summary of Follow-Up TEE Studies
TIE Treatment
pH adjustment
PBO addition
CIS SPE
HPLC fractionation
Sample stability
studies
Effect on Organophosphate Insecticides
Diazinon is degraded rapidly at pH 3, but is relatively stable at pH 1 1
PBO at 100-700 ug/L effectively protects against three times LC50 concentration of
diazinon and chlorpyrifos (1.6 and 0.24 ug/L, respectively). Effectiveness of PBO is
affected by the matrix; therefore, use a range of PBO additions (USEPA, 1991a, 1993a).
Diazinon is well recovered (80-100%) from CIS SPE columns
Diazinon elutes sharply in specific methanol/water fractions: 75-80% methanol fractions
for CIS SPE columns
Chlorpyrifos is poorly recovered from CIS SPE columns (40-50% recovery)
Chlorpyrifos tends to elute in broad bands: 80-95% methanol fractions for CIS SPE
columns
Diazinon is well recovered in specific fractions from C18 HPLC columns
Recovery of chlorpyrifos from C18 HPLC columns is highly variable (20-60% recovery)
Significant amounts (20-40%) of diazonin and chlorpyrifos are lost from influent and
effluent samples stored in either glass or plastic containers for 48 hours
Effluent samples should be analyzed or extracted within hours of collection
The follow-up studies provided additional insight into
the initial Phase I and II TIE results.  The instability of
diazinon at pH 3 is  consistent with the reduction in
effluent toxicity after pH 3 treatment. Diazinon is well
recovered  through the Phase n concentration and
fractionation steps (Bailey etal, 1996); therefore, toxic
fractions corresponding to those produced by diazinon
standards  should  be present in all  toxic  effluent
samples, as was demonstrated in the TIE.

On  the  other  hand, the low overall recovery of
chlorpyrifos from C18 SPE columns would explain the
failure to detect chlorpyrifos toxicity in the effluent
C18 SPE and HPLC fractions. For example, using the
values in Table F-3, the  recovery of chlorpyrifos in
HPLC fractions could be as low as 8% (i.e., 40%
recovery from 3 mL SPE column x 40% recovery from
1  mL SPE column x 50% recovery from HPLC
column). This level of recovery would require an add-
back of more than 12X to ensure that concentrations of
chlorpyrifos in  the HPLC fractions  and the effluent
samples were comparable.  This study indicated that
add-backs of fractions at levels substantially greater
than 1.5X should be avoided because of the potential
to amplify the toxicity due to toxicants that are below
the toxic threshold  in the effluent,  but are well
conserved through the TIE process. This could lead to
erroneous identification of chemicals that do not have
a causal role in the effluent toxicity.
 A  critical issue  facing  the  investigator is how to
 identify toxicants that are not well recovered through
 the TIE process.   Recently, procedures have been
 developed  to selectively  remove  diazinon   and
 chlorpyrifos from effluent samples using antibody-
 mediated processes (Miller et al., 1996; Miller et al.,
 1997).  This process involves treating the effluent
 sample with the chemical-specific antibody preparation
 that selectively  removes up to 95%  of the target
 chemical  (either  diazinon  or  chlorpyrifos).    By
 conducting effluent toxicity tests before and after the
 antibody treatment, the exact contribution of the target
 chemical to  the overall toxicity can be determined. In
 addition,  use of sequential antibody  treatments to
 remove both diazinon  and  chlorpyrifos  from  the
 effluent matrix can  indicate  the extent to which
 toxicity is not due to either compound. The residual
 toxicity can  be further characterized through the TIE.

Alternative Analytical Procedures
A major limitation of the TIE study  was obtaining
 accurate and timely analytical information on levels of
insecticides  in effluent samples and TIE treatments.
The GC/MS methods that were available involved
tedious extractions, clean-up, and the use of expensive
analytical equipment  that was fully  scheduled  for
compliance-related purposes. ELIS A procedures were
evaluated as an alternative analytical method for  the
analysis of diazinon and chlorpyrifos  in subsequent
Phase in TIE and source identification studies.
                                                123

-------
Commercially  available  ELISA  kits   (Beacon
Analytical, Scaresborough, Connecticut) have some
distinct advantages  over GC or GC/MS methods,
including cost ($40-70 versus $250-500 per sample),
sample volumes (100 uL versus liters),  sample turn-
around (hours versus days or weeks), and equipment
costs ($3,000 versus >$50,000). The detection limit
for ELISA  kits for  diazinon and  chlorpyrifos
(0.030 ug/L) is also comparable to that for GC/MS.
An interlaboratory study involving 6 laboratories and
a  total  of 19 influent  samples  was conducted  to
compare the performance of ELISA, GC, and GC/MS
procedures for diazinon and chlorpyrifos.  The study
showed that ELISA values for both insecticides were
highly correlated (R2 >0.95) with GC  and GC/MS
results for those laboratories (Singhasemanon et al.,
1997). The results were comparable over a wide range
of concentrations (i.e., 0.030 to 31.5 ug/L for diazinon
and 0.030 to 9.8 ug/L for chlorpyrifos).

Based on the excellent  performance  of the ELISA
procedures  in  the interlaboratory study,  ELISA
procedures  were used  to  monitor  diazinon and
chlorpyrifos concentrations in the CCCSD influent and
effluent samples during  follow-up studies, including
source identification, POTW influent removal studies,
and monitoring the  effectiveness of public outreach
programs.

Source Identification Studies
Source Study 1
A reconnaissance study was conducted in August 1995
to identify  potential  sources  of  diazinon and
chlorpyrifos in wastewater  from selected residential
and commercial sources within the CCCSD collection
system. A total of 36 24-hour composite  samples of
influent were analyzed  for the two insecticides  by
ELISA.   The samples included daily and/or hourly
composite  samples  collected from a  residential
community, and from selected businesses within the
CCCSD collection system, including self-service pet
grooming facilities, operations centers for pest control
operators, and kennels.

The measured levels of diazinon and chlorpyrifos were
coupled with estimated flows from the various sources
to provide estimates of overall contribution of the two
insecticides to the CCCSD's influent. The results are
shown in Table F-4.

Diazmon  and  chlorpyrifos concentrations in the
wastewater from the residential sources were highly
variable (0.050-0.720 ug/L and <0.050-0.520 ug/L,
respectively). Peak concentrations of both insecticides
in the residential samples were measured  in the
samples collected on Saturday afternoon. The cause of
the  spikes of  the insecticides  in the residential
wastewater is under further study and may be related to
home use and/or improper disposal of these chemicals
during weekend activities (e.g., lawn care operations
for diazinon and pet flea control for chlorpyrifos).

Diazinon  and  chlorpyrifos  levels in  wastewater
samples collected from commercial sources also were
highly variable  (<0.030-16.0 ug/L and 0.040-5.4
ug/L, respectively). The highest concentrations of both
insecticides  were measured in wastewater samples
from a commercial kennel.
                 [
Overall, the reconnaissance study showed that although
high levels of diazmon and chlorpyrifos were detected
in some of the wastewater samples from commercial
sources,  the  vast majority of  the loading  of the
insecticides into CCCSD influent during the sampling
period was from residential sources.  This finding
agrees with an earlier study of sources of diazinon in
Fayetteville, NC (Fillmore et al., 1990).
 Table F-4. Diazinon and Chlorpyrifos Concentrations in Wastewater Samples from Selected Residential and Commercial
 Sources in the CCCSD
Source
Residential
Commercial: Pest control operators
Pet groomers
Kennels
Diazinon ;
ug/L
0.050-0.720
<0.03-1.10
<0.03-0.10
0.070-16.00
% of Total
Influent Loading
101
3
<1
2
Chlorpyrifos
ug/L
<0.05-0.52
0.060-1.80
0.04-7.00
3.10-5.40
% of Total
Influent Loading
94
4
2
1
                                                 124

-------
 Source Study 2
 Results of the reconnaissance study were used by the
 CCCSD and the California Department of Pesticide
 Registration (CADPR) to develop a plan for a more
 definitive study that  was conducted from June to
 September 1996 (Singhasemanon et al., 1997). m this
 study, over 200 flow-proportional 24-hour composite
 samples were collected from each of 5 residential areas
 and  12 businesses  (pet groomers,  pest  control
 operators, and kennels) within the CCCSD collection
 system.  How measurements were made at selected
 sampling points in order to calculate mass loadings of
 diazinon and chlorpyrifos.   The measured flows in
 residential areas were compared with modeled flow
 data  obtained  from a  computer program  [Sewer
 Network Analysis Program (SNAP) 1989, developed
 by the CCCSD]. The SNAP program applies modeled
 land use, groundwater infiltration,  and CCCSD plant
 influent data to determine flow rates from the sampled
 areas.   Concentrations  of  the insecticides were
 measured  using  ELISA,   GC,   and/or   GC/MS
 procedures.  The loading of diazinon and chlorpyrifos
 in the CCCSD influent from residential sources was
 estimated  by  multiplying  the mean  insecticide
 concentrations measured from the residential sites by
 the SNAP flow rates from the sampled sources.  The
 commercial loading was estimated by multiplying the
 mean insecticide  concentrations measured  at each
 business by the measured flows and the number  of
 similar businesses in the sewer service area.  The data
 were analyzed using a computer program (SASฎ, SAS
 Institute,  Inc,  1994,  Version  6.1, Gary,  North
Carolina), which calculated the Uniformly Minimum
Variance Unbiased Estimator (UMVUE) for the mean
influent loading concentrations  for the  insecticide
(Singhasemanon et al., 1997).  The mean UMVUE
influent concentrations and associated loading for
diazinon was 0.230 [ig/L and 34.7 g/day, respectively.
Corresponding values for chlorpyrifos were 0.145 |ig/L
and 15.0 g/day.  The percentage of the total loading
contributed by residential, commercial and unknown
sources is shown in Figure F-2.

The CADPR study concluded that:

  • Levels of diazinon and chlorpyrifos were highly
   variable  in wastewater  samples  from  both
   residential and commercial sources.
  • Residential   neighborhoods   contributed   the
   majority  of diazinon  and chlorpyrifos  to the
   CCCSD's influent.
  • Although relatively high concentrations of both
   insecticides  were found at commercial sources,
   low flows from these sources resulted in relatively
   small mass loadings.
  • A mass balance showed that a significant mass of
   chlorpyrifos  and,   particularly,  diazinon  was
   unaccounted for. Uninvestigated sources such as
   restaurants,  nurseries,  and industrial facilities
   should be sampled in future studies.
  • Future source reduction strategies should focus on
   residential customers  to  identify  and correct
   behaviors   that  contribute   to  disposal   of
   organophosphate insecticides to the sewer system.
                   Diazinon
  Unknown
    (42%)
         Commercial
            (6%)
                                 Residential
                                   (52%)
                   Chlorpyrifos
   Unknown
     (25%)
                                  Residential
                                    (60%)
                                                   Commercial
                                                      (15%)
Figure F-2. Percent mass contribution of sources to the CCCSD influent.

                                               125

-------
As a result of the source identification studies, the
CCCSD:

  •  Initiated a multi-faceted public outreach program
    targeting residential costumers to increase public
    awareness of the  proper  use  of  disposal of
    insecticides. The initial program included point of
    sale  information  sheets,  newspaper  articles,
    television ads, and billboards.   A program to
    enhance public awareness of proper insecticide use
    by  promoting  integrated  pest  management
  .  practices is on-going.
  •  Shared study information with interested POTWs
    and State and Federal regulatory agencies.
  •  Initiated frequent effluent monitoring of diazinon
    and chlorpyrifos coupled with an effluent toxicity
    program to monitor the success of the public
    outreach program.
  •  Planned further studies to identify homeowner
    practices that contributed to the  discharge of
    insecticides to the collection system.
  •  Reviewed disposal practices with  pest control
    operators, pet care businesses, and kennels within
    the District.
  •  Conducted a study to identify the toxicity of
    alternative products for pet flea control.

 Loading and Removal of Diazinon and
 Chlorpyrifos
 Study 1
 As  an  ancillary   part  of  the  CADPR source
 identification study, diazinon and chlorpyrifos were
 measured in  seven consecutive daily samples of
 influent and effluent from CCCSD and two nearby
 POTWs [Union Sanitary District (USD),  Fremont,
 California, and the  Regional Water Quality Control
 Plant (RWQCP), Palo Alto, California]. The purpose
 of the study was to  assess differences in loading and
 removal efficiencies for the POTWs.  The three
 POTWs had similar influent flows (25-38 mgd),
 aeration detention times (3.8-5.6 hours), and clarifier
 detention times (2.0-4.2 hours). However, the CCCSD
 and the USD had shorter MCRTs (1.6-1.8 days versus
 11.6 days) and shorter chlorine contact time (30-50
 minutes versus 90  minutes) when  compared to the
 RWQCP. In addition, the RWQCP treatment process
 incorporates two-stage aeration and dual media
 filtration to optimize particulate removal. The results
 of the study are shown in Figure F-3.

 Daily concentrations of both diazinon and chlorpyrifos
 in the three POTWs varied widely during the sampling
 period.  The CCCSD consistently  had the highest
influent  and  effluent  concentrations   of  both
insecticides, followed by the USD and the RWQCP.
The  CCCSD and the USD, which have  similar
treatment processes, had similar removal efficiencies
for  diazinon  (32 and 24%,  respectively),  and
chlorpyrifos (53  and  49%,  respectively).    The
RWQCP, which has longer chlorine contact time, two-
stage aeration, and dual media filtration had the highest
removal  efficiencies   for  diazinon  (82%)  and
chlorpyrifos (71%). The effect of these parameters on
the  removal  and/pr  degradation of diazinon  and
chlorpyrifos  in   municipal  influent  was  further
evaluated in a subsequent study.

Study 2
A larger scale study was conducted to confirm the
findings of the CADPR study, which suggested that
there may be  demographic  and/or microclimatic
differences in influent loadings  of diazinon  and
chlorpyrifos to POTWs within the same region and
moreover,  there  may be  differences  in  removal
efficiencies of the two insecticides in POTWs using
different treatment systems.  Seven daily  24-hour
composite  samples of influent and effluent were
collected from 9  Bay Area POTWs during August
 1997. The POTWs included the CCCSD and the cities
of Fairfield-Suisun, Hayward, Palo  Alto, Petaluma,
San Francisco, San Jose, Union City, and  Vallejo.
Samples were analyzed for diazinon and chlorpyrifos
within 24 hours of collection using ELISA (AQUA-
Science, 1997).   The results  for diazinon and
chlorpyrifos are shown in Figure F-4. Information on
 the characteristics of each POTW treatment system is
 shown in Attachment 1.

 The results of this study confirmed and extended the
 findings of the previous study. A summary is provided
 below.

