United States
Environmental Protection
Agency
Office of Water
4301
EPA-820-B-95-001
March 1995
Water Quality
Guidance for the
Great Lakes
System:
Supplementary
Information
Document  (SID)

-------
                               DISCLAIMER

    This document has been reviewed  by the Office of Science and
    Technology,  U.S. Environmental Protection Agency, and approved
    for publication as a support document for the final Water
    Quality Guidance for the Great Lakes System.  Mention of trade
    names and commercial products does not constitute endorsement
    of their use.
                          AVAILABILITY NOTICE

    This document i's available for a  fee  upon  written request or
    telephone call to:
fA
             National Technical  Information Center  (NTIS)
                      U.S. Department of  Commerce
                          5285  Port  Royal  Road
                        Springfield,  VA  22161
                             (800) 553-6847
                             (703) 487-4650
                   NTIS Document Number:   PB95187266

                                  or

       Education Resources  Information Center/Clearinghouse  for
    Science, Mathematics, and  Environmental Education  (ERIC/CSMEE)
                     1200 Chambers  Road,  Room 310
                          Columbus,  OH  43212
                             (800) 276-0462
                             (614) 292-6717
                          ERIC Number:  D046
                                           U.S. Environmental Protectfon Aginey
                                           Region 5, Library (PL-12J)
                                           77 West Jackson Boulevard, 12th Fl§§f
                                           Chicago, IL 60604-3590

-------
                             FOREWORD
     On March 23', 1995, EPA published the final Water Quality
Guidance for the Great Lakes System in the Federal Register.  In
addition to the Guidance itself, the Federal Register publication
included: a brief summary of the process for developing the final
Guidance; a summary of the final Guidance; a summary of the
consultation with the U.S. Fish and Wildlife Service under the
Endangered Species Act; a summary of EPA's analyses pursuant to
the Regulatory Flexibility Act and the Paperwork Reduction Act;
and a summary of EPA's response to Executive Order 12866.

     The purpose of this Supplementary Information Document (SID)
is to provide a more detailed description of the final Guidance,
including a statement of the basis and purpose of the final
Guidance; a discussion of the major changes from the proposal;
and an analysis of issues raised in comments received on the
proposed Guidance and subsequent documents in the Federal
Register.

     The SID is divided into 12 major sections and further
divided within the sections into numerous subsections (see list
of major sections below and the table of contents).   Each
subsection presents a summary of the proposal for that issue,
significant comments received  (with EPA's response to each), and
the requirements for the final Guidance for that issue.
Responses to all comments received are provided in the Response
to Comment document with is available in the docket for the final
Guidance.
     Section I
     Section II
     Section III
     Section IV
     Section V
     Section VI
     Section VII
     Section VII-I
     Section IX
     Section X
     Section XI

     Section XII
Background
Regulatory Requirements
Aquatic Life
Bioaccumulation Factors
Human Health
Wildlife
Antidegradation
Implementation Procedures
Executive Order 12866
Regulatory Flexibility Act
Enhancing the Intergovernmental Partnership
under Executive Order 12875
Paperwork Reduction Act

-------
                        TABLE OF CONTENTS

I.  BACKGROUND	   1

A.   Description of Resource  	   1

B.   Environmental Problems in the Great Lakes System  ....   2
     1.   Historical Nature of Environmental Problems  in the
          Great Lakes System	   2
     2.   Current Trends of Environmental Problems in  the
          Great Lakes System	   3
          a.   Impacts From Chemical Contamination   	   3
          b.   Trends in Contaminant Levels 	   7
          c.   Fish Consumption Advisories  	   8

C.   History of the Great Lakes Water Quality Guidance  ...  11
     1.   The Great Lakes Toxic Substances Control
          Agreement	11
     2.   The Great Lakes Water Quality Initiative   	  11
     3.   The Great Lakes Critical Programs Act of 1990 ...  12
     4.   Principles Underlying the Final Water Quality
          Guidance for the Great Lakes System	12
          a.   Use the Best Available Science to Provide
               Protection to Human Health, Wildlife, and
               Aquatic Life	12
          b.   Recognize the Unique Nature of the Great
               Lakes Basin Ecosystem	13
          c.   Promote Consistency in Standards and
               Implementation Procedures While Allowing
               Appropriate Flexibility to States and Tribes  .  13
          d.   Establish Equitable Strategies to Control
               Pollution,Sources  	  16
          e.   Promote Pollution Prevention Practices  ....  17
          f.   Provide Accurate Assessment of Costs and
               Benefits	17

D.   Progress on Other Programs to Protect and Restore the
     Great Lakes System	17
     1.   The Great Lakes Toxic Reduction Effort  	  18
          a.   Pathway Track	18
          b.   Virtual Elimination Project	18
          c.   Lake Michigan Enhanced Monitoring Program. .  .  19
     2.   Clean Air Act Amendments	19
     3.   ARARs and the Superfund Program	  .  21
     4.   RAPs and LaMPs	21
          a.   RAPs	21
          b.   LaMPs	23
     5.   Sediments	24

E.   Science Advisory Board Review  	  26

F.   References	26

II.  REGULATORY REQUIREMENTS  	  29

                                ii

-------
A.   Scope and Purpose	29

B.   Definitions	29
     1.   Proposal	29
     2.   Comments	29
     3.   Final Guidance	30

C.   Adoption and Application of Criteria, Methodologies,
     Policies, and Procedures 	  32
     1.   Adoption of Tier I Criteria and Methodologies  ...  32
          a.   Proposal	  .  32
          b.   Comments	32
          c.   Final Guidance	37
     2.   Adoption and Application of Tier II Methodologies  .  37
          a.   Proposal	  37
          b.   Comments	37
          c.   Final Guidance	40
     3.   Application of Anti-backsliding Provisions of  the
          Clean Water Act	41
          a.   Proposal	41
          b.   Comments	41
          c.   Final Guidance	42
     4.   Basin-wide Application of Criteria and Values  ...  46
          a.   Proposal	46
          b.   Comments	46
          c.   Final Guidance	48
     5.   Pollutants Subject to Federal, State, and Tribal
          Requirements	48
          a.   Proposal	48
          b.   Comments	48
          c.   Final Guidance	56
     6.   Scientific Defensibility Exclusion   	  58
          a.   Proposal	58
          b.   Comments	59
          c.   Final Guidance	59
     7.   Wet Weather Exclusion	59
          a.   Proposal	59
          b.   Comments	60
          c.   Final Guidance	60
     8.   Bioaccumulative Chemicals of Concern  	  61
          a.   Proposal	61
          b.   Comments	62
          c.   Final Guidance	67
     9.   Potential Bioaccumulative Chemicals of Concern  .  .  68
          a.   Proposal	68
          b.   Comments	68
          c.   Final Guidance	69
     10.  Pollutants of Initial Focus 	  69
          a.   Proposal	69
          b.   Comments	70
          c.   Final Guidance	71

D.   Procedures for Adoption and EPA Review	71
     1.   Adoption Procedures 	  71

                               iii

-------
          a.   Proposal	71
          b.   Comments ...."'	71
          c.   Final Guidance	.  .  73
     2.   Interpretation of "Consistent With" 	  74
          a.   Proposal	74
          b.   Comments	,	>.  74
          c.   Final Guidance	77
     3.   Indian Tribes	•	79
          a.   Proposal	79
          b.   Comments	79
          c.   Final Guidance	  80

E.   Amendments to NPDES and Water Quality Standards Program
     Regulations	80
     1.   Proposal	80
     2.   Final Guidance	80

F.   Precedential Effects of Elements of the Guidance ....  81
     1.   Proposal	81
     2.   Comments	81
     3.   Final Guidance	82

G.   Implementation of Endangered Species Act	82
     1.   Proposal	82
     2.   Consultation with the U.S. Fish and Wildlife
          Service	83
          a.   Protection of Endangered Mussel Species   ...  83
          b.   Wildlife Criteria Methodology	85
          c.   Other Issues	86
     3.   Comments	88
     4.   Final Guidance	88

III.  AQUATIC LIFE	93

A.   Summary of Final Rule	93

B.   Final Tier I Methodology	93
     1.   Required Data	94
          a.   Proposal	94
          b.   Comments	94
          c.   Final Guidance	94
     2.   Commercially and Recreationally Important Species  .  95
          a.   Proposal	95
          b.   Comments	95
          c.   Final Guidance	  .  95
     3.   Ecologically Important Species  	  95
          a.   Proposal	95
          b.   Comments	95
          c.   Final Guidance	96
     4.   Elimination of Final Residue Value	  .  96
          a.   Proposal	96
          b.   Comments	96
          c.   Final Guidance	96
     5.   Acute-Chronic Ratios	96

                                iv

-------
          a.   Proposal	96
          b.   Comments	.  .  .  .	97
          c.   Final Guidance	97
     6.   Bioavailability	97
          a.   Proposal	97
          b.   Comments	98
          el1   Final Guidance	  102
     7.   Averaging Period/Frequency of Exceedance   	  102
          a.   Proposal	102
          b.   Comments	102
          c.   Final Guidance	103
     8.   Final Tier I Criteria	103
          a.   Proposal	103
          b.   Comments	••	104
          c.   Final Guidance	104
     9.   Tier I Criteria	109
          a.   Proposal	  .  109
          b.   Comments:   	109
          c.   Final Guidance	109
     10.  Potential Changes to National Guidelines   	  110
          a.   Proposal	110
          b.   .Comments	Ill
          c.   Final Guidance	Ill

C.   Final Tier II Methodology	Ill
     1.   Requirement for use in Interpreting the Narrative
          Toxics Criterion  	  Ill
          a.   Proposal	ill
          b.   Comments	112
          c.   Final Guidance	112
     2.   Data Requirements	113
          a.   Proposal	113
          b.   Comments	113
          c.   Final Guidance	113
     3.   Other Methods  for Tier II Values	113
          a.   Proposal	113
          b.   Comments	113
          c.   Final Guidance	114
     4.   Adjustment Factors	114
          a.   Proposal	114
          b.   Comments	114
          c.   Final Guidance	114
     5.   Assumed ACRs	117
          a.   Proposal	117
          b.   Comments	117
          c.   Final Guidance	117

D.   Comparison with the CWA and EPA's National Guidance   .  .  117

E.   Conformance* with the Great Lakes Water Quality
     Agreement	118
     1.   Tier I Aquatic Life Criteria and Methodology   .  .  .  118
     2.   Tier II Values and Methodology	119

-------
IV.  BIOACCUMULATION FACTORS	121

A.   Summary of Final Rule	-121

B.   Explanation of Final Provisions	  . 121
     1.   BAFs  •	121
     2.   Measured and Predicted BAFs	'.  . 122
          a.   Hierarchy of Methods	122
               i.   Proposal:	122
               ii.  Comments	123
               iii. Final Guidance	127
          b.   Field-Measured BAFs	127
               i.   Proposal	127
               ii.  Comments	127
               iii. Final Guidance	128
          c.   Field-Measured BSAFs	128
               i.   Proposal	128
               ii.  Comments	128
               iii. Final Guidance	129
          d.   Measured and Predicted BCFs	129
               i.   Proposal	129
               ii.  Comments	130
               iii. Final Guidance	131
          e.   Inorganic Chemicals	131
               i.   Proposal	131
               ii.  Comments	131
               iii. Final Guidance	131
     3.   Lipid Values	132
          a.   Lipid Value for Human Health BAFs	132
               i.   Proposal	132
               ii.  Comments	132
               i4.i. Final Guidance	133
          b.   Lipid Value for Wildlife BAFs	133
               i.   Proposal	133
               ii.  Comments	134
               iii. Final Guidance	134
     4.   FCMs	134
          a.   Proposal	134
          b.   Comments	134
          c.   Final Guidance	136
     5.   Accounting for the Effect of Metabolism  in
          Predicted BAFs	136
          a.   Proposal	136
          b.   Comments	137
          c.   Final Guidance	137
     6.   Bioavailability	137
          a.   Proposal	137
          b.   Comments	138
          c.   Final Guidance	139
     7.   Calculation of Baseline BAFs	139
          a.   Proposal	139
          b.   Final Guidance	139
     8.   Other Uses of BAFs	141
     9.   Individual BAFs	141

                                vi

-------
          a.   2,3,7,8-TCDD	141
               i.   Proposal   . .'.	141
               ii.  Comments	141
               iii. Final Guidance	141
          b.   PCBs	142
               i".   Proposal	•  142
               ii.  Comments	142
               iii. Final Guidance	•	143
          c.   Mercury	143
               i.   Proposal	143
               ii.  Comments	143
               iii. Final Guidance	144

C.   Conformance to the  CWA, Great Lakes Water Quality
     Agreement and Great Lakes Critical Programs Act  of
     1990	144

D.   Adoption of Water Quality Standards Consistent with  the
     Final Guidance	144

E.   References	145

V.  HUMAN HEALTH	147

A.   Summary of Final Rule	147

B.   Explanation of Final Provisions   	  147

C.   Criteria Methodologies  	  148
     1.   Endpoints Addressed	148
          a.   Proposal	148
          b.   Comments	148
          c.   Final Guidance	148
     2.   Mechanism of Action	148
          a.   Cancer	148
               i.   Proposal	148
               ii.  Comments	149
               iii. Final Guidance	149
          b.   Noncancer	150
               i.   Proposal	150
               id.  Comments	150
               iii. Final Guidance	150
     3.   Choice of Risk Level	151
          a.   Proposal	151
          b.   Comments	151
          c.   Final Guidance	151
     4.   Acceptable Dose	153
          a.   RAD	153
               i.   Biologically Relevant versus Sensitive
                    Species	153
                    (A) . Proposal	153
                    (B) . Comments  .  .•	153
                    (C) . Final Guidance	153
               ii.  Less than Lifetime Adjustment Factor.  .  .  154

                               vii

-------
               (A) .  Proposal	154
               (B) .  Comments	154
               (C).  Final Guidance 	  .....  154
          iii. Species Scaling Factor   	  154
               (A) .  Proposal	154
               (B).  Comments 	  .....  154
               (C) .  Final Guidance	  155
     b.   ADE	155
          i.   List of Deleterious Effects	155
               (A) .  Proposal	155
               (B) .  Comments	155
               (C) .  Final Guidance	155
          ii.  Uncertainty Factors	156
               (A) .  Proposal	156
               (B) .  Comments	  156
               (C) .  Final Guidance	156
     c.   IRIS	157
          i.   Proposal	157
          ii.  Comments	157
          iii. Final Guidance	157
5.   Exposure Assumptions	  .  158
     a.   Body Weight	158
          i.   Proposal	158
          ii.  Comments	158
          iii. Final Guidance	158
     b.   Duration of Exposure	158
          i.   Proposal	158
          ii.  Comments	159
          iii. Final Guidance	159
     c.   Incidental Exposure  	  159
          i.   Proposal	  .  159
          ii.  Comments	  159
          ij.i. Final Guidance	160
     d.   Drinking Water Consumption	160
          i.   Proposal	160
          ii.  Comments	161
          iii. Final Guidance	161
     e.   Fish Consumption	161
          i.   Proposal	161
          ii.  Comments	162
          iii. Final Guidance:	165
     f.   BAFs	165
     g.   Relative Source Contribution	166
          i.   Proposal	 .  .  166
          ii.  Comments	166
          iii. Final Guidance	167
6.   Minimum Data Requirements/Tier I and Tier II   ...  167
     a.   Carcinogens	168
          i.   Proposal	168
          ii.  Comments	168
          i-ii. Final Guidance	169
     b.   Non-carcinogens	170
          i.   Proposal	170
          ii.  Comments	170

                          viii

-------
               iii. Final Guidance	171

D.   Criteria Derivation	171
     1.   Proposed Criteria and Values   	  172
          a.   PCBs  (Human Cancer Value)	173
               i.   Proposal	173
               ii.  Comments	173
               iii. Final Guidance:	174
          b.   TCDD	175
               i.   TCDD - Noncancer Criterion	175
                     (A) . Proposal	175
                     (B) . Comments	175
                     (C) . Final Guidance	175
               ii.  TCDD - Cancer Criterion  .........  175
                     (A)  Proposal	175
                     (B) . Comments	175
                     (C) . Final Guidance	176
          c.   Mercury	176
               i.   Proposal	176
               ii.  Comments	176
               iii. Final Guidance	176

E.   Relationship of the Great Lakes Initiative Guidelines
     to National Guidelines Revisions  	  176
     l.   Proposal	176
     2.   Comments	177
     3.   Final Guidance	177

F.   Comparison with the CWA and Great Lakes Water Quality
     Agreement	177
     1.   Tier I Criteria/Methodology	177
          a.   Comparison with the CWA	177
          b.   Conformance with the GLWQA	178
     2.   Tier IJ Criteria/Methodology	178
          a.   Comparison with the CWA	178
          b.   Conformance with the GLWQA	178

G.   References	178

VI.  WILDLIFE	181

A.   Introduction	181

B.   Scope of Methodology	182
     1.   Proposal	182
     2.   Discussion of Comments	182
     3.   Final Guidance	185

C.   Effect Component  	  185
     1.   Minimum Data Requirements	185
          a.   Proposal	185
          b.   Comments	186
          c.   Final Guidance	187
     2.   UFs	187

                                ix

-------
          a.   Proposal	187
          b.   Comments	187
          c.   Final Guidance	189

D.   Exposure Component 	 . 	 ..... 190
     1.   Representative Species   	 190
          a.   Proposal .	 190
          b.   Comments	191
          c.   Final Guidance	191
     2.   Exposure Parameters	192
          a.   Proposal	192
          b.   Comments	192
          c.   Final Guidance	194

E.   Protection of Individual Members of a Population .... 194
     1.   Proposal	194
     2.   Comments	194

F.   Wildlife Criteria	195
     1.   Proposal	195
     2.   Discussion	195
          a.   Mercury	195
          b.   DDT	196
          c.   PCBs	197
          d.   2,3,7,8 TCDD	197
     3.   Final Guidance	198

G.   Comparison of Wildlife Criteria and Methods to National
     Program and to Great Lakes Water Quality Agreement  .  .  .198

H.   References	198

VII.  ANTIDEGRADATION	203

A.   General Discussion/Background  	 203
     1.   History of the Great Lakes Antidegradation
          Guidance	203
     2.   Summary of the Proposed Guidance	204
          a.   The Antidegradation Standard 	 204
          b.   Antidegradation Implementation Procedures   .  .205
          c.   Antidegradation Demonstration  	 205
          d.   Antidegradation Decision 	 206

B.   Overview of. the Final Guidance	206

C.   Detailed Discussion of the Final Regulation   	 207
     1.   The Antidegradation Standard  	 207
     2.   Antidegradation Implementation Procedures 	 207
          a.   De Minimis Lowering of Water Quality	207
               i.   Background	207
               ii.  Discussion of  Significant Comments   .  .  . 208
               iii. The Final Guidance	209
          b.   High Quality Waters	210
               i.   Background	210

-------
               ii.  Discussion of Significant Comments   .  .  . 210
               iii. The Final Guidance	212
          c.   Lake Superior Basin - Outstanding
               International Resource Waters  	 212
          d.   Outstanding National Resource Water  	 212
               i.   Background	: 212
               ii.  Discussion of Significant Comments   .  .  . 212
               iii. The Final Guidance  . . .	213
          e.   Significant Lowering of Water Quality  .... 213
               i.   Background	213
               ii.  Discussion of Significant Comments   .  .  . 213
               iii. The Final Guidance	214
          f.   Deleted Definitions  	 214
          g.   Implementation Procedures  	 215
              . i.   Background	215
               ii.  Discussion of Significant Comments   .  .  . 216
               iii. The Final Guidance	218
     3.   Antidegradation Demonstration 	 219
          a.   Background	219
          b.   Discussion of Significant Comments 	 219
          c.   The Final Guidance	221
               i.   Identification of Cost-Effective
                    Pollution Prevention Alternatives to
                    Prevent or Reduce the Significant
                    Lowering of Water Quality 	 221
               ii.  Alternative or Enhanced Treatment to
                    Eliminate the Significant Lowering of
                    Water Quality	222
               iii. Important Social and Economic
                    Development	223
     4.   Antidegradation Decision  	 224
          a.   Background	224
          b.   Discussion of Significant Comments 	 224
          c.   The Final Guidance	225

VIII.  IMPLEMENTATION PROCEDURES	227

A.   Site-Specific Modifications to Criteria and Values  .  .  .227
     1.   General	227
     2.   Aquatic Life	227
          a.   Proposal	227
          b.   Comments	227
          c.   Final Guidance	228
     3.   Wildlife	  . 229
          a.   Site-specific Modifications  	 229
               i.   Proposal	229
               ii.  Comments	229
                    (A). Less Stringent Site-specific
                         Modifications  	 229
                    (B). More Stringent Site-specific
                         Modifications  	 230
               iii. Final Guidance	230
          b.   Protection of Endangered or Threatened
               Species	230

                               xi

-------
               i.   Proposal	230
               ii.  Comments  .	230
               iii. Final Guidance	 231
     4.   Bioaccumulation Factors 	 231
          a.   Proposal .	231
          b.   Comments	232
          c:   Final Guidance	233
     5.   Human Health	233
          a.   Proposal	233
          b.   Comments	233
               i.   Fish Consumption Rate	234
               i-i.  Percent Fish Lipid	234
               iii. BAF	235
          c.   Final Guidance	•	235

B.   Variances from Water Quality.Standards for Point
     Sources	 237
     1.   Applicability	237
          a.   Proposal	237
          b.   Discussion of Significant Comments  	 238
          c.   Final Guidance	239
     2.   Maximum Timeframe for Variances	239
          a.   Proposal	239
          b.   Discussion of Significant Comments  	 239
          c.   Final Guidance	239
     3.   Conditions to Grant a Variance	240
          a.   Proposal	240
          b.   Discussion of Significant Comments  	 240
          c.   Final Guidance	241
     4.   Submittal of Variance Application 	 242
          a.   Proposal	242
          b.   Discussion of Significant Comments  	 242
          c.   Final Guidance	242
     5.   Public Notice	242
          a.   Proposal	242
          b.   Discussion of Significant Comments  	 242
          c.   Final Guidance	243
     6.   Final Decision on Variance Request  	 243
          a.   Proposal	243
          b.   Discussion of Significant Comments  	 243
          c.   Final Guidance	244
     7.   Incorporating Variances into NPDES Permits   .... 244
          a.   Proposal	244
          b.   Discussion of Significant Comments  	 244
          c.   Final Guidance	244
     8.   Renewal of Variances	245
          a.   Proposal	245
          b.   Discussion of Significant Comments  	 245
          c.   Final Guidance	245
     9.   EPA Approval	245
          a.   Proposal	 245
          b.   Discussion of Significant Comments  	 246
          c.   Final Guidance	246
     10.  State or Tribal Water Quality Standards  Revisions  . 246

                               xii

-------
          a.   Proposal	246
          b.   Discussion of Significant Comments 	 246
          c.   Final Guidance	-246

C.   Total Maximum Daily Loads	247
     1.   Background	.  . 247
     2.   Overview of Proposed Procedures 3A and 3B . .  .  .''  . 249
          a..   Proposal	249
          b.   Comments	249
          c.   Final Guidance	250
     3.   General Conditions of Application 	 251
          a.   General Condition 1 - TMDLs Required 	 252
               i..   Proposal	252
               ii.  Comments	252
               iii. Final Guidance	252
          b.   General Condition 2 - Attainment of Water
               Quality Standards  	 253
               i.   Proposal	253
               ii.  Final Guidance	253
          c.   General Condition 3 - TMDL Allocations .... 254
               i.   Proposal	254
               id.  Comments	254
               iii. Final Guidance	254
          d.   General Condition 4 - WLA Values	255
               i.   Proposal	255
               ii.  Final Guidance	256
          e.   General Condition 5 - Margin of Safety .... 256
               i.   Proposal	256
               ii.  Comments	256
               iii. Final Guidance	256
          f.   General Condition 6 - More Stringent
               Requirements	257
          g.   General Condition 7 - Accumulation in
               Sediments	257
               i.   Proposal	257
               ii.  Comments	257
               iii. Final Guidance	258
          h.   General Condition 8 - Wet Weather Events  .  .  . 259
               i.   Proposal	259
               il.  Comments	259
               iii. Final Guidance	259
          i.   General Condition 9 - Background
               Concentrations of Pollutants 	 260
               i.   Choice of Data Set	260
                    (A)   Proposal	260
                    (B)   Comments	260
                    (C)   Final Guidance	261
               ii.  Geometric Mean	262
                    (A) . Proposal	262
                    (B) . Final Guidance	262
               iii. Data Points Above and Below Detection  .  . 263
                    (A)   Proposal	263
                    (B)   Comments	263
                    (C)   Final Guidance	264

                              xiii

-------
     j .    General Condition 10 - Effluent Flow	265
     k.    General Condition 11 - Reserved Allocations   . 266
          i.   Proposal	266
          ii.  Comments	266
          iii. Final Guidance	266
4.   Special Provisions for BCCs	266
     a.    Proposal	266
     b.    Comments	266
     c.    Final Guidance	268
5.   TMDLs for Open Waters of the Great Lakes	273
     a.    Mixing Zones for non-BCCs	'  . 273
          i.   Proposal	273
          ii.  Comments	274
          iii. Final Guidance	274
     b.    Calculating Load Allocations	  . 275
     c.    Protection from Acute Effects  	 275
          i.   Proposal	275
          ii.  Comments	275
          iii. Final Guidance	276
6.   TMDLs for Discharges to Tributaries 	 276
     a.    Steady State	277
          i\   Proposal	277
          ii.  Comments	277
          iii. Final Guidance	277
     b.    Stream Design Flows	277
          i.   Proposal	277
          ii.  Comments	278
          iii. Final Guidance	278
          iv.  Wildlife	279
               (A)  Proposal	279
               (B)  Comments	279
               (C)  Final Guidance	279
          v.   Chronic Aquatic Life	281
               (A)  Proposal	281
               (B)  Comments	281
               (C)  Final Guidance	281
          vi.  Acute Aquatic Life	281
               (A)  Proposal	281
               (B)  Comments	281
               (C)  Final Guidance	281
          vii. Human Health	282
               (A)  Proposal	282
               (B)  Comments	282
               (C)  Final Guidance	282
     c.    Mixing Zones for Non-BCCs	283
          i.   Proposal	283
          ii.  Comments	283
          iii. Final Guidance	284
7.   Procedures for High Background Concentrations  .  .  . 285
     a.    Proposal	285
     b.    Comments	285
     c.    Final Guidance .  . .  :	285
8.   Pollutant Degradation  	 286
     a.    Proposal	286

                          xiv

-------
          b.   Comments	286
          c.   Final Guidance .  .	286
     9.   Mixing Zone Studies	287
          a.   Proposal	287
          b.   Comments	287
          c.   Final Guidance	287
     10.  Pollution Trading Opportunities 	 288

D.   Additivity	289
     1.   Background	289
     2.   Additivity Considerations in Other EPA Programs .  . 290
     3.   Existing State Water Quality Standards  	 290
     4.   Proposed Guidance Overview  	 291
     5.   Aquatic Life	•	291
     6.   Human Health	291
          a.   Carcinogens	291
               i-.    Proposal	291
               ii.  Discussion of Significant Comments on
                    the Assumption of Additivity for
                    Carcinogens	292
               iii. Discussion of Significant Comments on
                    Application of Additivity Provision for
                    Carcinogens	293
               iv.  Final Guidance	295
          b.   Non-carcinogens	296
               i'.    Proposal	296
               ii.  Discussion of Significant Comments  . .  . 297
               iii. Final Guidance	298
     7.   Toxicity Equivalency Factors/Bioaccumulation
          Equivalency Factors 	 298
          a.   Proposal	298
          b.   Discussion of Significant Comments 	 299
          c.   Response for Application of TEFs/BEFs to
               Human Health	299
          d.   Response to Application of TEFs to Wildlife   . 300
          e.   Final Guidance:  	301
     8.   Wildlife	301
          a.   Proposal	301
          b.   Comments	302
          c.   Final Guidance:  	302
     9.   References	304

E.   Reasonable Potential for Exceeding Numeric Water
     Quality Standards	 .  . 307
     1.   Existing National Rules and Guidance  	 307
     2.   General Requirements of Procedure 5 	 309
          a.   Developing Preliminary Wasteload Allocations  . 311
               i.    Proposal	311
               ii.  Comments	311
               iii. Final Guidance	311
          b.   Developing Preliminary Effluent Limitations   . 312
               i,.    Proposal  .  .  .	312
               ii.  Comments	313
               iii. Final Guidance	315

                               xv

-------
     c.    Determining Reasonable Potential to Exceed
          the Preliminary Effluent Limitations Using
          Pollutant Concentration Data 	 315
          i.    Proposal	.316
          ii.  Comments	316
          iii.  Final Guidance:	318
     d.    Determining Reasonable Potential Using
          Pollutant Concentration Data Where the
          Effluent Flow Rate is Equal to or Greater
          than the 7Q10	324
          i.    Proposal:	324
          ii.  Comments	325
          iii.  Final Guidance	325
     e.    D.etermining Reasonable Potential in the
          Absence of Facility Specific Effluent
          Monitoring Data	.326
          i.    Proposal	326
          ii.  Comments	326
          iii.  Final Guidance	327
     f.    Determining Reasonable Potential for
          Pollutants When Tier II Values are Not
          Available	327
          i..    Proposal	327
          ii.  Comments	329
          iii.  Final Guidance	331
     g.    Determining Reasonable Potential Using Fish
          Tissue Data	333
          i.    Proposal	333
          ii.  Comments	333
          iii.  Final Guidance	334
     h.    Basis for Effluent Limitations	336
3.   Consideration of Pollutants in Intake Water .... 339
     a.    Introduction	339
     b.    Existing Mechanisms	340
     c.    Summary of Proposal	341
          i.    Proposed Guidance 	 341
          ii.  Other Options	 343
               (A)  Option 1	343
               (B)  Option 2	343
               (C)  Option 3	344
               (D)  Option 4	344
4.   Summary of Intake Pollutant Considerations in
     Final Guidance and Overall Rationale	345
     a.    Intake Pollutant Reasonable Potential
          Procedure	345
     b.    No Net Addition Interim Approach for Setting
          WQBELS	346
     c.    Consideration of Intake Pollutants from a
          Different Body of Water	349
     d.    Combined Approach for Multiple Intake
          Sources   	  ..... 352
5.   Legal Authority	353
                          xvi