   •  Mean  influent concentrations for both diazinon
     and chlorpyrifos were highly variable and ranged
     from 0.278-1.211 |ig/L and 0.030-0.176 ug/L,
     respectively.  These results suggest-that there are
     regional demographic, and possibly,  climatic
     differences in use and disposal practices for the
     insecticides.
   •  All the POTWs achieved substantial removal of
     the two insecticides from influent (up to 98% for
     diazinon and up to 86% for chlorpyrifos). These
     removal rates  are generally higher  for  both
     insecticides than  were  observed in the previous
     study.   The highest levels  of removal  were
                                                126

-------
                     +0.474 +0.26
                           CCCSD                 USD

                                   Influent and Effluent Samples

               | Diazinon Influent  03 Diazinon Effluent
                     RWQCP
Chlorpyrifos Influent  l~l Chlorpyrifos Effluent
 Figure F-3. Mean diazinon and Chlorpyrifos concentrations (ฑstd) in influent and effluent from three Bay Area POTWs.
    associated with POTWs that had filtration, longer
    MCRTs  and chlorine contact times,  and long
    retention in ponds.
  • Mean effluent concentrations for diazinon  and
    Chlorpyrifos ranged from <0.030-0.241 ug/L and
    <0.030-0.085 |ig/L, respectively.  The combined
    mean effluent concentrations for both insecticides
    exceeded 1.0 TUa in  only three of the nine
    POTWs sampled (including the CCCSD).
  • Overall, the results showed that all the POTWs
    sampled during this period had potentially toxic
    levels  of diazinon and  Chlorpyrifos in their
    influents.  However, all  the POTWs achieved
    substantial removal of both insecticides.

Another round of sampling was scheduled for February
1998 to assess seasonal effects on influent levels and
removal rates from the POTWs.

Alternative Pet Flea Control Products
Toxicity source investigations  by   the  CCCSD
suggested that pet flea control products were a major
source of Chlorpyrifos in the influent (AQUA-Science
1995a  and  1995b).    Before the CCCSD  could
 recommend alternative products, it was necessary to
 conduct studies to determine the toxicity of several
 commonly used pet flea dips and shampoos. The acute
 toxicity of  six flea shampoos and  four dips was
 evaluated with C.  dubia (AQUA-Science,  1995a;
 Miller et  al., 1994).  Although  the products tested
 varied widely in toxicity, shampoos were generally less
 toxic than the dips. The most toxic products tested
 contained Chlorpyrifos (IC25s of 0.800 to 2.30 ug/L as
 product),  which were  2,500-7,000  times more toxic
 than the least toxic product tested, which contained
 D-limonene (IC25 of 5.687  ug/L).   The products
 containing pyrethrins and permethrin had intermediate
 levels  of toxicity  (IC25s  of 0.149-4.683  ug/L).
 Calculations (with the  associated assumptions on use
 rate, system losses, and dilution) indicated that only
 flea dip  products  containing  Chlorpyrifos  were
 sufficiently  toxic to  produce measurable  effluent
 toxicity to C. dubia.

Effects of Household Bleach on Aqueous
 Concentrations of Diazinon and Chlorpyrifos
A study was conducted to determine if household
bleach could be recommended to residential customers
                                                127

-------
                      +0.099
                                                                                              +0.001
                       CCCSD   Fairfield  Hay ward  Palo Alto  Petaluma    San   ,  San Jose     Union
                                                                   Francisco
                                                                    +1.15
Vallejo
                0.14
                        CCCSD   Fairfield  Hayward  Palo Alto  Petaluma    San    San Jose    Union    Vallejo
                                                                    Francisco
                                              Influent and Effluent Samples
                         I Influent   [J Effluent
Figure F-4. Mean chlorpyrifos and diazinon concentrations (ฑstd) in influent and effluent from mine Bay Area POTWs during
August 1997.
                                                       128

-------
 as a measure to degrade diazinon in spray container
 rinsate and chlorpyrifos from pet flea washes prior to
 disposal into the sewer. Samples of tap water were
 spiked  with  high   concentrations   of  diazinon
 (60.0 fig/L) and chlorpyrifos (10.0 |J.g/L) and treated
 with either 0.005 or 5% solutions of household bleach
 for 24 hours. After neutralization, concentrations of
 the insecticides were measured by ELISA (AQUA-
 Science, 1995a). Both bleach concentrations reduced
 concentrations of the insecticides by 86-92%.  The
 study  suggested that  household bleach may be a
 effective pretreatment for waste solutions of diazinon
 and chlorpyrifos prior to disposal. Additional studies
 are planned to further define bleach exposure times and
 concentrations under actual  use conditions, and to
 characterize the chemical oxidation products produced
 by the chlorine treatment.

 Diazinon and Chlorpyrifos Concentrations in
 Water Samples from Restaurant Grease Traps
 The CADPR source identification study recommended
 follow-up  studies to  determine  concentrations  of
 diazinon  and chlorpyrifos  in  wastewater  from
 restaurants. Water samples were collected from the
 grease traps of eight restaurants in the CCCSD service
 area (AQUA-Science,  1997).  ELISA  was used to
 measure concentrations  of  the  two  insecticides.
 Diazinon and chlorpyrifos concentrations ranged from
 0.192-4.197 jig/Land 0.265-4.313 ng/L, respectively.
 The highest concentrations of both insecticides were
 found in wastewater from the same restaurant. The
 uses that contributed to these insecticide residues in the
 wastewater are currently being investigated by the
 CCCSD.

 Regulatory Activities
 Chlorpyrifos-Related
In January 1997, Dow-Elanco, as part of an agreement
with USEPA, announced  the  following actions
associated with the registered uses of chlorpyrifos (L.
Goldman,   USEPA  Assistant  Administrator  for
Prevention, Pesticides and Toxic Substances.  Press
Release on January 16, 1997):

 •  Withdrawal of chlorpyrifos from indoor broadcast
    and fogger flea control markets.
 •  Withdrawal of chlorpyrifos from direct application
    pet-care uses (shampoos, dips, and sprays).
 •  Increase marketing of ready-to-use products to
    replace concentrated formulas.
 •  Increase training and supervision of pest control
    operators.
   • Revise chlorpyrifos labels to limit retreatment
     intervals.

 If the chlorpyrifos in POTW influent loading is due to
 indoor and pet-care uses and/or misapplications by pest
 control  operators, these actions should substantially
 reduce influent loadings of this chemical.

 Diazinon-Related
 In 1996, Novartis Crop Protection, Inc., the major U.S.
 registrant of  diazinon, submitted voluntary label ,
 changes to USEPA to warn users not to dispose of this
 product  into sanitary or storm drains.  Novartis also
 developed educational materials with this message and
 provided the materials to selected cities in Texas and
 California. In 1997, Novartis completed a 4-year study
 with several POTWs in USEPA Region VI on diazinon
 occurrence and treatability (Novartis, 1997). Afollow-
 up study is  on-going with a California POTW  to
 identify  treatment processes that consistently optimize
 removal  of  diazinon   (D.   Tierney,   personal
 communication, Novartis Crop Protection, 1997).

 Discussion
 In this case study, USEPA TIE procedures were used
 to identify organophosphate insecticide toxicity in a
 POTW  effluent.  Phase  I and n TIE procedures
 identified diazinon as a candidate toxicant. Phase HI
 TIE  studies  determined  that  effluent   diazinon
 concentrations  were  significantly correlated with the
 extent of  the  effluent  toxicity, but  diazinon only
 accounted for approximately half of the effluent's
 toxicity.    The follow-up  TIE  studies identified
 chlorpyrifos at potentially toxic concentrations in the
 toxic effluent samples. ELISA procedures were shown
 to provide sensitive and accurate measurements of the
 two insecticides in samples of  POTW influent  and
 effluent,  and these procedures were used extensively in
 follow-up  TIEs and source  identification  studies.
 Additional TIE experiments found chlorpyrifos to be
 poorly recovered through the Phase  I and n TIE
 processes, which may explain why it has not been
 identified as a toxicant in other effluent TIEs.

 The source identification studies at the CCCSD and
 other Bay Area POTWs showed that the influents
 contained highly variable, and often potentially toxic,
 levels of diazinon and chlorpyrifos, which appearto be
 originating primarily from residential rather than
 commercial sources. However, only a relatively small
number of commercial sources have been sampled to
date. Thus, it is possible that certain business types
                                                129

-------
(e.g., restaurants and nurseries) may be significant
contributors of the two insecticides into wastewater.
All of  the POTWs that were sampled to date have
demonstrated substantial removal of both insecticides
from their influents.  This was surprising because it
was generally believed that these insecticides were
poorly  treated  by POTWs  (J.L.  Miller,  personal
communication, Aqua-Science, Inc., Davis, California,
April 1998). The available data suggest that there were
substantial differences in influent loadings of diazinon
and chlorpyrifos  between POTWs  within the San
Francisco Bay region. Further studies are planned to
explore the demographic basis for these differences to
evaluate patterns of insecticide use. Seasonal trends in
insecticide  removal efficiencies are currently being
monitored in nine Bay Area POTWs. Public outreach
programs, supported, in part, by the manufacturers of
diazinon and chlorpyrifos, have been implemented by
the CCCSD and other POTWs across the country to
increase  awareness of the proper use and disposal of
insecticides. Recent regulatory actions have resulted
in the  withdrawal of chlorpyrifos from the pet flea
control market, and this  action,  coupled  with the
enhanced training of applicators and the increased use
of prediluted insecticide products, may eventually
reduce the influent loadings. Monitoring studies are in
place at  the CCCSD and elsewhere to determine if
these programs  will result in reduced influent loadings
and decreased incidences of insecticide-related effluent
toxicity.

Acknowledgments
This work was supported wholly or in  part by the
CCCSD, Martinez, California.  Bart Brandenburg,
Bhupinder Dhaliwal, and Jim Kelly of the CCCSD
managed the various aspects of this project.  The TRE
studies were conducted  at  AQUA-Science, Davis,
California, under the direction of J.L. Miller and MJ.
Miller.

 References
 Amato,  J.R.,  D.I. Mount,  EJ.  Durham, M.T.
    Lukasewycz,  G.T. Ankley,  and E.D.  Robert.
     1992.   An  Example of the Identification of
     Diazinon as  a Primary Toxicant in  an Effluent.
     Environ. Toxicol. Chem.  11: 209-16.

 Ankley, G.T.,  J.R. Dierkes, D.A. Jensen, and G.S.
     Peterson.  1991. Piperonyl Butoxide as a Tool in
     Aquatic  Toxicological  Research   with
     Organophosphate   Insecticides.    Ecotoxicol.
     Environ. Saf.  21: 266-74.
AQUA-Science.  1992.  Phase III Effluent Toxicity
    Identification   Evaluation  Studies   with
    Ceriodaphnia dubia.  Report for Central Contra
    Costa Sanitary District, Martinez, California.

AQUA-Science. 1995a. Diazinon and Chlorpyrifos in
    Wastewater from Residential and Commercial
    Sources. Report for Central Contra Costa Sanitary
    District, Martinez, California.

AQUA-Science.   1995b.   Flea Control Products
    Toxicity Assessment  Study with Ceriodaphnia
    dubia.  Report for Central Contra Costa Sanitary
    District, Martinez, California.

AQUA-Science.  1997.  Diazinon and Chlorpyrifos
    Concentrations   in Wastewater Samples from
    Restaurant Grease  Traps.  Report for Central
    Contra  Costa   Sanitary  District,   Martinez,
    California.   .

Bailey, H.C.,  C.  Digiorgio, K. Krull, J. Miller,  D.
    Hinton, and G. Starrett. 1996. Development of
    Procedures for Identifying Pesticide Toxicity in
    Ambient  Waters:   Carbofuran, Diazinon and
    Chlorpyrifos.  Environ. Contamin. Toxicol.  15:
    837-45.

Bailey, H.C., J.L. Miller, MJ. Miller, L.C. Wiborg, L.
    Deanovic, arid T. Shed. 1997.  Joint Toxicity of
    Diazinon  and Chlorpyrifos under Conditions of
    Acute Exposure to Ceriodaphnia dubia. Environ.
    Toxicol. Chem.   16: 2304-08.

Fillmore, L.B., T.L. Morris, T.L. Champlin, M.C.
    Welch, and J.A.  Botts.  1990. Toxicity Reduction
    Evaluation at the City of Fayetteville Cross Creek
    Wastewater Treatment Plant, Draft Report. Risk
    Reduction Engineering Laboratory, Cincinnati,
    Ohio.  USEPA Contract No. 68-03-3431.

 Miller, J.L.,H.C. Bailey, and MJ. Miller. 1994. Fate
    and Toxicity of Diazinon  and Chlorpyrifos in
    Municipal Effluent. Published Abstracts.  15th
    Annual  Meeting,  Society  of  Environmental
    Toxicology and Chemistry, Denver, Colorado.
    October 30-November 3.

 Miller, J.L., MJ.' Miller, and B. Silverman.  1996.
    Selective  Removal of Diazinon and Chlorpyrifos
    from Aqueous Matrices.   Published Abstracts.
    17th Annual Meeting, Society of Toxicology and
     Chemistry, Washington, D.C., November 17-21.
                                                 130

-------
Miller, J.L., MJ. Miller, B. Silverman, C. Foe, and V.
    DeVlaming.   1997.   Selective  Removal of
    Diazinon   and  Chlorpyrifos  from  Aqueous
    Matrices Using Antibody-Mediated Procedures.
    Published  Abstracts.   18th  Annual Meeting,
    Society of Toxicology and  Chemistry,  San
    Francisco, California, November 16-20.

Norberg-King, T.J., M Lukasewcyz, and J.  Jenson,
    1989.  Results of Diazinon  Levels in POTW
    Effluents in the U.S.  NETAC Technical Report
    14-89. USEPA, Duluth, Minnesota.

Novartis Crop Protection and Makhteshim-Agan of
    North America. 1997. Investigation of Diazinon
    Occurrence, Toxicity and Treatability of Southern
    United States Publicly Owned Treatment Works.
    Executive  Summary.   Technical  Report:  3-97.
    Greensboro, North Carolina.

Singhasemanon, N., C. Nordmark, and T. Barry. 1997.
    Diazinon and  Chlorpyrifos in Central  Contra
    Costa Sanitary District Sewer System, Summer
    1996. State of California Environmental Protection
    Agency, Department of Pesticide  Regulation.
    Sacramento, California. Draft Report.

USEPA.   1988a.  Methods For Aquatic Toxicity
    Identification Evaluations:   Phase I Toxicity
    Characterization Procedures. EPA/600/3-88/034.
    National Effluent Toxicity Assessment  Center,
    Duluth, Minnesota.