-------
          a.   Discharge of Intake Pollutants is an Addition
               of Pollutants Under-the CWA	353
          b.   EPA's Authority and Rationale for
               Establishing Interim Permitting Procedures
               Allowing "No Net Addition" Limitations for
               Intake Water Pollutants  	 359
     6.   Continued Availability of Existing Mechanisms . .   . 362
     7.   Final Intake Credit Provision and Response to
          Major Comments	365
          a.   General Issues	365
               i.   Pollutant-by-Pollutant, Outfall-by-
                    Outfall Analysis	365
               ii.  Pollutants Covered  	 365
               iii. Required Demonstration  	 367
               iv.  Definition of Same Body of Water  .... 368
                    (A). General	369
                    (B). Intermediate Use/Public Water
                         Supply	372
                    (C) . Ground water	374
               v.   100 Percent Requirement	375
               vi.  No Increase in Concentration
                    Requirement	376
               vii. Chemical/Physical Alteration
                    Requirement	379
               viii. Timing/Location Requirement  	 380
               ix.  Relationship Between this Section and
                    Other Reasonable Potential Sections . .   .381
          b.   Issues Specific to the Intake Pollutant
               Reasonable Potential Procedure 	 382
               i..   No Mass Added Requirement	382
               ii.  Other Requirements	386
                    (A) . Documentation	386
                    (B) . Monitoring	387
                    (C) . Reopener	388
          c.   Issues Specific to Consideration of Intake
               Pollutants in the Derivation of WQBELs .... 388
               i.   Availability Only for Non-Attainment
                    Waters	388
               ii.  Partial Credit at Discretion of
                    Permitting Authority  	 390
               iii. Developing the Limit	392
                    (B). Expression of a Numeric Limit  . .   . 392

F.   Whole Effluent Toxicity  	 395
     1.   Background	395
     2.   Criteria for WET	396
          a.   Proposal	396
          b.   Comments	396
          c.   Final Guidance	397
     3.   Acute Toxicity Control	397
          a.   Proposal	397
          b.   Comments	397
          c.   Final Guidance	398
     4.   Chronic Toxicity Control  	 398

                              xvii

-------
          a.   Proposal	398
          b.   Comments	399
          c.   Final Guidance	 399
     5.   Numeric and Narrative Criteria	.  . 400
          a.   Proposal	400
          b.   Final Guidance	: 400
     6.   WET Test Methods	400
          a.   Proposal	•	400
          b.   Comments	400
          c.   Final Guidance	401
     7.   Permit Conditions 	 401
          a.   Data Indicates Reasonable Potential	401
               i.   Proposal	401
               ii.  Final Guidance	401
          b.   Insufficient Data to Determine Reasonable
               Potential	402
               i.   Proposal	402
               ii.  Comments	402
               iii. Final Guidance	403
          c.   Data Indicates No Reasonable Potential .... 403
     8.   Reasonable Potential Determinations 	 404
          a.   Proposal	404
               i.   Characterization of the Discharge .... 404
               ii.  Specific Acute WET Procedure  	 404
               iii. Specific Chronic WET Procedure  	 404
          b.   Comments	404
          c.   Final Guidance	406
               i.   Reasonable Potential Equation for Acute
                    WET	407
               id.  Reasonable Potential Equation for
                    Chronic WET	407

G.   Loading Limits	411
     1.   Proposal	411
     2.   Comments	411
          a.   Need for Loading Limits	411
          b.   Wet Weather Discharges	412
     3.   Final Guidance	412

H.   WQBELs Below the Level of Quantification 	 415
     1.   Expressing a WQBEL Below the Minimum
          Quantification Level  	 415
          a.   Proposal	415
          b.   Comments	415
          c.   Final Guidance	415
     2.   Compliance Issues 	 417
          a.   Proposal	417
          b.   Comments	417
          c.   Final Guidance	418
     3.   Compliance with the CEL	421
          a.   Proposal	421
          b.   Comments	:	422
          c.   Final Guidance	422
     4.   Pollution Minimization Program	422

                              xviii

-------
          a.   Proposal	422
          b.   Comments	423
          c.   Final Guidance	423
     5.   BCC Requirements	427
          a.   Proposal	427
          b.   Comments	427
          c:-   Final Guidance	427

I.   Compliance Schedules 	 429
     1.   Proposal	429
     2.   Final Guidance	429
          a.   Eligibility	429
          b.   Duration of Compliance Schedules . 	 431

IX.  EXECUTIVE ORDER 12866  	 435

A.   Introduction and Rationale for Estimating Costs and
     Benefits for the Great Lakes Water Quality Guidance  .  . 435

B.   Summary of Proposal	436
     1.   Costs	436
          a.   Method	436
          b.   Results	440
     2.   Cost-Effectiveness  	 441
          a.   Method	441
          b.   Results	442

C.   Revisions to Projected Costs	442
     1.   Summary of Revised Approach to Estimate Costs and
          Pollutant Load Reductions 	 442
          a.   Revised Criteria and Implementation
               Procedures	445
               i.   Dissolved Metals Water Quality Criteria  . 445
               ii.  Intake Credits	445
               iii. Additivity/TEFs	446
               iv.  Acilte Mixing Zones for WET	447
          b.   Criteria for Tier I and Tier II Pollutants .  . 447
          c.   Data for the Sample Facilities	450
          d.   Compliance Cost Decision Matrix  	 451
          e.   Pollution Prevention/Waste Minimization
               Costs	452
          f.   Indirect Dischargers 	 452
          g.   Revised Toxic Weights  	 453
     2.   Results	453
          a.   Estimated Compliance Costs 	 454
               i.   EPA's Low-End Estimate  	 454
               ii.  EPA's High-End Estimate 	 456
               iii. Comparison of Estimated Costs for the
                    Final Guidance to Costs of Proposed
                    Guidance	456
          b.   Estimated Pollutant Reductions 	 458

D-.   Major Is sues/Comments and Responses Related to
     Estimated Costs  	 462

                               xix

-------
          1.   Use of Pollution Prevention/Waste
               Minimization to Abhieve Guidance-Based
               Limits	.	-462
               a.   Comments	462
               b.   Response	463
          2.   Future Impact of Detection Levels	 463
               a.   Comments	.'  . 463
               b..   Response	464
          3.   Intake Credits	465
               a.   Comments	465
               b.   Response	465
          4.   Tier II Criteria	466
               a.   Comments	, .  . 466
               b.   Response	466
          5.   Wildlife Criteria/Mercury Criteria 	 466
               a.   Comments	466
               b.   Response	 . , .  . 466
          6.   Elimination of Mixing Zones for BCCs	467
               a.   Comments	467
               b.   Response	467
          7.   Antidegradation	468
               a.   Comments	468
               b.   Response	468
          8.   Additivity	469
               a.   Comments	469
               b-.   Response	469
          9.   Other Study Costs  	 470
               a.   Comments	470
               b.   Response	470

E.   Benefits	471
     1.   Summary of Proposal	471
          a.   Introduction	471
          b.   Qualitative Assessment of Benefits	471
          c.   Quantitative Assessment of Benefits Analysis  . 472
               i.   Economic Concepts Applicable to the
                    Benefits Analysis 	 472
               ii.  Benefits Methodologies   	 473
               iii. Results	474
     2.   Major Issues/Comments and Responses 	 477
          a.   Attribution of Benefits to the Guidance  , .  . 477
               i.   Comments	477
               ii.  Response	478
          b.   Risk Assessment	478
               i.   Comments	478
               ii.  Response	 478
          c.   Valuation Approaches Used for the Proposed
               Benefits Analysis  	 479
               i.   Comments	479
               ii.  Response	479
          d.   Limitations of the Case Study Approach and
               Specific Case Study Critiques  	 480
               i.   Comments	480
               ii.  Response	480

                                xx

-------
          e.   Economic Impacts of the Guidance	481
               i'.   Comments  . .".	481
               ii.  Response	482
     3.   Revised Benefits Estimates  	 482
          a.   Introduction	482
          b.   Updated Risk Assessments for Great Lakes
               Anglers	484
               i.   Sport Angler Risk Assessment  	 484
               ii.  Native American Risk Assessment	487
          c.   Revised Case Study Benefit-Cost Analyses for
               the Final Guidance	489
               i.   Fox River and Green Bay Case Study  .  .  . 489
               ii.  Saginaw River/Saginaw Bay Case Study   .  . 489
               iii. Black River Case Study	489
               iv.  Comparison of Benefits and Costs for the
                    Case Studies	490
               v.   Case Study Representativeness 	 490

X.  REGULATORY FLEXIBILITY ACT  	 493

XI.  ENHANCING THE INTERGOVERNMENTAL PARTNERSHIP UNDER
     EXECUTIVE ORDER 12875  	 495

XII.  PAPERWORK REDUCTION ACT	497
                               xxi

-------
                                Section I: Background
                         I.   BACKGROUND
A.    Description of Resource

      The Great Lakes are a unique natural resource that have played a vital
role  in the history and development of the United States and Canada.  The
Great Lakes consist of Lakes Superior, Huron, Michigan, Erie and Ontario and
their connecting channels  (i.e., the Saint Mary's River, Saint Clair River,
Detroit River, Niagara River and the Saint Lawrence River to the Canadian
Border).  The Great'Lakes plus all of the streams, rivers, lakes and other
bodies of water that are within the drainage basin of the Lakes collectively
comprise the Great Lakes System.

      The Great Lakes span over 750 miles across eight States--New York,
Pennsylvania, Ohio, Michigan, Indiana, Illinois, Wisconsin and Minnesota--and
the Province of Ontario.  The Lakes contain approximately 18 percent of the
world's and 95 percent of the United States' fresh surface water supply.  The
Great Lakes are a source of drinking water and energy, and are used for
recreational, transportation, agricultural and industrial purposes by the more
than  46 million Americans and Canadians who inhabit the Great Lakes region,
including 29 Native American tribes.  Over 1,000 industries and millions of
jobs  are dependent upon water from the Great Lakes.  The Great Lakes System
also  supports hundreds of species of aquatic life, wildlife and plants along
more  than 4,500 miles of coastline which boast six National Parks and
Lakeshores, six National Forests, seven National Wildlife Refuges, and
hundreds of State parks, forests and sanctuaries.  The prominent features of
the Great Lakes can be seen when the Earth is viewed from outer space.

      Although each of the Lakes is an interrelated and interdependent
component of the Great Lakes System, the Lakes vary in size, retention times,
and existing water quality.  Lake Superior, the northernmost Lake, is the
largest, deepest, coldest and most pristine of the Great Lakes, while Lake
Erie  is the smallest, shallowest, and the most susceptible of all of the Great
Lakes to the effects of urban and agricultural activities.  Lake Superior also
has the longest retention time--the average time it takes for one molecule of
water to exit the system--of 173 years, while Lake Erie has the shortest at
2.7 years.  Lake Huron alone acts as a drain for 51,700 of the 201,000 square
miles comprising the Great Lakes basin.  Lake Ontario, the easternmost Lake,
eventually receives all of the outflow from the other Great Lakes.  The waters
of four of the Great Lakes--Superior, Huron, Erie and Ontario--are shared by
the United States and Canada; only Lake Michigan is located wholly within the
United States.

      State and provincial Natural Heritage programs,  which have been
established throughout the Great Lakes basin for the purpose of identifying
elements (i.e., natural ecological communities or individual species)  of
biological diversity requiring protection, have identified 131 elements of
global significance within the Great Lakes basin that are crucial to
maintaining the region's biological diversity (The Nature Conservancy,  1994).
Of these,  31 are natural ecological community types.  The remaining 100
consist of plants,  insects, mollusks, fish, birds, reptiles and one mammal.
All of these species are either critically imperiled,  imperiled or rare on a
world-wide basis.  Nearly half (47 percent) are present either exclusively or
predominantly within the basin,  or are the best examples of their respective
species.  All 131 elements provide some of the most visible examples of the
biological character of the Great Lakes System.   The continued existence of
these species depends upon their ability to survive in the Great Lakes basin.

-------
2      Water Quality Guidance for the Great Lakes System — Supplementary Information Document

B.    Environmental Problems in the Great Lakes System

1.    Historical Nature of Environmental Problems in the Great Lakes System

      The Great Lakes Basin Ecosystem--the interacting components of air,
land, water and living organisms, including humans, that live within the Great
Lakes drainage basin--is a very young,  yet remarkable, ecosystem in geological
terms due to the physical and biological characteristics discussed above.  The
species which inhabit the Great Lakes basin reside in a wide range of
habitats.  Many of the features and processes that provided the environment
for these species to initially migrate to this region, adapt to its physical
characteristics, and evolve into even more unique and diverse species are also
susceptible to the multiple environmental stressors that humans have placed,
and continue to place, upon this important ecosystem.

      The Tribal relationship with the Great Lakes has been well documented
since being noted by the earliest Europeans to come into contact with the
indigenous people of the area.  The Great Lakes System played an integral role
in Tribal sustenance, stories, culture and spirituality to an extent unmatched
by any other group.

      Early settlement and related economic activities drastically changed
portions of the Great Lakes System.  The clearing of land for agricultural
purposes, commercial fishing, and industrialization placed a combination of
physical, biological and chemical stresses upon the Great Lakes Basin
Ecosystem, some of which continue today.

      By the late 1800s, all of the available agricultural lands within the
basin had been cleared and settled.  This harvesting and clearing of the Great
Lakes basin landscape imposed the first significant human stress condition
upon the Great Lakes Basin Ecosystem.  Temperatures in tributary streams rose
and entire forest systems, as well as wetlands and prairies, were frequently
lost as a result of these early logging and agricultural practices.  Also lost
were thousands of acres of critical habitat for wildlife.  The loss of  this
habitat caused many species to migrate to other areas or simply to perish due
to this physical alteration of the Great Lakes Basin Ecosystem  (The Nature
Conservancy, 1994).  In addition  to the above factors, dams and watercourse
modifications, human disturbance  of breeding and nesting locations, soil
erosion and the silting of spawning areas, primarily due to urbanization and
shoreline development pressures,  were responsible for reductions in suitable
fish, avian and wildlife habitat.

      The construction of  several canals from the beginning of the 1800s and
ending with the construction of the St. Lawrence Seaway in the 1950s
connecting the various Great Lakes, several  of their waterways and the  St.
Lawrence River during this same time period also impacted the Great Lakes
System.  Although these canals provided a valuable form of transportation for
the industries and populations located within and around the Great Lakes
basin, they also provided  a mechanism for the unintentional introduction of
exotic  (i.e., non-indigenous) species such  as the alewife  (Alosa
psuedoharengus), the  sea lamprey  (Petromyzon marinus), the Euroasian river
ruffe  (Gymnocephalus  cernuus), and more recently, the zebra mussel  (Dreissena
polymorpha).  Once these exotic  species entered the Great Lakes System, they
altered the natural biotic system by not only competing with native  species
for sustenance, but also by  serving as those species' prey.

        Great Lakes' fish were an important  resource valued by both  the  early
Native American  and European settlers.  Commercial fishing flourished
throughout  the mid- to  late-1800s.  Near the end of the century, however,  the
commercial  fishing industry  experienced a multitude of  stresses from the
introduction of  exotic  species,  overfishing, loss  of  habitat, and pollution.
This combination of physical, biological and chemical stressors eventually
impacted  the commercial fishing  industry in the Great Lakes beginning  in the
late nineteenth  century.

-------
                                Section I: Background
      Despite the amount of general social and economic changes, wild rice
harvesting, hunting, gathering, inland'shore and landlocked fishing continue
to be viable economic and cultural pursuits for modern Great Lakes Native
American communities.  There are presently more than 100 Tribal commercial
fishermen operating-on the Great Lakes.  In 1992, the Tribal commercial fish
harvest on U.S. waters of Lake Superior was more than 600 tons.
                  •
      Additionally, traditional wild rice harvesting of landlocked, flowing,
and inland Lake Superior shore waters is an economic, subsistence and cultural
mainstay of the annual cycle of life for Native American  (and non-Native
American) people in Minnesota, Wisconsin and Michigan.  Hunting of waterfowl
and small and large mammals is also very important to Tribal peoples for
subsistence and cultural reasons.

        The wastes that resulted from industrial and other activities that
prevailed in the basin during the late 1800s and early to mid-1900s placed
additional stresses upon the Great Lakes.  Iron and steel manufacturing, salt
brine production, copper and iron mining, paper making, power generation and
chemical manufacturing facilities utilized the natural resources and waterways
of the Great Lakes in their industrial and transportation processes.  Metals,
organic compounds and other chemical substances were introduced into the Great
Lakes System as a result of this industrialization.  In addition, the growth
of the human population in the Great Lakes basin and the corresponding
disposal of domestic wastes also contributed to the overall degradation of
Great Lakes water quality.

      By the mid-1900s, both the fishing quality and the water quality in the
Great Lakes basin were succumbing to these physical, biological and chemical
stresses.  For example, increased loadings of nutrients to the Great Lakes had
dramatically stimulated the growth of green plants and algae.  When these
materials decomposed, the Lakes experienced decreased levels of dissolved
oxygen in bottom waters, which in turn, caused the displacement of certain
species of insects and fish from the affected areas of the Great Lakes Basin
Ecosystem.  Environmental managers determined that a lakewide approach to
limit the loadings of phosphorous was the key to controlling the excessive
algal growth in the Great Lakes.

      The presence of environmentally persistent, bioaccumulative, chlorinated
organic contaminants was observed and identified as a serious environmental
threat in the 1960s  (Tanabe, 1988; Stevens et al., 1989).  Beginning in 1963,
adverse environmental impacts in the form of poor reproductive success and
high levels of DDT/DDE were observed in herring gulls in Lake Michigan
(Environment Canada, 1991) .  Expanded pollution control programs were
instituted in an attempt to stem and reverse the adverse effects posed by the
presence of these chemicals.  Although individual and concerted actions taken
by citizens, industries, governments and private organizations helped to
reduce the threats posed by these stressors on the Great Lakes Basin
Ecosystem, human activities still adversely impact this important natural
resource.

2.     Current Trends of Environmental Problems in the Great Lakes System

      a.    Impacts From Chemical Contamination.  In spite of the fact that
the Great Lakes contain 5,500 cubic miles of water that cover a total surface
area of 94,000 square miles, the Lakes are sensitive to the effects of a wide
range of pollutants from all sources, both point and nonpoint.  In particular,
the Great Lakes are susceptible to relatively non-degradable, bioaccumulative
chemicals because of the Lakes' unique physical, chemical and biological
characteristics.  Examples of these characteristics include:  (1) long
hydraulic retention times (i.e., relatively closed systems);  (2) low
biological productivity;  (3) low suspended solids concentrations;  (4) great
depth; and  (5) the presence of self-contained fish and wildlife populations
dependent on the Great Lakes System for water, habitat and food.  Taken
together, these characteristics result in such pollutants remaining in the

-------
4      Water Quality Guidance for the Great Lakes System — Supplementary Information Document

system for long periods of time and bioaccumulating in fish and  wildlife
tissue at concentrations which are orders of magnitude above the ambient
concentrations in the water column.  Once these pollutants are released into
the Great Lakes System, they will cycle within the System for decades,
exerting biological-effects and presenting relatively high levels of  risk  to
aquatic life, wildlife and humans which inhabit the Great Lakes  basin.

      The internal responses and processes that operate in the Great  Lakes
because of their depth and long hydraulic residence times mean that pollutants
recycle between biota, sediments and the water column.  As a result,  the Great
Lakes may still be experiencing the effects of historic discharges.   In
certain circumstances, for example, water column concentrations  may be  more
affected by the total mass of pollutants in the Lakes rather than the loadings
from point sources.  Additional discussion of the characteristics and
processes affecting contaminant levels in the Great Lakes is provided in the
preamble to the proposed Guidance  (58 PR at 20807-20817) .

      Scientists have detected 362 contaminants in the Great Lakes System.  Of
these, approximately one third have toxicological data showing that they can
have acute or chronic toxic effects on aquatic life, wildlife and/or  human
health (IJC, 1991).  Chemicals that have been found to bioaccumulate  at levels
of concern in the Great Lakes include, but are not limited to, PCBs,  mercury,
DDT, dioxin, chlordane, and mirex.  The main route of exposure to these
chemicals for humans is through the consumption of Great Lakes fish  (Colborn
et al., 1990).

      Potential adverse human health effects resulting from the  consumption of
contaminated fish include both the increased risk of cancer and  the potential
for systemic or noncancer risks such as kidney damage.  EPA has  calculated
health risks to populations in the Great Lakes System from consumption  of
contaminated fish based on exposure to eight bioaccumulative pollutants:
chlordane, DDT, dieldrin, hexachlorobenzene, mercury, PCBs, 2,3,7,8-
tetrachloro-p-dioxin  (TCDD), and toxaphene.  These chemicals were chosen based
on their potential to cause adverse human health effects  (i.e.,  cancer  or
disease)  and the availability of information on fish tissue contaminant
concentrations from the Great Lakes.

      Based on this data, EPA estimates that the lifetime cancer risks  for
Native Americans in the Great Lakes System due to ingestion of contaminated
fish at current concentrations range from 1.8 times 10"3 (1.8 in  one thousand)
for Lake Superior, to 3.7 times 10'2 (3.7 in one hundred) for Lake Michigan.
Estimated risks to low income minority sport anglers range from  2.5 times  10'3
for Lake Superior, to 1.2 times 10'2 for Lake Michigan.  Estimated risks for
other sport anglers range from 9.7 times 10"4 for Lake Superior,  to 4.5  times
10'3 for Lake Michigan (Regulatory Impact Analysis of the Final Great  Lakes
Water Quality Guidance. 1995. EPA  820-B-95-011)  (RIA), Chapter 6).  In
comparison, EPA has- long maintained that one times 10~*  (one in ten thousand)
to 1 times 10"6 (one in one million) is an appropriate range of risk to  protect
human health.

      EPA also estimates a high potential risk of systemic injury to
populations in the Great Lakes basin due to ingestion of fish contaminated
with these pollutants at current concentrations  (RIA, Chapter 6).  The
systemic adverse health effects associated with the assessed contaminants  are
described in appendix E of the RIA.

      While it is npt possible to  state conclusively that low-level exposure
to these chemicals  is or is not associated with adverse human health
reproductive effects, evidence from wildlife studies  (Fox, 1992; Flint  and
Vena, 1991) and epidemiological investigations of occupational exposures  to
these chemicals indicates that they may be able to alter human reproduction.
Epidemiological studies that have  addressed adverse pregnancy outcomes  in
populations in the Great Lakes have shown some potential effects of  concern

-------
                                Section I: Background
 (Swain, 1991; Fein et al., 1984; Smith, 1984), while other studies have  shown
 little or no effects  (Dar et al., 1992).

      Additionally, new research into the ability of several organochlorine
 compounds to mimic hormones has  recently been published  (Environment Canada,
 1991).  Studies have shown that  PCBs  (Korach et al., 1988) and DDE and DDT
 (McLachlan et al., 1987) have been found to have weak estrogenic abilities.
 There has been speculation among researchers that an increased incidence of
 breast cancer may occur among women exposed to organochlorines  (Manz et  al.,
 1991; Falck et al., 1992) .  However, the link between these compounds and
 breast cancer has not been clearly established.

      As stated earlier, adverse effects observed in fish, birds and mammals
 in the Great Lakes have been attributed to contaminants.  These effects
 include death, eggshell thinning, reduced hatching  success, abnormal behavior,
 deformities  (such as crossed beaks and club feet) and population declines.
 For several species of birds (e.g., double=crested  cormorants, Caspian terns,
 Forster's terns, common terns, ring-billed gulls, herring gulls) in the  Great
 Lakes, a suite of effects including embryo mortality, edema, growth
 retardation, and deformities have been reported and considered similar to
 chick edema syndrome associated  with some classes of PCBs  (Gilbertson et al.,
 1991; Fox et al., 1991; Tillitt  et al., 1992; Environment Canada, 1991;  Mora
 et al., 1993; Kurita and Ludwig, 1988; Yamashita et al., 1993).  More subtle
 effects, such as abnormalities in the thyroid, endocrine and liver systems,
 have also been observed in organisms within these populations  (Gilbertson et
 al., 1991; Colborn, 1991) .  Table 1-1 below summarizes the fish and wildlife
 species thought to be affected by contaminants in the Great Lakes and is
 available in the docket for this rulemaking.

      Of the eleven wildlife species which showed evidence of contaminant
 impacts in the past, three of these species (bald eagles, cormorants and
 herring gulls) are providing some evidence of recent improvements  (Giesy et
 al., 1994).  Current concentrations of organochlorine compounds in Great Lakes
 fish and fish-eating birds are less than during the 1960s and 1970s  (Giesy et
 al., 1994).  Although these trends for these species are encouraging,
persistent toxic chemicals still exist at levels that continue to produce
 adverse effects on fish-eating wildlife and fish  (Giesy et al., 1994) .

      With respect to Great Lakes fish populations  and fish communities, some
 researchers argue that the effects of contaminants  on fish populations and
 communities are difficult to separate from the effects of overfishing, habitat
 loss and the introduction of exotic species (Environment Canada, 1991) .  Great
Lakes lake trout populations, for example, have been devastated by
overharvesting and the introduction of the sea lamprey, as well as subjected
to a variety of impairments from toxic pollutants (Environment Canada, 1991).

      EPA recognizes that chemical contaminants are only one of the threats to
the health of the Great Lakes Basin Ecosystem.  The continued decline of
physical habitat and the presence of exotic (i.e.,  non-indigenous) species,
for example, are also of concern.  The final Guidance provides a consistent
approach for reducing the threat from chemical stressors to the Great Lakes
Basin Ecosystem.  Other programs and activities are currently being
implemented by EPA and other Federal and State agencies to address biological,
chemical and physical problems in the Great Lakes (see section I.D below).

-------
i
8
I I
V f
o J5
tt CJ
|
1 1
8><2
11
IE
I «
Is
l|
oo 13
W P

1
o
s
e
o
•8
3

a-!
* w



•1 I
1 2




§ .
•^ eo
~ §

Qd i?
0 S
fiu Q
GO
!


w


u



u
z

z

X







X




-





X




1


w


w



w

a














-





X







X


g





X









X




-





X




Double-crested
Cormorant |


X


X





X









X




C"





X




Black-crowned Night
Heron |


W


w





X

u







X




-





X




Bald Eagle


X


X



X

X

X







X















Herring Oull


W
z


X







X























Ring-billed Oull |


W


X



W
z











X




X










Caspian Tern


X


X





X









X















Common Tern


X


X



X











X















Forster's Tern j


a


X



a
z

<

u







X




w





z




.1
CO


X








Z









X















Lake Trout


X








Z

























•o
1
e
CO


X


X





Z

























White Sucker






















1
1
0
•s
•2
1
i §
1?
f ^
u
5 e-





= effects documented
A = not applicable


-------
                                Section I: Background
      b.    Trends in Contaminant Levels.  Concentrations of certain
bioaccumulative contaminants, such as PCBs, DDT, dieldrin, and oxychlordane,
have declined significantly over the past 20-30 years, as evidenced by basin-
wide decreases of these pollutants in water, fish, bird eggs, and  sediments
 (DeVault, 1993a; DeVault 1993b; Environment Canada, 1991).  These  declines are
believed to be attributable to bans and restrictions that were placed on  the
manufacture and use*of these chemicals from the late 1960s through the mid-
1970s.  Decreased chemical levels in the Great Lakes are also attributable, in
part, to existing regulatory controls, industrial source controls, and the
Lakes' ability to respond to changes in loads following remedial actions
 (Deposition of Air Pollutants to the Great Waters: First Report to Congress,
EPA 453-R-93-055.D.S. EPA, 1994) (Great Waters Report).

      Recent water column data for both Lakes Superior and Michigan indicate
declines for PCBs are continuing.  Pollutant levels for PCBs, for  example,
declined from 1.73 ng/I> to 0.18 /ig/L in Lake Superior between 1978 and 1992
 {DeVault, 1993).  Between 1980 and 1993, PCB levels in southern Lake Michigan
also declined, from 1.8 /xg/L to 0.2 ^g/L.  Similar results were observed  for
DDT, 2,3,7,8-TCDD and 2,3,7,8-tetrachloro-dibenzofuran  (TCDF) across the  Great
Lakes basin, and in levels of mirex found in Lake Ontario.  Significant
declines in both net loadings and environmental concentrations of  lead and
mercury have also occurred since the mid-1970s.

      Due to difficulties with reliably measuring chemicals at extremely  low
concentrations in water, scientists have often relied upon fish tissue
residues in some species, such as lake trout and walleye, to provide an
indication of chemicals in the water column as well as to indicate any changes
in water column pollutant concentrations.  In the preamble to the  proposed
Guidance, EPA provided environmental information regarding several
bioaccumulative chemicals (e.g., PCBs, DDT) in game fish  (58 FR 20809-20816).
EPA stated that contaminant levels in fish had showed dramatic declines
through the mid-1980s, but were currently fluctuating around a lower level.
This conclusion was based on data showing that earlier declines may have  been
leveling off in recent years (DeVault, 1993).

      The lake trout data which were presented in the preamble to  the proposed
Guidance illustrated that existing PCB levels in such fish were too high  to
ensure adequate projection of human health and that the rate of decline in
fish tissue levels had slowed.   Similarly, the data for coho salmon showed
that PCB levels in this species remained above the fish tissue concentrations
of concern and that the trend of PCBs in this species between 1984 through
1990 had leveled off in both Lakes Michigan and Erie.

      In addition to the data presented in the preamble to the proposed
Guidance, lake trout data for Lakes Superior and Michigan, and walleye data
for Lake Erie, indicate that concentrations of PCBs and DDT initially declined
rapidly, but leveled off in the mid-1980s.  Similarly, concentrations of  PCBs
and DDT in Lake Erie walleye showed no significant changes since 1982.  PCB
and DDT concentrations in lake trout from Lakes Michigan and Superior had not
changed significantly since 1986 (DeVault, 1993).

      Current trends indicate that many water quality objectives and fish
tissue criteria for the protection of human health are still being exceeded
(DeVault et al.,  1994).  In Lake Erie, for example, water column
concentrations of both PCBs and DDT declined from 1980 through 1984, and  have
not changed significantly through 1992, with levels of PCBs still
substantially above the fish tissue concentrations of concern (DeVault et al.,
1994).   PCB and DDT concentrations in coho salmon filets from Lake Michigan
showed rapid declines from 1980 through 1983, but increased between 1983  and
1992.  See Figures i-1 and 1-2.  Thus, even though water column concentrations
of PCBs continued to decline in Lakes Superior and Michigan through the early
1990s,  an increase or lack of decline of PCBs has been observed in Lake
Superior and Lake Michigan fish tissue residues since the mid-1980s.