USEPA.  1988b.  Methods For  Aquatic Toxicity
    Identification Evaluations:  Phase II Toxicity
    Identification  Procedures.   EPA/600/3-88/035.
    National Effluent Toxicity Assessment  Center,
    Duluth, Minnesota.
 USEPA.   1988c   Methods  For Aquatic Toxicity
    Identification Evaluations:   Phase III Toxicity
    Confirmation  Procedures.  EPA/600/3-88/036.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

 USEPA.   1991.   Methods  for Aquatic Toxicity
    Identification Evaluations:   Phase  I, Toxicity
    Characterization Procedures.  Second Edition.
    EPA/600/6-91-003.  National Effluent Toxicity
    Assessment Center, Duluth, Minnesota.

 USEPA.  1992.  Toxicity Identification Evaluations:
    Characterization of Chronically Toxic Effluents,
    Phase I. EPA/600/6-91-005F. National Effluent
    Toxicity Assessment Center, Duluth, Minnesota.

 USEPA.   1993a.   Methods  for Aquatic Toxicity
    Identification Evaluations:  Phase  II Toxicity
    Identification Procedures for Samples Exhibiting
    Acute and Chronic Toxicity. EPA/600/R-92-080.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

 USEPA. 1993b.  Methods for Aquatic Toxicity Identi-
    fication Evaluations. Phase III Toxicity Confirma-
    tion Procedures for Samples Exhibiting Acute and
    Chronic Toxicity. EPA/600/R-92-081. National
    Effluent Toxicity Assessment  Center, Duluth,
    Minnesota.

USEPA.   1996.   Marine Toxicity  Identification
    Evaluation (TIE) Guidance Document, Phase I.
    EPA/600/R-96/054.   Office  of Research  and
    Development, Washington, D.C.
                                               131

-------
                                          Attachment I
                       Summary of POTW Treatment System Characteristics
CCCSD
39mgd
Primary sedimentation
Air-activated sludge (MART 1.6 days)
Secondary clarification
UV disinfection

Fab-field
13 mgd
Primary sedimentation
Oxidation towers with clarification
Air-activated sludge (MART 12-14 days)
Secondary clarification
Tertiary filtration with dual media
Chlorine disinfection (90-120 minutes)

Hayward
12 mgd
Valuators
Primary sedimentation
Fixed film reactors (sludge age n/a)
Anaerobic digester
Final clarifiers
Chlorine disinfection (~100 minutes)

Palo Alto
26 mgd
Primary sedimentation
Fixed  film reactor to  mixed aeration basins with
    activated sludge (MART 11.6 days)
Secondary clarifiers
Mixed media filtration
Chlorine disinfection (90 minutes)

Petaluma
6 mgd
Primary clarification
41% to activated sludge
32% to trickling filter
27% bypasses to ponds where retention time is about
    100 days       :

San Francisco
17 mgd
Primary sedimentation
Air-activated sludge (MART ~ 0.86 days)
Secondary clarification
Sodium hypochlorite disinfection

San Jose
137 mgd
Primary sedimentation
Air-activated sludge (MART ~ 4 days)
Secondary clarification
Nitrification and clarification (MART -11 days)
Tertiary filtration with backwash to clarification (for
    flow equilibrium)
Chlorine disinfection (40-60 minutes)

Union
31 mgd
Primary sedimentation
Air-activated sludge (MART ~ 1.75 days)
Secondary clarifiers
Chlorine disinfection (30 minutes)
                  I
Vallejo
12 mgd
Primary sedimentation
Biological filters
Aeration basins (MART ~ 3 days)
Clarification
UV disinfection and sodium hypochlorite  contact
    (8 minutes)
                                                132

-------
                                        Appendix G

                                     TRE Case Study:
                 Linden Roselle Sewerage Authority, New Jersey
Abstract
TRE Goal:

Test Organism:
TRE Elements:
Toxicants Identified:
Toxicity Controls:
96-hour LC50;> 50%
Interim goal of LC50 ^30%
M. bahia
Facility performance
evaluation,  TIE,  toxicity
source evaluation
Ammonia,  non-polar
organic   compounds,
surfactants
Pretreatment limits
Summary
Ammonia was confirmed as the  primary cause of
toxicity, and pretreatment limits were developed to
reduce effluent ammonia concentrations.  Secondary
causes of toxicity were complex and highly variable.
Toxicity-based procedures were  used to identify
industrial sources of toxicity and develop pretreatment
limits to control secondary causes of toxicity.

In 1997, a major source of ammonia was eliminated.
An acute toxicity test performed since then showed a
reduction  in  effluent  toxicity  (LC50 = 72%) to
compliance levels (i.e., LC50 >50%). Additional tests
are planned to confirm this initial result.

Key Elements
1. TIE procedures  may need to be modified to
   evaluate multiple causes of effluent toxicity. In
   this study, it  was  necessary  to  remove  toxic
   effluent concentrations of ammonia in the TIE
   before other causes of toxicity could be identified
   and confirmed.
2. If  TIE analyses are  successful  in confirming
   causes  of effluent  toxicity  (e.g., ammonia),
   chemical-specific analyses can be used to identify
                                                  4.
sources and pretreatment limits can be developed
for controllable toxicants.
If the TEE is inconclusive or the causes of toxicity
are variable and complex, the RTA approach can
be used to track the industrial sources of toxicity in
the collection system. Once identified, the toxic
dischargers can be required to meet pretreatment
limits for toxicity.
If effluent toxicity is contributed by controllable
industrial sources, pretreatment controls are more
practical than in-plant controls.
                           Introduction
                           Permit Requirements
                           The  LRSA   New  Jersey  Pollutant  Discharge
                           Elimination System (NJPDES) permit contains an
                           acute whole effluent toxicity limit of LC50 >50%
                           effluent.  A 96-hour static renewal M. bahia (mysid)
                           test is used to monitor compliance with the  limit.
                           Based on observed toxicity to mysids, the NJPDES
                           permit was amended to include a  requirement to
                           perform a TRE. In July 1992, the LRSA entered into
                           an administrative consent order (AGO) with the New
                           Jersey Department  of  Environmental  Protection
                           (NJDEP) to establish a  compliance  schedule for
                           reducing acute effluent toxicity. The AGO established
                           a compliance date of October 31,1996, if pretreatment
                           controls are implemented and a compliance date of
                           December  31,  1997,  if  in-plant  controls  are
                           implemented. The AGO also includes TRE milestones
                           and an interim whole effluent toxicity limitation of an
                           LC50 of 30%.  The acute effluent toxicity limit of an
                           LC50 of 50% becomes effective on May 1, 2000.

                           Description of the Treatment Plant
                           The LRSA POTW serves a 13-square-mile area in
                           northeastern New Jersey.  The POTW has a design
                                              133

-------
flow of 17 mgd and is presently treating a wastewater
flow of about  13 mgd. Approximately 20% of the
influent flow is contributed by 40 industrial users.
Primary treatment consists of screening and degritting
followed by primary  sedimentation.  The primary
effluent is then treated by roughing (trickling) filters
and   conventional  activated  sludge   treatment.
Following secondary  clarification,  the  effluent is
disinfected with chlorine and then discharged to the
Arthur Kill estuary. The NJPDES permit specifies that
samples for toxicity testing be collected  prior to
chlorination.

Plant Performance Evaluation
A limited  POTW  performance  evaluation  was
conducted during a USEPA TRE research study to
determine if POTW operations or performance was
contributing to the observed acute  toxicity.  The
POTW performance evaluation findings showed that
industrial wastewater contributions have a significant
effect on the variability and concentration of influent
constituents.  For example, in 1987, influent BOD5
varied from 292 to 636 mg/L, oil and grease ranged
from 11 to 132 mg/L, and ammonia-nitrogen varied
from 17  to  119  mg/L (Morris et al., 1990).  The
influent variability  requires  the  LRSA  to  make
significant modifications to plant operations, such as
operating  one or two aeration basins, to  maintain
optimum treatment. Despite this variability, the LRSA
has consistently met NJPDES permit effluent limits for
conventional pollutants.

Overall, the POTW performance evaluation indicated
that  the operation and performance  of the LRSA
POTW was  satisfactory and the  treatment processes
did not appear to be contributing to effluent toxicity
(Morris et al., 1990).   The POTW performance
evaluation  also   indicated  that   the   ammonia
concentrations observed  in the effluent warranted
further evaluation as a cause of effluent toxicity.

Pretreatment Program Review
Monthly average influent ammonia concentrations at
the LRSA have been as high as 150 mg/L. A review of
the influent ammonia data indicated consistently lower
ammonia levels in July of each year (LRSA, 1990a).
The decreased ammonia concentrations were related to
the temporary shutdown of a manufacturing process at
a major industrial contributor.

Toxicity Identification Evaluation
An objective of the LRSA TRE was  to identify the
causes of effluent toxicity in order to select controls for
reducing toxicity.  Initial TIE Phase I and Phase n
testing was performed in 1989 using C. dubia as a
surrogate test species. C. dubia were used because
little information was available at the time for using
mysids as a TEE test organism. Subsequent TIE testing
in 1991 was performed using mysids to confirm that
the causes of toxicity identified using C. dubia were
also causes of toxicity to mysids.

TIE Phase I
During the USEPA study, three effluent samples were
tested using the TIE Phase I procedures (USEPA,
1988). The Phase I results and ammonia data indicated
that ammonia was a primary cause of effluent toxicity.
Toxicity reduction by CIS  SPE suggested that non-
polar organic compounds were also contributing to
effluent toxicity (Morris et al., 1990).

TIE Phases II and III
TIE Phase H (USEPA, 1989b) and Phase HI (USEPA,
1989c) analyses were performed using C. dubia and
mysids to identify arid confirm ammonia and non-polar
organic toxicants as causes of effluent toxicity (LRSA
1990b, 1991; Morris et al., 1992). It was necessary to
remove ammonia toxicity  in the TIE before other
toxicants  could be  evaluated.   A serial treatment
approach  was used,  to evaluate the contribution of
non-polar  organic  toxicants to acute effluent toxicity.
Effluent samples  were first treated with zeolite to
remove ammonia and then non-polar organic toxicity
was evaluated using CIS SPE column treatment and
GC/MS analyses.  A separate CIS SPE column test
was performed using whole effluent to determine if
zeolite  treatment  had removed non-polar organic
toxicity.

Results of the non-polar organic toxicant confirmation
tests, presented in Table G-l, show that filtration, CIS
SPE column treatment, and zeolite treatment reduced
toxicity to both mysids and C. dubia. The combined
treatment steps removed all of the acute toxicity to both
species.    Following filtration,  zeolite  treatment
removed 1.3 to 2.0 TUa, while the CIS SPE column
removed 1.5 to 4.3 TUa. Acute toxicity to both species
was  recovered  in the 80 to 100% methanol/water
fractions from the C18 SPE column. Although only
0.3 TUa were recovered from the  column, previous
tests had shown greater recovery (>2 TUa). The lower
recovery of non-polar organic toxicity in this sample
may be due  to the presence of toxicants that are
difficult to elute  from the  C18 SPE column (e.g.,
surfactants were indicated as a possible toxicant based
                                                134

-------
Table G-l. TEE Phase III Results; Non-Polar Organic Compound Confirmation (LRSA POTW)
Sample Description*
Baseline toxic units
Post-filtration treatment
Aliquot No. 1
Post-filtration and CIS SPE column treatment (original pH)
Combined toxic methanol/water
CIS SPE column fractions*
Aliquot No. 2 '
Post-zeolite treatment
Post-zeolite and CIS SPE column treatment
Combined toxic methanol/water fractions from zeolite/
CIS SPE column treatment*
TUa (100/LC50)
C. dubiaf
4.3
2.8

100 ง
0.3

1.5
<1.0
0.3
M, bahia$
8.5
6.3

100 ง
0.3

4.3
<1.0
0.3
 * Effluents of serial treatment steps.
 t 48-hour C. dubia acute toxicity test.
 t 96-hour M bahia acute toxicity test.
 ง Percent mortality in 100% sample after 48 and 96 hours for C. dubia and M. bahia, respectively.
 # Methanol/water fractions were evaluated at 5 times and 2.5 times whole effluent concentration for C. dubia and M. bahia,
   respectively.
on the toxicity removed by filtration).  Overall, the
results showed that mysids were sensitive to the same
non-polar organic toxicity as C. dubia.  These tests
confirmed non-polar organic toxicants as a cause of
effluent toxicity to mysids.

Difficulties were encountered in trying to identify and
confirm the specific non-polar organic toxicants. TIE
Phase H procedures (USEPA, 1989b), which included
HPLC separation  and GC/MS analyses, tentatively
identified more than 20 non-polar organic compounds
as potential causes of toxicity.  In addition,  many
potentially toxic unknown compounds were detected.
The  results  suggested  that  the  majority  of  the
compounds were related to industrial sources because
the compounds are not typically found  in domestic
wastewater.    Further work was not performed  to
identify  the  toxic non-polar  organic  compounds
because:

  • Little or no toxicity data were available for most of
   the non-polar organic compounds identified in the
   effluent (e.g., no LC50 values for the specific non-
   polar organic compounds); therefore, it was not
   possible to determine if the concentrations present
   in the effluent were acutely toxic.
  • The non-polar  organic  toxicants varied from
    sample to sample, which made it  difficult to
    determine consistent causes of non-polar organic
    toxicity.
  • Many of the compounds detected were unknowns.

The TIE results indicated that, in addition to ammonia,
non-polar organic toxicity may need to be controlled to
achieve compliance with the acute toxicity limit. Due
to the difficulty in determining the non-polar organic
toxicants, the LRSA decided to use a toxicity-based
approach to identify the sources of non-polar organic
toxicity and other non-ammonia effluent toxicity.

Toxicity Source Evaluation
The available information indicated that both ammonia
and non-ammonia (e.g., non-polar organic) toxicity
was being  contributed  by  controllable industrial
sources.    Therefore,  pretreatment controls  were
deemed to be feasible and source evaluation studies
were performed to identify the sources of ammonia and
non-ammonia  toxicity.  Sources  of ammonia were
identified by a chemical-specific approach and sources
of  non-ammonia  toxicity  were  identified by a
toxicity-based  approach.  The resulting information
was used to develop appropriate pretreatment limits.
                                                 135

-------
Chemical-Specific Source Evaluation
The LRSA conducted studies to locate the major
sources of ammonia in the collection system.  Key
manholes and industrial discharges were sampled and
tested for total ammonia from 1990 through 1992. the
results  indicated  one  major industrial source of
ammonia in the collection system. Based on the survey
results, the  LRSA  developed  and  implemented
pretreatment  limits  to reduce  effluent ammonia
concentrations (LRSA,  1993a).