-------
8      Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      The difference between the declining rates of PCBs in the water column
concentrations and the fish tissue residues in Lakes Superior and Michigan
indicate that these two systems have not reached an equilibrium of existing
loads and bottom sediments.  One possible explanation provided by DeVault et
al.  (1994) is that several recent changes to the zooplankton and forage fish
communities in the Great Lakes System may be responsible.  These changes were
observed concurrently with the slowing and reversals in contaminant declines
discussed above.

      Despite the trends discussed above, the fish tissue concentrations of
PCBs described above show that the acceptable fish tissue concentration of 0.1
pig/g for the protection of biological resources is still being exceeded across
most of the Great Lakes basin  (DeVault et al., 1994).  These contaminant
concentration levels continue to result in exceedances of State and Provincial
human health criteria, potential risks to human health from cancer and
noncancer systemic injuries, and fish consumption advisories for PCBs in each
of the Great Lakes.  Based on this information, further decreases in loadings
to the Great Lakes are necessary in order to meet both existing and proposed
future water quality objectives and criteria.  EPA and the Great Lakes States
and Tribes are currently addressing residual pollutant problems in the Great
Lakes through a variety of regulatory and voluntary programs, some of which
are discussed in section I.D below.

      c.    Fish Consumption Advisories.  Due to the presence of contaminants
at levels above current standards or guidelines, elevated contaminant levels
in wildlife, and contaminated sediment hot spots (DeVault et al., 1994), fish
consumption advisories exist in all of the Great Lakes States, including
various waters located in the Great Lakes basin.  The Great Lakes States issue
these fish contaminant advisories based on a system incorporating and weighing
such factors as the type of contaminants found in Great Lakes fish flesh,
contaminant levels in fish of various sizes and species, the typical
consumption rates of sensitive populations such as sport anglers, nursing and
pregnant women, and an evaluation of the human health risks of the potential
impacts of these substances.

      Pollutants for which these fish advisories are issued include eight of
the 22 bioaccumulative chemicals of concern  (BCCs)  identified in the final
Guidance.  High-risk groups, which fish consumption advisories are established
to protect, include mothers who breastfeed their young, due to the presence of
pollutants in human tissue to which babies continue to be exposed after they
are born.  Also at risk of greater exposure are those sport anglers, Native
Americans, and the urban poor who may consume more Great Lakes fish than the
average consumption of the basin population as a whole.  The Federal
government is examining the impacts to these higher-risk populations through a
number of studies initiated during 1993 and 1994 by the Agency for Toxic
Substances and Disease Registry, within the Department of Health and Human
Services.  As a result of the factors mentioned above, impacts from fish
consumption to higher-risk populations are being taken into consideration in
environmental regulation activities,  including the final Guidance.

-------
                      Section I: Background
0)
   (0

   §
 O '4=
o
 •
I
o
CO
<

o
i
i
|

i
o
           w
                                 =   '    '- L
                                 E  /*    if
                                 —  .  \   -" '
o>

OJ



O)

o
O)

O)
00

00
00

r-
00

(0
00

in
oo


oo

«
00

01
00


00

o
00

o>
              €0
                •
              o
                                  o
                                   »
                                  o
                                        tr
                                        <
                                        Hi
        6/Bn    NOIlVaiN30NOO 1QQ
FIGURE 1-1  (Devault et al. 1986; Devault, 1993)

-------
10
Water Quality Guidance for the Great Lakes System - Supplementary Information Document
i*
"5 3
CO
    CO
0  £
-S -2
O -M
o  2
   *-•
»  £
4)  

O) o> o 00 0> 00 00 00 N. 00 • oo 00 CO 00 CM 00 00 o 00 0> q o B/Bn NOIlVaiNBONOO 3Od FIGURE 1-2 (Devault et al. 1986; Devault, 1993)


-------
                                Section I: Background                             11
C.    History of the Great Lakes Water Quality Guidance

l.    The Great Lakes Toxic Substances Control Agreement

      Against this background of environmental concerns, the Governors of the
eight Great Lakes States signed the Great Lakes Toxic Substances Control
Agreement (Governors' Agreement) pledging the States' cooperation in studying,
managing and monitoring the Great Lakes as an integrated ecosystem in 1986.
The purpose of the Governors Agreement was to establish a framework for
coordinated regional action in controlling toxic substances entering the Great
Lakes System; to further the understanding and control of toxic contaminants;
and to develop common goals, management practices and control strategies for
toxics to ensure a cleaner and healthier Great Lakes Basin Ecosystem.
      The Governors believed that maintaining and improving the quality of
Great Lakes waters would sustain water supply systems as well as commercial,
manufacturing and recreation industries, while creating new economic
development opportunities.  They agreed to preserve the value of the Great
Lakes by striving to maintain a high standard of water quality when
establishing regulatory standards, and committed to continue reducing toxics
in the Great Lakes System to the maximum extent possible.  For further
discussion of the principles outlined in the Governors' Agreement, see the
preamble to the proposed Guidance (58 PR 20820) .
      The Great Lakes Governors also realized that a number of toxic
substances control programs and measures had been responsible for achieving
significant reductions in certain organic pesticides and metals.  These
reductions are discussed in more detail in section I.B.2 above.  However, the
Governors recognized that the complexity of the pollutant problem in the Great
Lakes System called for continuing action.  In many cases, the Governors
believed new and creative initiatives would be required to achieve the goal of
prohibiting the discharge of toxic pollutants in toxic amounts in order to
protect the water quality of the Great Lakes.
      To implement the goals of the Governors' Agreement, the signatory States
directed their environmental administrators to jointly develop an agreement
for coordinating the control of toxic releases within the Great Lakes System.
This coordinated effort between the Great Lakes States contributed to the
development of the Great Lakes Water Quality Initiative.

2.    The Great Lakes Water Quality Initiative

      EPA and the Great Lakes States initiated the Great Lakes Water Quality
Initiative (Initiative) in 1989 to further address the environmental concerns
identified in the Governors Agreement.  The Initiative was intended to provide
a forum for Great Lakes States and EPA to develop uniform water quality
criteria and implementation procedures for the Great Lakes basin.  The
participants planned to use the results of this effort as a basis for revising
State water quality standards and permit programs pursuant to sections 303(c)
and 402 of the Clean Water Act  (CWA).     Three committees were formed under
the Initiative.  A Steering Committee, consisting of directors of water
programs from EPA's National and Regional offices and the Great Lakes States'
environmental agencies, as co-regulators of CWA water quality programs,
discussed policy, scientific and technical issues and directed the work of the
Technical Work Group.  The Technical Work Group, consisting of technical staff
from the Great Lakes States environmental agencies, EPA, the U.S. Fish and
Wildlife Service and the U.S. National Park Service, prepared proposals for
submission to the Steering Committee.  The Public Participation Work Group,
consisting of representatives from environmental groups, municipalities,
industry and academia, observed the deliberations of the other two groups,
advised them of the public's concerns, and kept its various constituencies
apprised of Initiative activities.  These committees are collectively known as
the Initiative Committees.

      Of particular' concern to the Steering Committee were those pollutants
which exhibit the potential to produce System-wide impacts.  Based upon
observed impairments to the Great Lakes, the States believed that these

-------
12     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

pollutants pose one of the greatest threats to the health of the Great  Lakes
Basin Ecosystem.  The Steering Committee believed that further reductions  in
loadings of such pollutants from all sources should be pursued.  The Steering
Committee wanted actions to be taken to ensure that problems with pollutants
which could potentially cause impairment of beneficial uses in the Great Lakes
Basin Ecosystem would not develop in the future.  Therefore, the Steering
Committee charged EPA and State staff on the Technical Work Group to define
the pollutants that,warrant more stringent controls, and to draft additional
control approaches for those pollutants.  Further discussion on the factors
chosen for identifying and controlling such pollutants is provided in section
II.C.8 of this document.

3.     The Great Lakes Critical Programs Act of 1990

      The enactment of the Great Lakes Critical Programs Act  (CPA) of 1990
(Public Law 101-596, November 16, 1990) codified the ongoing Initiative effort
into the CWA.  Section 101 of the CPA  (CWA section 118(c)(2)) requires  EPA to
publish proposed and final water quality guidance for the Great Lakes System
which conforms with-the objectives and provisions of the Great Lakes Water
Quality Agreement (GLWQA) and is no less restrictive than provisions of the
CWA and National water quality criteria and guidance.  The final Guidance must
specify minimum requirements for the waters of the Great Lakes System in three
areas: (a) water quality criteria, including numerical limits on pollutants in
ambient Great Lakes waters to protect human health, aquatic life and wildlife;
(b) antidegradation policies; and (c) implementation procedures.  This  section
also requires Great Lakes States to adopt water quality standards,
antidegradation policies and implementation procedures consistent with  the
final Guidance within two years of EPA's publication of the final Guidance, or
be subject to EPA promulgation of the provisions within the same two-year
period.  (CWA section 118(c)(2)(C)).  Further discussion of the CPA is provided
in the preamble to the proposed Guidance (58 FR 20823).

      The substantive scope of the final Guidance is consistent with the
initial 1989 expectations for the original Initiative.  The final Guidance
includes water quality criteria and methodologies, an antidegradation policy,
and implementation procedures that were initially developed through an  open,
collaborative process by staff from the Great Lakes States and EPA with input
from other stakeholders in the basin.  The principal changes in the scope  of
the final Guidance that have occurred since the initial planning for the
Initiative resulted, from the enactment of the CPA.

4.     Principles Underlying the Final Water Quality Guidance for the Great
      Lakes System

      The final Guidance is the culmination of a six-year cooperative effort
that included participation by the eight Great Lakes States, the environmental
community, academia, industry, municipalities and EPA Regional and National
offices.  EPA's development of the final Guidance continued the work begun by
the Initiative Committees and was guided by the following six general
principles:

      a.    Use the Best Available Science to Provide Protection to Human
Health. Wildlife, and Aquatic Life.  EPA and the Initiative Committees  have
been committed to using the best available science to develop programs  to
protect the Great Lakes System since the beginning of this ambitious project.
In the 1986 Governors' Agreement, the Governors of the Great Lakes States
recognized that the problem of persistent toxic substances was the foremost
environmental issue confronting the Great Lakes.  They also recognized  that
the regulation of toxic contaminants was scientifically  complex because the
pollutants are numerous, their pathways into the Lakes are varied, and  their
effects on the environment, aquatic life and human health were not completely
understood.  Based  on the  importance  of the Great Lakes  Basin Ecosystem and
the documented adverse  effects from toxic contamination,  however, the
Governors directed  their environmental administrators  to jointly  develop an

-------
                                Section I: Background                             13
agreement and procedure for coordinating the control of toxic releases and
achieving greater uniformity of regulations governing such releases within the
Great Lakes basin. •

      As discussed further above, the Initiative was subsequently created to
begin work on these goals.  EPA and the Great Lakes States, with input from
interested parties- in the basin, began collecting and analyzing data,
comparing regulatory requirements and technical guidance in their various
jurisdictions, and drafting specific methodologies and prpcedures to control
the discharge of toxic contaminants.  The provisions of the final Guidance
were based in large part on these prior efforts of the Initiative Committees,
and incorporate the best available science to protect human health, wildlife
and aquatic life intthe Great Lakes System.  For example, the final Guidance
includes new criteria and methodologies developed by the Initiative Committees
to specifically protect wildlife; incorporates recent data on the
bioavailability of metals into the aquatic life criteria and methodologies;
incorporates Great Lakes-specific data on fish consumption rates and fish
lipid contents into the human health criteria; and provides a better
methodology to determine the bioaccumulation properties of individual
pollutants.

      b.    Recognize the Unique Nature of the Great Lakes Basin Ecosystem.
The final Guidance also reflects the unique nature of the Great Lakes Basin
Ecosystem by establishing special provisions for BCCs.  The special provisions
for BCCs initially developed by the Initiative Committees and incorporated
into the final Guidance include more stringent antidegradation procedures, to
ensure that these problems are minimized; phase out and elimination of mixing
zones for existing discharges of BCCs after 12 years, with limited exceptions,
to reduce overall loadings to the Lakes;  more extensive data generation
requirements to ensure that they are not underregulated for lack of data; and
the development of water quality criteria that will protect wildlife that feed
on aquatic prey.  As discussed in sections I.A and I.B above, it is reasonable
and appropriate to establish special provisions for these chemicals because of
the physical, chemical and biological characteristics of the Great Lakes
System, and the documented environmental harm to the ecosystem from the past
and continuing presence of these types of pollutants.

      The final Guidance is not only designed to address existing problems,
but also to prevent emerging and potential problems posed by additional
chemicals in the future which may damage the overall health of the Great Lakes
System.  The experience with such pollutants as DDT and PCBs indicates that it
takes many decades to overcome the damage to the ecosystem caused by even
short-term discharges, and that prevention would have been dramatically less
costly than cleanup.   The elements of the final Guidance provide a coordinated
ecosystem approach for addressing possible pollutant problems before they
produce adverse and long-lasting basinwide impacts, rather than waiting to see
what the future impacts of the pollutants might be before acting to control
them.  The comprehensive approach used in the development of the final
Guidance provides regulatory authorities with both remedial and preventive
ways of gauging the actions and potential effects of chemical stressors upon
the Great Lakes Basin Ecosystem.  The methodologies, policies and procedures
contained in the final Guidance provide mechanisms for appropriately
addressing both pollutants that have been or may in the future be documented
as chemicals of concern.  Additional discussion of the characteristics and
processes affecting contaminant levels in the Great Lakes and special controls
for BCCs is also provided in the preamble to the proposed Guidance (58 PR at
20807-23 and 20844-45) and in section II.C.8 of this document.

      c.    Promote Consistency in Standards and Implementation Procedures
While Allowing Appropriate Flexibility to States and Tribes.  Promoting
consistency in standards and implementation procedures while providing for
appropriate State and Tribal flexibility was• the third principle in EPA's
development of the final Guidance.  A primary impetus for the Governors'
Agreement, the Initiative, and the requirements set forth in the CPA was a

-------
14     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

recognition of the need to promote consistency in the minimum water quality
standards, antidegradation policies, and implementation procedures adopted by
the Great Lakes States to protect human health, aquatic life and wildlife.
Although provisions in the CWA provide for the adoption of and periodic
revisions to State water quality criteria, such provisions do not necessarily
ensure that water quality criteria of adjoining States are consistent within a
shared waterbody.  For example, State acute ambient water quality criteria in
place in six of the.eight Great Lakes States include a range of 1.79 to 15.0
/ig/L for cadmium in order to protect against acute effects for aquatic  life,
and from 0.21 to 1.33 /zg/L for dieldrin.  Other examples of variations  in
acute ambient water quality criteria include: nickel, which ranges from 290.30
jjg/L -to 852.669 jig/L; lindane, with a range of no criteria in place to  1.32
fjg/L; and mercury, ranging from 0.5 M9/L to 2.4 jig/L.  Similar ranges and
disparities exist in the chronic ambient water quality criteria and human
health criteria adopted by the Great Lakes States.

      Disparities also exist among State procedures to derive individual
discharge permits from water quality criteria.  Wide variations exist,  for
example, in procedures for granting mixing zones, interpretation of background
levels of pollutants, consideration of pollutants present in intake waters,
controls for pollutants present in concentrations below the level of
detection, and determination of appropriate levels for the discharge of
multiple pollutants.  Additionally, when calculating exposure factors in fish
that will be consumed by humans and wildlife, some States consider
accumulation through multiple steps in the food chain  (bioaccumulation) while
others consider only the single step of concentration from the water column
(bioconcentration).  Further disparities exist in different translator
methodologies in deriving numeric values for implementing narrative water
quality criteria; different assumptions when calculating total maximum  daily
loads (TMDLs) and wasteload allocations, including different assumptions about
background concentrations, mixing zones, receiving water flows, or
environmental fate; and different practices in deciding what pollutants need
to be regulated in a discharge, what effect detection limits have on
compliance determinations, and how to develop whole effluent toxicity
limitations.

      These inconsistencies in State standards and implementation procedures
have resulted in the disparate regulation of point source discharges, which
may have led to disputes in the past among the Great Lakes States.  In  the
Governors' Agreement, however, the Governors recognized that the water
resources of the basin transcend political boundaries and committed to  taking
steps to manage the Great Lakes as an integrated ecosystem.  The Great  Lakes
States, as part- of the Initiative Committees, recommended provisions that were
ultimately included in the proposed Guidance for coordinated review and
development based on their extensive experience in administering State  water
programs and knowledge of the significant differences in these programs within
the basin.  The final Guidance incorporates the work begun by the Initiative
Committees to identify these disparities and improve consistency in water
quality standards and permit procedures in the Great Lakes System.

      Although improved consistency in State water programs is a primary goal
of the final Guidance, it is also necessary to provide appropriate flexibility
to States and Tribes in the development and implementation of water programs.
In overseeing States' implementation  of the CWA, EPA has found that reasonable
flexibility is not only necessary to  accommodate site-specific situations and
unforeseen circumstances, but is also appropriate to enable innovation  and
progress as new approaches and information become available.  Many commenters
urged EPA to evaluate the appropriate level of flexibility provided to  States
and Tribes in the proposed Guidance provisions.  EPA reviewed all sections  of
the proposed Guidance and all comments received to determine the appropriate
level of flexibility needed to address these concerns  while still providing a
minimum level of  consistency between  the  State and Tribal programs.  Based  on
this review, the  final Guidance provides  flexibility for State and Tribal

-------
                                Section I: Background                             15
adoption and implementation of the final Guidance provisions in many areas,
including the following:

          Antidegradation:  Great Lakes States and Tribes may develop their
own approaches for implementing the prohibition against deliberate actions of
dischargers that increase the rate of mass loading of BCCs without an approved
antidegradation demonstration.  Furthermore, States and Tribes have
flexibility in adopting antidegradation provisions regarding non-BCCs.

          TMDLs:  Great Lakes States and Tribes may use assessment and
remediation plans for the purposes of appendix F of part 132 if the State or
Tribe certifies that the assessment and remediation plan meets certain TMDL-
related provisions in the final Guidance and public participation requirements
applicable to TMDLs, and if EPA approves such plan.  Thus, States have the
flexibility in many cases to use Lakewide Management Plans (LAMPs), Remedial
Action Plans (RAPs) and State Water Quality Management Plans in lieu of TMDLs.

          Intake Credits:  Great Lakes States and Tribes may consider the
presence of intake water pollutants in establishing water quality-based
effluent limits (WQBELs) in accordance with procedure 5 of appendix F to part
132, as discussed further in section VIII.E of this document.

          Site-Specific Modifications: Great Lakes States and Tribes may adopt
either more or less'stringent modifications to human health,  wildlife, and
aquatic life criteria and BAFs based on site-specific circumstances specified
in procedure 1 of appendix F to part 132, as discussed further in section
VIII.A of this document.  All criteria, however, must not be likely to cause
jeopardy to threatened or endangered species listed or proposed to be listed
under the Federal Endangered Species Act  (ESA) .

          Variances:  Great Lakes States and Tribes may grant variances from
water quality standards based on the factors identified in procedure 2 of
appendix F to part 132, discussed further in section VIII.B of this document.

          Compliance Schedules:  Great Lakes States and Tribes may allow
existing Great Lakes dischargers additional time to comply with permit limits
in order to collect data to derive new or revised Tier I criteria and Tier II
values in accordance with procedure 9 of appendix F to part 132, as discussed
further in section VIII.I of this document.

          Mixing Zones:  Great Lakes States and Tribes may authorize mixing
zones for existing discharges of BCCs after the 10-year phase-out period in
accordance with procedure 3.B of appendix F to part 132, if the permitting
authority determines, among other things, that the discharger has reduced its
discharge of the BCC for which a mixing zone is sought to the maximum extent
possible.  Water conservation efforts that result in overall reductions of
BCCs are also allowed even if they result in higher effluent concentrations.
These provisions are discussed further in section VIII.C.4 of this document.

          Scientific Defensibility Exclusion:  Great Lakes States and Tribes
may apply alternate procedures consistent with Federal, State, and Tribal
requirements upon demonstration that a provision in the final Guidance would
not be scientifically defensible if applied to a particular pollutant in one
or more sites.   This provision, in § 132.4(h) of the final Guidance, is
discussed further in section II.C.6 of this document.

          Reduced Detail:  In many instances, EPA has revised the proposed
Guidance to reduce the amount of detail in the provisions without sacrificing
the objectives of the provisions.  Examples of such revisions include
simplification of procedures for developing TMDLs in procedure 3 of appendix F
to part 132,  discussed further in section VIII.C of this document, and
simplification of procedures for determining reasonable potential to exceed
water quality standards in procedure 5.B of appendix F to part 132, discussed
further in section VIII.E of this document.

-------
16     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

          Other Provisions:  Flexibility is also present in provisions for the
exercise of best professional judgment toy the Great Lakes States and Tribes
when implementing many individual provisions in the final Guidance including:
determining the appropriate uncertainty factors in the human health and
wildlife criteria methodologies; selection of data sets for establishing water
quality criteria; identifying reasonable and prudent measures in
antidegradation provisions; and specifying appropriate margins of safety when
developing TMDLs.  In all cases, of course, State and Tribal provisions would
need to be scientifically defensible and consistent with all applicable
regulatory requirements.

      d.    Establish Equitable Strategies to Control Pollution Sources.  Many
commenters argued that the proposed Guidance unfairly focused on point source
discharges.  They asserted that nonpoint sources or diffuse sources of
pollution, such as air emissions, are responsible for most of the loadings of
some pollutants of concern in the Great Lakes, that increased regulation of
point sources will be inequitable and expensive, and that the final Guidance
will not result in any environmental improvement given the large, continuing
contribution of toxic pollutants by nonpoint sources.

      EPA recognizes that regulation of point source discharges alone cannot
address all existing or future environmental problems from toxic pollutants in
the Great Lakes.  In addition to discharges from point sources, toxic
pollutants are also,contributed to the Great Lakes from industrial and
municipal emissions to the air, resuspension of pollutants from contaminated
sediments, and urban and agricultural runoff, hazardous waste and Superfund
sites, and spills.  Restoration and maintenance of a healthy ecosystem will
require significant efforts in all of these areas.  As discussed further in
section I.D below, EPA, Canada and the Great Lakes States and Tribes are
currently implementing or developing many voluntary and regulatory programs to
address these and other nonpoint sources of environmental contaminants in the
Great Lakes.

      Additionally, EPA intends to use the scientific data developed in the
final Guidance and fiew or revised water quality criteria subsequently adopted
by Great Lakes States and Tribes in evaluating and determining appropriate
levels of control in other environmental programs.  For example, EPA's future
biennial reports under section 112(m) of the Clean Air Act will consider the
extent to which air discharges cause or contribute to exceedances of water
quality criteria in assessing whether additional air emission standards or
control measures are necessary to prevent serious adverse effects.  Similarly,
once the provisions of the final Guidance are adopted by the Great Lakes
States or Tribes, they will serve as applicable or relevant and appropriate
requirements  (ARARs) for on-site responses under the Comprehensive
Environmental Response, Compensation and Liability Act  (CERCLA).  EPA will
also consider the data and criteria developed for the final Guidance,
including the information on bioaccumulative pollutants, in developing or
evaluating LaMPs and RAPs under  section 118  of the CWA and Article VI, Annex 2
of the GLWQA; determination of corrective action requirements under sections
3004(u), 3008(h), or 7003 of the Solid Waste Disposal Act; new or existing
chemical reviews under the Toxic Substances  Control Act  (TSCA); pesticide
reviews under the Federal  Insecticide, Fungicide and Rodenticide Act  (FIFRA) ,-
and reporting requirements for toxic releases under the Emergency Planning and
Community Right-to-Know Act  (EPCRA).

      The final  Guidance also  includes provisions to address the contribution
of pollutants by nonpoint  sources.  First,  the water quality criteria  to
protect human health, wildlife and aquatic  life, and the antidegradation
provisions  apply to the waters  in  the Great Lakes System regardless of whether
discharges  to the water are  from point or nonpoint sources.  Accordingly,  any
regulatory programs for nonpoint sources that require compliance with  water
quality standards would also be  subject to  the  criteria and antidegradation
provisions  of the final Guidance once they  are  adopted  into State or Tribal
standards.

-------
                                Section I: Background                             17
      Second, several elements of the final Guidance would, after State,
Tribal or Federal promulgation, require or allow permitting authorities to
consider the presence of pollutants in ambient waters--including pollutants
from nonpoint source dischargers--in establishing WQBELs for point sources.
For example, permit'authorities may consider the presence of other point or
nonpoint source discharges when evaluating whether to grant a variance from
water quality .criteria under procedure 2 of appendix F to part 132.
Additionally, the provisions for TMDLs in procedure 3 of appendix F to part
132 address nonpoint sources by requiring allocation of the available load
capacity of receiving waters that do not meet water quality criteria among all
sources of the pollutant, including nonpoint sources.  The development of
TMDLs is the preferred mechanism for addressing equitable division of the
loading capacities of these nonattained waters.  Because TMDLs have not been
completed for most nonattained waters, however, the final Guidance promotes
the development of TMDLs through a phased approach, where appropriate, and
provides for short-term regulatory relief to point source dischargers in the
absence of TMDLs through intake credits, variances, and other water quality
permitting procedures.  These provisions are discussed in sections VIII.B,
VTII.C, and VIII.F of this document.

      e.    Promote Pollution Prevention Practices.  The final Guidance also
promotes pollution prevention practices consistent with EPA1 s National
Pollution Prevention Strategy and the Pollution Prevention Action Plan for the
Great Lakes.  The Pollution Prevention Act of 1990 declares as National policy
that reducing the sources of pollution is the preferred approach to
environmental protection (Pub. L. 101-508, section 6601-6610 104 Stat. 1388,
codified at section 13101-13109 West Supp. 1991) .  When source reductions are
not possible, however, recycling, treating and properly disposing of
pollutants in an environmentally safe manner complete the hierarchy of
management options designed to prevent pollution from entering the
environment.

      Consistent with the goals of the Pollution Prevention Act, EPA developed
the Great Lakes Pollution Prevention Action Plan (April, 1991) .  The Great
Lakes Pollution Prevention Action Plan highlights how EPA, in partnership with
the Great Lakes States, will incorporate pollution prevention into actions
designed to reduce'the use and release of toxic substances in the Great Lakes
basin.  The Great Lakes Pollution Prevention Action Plan is discussed in more
detail in the preamble to the proposed Guidance (58 PR 20827-28).

      The final Guidance builds upon these two components of the Great Lakes
program by promoting the development of pollution minimization programs in the
level of detection,  mixing zone elimination, and antidegradation sections of
the final Guidance.   Also,  the decision to provide special provisions for BCCs
implements EPA's commitment to pollution prevention by reducing the discharge
of these pollutants in the future.  This preventive step not only makes good
environmental management sense, but is appropriate based on the documented
adverse effects that the past and present discharge of these pollutants have
produced in the Great Lakes System.

      f.    Provide Accurate Assessment of Costs and Benefits.  In developing
the final Guidance,  EPA identified and carefully evaluated the anticipated
costs and benefits of the major provisions.  EPA received many comments on the
draft cost and benefit studies conducted as part of the proposed Regulatory
Impact Analysis (RIA)  required by Executive Order 12291, and its successor,
Executive Order 12866.  Based upon consideration of those comments and further
analysis, EPA has revised the RIA.  The results of this analysis are
summarized in section IX of this document.

D.    Progress on Other Programs to Protect and Restore the Great Lakes System

      The final Guidance is only one component of many Federal, State and
Tribal programs designed to address environmental contamination in the Great
Lakes System.  Unacceptably high concentrations of bioaccumulative pollutants,

-------
18     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

including PCBs, currently exist in fish tissue and sediments through the
basin.  As discussed further above, these pollutants have the potential to
cause significant adverse human health effects, including increased risk of
cancer and systemic injuries, as well as to adversely impact the aquatic life
and wildlife in this ecosystem.

      EPA recognizes that implementation of the final Guidance alone will not
solve the existing environmental problems from chemical pollutants in the
Great Lakes.  Complete restoration  of the Great Lakes and achievement of
water quality criteria will require significant time, resources, and the
scientific expertise of many parties.  Other regulatory and voluntary programs
are currently underway, however, to prevent or reduce future pollutant
loadings and to remediate the averse effects associated with past pollutant
discharges to the Great Lakes System.  Implementation of the provision in the
final Guidance, in conjunction with the activities described below, will
provide EPA and the"Great Lakes States and Tribes with a comprehensive
framework for reducing toxic loadings.  In addition, these activities will
enable EPA and the Great Lakes States and Tribes to develop cost-effective
strategies to further the goals and requirements of the CWA, Great Lakes Toxic
Reduction Effort (GLTRE)  and GLWQA, and attain the level of water quality
necessary to fully protect human health, wildlife and aquatic life in the
Great Lakes System.

1.    The Great Lakes Toxic Reduction Effort

      EPA and the Great Lakes States and Tribes have established a multi-media
strategy called the GLTRE to achieve further reductions in the use and release
of toxic substances to the Great Lakes System.  The GLTRE emphasizes
addressing nonpoint sources and wet weather point sources of pollution, and is
consistent with the Great Lakes 5-year Strategy discussed in the preamble to
the proposed Guidance  (58 FR 20826).