Toxicity-Based Source Evaluations
The toxicity-based approach used RTA procedures that
involved treating  industrial wastewater  samples in
bench-scale, batch simulations of the POTW activated
sludge process and measuring the resulting  toxicity
(USEPA,  1989a).  The toxicity remaining after batch
treatment,  referred  to  as   "refractory"  toxicity,
represented the toxicity that passes through the POTW
and causes effluent toxicity. As shown in Figure 5-2
(Section 5), two types of batch reactors are tested. A
control reactor simulated the treatment plant and
treated only the POTW influent. The second reactor
evaluated the addition of the industrial discharge to the
POTW by treating industrial wastewater spiked into
the POTW influent. An industrial discharge would be
considered a source of toxicity if effluent from the
spiked reactor was more toxic than the control reactor
effluent.

Initial RTA tests conducted during the USEPA study
indicated  that refractory toxicity was limited to an
industrialized area of the collection system. Following
the USEPA study, ammonia  was confirmed as the
primary cause of effluent toxicity and the major source
of ammonia was identified. Accordingly, subsequent
RTA  tests  focused   on  identifying  sources of
non-ammonia toxicity.  In 1992, RTA testing was
performed to evaluate  sources of non-polar organic
toxicity because non-polar organic compounds had
been identified as a major cause  of non-ammonia
toxicity.

The procedure for  measuring  non-polar  organic
toxicity involved passing the RTA batch  effluent
samples through  a  CIS  SPE column,  eluting the
column with methanol, and performing a toxicity test
on  the methanol  elution (LRSA,  1992a).   This
procedure provided a direct means of measuring non-
polar organic toxicity and it eliminated interferences
associated with  toxic  ammonia concentrations
(i.e., ammonia was not captured by or eluted from the
CIS SPE column).

The toxicity source evaluation identified two industrial
dischargers of  non-polar organic toxicity (LRSA,
1992b). Nonpolar organic toxicity tests performed on
the effluent during this period suggested that non-polar
organic toxicity was variable and that there may be
other causes  of non-ammonia  toxicity.   Therefore,
further RTA testing was conducted in 1993 to identify
sources of non-ammonia toxicity that may be caused by
non-polar organic compounds and other unidentified
compounds.

The ammonia pretreatment limits were not to become
effective until after July 1995;  therefore, the LRSA
influent and effluent ammonia concentrations remained
high  during  1993.   It was necessary  to remove
ammonia toxicity in RTA testing in order to identify
sources of non-ammonia toxicity (LRSA, 1993b).
Zeolite  treatment  of  the  batch effluent  samples to
remove ammonia was considered, but previous studies
indicated that zeolite also may remove non-ammonia
toxicity.  Therefore, two alternative approaches were
used to remove ammonia toxicity in the RTA.  First,
testing was conducted during periods of low influent
ammonia concentrations, which occurred during the
annual summer shutdown of the ammonia-contributing
industrial  process.  During this period, ammonia
concentrations werb not acutely toxic; therefore, RTA
testing  would  provide a  direct measure of the
non-ammonia toxicity contributed to the POTW. The
second  approach  was  used  when  the ammonia
contributing process was fully operational and involved
using a simulated plant  influent (SPI).    The SPI
consisted of sewer wastewater collected from all major
trunk lines except the sewer line serving the ammonia
discharger. It was also necessary to wash  the RAS
used in the RTA to reduce the ammonia concentrations
associated with the RAS (LRSA, 1993c).

The 1993 RTA testing was intended to identify those
industries that would be required to meet pretreatment
requirements   to  control  non-ammonia  toxicity.
Thirty-two of the 40 industrial  users were evaluated
either directly or indirectly by testing sewer wastewater
samples collected from key manholes. Previous RTA
results and information  obtained in an industrial user
waste survey  were used to select the industries to be
tested.
                                                136

-------
The results  of RTA  tests performed  in July  and
October 1993 are presented in Table  G-2.  If the
effluent toxicity of the sewer wastewater spiked reactor
was greater  than  that of the control reactor on  two
occasions, the discharge was considered a source of
toxicity. Industries A, B, E,  and F were indicated as
sources of non-ammonia toxicity based on the results
of direct testing of their industrial discharges. These
results support the findings of the USEPA study, which
identified industries A, B, and E as sources of toxicity,
and the 1992 study, which identified industries B and
E as sources of non-polar organic toxicity. Six other
industries were identified as suspected sources based
on  the results  obtained for key manholes 9 and 12.
LRSA plans to test these suspected sources directly to
determine which industries are contributing toxicity.
Table G-2. Results of Refractory Toxicity Assessment, July and October 1993*
RTA Reactor
Effluent
Control Reactor
Spiked Reactors
Industry A
Industry B
Industry C
Industry D
Industry E
Industry E 5x f
Industry F
Industry G
Industry H
Industry I
Industry J
Key manhole 1
Key manhole 3
Key manhole 4
Key manhole 7A
Key manhole 9 ง
Key manhole 10
Key manhole 12 #
Key manhole 14
Key manhole 15
Roselle flume
96-Hour Mysid TUa (100/LC50)
JullS
<1.0

<1.0
1.45
<1.0
NT
NT
NT
NT
NT
NT
NT
NT
<1.0
NT
NT
<1.0
1.1
NT
1.33
NT
NT
NT
Jull6
<1.0

NT
NT
NT
<1.0
<1.0
NT
<1.0
NT
NT
NT
NT
NT
NT
<1.0
NT
NT
NT
NT
NT
<1.0
<1.0
Jul22
1.63

1.92
1.89
NT
NT
NT
NT
NT
<1.0
NT
NT
NT
NT
NT
NT
NT
NT
1.33
1.81
1.33
NT
NT
Jul23
1.05

NT
NT
NT
NT
1.19
4.0
2.18
NT
NT
NT
NT
NT
1.12
NT
<1.0
<1.0
NT
NT
NT
NT
NT
Octl9
2.0

3.39
NT
NT
NT
NT
NT
1.55
NT
2.28
NT
NT
NT
t
NT
NT
6.1
NT
1.71
NT
NT
NT
Oct20
1.75

1.22
NT
NT
NT
1.75
NT
1.86
NT
NT
1.29
1.81
NT
NT
NT
NT
NT
NT
1.63
NT
NT
NT
Source of Refractory
Toxicity?0
n/a

YES
YES
NO
NO
YES
YES
YES
NO
NO
NO
NO
NO
NO
NO
NO
YES
NO
YES
NO
NO
NO
 *   Spiked reactor results shown in bold indicate greater TUa than the control.  Increased toxicity in the spiked reactor
     effluent compared to the control indicates a source of refractory toxicity.
 t   Tested at five times the normal flow contribution to evaluate anticipated increase in flow.
 t   Toxicity test was invalid based on unacceptable control survival.
 ง   Key manhole 9 receives wastewater from three industries.
 #   Key manhole 12 receives wastewater from three industries.
 w   If a spiked reactor result was greater than that of the control on two occasions then the discharge was considered a source
 of  refractory toxicity.
 NT Not tested.
                                                    137

-------
Toxicity Control Evaluation
The LRSA evaluated control options for ammonia and
non-ammonia toxicants. The objective was to identify
and assess the available options and to determine the
most  cost effective and  pragmatic approaches for
reducing effluent toxicity to acceptable levels.

Ammonia Toxicity Control Evaluation
A modified  acute  toxicity  test  procedure  was
developed by the LRSA and approved by the NJDEP
to control pH drift in the toxicity test.  The pH in
previous LRSA compliance tests typically drifted up to
8.0 to 8.5, which resulted in an overestimation of
ammonia toxicity (i.e., unionized ammonia concen-
trations increase as pH increases). The modified test
procedure maintains pH  in the toxicity test at the
receiving  system pH of 7.4.   This modification
provides a more accurate measurement of instream
ammonia toxicity.

Using ammonia toxicity values for mysids published
by USEPA (1989d), a linear regression model was
prepared to predict the concentration of ammonia in
the effluent which, in the absence of other toxicants,
should result in compliance with the acute toxicity
limit.  The ammonia value generated by the model
accounts for  toxicity test conditions  that affect the
concentration  of unionized  ammonia  (e.g.,  pH,
temperature, and salinity). The model determined that
the acute toxicity limit could be met with an effluent
ammonia concentration of 35 mg/L (LRSA, 1991).

Several options forin-plant treatment of ammonia were
evaluated to achieve the  ammonia  target level.  As
shown in Table 6-1 (Section 6), none of the six options
evaluated was practical based on technical and cost
considerations.  In addition, significant inhibition of
nitrification was observed during treatability tests,
indicating that inhibitory compounds would need to be
controlled if  nitrification was selected as a control
option (LRSA, 1991).  Based on these results and the
results of the ammonia source evaluation, chemical-
specific pretreatment limits were selected as the best
approach for controlling toxicity caused by ammonia
(LRSA, 1993a).                               \
                                              i
Non-Ammonia Toxicity Control Evaluation
The TIE indicated that the causes  of non-ammonia
toxicity were complex and highly  variable and the
specific compounds causing non-ammonia toxicity
could not be identified and confirmed. Consequently,
the necessary information was not available to develop
chemical-specific pretreatment limits.

As an alternative to pretreatment  limits, activated
carbon treatment at the POTW was evaluated based on
its effectiveness in reducing effluent toxicity caused by
a variety of compounds including non-polar organic
toxicants.  Both  PAC  and GAC  treatment  were
considered and found to be cost prohibitive  (T.L.
Morris, Technical Memorandum to LRSA, Evaluation
of Granular Activated Carbon at LRSA, January 19,
1993). It also was determined that the use of PAC
treatment would result in unacceptable sludge quality.

The LRSA elected to implement pretreatment controls
because  controllable  industrial  sources  of  non-
ammonia toxicity had been identified  and  practical
in-plant treatment options were not available. It was
determined  that the pretreatment  limits must  be
toxicity-based  because  of  the  lack  of  specific
information on the causes of non-ammonia toxicity.
The proposed pretreatment approach involved  RTA
testing to determine which industries should be issued
limits and which industries should  be  monitored to
assess the need for future limits (LRSA, 1993c).

Implementation Of Toxicity Controls
Ammonia Pretreatment Limits
The approach used to develop pretreatment limits for
ammonia was relatively straightforward. As required
by the AGO, the  LRSA submitted  a work plan for
developing ammonia pretreatment limits to the NJDEP
in April 1992 and the plan was approved in May 1992
(LRSA, 1992c). Using the target ammonia level of 35
mg/L and the ammonia survey data,  an allowable
headworks loading  approach (USEPA,  1987) was
followed to  develop draft pretreatment limits.  The
LRSA published the draft limits for public notice and
comments were received and reviewed.  In January
1993, the proposed ammonia pretreatment limits and
the LRSA's response to public comments  were
submitted to the NJDEP. The limits were approved in
March 1993  and industrial users were to comply with
the limits by July 1995 (LRSA,  1993a).

Toxicity-Based Pretreatment Limits for
Non-Ammonia Toxicity
The LRSA is one of the first municipalities to develop
toxicity-based  pretreatment  limits  to  control  non-
ammonia toxicity. At the time of this study, toxicity-
based pretreatment limits  had not  been applied
                                                138

-------
elsewhere and. there was  no specific guidance on
developing such  limits.  The selected approach was
based on the available TRE information and involved
several  aspects of various pretreatment approaches
recommended by USEPA (1987).

The LRSA submitted a work plan for development of
the limits to the NJDEP in June 1993 (LRSA, 1993b).
The proposed approach was designed to address both
major and minor sources  of non-ammonia toxicity
(LRSA, 1993c) and to  ensure compliance without
unnecessary controls. The proposed limits will consist
of the following components referred to collectively as
a toxicity management program (TMP):

  •  Narrative local pretreatment limit of "no discharge
    of refractory toxicity."
  •  Pass/fail  toxicity-based  limit using  the  RTA
    procedure as a compliance test (i.e., the effluent
    LC50 of the industrial user spiked reactor may not
    be  less  than the LC50  of the  control reactor
    effluent).
  •  Industrial  user  (if toxicity  is found)  may be
    required  to   implement  a  toxicity  reduction
    program  comprising  requirements to  identify
    causes  and   sources  of  toxicity,  implement
    industrial user managementpractices, and evaluate
    and establish other controls to ensure compliance
    with the toxicity-based limits.
  •  RTA  monitoring requirements  and  decision
    criteria for determining if an industrial user needs
    to continue with the TMP.
  •  Provisions to allow industries to be relieved from
    the TMP requirements if toxicity requirements are
    met.
  •  Compliance  schedule  including milestones and
    progress reports.
  •  Reopener  clause stating that the pretreatment
    permit will  be modified to include  chemical-
    specific limits  if the causes of  toxicity  are
    identified.

The proposed pretreatment limit approach falls under
the case-by-case/best professional judgment approach
described by  USEPA  (1987),  but  also includes
toxicity-based  requirements,   industrial  user
management  practice,  and   chemical-specific
components.  The TMP  approach is consistent with
USEPA  recommendations   for  monitoring  and
controlling effluent toxicity through the NPDES.
The RTA procedures had not been used for compliance
monitoring purposes  in New Jersey.  Therefore, a
site-specific  RTA protocol  (LRSA,  1994)  was
submitted to the NJDEP for review and approval prior
to development of the draft pretreatment limits. The
RTA protocol was approved by the NJDEP in June
1996.   Pretreatment program permits for several
industries   were   modified  to  include  the  TMP
provisions. These industries are currently required to
conduct quarterly monitoring using the RTA protocol.

Discussion
Chemical-specific  pretreatment  limits  are  being
implemented to control toxicity caused by ammonia
and toxicity-based pretreatment limits are in  place to
control non-ammonia toxicity.  The major source of
ammonia ceased its discharge of the ammonia-laden
waste stream in 1997. As a result, effluent ammonia
concentrations at the LRSA treatment plant decreased
to about 30 mg/L!_A_co.mpliance test performed after
the ammonia source was eliminated showed improved
effluent quality (i.e.-, LC50 = 72%).  Additional tests
are planned to confirm this initial result.

It is possible that the ammonia pretreatment  limits
alone will  achieve compliance with the acute effluent
toxicity limit.  However, due to the complex and
variable nature of the non-ammonia toxicity,  it  is not
possible to accurately predict if the ammonia reduction
will achieve consistent-compliance with the permit
limit LC50 ;>50%).    The LRSA  has  established
pretreatment requirements for non-ammonia toxicity to
ensure full and timely compliance with the toxicity
limit.    The need for industrial  users  to  control
non-ammonia toxicity is ultimately tied to compliance
with the acute effluent toxicity limit.  If necessary,
industrial  users   may  request  relief  from  these
requirements if the effluent consistently complies with
the acute effluent toxicity limit.