      The GLTRE has three multi-media tracks.  The Pathway track focuses on
the primary nonpoint source paths and wet weather sources through which BCCs
enter the Great Lakes System.  The Virtual Elimination Project focuses on
identifying the ongoing uses and sources of the BCCs and identifies specific
actions designed to* achieve further reductions of these pollutants.  The third
track, the Lake Michigan Enhanced Monitoring Program, being pursued as part of
the Lake Michigan LaMP, is designed to develop a sound scientific basis to
guide future pollution prevention and reduction efforts to address toxic
pollutants  in the Great Lakes.

      a.    Pathway Track. ' The first track of the GLTRE has identified five
primary wet weather and nonpoint  source pathways for BCCs: air deposition;
Combined Sewer Overflow  (CSO)/stormwater/runoff; contaminated sediments;
storage, handling and transport  (spills); and  leaking waste storage sites.
The GLTRE will address any gaps or barriers  in existing regulatory and non-
regulatory programs designed to prevent and reduce the introduction of BCCs
through these five sources.  The  final product of the GLTRE Pathway Track will
be a menu of actions and recommendations aimed at focusing current and
emerging program authorities on preventing,  controlling and reducing loadings
of BCCs; improving reporting, education and outreach; improvements in
monitoring  and modeling; and risk communication techniques.  Actions and
recommendations that enhance media-specific  or multi-media regulatory gaps and
eliminate barriers to effective regulation,  including use of the ambient water
quality criteria in the final Guidance  in other media regulatory programs,
will also be addressed.  The Pathway track will encourage the prevention of
BCC use, recycling and proper disposal  of BCCs, as well as the  replication of
exemplary or innovative prevention and  reduction programs.  Prevention
measures may also include  recommendations for  bans or sunsets for  certain
BCCs, if appropriate.

      b.    Virtual Elimination Pro-ject.  The  goal of the second component of
the GLTRE is the virtual elimination of bioaccumulative, persistent toxic

-------
                                Section I: Background                             19
substances from the Great Lakes basin.  The project identifies the use and
release of specific toxic substances in the Great Lakes basin and examines the
existing regulatory framework that applies to each substance.  The project, is
designed to provide, recommendations for voluntary and incentive-based changes
to increase the pace and level of reductions of bioaccumulative toxics.  This
project is initially focusing on PCBs and mercury for the purpose of finding
opportunities to achieve virtual elimination of discharges through voluntary
source reductions.

      c.    Lake Michigan Enhanced Monitoring Program.  As discussed in
section I.D.4.b below, this program will develop a sound, scientific base of
information to guide toxic pollutant load reduction efforts at the State and
Federal levels.  The Lake Michigan Enhanced Monitoring Program will help
determine: loadings of contaminants from tributaries, the atmosphere and open
lake sediments; concentrations and fluxes of toxic chemicals in the food web;
and the magnification of toxic chemicals concentrations through representative
food chains.  This work will result in a better understanding of the relative
sources and fate of toxic pollutants in Lake Michigan, and will be of use in
addressing pollutant concerns in all of the Great Lakes.  The results of this
program will also be used to make recommendations on regulatory and
nonregulatory changes needed to fully achieve water quality standards.

      2.    Clean Air Act Amendments

      Implementation of the major provisions of the Clean Air Act Amendments
of 1990 (CAAA) is ah integral part of EPA's broader program to protect and
restore the water quality of the Great Lakes.  Specifically, in response to
information indicating that air pollution contributes significantly to water
pollution, Congress included section 112(m), referred to as the Great Waters
program, in the CAAA.  The purpose of the Great Waters program is to evaluate
the atmospheric deposition of air pollutants to the Great Lakes, Lake
Champlain, Chesapeake Bay, and coastal waters.  As required by this provision,
EPA completed the Great Waters Report in May, 1994.  The Great Waters Report
includes information on the contribution from atmospheric deposition of
pollutant loadings, the environmental or public health effects of such
pollution, the source or sources of such pollution, and a description of
regulatory steps EPA plans to initiate under applicable Federal laws aimed at
the protection of human health and the environment.

      The Great Waters Report observed that water quality conditions in the
Great Lakes, among other waterbodies, are greatly improved compared to a few
decades ago.  Although these improvements are credited to environmental
regulatory programs and public and industrial cleanup efforts that are
addressing primarily waterborne pollution, the Great Waters Report cautions
that the Great Waters ecosystems are far from fully recovered.  The Great
Waters Report adds that it is now necessary to address the more diffuse
sources of pollution, including the air component, in order to attain water
quality goals and to ensure protection of human health and the environment.

      The Great Waters Report concluded that although specific data may be
available for some waterbodies, such as Lake Superior, insufficient data are
available to generalize the atmospheric loadings to all waters.  Overall,
scientists estimate that from 35 to 50 percent of current yearly inputs of a
variety of toxic chemicals to the Great Lakes may be from the air.  Adverse
effects of the chemicals of concern addressed in the Great Waters Report are
evident and studies of selected waters show significant proportions of toxic
pollution coming from the atmosphere.  Although the Great Waters Report noted
that uncertainties in current information are significant, and further
research is needed to better characterize the information for decisionmakers,
adequate information is available to lead EPA to the conclusion that some
actions are justified and necessary at this time.  Because the linkage between
specific sources and subsequent deposition and effects has yet to be
demonstrated,  however, the kinds of action recommended in the Great Waters

-------
20     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

Report focus on the chemicals of concern   (e.g., mercury, PCBs,  dioxins)
rather than on specific sources.

      EPA considered the implications of action and of inaction, while
recognizing that section 112(m) of the CAAA mandates that EPA should act to
"prevent" adverse effects and to "assure protection of human health and the
environment."  EPA's recommendation is that specified reasonable actions are
justified, based on evaluation of the scientific information currently
available, and should now be taken, and that research should continue.

      Most of the initial actions EPA is undertaking focus on utilizing
regulatory mechanisms in the CAAA that are intended to address the most
hazardous chemicals.  The characteristics of toxicity, persistence, and
tendency to bioaccumulate warrant special treatment of the Great Waters
pollutants of concern.  This is consistent with Congressional intent for those
regulatory mechanisms and for section 112(m) of the CAAA.

      The Great Waters Report recommended that for specific source 'categories
of particular concern in the Great Lakes area, CAAA section 112(d) maximum
achievable control technology  ("MACT") standards should be issued ahead of
schedule, where possible.  It also recommended evaluating whether lesser
quantity emission rates, as defined by section 112(a) of the CAAA, should be
set for pollutants of particular concern in the Great Lakes System, and
considering setting area source standards for some source categories.
Following the initial Great Waters Report,  the CAAA requires additional
reports on a biennial basis.  Based on the reports required by section  112(m)
of the CAAA, EPA is required by November 15, 1995, to promulgate further
emission standards or control measures under section 112 of the  CAAA, if
necessary to prevent serious adverse effects, or recommend necessary
regulatory changes under other applicable Federal legislation.

      Between 1992 and 2000, EPA must promulgate technology-based emission
standards for all categories of major sources emitting the 189 toxic air
pollutants listed in section 112(b) of the CAAA.  Area sources will be
regulated in those cases that the Administrator determines are justified in
order to protect health and the environment.  Within eight years of the
promulgation of such standards, EPA will evaluate whether the health and
environmental risk remaining after the application of the control requirements
is significant enough to warrant additional regulatory requirements.  To date,
EPA has promulgated final technology-based standards for Resource Conservation
and Recovery Act (RCRA) treatment, storage and disposal facilities, the
synthetic organic chemical manufacturing industry, coke ovens, industrial
cooling towers, the dry cleaning industry,  halogenated solvent cleaning,
magnetic tape production, commercial ethylene oxide sterilizers, and has
proposed several more standards including those covering epoxy and polymer
resin production, the pulp and paper industry, marine vessels, off-site waste
treatment, petroleum refineries, secondary lead smelters, ship building, wood
furniture manufacturing, municipal waste combusters, chromium electroplating,
and gasoline distribution.    New source performance standards have been
proposed for several industry categories including synthetic organic
chemicals, wastewater and cold-cleaning digesters.  The standard for the
synthetic organic chemical manufacturing industry alone is estimated to reduce
toxic emissions by 510,000 tons a year  (59 FR 19410, April 22, 1994).
Proposed new source performance standards and emissions guidelines have been
published for municipal waste combusters sources greater than 35 MG/day (56 FR
5514), and are scheduled for final promulgation in September, 1995.

      Under section 112 of the CAAA, EPA may add additional substances  to  the
list of hazardous air pollutants, including pollutants of concern in the Great
Lakes, provided the criteria for additions as identified in CAAA section
112(b) are met.  EPA also has discretion to add pollutants when  scientific
information dictates -additions are warranted:  Once a pollutant  is listed, EPA
must set technology-based standards for sources of that pollutant by the year
2000 or within two years of  listing, whichever  is  later.

-------
                                Section I: Background                              21
      Section 112(c)(6) of the CAAA provides for a reduction  in  sulfur  dioxide
 (S02) emissions from utilities of approximately 10 million  tons  by 2010.   The
 first phase of these reductions goes  into effect in  1995.   Not only will  a
 portion of this reduction occur within  the Great Lakes, but many of the
 control technologies likely to be used  to achieve these reductions may  reduce
 toxic air pollutants, such as mercury,  that are specifically  of  concern in the
 Great Lakes System.  In addition, under section 112(n)(1)(B)  of  the CAAA,  EPA
 is conducting-a study of mercury emissions from utilities and various other
 sources, which will include an assessment of the health and environmental
 effects of such emissions.

      EPA will continue ongoing efforts to implement  section  112 of the CAAA
 and other sections of the CAAA and use  the results of the Great  Waters  Report
 in the development of policy and regulations that will reduce emissions of
 Great Waters pollutants of concern.   EPA also recognizes the  need for an
 integrated multimedia approach to the problem of air  deposition  and will
 utilize authorities beyond the CAAA to  reduce human and environmental exposure
 to pollutants of-concern.

 3.    ARARs and the Superfund Program
                   *
      Section 121 of the CERCLA  (or Superfund) establishes  cleanup standards
 for remedial actions.  Under CERCLA,  remedial actions must  attain a degree of
 cleanup that assures protection of human health and the environment.  For  any
 hazardous substances remaining on-site,  the remedy must attain any applicable
 or relevant and appropriate standard, requirement, criteria,  or  limitation
 promulgated under any Federal or State  environmental  law, or  any promulgated
 State standard requirement, criteria  or limitation that is  more  stringent  than
 the Federal regulations.  These requirements are commonly referred to as
 ARARs.  Under CERCLA, a requirement is  applicable if  the environmental  law or
 regulation directly addresses the circumstances at a  site.  If not applicable,
 then a requirement may be relevant and  appropriate if circumstances at  a site
 are sufficiently similar to the problems or situations regulated by the
 requirement that it is well-suited to the site.

      Among the CWA regulations that  may be ARARs for CERCLA  actions are the
 requirements governing discharges of  pollutants to surface  waters,  including
 State, Tribal or Federal water quality  standards.  To be an ARAR,  a standard
 must be promulgated and legally enforceable.  The final Guidance is not, by
 itself, enforceable, and is, therefore,  not an ARAR,  although it could  be
 treated as a nonbinding policy to be  considered in CERCLA actions.   Provisions
 consistent with the,final Guidance will  become enforceable  only  when adopted
 by a State or Tribe as part' of its National Pollutant Discharge  Elimination
 System (NPDES) or water quality standards programs, promulgated  by EPA  in  the
 absence of State or Tribal action, or when included in an NPDES  permit.  When
 such adoption or promulgation occurs, those provisions will be ARARs, as
 appropriate,  for on-site CERCLA discharges.  Off-site CERCLA  discharges must
 comply fully with all applicable Federal, State or Tribal requirements.

 4.    RAPS and LaMPs

      The United States and Canadian  Governments agreed to  develop and
 implement RAPs and LaMPs pursuant to Article VI, Annex 2 of the  GLWQA.  The
 development of RAPs and LaMPs is also required by section 118(c)(3) and (4)  of
 the CWA.   RAPs,  which are designed to address impairments to  any one of 14
 specified beneficial uses, are being  developed for each of  43 designated Areas
 of Concern (AOCs).  LaMPs, which are  designed to address loadings of critical
pollutants that are currently impairing or have the potential to impair the
 beneficial uses of the open waters of the Great Lakes, are  being developed by
 EPA in conjunction with the States, Tribes and Canada on a  phased basis.

      a.     RAPs.   The RAP process consists of three  stages:  problem
 definition and identification of sources and causes of contamination;
 identification of remedial and preventive actions designed  to restore uses;

-------
22     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

and implementation of the RAP and the actual restoration of beneficial  uses.
RAPs are: currently under development in: all 26 U.S., 12 Canadian and five
binational AOCs.  To date, 40 of the 43 AOCs have established stakeholder
groups, coordinating committees, public advisory councils or some other forum
representative of local societal, economic, and environmental concerns.  These
infrastructures help to facilitate public participation in the RAP process, as
well as to coordinate RAP development, implement an ecosystem approach, and
build the institutional capacity to restore beneficial uses.

      Each RAP is tailored to address the specific pollutant problems that are
or may be present at the AOC.  The Grand Calumet River/Indiana Harbor Ship
Canal and nearshore Lake Michigan, for example, is the only AOC in which all
14 beneficial uses are impaired.  Several of these use impairments are  caused
by the presence of millions to cubic meters of contaminated sediments in the
waterway.  RAP efforts to eliminate the causes of these use impairments
include projects designed to clean up individual segments of the waterway and
to prevent their recontamination.  The RAP process has developed and obtained
funds for a Toxic Pollution Prevention Program on the waterway and the  Gary,
Hammond and East Chicago Sanitary Districts have formally adopted the RAP's
Common Policy on Toxic Pollution Prevention.

      Restoration of impaired uses in the Grand Calumet River\Indiana Harbor
Ship Canal has progressed mostly by implementing cleanup, pollution
prevention, and restoration projects outside of the waterway proper.  The RAP
process has involved the Indiana Department of Environmental Management's
(IDEM) pollution prevention staff, local industry and the general public in
implementing a Household Hazardous Waste Collection Project, which began in
April of 1994.  Agricultural clean sweeps to collect and properly dispose of
pesticides were also conducted in 1992 and 1993 in several Northern Indiana
counties.  Local educators have helped IDEM develop an Enviromobile which
stops at area schools to educate school children on ways to prevent pollution
while increasing their environmental consciousness.  All of these projects
directly reduce the causes of several use impairments in the AOC, and some
will help to preserve and protect the AOC's globally endangered habitats and
their outstanding biodiversity which is among the highest in the Great  Lakes.

      The Niagara River AOC is located in Erie and Niagara Counties in  western
New York.  Past municipal and industrial discharges and waste disposal  sites
have been a source of contaminants to the Niagara River.  A long history of
development has also changed the Niagara shoreline along much of the river,
affecting fish and wildlife habitat.  Habitat impairment and survival of
aquatic life in the.AOC have been attributed to PCBs, mirex, chlordane,
dioxin, dibenzofuran, hexachlorocyclohexane, polyaromatic hydrocarbons  (PAHs),
and other pesticides.  Contaminated sediments contain metals and cyanides and
are a source of use impairments  to the AOC.

      The Niagara River RAP represents a comprehensive and focused corrective
action strategy to: remediate contaminated sediments and hazardous waste
sites; continue and enhance monitoring activities; continue point and nonpoint
source control programs; and improve fish and wildlife habitat.  In
particular, EPA and the New York State Department of Environmental
Conservation (NYSDEC) are overseeing remediation at three sites along the
Niagara River that are considered sources of contaminants causing use
impairments in the river.  Other efforts currently being pursued under  this
RAP process include stream water quality monitoring to estimate pollutant
loadings, investigations of inactive hazardous waste sites, developing  a
nonpoint source loading estimation methodology for surface runoff, groundwater
migration and atmospheric deposition in conjunction with Environment Canada
and the Ontario Ministry of the  Environment and Energy  (OMEE), developing a
water quality enhancement and protection policy to include discharge
restriction categories, antidegradation and substance bans; working with the
Buffalo Sewer Authority to continue the model development process for system
sub-basins; and developing a comprehensive  inventory of fish and wildlife
populations and habitat.  Pursuant to the Niagara River RAP, a. large number of

-------
                                Section I: Background                             23
sediment remediations/removals, waste site remediations and enforcement
actions on the U.S. side of the Niagara River have resulted in significant
reductions (i.e., 98 percent of priority pollutants) of loadings to the river.
For further examples of RAP activities, see "Progress in Great Lakes Remedial
Action Plans"  (EPA 905-R-24-020, 1994) available in the docket for the final
Guidance.

      b.    LaMPs.  The LaMP process integrates Federal, State and local
programs that address loadings of critical pollutants, assess whether these
programs ensure attainment of beneficial uses, and recommend media-specific
program enhancements to reduce loadings of critical pollutants to the open
waters of the Great Lakes as necessary to attain beneficial uses.  LaMPs also
address pollutants which might impair waters that currently attain beneficial
uses.

      As part of the Lake Michigan LaMP process, the Lake Michigan States
 (Wisconsin, Illinois, Indiana, Michigan) are evaluating the beneficial use
impairments due to all stressors, including toxics, habitat quantity and
quality, exotic species, and human influences.  Multi-media pollution
prevention projects have recently been completed or are ongoing in greater
Chicago, western Michigan, and Milwaukee.  Sediment assessment and remediation
projects are proceeding in Illinois, Michigan and Indiana.  Agricultural
"clean sweeps" to properly collect and dispose of unused pesticides are
continuing in Indiana, Michigan and Wisconsin, and urban "clean sweeps" are
taking place in Northwest Indiana.  The Lake Michigan Forum (i.e., the public
participation component of the Lake Michigan LaMP), along with the Federal and
State governments, also continues to foster public participation.  The Forum
has finalized an action plan for enhancing public outreach, education, and
participation in the LaMP process.

      Implementation of the Lake Michigan Enhanced Monitoring and Mass Balance
Project is also proceeding in the Lake Michigan basin. This project is a
cooperative effort among EPA, the U.S. Geological Survey, U.S. Fish and
Wildlife Service, National Oceanic and Atmospheric Administration, National
Biological Survey and the four Lake Michigan States, and supports the Lake
Michigan LaMP process, GLTRE, and CAAA requirements by determining the
relative loadings of several LaMP Critical Pollutants from both air and water
sources to the Lake.  Modeling these pollutant inputs, and determining their
fate and transport, will allow the participating agencies to predict Lake
Michigan system responses to various load reduction alternatives as well as to
target limited resources to those pollutant sources posing the greatest risk
to the health of the Lake's ecosystem, and will provide important basinwide
information for use in other Great Lakes efforts.  A revised Lake Michigan
LaMP will be published in 1995.

      On February 15, 1994, EPA published a notice of availability in the
Federal Register of•the proposed Stage 1 Lake Superior LaMP (50 FR 7252).  The
draft Lake Superior LaMP puts forward a framework for zero discharge that will
further the goals of virtual elimination in the Lake Superior Basin.  EPA and
the Lake Superior Basin States have developed a Pollution Prevention Strategy
that focuses on all sources  (i.e., point and nonpoint) of the nine zero
discharge pollutants  (PCBs, dieldrin, chlordane, DDT and metabolites, mercury,
dioxin (2,3,7,8-TCDD), toxaphene, hexachlorobenzene, and octachlorostyrene),
and that identifies pollution prevention opportunities, implementation
methods, and measures of success.  Participating agencies have agreed to
develop a Binational Pollution Prevention Strategy, using the EPA's National
Pollution Prevention Strategy as a template.  A final Lake Superior LaMP will
be complete in 1995*.

      The workplan for the Lake Ontario Lakewide Management Plan for Critical
Pollutants was finalized in November 1993.  Target dates include the
completion of the development of this LaMP in three stages: Stage I is
anticipated to be completed in April 1995; Stage II by December 1995; and
Stage III will be complete by the end of December 1996.  Also, a work group

-------
24	Water Quality Guidance for the Great Lakes System - Supplementary Information Document

has been formed to develop a Lake Ontario LaMP Public Forum.  The Coordination
Committee  (water division managers frofli EPA, NYSDEC, Environment Canada and
the OMEE) and Secretariat  (technical staff from these same agencies) members
are discussing options for the LaMP management structure and incorporating
other agencies into the LaMP process when dealing with issues beyond Critical
Pollutants.

      In the spring of 1993, EPA and Environment Canada formed a temporary
Binational Implementation Committee to develop a concept .paper that would
provide a framework for the development of the Lake Erie LaMP, and to organize
a LaMP Management Committee to oversee LaMP development.  Representatives of
the U.S. and Canadian Federal, State, and Provincial governments met in
September 1994 to formally convene a binational Management Committee to
oversee development of the Lake Erie LaMP.  The Management Committee approved
the Lake Erie LaMP concept paper, which describes the proposed focus, scope,
and management structure of the Lake Erie LaMP.  As outlined in the concept
paper, the Lake Erie LaMP will address a number of environmental issues above
and beyond toxic substances, including habitat destruction, exotics, and
nutrient loadings.  The Management Committee also charged a staff-level work
group to initiate a variety of LaMP activities.

      In addition, EPA expects any new loadings data obtained during the LaMP
process will be incorporated by the States and Tribes when establishing or
revising TMDLs and wasteload allocations  (WLAs) in the Great Lakes System.
These new TMDLs and. WLAs will then be appropriately reflected in subsequent
revisions to NPDES permits.

      5.    Sediments

      Contaminated sediments are a significant source of loadings of toxic
pollutants at harbors and river mouths throughout the Great Lakes System.  To
address this source of toxic pollutant loadings, the 1987 CWA Amendments
authorized a five-year demonstration program to identify and develop
assessment and treatment technologies for contaminated sediments in the Great
Lakes basin (CWA section 118(c) (7)) .  This program, known as the Assessment
and Remediation of Contaminated Sediments  (ARCS) Program, was designed  to
evaluate appropriate assessment and treatment methodologies for the cleanup of
toxic pollutants in Great Lakes contaminated sediments.  The information
developed through the ARCS Program will help support implementation of  RAPs by
providing guidance to effectively address the contaminated sediment problems
in designated AOCs.

      Contaminated sediments have been identified as environmental problems in
42 of the 43 AOCs.  The 1987 CWA Amendments also specified five AOCs for
priority consideration in conducting the demonstration projects.  These AOCs
are Saginaw Bay, Michigan; Sheboygan Harbor, Wisconsin; Grand Calumet River,
Indiana; Ashtabula River, Ohio; and Buffalo River, New York.   (Id.)

      EPA successfully completed the ARCS Program in 1994.  Approximately 40
documents will be published by early 1995 which discuss the findings of the
sediment assessment work, risk assessment studies, mass balance modeling
results, and results of the pilot demonstrations.  Three guidance documents on
risk assessment, sediment assessment, and treatment technologies have been
published.  Technology transfer workshops, which covered such topics as
sediment assessments, mass balance modeling, and determining contaminant
losses from dredging and disposal projects, have been held at a  number  of
locations throughout the Great Lakes basin.

      A key component of the ARCS Program was  the public outreach and
involvement that was maintained throughout the demonstration process.
Information on ARCS Program  activities has been widely distributed  to the
public in the form of ARCS Update fact sheets, news releases, a  slide show,
and public meetings.  The ARCS Program has also made all written documentation

-------
                                Section I: Background                             25
accessible by setting up repositories for ARCS Program materials in local
libraries at the five ARCS priority locations.

      The final summary report on ARCS program results was presented to
Congress in the fall of 1994 and is available in the docket for the final
Guidance.  The conclusions contained in the final summary report discuss the
state-of-the-art methods the ARCS Program demonstrated for the assessment of
contaminated Sediments, especially in the area of toxicity testing, and the
new ground broken by the ARCS Program in the application of the mass balance
modeling approach.  The ARCS Program made significant contributions to the
knowledge base on contaminated sediment remediation by selecting promising
treatment technologies, taking them out into the field, and demonstrating
their effectiveness on site.

      The major findings of the ARCS Program consisted of the need to perform
thorough, integrated sediment assessments; the importance of mass balance
modeling in the evaluation of remediation scenarios; the identification and
demonstration of several feasible sediment treatment technologies; and the
recognition and success of public involvement and active participation in
sediment assessment and remediation projects.  The field assessment,
contaminant fate modeling, risk assessment, and remediation technology
techniques demonstrated in the ARCS Program have improved the knowledge base
and will enable Federal, State and local officials to make scientifically
sound remediation decisions.

      The products of the ARCS Program will not, by themselves, eliminate the
problems posed by contaminated sediments, nor do they propose one "cure all"
treatment technology for their remediation.  They do, however, provide
guidance for the selection of sediment assessment and treatment technologies
as well as recommendations for future full-scale applications.  The results
and products from the ARCS Program will have far-reaching implications for the
remediation of contaminated sediments within the Great Lakes as well as on a
nationwide basis.

      At the time the proposed Guidance was published, EPA was also preparing
to publish for public comment proposed sediment quality criteria for five
organic chemicals.  EPA's proposed sediment quality criteria are intended to
protect benthic organisms from unacceptable effects of chemicals associated
with sediments.  The proposed sediment quality criteria, however, do not
protect against additive, synergistic or antagonistic effects of contaminants,
or bioaccumulative effects to aquatic life or humans.  Implementation of the
final Guidance will complement the proposed sediment quality criteria in these
areas.

      In January 1994, EPA announced in the Federal Register the availability
of the draft sediment quality criteria for three polycyclic hydrocarbons
(acenaphthene, fluoranthene and phenanthrene) and two pesticides (endrin and
dieldrin).  The public comment period for these criteria documents ended in
June 1994, and EPA is now in the process of developing final sediment
criteria.

      At the time of the proposal, EPA was also in the process of developing a
proposed National Contaminated Sediment Management Strategy.  On August 30,
1994, EPA announced in the Federal Register the availability of the proposed
comprehensive, multimedia Strategy.  The proposed Strategy describes specific
actions that EPA will take to reduce environmental and human health risks
associated with contaminated sediments.  The Strategy also defines a plan of
action to address the National problem of contaminated sediments and to
streamline decision-making within and among EPA programs.  EPA accepted
comments on the Strategy through October 31, 1994, and is currently reviewing
the comments.   The proposed Strategy is available in the docket for the final
Guidance.

-------
26     Water Quality Guidance for the Great Lakes System - Supplementary Information Document

      Implementation of EPA's National Strategy has been proceeding in  the
Great Lakes basin, with emphasis on the Great Lakes AOCs.  Programs in  EPA and
the States have been assessing sediment contamination and utilizing a rangg of
regulatory tools to' address the most contaminated sites.  Some examples of
remediation activity are the sediment cleanups that have taken place at
Waukegan Harbor in Illinois, the Black River in Ohio, Cedar Creek near
Milwaukee, Wisconsin, and in East Chicago, Indiana.  Settlements have also
been reached in enforcement actions with responsible parties for sediment
cleanup for sites in Gary and East Chicago, Indiana.  Among the actions being
pursued are those for: Manistique River/Harbor in Manistique, Michigan; the
River Raisin near Monroe, Michigan; Ashtabula River and Fields Brook near
Ashtabula, Ohio; Grand Calumet River in Hammond, Indiana; and Sheboygan
River/Harbor near Sheboygan, Wisconsin.

E.    Science Advisory Board Review

      Information concerning EPA's Science Advisory Board  (SAB) and their
review of the draft Water Quality Guidance for the Great Lakes System prior to
proposal is contained in the preamble to the proposed Guidance  (58 FR at
20826) .  During the process of developing the final Guidance, EPA met with the
SAB from April 27-29, 1994, to discuss the development of a National
methodology for developing wildlife criteria and bioaccumulation factors.  EPA
considered and addressed the SAB's comments pertaining to the Great Lakes
System in preparing the final Guidance.  Some of these comments are discussed
in the applicable sections of this document that follow.  The final SAB report
is also included in the docket for the final Guidance.

F.    References

      Colborn, T.E., Davidson, A., Green S.N., Hodge, R.A., Jackson, C.I., and
Liroff, R.A.  1990.  Great Lakes, Great Legacy?  Washington, D.C.: The
Conservation Foundation; and Ottawa, Ontario: The Institute for Research on
Public policy; 174.

      Colborn, T.I.  1991.  Epidemiology of Great Lakes bald eagles.  J.
Environ. Health Toxicol. 4:395-453.

      Dar, E., Kanarek, M., Anderson, H., and Sonzogni, W.  1992.  Fish
consumption and reproductive outcomes  in Green Bay, Wisconsin.  Environ. Res.
59: 189-201.

      DeVault, D.S., W.A. Willford, R.J. Hesselberg, D.A. Nortrupt, D.A.
Rundberg, A.K. Alwan, and C. Bautista.  1986.  Contaminant trends  in lake
trout  (Salvelinus namaycush) from the upper Great Lakes. Arch. Environ.
Contain. Toxicol. 15:349-356.

      DeVault, D.S., J.M. Clark, G. Lahvis, and J. Weishaar.  1988.
Contaminants and trends  in fall run coho salmon.  J. Great Lakes Res.   14:23-
33.

      DeVault, D.S., USEPA, GLNPO, Personal Communications.

      DeVault, D.S.  1993a.  Data on  contaminant trends  in Lake  Trout
 (Unpublished).  Contains data for 1984-1990.

      DeVault, D.S.  1993b.  Data on  contaminant trends  in fall  run  coho
salmon  (Unpublished).   Contains data  for 1986-1990.

      DeVault, David S., and R. Hesselberg.   1994.   Contaminant  trends  in lake
trout and walleye from  the St. Lawrence Great Lakes.   (Final Draft)

      Environment Canada.   1991.  Toxic Chemicals  in the Great  Lakes and
Associated  Effects", Vols.  I and II.   Environment Canada,  Department  of
Fisheries and Oceans, Health and  Welfare  Canada.

-------
                                Section I: Background                             27
      Falck, F., A. Ricci, M. Wolff, J. Godbold, and P. Deckers.   1992.
Pesticides and polychlorinated biphenyl residues in human breast  lipids  and
their relation to breast cancer.  Arch. Environ. Health 47(2):  143-146.

      Fein, G.G., J.L. Jacobson, S.W. Jacobson, P.M. Schwartz,  and J.K.
Dowler.  1984b.  Prenatal exposure to polychlorinated biphenyls:  Effects on
birth size and gestational age.  J. Pediatr. 105(2): 315-320.

      Flint, R.W. and J. Vena  (eds).  1991.  Human Health Risks from Chemical
Exposure: The Great Lakes Ecosystem.  Chelsea, Michigan:  Lewis Publishers,
Inc.

      Fox, G.A.  1991.  Practical causal inference for epidemiologists.  J.
Toxicol. Env. Health.  33: 359-373.

      Fox, G. A., B. Collins, E. Hayakawa, D.V. Weseloh, J.P. Ludwig, T.J.
Kubiak and T.C Erdman.  1991.  Reproductive outcomes in colonial  fish-eating
birds: A biomarker for developmental toxicants in Great Lakes food chains.
II.  Spatial variation in the occurrence and prevalence of bill defects  in
young double-crested cormorants in the Great Lakes, 197901987.  J. Great Lakes
Res: 17(2): 158-167.

      Fox, G.A.  1992.  In: Chemically-Induced Alterations in Sexual and
Functional Development: The Wildlife/Human Connection  (Colborn, T., and
Clement, C., eds.).  Princeton, NJ: Princeton Scientific Publishing Co.,  Inc.,
1992; 147-158.

      Giesy, J.P., J.P. Ludwig, and D.E. Tillitt.   (In press).  In: Dioxins
and Health.  A. Scheckter (Ed).  Penum: New York.

      Giesy, J.P., J.P. Ludwig, and D.E. Tillitt.  1994.  Deformities in birds
of the Great Lakes region: Assigning causality.  Environ. Sci.  Technol.,
28(3): 128A-135A.

      Gilbertson, M., T. Kubiak, J. Ludwig, and G. Fox.  1991.  Great Lakes
embryo mortality, edema, and deformities syndrome  (GLEMEDS) in  colonial  fish-
eating birds: similarity to chick-edema disease.  J. Toxicol. Environ. Health,
33: 455-520.

      International Joint Commission.  1991.  Cleaning Up the Great Lakes.
Windsor, Ontario: International Joint Commission.

      Korach, K.S., P. Sarver, K. Chae, J.A. McLachlan, and J.D. McKinney.
1988.  Estrogen receptor binding activity of polychlorinated hydroxybiphenyls:
conformationally restricted structural probes.  Mol. Phar. 33:120-126.

      Kubiak, T.J., H.J. Harris, L.M. Smith, T.R. Schwartz, D.L.  Stalling,
J.A. Trick, L. Sileo, D.E. Docherty, and T.C. Erdman.  1989.
Microcontaminants and reproductive impairment of the Forster's  tern on Green
Bay, Lake Michigan-:1983.  Arch. Environ. Contain. Toxicol.  18: 706-727.

      Kurita, H. and J.P. Ludwig.  1988.  Embryonic teratologies and
abnormalities assessed in naturally-incubated eggs and double-crested
cormorants (Phalacrocorax auritus) and Caspian terns (Hydroprogne  caspia) from
Michigan Great Lakes colonies in 1988.  Reports to the Michigan Audubon
Society on the 1986-1988 Findings of the Michigan Colonial Waterbird
Monitoring Project.  Dec. 15, 1988, Ecological Research Services,  Inc.

      Manz, A., J. Berger, J.H. Dwyer, D. Flesch-Janys, S. Nagel,  and H.
Waltsgott.  1991.  Cancer mortality among workers in chemical plant
contaminated with dioxin.  Lancet. 338:959-964.

      McLachlan, J.A., R. Newbold, K.S. Korach, and M. Hogan.   1987.  Risk
assessment considerations for reproductive and developmental toxicity of

-------
28     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

oestrogenic xenobiotics.  In: Human Risk Assessment: The Roles of Animal
Selection and Extrapolation  (Roloff, M.V. and A.W. Wilson, eds.).  London:
Taylor and Francis, Ltd., 1987: 187-193.

      Mitchell, C.A. and T.W. Custer.  1986.  Hatching success of Caspian
terns nesting in the lower Laguna Madre, Texas, USA.  Colonial Waterbirds.   9:
86-89..

      Mora, M.A. , H.J. Auman, J.P. Ludwig, J.P. Giesy, D.A- Verbrugge,  and
M.E. Ludwig.  1993.  Polychlorinated biphenyls and chlorinated insecticides  in
plasma of Caspian terns: relationships with age, productivity, and colony site
tenacity in the Great Lakes.  Arch. Environ. Contam. Toxicol.  24: 320-331.

      Nature Conservancy Great Lakes Program.  1994.  The Conservation  of
Biological Diversity in the Great Lakes Ecosystem: Issues and Opportunities.
Chicago, Illinois.

      Soikkeli, M. -1973.  Breeding success of the Caspian tern  in Finland.
Bird-Banding.  44:196-204.

      Stevens, R.J., and M.A. Neilson.  1989.  Inter- and intralake
distributions of trace organic contaminants in surfacewaters of  the Great
Lakes.  J. Great Lakes Res. 15(3): 377-393.

      Swain, W.R.  1991.  Effects of organchlorine chemicals on  the
reproductive outcomes of humans who consumed contaminated Great  lakes fish:  an
epidemiologic consideration.  J. Toxicol. Environ. Health 33: 587-639.

      Tanabe, S.  1988.  PCB problems in the future: foresight from current
knowledge.  Environmental Pollution  (50) 5-28.

      Tillitt, D.E., G.T. Ankley, J.P. Giesy, J.P. Ludwig, H. Kurita-Matsuba,
D.V. Weseloh, P.S. Ross, C.A. Bishop, L. Sileo, K.L. Stromborg,  J. Larson, and
T. Kubiak.  1992.  Polychlorinated biphenyl residues and egg mortality  in
double-crested cormorants from the Great Lakes.  Env. Toxicol. Chem. 11: 1281-
1288.

      U.S. EPA.  1994.  Assessment and Remediation of Contaminated Sediments
(ARCS) Program, EPA. 905-S-94-001.

      U.S. EPA.  1994.  Deposition of Air Pollutants to the Great Waters:
First Report to Congress, EPA 453-R-93-055.

      U.S. EPA.  1994.  EPA's Contaminated Sediment Management Strategy, EPA
823-R-94-001.

      U.S. EPA.  1994.  Progress  in Great Lakes Remedial Action  Plans,  EPA
905-R-24-020.

      Yamashita, N.', S. Tanabe, J.P. Ludwig, H. Kurita, M.E. Ludwig, and R.
Tatsukawa.  1993.  Embryonic abnormalities and organochlorine contamination  in
double-crested cormorants  (Phalacrocorax Auritus) and Caspian terns
(Hydroprogne caspia) from the upper Great Lakes in 1988.  Environ. Poll.
79:163-173.

-------
                          Section II: Regulatory Requirements                        29
          H.   REGULATORY REQUIREMENTS


A.    Scope and Purpose

      Section 118(c)(2) of the Clean Water Act  ("CWA")  (Pub. L. 92-500 as
amended by the Great Lakes Critical Programs Act  of  1990, Pub. L. 101-596)
requires EPA to publish final water quality guidance on minimum water quality
standards, antidegradation policies,  and implementation procedures for the
Great Lakes System.  Part 132, including appendixes  A through F, constitutes
the Water Quality Guidance for the Great Lakes  System required by section
118 (c) (2).

      The Guidance fulfills the requirements of section 118(c)(2) and is
generally organized as follows:

          Water quality standards; The Guidance contains numerical water
quality criteria for 29 pollutants, listed in Tables 1 through 4 of part 132.
The Guidance also contains methodologies for the  development of water quality
criteria and water quality values for the  protection of aquatic life, human
health,  and wildlife,  and a methodology for development of bioaccumulation
factors, in appendixes A through D.  Together,  these criteria and
methodologies specify minimum numerical limits  on pollutants in ambient Great
Lakes waters to protect human health,  aquatic life,  and wildlife.

          Antidearadation policy; The minimum antidegradation policy,
including an antidegradation standard,  implementation procedures,
demonstration provisions, and decision provisions are contained in appendix E.

          Implementation procedures:  Minimum implementation procedures are
contained in appendix F to part 132.

B.    Definitions

      1.    Proposal.   Section 132.2  of the proposed Guidance contained
definitions of 71 terms that would apply to part  132.  EPA proposed the
definitions as a partial list of terms which need to be defined for consistent
interpretation of the Guidance.  Proposed § 132.4(a) provided that Great Lakes
States and Tribes were to adopt requirements applicable to waters of the Great
Lakes System for the purposes of sections  118,  303,  and 402 of the CWA that
are consistent with these definitions,  or  be subject to EPA promulgation under
§ 132.5.  Other definitions, bearing on the individual portions of the
Guidance,  were contained in proposed appendixes A through F.  Generally, where
terms were applied in the proposed Guidance in  the same manner as in previous
National regulations,  such as those in 40  CFR 122.2, 130.2, and 131.3, the
proposal did not provide a duplicate  definition.  In some cases, however, the
proposal provided a duplicate definition to assist the reader.

      2.    Comments.   Several comments recommended  that EPA resolve
differences between definitions proposed in § 132.2  and those proposed in
various  appendixes to part 132.

      A number of comments were received on definitions that are discussed
elsewhere in this document.  These include: the definition of "bioaccumulative
chemical of concern,11  discussed further in section II.C.8 of this document;
and the  definition of "existing discharger," discussed further in section
VTII.I of this document.

      One comment recommended deleting the definition of "reasonable
potential."  EPA agrees.  Since the definition  of "reasonable potential" is
not necessary to implement the Guidance,  EPA has  deleted the definition.

-------
30     Water Quality Guidance for the Great Lakes System - Supplementary Information Document

      Technical and editorial comments were also received concerning  a  number
of definitions.

      EPA agrees that definitions should be consistent throughout  the
Guidance.  In some cases, as discussed below under the Final Guidance section,
definitions have been modified or deleted from § 132.2 to eliminate such
inconsistencies.

      EPA agrees with the majority of technical and editorial comments
received, and has made changes accordingly.  Some of these changes are
discussed further under the Final Guidance section below.  EPA's responses  to
individual technical and editorial comments on definitions in §  132.2 are
included in the docket for this rulemaking.

      3.    Final Guidance.  Section 132.2 of the final Guidance includes 56
definitions, reflecting deletion of 17 proposed definitions, and addition of
two new definitions.  Thirty six definitions were clarified or modified, and
18 were unchanged.

      The following definitions have been deleted from § 132.2 because  they
are defined adequately in one or more of the appendixes: "acceptable  daily
exposure (ADE)," "acute toxic unit (TU.) , "  "biomagnification,"  "chronic toxic
unit  (TU0) ,"  "depuration,"  "dilution  fraction,"  "linear multi-stage model,"
11 octanol/water partition coefficient  (KQW) , " "relative source contribution
(RSC)," and "slope factor."

      The definition of "allowable dilution flow  (Q^)" has been deleted from
§ 132.2.  In response to comments, changes have been made to the final
procedure 3* of appendix F that made this definition unnecessary.   This  change
is discussed further in section VIII.C of this document.

      The definition of "compliance evaluation level  (CEL)" has been  deleted
because the term is no longer used.  This change is discussed further in
section VIII.H of this document.

      The definition of "existing uses" has been deleted to because it
duplicates the definition in 40 CFR 131.3.

      EPA deleted the proposed definition of "noncarcinogen" because  it should
be clear that a noncarcinogen is any substance which is not a carcinogen.
Furthermore, EPA found that the proposed definition could have been confusing
when  compared to the definition of "carcinogen."

      The definition of "reasonable potential" has been deleted  from  §  132.2
as discussed in Comments above.

      The definitions of "steady-state BAF/BCF" and "superlipophilic
chemicals" were deleted, since revisions to the methodology for  development of
BAFs made the terms unnecessary.  These changes are discussed further in
section IV of this document.

      The definition of "biota-sediment accumulation factor  (BSAF)" has been
added, and is discussed in section IV of this document.

      The definition of "new Great Lakes discharger" has been added,  and the
definition of "existing discharger" has been modified and redefined as
"existing Great Lakes discharger."  Together, the two definitions  categorize
all discharges to waters of the Great Lakes System depending on  whether the
construction of buildings, structures, facilities, or  installations generating
the discharge commenced before  ("existing") or after  ("new") March 23,  1997.
Under procedure 3 of appendix F, mixing zones for BCCs are not available to
new Great Lakes dischargers after March 23, 1997; for  existing Great  Lakes
dischargers, mixing zones  for BCCs are generally not available after  March  23,
2007  except under specified circumstances.  These provisions are discussed

-------
                           Section II: Regulatory Requirements                        31
more fully in sectipn VTII.C of this document.  Under procedure 9 of
appendix F, NPDES permits may allow compliance schedules under certain
circumstances to existing Great Lakes dischargers but not to new Great Lakes
dischargers.  These provisions are discussed more fully in section VIII.J of
this document.

      .The definition of "bioaccumulative chemical of concern  (BCC)" has been
modified, and is discussed in section II.C of this document.

      The definition of "carcinogen" has been clarified to add a reference to
a discussion of the classification of carcinogens in section II.A of appendix
C to the final Guidance.

      The definition of "chronic toxicity" has been clarified to add reference
to "delayed adverse effects" to reflect EPA's concern that pollutants such as
2,3,7,8-tetrachlorodibenzo-p-dioxin (2378-TCDD) have been observed to cause
delayed adverse effects days, weeks, or even months after exposure.  See, for
example, "Interim Report on Data and Methods for Assessment of 2378-TCDD Risks
to Aquatic Life and Associated Wildlife," EPA, March 1993  (EPA 600/R-93/055),
which is available in the docket for this rulemaking.

      The definition of "Great Lakes States and Great Lakes Tribes" has been
modified to clarify that an Indian Tribe in the Great Lakes basin would be a
"Great Lakes Tribe" subject to the provisions of the final Guidance only if
EPA has approved water quality standards for the Tribe, or if EPA has
authorized the Tribe to administer an NPDES program.  This clarification was
necessary because the Guidance provisions for water quality standards would be
meaningless unless at least a basic set of standards are in place, including
designated uses, narrative criteria, and legal authority references.
Similarly, the Guidance provisions for NPDES permits would be meaningless
unless the Tribe has been authorized to administer an NPDES program.
Provisions applicable to Tribes are discussed further in section II.D.3 of
this document.

      The definition of "Tier II value" has been modified, and is discussed in
sections II.C.2 and VT of this document.

      The definition of "wet weather point source" has been modified to
incorporate the existing definition of point source contained in § 122.2, for
the convenience of the reader.  Furthermore, in response to comments and to
clarify possible ambiguities in the proposal, EPA has modified the definition
in the final Guidance to clarify that the definition includes only certain
types of discharges, to specify the types of excluded discharges more
explicitly, and to delete an unnecessary definition of combined sewer
overflow.  The exclusion for wet weather point sources is discussed in section
II.C.7.

      The following definitions were modified to make technical clarifications
and corrections: "acute toxicity," "acute-chronic ratio (ACR),"
"bioaccumulation," "bioaccumulation factor  (BAF)," "bioconcentration,"
"bioconcentration factor (BCF)," "chronic toxicity," "criterion continuous
concentration (CCC)," "criterion maximum concentration (CMC)," "EC50,"
"endangered or threatened species," "final acute value (FAV)," "final chronic
value (FCV)," "final plant value (FPV)," "Great Lakes System," "human cancer
value (HCV)," "human noncancer value  (HNV)," "LC50," "minimum level  (ML)," "no
observed adverse effect level (NOAEL)," "no observed effect concentration
(NOEC)," "quantification level," "quantitative structure activity relationship
(QSAR)," "risk associated dose (RAD)," "species mean acute value  (SMAV),"
"species mean chronic value  (SMCV)," "stream design flow," "uncertainty factor
(UF)," "uptake," and "wasteload allocation  (WLA)."

      The following- definitions are the same as proposed,  or the same with
editorial corrections: "adverse effect," "connecting channels of the Great
Lakes," "detection level," "Federal Indian Reservation," "genus mean acute

-------
32     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

value (GMAV)," "genus mean chronic value  (GMCV)," "Great Lakes," "human cancer
criterion (HCC)," "human noncancer criterion (HNC)," "Indian Tribe," "load
allocation  (LA)," "loading capacity," "lowest observed adverse effect level
(LOAEL)," "open waters of the Great Lakes," "threshold effect," "Tier I
criteria,"  "total maximum daily load (TMDL)," and "tributaries of the Great
Lakes System."

C.    Adoption-and Application of Criteria. Methodologies. Policies, and
      Procedures

1.    Adoption of Tier I Criteria and Methodologies

      a.     Proposal.  The proposed Guidance included a two-tiered approach,
consisting of methodologies to develop water quality criteria  (Tier I), and
methodologies to calculate water quality values (Tier II) with fewer data than
the full minimum data required for a Tier I criterion calculation.  The
purpose of Tier II methodologies is to provide a uniform approach for
evaluating and controlling pollutants when  there are insufficient data to
develop Tier I criteria.  In the absence of State- or Tribal-adopted Tier I
criteria for a pollutant, the Tier I and Tier II methodologies provide
mechanisms with which to interpret and ensure that the States' and Tribes'
narrative criteria prohibiting the discharge of toxic pollutants in toxic
amounts are reflected in water quality-based effluent limits.  The preamble to
the proposed Guidance described the two-tiered approach in more detail (58 FR
20835; April 16,  1993).

      Proposed § 132.4(a) provided that Great Lakes States and Tribes were to
adopt Tier  I methodologies for developing numeric water quality criteria to
protect aquatic life, human health, and wildlife consistent with those
specified in appendixes A through D, or be  subject to EPA promulgation under
§ 132.5.

      Upon State or Tribal adoption or EPA promulgation of the Guidance,
proposed §  132.4(b) specified that Great Lakes States or Tribes would have
used the Tier I methodologies in appendixes A, C, and D and the methodology in
appendix B when adopting or revising numeric water quality criteria.  In
addition, if a Great Lakes State or Tribe had not adopted a numeric water
quality criterion for a pollutant in its water quality standards, but enough
data existed to meet Tier I minimum data requirements, proposed § 132.4(c)
provided for use of the Tier I methodologies for any development of numerical
criteria to implement narrative criteria.   Such implementation would include
development of water quality-based effluent limits, where appropriate.

      The proposed Guidance also included numeric criteria in Tables 1 through
4 of part 132 that were derived using the Tier I methodologies.  Proposed
§ 132.3 would have required Great Lakes States and Tribes to adopt these
specific numeric criteria or more stringent criteria into their water quality
standards for the Great Lakes System or be  subject to EPA promulgation under
§ 132.5.

      b.    Comments.  EPA received a number of comments on the requirements
for adopting additional Tier I criteria beyond those listed in Tables 1
through 4 of part 132, and for revising the criteria in  the Tables.  Many
commenters  supported the proposed approach  because it would ensure  increased
consistency among the water quality standards of Great Lakes States and
Tribes.  Other commenters urged EPA to provide more flexibility to  States  and
Tribes in adopting criteria consistent with the Guidance, including the
flexibility to modify the criteria in Tables 1 through 4 to reflect new
scientific  findings and data.  Many commenters argued that the criteria
methodologies and the criteria in Tables  1  through 4 should be provided  as
non-binding guidance.

      A large number of comments  also suggested that EPA should be  responsible
for developing any new  or revised Tier I  criteria as additional data  become

-------
                           Section II: Regulatory Requirements                        33
available and are evaluated.  These commenters believe that it is more
efficient for EPA to develop criteria than for States or dischargers who may
not have the expertise or resources to develop criteria independently.  Some
of these commenters also suggested that EPA should ensure full public review
of such new or revised criteria.  Some commenters also urged EPA to amend
Tables 1 through 4 in future EPA rulemakings.  Additionally, a few commenters
were concerned that the Guidance as proposed may not provide sufficient
protection to endangered or threatened species from pollutants that may be
found to have particularly potent effects.
      EPA agrees that it is important for States and Tribes to have the
flexibility to modify criteria and values, including the criteria in Tables 1
through 4, in appropriate circumstances when new scientific findings and data
become available.  EPA has determined that the proposed minimum requirements,
together with changes to the procedure for site-specific criteria
modifications in the final Guidance, as well as EPA's planned approach to
assist States and Tribes in implementing the Guidance, provide adequate
flexibility for incorporating new information without sacrificing the improved
consistency in Great Lakes water quality standards envisioned in the Great
Lakes Critical Programs Act ("CPA").

      Site-specific criteria modifications provide a degree of flexibility to
incorporate new scientific findings and data as they may affect a specific
waterbody.  EPA has retained the provisions allowing site-specific
modifications to criteria developed under the Guidance, including criteria in
Tables 1 through 4, and has expanded the flexibility of the provisions to
allow less restrictive site-specific modifications for aquatic life, human
health and wildlife criteria under certain conditions using the final
procedure 1 of appendix F.  These provisions are discussed further in section
VTll.A of this document.

      Additionally,'there are several steps within the criteria methodologies
where flexibility is available to reflect new findings and data.  For example,
although EPA states a preference for using the linearized multistage model
when deriving human health criteria, the final Guidance allows the use of
different models if the data support their use.  In addition, the final
Guidance recognizes that the EPA methodology for conducting cancer risk
assessments is currently under review and that any changes adopted by EPA can
be incorporated by the States and Tribes.  Also, the section VIII.A of
appendix B provides that BAFs derived in accordance with the BAF methodology
in the final Guidance should be modified if changes are justified by available
data.

      EPA's planned approach for implementing the final Guidance will further
facilitate flexibility to incorporate new scientific findings and data.  EPA
intends to expand the future assistance it provides to Great Lakes States and
Tribes over that envisioned at the time of the proposal.  EPA's planned
implementation is as follows:

          EPA Region 5, in cooperation with EPA Regions 2 and 3 and
Headquarters offices,  and the Great Lakes States and Tribes, will establish a
Great Lakes Initiative (GLI)  Clearinghouse to assist States and Tribes in
developing numeric Tier I water quality criteria and Tier II water quality
values.  Further information about the GLI Clearinghouse is available from the
address listed at the beginning of this document.  As additional toxicological
data and exposure data become available or additional Tier I numeric criteria
and Tier II values are calculated by EPA, States, or Tribes, Region 5 will
ensure that this information is disseminated to the Great Lakes States and
Tribes.

            EPA Region 5 will work with the States and Tribes, EPA Regions 2
and 3,  EPA Headquarters offices, and EPA research laboratories to review new
toxicological and exposure data.  The review will include consideration of
data quality and appropriateness for use with the Guidance methodologies.  For
pollutants of especially high interest and/or concern, Region 5 and the other

-------
34	Water Quality Guidance for the Great Lakes System - Supplementary Information Document

EPA offices identified above intend from time-to-time to use the Clearinghouse
information to develop GLI criteria guidance documents similar to those
supporting the proposed and final Guidance.  EPA will then publish a notice in
the Federal Register announcing the availability of such documents and
inviting public comment on them.  After reviewing the comments, EPA will
finalize the GLI criteria guidance documents and make them available as
guidance to Great Lakes States and Tribes.  The GLI criteria guidance
documents would represent EPA's best current information about effects  of the
pollutants in the Great Lakes System.  The GLI criteria guidance documents
could address either or both Tier I criteria and Tier II values.

            Regions 2, 3 and 5, through their review and approval of State
water quality standards under section 303 of the CWA, will work with the
States to ensure that all Great Lakes States and Tribes maintain a minimum
consistent level of protection for aquatic life, human health, and wildlife
throughout the Great Lakes System consistent with the methodologies in  the
final Guidance.  EPA will also make a special effort to encourage Great Lakes
States and Great Lakes Tribes to adopt criteria based on any newly calculated
GLI criteria guidance documents in the next triennial review of water quality
standards under section 303.  If such efforts are not successful for all
States, EPA could evaluate use of section 303(c)(4), if appropriate, to
determine that a particular State or Tribe needed new or revised water  quality
standards reflecting the new criteria,  and promulgate Federal criteria  if
necessary for that State or Tribe to ensure minimum consistent criteria are
present in all States and Tribes.

      EPA believes the above plans will provide the flexibility to incorporate
new data into water quality criteria development.  For example, if new  data
become available that would result in significant changes to criteria in
Tables 1 through 4, EPA may use the above process to develop one or more
revised GLI criteria guidance documents.  EPA would then work with the  States
and Tribes in their adoption of the revised criteria.  If the revised criteria
are more stringent than the corresponding criteria in Tables 1 through  4 of
part 132, States and Tribes would be able to adopt them without further EPA
rulemaking.  If the'revised criteria are less stringent than the corresponding
criteria in Tables 1 through 4 of part 132, EPA would consider initiating a
rulemaking action to delete or revise criteria in the Tables if necessary to
allow or facilitate State and Tribal adoption of the less stringent criteria.

      It is also possible that the new information could be so substantial
that it makes some criteria in Tables 1 through 4 scientifically indefensible.
In this situation, States or Tribes could utilize the provisions in § 132.4(h)
to adopt new criteria even if they were less stringent than criteria in Tables
1 through 4 without further EPA rulemaking.  EPA expects that this situation
would occur rarely,.if at all.

      The above plans will also result in an efficient use of resources to
develop additional criteria and values where they are most needed.
Furthermore, this approach provides more flexibility than the alternative of
EPA conducting a rulemaking every time any additional data could support a
modification, to a criterion in Tables 1 through 4.  Finally, the final
Guidance retains the existing State and Tribal responsibility to adopt  water
quality standards as necessary under the CWA without waiting for EPA to
develop detailed criteria guidance.  At the same time, however, the
involvement of EPA in facilitating the review of data, developing GLI criteria
guidance documents as appropriate, in conjunction with continued oversight of
the section 303 water quality standards programs will contribute to more
consistent criteria in the future throughout the Great Lakes System.

      EPA considered but rejected the suggestion of some commenters that the
criteria methodologies and the numeric criteria in Tables 1 through 4 of part
132 should be provided as guidance but not  specified as minimum requirements
for State and Tribal adoption.  First, EPA  does not believe that the suggested
approach would satisfy the requirements of  section  118(c)(2).  That section

-------
                           Section II: Regulatory Requirements                        35
not only directs EP£ to provide guidance to the Great Lakes States on
"minimum" water quality standards, antidegradation policies, and
implementation procedures for the Great Lakes System, but also requires the
States to adopt provisions consistent with the guidance or be subject to EPA
promulgation.  EPA believes that whether States and Tribes adopt standards,
policies, and procedures consistent with the final Guidance, or whether EPA
promulgates them, Congress intended that the final Guidance would establish
minimum requirements for the Great Lakes System.  This issue is discussed
further in section II.D.2 of this document.

      Second, the suggested approach could result in significant variations
between State and Tribal programs submitted under section 118(c)(2)(C).  As
discussed in the preamble to the proposed Guidance, the establishment of more
uniform control of water pollution in the Great Lakes System was one of the
most important goals of the CPA, and the 1986 Great Lakes Toxic Substances
Control Agreement signed by the eight Great Lakes Governors (58 PR 20820-23,
and 20838-39).  Progress toward this goal will not be achieved unless the
Guidance specifies the "minimum" requirement for the Great Lakes System.

      Similarly, EPA considered but rejected an alternative approach that
would retain the provisions for States and Tribes to adopt criteria consistent
with the criteria in Tables 1 through 4, but would provide the methodologies
in appendixes A through D as guidance but not specified as minimum
requirements for State and Tribal adoption.  EPA believes the criteria in
Tables 1 through 4 are necessary to protect human health, wildlife, and
aquatic life in the Great Lakes System, but is concerned that these criteria
alone would not be sufficient.  The pollutants in Tables 1 through 4 are good
examples of pollutants in the Great Lakes System, but they are clearly not the
only criteria necessary to protect human health, wildlife and aquatic life in
the Great Lakes System.  For example, in sections I and II of this document
EPA identifies BCCs as a category of pollutants of high concern in the System,
yet because of the limited time and resources available to develop the
proposed and final Guidance, Tables 1 through 4 do not contain criteria for
all 22 currently-identified BCCs.  Also, in some cases a pollutant that
appears in only one of the four Tables would be likely to have a lower
criterion in another of the Tables if the other criteria were developed.  For
these reasons, EPA believes that the final Guidance must also continue to
provide for State and Tribal adoption of criteria methodologies consistent
with appendixes A though D in order to increase the consistency between water
quality criteria and permit limits throughout the Great Lakes System and to
satisfy EPA's obligation to specify minimum requirements necessary to protect
human health, wildlife, and aquatic life.

      EPA also considered but rejected an alternative approach under which
States and Tribes could use the Guidance's criteria methodologies to adopt and
submit for EPA review criteria less stringent than those in Tables 1 through 4
solely on the basis of new data or information.  Such an approach would have
two major problems.  First, it would be difficult to develop an operational
definition for "new" data that would define all factors necessary for
concluding that data and information were sufficiently new or significant to
justify EPA approval of a submitted criterion less stringent than in Tables 1
through 4.  States, Tribes and EPA would also face an additional
administrative burden of determining whether particular data sets met the
definition.

      Second, although this approach would enable new data to be incorporated
rapidly,  EPA is concerned that it could result in significant departures from
the increased uniformity supported by EPA and the States and envisioned in the
CPA.  For example, one State could use a new toxicological study to develop
and adopt less stringent criteria before EPA and other experts participating
in the GLI Clearinghouse described above could review the quality and
appropriateness of the study.  EPA believes such peer review is desirable for
all pollutants, but especially for pollutants such as those in Tables 1
through 4 that have benefitted from extensive EPA, State, and public review,

-------
36     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

and include pollutants of widespread concern and interest in the Great Lakes
System.  These numeric criteria and methodologies were initially developed by
technical experts from EPA, the Great Lakes States, the U.S. Fish and Wildlife
Service, and other participants in the Great Lakes Water Quality Initiative,
who are familiar with the data form and the unique characteristics of the
Great Lakes System.  During the development of the Guidance, this cooperative
effort to review and process toxicological data also resulted in a high degree
of consensus for the numeric criteria in Tables 1 through 4.  If EPA and the
other participants had unlimited time and resources, it would be desirable to
utilize the same process for all pollutants requiring numeric criteria.

      EPA believes that the implementation plans discussed above provide a
reasonable and systematic way to incorporate the effects of new data into
water quality criteria development without the disadvantages of the
alternative approach.  Additionally, as discussed further below, the Guidance
provides adequate flexibility to the States and Tribes to modify the criteria
in Tables 1 through 4.  This flexibility is provided through § 132.4 (i), which
preserves State and Tribal authority to adopt more stringent provisions;
through § 132.4(h) which provides for more or less stringent modifications
based on scientific indefensibility and is discussed in section II.C.6 of this
document; and through § 132.3 which provides for more or less stringent site-
specific modifications of the criteria in the circumstances specified in
procedure 1 of appendix F.

      EPA also does not believe that it is necessary or appropriate for EPA to
be solely responsible for developing new or revised Tier I criteria, or to
make State or Tribal adoption of new or revised criteria contingent on EPA
amendment of Tables 1 through 4 of part 132 in future rulemakings.  The CWA
has always placed the primary responsibility for developing and adopting
criteria on the States and Tribes approved to administer water quality
standards programs.  If the final Guidance were to alter this relationship and
make EPA responsible in the future for developing and/or conducting rulemaking
for all new or revised criteria. State and Tribal adoption of criteria that
might otherwise be necessary to protect aquatic life, human health, or
wildlife and for which data were available could be significantly delayed.
EPA believes that the provision for States to adopt the methodologies found in
appendixes A through D, together with the information exchange and peer review
that will occur through the operation of the GLI Clearinghouse, will provide
the States and Tribes with an adequate framework for ensuring consistent and
timely interpretation of narrative standards.  At the same time, EPA
recognizes that there are some situations, especially with toxic pollutants of
high concern and interest, where additional EPA involvement in criteria
development is desirable.  'Furthermore, EPA's planned involvement in
coordinating and disseminating information will maximize efficient use of
limited State, Tribal, and Federal resources.  For this reason, EPA is
prepared to participate actively in operating the GLI Clearinghouse described
above, and is committed to working with States and Tribes to develop, review,
analyze, and disseminate data, and to develop GLI criteria guidance documents
as necessary in accordance with available resources.