Acknowledgments
The USEPA research study was funded by the Office
of Research   and Development,  Risk  Reduction
Engineering Laboratory in  Cincinnati,  Ohio.   The
LRSA assisted USEPA and its technical consultant,
Engineering-Science,  Inc.  (Fairfax,  Virginia),  in
conducting this study. TIE and toxicity analyses for
the USEPA study were performed by EA Engineering,
Science and Technology, Inc. (Sparks, Maryland), and
the USEPA's National Effluent Toxicity Assessment
                                                139

-------
Center  (NETAC).   Subsequent  TRE work was
performed by Engineering-Science, Inc., in association
with EA. Engineering-Science, Inc., and the LRSA
acknowledge the assistance of Gary Fare and Judy
Spadone (LRSA), the LRSA Board members, William
Goodfellow (EA), and John Botts, Mark Collins, Tim
Morris, and Tim Schmitt (Engineering-Science, Inc.).

The  material presented in this appendix  includes
copyrighted   information  published in  text  by
Technomic Publishing Company, Lancaster, PA (Botts
et al., 1992). Technomic Publishing Company has
granted  permission to include the information in this
document.

References
Botts, J.A., T.L. Morris,  J.E.  Rumbo, and  C.H.
    Victoria-Rueda.     1992.     Case  Histories-
    Muncipalities. In Toxicity Reduction: Evaluation
    and Control.    D.L.  Ford,   ed.  Technomic
    Publishing Co., Lancaster, Pennsylvania.

Engineering-Science, Inc. 1991.  Memorandum from
    Tim Morris (ES) to Gary Fare and Judy Spadone
    (LRSA)  -  1991 LRSA  Toxicity Identification
    Evaluation Results. July 1991.

Linden Roselle Sewerage Authority.  1990a.  Linden
    Rosclle Sewerage Authority Annual Pretreatment
    Program Report.  Submitted to the New Jersey
    Department  of  Environmental  Protection,
    November 30,1990.

Linden Roselle Sewerage Authority.  1990b.  Linden
    Roselle   Sewerage  Authority   1990  Toxicity
    Reduction Evaluation Project Report. Prepared by
    Engineering-Science, Inc., Fairfax, Virginia, in
    association  with EA Engineering, Science and
    Technology, Inc., Sparks, Maryland.

Linden Roselle Sewerage Authority.  1991.  Linden
    Roselle   Sewerage  Authority   1991   Toxicity
    Reduction Evaluation Project Report. Prepared by
    Engineering-Science, Inc., Fairfax, Virginia, in
    association  with EA Engineering, Science and
    Technology, Inc., Sparks, Maryland.

Linden Roselle Sewerage Authority.  1992a.  Linden
    Roselle Sewerage Authority NPO Toxicity Source
    Evaluation Test Plan. Prepared by Engineering-
    Science, Inc., Fairfax, Virginia.              i
Linden Roselle Sewerage Authority. 1992b. Linden
    Roselle Sewerage Authority 1992 NPO Toxicity
    Source Evaluation Summary Report. Prepared by
    Engineering-Science, Inc., Fairfax, Virginia, in
    association  with EA Engineering, Science  and
    Technology, Inc., Sparks, Maryland.

Linden Roselle  Sewerage Authority.  1992c.  1992
    LRSA Toxicity Reduction Program: Work Plan for
    Developing  Ammonia  Pretreatment  Limits.
    Prepared by Engineering-Science, Inc., Fairfax,
    Virginia.    Approved  by  the  New  Jersey
    Department of  Environmental Protection  and
    Energy, May 12, 1992.

Linden Roselle Sewerage Authority. 1993a. Proposed
    Ammonia Pretreatment  Limits.   Prepared by
    Engineering-Science,  Inc.,  Fairfax,  Virginia,
    November 1992. Approved by the New Jersey
    Department of  Environmental Protection  and
    Energy, March 10, 1993.

Linden Roselle  Sewerage Authority.  1993b.  7993
    LRSA Toxicity Reduction Program: Work Plan for
    Developing Secondary Toxicant  Pretreatment
    Limits. Prepared by Engineering-Science, Inc.,
    Fairfax, Virginia.

Linden Roselle  Sewerage Authority.  1993c.  LRSA
    Toxicity Reduction  Program  1993:  Toxicity
    Source Evaluation Final Report.  Prepared by
    Engineering-Science, Inc., Fairfax, Virginia.

Linden Roselle  Sewerage Authority.  1994.  Draft
    Protocol for  Conducting Refractory Toxicity
    Assessment  Testing  Under the Linden Roselle
    Sewerage  Authority   Industrial  Pretreatment
    Program. Prepared by Engineering-Science, Inc.,
    Fairfax, Virginia.

Morris, T.L., J.A. Botts, J.W. Braswell, M.C. Welch,
    and W.L. Goodfellow.  1990. Toxicity Reduction
    Evaluation  at the Linden  Roselle  Sewerage
    Authority Wastewater  Treatment Plant.   Draft
    Report.  Prepared for the USEPA, Office of
    Research  and  Development,  Risk Reduction
    Engineering  Laboratory,  Cincinnati,   Ohio.
    USEPA Contract No. 68-03-3431.

Morris, T.L., G. Fare, and J. Spadone. 1992. Toxicity
    Reduction  Evaluation  at the Linden  Roselle
                                               140

-------
    Sewerage Authority Wastewater Treatment Plant.
    Water, Environment, and Technology, June: 8-16.

USEPA. 1987. Guidance Manual on the Development
    and  Implementation  of  Local  Discharge
    Limitations  Under the  Pretreatment Program.
    EPA Office of Water Enforcement and Permits,
    Washington, D.C.

USEPA.   1988.   Methods  for  Aquatic   Toxicity
    Identification   Evaluations:  Phase I  Toxicity
    Characterization Procedures. EPA/600/3-88/034.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.   1989a.   Toxicity  Reduction  Evaluation
    Protocol for Municipal Wastewater Treatment
    Plants. EPA/600/2-88/062.  Office of Research
    and Development, Risk Reduction Engineering
    Laboratory, Cincinnati, Ohio.

USEPA.  1989b.  Methods for  Aquatic  Toxicity
    Identification  Evaluations:  Phase  II  Toxicity
    Identification  Procedures. EPA/600/3-88/035.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.  1989c.  Methods for  Aquatic  Toxicity
    Identification  Evaluations: Phase III  Toxicity
    Confirmation  Procedures. EPA/600/3-88/036.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA. 1989d. Ambient Water Quality Criteria for
    Ammonia (Saltwater). EPA/440/5-88/004. Office
    of Water Regulations and Standards, Criteria and
    Standards Division, Washington, D.C.
                                               141

-------
                                         Appendix H

         Toxicity Control Options for Organophosphate Insecticides
Organophosphate insecticides, including diazinon,
chlorpyrifos, malathion, and chlorfenvinphos, have
been  found to cause effluent toxicity at POTWs
throughout the United  States (Norberg-King et al.,
1989; Amato et al., 1992; USEPA, 1987; Botts et al.,
1992; Kllmore et al., 1990).  A case study of the
occurrence of Organophosphate insecticide toxicity at
POTWs in the San Francisco Bay area is presented in
Appendix F.  Although procedures are available for
identifying Organophosphate toxicants, less is known
about how to control Organophosphate insecticides in
POTW effluents. This  section describes approaches
for Organophosphate toxicity control that have been
successfully implemented at POTWs. Information is
also presented on ongoing research  into  POTW
operational improvements that may reduce  effluent
concentrations of Organophosphate toxicants.

A review of the literature suggests that two approaches
may be  successful in  reducing  Organophosphate
compQunds at POTWs:

  •  Public education  to  limit  the  discharge  of
    Organophosphate compounds to the POTW.
  •  POTW  modifications,   particularly  involving
    enhancements to the biological treatment and
    chlorine disinfection processes.

The latter approach has been the subject of a research
study being funded by the two principal manufacturers
of Organophosphate compounds in North America:
Novartis Crop Protection, Inc., and Makhteshim-Agan
of North America, Inc.

Public Education Approach
Organophosphate insecticides are used widely for pest
control by homeowners, restaurants, veterinarians, and
other commercial businesses.  These sources are not
readily controlled by pretreatment program regulations.
Alternative efforts to minimize the use or disposal of
Organophosphate insecticides must have broad appeal
to the public at large.

Organophosphate  insecticide control measures  that
have been considered by POTW staff include public
outreach and education programs and approaches to
restrict  the use  of  Organophosphate  compound
applications.  Efforts  to ban or restrict the use of
Organophosphate insecticides have not been successful,
largely because of concern about legal issues and the
difficulty in controlling the  sale of Organophosphate
compounds outside of the community.

Restrictions on Organophosphate Insecticide
Use             '
In 1990, the City of Largo, Florida, evaluated the
feasibility of banning the use of diazinon and other
Organophosphate   insecticides  (malathion   and
chlorfenvinphos)  to  control effluent  toxicity  (C.
Kubula,  personal  communication,  City  of Largo,
Florida, 1992). It was determined that a diazinon ban
would likely increase the use of other, equally toxic,
insecticides.   For  example,   Dursbanฎ, a likely
alternative insecticide,  contains chlorpyrifos, which
has been found to be more toxic than diazinon. Also,
restrictions on diazinon use would apply only to new
supplies, not to insecticides already in stock at stores.
The  City of Largo estimated that the  stockpiled
diazinon would last for more than a year. An effective
control program would also require the cooperation of
neighboring communities in limiting the purchase of
diazinon outside of the community.  In addition, the
local banning of federally approved insecticides would
be controversial.   It was anticipated that insecticide
manufacturers and distributors  would challenge the
City's authority to implement such controls. Based on
                                               142

-------
this analysis, the City of Largo determined that banning
diazinon would not be a practical control option.

Public Education Campaigns
Based on the impracticality of insecticide bans, the
City of Largo elected to pursue a public awareness
approach to control diazinon toxicity.  The City of
Greenville,  Texas,  also   implemented   a  public
education program in 1990 (City of Greenville, 1991).
The first year of the program focused on determining
significant users of  the insecticide and developing
educational materials.  The  following years  have
involved distributing the materials and conducting
other informational activities.

The City of Greenville initially identified nine groups
of diazinon users: pest control businesses, lawn care
businesses, veterinarians, animal  shelters, janitorial
services, apartment complexes, restaurants, hotels, and
retail stores (City of Greenville, 1991). The residential
population also was added as a target user group. The
City  service  area was divided into sections,  and a
telephone survey was conducted.  Information was
gathered on diazinon use, including existing supplies
and  application  and waste disposal practices, and
business owners and homeowners were notified of the
importance of controlling diazinon  wastes.   The
program involved the  following  public  education
activities:

  • Brochures and handouts
  • Pest control fact sheets describing integrated pest
    management  methods,  which   focused  on
    minimizing insecticide usage
  • Mass mailings
  • Newspaper articles
  ซ Public service announcements
  • Occasional talk shows on local radio stations
  • Biweekly presentations to schools and business
    groups      •
  • A telephone information line.

The City of Greenville also enacted an ordinance to
encourage environmentally sound use of insecticides.
The  ordinance requires retail vendors, pest control
services,  and  apartment  managers  to  distribute
educational material to customers  and  to periodically
report insecticide applications to the City.

The results of the Greenville education campaign are
encouraging.   Beginning  in December  1993,  the
treatment plant effluent was not toxic to C. dubia for 3
consecutive months. The public awareness effort is
continuing and the City will monitor its effect on
toxicity reduction.

The  City of  Largo  initiated a  public  education
campaign in  1992.  An information brochure was
prepared and distributed in 1993.  Effluent toxicity
decreased; however, it was not known if the reduction
is related to the public education program.  A strong
emphasis has not been placed on the program because
the City has opted for a land irrigation treatment
system in lieu of continued effluent discharge.

As noted in Appendix A of this manual, diazinon and
its  toxic metabolite  diazoxon  were  tentatively
identified as effluent toxicants at the City of Lawton
POTW.  The City decided to implement a public
awareness program in 1993 to control the discharge of
insecticides to the POTW (Engineering Science, 1993).
Information  on the   proper  use  and disposal  of
insecticides was printed in newspaper articles and on
monthly water bills. An electronic message sign with
insecticide information also  was located  at major
intersections. Since August 1993, the POTW effluent
has  met the  toxicity  permit  limit (NOEC >96%
effluent) with the exception of 2 months in 1994 and
several  months  in 1995 (as of September 1997).
Although diazinon was not confirmed as an effluent
toxicant, the City's ongoing insecticide control effort
appears  to  have  been  successful  in  achieving
compliance with the chronic toxicity limit.

POTW Operational Improvements
Diazinon Treatment
In 1992, Novartis Crop Protection, Inc., in cooperation
with  Makhteshim-Agan of  North America,  Inc.,
initiated a study on diazinon and its relationship to
effluent  toxicity at POTWs (Novartis, 1997).  A
principal objective of the study was to determine the
treatability of diazinon and assess its fate in POTWs.
Research on this subject included a survey of POTWs
in which organophosphate insecticide toxicity was
observed and  bench-scale  treatability  tests  were
conducted to evaluate diazinon removal by various
treatment methods  and operating conditions.

Two types of POTW biological treatment  processes
were investigated in  the Novartis study:  fixed film
(trickling filter and  RBC)  and activated  sludge.
Influent and effluent concentrations at several POTWs
in the southwestern United States were compared to
determine removals  of diazinon and  chlorpyrifos.
                                                 143

-------
 Overall, the data indicated that diazinon reduction
 could be achieved in conventional POTW treatment
 processes.  A statistical analysis of the data showed
 that the fixed film process had a significantly lower
 percent removal  (p=0.95) for diazinon than  the
 activated sludge process or a combined fixed film/
 activated sludge process. A similar trend was observed
 for chlorpyrifos, although no significant differences
 were found between the process types.

 Bench-scale  treatability  testing was conducted  to
 further evaluate the fate of diazinon in typical POTW
 processes.  These tests considered the effect of design
 and operating conditions  for  biological treatment
 processes on diazinon removal and effluent toxicity.
 Additional tests  were performed to investigate  the
 effect  of  physical/chemical  processes,  including
 chemical precipitation, chlorination/dechlorination, and
 post aeration on diazinon concentrations and toxicity.

 As shown in Figure H-l, a correlation was found to
 exist between diazinon removal and sludge retention
 time  (SRT),  HRT,  and  MLSS  concentration  in
 activated sludge treatment tests.  The primary removal
 mechanism in the activated sludge tests was adsorption
 onto the biological solids. These results  suggest that
 diazinon removal may be improved by increasing the
 SRT, HRT,  and/or MLSS  concentration  of the
 treatment process.

 Auxiliary  process   studies  provided  additional
 information on treatment of diazinon (Novartis, 1997).
 Chemical  precipitation using  ferric chloride and
 polymer only slightly reduced diazinon  levels. No
 major change in diazinon concentrations was observed
 whether the  coagulants  were  added  to  primary
 wastewater or secondary  treated wastewater prior  to
 clarification.  Chlorination treatment was effective  in
 reducing diazinon from secondary clarifier effluent;
 however, chronic toxicity was unchanged. Qualitative
 results suggest that the chlorine oxidized diazinon  to
 diazoxon, a  by-product  that exhibits similar toxic
 effects as diazinon. Post aeration of secondary clarifier
 effluent also reduced diazinon levels; however, once
 again, chronic toxicity was not significantly changed.
 Again, it was assumed that diazinon was oxidized  to
 diazoxon.                                        !

Additional  tests evaluated  the  fate  of diazinon  in
 POTWs (Novartis, 1997). Anecdotal evidence from
other studies (Fillmore et al., 1990) and the treatability
studies suggested that adsorption onto solids was the
dominant removal mechanism.  Therefore, the tests
focused on partitioning of diazinon and chlorpyrifos
onto primary and  mixed liquor solids.  These tests
showed that about 30% of the diazinon and 85 to 90%
of the chlorpyrifos present in POTW primary influent
samples  is adsorbed onto primary influent solids.
Mixed  liquor   adsorption  results  revealed   that
approximately 65 to 75% of the diazinon added to the
mixed liquor adsorbed onto the biomass.  Diazinon
adsorption was greater for a 30-day SRT biomass than
for a 15-day biomass. Chlorpyrifos strongly adsorbed
to the biomass; 100% was removed.

Summary
Studies have shown that organophosphate compounds
can be effectively controlled through public education
(City of Greenville, 1991; Engineering Science, Inc.,
1993).  This effort may vary from the distribution of
educational materials to the enactment of ordinances
that require strict accounting of insecticide use. The
studies conducted to date indicate that characterization
of the sources of organophosphate compounds is key
to the development of a successful toxicity control
program.

Recent  information  shows  that  relatively simple
enhancements to  POTWs  may  help to  reduce
organophosphate   compounds.    Factors  affecting
diazinon and chlorpyrifos removal include the SRT,
HRT, and MLSS concentrations in activated sludge
processes,  chlorination/dechlorination,  and  post
aeration.  Further  studies are  in  progress to better
define  the operating conditions that will  promote
organophosphate  compound removal (D.  Tierney,
personal  communication, Novartis Crop Protection,
Inc., 1997).
                                                144

-------
    70 -r
_  65
ฃ  60--
1  55 --
ts  50
S  45 +
ฃ  40 +
    35 --
    30
       I
       ง
       I
           70 T
           30
            1,000
    70 -•
    65 -•
    60 -•
    55 -•
    50 --
    45 -•
    40 ••
    35 --
    30 --
                                             R2 = 0.543
                                                                  10
                                                                        12
                                       HRT(hrs)
                      1,500
      2,000
MLSS (mgfl)
2,500
                                                                       3,000
                                              R2 = 0.9956
                                               10
                                                               15
                                                                               20
                                       SRT (days)
Figure H-1. Diazinon removal as a function of SRT, HRT, and MLSS concentration (reprinted with the permission
of Novartis Crop Protection, Inc.) (Source: Novartis, 1997).
                                            145

-------
References
Amato,  J.R.,  D.I.  Mount,  EJ.   Durban,  M.L,
    Lukasewycz,  G.T. Ankley,  and E.D.  Robert!
    1992.   An Example  of the Identification of
    Diazinon as a Primary Toxicant in an Effluent;
    Environ. Toxicol. Chem.  11: 209-16.           ;

Bolts,  J.A., T.L. Morris, J.E.  Rumbo, and  C.H.
    Victoria-Rueda. 1992. Case Histories - Munici-
    palities.  Toxicity Reduction:  Evaluation and
    Control.   D.L. Ford,  ed.    Lancaster,   PA:
    Technomic Publishing Co.

City of Greenville. 1991. TRE Phase B Final Report.
    City of Greenville, North Carolina.
                                               |
Engineering-Science, Inc. 1991.  Toxicity Reduction
    Evaluation Phase I Report. Submitted to the City
    of Lawton, Oklahoma, September 1991. Parsons
    Engineering Science, Inc., Fairfax, Virginia.
                                               j
Engineering-Science, Inc. 1992.  Toxicity Reduction
    Evaluation: Toxicant Identification and Confirma-
    tion. Submitted to the City of Lawton, Oklahoma,
    Parsons  Engineering  Science,  Inc.,  Fairfax,
    Virginia.

Engineering-Science, Inc.  1993. Review of Diazinon
    Control Options. Submitted to the City of Lawton,
    Oklahoma, April 1993.  Parsons Engineering
    Science, Inc., Fairfax, Virginia.

Fillmore, L.B., T.L. Morris, T.L. Champlin,  M.C.
    Welch,  and J.A. Bolts.  Draft  1990.  Toxicity
    Reduction Evaluation at the City of Fayetteville
    Cross Creek Wastewater Treatment Plant.  Draft
    Report.  USEPA Risk Reduction Engineering
    Laboratory, Cincinnati, Ohio.

Norberg-King, T.J.,  M. Lukasewycz, and J.  Jensen.
    1989. Results of Diazinon  Levels  in  POTW
    Effluents in the  United States, Technical Report
    14-89.   USEPA,  National  Effluent  Toxicity
    Assessment Center, Dululh, Minnesola.
Novartis Crop Protection, Inc., and Makhteshim-Agan
    of North America, Inc.  1997.  Investigation of
    Diazinon Occurrence, Toxicity, and Treatability in
    Southern United States Publicly Owned Treatment
    Works.  Technical Report 3-7, Environmental
    Affairs Departmenl, Greensboro, North Carolina.

USEPA. 1987. Toxicity Identification Evaluation of
    the Largo, Florida POTW Effluent. J.R. Amalo,
    D.I. Mounl, M. Lukasewycz, E. Durham, and E.
    Robert.  National Effluent Toxicity Assessment
    Center, Duluth, Minnesola.

USEPA.  1989a.  Methods  for  Aquatic  Toxicity
    Identification Evaluations: Phase II:  Toxicity
    Identification  Procedures. EPA/600/3-88/035.
    Office of  Research  and Development Duluth,
    Minnesola.

USEPA.  1989b.  Methods  for  Aquatic  Toxicity
    Identification   Evaluations,     Phase   HI.
    EPA/600/3-88/036.   Toxicily   Confirmalion
    Procedures. Office of Research and Development,
    Dululh, Minnesola.

USEPA. 1992. Toxicity  Identification Evaluation:
    Characterization  of Chronically Toxic Effluents,
    Phase I. EPA/600/6-91-005F. National Effluenl
    Toxicily Assessmenl Center, Dululh, Minnesota

USEPA.  1993a.  Methods  for  Aquatic  Toxicity
    Identification  Evaluations: Phase II  Toxicity
    Identification Procedures for Samples Exhibiting
    Acute and  Chronic Toxicity. EPA/600/R-92-080.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.

USEPA.  1993b.  Methods  for  Aquatic  Toxicity
    Identification Evaluations: Phase III  Toxicity
    Confirmation Procedures for Samples Exhibiting
    Acute and  Chronic Toxicity. EPA/600/R-92-081.
    National Effluent Toxicity Assessment Center,
    Duluth, Minnesota.
                                               146

-------
                                         Appendix I

                     Pretreatment Program Chemical Review
Introduction
It may be possible in limited cases to identify the toxic
influent sources by comparing pretreatment program
data on suspected sources to chemical-specific and
toxicity data on the POTW effluent. The objective of
the PPCR is to determine the sources of toxicity by
comparing chemical data on industrial dischargers to
toxicity data  reported  in the   literature.     The
pretreatment program information should include flow
and chemical monitoring data on the industrial users,
descriptions and schedules of industrial production
campaigns, and inventories  of chemicals used in
production. The final outcome of this review should
be  an improved understanding  of the industries'
processes  and chemical  usage,   and the possible
identification  of  sources  of  toxicity.    Source
identification through the PPCR approach has been
successful in reducing effluent toxicity at POTWs with
a limited number and type of industrial inputs (Diehl
and Moore, 1987).

General Procedure
The main  steps in a PPCR are to:

  • Gather the pertinent pretreatment program data
  • Compare  the data to POTW effluent  toxicity
    results and/or TIE data
  • Identify potential influent source(s) of toxicity
  • Evaluate  and recommend a toxicity  control
    option(s).

A brief description of each of these steps follows.

Collect Data on Industrial Users
Data on all categorical, significant non-categorical and
other potential toxic dischargers (e.g., industrial users
with local limits and RCRA and CERCLA inputs)
should be collected.  A list of pertinent information
that should be considered in a PPCR is presented in
Table 2-3. The data collection effort should include a
survey of each industrial  user, using the example
checklist shown in Table 1-1.

Information  on  chemicals that  may be used in
manufacturing processes can be  obtained from the
Encyclopedia of Chemical Technology (Kirk-Othmer,
1982).  Although OSHA regulations require  that
information on hazardous  chemicals is to be made
available to  the public on MSDSs, information on
various "specialty" chemicals can be difficult to obtain.
When data on a "specialty" chemical are not disclosed,
a literature review can be performed to determine the
chemical's acute toxicity and biodegradability. This
information allows assumptions to be made concerning
the biodegradability of the chemical at the POTW and
the potential for the chemical to.cause effluent toxicity.
An initial indication of the possible toxic pollutants
causing effluent toxicity can be made by comparing
expected or actual effluent concentrations to  toxicity
values provided in the literature.

Compare PPCR Data to POTW Effluent Toxicity
Results
Information on the magnitude, variability, and nature
of the POTW effluent toxicity can be compared with
the PPCR data to determine the soufces(s) of possible
problem chemicals.  This comparison can be made
using statistical analyses to determine if the variability
in the source characteristics can be  related to the
variability in  the POTW  effluent  toxicity.    A
description of data analysis techniques for comparing
POTW and industry pretreatment data follows.

Two types  of statistical  analyses can be  used  to
compare the pretreatment program chemical data and
POTW effluent toxicity data: linear regression (Draper
and Smith, 1966) and cluster analysis (Pielou, 1984;
Romesburg, 1984).
                                                147

-------
Table M. PPCR Data Sheet
  Industry Name
      Notes:

  Address
      Notes:

  Industrial Category (SIC Code)
      Notes:

 TRE Objectives
      Notes:

 Manufactured Products
      Notes:

 Chemicals Used
      Notes:

      Amounts (write on MSDS)
      Notes:

      MSDS

      Process in which chemical is used
      (write on each MSDS)
      Notes:

      Aquatic toxicity/biodegradability information
      on all chemicals used. Review MSDS,
      supplier information, and literature
      Notes:

 Engineering drawings of facility
     Notes:

     Production flowchart and line schematic
     Notes:

     All floor and process drains with schematic
     Notes:

     Wastewater pretreatment system schematic
     Notes:

 Facility records
     Notes:

     Water usage, water bills
     Notes:

     DMRs for 24 months
     Notes:

     Pretreatment system operations data
     Notes:

     Pretreatment system operator interview
     Notes:

     Spilt prevention control plan
     Notes:

     RCRA  reports, hazardous waste manifests
     Notes:
  All Attached
  None
_ Available


. Available


. Available





. Available


. Available


. Available


  Available


 Available


 Available
                          Partial Available
                          Some
 No
 No
No
No
No
No
No
No
                         No
                                                           148

-------
Linear regression analysis is used to find correlations
among the variables in the data base and to relate
changes in POTW effluent toxicity to the variables. A
cluster analysis using pattern recognition software can
weigh and evaluate the significance of toxics/toxicity
correlations.   The  determination of concentration
/response  relationships through  statistical  analysis
should not be considered as a definitive answer to
toxicity tracking because of the complexity of the
factors contributing to toxicity in POTW effluents.

The following  example illustrates how a stepwise
linear regression technique can be used in a PPCR
assessment. The technique is used  to identify how
changes in several variables can impact the presence
and variability of effluent toxicity. Table 1-2 presents
an  example data sheet for a POTW serving  one
manufacturing plant.  In this example, only a  few
POTW effluent industry variables were used in the
linear regression   analysis;  however,  additional
variables  also  could be  added in  the  regression
analysis.

The following variables are the "X" variables:

Industry variables:

  •  LBS  is the manufactured  product per month
    (millions of pounds).
  •  INFLOW is the discharge flow based on water
    usage (mgd).

Table 1-2. Data Sheet for Regression Analysis
POTW effluent variables:

  •  OELOW is the recorded effluent flow (mgd).
  •  COD is the chemical oxygen demand concentra-
    tion (mg/L).
  •  BOD5 is the biochemical oxygen demand concen-
    tration (mg/L).
  •  Cu is the copper concentration (mg/L).
  •  Cr is the chromium concentration (mg/L).
  •  Zn is the zinc concentration (mg/L).

The following variable is the "Y" variable:

  •  LC50 is the acute LC50 as percent effluent. •

By  applying standard stepwise linear regression, the
variables OELOW, BOD5, Cr, and Cu were eliminated
because they were insignificant to toxicity.  Stepwise
linear regression showed that  the remaining  (X)
variables were significant as regressed versus  (Y)
LC50. This analysis indicated that Zn, COD, LBS, and
INFLOW   were  correlated  with  POTW  effluent
toxicity.

Identify Source(s) of Toxicity
Based on the  data  analysis, a  list of the possible
contributors to effluent toxicity at the  POTW can be
developed. Sources of suspected toxicants should be
selected  based  on  toxicant loading  calculations.
Industrial  users  who contribute  potentially toxic
Month
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Parameter
LBS
0.80
1.01
1.20
1.25
1.16
0.90
0.90
1.20
1.30
1.27
1.10
0.90
INFLOW
1.2
1.5
1.7
1.7
1.6
1.2
1.2
1.6
1.8
1.7
1.6
1.2
OFLOW
1.0
'1.2
1.4
1.5
1.4
1.0
0.9
1.4
1.6
1.4
1.4
1.0
COD
30
33
41
39
30
28
25
23
25
26
30
40
BOD5
10
11
15
14
12
11
10
9
15
18
17
21
Cu
0.73
0.61
0.78
0.65
0.66
0.68
0.71
0.72
0.69
0.72
0.71
0.75
Cr
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
Zn
1.6
1.9
2.0
1.6
1.5
1.4
1.8
1.9
2.0
2.1
1.9
2.0
LC50
20
20
18
18
22
30
40
38
40
33
28
22
                                                149

-------
loadings of suspected toxicants would be candidates
for a toxicity control evaluation.