      EPA also agrees with comments that EPA could help facilitate public
review of the adoption of new or revised criteria, and implementation of the
criteria in NPDES permits.  Dnder 40 CFR 131.20, States and Tribes are to
provide an opportunity for public participation in adoption or revision of
water quality standards.  Under 40 CFR 123,25, States implementing the NPDES
program are to provide an opportunity for public participation during
development of discharge permits.  EPA believes the GLI Clearinghouse will
assist States and Tribes in  implementing these responsibilities by providing
an additional opportunity for the public to review data that may be used in
the development of future water quality standards and NPDES permit limits.
Furthermore, EPA intends to  provide an opportunity for public review and
comment before finalizing any future GLI criteria guidance  documents.

-------
                           Section II: Regulatory Requirements                        37
      Finally, EPA agrees with commenters that it is important to incorporate
into regulatory programs any new toxicblogical data which shows that a
threatened or endangered species--or a surrogate for such species--is
particularly sensitive to a pollutant.  EPA believes the implementation plans
described above will ensure that endangered or threatened species are
adequately protected.  For example, when relevant new data become available,
the data will be placed in the GLI Clearinghouse where they will be available
to States, Tribes, and other interested persons.  EPA will place special
emphasis in the Clearinghouse on data relevant to protection of endangered or
threatened species, and will alert States and Tribes to data that indicate
unusual sensitivity of these species or their surrogates in the Great Lakes
basin.  EPA and the U.S. Fish and Wildlife Service will also work with States
and Tribes to identify sites where special protection may be recommended.
States and Tribes must use such data wherever appropriate in their water
programs, and EPA will evaluate these actions during the triennial reviews of
water quality standards under section 303(c) of the CWA.  Before approving
State and Tribal water quality standards, EPA will ensure that they are not
likely to jeopardize the continued existence of any listed endangered or
threatened species, or result in the destruction or adverse modification of
such species' critical habitat.  EPA could not approve State or Tribal
adoption of a criterion that does not adequately protect endangered or
threatened species.  EPA's implementation of the Endangered Species Act is
discussed further in section II.G of this document.

      c.    Final Guidance.  For the reasons above, the proposed minimum
requirements regarding Tier I criteria and methodologies generally have been
retained in the final Guidance.

      In response to issues and comments discussed elsewhere in this document,
the minimum required application of the Tier I methodology for development of
wildlife criteria has been limited to BCCs.  This change is reflected in §
132.4(a)(5), § 132.4(d)(4), and appendix D to part 132, and is discussed
further in section VI of this document.  The final Guidance retains the
proposed provisions to apply the Tier I methodologies for human health and
aquatic life to all pollutants except those in Table 5 of part 132.

2.    Adoption and Application of Tier II Methodologies

      a.    Proposal.  The proposed Guidance included a mechanism to interpret
and ensure that existing narrative prohibitions against the discharge of toxic
substances in toxic amounts are reflected in water quality-based effluent
limitations in a consistent way throughout the Great Lakes System, even where
data are limited.  If a State or Tribe has not adopted a numeric water quality
criterion for a pollutant and insufficient data exist to meet Tier I minimum
data requirements, proposed § 132.4(c) would provide for application of Tier
II methodologies to develop Tier II values to implement the narrative
criteria.   Additionally, if sufficient data to calculate a Tier II value for a
pollutant on Table 6 of part 132 do not exist, procedure 5 of the proposed
implementation procedures (appendix F to part 132) provided that the
permitting authority,  under specified circumstances, would generate or require
the permittee to generate the data necessary to derive Tier II values.

      The above approach is consistent with existing EPA national regulations
at 40 CFR 122.44(d)(vi) which require permit authorities to establish effluent
limits for individual pollutants or indicator parameters under specified
circumstances when the pollutant is present in the effluent at a concentration
that causes, has the reasonable potential to cause, or contributes to an
excursion above any water quality standard, including numeric and narrative
criteria for water quality.  This approach is discussed further in section
VIII.E of this document.

      b.    Comments/  EPA received many comments objecting to the inclusion
of Tier II methodologies in the Guidance.  Among the concerns raised were that
the Tier II methodologies are not scientifically valid for use in determining

-------
38     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

water quality-based effluent limits, that it is inappropriate to encourage
dischargers to generate data to be used in development of Tier II values, that
such data generation is too costly and time-consuming, and that the Tier II
methodologies are overly restrictive.  A large number of comments singled out
the Tier II methodology for aquatic life as a particular concern because it
allows use of fewer data than the established national aquatic criteria
methodology, because values are adjusted to be more restrictive when fewer
data points are available, and because of possible overlap with the
implementation procedure for whole effluent toxicity.  Many comments singled
out the Tier II methodology for wildlife because of concerns about the
appropriateness of the scientific approach.  Some commenters suggested  the use
of biological tests such as whole effluent toxicity tests and bioconcentration
testing in place of' the Tier II methodologies for numerical pollutant-specific
values.  Some commenters suggested that anti-backsliding and antidegradation
requirements would prevent future adjustments in Tier II numbers when more
data become available.  Some commenters suggested that EPA should have  the
responsibility to generate Tier II values.

      EPA also received a number of comments supporting the proposed use of
Tier II methodologies in the Guidance.  These comments indicated that it is
critical for the waters of the Great Lakes System that toxic substances not
remain unregulated simply because adequate data to calculate a Tier I
criterion are not available.  These commenters also believe that the proposed
approach properly places the burden of proof on dischargers to demonstrate
that their discharges will not damage the aquatic ecosystem or human health.

      With regard to the aquatic life methodology, EPA carefully reviewed the
concerns of commenters about potential overlap between Tier II requirements
for aquatic life and the appendix F requirements for whole effluent toxicity
testing.  The final Guidance clarifies and elaborates on the option of  using
indicator parameter limits under 40 CFR 122.44(d)(1)(vi)(c).  A full
discussion of the use of the indicator parameter option and its role in the
final Guidance appears in section VIII.E of this document.  As discussed in
that section of the*document, the State or Tribe is still required to adopt
the Tier II methodology for aquatic life.  As described in procedure 5  of the
final Guidance, States and Tribes will then be required under this Guidance to
implement the Tier II values through water quality-based effluent limitations
(WQBEL) based on such values when, under procedure 5, such limits are
determined to be required.  When deriving limits to meet Tier II values.
States and Tribes have the option of using an indicator parameter limit,
including use of a WET limit under appropriate conditions, in lieu of a Tier
II-based limit.  If an indicator parameter is used, the State or Tribe  must
ensure that the indicator parameter will attain the  "applicable water quality
standard"  (as described in 40 CFR 122.44(d)(1)(vi)(C).  The "applicable water
quality standard" in this instance would be the State's or Tribe's narrative
water quality standard that protects aquatic life, as interpreted using the
Tier II methodology.

      EPA does not agree, however, with comments that questioned the
scientific validity of the proposed aquatic life Tier II methodology.   EPA has
reviewed commenters1 concerns, and concluded that the Tier II approach  is
scientifically sound and necessary for the Great Lakes System, and is
appropriate for development of water quality-based effluent limits.
Furthermore, EPA does not agree with comments that the methodology is overly
restrictive.  For these reasons,  the proposed aquatic life Tier II methodology
is retained in the final Guidance.  Further discussion of this issue is found
in section III of this document.

      EPA also does not agree with comments  that  the proposed human health
Tier II methodology is not  scientifically valid.  A  full discussion of  this
issue is found in section V of this document.

      EPA does agree with  some of the concerns expressed about the
applicability of the wildlife methodology for developing wildlife  criteria  or

-------
                           Section II: Regulatory Requirements                        39
values beyond the BCCs.  As a result, EPA has modified the final Guidance  to
no longer require Great Lakes States arid Tribes to adopt provisions mandating
the use of the proposed wildlife Tier II methodology when developing water
quality-based effluent limits, nor to require the use of the wildlife Tier I
methodology for pollutants other than BCCs.  These issues are discussed
further in section VI of this document.

      EPA also agrees with comments that it is appropriate for dischargers to
share the cost of developing data on pollutants for which there are no Tier I
criteria and/or Tier II values, and does not agree with comments that it is
inappropriate.  EPA recognizes that the ultimate statutory responsibility  for
developing, adopting, and approving water quality standards rests with States,
Tribes, and EPA.  The CWA, however, also makes dischargers ultimately
responsible for the content of their discharges, and gives broad authority to
the Administrator and the States for data gathering and reporting concerning
such discharges  (CWA sections 308, 402).  As a matter of public policy, EPA
believes it is appropriate to provide incentives to reduce the presence of
toxic pollutants in discharges.  Procedure 5.C of appendix F of the final
Guidance provides such an incentive.  Upon adoption or promulgation, it would
require NPDES permit authorities to generate toxicological data sufficient to
calculate Tier II values in certain circumstances, but would retain their
current ability to require permittees to generate such data.

      At the same time, EPA does not want to impose an undue burden on
dischargers, and has reviewed carefully the comments of those concerned about
the cost and time required to generate Tier II data.  EPA has concluded that
because of the amount of existing data already available for the GLI
Clearinghouse, the potential burden to generate required Tier II data in
specified circumstances will be relatively insignificant.  EPA's analysis  is
as follows.

      Upon adoption or promulgation, the Guidance would provide that
dischargers could be required, under conditions and with some exceptions
specified in procedure 5.C of appendix F, to generate data to derive Tier  II
values if sufficient data do not exist for the 138 pollutants selected by  the
States and EPA for the initial focus of the Initiative  (listed in Table 6  of
part 132).  This means that a permitting authority, when faced with the need
to develop an effluent limit for a pollutant for which there is no Tier I
criterion or Tier II value, would need to generate or require the permittee to
generate data to develop a Tier II value for that pollutant.  The conditions
and exceptions for this procedure are described further in section VIII.E  of
this document.  For human health, EPA has already developed Tier I criteria
for 18 of the 138 pollutants, and has tentatively determined that there is
currently enough toxicological information available to calculate at least 101
more Tier I criteria, leaving only 14 pollutants for which there is not enough
data to calculate at least a Tier II human health value, and 5 pollutants
which are currently under review by EPA.  For aquatic life, EPA has developed
Tier I criteria for 15 of the 138 pollutants, and in addition has tentatively
determined that there is currently enough information available to calculate
at least 98 more Tier I criteria or Tier II values.  This means that the
maximum potential burden for dischargers to generate required data would be
for 14 human health values and 25 aquatic life values.  For several of the
remaining 25 pollutants, however, it is possible that the pollutant is
insoluble in water at levels that are acutely toxic to daphnids.  In these
situations, under the scientific defensibility exclusion of § 132.4(g) there
would be no need to conduct the toxicity tests.

      This maximum potential burden may never be realized by dischargers for
several reasons.  First, in operating the GLI Clearinghouse, EPA and the
States may identify additional data on the remaining pollutants.  Second,  EPA
may decide to generate the necessary data for one or more of the remaining
pollutants.  Third, one or more States may decide to generate the data, and
not require dischargers to generate the data.  Finally, the pollutants may not
be in an individual discharger's effluent, and therefore would not trigger the

-------
40     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

need to generate data under procedure 5 of appendix F.  Because of the above
mitigating factors, EPA believes the burden to generate required Tier II data
is likely to be small.

      EPA is also aware that because the Tier II methodologies generally yield
more conservative numbers than Tier I,  to reflect the greater uncertainty
related to the absence of complete data sets, an incentive is created for
dischargers to generate additional toxicological data to enable generation of
a new Tier II value or a Tier I criterion.  To the extent that dischargers
choose to conduct studies to generate new data, this is an optional activity
and not a requirement of the final Guidance.  Additionally, the cost of such
testing could be offset by any reduction in treatment costs associated with
less stringent permit limits based on Tier I criteria.  EPA has therefore not
estimated the potential cost if any of this optional activity.

      Additionally, in situations where dischargers are required to generate
data to derive Tier I criteria or Tier II values, the final Guidance continues
to specify that permit authorities may grant a reasonable period of time in
which to provide additional studies necessary to develop a Tier I criterion or
to modify the Tier II value.  These provisions are discussed further in
section VTII.H of this document.

      EPA also agrees in part with commenters who recommended allowing the use
of established biological tests in place of the Tier II methodologies for
numerical pollutant,-specific values.  As described above, when deriving limits
to meet Tier II values, States and Tribes have the option of using an
indicator parameter limit, including use of a WET limit under appropriate
conditions, in lieu of a Tier II-based limit.  EPA does not agree, however,
with suggestions by commenters that the entire Tier II approach be replaced by
reliance on established biological tests.  Although whole effluent toxicity
testing and permit limits are established procedures that EPA believes are
effective in controlling toxicity to aquatic organisms in many circumstances,
such testing is not designed to address toxic effects on human health.
Furthermore, the whole effluent toxicity approach has some limitations for
aquatic life protection as discussed in the section III of this document.
Similarly, other ecological testing suggested by commenters, such as rapid
bioassessment and biological criteria evaluations, do not address human health
concerns.  Other suggestions, such as effluent and chemical-specific
bioconcentration testing, and fish tissue residue studies of receiving waters,
have been adopted into the Guidance as part of the methodology to develop
bioaccumulation factors in section IV of appendix B, and as part of procedure
8.F of appendix F.  They are measurement methods, however, not regulatory
approaches, and therefore should not be used as separate regulatory controls
to replace the Tier II approach as suggested by some commenters.  Furthermore,
they do not directly measure end points of concern in protecting aquatic life,
and would likely nof be acceptable indicators under existing
§ 122.44(d)(1)(vi).

      Finally, EPA does not agree with the concern of many commenters that the
CWA's anti-backsliding provisions will prevent the future adjustment of water
quality-based effluent limits based on Tier II values.  Application of anti-
backsliding provisions of the CWA are discussed in section II.C.3 of this
document.

      c.    Final Guidance.  For the reasons above, EPA has made specific
modifications and exceptions to the proposed minimum requirements for States
and Tribes to adopt-and use Tier II methodologies.  These changes include:

            The application of the Tier II methodology for aquatic life has
been modified to clarify that States and Tribes have the flexibility to use
either Tier II values or indicator parameters  including whole effluent
toxicity, where appropriate and justified,  consistent with the current
national NPDES permit program.  This change is reflected in changes to

-------
                           Section II: Regulatory Requirements                        41
procedure 5 of appendix F to part 132, and is discussed further  in VIII.E  of
this document.

            EPA has eliminated the proposed minimum requirement  to use  a Tier
II methodology for deriving wildlife values in the development of water
quality-based effluent limits.  This change is reflected by removing  the Tier
II wildlife methodology from appendix D to part 132, and retitling the
appendix; by removing requirements to adopt and ufie the methodology in
§§ 132.4(a)(5), 132.4(c), and 132.4(d)(4); and removing wildlife values from
the definition of Tier II value in § 132.2.  These changes are discussed
further  in section VI of this document.  Removing these requirements  from  part
132, however, does not relieve permit authorities from their responsibilities
under 40 CPR 122.44(d)(1) to ensure that narrative water quality standards are
implemented to protect wildlife.

3.    Application of Anti-backsliding Provisions of the Clean Water Act

      a.    Proposal.  The preamble to the proposed Guidance  (58 PR 20837)
discussed why in most cases the anti-backsliding provisions of the CWA  will
not prevent adjustments to either Tier I criteria or Tier II values.   ("Anti-
backsliding requirements" refers to the need to make certain showings to
justify making an effluent limitation less stringent.)  First, because  anti-
backsliding requirements do not apply to changes made in an effluent
limitation prior to its compliance date, they would not apply where permit
limits based on Tier II values are revised as a result of studies performed
under a compliance schedule pursuant to proposed procedure 9 of  appendix F.
Second, even where anti-backsliding requirements do apply  (e.g., where
effluent limitations based on Tier I criteria or Tier II values  change  after
the compliance date), dischargers may be able to meet the requirements  of
section 303(d)(4) and therefore be allowed to have less stringent effluent
limitations.

      b.    Comments

      Comment;  Comments indicated that EPA should exempt Tier II based limits
from anti-backsliding requirements regardless of the time period involved,
even after the limits become effective.

      Response:  The statute does not provide the flexibility to exempt Tier
II limits from anti-backsliding requirements after the limits become
effective.  However, as discussed below, even where anti-backsliding
requirements do apply, they do not prevent all changes for effluent
limitations.

      Comment:  Some comments asked for the statutory justification for
indicating that anti-backsliding requirements do not apply until the
compliance date or if a limitation is appealed.

      Response:  EPA has reviewed the statutory language contained in section
402(o)(1), which contains the general prohibition against backsliding for
WQBELs.  The statute refers to effluent limitations that have been
"established."  Restrictions on backsliding do not apply to challenged  permit
limits which have been stayed pending final agency action as these limitations
have not been "established."  For example, where a limit is challenged  in  an
evidentiary hearing or administrative appeal, the limit may be made more or
less stringent than the initially proposed limit, without the change  being
subject to the anti-backsliding requirements.  EPA has also determined  that
the anti-backsliding requirements do not apply to limits with a  delayed
compliance date, until the date of compliance, as the limitation again  is  not
yet "established" until that date.  EPA has codified in procedure 9.C of
appendix F of the final Guidance its interpretation that a limitation is not
established for the purposes of section 402(o) until its effective date.

-------
42     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      None of the Great Lakes States has indicated to EPA that  they have anti-
backsliding requirements that are more stringent than the EPA interpretation.

      Comment:  Commenters stated that it will be very  difficult  to satisfy
the anti-backsliding requirements.

      Response:  There is flexibility provided by EPA's interpretation of
sections 402(o) and 303(d)(4) to allow relaxation of effluent limitations in
many of the circumstances described in the comments.  Specifically,  EPA does
not interpret section 402(o) to require compliance with both sections
402(o)(2) and 303(d)(4), but only with one or the other (as well  as 402(o)(3))
in order to establish less stringent limitations.

      Comment:  Commenters stated that EPA should issue a regulation
interpreting the anti-backsliding requirements.

      Response:  EPA is not issuing a regulation defining national  anti-
backsliding policy in this rule, since such a regulation would  have much
broader applicability than the Great Lakes basin.  However, EPA is  repeating
and clarifying its interpretation of the anti-backsliding requirements of
sections 402(o) and 303(d)(4) in this document.  See the Final  Guidance
section below for a detailed explanation of EPA's interpretation.

      Comment:  Commenters stated that backsliding should not be  allowed from
current effluent limitations and existing conditions.

      Response:  The statute or existing regulations do not prohibit a change
from existing limitations, so long as the exceptions are met and  there is
compliance with the requirements contained in section 402(o)(3).

      Comment:  A number of comments dealt with the interaction of  anti-
backsliding and antidegradation requirements.

      Response:  Antidegradation requirements do apply  to an analysis  of the
anti-backsliding requirements, as provided under section 402(o)(3).  However,
this is consistent with the general requirement under the CWA that  permits
comply with water quality standards including antidegradation.   See the NPDES
regulations (40 CFR 122.44(b) and  (d)) on establishment of WQBELs to ensure
compliance with water quality standards, including antidegradation  policies.
Specific issues concerning compliance with antidegradation in the Great Lakes
are addressed in section VTI of this document.

      Comment:  Some Commenters expressed concern that  the antibacksliding
requirements of section 303(d)(4)(A) of the CWA would make it difficult to
revise limits based on Tier II values if studies were completed after  the
compliance date for the limit in question.  For example, if future  studies
resulted in a Tier II value being changed to a less stringent Tier  II  value or
Tier I criterion, after the effective date of an initial Tier II-based limit,
the concern is that antibacksliding would prohibit the  relaxation of the Tier
II-based limit.

      Response:  Section 303(d)(4)(A) requires that the limit which is to be
relaxed have been based on a TMDL or other wasteload allocation established
under section 303(d).  Since any effluent limit based on a Tier II  value will
also be based on procedure 3 of appendix F  (i.e., on an approved TMDL, WLA in
the absence of a TMDL, or assessment and remediation plan),  such effluent
limitations will satisfy the requirement that they be based  on  a TMDL  or other
wasteload allocation established under section 303(d).  EPA  therefore  believes
that permittees will be able to satisfy the requirements of  section
303(d) (4) (A) .

      c.    Final Guidance.  EPA has retained the proposed provisions
concerning the application of CWA anti-backsliding requirements to  the
implementation of the final Guidance.  Final procedure  9.C of appendix F

-------
                           Section II: Regulatory Requirements                        43
states explicitly that anti-backsliding requirements contained  in  section
402(o) do not apply to changes made in an effluent  limitation prior  to its
compliance date.  In addition, there is adequate flexibility contained in
EPA's interpretation of the anti-backsliding requirements of the CWA to allow
adjustments to either Tief I criteria or Tier II values  in many situations.

      The approach in the final Guidance regarding  the applicability of anti-
backsliding requirements will provide the greatest  degree of uniformity among
the States and Tribes in terms of when Tier II values would become a part of  a
final permit.  In addition, the approach of the final Guidance  will  likely
provide the greatest degree of environmental protection  in the  short term
because there will be a shorter time for completion of the studies and
issuance of a permit with final effluent limitations.  This approach also is
consistent with the language of the CWA itself.

      The following discussion clarifies and expands EPA's interpretation of
sections 402(o) and 303(d)(4), and explains why the States and  Tribes have
considerable flexibility to revise WQBELs.

      Section 402(o) of the CWA, added by the Water Quality Act of 1987 (WQA),
for the first time establishes express statutory language prohibiting the
backsliding of permit limits.  Section 402(o) consists of three main parts.
First, section 402(o)(1)  prohibits  (subject to exceptions in sections
303(d)(4) and/or 402(o)(2)) backsliding of two types of  permit  limits:  (1)
technology-based effluent limitations based on best professional judgment
(BPJ) being revised to reflect subsequently promulgated  effluent guidelines
which are less stringent, and  (2) water quality-based effluent  limitations
established on the basis of sections 301(b)(1)(C) or 303(d) or  (e).   Second,
section 402(o)(2) outlines six specific exceptions  to the prohibition
contained in section 402(o)(1).  Third, section 402(o)(3) outlines a baseline
requirement that must be met before any limits can  be relaxed,  namely that the
new limit must ensure compliance with applicable effluent guidelines and water
quality standards.

      EPA's pre-WQA anti-backsliding regulations were revised on January 4,
1989  (54 FR 246), to reflect the prohibition imposed by  section 402(o)  for the
first situation: revision of existing BPJ-based permit limitations to reflect
subsequently issued effluent guidelines  (40 CFR 122.44(1} (2)) .  EPA's current
anti-backsliding regulations have not been revised  to reflect the  1987 WQA
prohibition on the backsliding of the second situation:  relaxation of effluent
limitations established on the basis of sections 301(b)(1)(C) or 303(d)  or
(e).  However, EPA believes these provisions must be implemented based upon
the CWA in the meantime.

      All other types of backsliding--for example,  backsliding  from  effluent
guideline derived limits, from new source performance standards, from existing
BPJ limits to new BPJ limits, or from water quality-related standards or
conditions (except for effluent limitations)--remain unaffected by the 1987
WQA amendments and EPA's existing regulations at 40 CFR  122.44(1)(1)  will
continue to govern them.   This is because section 402(o) only prohibits the
backsliding of "effluent limits," not other standards or conditions  such as
monitoring frequency or changes in species or protocol for whole effluent
toxicity (WET) testing.  In addition, the requirements contained in  CWA
section 402(o) do not lend themselves to application to  changes of such
standards and conditions.  The relaxation of all other types of standards or
conditions contained in a permit are, however, subject to EPA's existing
backsliding regulations at 40 CFR 122.41(1) (1).  Under these regulations, a
permittee must meet a cause for modification in order to allow  relaxation.

      As indicated in the preamble to the proposed  Guidance  (58 FR 20837), EPA
has consistently interpreted section 402(o) of the  CWA to allow relaxation of
WQBELs if either the requirements of section 402(o)(2) or section  303(d)(4)
are met.  These are independent exceptions to the prohibition against
relaxation of water*quality-based permit limitations.  In dealing  with anti-

-------
44     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

backsliding issues under the final Guidance, section 303(d)(4) will,  in most
cases, provide the flexibility necessary for permitting authorities  to  issue
permits reflecting adjustments to Tier I criteria or II values.

      Section 402(o)(1) provides that backsliding from WQBELs  is prohibited
except in compliance with section 303(d)(4).  Section 303(d)(4) has  two parts
that must be considered, along with an identification requirement: paragraph
(A) which applies to "non-attainment waters" and paragraph  (B) which applies
to "attainment waters."

      Section 303(d) (4) (A) allows establishment of a less stringent  WQBEL when
the receiving water has been identified under section 303(d) (1) (A) and  where
applicable water quality standards are not being met  (i.e.,  a  "non-attainment
water"), if the permittee meets two conditions.  First, a permittee  may seek a
less stringent effluent limitation under section 303(d) (4) (A)  only if the
existing permit limitation was based on a total maximum daily  load  (TMDL)  or
other wasteload allocation  (WLA) established under section  303.  Second,
relaxation of a WQBEL is only allowed if attainment of water quality standards
is ensured, or if the designated use which is not being attained is  removed in
accordance with 40 CFR part 131.

      Section 303(d)(4)(B) applies to waters where the water quality equals or
exceeds levels necessary to protect the designated use, or  to  otherwise meet
applicable water quality standards  (i.e., an "attainment water").  Under
section 303(d)(4)(B), permit limitations based on a section 303 TMDL/WLA,  on
any water quality standards established under section 303,  or  on any other
permit standard may be relaxed only where this is consistent with a  State's
antidegradation policy  (see 40 CFR 131.12).

      Section 402(o)(2) also outlines exceptions to the general prohibition
against backsliding from WQBELs.  These exceptions are independent of the
section 303(d)(4) exception discussed above and are also applicable  to  the
backsliding of BPJ limits to reflect subsequently promulgated  less stringent
guidelines.

      Regardless of whether any of the backsliding exceptions  are applicable
and met, section 402(o)(3) acts as a floor and restricts the extent  to  which
WQBELs  (and BPJ limits) may be relaxed.  Specifically, section 402(o)(3)
prohibits the relaxation of such permit limitations below applicable
technology-based effluent limitation guidelines in effect at the time the
permit is renewed, reissued or modified.  In addition, it prohibits  the
relaxation of limits if such relaxation would result in a violation  of
applicable water quality standards, which include antidegradation
requirements.

      EPA is providing four examples of the application of  anti-backsliding
requirements.   (EPA has not provided the analysis of these  examples  under
section 402(o)(2), because section 303(d)(4) in almost all  cases is  more
flexible.)

            Examole  1
            Scenario; A publicly owned  treatment  works  (POTW)  seeks to relax
      its WQBEL for pollutant X.  The current permit  limitation is based on
      the TMDL and WLA for the POTW  developed in  accordance  with 40 CFR 130.7.
      The POTW is in compliance with its  existing limitation.   The applicable
      water quality standards for pollutant X is  attained.   The POTW has
      developed new models with new  river flow  information,  which indicate
      that the water quality standards  for pollutant  X  would be maintained
      with a relaxed permit limitation.   The permitting authority can revise
      the permittee's WLA to allow a larger discharge of pollutant X because
      another discharger to the TMDL ceased the discharge of pollutant X.  May
      the effluent limitation for pollutant X be  relaxed?

-------
                     Section II: Regulatory Requirements                        45
      Answer: Section 303(d)(4) may justify the requested relaxed permit
limitation.  Section 303(d)(4)(B) is the applicable provision because
the water quality standards for pollutant X is currently attained.
Under section 303(d)(4)(B), a permit limitation can be relaxed  if
antidegradation requirements are met.  Finally, the permitting  authority
may only allow backsliding if the relaxed limitation would not  result  in
violation of any effluent limitations guideline or other water  quality
standards.

      Example 2

      Scenario; On June 30, 1991, the State issued a NPDES permit to an
industrial permittee which for the first time included a WQBEL  for
pollutant Y.  The limitation for pollutant Y is a delayed effective date
limitation which is effective on June 30, 1994.  The WQBEL is derived
from the State's existing water quality criteria.  The State conducted
additional water quality studies on pollutant Y during its triennial
review in 1993 and relaxed the water quality criterion for pollutant Y.
On January 30, 1994, the permittee seeks to modify its permit to relax
the effluent limitation for pollutant Y, based upon the new State water
quality criterion.  Will the anti-backsliding provisions of the CWA and
NPDES regulations prevent relaxation of the permit limitation?

      Answer: No.  In this case, the permittee seeks to revise  an
effluent limitation which is not yet effective.  The anti-backsliding
provisions of the CWA and NPDES regulations do not apply to a delayed
effective date limitation until it is effective.  Prior to relaxing the
permit limitation, however, the permitting authority will need  to ensure
that the action is consistent with antidegradation provisions.

      Example 3

      Scenario; The State has a narrative water quality criterion of "no
toxics in toxic amounts."  On the basis of WET testing data or  other
information, the State finds reasonable potential to exceed the
narrative water quality criterion and imposes a WET limitation  under 40
CFR 122.44(d)(1)(v).  The permittee determines that pollutant Z is the
cause of the WET in its discharge.  The permittee can demonstrate
through sufficient data  (including WET testing data) that an effluent
limitation for pollutant Z will assure compliance with the narrative
water quality standards as well as the State's numeric criteria for
pollutant Z as required by 40 CFR 122.44(d)(1)(v).  May the State modify
the permit to delete the WET limitation and to add the limitation for
pollutant Z?

      Answer: Section 303(d)(4) may justify this action.  The applicable
provision of section 303(d)(4) is section 303(d)(4)(B) because  the
narrative water quality standards is currently attained.   (The  permittee
is currently complying with the existing WET limitation to attain and
maintain the State's narrative water quality standards.)  Under section
303(d)(4)(B) , the permittee may backslide so long as antidegradation
requirements will be met, and the relaxed limitation will not cause a
violation of any effluent limitations guidelines and water quality
standards applicable to the discharge.  In this case, this appears
likely because the discharger can demonstrate that the new limitation
for pollutant Z will assure compliance with applicable narrative as well
as numeric water quality standards.