Recommend Toxicity Control Option(s)
Of the potential toxicity control options, toxic chemical
substitution  or  elimination  is usually  the most
pragmatic approach. Thus, a follow-up interview with
the toxic discharger(s) should be conducted to develop
information concerning techniques for the preferred
use of problem chemicals. A list of useful interview
questions is shown in Table 1-3. These questions may
enable the industry to identify problem areas and
possible  corrective actions in  the use  of  toxic
chemicals  in manufacturing.  Source  control  may
include  substitution  or elimination  of  problem
chemicals,  flow reduction, equalization, spill control,
and manufacturing process changes.
Table 1-3. Summary of the PPCR Chemical Optimization Procedure
 1.   Objectives

     a.   Optimize chemical usage amounts in production and water treatment processes.

     b.   Optimize chemical structures in process chemicals ensuring biodegraclability or detoxification is possible.

     c.   Establish process controls over incoming raw materials, measuring possible toxic components.  Example:
         corrosion-resistant finish put on steel by manufacturer that must be removed prior to part fabrication.
 2.   Strategy

     a.   Determine the role of each chemical in the process. This is done by supplier interviews and review of data
         gathered during the initial survey. Ask the questions:

         Can less of this chemical be used?

         Has the optimum amount been determined for each process?         ;

         Do other suppliers offer compounds that will perform as well at lesser concentrations?

         Is the compound in reality a part of the manufacturer's water treatment system and independent of product
         production?

         OBJECTIVE: Use less chemicals per pound of product produced.

     b.   Discover the biodegradability and toxicity of the process chemical. This is done by supplier interview, review
         of MSDS information, and literature search. Suppliers may not want to supply exact chemical formulations.
         In this case, ask industry to request supplier to perform tests to develop needed data. Questions to ask:

         What are the components in the product?

         What is its aquatic toxicity?

         Is the product biodegradable?             :

         What is the rate of biodegradation or half-life?

         Are there other component chemicals on the market that meet manufacturing requirements, but are low in
         toxicity and highly biodegradable?

         OBJECTIVE:  Use chemicals that will not create or contribute to toxicity problems.

    c.   Establish process controls over incoming raw materials.  Many raw materials have chemicals used in their
         manufacturing that are removed in the production of the final product.  Many raw materials may have trace
         contaminants that may cause toxic problems.  Questions to ask:

        What chemicals are used in the manufacturing of the raw material?

        What are the residual amounts of these raw material contaminants or by-products?

         Are there quality-control procedures that measure the amounts of these chemicals?

  	What are the statistical process measures used in the monitoring of these chemicals in the raw materials?
                                                 150

-------
Table 1-3. Summary of the PPCR Chemical Optimization Procedure (continued)
         If these chemicals are required to be removed before the raw materials can be used in manufacturing the
         final product, what purpose do the chemicals serve in raw material manufacturing?
         Can they be eliminated?
         Can they be made less toxic or more biodegradable?
         OBJECTIVE: Understand all raw materials being used and encourage development of QA procedures
         to monitor toxic chemicals removed during processing.
 3.  Outcome of Investigations
     a.  There will be a list of all chemicals used in processing and manufacturing of products. Included will be the
         amounts used, why the chemicals are used, and if optimization is being practiced.
     b.  MSDS sheets for all chemicals used will be on file.
     c.  A list of chemicals applied or used in the manufacturing of all raw materials will be on file under that raw
         material with the residual amounts noted if possible.
     d.  There will be a list of all chemicals and raw materials purchased on a monthly basis and the amount of
         product produced.
         OBJECTIVE: Hard information to be used in data analysis.
 4.  Use of opportunities available due to past experience
     a.  With experience in various industries, certain chemicals will become "known" as typically used in some
         process of manufacturing.
     b.  These known compounds can be categorized and toxicity determinations made. Once found toxic, the first
         information the industry must supply to the POTW staff conducting the TRE is whether these chemicals
         are used in its manufacturing process, in raw materials, or in water treatment processes.
     c.  Letters  also are sent to raw material suppliers  asking if these compounds are used in raw material
         production. If they are, the supplier is asked to  submit prototype alternative raw materials that do not
         contain these compounds.
     d.  This can be done at the beginning of the TRE for known problem chemicals. Indeed, control regulations
         also usually involve establishing limits for selected known toxics in industrial operations.
     e.  What is accomplished by this process can be remarkable. First, the supplier is alerted that these compounds
         can cause his or her customers problems, resulting in a search for an alternative raw material source that
         is free of these objectionable chemicals.  A successful market search reduces the market demand for
         contaminated or objectionable raw material.
 5.  Tests to help assess toxicity/biodegradability on speciality formulated  chemicals and mixtures and to help
     evaluate competitive products
     a.  BOD?,BOD20.
     b.  BOD5, BOD20 performed at LC50 concentration with toxicity test performed on settled effluent from test.
     c.  COD before and after BOD5, BOD20 at LC50, EC50 concentrations.
     d.  Estimate biodegradability by  using BOD5 and COD tests and the calculation (BOD5 -  COD)/COD x 100
         of 10 or 20 mg/L solutions of chemical; this can be repeated at a 20-day BOD.
     e.  Biomass inhibition tests (see detailed procedures given in Section 5).
     f.  LC50 on products; screening dilutions 1-10,000 ppm.
         OBJECTIVE:  Help industry determine relative biodegradability and toxicity of various raw materials,
         products, and by-products.
                                                  151

-------
References
Diehl, R., and S. Moore.  1987.  Case History: A
    North Carolina Municipal TRE. Toxicity Iden-
    tification/Reduction Evaluation Workshop, Water
    Pollution   Control  Federation  Conference,
    Philadelphia, Pennsylvania.

Draper and Smith. 1966. Step-wise Multiple Regres-
    sion:  Applied Regression Analysis. New York,
    New York: John Wiley and Sons.
Kirk-Othmer.   1982.   Encyclopedia of Chemical
    Technology.  New  York, New  York: Wiley
    Interscience.

Pielou.   1984.   Cluster Analysis  Techniques:  The
    Interpretation of Ecological Data. New York:
    Wiley Interscience.

Romesburg,  H.C.   1984.   Cluster Analysis for
    Researchers.  Lifetime  Learning Publications.
    Gelmont, California.
                                               152

-------
                                        Appendix J

                     Refractory Toxicity Assessment Protocol:
                                Step-by-Step Procedures
The   following  protocol  provides  step-by-step
procedures for designing and executing RTA studies to
track sources of acute and/or chronic toxicity in POTW
collection systems.  This  protocol describes the
following steps:

  •  Using characterization data to  evaluate waste
    streams of concern.
  •  Accounting  for toxicity in the activated  sludge
    biomass to be used in testing. .
  •  Adapting and calibrating the protocol to site-
    specific conditions.
  •  Collecting and analyzing samples to be used in
    testing.
  •  Preparing RTA test mixtures.
  •  Performing RTA tests.
  •  Evaluating  the  inhibitory  potential of waste
    streams.
  •  Performing TIE Phase I tests on RTA effluents
    (optional).

The RTA protocol was first developed in the USEPA
TRE research study at the City of Baltimore's Patapsco
POTW (Botts et al., 1987) to evaluate the potential for
indirect dischargers to contribute refractory toxicity.
Additional USEPA TRE research studies in Linden,
New Jersey; High Point, North Carolina; Fayetteville,
North Carolina; and Bergen County, New Jersey were
conducted to improve the RTA approach (Morris et al.,
1990; DiGiano, 1988; Fillmore et al., 1990; Collins et
al., 1991).  The RTA protocol described below is a
refined version of the method given in the first edition
of the Municipal TRE Protocol (USEPA,  1989).

The RTA procedure has been used to track sources of
acute and chronic toxicity using both freshwater and
estuarine/marine species (Morris et al., 1990; Botts et
al., 1992, 1993, 1994).  Examples of RTA studies are
presented in Appendices C, D, and G.  The RTA
protocol has been designed to simulate conventional
activated sludge processes, although it has also been
adapted to other POTW treatment processes including
single and two-stage nitrification systems (Collins, et
al. 1991), BNR processes (Appendix D), and filtration
treatment systems (Appendices C and D).
A.  POTW Wastewater Profile
Characterization data are generated for each waste
stream to be tested in the RTA.

    1.   Collect grab  samples of RAS and 24-hour
        composite samples of POTW primary effluent
        and selected sewer wastewaters (i.e., sewer
        line wastewater or indirect discharges).

    2.   Analyze RAS samples (filtrate) for TSS, VSS,
        NH3-N, and pH.

    3.   Analyze  primary  effluent  and  sewer
        wastewater samples for BOD5, COD, TSS,
        TKN, TP, NH3-N, and pH.

    4.   Determine the type of unit processes, type of
        discharge   (e.g.,   continuous   versus
        intermittent), operations schedule, and flow
        rate for the  discharge points selected  for
        evaluation (see Section 5).

    5.   Repeat above steps on several samples to
        characterize variability over time.
                                               153

-------
B. Biomass Toxicity Measurement
Biomass toxicity is measured to evaluate the potential
for toxicity interferences in the RTA.

    1.  Collect  5 liters  of fresh RAS and aerate
        vigorously for 15 minutes.

    2.  Prepare glass fiber filter [same type used for
        TSS analysis (APHA, 1995) by rinsing two 50
        ml volumes of high purity water through the
        filter.
                                             I

    3.  Rlter sufficient volume of RAS for two acute
        or chronic toxicity tests.1
                                             I
    4.  Centrifuge a portion of the RAS filtrate at
        10,000 xg for 10 to 15 minutes. Alternatively,
        filter RAS filtrate through a 0.2 um pore-size
        filter if blank tests show that the filter does not
        remove  soluble  toxicity or add artificial
        toxicity (see Section 5).1

    5.  Test RAS  filtrate  and RAS  centrate/fine
        filtrate for acute toxicity using procedures
        described by USEPA (1991a, 1991b) or for
        chronic toxicity using limited-scale methods
        provided by USEPA (1992a, 1992b,  1992c,
        1995,1996).

    6.  Repeat above steps on several RAS samples
        to characterize variability over time.

    7.  If RAS filtrate is more  toxic than the RAS
        centrate/fine filtrate, obtain non-toxic biomass
        (e.g.,  another  POTW  biomass  or  a
        freeze-dried preparation) (see Section 5).

C. RTA Reactor Calibration Testing2
Calibration tests are performed to select the RTA test
operating conditions  that most closely simulate the
POTW operation and performance.
1    Positive pressure filtering is recommended. Chronic
    toxicity  measurement will require  larger filtrate
    volumes than acute tests.
2    RTA calibration is recommended. If resources are
    limited, POTW staff may select test conditions that
    reflect POTW operating conditions. However, RTA
    reactor performance should be compared to POTW
    performance  to ensure that  the RTA procedure
    effectively simulates the POTW processes.
 1.  As described in Section 5, estimate MLVSS
    concentration  for  RTA  batch  tests  using
    mathematical models (Grady and Lim, 1980;
    Kornegay,  1970).   Alternatively,  use the
    average MLVSS concentration for the POTW.

 2.  Select  a series of MLVSS concentrations
    (e.g., four) that includes the model MLVSS
    concentration.  Calculate the volumes of RAS
    (Vr)   needed  to   yield  the  MVLSS
    concentrations in the batch  reactors.  If the
    RAS was found to be toxic (i.e., RAS filtrate
    is more toxic than RAS centrate in step B-5
    above), also select appropriate  volumes of
    non-toxic biomass (Vnb). An equation for
    calculating Vb and Vnb is:

 Vb orVnb(L)= TargetMLVSS (mg/L)
                  RASVSS(mg/L)

    where:  Vr is the reactor test volume.

 3.  Add each RAS volume  (Vb and Vnb, if
    needed) to pre-cleaned glass or clear plastic
    containers.  Add diffused air using air stones
    and gently aerate.   Note  that it  may be
    necessary to filter the air supply to prevent
    contamination  (e.g., compressor oil) of the
    reactor mixed liquors.

4.  Add primary effluent (Vpe) to each reactor
    containing  Vb and  Vnb.    Vpe  can be
    calculated using the following equation:

                Vpe = (Vr-Vb).

5.  Adjust aeration  rate to  maintain DO  at
    concentrations  typically observed in POTW
    activated sludge process. Mechanical mixing
    using a magnetic stirrer and teflon-coated stir
    bars may be required to ensure complete
    mixing. Periodically check and adjust DO
    level.

6.  Periodically  check the batch reactor pH.
    Adjust pH to 6-9 range, if necessary.

7.  Periodically collect 50-100 ml  samples of
    batch reactor mixed liquor from each reactor
    (e.g., 1- to 2-hour intervals).

8.  Allow mixed liquor samples to settle for 15
    minutes. Rinse glass fiber filters as stated in
                                                154

-------
       step B-2 above.  Filter each mixed liquor
       supernatant using separate filters.

   9.  Stop  aeration after  the required  reaction
       period, allow the Vb (and Vnb) to settle for 15
       minutes, and filter the clarified batch effluents
       as described in step C-8.

   10. Analyze filtered batch mixed liquor  and
       effluent samples to determine COD removal
       over time.

   11. Decant additional clarified batch effluent for
       toxicity analysis.  Filter each batch  effluent
       using rinsed filters.1  Wash  filter apparatus
       between each sample filtration using high-
       purity water.

   12. Batch filtrates that were treated with toxic
       biomass (Vb) must be centrifuged at 10,000
       xg  for 10 to 15 minutes to remove colloidal
       size particles.  Viscous  mixtures may require
       faster or longer centrifugation (ASM, 1981).
       Alternatively, the batch  filtrates  may be
       filtered through a 0.2 |im pore-size filter if the
       filter does not remove soluble toxicity (see
       Section 5).3 Filter blank analyses should be
       performed  for each  filter type using high-
       purity water.

    13. Analyze the batch effluent filtrates, centrates,
       and filter blanks for acute or chronic toxicity
       using the procedures referenced in step B-5.

    14. Calibration test results can be used to select a
       batch MLVSS concentration that achieves a
       level of COD and toxicity removal similar to
       that provided by the POTW activated sludge
       process (see Section 5).

D.  Sample Collection
Representative samples are collected from each waste
stream to be tested in the RTA.

    1.  Upon completion of the RTA calibration, tests
       can be conducted to evaluate the refractory
       toxicity of sewer wastewaters.
    2.  Obtain 24-hour, flow-proportioned composite
       samples of sewer wastewater (i.e., sewer line
       wastewater or indirect discharger  effluent)
       and POTW primary effluent. If possible, lag
       collection of the primary effluent sample by
       the  estimated travel  time of  the sewer
       wastewater to the POTW.

    3.  Collect 10 liters  of RAS  (and non-toxic
       biomass, if needed) on day of test.

E.  Sample Characterization (performed on
    day of sample collection)
Sample characterization data are  collected to set the
operating conditions for the RTA.

    1.  Analyze sewer wastewater for BOD5, COD,
       TSS, TKN, TP, NH3-N,  andpH.

    2.  Prepare glass fiber filter as stated in step B-2.
       Filter RAS  and test filtrate  for  acute or
       chronic  toxicity   using  the  procedures
       referenced in step B-5.4

    3.  Determine   percent   volume  of  sewer
       wastewater in POTW influent based on flow
       data gathered in the wastewater profile  (step
       A above).