      Example-4

      Scenario: An industrial permittee seeks to revise its WQBEL of
1000 mg/L for a pollutant to 6000 mg/L, its actual discharge level.  The
permittee has installed and properly operated and maintained its
treatment facilities, but has been unable to achieve the effluent

-------
46     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      limitation of 1000 mg/L.  The current permit limitation is based upon a
      TMDL and WLA for the permittee, which were developed in accordance with
      40 CFR 130.7.  The water quality standards for the pollutant is not
      being attained.  New modeling information shows that the water quality
      standards for the pollutant will be attained with a permit limitation of
      4000 mg/L.  The permitting authority is able to revise the discharger's
      WLA to allow a permit limitation of 4000 mg/L.  May the permit
      limitation be revised from 1000 mg/L to 6000 mg/L?

            Answer: No.  However, under sections 303(d)(4), the permit
      limitation may potentially be relaxed to 4000 mg/L.  The water quality
      standards for the pollutant is not currently being attained.  Therefore,
      the applicable 303(d)(4) provision is 303(d)(4)(A).  In this case, the
      permitting authority may allow backsliding to 4000 mg/L under section
      303(d)(4)(A) because the existing effluent limitation is based upon a
      TMDL/WLA and the data shows that attainment of the water quality
      standards is assured with a permit limitation of 4000 mg/L  (but not with
      a limitation of 6000 mg/L).  The permitting authority may also revise
      the discharger's WLA to allow a discharge limitation of 4000 mg/L.
      However, before backsliding to a limitation of 4000 mg/L will be allowed
      in this case, the permitting authority must also find that the relaxed
      limitation will not result in violation of applicable water quality
      standards  (including antidegradation requirements) and effluent
      limitations guidelines for the discharge.

4.    Basin-wide Application of Criteria and Values

      a.    Proposal.  With exceptions discussed below, the proposed Guidance
generally provided for State and Tribal application of the criteria, values
and methodologies in the Guidance to all waters of the Great Lakes System
regardless of current use designations.  This approach was selected in order
to provide the integrated Great Lakes ecosystem a consistent approach to
pollution control across the entire basin  (see 58 PR 20838-40).

      The proposal contained four exceptions to this approach: First, Great
Lakes States or Tribes could apply more stringent numeric criteria or values
to any waters of the Great Lakes System within their borders.  Second, Great
Lakes States or Tribes could develop less stringent site-specific
modifications to the criteria and values for aquatic life for specific waters
of the Great Lakes System in certain limited circumstances.  Third, Great
Lakes States and Tribes would be required to adopt different types of human
health criteria depending in part on the designated uses of the waters:
"drinking" criteria and values would apply  to open waters and connecting
channels of the Great Lakes, and to other waters designated for use as public
water supplies;  "nondrinking" criteria and  values would  apply to all other
waters.  Fourth, the Guidance provided general exceptions for 16 pollutants
listed in Table 5 of the proposed Guidance; for discharges from wet weather
point sources; and for situations where the criteria methodologies are not
scientifically defensible.

      b.    Comments.  A few commenters asserted  that  the application of human
health and/or wildlife criteria/values regardless of current use  designations
throughout the Great Lakes System seems overly restrictive.  In particular,
these commenters are concerned  that this practice is not justified for non-
bioaccumulative  substances because these pollutants may never reach the  lakes,
actual drinking water supplies,  or appropriate wildlife  habitats  when
discharged into upstream waters.  Other commenters  stated that Congress  did
not intend to take away the Great Lakes States' ability to develop use
designations and to  develop water quality  standards protective of those  uses.
Other commenters stated that  the approach  fails to  recognize the  ecological
diversity of the Great Lakes  ecosystem.

-------
                           Section II: Regulatory Requirements                        47
      A few comments supported the proposed approach which allows application
of site-specific criteria as a more appropriate mechanism for developing
criteria and values than through a use-designation mechanism.

      EPA carefully considered the concerns expressed about the proposed
approach, but continues to believe the proposed approach is appropriate.
First, section 118(c) of the CWA requires the Guidance to specify minimum
numerical limits on pollutants to protect human health, aquatic life, and
wildlife in the Great Lakes System.  EPA believes it is a reasonable
interpretation of this requirement to develop criteria that are generally
applicable to the entire Great Lakes System in order to improve consistency of
water quality criteria while providing sufficient flexibility to address site-
specific circumstances.   (See 58 FR 20837-40.)  This interpretation is also
reasonable in light of the short statutory deadline established by Congress to
complete the Guidance.

      Second, the approach was not developed because of an assumption that
pollutants move freely throughout the ecosystem, although both bioaccumulative
and non-bioaccumulative pollutants can become widely dispersed.  Rather, as
explained in the preamble to the proposed Guidance, EPA believes that the
Great Lakes are an integrated ecosystem necessitating a more consistent
approach to pollution control across the entire basin.

      Third, following the current National program of differing criteria
based on differing use designations could seriously hinder--and perhaps
prevent--the attainment of the goals of the CPA.  Contrary to the views of the
commenter above, EPA believes Congress did intend to restrict some of the
current flexibility in the national program in order to achieve more uniform
protection of the Great Lakes System through enactment of the special
requirements in section 118(c) for this ecosystem.

      Fourth, EPA believes that uniform minimum water quality standards can
help simplify implementation and avoid costly duplication of research and
standard-setting by EPA and the Great Lakes States and Tribes.

      Fifth, as a practical matter, designated uses currently do not exhibit
wide variation across the basin.  For example, use designations for most
waters within the Great Lakes System currently include protection of aquatic
life and recreational uses.  No comments were received in response to EPA's
request for comments on any waters within the Great Lakes System that are not
currently designated to protect these uses.

      Finally, to the extent that there may be unique local situations not
amenable to strict application of basin-wide criteria, the final Guidance
contains several areas of flexibility, including: the scientific defensibility
exclusion in § 132.4(h), discussed further in section IX.C.6 of this document;
site-specific criteria modifications available through procedure 1 of appendix
F, discussed further in section VIII.A of this document; and variances
available through procedure 2 of appendix F, discussed further in section
VIII.B of this document.

      EPA agrees with the comment, however, that the Guidance should recognize
the ecological diversity of the Great Lakes ecosystem.  It is reasonable to
provide differential criteria or flexibility in implementing the criteria to
reflect variations due to local physical, chemical, and biological factors.
The proposed Guidance provided some different criteria--for example, different
aquatic life criteria depending on water hardness, and different human health
criteria for open waters of the Great Lakes--to reflect common variations.  It
also provided flexibility of implementation, including site-specific
modifications to criteria to address more specific or localized differences.

      The final Guidance also includes additional flexibility in applying
criteria and values to further recognize ecological diversity within the Great
Lakes basin.  Procedure 1 of appendix F now allows site-specific modifications

-------
48     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

of human health and wildlife criteria and values that can be either more
stringent or less stringent to reflect site-specific information on
bioaccumulation factors.  This change,  together with the flexibility already
provided in the proposal allowing both more stringent and less stringent site-
specific modifications for aquatic life criteria/values, should provide
sufficient flexibility to reflect site-specific conditions and the ecological
diversity of the Great Lakes basin.  Changes to the procedure for site-
specific modifications are discussed further in section VIII.A of this
document.  In addition, changes were made in the definition of high quality
waters in the antidegradation policy to exclude certain waters from an
antidegradation review depending on their ecological, recreational, or
aesthetic significance.  Changes to the antidegradation policy are discussed
further in section VII of this document.

      c.    Final Guidance.  With one exception, the proposed provisions of
§ 132.4(d) specifying the applicability of  criteria and values have not been
changed in the final Guidance.  The exception is that § 132.4(d)(4) has been
modified to eliminate the requirement for States and Tribes to adopt
provisions consistent with the Tier II wildlife methodology, as discussed
above and in section VI of this document.

5.    Pollutants Subject to Federal. State, and Tribal Requirements

      a.    Proposal.  The proposal left to the discretion of the States and
Tribes whether to adopt provisions requiring the use of the Guidance's
criteria development methodologies or implementation procedures 1, 2, 3, 4, 5,
7, 8, and 9 in appendix F for a pollutant if it was listed in Table 5 of the
proposed Guidance.  Proposed Table 5 listed 16 pollutants selected by the
Great Lakes States and EPA during the Great Lakes Initiative process:
alkalinity, ammonia, bacteria, biochemical  oxygen demand, chlorine, color,
dissolved oxygen, dissolved solids, hydrogen sulfide, pH, phosphorus,
salinity, sulfide, temperature, total and suspended solids, and turbidity.
These pollutants would continue to be subject to existing water programs,  such
as State or Tribal programs implementing 40 CFR part 131 for the development
and adoption of water quality standards and criteria, 40 CFR part 122 for
development of NPDES permits, and other appropriate requirements and guidance
under the CWA or State or Tribal law.  They would also be subject to the
antidegradation policy in appendix E.  The  proposal did not need to exempt the
Table 5 pollutants from procedure 6, since  procedure 6 applies to whole
effluent toxicity not to individual pollutants.

      As discussed more fully in the preamble to the proposed Guidance  (58 FR
20842-43), the Initiative Committees believed that regulatory authorities
should retain the flexibility to address these pollutants in their existing
water quality programs.

      b.    Comments.  Many commenters supported the proposed listing of
pollutants in Table 5 that would only be subject to existing Federal, State,
and Tribal requirements.  In particular, several commenters supported the
proposed listing of chlorine in Table 5, pointing out that States need
continuing flexibility  in their programs to deal with the special uses of
chlorine as a disinfectant in water and wastewater treatment.  Other
commenters supported the proposed listing of ammonia, citing its role as a
nutrient as well as its toxicity.

      Other commenters urged EPA  to delete  ammonia and  chlorine from Table 5.
These commenters believe that ammonia and chlorine, although regulated by
states  in the basin, should be controlled more  consistently because of their
potential adverse effects on aquatic biota. Commenters  also argued that the
Guidance methodologies  and procedures can technically be applied to ammonia
and  chlorine, and that  the States are not consistently  regulating  these
pollutants within the basin.

-------
                           Section II: Regulatory Requirements                        49
      Various individual commenters suggested: deleting hydrogen sulfide and
sulfide from Table 5; deleting salinity from Table 5; adding non-toxic
pollutants to Table 5; and adding common inorganic constituents such as
chloride, sulfate, sodium, and calcium to Table 5.

      EPA agrees with comments that Table 5 should be retained in the final
Guidance, because for the pollutants in the final Table 5 it would be
scientifically and technically inappropriate to require use of some or all of
the Guidance's methodologies and procedures.  EPA agrees with comments that
the proposed Table 5 should be modified, however, to remove two pollutants--
hydrogen sulfide and sulfide.  The reasons for listing or removing pollutants
in Table 5, together with EPA's response to comments concerning specific
pollutants, are as follows.

                Alkalinity.  The Guidance methodologies and implementation
      procedures are not scientifically and technically appropriate for
      developing and applying water quality criteria and values for
      alkalinity:  The Guidance methodologies for numeric criteria and values
      and the corresponding implementation procedures are designed to be used
      with specific chemicals.  Alkalinity is not a specific chemical, but the
      combined effect of several substances such as carbonates, bicarbonates,
      hydroxides, borates, silicates, and phosphates.  Furthermore, the
      Guidance methodologies for numeric criteria and values and the
      corresponding implementation procedures are designed to be used with
      chemicals that exert an adverse effect as the concentration of the
      chemical becomes too high, while alkalinity exerts adverse effects on
      aquatic life if it is too low, and human health--irritation to swimmers
      by altering the pH of the lacrimal fluid around the eye--if it is too
      high.  Additionally, some components of alkalinity such as carbonate and
      bicarbonate can have a beneficial effect on water quality by complexing
      some toxic heavy metals and reducing their toxicity to aquatic life.  It
      has also been noted that some waterfowl habitats are more productive
      with higher alkalinities.  The criteria development methodologies and
      implementation procedures in the final Guidance do not address these
      types of issues or end points.  The final Guidance also does not address
      other end points of concern commonly identified for low alkalinity,
      including adverse impacts on industrial uses such as food and beverage
      production.  Therefore, EPA believes it is not appropriate to apply the
      Guidance methodologies and implementation procedures in the case of
      alkalinity.

                Ammonia.  E.PA considered carefully the concerns of some
      commenters that ammonia be removed from Table 5 and made subject to all
      provisions of the final Guidance.  EPA shares concerns of some
      commenters that there may be inconsistencies among State water programs
      addressing ammonia.  Nevertheless, EPA found that there would be
      significant problems in applying the aquatic life criteria methodology
      and corresponding implementation procedures for ammonia, and that even
      if these could be overcome, there would likely be no significant
      improvements in consistency of permit limits for this pollutant
      throughout the Great Lakes basin.

            For reasons described in section II.C.4 of this document, the
      aquatic life methodology in the final Guidance was developed to provide
      generally a single set of criteria (or a single equation, in the case of
      pollutants that are adjusted for a water quality characteristic) to
      protect aquatic life throughout the basin regardless of the specific
      species present in different waterbodies.  For ammonia, however, EPA's
      best current information is that a single set of water quality criteria
      to protect aquatic life is not appropriate.  In January 1985 EPA issued
      a water quality criteria document for ammonia,  "Ambient Water Quality
      Criteria for Ammonia - 1984."  The criteria document was supplemented by
      additional guidance to EPA Regional water quality standards coordinators
      in July 1992 in a memorandum, "Revised Tables for Determining Average

-------
50	Water Quality Guidance for the Great Lakes System ~ Supplementary Information Document

      Freshwater Ammonia Concentrations."  These documents are available in
      the docket for this rulemaking.  In these documents, two sets of
      criteria are presented: one set to protect aquatic life when salmonids
      or other sensitive coldwater species are present, and one set to protect
      aquatic life when salmonids or other sensitive coldwater species are
      absent.  Both sets include separate chronic and acute criteria for
      temperatures ranging from 0° C to 30° C.  The two sets of chronic
      criteria differ from each other by factors of approximately 1.4 at
      temperatures above 15° C, and are the same at temperatures 15° C or
      below.  The two sets of acute criteria differ from each other by factors
      of approximately 1.4 at temperatures above 20° C, and are the same at
      temperatures 20° C or below.

            Furthermore,  EPA has determined that even if ammonia were removed
      from Table 5, there would likely be no significant change in consistency
      among States and Tribes in implementing the criteria.  The EPA 1984
      ammonia criteria document provides general recommendations on
      implementation which emphasizes consideration of site-specific factors.
      The document recommends including site-specific factors when conducting
      wasteload allocation modeling in those situations, including
      consideration of effluent variability, and the selection of different
      design flows for steady-state wasteload allocation modeling depending on
      whether the systems are stressed or unstressed.  Accordingly, States and
      Tribes would likely develop site-specific criteria modifications and/or
      site-specific adaptations of wasteload allocation models for the
      majority of waters throughout the basin.

            Because the implementation procedures in the final Guidance are
      designed to be applicable to a wide range of pollutants, they do not
      provide details that would assist States and Tribes in making site-
      specific modifications and adaptations in the specific case of ammonia.
      For example, appendix F of the final Guidance: provides no direction
      concerning what site-specific information should be developed for
      wasteload allocation modeling in the above situation or how the
      information should be used; does not include direction on how to
      evaluate whether systems are stressed or unstressed; and does not
      describe how design flows should be selected in stressed or unstressed
      systems.  Therefore, in the absence of such direction, States and Tribes
      would simply be implementing current laws and national program guidance.

            EPA considered the possibility of developing more detailed
      guidance concerning the site-specific modifications and adaptations
      recommended for ammonia.  EPA also considered developing site-specific
      numeric criteria and/or site-specific wasteload allocation modeling
      parameters for ammonia for all waters in the entire basin.  Neither
      approach would be possible or administratively feasible within the time
      EPA had available to complete the final Guidance.

            Another problem raised by commenters is that ammonia not only
      produces toxic effects in aquatic life but also is a nutrient that with
      other forms of nitrogen can contribute to accelerated eutrophication of
      lakes and other waters.  EPA agrees that ammonia can contribute to
      eutrophication problems, although if it is controlled to prevent toxic
      effects, its contribution to eutrophication will be less.  Nevertheless
      ammonia needs, to be considered because of its interaction with other
      forms of nitrogen when performing total maximum daily loads and
      wasteload allocations to address eutrophication problems.  The Guidance
      methodologies and procedures do not reflect eutrophication as an end
      point.

            For these reasons, EPA decided to retain ammonia as a pollutant
      listed in Table 5 to part 132.  Nevertheless, because of inconsistencies
      among State programs in the Great Lakes basin in addressing ammonia, EPA

-------
                     Section H: Regulatory Requirements                        51
will take additional steps to review such programs.  These  steps are
discussed below under the Final Guidance section.

    •  EPA notes that under § 132.4 (e) (2), discharges that contain
pollutants on Table 5 such as ammonia are still subject to  the whole
effluent toxicity provisions contained in procedure 6 of appendix F of
the final Guidance.  These provisions are discussed further under the
Final Guidance section below and in section VTII.F of this  document.

          Bacteria.  The methodologies in appendixes B and  C and the
implementation procedures in appendix F are not scientifically and
technically appropriate for developing and applying criteria and values
for bacteria.  First, the concept of bioaccumulation in appendix B
cannot be applied in the case of bacteria.  Second, the human health
methodology in appendix C is based on toxicological analysis of dose-
response relationships, while the criteria for bacteria are based on
upper limits for densities of indicator bacteria in waters  that have
been associated with acceptable health risks for swimmers.  Finally, the
implementation procedures for individual pollutants in appendix F are
generally designed for chemicals whose effects are related  to their
concentration in the water column, whereas the effects of bacteria are
generally related to the density of bacterial colonies measured after
several days of culturing in the laboratory.  Neither steady state nor
dynamic water quality modeling used in appendix F has been  found to be
applicable to-developing total maximum daily loads, water quality-based
effluent limits, or loading limits for bacteria.  Therefore, EPA
believes it is not appropriate to apply the Guidance methodologies and
implementation procedures in the case of bacteria.

          Biochemical oxygen demand.  The Guidance methodologies and
implementation procedures are not scientifically and technically
appropriate for developing and applying water quality criteria and
values for biochemical oxygen demand  (BOD).  First, the Guidance
methodologies for numeric criteria and values and the corresponding
implementation procedures are designed to be used with specific
chemicals or related congeners.  BOD is not a specific chemical, but the
combined effect of many chemicals which lead to the depletion of oxygen
as they are degraded by aquatic organisms.  Second, BOD exerts no
measurable toxic effect itself on aquatic life, human health, or
wildlife, but rather is a precursor together with other factors of an
adverse effect on aquatic life, the lowering of dissolved oxygen.
Third, the water quality modeling needed to develop total maximum daily
loads and water quality-based effluent limits for BOD requires different
data inputs and different mathematical calculations than the modeling
generally required in appendix F for pollutants that exert  a measurable
toxic effect.- Therefore, EPA believes it is not appropriate to apply
the Guidance methodologies and implementation procedures in the case of
BOD.

          Chlorine.  EPA does not agree with comments that  chlorine
should be removed from Table 5.  Key portions of the Guidance
implementation procedures are not scientifically and technically
appropriate for applying water quality criteria and values  for chlorine.
First, chlorine exerts acute toxicity effects, but is chemically highly
reactive and degrades in receiving waters much more rapidly than most
other pollutants.  The half life of total residual chlorine can range
from 1 to 3 hours, which is far shorter than most other pollutants.
Procedures 3.C.4 and 3.D.3 of appendix F provide that wasteload
allocations to protect aquatic life from acute effects shall not exceed
the Final Acute Value (FAV) for the pollutant in order to provide an
adequate margin of safety, as a default assumption in the absence of
site-specific data.  The FAV cap may be exceeded; however,  if a site-
specific mixing zone demonstration is conducted and approved pursuant to
procedure 3 of appendix F.  EPA is concerned that because of chlorine's

-------
52     Water Quality Guidance for the Great Lakes System - Supplementary Information Document

      unique characteristics, the above procedures, together with related
      elements of procedure 5,  have the potential to result in overly
      stringent controls on discharges of chlorine in the Great Lakes, basin.
      Because the Initiative Committees excluded chlorine when developing the
      draft Guidance, State and EPA technical staff did not consider
      chlorine's unique characteristics when developing procedures 3.C.4 and
      3-D.3.  Because of these concerns, EPA does not believe it would be
      appropriate scientifically and technically to apply the Guidance
      implementation procedures without modifications to take into account
      chlorine's unique properties.  EPA considered the possibility of
      reviewing procedures 3.C.4 and 3.D.3 to evaluate and address these
      concerns, but concluded this approach would not be possible or
      administratively feasible within the time EPA had available to complete
      the final Guidance.

            Second,' in the Great Lakes basin there are a number of dischargers
      who practice deliberate,  controlled, repetitive, intermittent
      chlorination in order to control undesirable organisms in their
      production or treatment processes.  Under these scenarios, the total
      time chlorine is discharged is often limited to several hours, and the
      concentration of chlorine discharged over that time period is quite
      variable, such that the peak concentration may be experienced for only a
      limited period of time during those few hours.  The Guidance
      implementation procedures may not be scientifically and technically .
      appropriate for applying water quality criteria in waters affected by
      such discharges.  Furthermore, the restrictions in procedures 3.C.4 and
      3.D.4 concerning wasteload allocations based on acute aquatic life
      criteria were not designed for such situations, and if applied as
      specified in the final Guidance could result in wasteload allocations
      that may not be scientifically and technically appropriate.  Therefore,
      EPA believes it would be inappropriate to require States and Tribes to
      apply the aquatic life methodology and corresponding implementation
      procedures to chlorine.  EPA considered the possibility of developing
      more appropriate implementation procedures for chlorine under these
      circumstances.   EPA concluded that the effort to develop such procedures
      would not be possible or administratively feasible within the time EPA
      had available-to complete the final Guidance.

            For these reasons,  EPA decided to retain chlorine as a pollutant
      listed in Table 5 to part 132.  Nevertheless, because of inconsistencies
      among State programs in the Great Lakes basin in addressing chlorine,
      EPA will take steps to review such programs.  These steps are discussed
      below under the Final Guidance section.

            EPA notes that under § 132.4(e)(2), discharges that contain
      pollutants on Table 5 such as chlorine are still subject to the whole
      effluent toxicity provisions contained in procedure 6 of appendix F of
      the final Guidance.  These provisions are discussed further under the
      Final Guidance section below and in section VTII.F of this document.

            --  Color.  The Guidance methodologies and implementation
      procedures are not scientifically and technically appropriate for
      developing and applying water quality criteria and values for color.
      First, the Guidance methodologies for numeric criteria and values and
      the corresponding implementation procedures are designed to be used with
      specific chemicals or related congeners.  Color is not a specific
      chemical, but the result of degradation processes in the natural
      environment. -There is no agreement as to the chemical composition of
      color, and in fact the composition may vary chemically from place to
      place.  Second, the aquatic life methodology is designed to protect
      aquatic life from adverse toxicological effects, while the effects of
      color in water on aquatic life principally are to reduce light
      penetration and thereby generally reduce photosynthesis by phytoplankton
      and to restrict the zone for aquatic vascular plant growth.  The

-------
                     Section II: Regulatory Requirements                        53
criteria development methodologies in the final Guidance do not address
these end points.  Third, the water quality modeling needed to develop
total maximum daily loads and water quality-based effluent limits for
color would require different data inputs and different mathematical
calculations from the water quality models generally required to
implement the total maximum daily load procedures in the final Guidance.
EPA is not aware of any water quality modeling that is generally
available to support development of total maximum daily loads for color.
Therefore, EPA believes it is not appropriate to apply the Guidance
methodologies and implementation procedures in the case of color.

          Dissolved oxygen.  The Guidance methodologies and
implementation procedures are not scientifically and technically
appropriate for developing and applying water quality criteria and
values for dissolved oxygen.  First, the Guidance methodologies for
numeric criteria and values and the corresponding implementation
procedures are designed to be used with chemicals that exert an adverse
effect as the concentration of the chemical becomes too high, while
dissolved oxygen exerts adverse effects on aquatic life if it is either
too low or too high.  The Guidance methodology to develop aquatic life
criteria and values, and implementation procedures such as those for
total maximum daily loads and loading limits would require significant
revision to accommodate such a pollutant.  Second, the Guidance
implementation procedures are generally designed to develop water
quality-based effluent limits for the same pollutant that has reasonable
potential to exceed water quality standards, while to achieve a
dissolved oxygen level that is not too low requires water quality-based
effluent limits on different pollutants, primarily biochemical oxygen
demand.  The Guidance's implementation procedures would require
significant revision to accommodate these situations.  Third, to achieve
a dissolved oxygen level that is not too high generally requires
addressing eutrophication problems, which are not addressed by the
Guidance, or ensuring that a dam or other hydrologic modification does
not induce oxygen supersaturation through turbulent mixing, which may or
may not involve development of water quality-based effluent limits for
dissolved oxygen.  The Guidance's implementation procedures would
require significant revision to accommodate these situations.
Therefore, EPA believes it is not appropriate to apply the Guidance
methodologies and implementation procedures in the case of dissolved
oxygen.

      EPA considered the possibility of developing more appropriate
criteria development methodologies and implementation procedures for
dissolved oxygen.  EPA concluded that the effort to develop such
methodologies and procedures would not be possible or administratively
feasible within the time EPA had available to complete the final
guidance.

          Dissolved solids and salinity.  The Guidance methodologies and
implementation procedures are not scientifically and technically
appropriate for developing and applying water quality criteria and
values for dissolved solids and salinity.  First, the Guidance
methodologies for numeric criteria and values and the corresponding
implementation procedures are designed to be used with specific
chemicals or related congeners.  Dissolved solids and salinity are not
specific chemicals, but the combined effect of several unrelated
substances.  Second, the methodologies and implementation procedures are
designed to be used with chemicals that exert an adverse effect as the
concentration of the chemical becomes too high.  Although dissolved
solids and salinity can produce adverse effects at high levels, they can
also produce more subtle effects at lower concentrations that affect the
ecological balance of an ecosystem.  For example, increased dissolved
solids and salinity can favor aquatic species that are not native to an
ecosystem and can disrupt the community structure.  The Guidance

-------
54     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      methodologies and implementation procedures do not address these types
      of end points.  Third, the establishment of water quality criteria for
      dissolved solids and salinity often involves site-specific
      considerations that .are not addressed by the final Guidance.  Therefore,
      EPA believes it is not appropriate to apply the Guidance methodologies
      and implementation procedures in the case of dissolved solids and
      salinity.

                Hydrogen sulfide and sulfide.   EPA agrees.with comments that
      the Guidance methodologies and procedures are scientifically appropriate
      for developing and implementing criteria and values for hydrogen sulfide
      and sulfide, and has therefore removed hydrogen sulfide and sulfide from
      Table 5 in the final Guidance.  These pollutants were originally listed
      by the Initiative Committees because they were believed to have only
      organoleptic effects.  In response to comments, EPA reviewed the
      available scientific information, including EPA's Quality Criteria for
      Water, July 1976, and found that the pollutants have adverse effects
      that can be addressed by the Guidance methodologies and procedures.  EPA
      is therefore removing them from Table 5.

            --  pH.  The Guidance methodologies and implementation procedures
      are not scientifically and technically appropriate for developing and
      applying water quality criteria and values for pH.  First, the Guidance
      methodologies for numeric criteria and values and the corresponding
      implementation procedures are designed to be used with chemicals that
      exert an adverse effect as the concentration of the chemical becomes too
      high, while pH exerts adverse effects on aquatic life if it is either
      too low or top high.  The Guidance methodology to develop aquatic life
      criteria and values, and implementation procedures such as those for
      total maximum daily loads and loading limits do not accommodate such a
      pollutant.  Second, the implementation procedures in the final Guidance,
      including the procedures for total maximum daily loads, reasonable
      potential, and loading limits, are generally designed to develop water
      quality-based effluent limits for the same pollutant that has reasonable
      potential to exceed water quality standards.  Achieving a pH level that
      is not too low or too high, however, might require water quality-based
      effluent limits on different pollutants, such as specific acids, bases,
      or buffering compounds that affect the overall pH of the effluent and
      receiving water.  The Guidance's implementation procedures do not
      accommodate these situations.  Third, the implementation procedures for
      total maximum daily loads are designed for pollutants that are expressed
      as an ordinary concentration, while pH is expressed as the negative
      logarithm of the hydrogen ion concentration.  Therefore, EPA believes it
      is not appropriate to apply the Guidance methodologies and
      implementation procedures in the case of pH.

            EPA considered the possibility of developing more appropriate
      criteria development methodologies and implementation procedures for pH.
      EPA concluded that the effort to develop such methodologies and
      procedures would not be possible or administratively feasible within the
      time EPA had available to complete the final Guidance.

            By listing pH among the pollutants in Table 5, EPA does not  intend
      to exclude pH as a water quality factor in calculating criteria and
      values for other pollutants.  Where appropriate, pH should continue to
      be used as such a factor.

                 Phosphorus.  The Guidance methodologies and implementation
      procedures are not scientifically and technically appropriate for
      developing and applying water quality criteria and values for
      phosphorus.  The Guidance methodologies for numeric criteria and values
      and the  corresponding  implementation procedures  are designed to be used
      with  chemicals that  exert an adverse effect, while  the primary
      environmental concern with phosphorus is its role as a nutrient  in

-------
                           Section II: Regulatory Requirements                        55
      accelerating eutrophication of lakes and other waterbodies.  The
      Guidance methodologies and procedures do not reflect eutrophication as
      an end point.  Therefore, EPA believes it is not appropriate to apply
      the Guidance methodologies and implementation procedures in the case of
      phosphorus.

                Temperature.  The Guidance methodologies and implementation
      procedures are not scientifically and technically appropriate for
      developing and applying water quality criteria and values for
      temperature.  The Guidance methodologies for numeric criteria and values
      and the corresponding implementation procedures are designed to be used
      with chemicals that exert an adverse effect as the concentration of the
      chemical becomes too high, while both high and low temperatures can have
      both beneficial and adverse effects, depending on aquatic species, stage
      of life cycle, and other chemical and physical factors.  Furthermore,
      adverse effects from temperature often arise from abrupt spatial and
      temporal differences in temperature, rather than from temperature
      extremes.  The Guidance methodologies and implementation procedures do
      not take these types of factors into account.  Therefore, EPA believes
      it is not appropriate to apply the Guidance methodologies and
      implementation procedures in the case of temperature.

            By listing temperature among the pollutants in Table 5, EPA does
      not intend to exclude temperature as a water quality factor in
      determining criteria and values for other pollutants.  Where
      appropriate, temperature should continue to be used as such a factor.