F.  Preparation of RTA Test Mixtures
Two types  of batch reactors are prepared:   one
consisting of the POTW influent (primary effluent) and
RAS, which serves as a control, and another consisting
of the sewer wastewater spiked into the POTW influent
and RAS.

    1.   Calculate the volume of sewer wastewater
        (Vw) based on the sewer wastewater flow and
        the desired flow concentration factor (Fw).
        Information on selecting  an appropriate Fw is
        presented in Section 5. Vw can be calculated
        using the following equation:


               Vw(L)=-ฐ^-x(Vr-Vb)xFw,
                        Qi
    Positive pressure filtering is recommended.  Also,
    chronic  toxicity measurement will require  larger
    filtrate volumes than acute toxicity tests.
    Positive pressure filtering is recommended. Also,
    chronic toxicity measurement will require larger
    filtrate volumes than toxicity tests.
                                                155

-------
      where:  Qw is the sewer wastewater flow rate
             (mgd).
             Qi is the  average POTW influent
             flow rate (mgd).
             Fw is the sewer  wastewater  flow
             concentration factor (e.g., 1, 2,  10
             times the sewer wastewater flow).

  2.   Calculate the volume  of  primary effluent
      (Vpe) using the following equation:

                Vpe=(Vr-Vb-Vw).

  3.   Prepare  spiked  batch  reactor influent  by
      mixing Vw with Vpe and  measure Vpe for
      control batch reactor influent.

  4.   If necessary, add  nutrients  to adjust the
      BOD/TKN/TP  ratio of the  spiked batch
      influent to equal the average BOD5/TKN/TP
      ratio of the POTW influent (or 100:5:1). An
      equation for calculating BOD5, TKN, and TP
      concentrations in the spiked batch influent is:
BOD5,TKN.orTP(C,mg/L) =
    5
                             (Vpe+Vw)
     where:  Cpe is the BOD5 or nutrient concen-
             tration in primary effluent (mg/L).
             Cw is the BOD5 or nutrient concen-
             tration in sewer wastewater (mg/L).

 5.  If necessary, adjust pH of batch influents to
     pH range for POTW influent.

 6.  Test  sample   toxicity   (using   methods
     referenced in step B-5) after nutrient addition
     and pH adjustment to determine if the batch
     influent toxicity is changed by these steps.

 7.  Select volume  of RAS  (Vb) to yield  the
     MVLSS   concentration   determined  in
     calibration testing (step C above). If RAS is
     toxic (i.e., RAS filtrate is  more toxic than
     RAS centrate), also select appropriate volume
     of non-toxic biomass (Vnb). The equation for
     calculating Vb and Vnb is  provided in step
     C-2.
    8.   Add  each RAS volume (Vb and Vnb, if
        needed) to pre-cleaned glass or clear plastic
        containers.

    9.   Add spiked batch influent and control batch
        influent  to reactors  containing  Vb  (and
        reactors containing Vnb, if needed).

G. Performance of RTA Tests
The spiked batch  reactor influent and control batch
reactor influent are treated and the resulting effluents
are tested for toxicity.

    1.   Add diffused air to reactors using air stones
        and gently aerate.   Note  that  it  may be
        necessary to filter the air supply to prevent
        contamination (e.g., compressor oil) of the
        reactor mixed liquors.

    2.   Adjust  aeration rate  to maintain DO  at
        concentrations  typically observed in  the
        POTW activated sludge process. Mechanical
        mixing may be required to ensure  complete
        mixing. Periodically check and adjust the DO
        level.

    3.   Periodically check the batch reactor pH and
        adjust to pH 6-9 range, if necessary.

   4.   The treatment period for the control reactor
        should be equal to the average HRT of the
        POTW  aeration system.  For  the spiked
        reactor, calculate the required reaction period
        necessary to achieve a batch F/M ratio (F/Mb)
        equal to the nominal F/M ratio determined in
        calibration testing (step C above). F/Mb can
        be calculated using the following equation:


    TestPeri0d(dayS)^BatchInfluentCOD(mg/L),
                       (MLVSS (mg/L) x F/Mb
       where:  F/Mb is equal to the calculated F/M
               of  the  control  (primary  effluent)
               reactor.
               F/Mb  =  CODpe/(MLVSS x  test
               period, days).
                                            156

-------
    5.  Stop  aeration  after the  required  reaction
       period and allow the Vb (and Vnb) to settle
       for 1  hour.   Decant  the  clarified batch
       supernatant for toxicity analysis. Filter each
       batch supernatant using rinsed filters.5 Wash
       filter apparatus between each sample filtration
       using high-purity water.

    6.  Batch filtrates that were treated with  toxic
       biomass (Vb) must be centrifuged  at 10,000
       xg for 10 to 15 minutes to remove colloidal
       size particles (ASM, 1981). Alternatively, the
       batch filtrates may be filtered through a 0.2
       )im pore size filter  if the filter  does not
       remove  soluble  toxicity (see  Section 5).1
       Filter blank analyses should be performed for
       each filter type using high-purity water.

    7.  Analyze the batch filtrates, centrates, and
       filter blanks for acute or chronic toxicity using
       the procedures referenced in step B-5 above.

H. Synthetic Wastewater Testing (Optional)
Synthetic  wastewater can be used in lieu of POTW
influent (primary effluent) in the RTA to determine the
toxicity of the sewer wastewater.

    1.  Select   non-toxic   synthetic   wastewater.
       Confirm that  the  synthetic wastewater  is
       non-toxic  using  toxicity test procedures
       referenced in step B-5 above.

    2.  Prepare synthetic wastewater solution  with
        SCOD concentration equal  to  the  average
        SCOD of the POTW primary effluent.

    3.   Prepare  volume  of synthetic wastewater
        (Vsw) equal to the volume of primary effluent
        (Vpe)  used   above   for  the   sewer
        wastewater/primary effluent batch test.

    4.   Add Vw and Vsw to a reactor containing Vb
        (and a reactor containing Vnb, if needed).

    5.   After batch treatment, analyze batch effluent
        toxicity as described in step G above.
5   Positive pressure filtering is recommended. Also,
    chronic toxicity measurement will require larger
    filtrate volumes than acute toxicity tests.
I.  Inhibition Testing (Optional)
The RTA  protocol  can  be used to  evaluate  the
inhibitory potential of the sewer wastewater.

    1.  Add equal volumes of Vb to four reactors.
        Add diffused air and gently aerate.

    2.  Prepare a series  of four sewer wastewater
        concentrations (e.g., 100, 50, 25 and 12.5%
        wastewater)  by adding sewer wastewater to
        toxicity test dilution water (freshwater).

    3.  If  necessary, add nutrients to adjust batch
        influent BOD/TKN/TP ratio.

    4.  Add sewer wastewater volumes (e.g., Vw 100,
        Vw50, Vw25 and Vwl2.5) to the reactors.

    5.  Adjust  aeration  rate  to  maintain DO at
        concentrations  typically observed  in  the
        POTW activated sludge process. Mechanical
        mixing may be necessary to ensure complete
        mixing.  Periodically check and adjust  DO
        level.

    6.  Periodically check the batch reactor pH  and
        adjust to pH 6-9 range, if necessary.

    7.  Subsample 300 ml from each reactor at 30
        minutes and every 2  hours following  test
        initiation.   Immediately measure oxygen
        utilization  using the BOD bottle method
        (APHA, 1995). Return the subsamples to the
        reactors  immediately  following  oxygen
        utilization   measurement.     Alternatively,
        oxygen utilization  can be measured using
        respirometric techniques.

    8.  Subsample  50  ml  from each reactor at 5
        minutes and every 2  hours following  test
        initiation, and at completion of the test. Also,
        subsample 50 ml of the original undiluted
        RAS.  Filter the  subsamples through a 0.45
        jam pore-size filter and measure the SCOD of
        the filtrates.

    9.  Calculate oxygen and COD utilization rates,-
        as described in Section 5 of this manual, and
        plot  rates  versus   sewer   wastewater
        concentration.   Lower  oxygen and COD
        removal rates  with increasing wastewater
        concentration may indicate inhibition.
                                                 157

-------
 J.  Phase I Toxicity Characterization (Optional)

     1.  TIE Phase I tests may be conducted on RTA
        test  effluents  using   indirect  discharger
        wastewatef  spiked  into  primary  effluent.
        Additional  volumes are  required  for  TIE
        Phase I testing; therefore, the batch reactor
        volume will need to be increased accordingly
        (USEPA 199 la, 1992a,  1996).

     2.  TIE Phase  I tests should be performed on
        effluent filtrates  from RTA  tests that use
        non-toxic POTW biomass.

 References
 American Public Health Association (APHA). 1995.
     Standard Methods for the Examination of Water
     and  Wastewater.  American  Public  Health
     Association, Washington, D.C.

 American Society for Microbiology (ASM).  1981.
     Manual for Methods for General Bacteriology.
     Washington, D.C.

 Bolts, J.A., J.W. Braswell, E.G. Sullivan, W.C.
     Goodfellow, B.D. Sklar, and A.G. McDearmon,
     1987.   Toxicity Reduction  Evaluation  at the
    Patapsco Wastewater Treatment Plant.  Water
    Engineering Research  Laboratory, Cincinnati,
    Ohio. Cooperative Agreement No. CR812790-01-
     1. NTIS # PB 88-220 488/AS.

 Bolts, J.A., T.L. Morris, I.E.  Rumbo, and C.H.
    Victoria-Rueda.  1992.   Case Histories-Munci-
    palities. In Toxicity Reduction: Evaluation and
    Control.  D.L. Ford, ed. Technomic Publishing
    Co., Lancaster, Pennsylvania.

 Botts, J.A., T.L. Morris, M.A. Collins, T.M. Schmitt,
    and E.K. Wilson. 1993. Evaluating the Impact of
    Industrial   Discharges   to   Publicly  Owned
    Treatment   Works: The  Refractory  Toxicity
    Assessment Protocol. Annual Conference and
    Exposition of the Water Environment Federation,
    Anaheim, California.

Botts, J.A., T. Schmitt, E. Wilson, M. Collins, D.
    Waddell,  R.  Diehl,  and L.  Ehrlich.    1994.
    Refractory Toxicity Assessment: An Alternative
    Approach  for  Chronic  Toxicity Reduction
    Evaluations. Annual Conference and Exposition of
     the  Water Environment Federation, Chicago,
     Illinois. Paper # AC944404.

 Collins, M.A., T.L. Morris, J.A.  Botts, T. Norberg-
     King,  J.  Thompson, and D.I. Mount.   1991.
     Chronic Toxicity Reduction  Evaluation  at  the
     Bergen County Utilities  Authority Wastewater
     Treatment Plant,  Draft Report.  USEPA, Risk
     Reduction Engineering Laboratory,  Cincinnati,
     Ohio.  USEPA Contract No. 68-03-3431.

 DiGiano.F. 1988. Case History: Toxicity Reduction
     Evaluation  at  High  Point,  North  Carolina.
     Virginia Water Pollution Control Association
     Wastewater  Toxics  Management  Seminar,
     November 3, 1988, Richmond, Virginia.

 Fillmore, L.B., T.L. Morris,  T.L. Champlin, M.C.
     Welch, and J.A. Botts. 1990.  Toxicity Reduction
     Evaluation at the City ofFayetteville Cross Creek
     Wastewater Treatment Plant, Draft Report, 1990.
     USEPA, Risk Reduction Engineering Laboratory,
     Cincinnati, Ohio.  USEPA Contract No. 68-03-
     3431.

 Grady,  C.P.L., and H.P. Lira.   1980.  Biological
     Wastewater Treatment: Theory and Applications.
     Marcel Dekker, Inc., New York, New York.

 Kornegay, B.H.  1970.  The Use of Continuous Cul-
     ture Theory in the Selection of Biological Reactor
     Systems. 43rd Annual Conference of the  Water
     Pollution   Control  Federation,  Boston,
     Massachusetts, October.

 Morris, T.L., J.A. Botts, J.W. Braswell, M.C. Welch,
     and W.L. Goodfellow. 1990. Toxicity Reduction
    Evaluation at the Linden  Rosette  Sewerage
    Authority  Wastewater Treatment  Plant,  Draft
    Report, 1990. USEPA,  Office of Research and
    Development,  Risk  Reduction   Engineering
    Laboratory, Cincinnati, Ohio. USEPA Contract
    No. 68-03-3431.

USEPA.   1989.   Toxicity  Reduction Evaluation
    Protocol for Municipal Wastewater  Treatment
    Plants.  EPA/600/2-88/062. Office of Research
    and Development, Risk Reduction Engineering
    Laboratory, Cincinnati, Ohio.

USEPA.  1991a. Methods for Aquatic Toxicity Identi-
   fication Evaluations: Phase I Toxicity Character-
    ization Procedures. EPA/600/6-91-003. Second
                                               158

-------
    Edition.  National Effluent Toxicity Assessment
    Center, Duluth, Minnesota.

USEPA.  199 Ib.  Technical Support Document for
    Water Quality-Based  Control.   EPA/505-2-90-
    001.  Office of Water Enforcement and Permits,
    Washington, D.C.

USEPA.  1992a. Toxicity Identification Evaluations:
    Characterization of Chronically Toxic Effluents,
    Phase I. EPA/600/6-91-005F.  National Effluent
    Toxicity Assessment Center, Duluth, Minnesota.

USEPA.  1992b. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to Marine and Estuarine  Organisms.
    EPA/600/4-91/003. Second Edition. Environ-
    mental Monitoring  and  Support  Laboratory,
    Cincinnati, Ohio.
USEPA.  1992c. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters   to   Fresh-water  Organisms.
    EPA/600/4-91/002. ThirdEdition. Environmental
    Monitoring and Support Laboratory, Cincinnati,
    Ohio.

USEPA.  1993. Methods for Measuring the Acute
    Toxicity of Effluents and Receiving Waters to
    Freshwater   and  Marine  Organisms.
    EPA/600/4-90-027F. Fourth Edition. Environmen
    tal Monitoring Systems Laboratory, Cincinnati,
    Ohio.

USEPA.  1995. Short-Term Methods for Estimating
    the Chronic Toxicity of Effluents and Receiving
    Waters to West Coast Marine and Estuarine
    Organisms. EPA/600/R-95/136. National Expo-
    sure Research Laboratory, Cincinnati, Ohio.

USEPA.  1996. Marine Toxicity Identification Evalu-
    ation (TIE)   Guidance  Document, Phase  I.
    EPA/600/R-96/054.   Office  of Research and
    Development,  Washington, D.C.
                                               159

-------

-------