            --  Total and suspended solids, and turbidity.  The Guidance
      methodologies and implementation procedures are not scientifically and
      technically appropriate for developing and applying water quality
      criteria and values for total and suspended solids, and turbidity.
      First, the Guidance methodologies for numeric criteria and values and
      the corresponding implementation procedures are designed to be used with
      specific chemicals or related congeners.  Total and suspended solids,
      and turbidity are not specific chemicals, but the combined effect of
      several unrelated substances.  Second, the aquatic life methodology and
      corresponding implementation procedures are designed to be used with
      chemicals that exert a direct adverse effect on aquatic life, while
      these pollutants exert their adverse effects through both direct and
      indirect means.  For example, turbidity not only directly clogs gills,
      mats eyes, and impairs respiration, but also indirectly affects aquatic
      life by impairing visibility necessary for finding food, altering water
      temperature, and reducing primary productivity that serves as the base
      of the aquatic food chain.  Total and suspended solids produce many of
      the same effects as turbidity, and also affect reproduction of some
      aquatic life by blanketing spawning areas and smothering eggs in stream
      beds.  The Guidance's aquatic life methodology and implementation
      procedures do not reflect the indirect effects described above.  Third,
      the Guidance implementation procedures are generally designed to develop
      water quality-based effluent limits for the same pollutant that has
      reasonable potential to exceed water quality standards, while to achieve
      a turbidity level that protects aquatic life will require water quality-
      based effluent limits on different pollutants, such as total or
      suspended solids.  Therefore, EPA believes it is not appropriate to
      apply the Guidance methodologies and implementation procedures in the
      case of total and suspended solids and turbidity.

      EPA does not agree with the comments to add additional pollutants to
Table 5.  Some comments suggested adding "non-toxic" pollutants, including
common inorganic constituents such as chloride, sulfate, sodium, and calcium.
EPA has decided not to add such pollutants.  First, it would likely be
difficult to develop an operational definition of "non-toxic."  For example,
each of the inorganic constituents suggested by commenters has some degree of
adverse effects.  Furthermore, the effect of a pollutant is a function of both

-------
56     Water Quality Guidance for the Great Lakes System - Supplementary Information Document

toxicity and exposure, not toxicity alone.  Listing a pollutant in Table 5
solely on the basis of low toxicity, however defined, would not be
appropriate, because even pollutants with relatively low toxicity could have
adverse effects in the environment if present in high concentrations.  Second,
EPA believes that the final Guidance's criteria development methodologies and
implementation procedures are designed to develop appropriate criteria, total
maximum daily loads, and water quality-based effluent limits over a wide -range
of toxicities, including pollutants with lower toxicities.  For example, under
procedure 5 of appendix F, a discharge containing relatively small amounts of
a pollutant with low toxicity might not have reasonable potential to exceed
water quality standards, and therefore would not require water quality-based
effluent limits.  If the pollutants did have reasonable potential to exceed
water quality standards, the Guidance methodologies and implementation
procedures would be used to develop water quality-based effluent limits.  If a
pollutant identified in the future has unique "non-toxic" features making one
or more of the Guidance methodologies or procedures scientifically
indefensible in a particular situation, then § 132.4(h) of the final Guidance
could be used generally to  exempt that pollutant from selected provisions of
the Guidance.  Therefore, for all these reasons, it is neither appropriate nor
necessary to add "non-toxic" pollutants to Table 5.

      The scientific defensibility exclusion discussed in section II.C.6 of
this document is designed to achieve a purpose similar to although more
limited than Table 5.  EPA believes both provisions are useful and appropriate
in different circumstances.  The scientific defensibility exclusion is
available for a pollutant for which the State or Tribe demonstrates that a
methodology or procedure in this part is not scientifically defensible.  It
enables Great Lakes States and Tribes to apply an alternative methodology or
procedure acceptable under 40 CFR part 131 when developing water quality
criteria or implementing narrative criteria, or to apply an alternative
implementation procedure that is consistent with all applicable Federal,
State, and Tribal laws.  This provision would be used to provide exclusions
for reasons which are currently unidentified, or not broadly applicable.  The
Table 5 exclusions, in contrast, are useful for the 14 pollutants where valid
scientific and technical reasons for not regulating them under such provisions
are already available and broadly applicable as discussed above.  The use of
the Table 5 exclusion in § 132.4(g) promotes administrative efficiency and
conserves resources of States, Tribes, and dischargers by not repeating the
analysis of scientific defensibility for each occurrence or discharge of these
pollutants.

      c.    Final Guidance.  The final Guidance retains the proposed Table 5
of part 132.  Table 5 has been renamed "Pollutants Subject to Federal, State,
and Tribal Requirements" to provide a more accurate description.

      The final Guidance also retains the proposed provisions for excluding
these pollutants from certain specified provisions of part 132, but not from
all requirements in Federal, State, or Tribal water quality programs.
Sections 132.4(b), 132.4(c), and 132.4(h){l) of the final Guidance provide
that States and Tribes do not need to apply the methodologies for development
of criteria and values in appendixes A through D for pollutants in Table 5,
but instead must apply any methodologies and procedures acceptable under 40
CFR part 131 when developing water quality criteria or implementing narrative
criteria for these pollutants.  Sections 4(e)(2) and 132.4(h)(2) of the final
Guidance provide that States and Tribes do not need to apply implementation
procedures 1, 2, 3, 4, 5, 7, 8, and 9 of appendix F of part 132 for pollutants
in Table 5, but any alternative procedures used instead must be consistent
with all applicable Federal, State, and Tribal laws.

      EPA recognizes that some of the methodologies or implementation
procedures of the final Guidance  could technically be applied in establishing
controls on the discharge of some or all of  the pollutants listed in Table 5.
For example, procedure 2  (Variances from Water Quality Standards) could be
applied in determining whether to grant a variance from water quality

-------
                           Section II: Regulatory Requirements                        57
standards to a point source discharger of any pollutant.  Great Lakes  States
and Tribes may apply such methodologies or implementation procedures in
establishing water quality criteria or controls on the discharge of any
pollutant in Table 5 of the proposed Guidance.

      In applying alternative methodologies and procedures in the above
situations, States and Tribes are encouraged to apply technical guidance that
EPA has issued where appropriate to assist in developing and implementing
consistent water quality-based controls for these pollutants.

      Section 132.4(f) provides no exclusion for Table 5 pollutants from the
antidegradation provisions of the Guidance.  The antidegradation policy in
appendix E, however, provides for different requirements for BCCs than for
non-BCCs.  The antidegradation policy is discussed in section VII of this
document.

      For reasons discussed above, hydrogen sulfide and sulfide have been
removed from Table 5.  The other 14 pollutants listed in the proposal  have
been retained.  No new pollutants have been added.

      As discussed in the Comments section above, EPA is concerned about the
possibility of inconsistencies among State programs in the Great Lakes basin
in addressing ammonia and chlorine.  For example, there are differences in
numeric criteria adopted by different States for these two pollutants,
especially for ammonia.  There has not been a systematic evaluation, however,
of the implementation procedures used in applying the criteria in the  Great
Lakes States, nor has there been an evaluation of resulting water quality-
based effluent limits in NPDES permits.  In part this is because of the
complexity of comparing implementation procedures and effluent limits  from
State to State where States have flexibility in implementation of water
quality programs, including the authority to be more stringent than the CWA
and implementing regulations.

      In order to determine whether significant inconsistency exists in the
level of protection^ of aquatic life from adverse effects of ammonia and
chlorine, and to take corrective action if warranted, EPA will work with the
States in reviewing water quality standards and implementation of those
standards for ammonia and chlorine in the Great Lakes basin as part of EPA's
responsibilities under section 303(c) of the CWA.  Under section 303(c),
States and Tribes must review and revise their water quality standards every
three years, and EPA must review and approve or disapprove such standards.
For the next two triennial review cycles, EPA will give special attention to
working with the States with respect to standards and implementation
procedures affecting ammonia and chlorine in the Great Lakes basin to  ensure
that any inconsistencies are addressed.  The review will include the following
steps:

         EPA will coordinate with the States and Tribes in reviewing whether
water quality criteria for ammonia and chlorine have been adopted in all
appropriate waters.  Under 40 CFR part 131, States and Tribes must adopt
criteria necessary to protect designated uses.

         EPA will place existing toxicological data on ammonia and chlorine in
the GLI Clearinghouse, including the EPA criteria documents and supplementary
information issued in 1984 and 1992 for ammonia, and in 1985 for chlorine,
under section 304(a)  of the CWA.  It will include any new data or additional
valid data that is hot be reflected in the above documents.

         To assist the States in ensuring that all States adopt adequate and
consistent criteria,  EPA will review numeric criteria adopted by the States
and Tribes for ammonia and chlorine to determine whether the criteria  meet all
applicable requirements and are based on consideration of current available
scientific information, including data in the GLI Clearinghouse.  The  review

-------
58     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

will include consideration of any site-specific modification procedures used
by the States and Tribes.

         EPA will also assist the States by reviewing implementation
procedures in use by States and Tribes to develop total maximum daily loads,
wasteload allocations, and water quality-based effluent limits for these two
pollutants as-part of EPA's ongoing review of TMDLs and NPDES programs and
permits.  To assist in this review, State and Tribal procedures will be
compared with existing national policies and guidance.

      -  EPA will use the results of the review in fulfilling its
responsibilities to review the triennial submissions of State and Tribal water
quality standards.  If a State or Tribe does not adopt and submit appropriate
criteria for these pollutants, EPA will take appropriate steps including
disapproval of the water quality standards, and promulgation of Federal
standards for affected waters to ensure consistent and adequate standards for
all the Great Lakes System.

      The above review steps differ from triennial reviews normally conducted
by EPA in five ways.  First, they specify two pollutants for in-depth
evaluation.  That is, EPA plans to evaluate more detailed supporting
information concerning State and Tribal standards for these pollutants.
Second, EPA will give special attention to identifying and evaluating
differences among State and Tribal programs for these pollutants.  Third, EPA
will coordinate reviews of procedures used in developing total maximum daily
loads and water quality-based effluent limits with the triennial reviews, even
though they are not a part of the review and approval process of 40 CFR part
131.  Fourth, EPA will assist State and Tribal development of water quality
standards by providing information through the GLI Clearinghouse.  Fifth, the
review will be specific to one drainage basin, the Great Lakes basin.  Because
of differences in the timing of different States' and Tribes' review cycles,
and because of the complexities of the interstate comparisons, EPA expects the
full review of ammonia and chlorine standards to require up to two triennial
review cycles.  EPA will make every effort to complete the review no later
than 2003.

      As part of the consultation with the U.S. Fish and Wildlife Service
 (FWS) under section 7 of the Endangered Species Act concerning the Guidance,
the FWS raised concerns that inconsistencies in standards and how they are
applied for ammonia and chlorine may  be adversely affecting endangered or
threatened species in the Great Lakes basin.  In implementing the above
review, EPA will consult with the FWS concerning EPA' s approval of State and
Tribal water quality standards under  section 303(c) of the CWA.  In addition,
EPA will invite and encourage the States and Tribes to participate actively in
the consultations.  EPA and the FWS believe that the review described above,
together with the involvement of  FWS  in consultations with EPA and the States
and Tribes, will serve on an expedited basis to minimize inconsistencies in
controls for ammonia and chlorine.

6.    Scientific Defensibilitv Exclusion

      a.    Proposal.  The proposed Guidance at  § 132.4(g) provides that the
Great Lakes States and Tribes need not apply the proposed criteria
methodologies and implementation  procedures to any pollutant for which the
regulatory authority demonstrates that one  or more procedures in the Guidance
are not scientifically  defensible.  The reason for this exclusion  is that
there may be pollutants  identified in the  future for which  some of  the
methodologies or procedures  in the final Guidance may not be technically
appropriate.

      EPA specifically  invited comment on  whether the final  Guidance  should
specify minimum  requirements  for  use  of this exclusion, demonstration
elements, or procedures for  EPA review of  these  submissions.

-------
                           Section II: Regulatory Requirements                        59
      b.    Comments.  Some commenters recommended removal of the language  in
§ 132.4(g) allowing pollutants to be excluded from Guidance procedures and
State adoption based simply on a State demonstration, preferring that EPA
formally add such pollutants to Table 5.

      Several commenters suggested that no minimum requirement should be
specified for an exclusion demonstration, and that the Guidance should strive
to maintain the flexibility needed for consideration of all site-specific
possibilities.

      EPA has determined that the scientific defensibility exclusion proposed
in § 132.4(g) is necessary and appropriate to include in the final Guidance.
Based on long experience in implementing the CWA, EPA has found that no matter
how carefully a regulatory requirement is planned, there is no way to
anticipate all possibilities.  In the water quality standards program in
particular, new scientific information inevitably arises that cannot always be
accommodated within existing program structures.  Eliminating the exclusion
would likely require future rulemaking by EPA, States, and Tribes to adjust to
new scientific information.  Because rulemaking is often a long process,
eliminating the exclusion would make it very difficult to adapt quickly to new
information when necessary.

      EPA believes that the exclusion will be implemented in a way that will
maintain a reasonable consistency in State and Tribal programs.  As explained
in the preamble to the proposed Guidance (58 PR 20843), EPA Regional Offices
will work with the States and review State demonstrations during water quality
standards submissions, TMDL approvals, and NPDES program implementation.
Through this process, the Regional Offices and States will ensure that the
scientific defensibility exclusions, if approved, will be consistent with the
Guidance, other EPA regulations, and current EPA policy and guidance.

      EPA agrees with comments that implementation of the exclusion should
proceed without further detailed guidance on minimum requirements,
demonstration elements, or review procedures.  Since the nature of the
exclusions cannot be predicted in detail, such guidance would need to be
highly speculative and likely could not anticipate all the circumstances.  EPA
anticipates that the States and EPA Regions will be able to address most
situations in a reasonable way using professional judgment.

      c.    Final Guidance.  For the reasons above, EPA has retained the
exclusion for scientific defensibility in the final Guidance.

      During its review, EPA discovered that there may have been ambiguity
concerning the scope of the exclusion.  EPA intended that the exclusion be
limited to each specific element of the Guidance that was demonstrated to be
inappropriate if applied to a specific situation, and not to all other
elements.  For example, EPA intended that a pollutant for which a State
demonstrated that the Guidance's methodology for development of aquatic life
criteria could not be applied should still be subject to other provisions of
the Guidance adopted into State or Tribal law, such as other criteria
methodologies, site-specific modification procedures, implementation
procedures,  and antidegradation policies.  Accordingly, in order to improve
the clarity of the scientific defensibility exclusion, EPA has created a new
§ 132.4(h) with clarified wording, and deleted the corresponding text from
§ 132.4(g).   The new wording, similar to the proposal, is intended to allow
States or Tribes to use an alternative methodology or procedure that
corresponds to the methodology or procedure found to be scientifically
indefensible.

7.    Wet Weather Exclusion

      a.    Proposal.  The proposed Guidance allowed, where appropriate, but
did not require,  the Great Lakes States and Tribes to adopt provisions
consistent with any of the proposed implementation procedures for establishing

-------
60     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

controls on wet weather point source discharges.  Proposed § 132.4(e)(l)
provided that "any procedures applied in lieu of these implementation
procedures shall be consistent with all applicable Federal, State, and  Tribal
requirements."  Accordingly, even though permitting authorities would not be
required to apply State of Tribal requirements consistent with the part 132
implementation procedures in establishing controls on wet-weather point source
discharges, all permits would still be required to contain any limitations and
conditions necessary to ensure compliance with the CWA and implementing
Federal and State regulations.  The proposed wet weather exclusion did  not
apply to the development of water quality criteria and values, since water
quality criteria and values apply to waters of the Great Lakes System
regardless of the source of the pollutant.

      EPA proposed this exclusion from the Guidance implementation procedures
because they do not address the significant differences that can exist  between
wet weather point source discharges and dry-weather point source discharges.
The preamble to the proposed Guidance discussed these differences, including
the significant variability that exists during and immediately following wet
weather events in rates, durations, and composition of wet weather flows  (58
FR 20840-42).

      b.    Comments.  Several comments were received supporting retaining the
wet weather exclusion in the final Guidance, including comments that
application of stringent provisions in the Guidance would in essence prohibit
all combined sewer overflows.

      A few comments, while not opposing the exclusion on a temporary basis,
urged EPA to develop a mechanism for addressing pollution during wet weather
point source discharges, since commenters believe that wet weather discharges
contribute significant loadings of toxic pollutants to the Great Lakes  System
and must be stemmed.

      No comments on the definition of "wet weather point source" itself were
received, although some comments expressed concerns about whether specific
types of discharges were or were not excluded from the Guidance implementation
procedures.

      EPA agrees with the comments that the exclusion should be retained, for
the reasons given in the preamble to the proposal  (58 FR 20840-42).  EPA also
agrees that mechanisms are needed for addressing pollution during wet-weather
events.  Accordingly, procedure 3.B.8 of appendix F has been clarified  to
provide that States and Tribes must consider pollution resulting from wet
weather events, where appropriate, when developing TMDLs.  States and Tribes
retain flexibility, however, in determining how to account for such discharges
and are free to choose the specific procedures they deem most appropriate.

      EPA recognizes that the proposed definition of "wet weather point
source" should be clarified.  As a result, the technical modifications  to the
definition have been made to clarify which types of discharges are included in
the definition, to clarify which types of discharges are excluded, and  to
delete an unnecessary definition of combined sewer overflow.  The definition
has also been modified to use terminology from the existing definition  of
point source contained in § 122.2, for the convenience of the reader.

      c.    Final Guidance.  Section  132.4(e) of the final Guidance retains
the proposed wet weather exclusion.

      The exclusion applies to "wet weather point sources" as defined in  §
132.2.  For the reasons above, the definition in the final Guidance has been
modified to read as follows:

      Wet weather point source means  any discernible, confined and discrete
      conveyance from which pollutants are, or may be, discharged  as  the
      result of a wet weather event.  Discharges from wet weather point

-------
                           Section II: Regulatory Requirements                        61
      sources shall" include only: discharges of storm water from  a municipal
      separate storm sewer as defined at 40 CFR 122.26(b)(8);  storm water
      discharge associated with industrial activity as defined at 40  CFR
      122;26(b)(14); discharges of storm water and sanitary wastewaters
       (domestic, commercial, and industrial) from a combined sewer overflow;
      or any other  stormwater discharge for which a permit is  required under
      s.ection 402 (p) of the CWA.  A storm water discharge associated  with'
      industrial activity which is mixed with process wastewater  shall not be
      considered a  wet weather point source.

      A combined sewer overflow (CSO) is the discharge from a  combined sewer
system  (CSS) at a point prior to the headworks of a publicly owned treatment
works  (POTW) treatment plant.  A CSS is a wastewater collection system owned
by a State or municipality  (as defined by section 502(4) of the CWA)  which
conveys sanitary wastewaters  (domestic, commercial and industrial wastewaters)
and storm water through a single-pipe system to a POTW treatment  plant  (as
defined in 40 CFR 403.3(p).  CSOs consist of mixtures of domestic sewage,
industrial and commercial wastewaters, and storm water runoff.  CSOs  are point
sources subject to  NPDES permit requirements including both technology-based
and water quality-based requirements of the CWA.  The NPDES permit
requirements for CSOs are explained in EPA1s Combined Sewer Overflow  Control
Policy  (59 FR 18688', April 19, 1994).

      EPA would like to clarify that although Great Lakes States  and  Tribes
are not required to apply Guidance procedures in establishing  controls on the
discharge of pollutants by wet weather point sources  (except for  procedure
3.B.8 of appendix F which provides that TMDLs must consider discharges from
wet weather events, where appropriate), they may nevertheless  choose  to  do so.
Furthermore, the use of such procedures as variances and compliance schedules
are available under existing State programs pursuant to parts  122 and 131.

8.    Bioaccumulative Chemicals of Concern

      a.    Proposal.  The proposed Guidance identified a class of highly
bioaccumulative pollutants, termed BCCs, for special attention.   The  Great
Lakes Initiative Steering Committee believed that every reasonable effort
should be made to reduce loadings of all BCCs, because these pollutants  tend
to persist throughout the Great Lakes ecosystem and have a propensity to
bioaccumulate in the food chain, and have been associated with serious and
systemwide impacts.

      The BCCs were defined in general as those chemicals which bioaccumulate
in aquatic organisms by a human health bioaccumulation factor  (BAF) greater
than 1000, in accordance with the BAF methodology proposed in  appendix B to
part 132.  BCCs would include, but not be limited to the pollutants identified
as BCCs in Table 6.

      In the proposed Guidance, a pollutant found to be a BCC  would be subject
to the following special provisions upon State or Tribal adoption or  EPA
promulgation:

         Under the methodology for deriving non-cancer human health criteria,
the proposed Guidance assumed a relative source contribution (RSC) from
surface water pathways  (water and fish) of 80 percent for BCCs, to at least
partially account for exposures through other pathways.  For non-BCCs, the RSC
assumption was 100 percent.  Accordingly, the proposed methodology for
deriving the human health non-cancer criteria would be 20 percent more
conservative for BCCs than for non-BCCs.

         Under the antidegradation procedures of the proposed  Guidance,  any
action by a discharger that results in an increase in the baseline rate  of
mass loading of a BCC would be considered a significant lowering  of water
quality, and thereby trigger an antidegradation review.

-------
62     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      -  The proposed Guidance provided, in general, that mixing zones for
existing discharges'of BCCs would be eliminated within 10 years after
publication of the final Guidance.  For new sources of BCCs, mixing zones
would not be allowed beginning on the date of publication of the final
Guidance.

         The proposed Guidance provided that for BCCs, permit authorities must
generate or cause the discharger to generate the data necessary to derive Tier
II values to protect aquatic life in certain circumstances that would not
apply to non-BCCs.

         The proposed Guidance would establish a specific process to regulate
the discharge of any pollutant for which the water quality based effluent
limit was below a level that could be analytically quantified.  If the
pollutant were a BCC, the discharger would also have to determine if the BCC
were bioconcentrating or bioaccumulating in fish exposed to the effluent.  If
such monitoring revealed unacceptable accumulation in fish tissue, additional
actions would be required of the discharger.

      b.    Comments.  EPA received a number of comments suggesting that
establishing special provisions for BCCs is not warranted because
bioaccumulation of pollutants is already taken into account in the proposed
methodologies for developing criteria fully protective of the Great Lakes
basin.  EPA also received comments supporting the BCC provisions,  stating that
such provisions are necessary because of the long retention times of the Great
Lakes, and because of the widespread adverse effects to human health and
wildlife from these types of pollutants.

      A number of comments recommended that if special provisions were to be
established for BCCs, the BCCs should be limited to substances that have been
shown through environmental monitoring, including sampling of fish tissues, to
be present at concentrations of concern.  Other comments stated that such an
approach would be overly simplistic, and would not provide a preventive
approach for these pollutants.

      A large number of comments questioned various aspects of the definition
of BCCs.  Many were concerned that the definition did not include the concepts
of toxicity and persistence.  Of these, some comments recommended deleting
pollutants from the BCC list that are not strongly toxic or do not persist for
long periods even if they are highly bioaccumulative, while others supported
adding pollutants that are persistent even if they are not highly
bioaccumulative.  Many comments were received concerning the proposed BAF
cutoff level of 1000 for defining BCCs: many stated that the proposed level
was arbitrarily selected, or based on inadequate scientific analysis; some
felt the proposed cutoff was too low; some felt it was too high; and some
supported the proposed level.  Some comments recommended that only field-
measured BAFs, not predicted BAFs, should be used in determining BCCs; others
supported the proposed use of both field-measured and predicted BAFs.  Some
comments recommended considering the sediment route of exposure in
establishing BCCs.

      Some comments recommended that additional chemicals should not be
subject to the special provisions for BCCs until after formal public comment
in the Federal Register including review of the BAF.

      After careful consideration of the comments, EPA continues to believe
that the special provisions for BCCs are warranted.  EPA's continued support
of the special emphasis on BCCs parallels the position of the Great Lakes
States as initially expressed by State  representatives on the Initiative
Committees.  EPA believes that these special provisions for BCCs are a
reasonable approach to address the issue of persistent bioaccumulative
pollutants in the Great Lakes System, for the following reasons.  First,
persistence of toxic pollutants is a major  concern  in an aquatic system like
the Great Lakes, for the reasons discussed  in the preamble to the proposed

-------
                           Section II: Regulatory Requirements                        63
Guidance  (58 FR 20820-23, 58 FR 20844-45), and in section I.B of this
document.  Persistence is especially problematic for chemicals that are highly
bioaccumulative, because the most important exposure pathway for these
chemicals for humans and wildlife in the Great Lakes System is consumption of
fish and other aquatic organisms.  Persistent bioaccumulative chemicals will
result in high exposures to humans and wildlife for a long time to come.

      Second, proposed procedures 3A and 3B in appendix F for development of
total maximum daily loads  (TMDLs) envisioned  predominant use of a simple,
steady-state mass balance approach for both bioaccumulative and non-
bioaccumulative pollutants.  Although the final Guidance would allow use of
other approaches for developing TMDLs, EPA expects that the steady-state mass
balance approach will be the approach used in a majority of waters within the
Great Lakes System because of its ease of implementation.  The steady-state
mass balance approach is a method used to approximate the mass of pollutants
within a water body.  This approach assumes that the input of mass into the
system (e.g., through point and nonpoint source loadings, atmospheric
deposition, groundwater seepage) equals the loss of mass from the system plus
any losses due to transformation of mass within the system.  In other words,
the approach assumes that no mass accumulates in the system.  This method
provides a first approximation of allowable loading allocations.

      For persistent bioaccumulative pollutants, however, approximation based
on a steady-state mass balance approach will likely not be accurate.  As
discussed in section I.A of the preamble to the proposed Guidance  (58 FR
20822), there are significant interactive physical, chemical, and biological
processes that affect the long-term behavior of persistent bioaccumulative
pollutants in the Great Lakes System, resulting in fairly common occurrences
where such pollutants do accumulate in the system.  Additionally, although the
phased approach to TMDLs discussed in section VTII.C of this document
recommends subsequent monitoring to identify any shortcomings in the chosen
control approach, this approach may present a significant risk of allowing
persistent bioaccumulative pollutants to concentrate in the ecosystem above
ambient criteria levels before the control approach can be evaluated and
revised as necessary.  EPA believes the costs of future remediation actions to
address BCCs would be significantly more expensive than efforts to control the
BCCs before they enter the environment.  Accordingly, additional controls
intended to prevent concentrations of persistent bioaccumulative pollutants
from increasing to the level of criteria concentrations in Great Lakes waters
are reasonable.

      In the proposal, EPA requested comment on issues concerning the details
of the proposed special provisions for BCCs.  After analyzing those issues and
the comments received, EPA has modified several of the provisions in ways that
may in some cases reduce costs for the regulated community without
significantly increasing the risk from BCCs.  EPA believes that with these
modifications the provisions for BCCs will continue to address the concerns of
the Initiative Committees for controlling the discharges of BCCs.  These
modifications include:

          Modifying the methodology for deriving non-cancer human health
criteria to assume a relative source contribution  (RSC) from surface water
pathways of 80 percent for all pollutants, not just BCCs.  Therefore, the 80
percent RSC still applies to BCCs, but is no longer a "special" provision for
BCCs.  The reasons for this change are discussed in section V.C.5 of this
document.

          Changing the antidegradation provisions to replace numeric existing
effluent quality-based ("EEQ")  limits as a means of implementing
antidegradation for BCCs with a narrative description of the types of
activities that will trigger an antidegradation review, and to provide greater
flexibility in the implementation, demonstration and decision components.
These modifications are discussed further in section VILA of this document.

-------
64     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

          Providing an opportunity for dischargers to retain a limited mixing
zone for a BCC under certain limited circumstances if the discharger is
complying with all applicable technology-based and water quality-based
requirements and has reduced its discharge of the BCC to the maximum extent
possible, but is unable to meet water quality standards in the absence of a
mixing zone.  This modification is discussed further in section VIII.C.4 of
this document.

          Deleting the provision for dischargers to determine if BCCs were
bioconcentrating or bioaccumulating in fish exposed to the effluent in
situations where the water quality-based effluent limit was below a level that
could be analytically quantified.  This modification is discussed further in
section VIII.H.4 of'this document.

      EPA could not accept the concept put forth by commenters that pollutants
should not be regulated as BCCs until they are shown to be present at
concentrations of concern in the Great Lakes System.  As discussed above, EPA
is concerned about preventing concentrations of BCCs from increasing to the
level of criteria concentrations in Great Lakes waters.  The regulatory
approach, suggested by some commenters, that would not trigger preventive
action until some measurable concentration resulting in adverse conditions is
reached in the environment would not be effective in addressing this concern,
particularly because of the difficulties of measuring these pollutants at
levels of concern in the environment.  As discussed further in sections VII.B
and VIII.C.4 of this document, the special provisions for BCCs in the final
Guidance will take full effect over the next twelve years (two years for
State/Tribal adoption or promulgation,  plus ten year phase-in period).  A
program requiring systematic environmental monitoring followed by a regulatory
process to designate BCCs could significantly delay implementation of these
provisions and allow build-up of new persistent, bioaccumulative pollutants in
the Great Lakes System.  The risks to the Great Lakes ecosystem of such a
delay are too great to warrant such an approach.

      EPA agrees with comments that toxicity and persistence should be
included in the definition of BCC.  As discussed in the preamble to the
proposed Guidance (58 FR 20807, 58 FR 20820), toxic pollutants that are
persistent and bioaccumulate are of particular concern in the Great Lakes
System.  With regard to toxicity, EPA has amended the definition of BCC in the
final Guidance to provide that a chemical must also have "the potential to
cause adverse effects" in order to be a BCC.  Under this revised definition,
if data become available showing that a chemical that otherwise meets the BCC
definition does not have the potential to cause an adverse effect, State or
Tribal authorities would not have to apply any adopted or promulgated
provisions for BCCs to that chemical.  EPA expects that very few pollutants,
if any, would be excluded as BCCs in this way, since most substances have
potential adverse effects at some level of concentration.  Nevertheless, if
scientifically valid experimental evidence is provided which demonstrates that
a chemical has no potential for producing adverse effects, then a State or
Tribe could find that the special provisions for BCCs need not be applied for
that pollutant.

      EPA has also made changes to the final definition of BCCs with regard to
persistence.  As discussed in the preamble to the proposed Guidance  (58 FR
20821), the proposed definition of BCC was based principally on the concept of
bioaccumulation.  EPA and the Initiative Committees had considered including
persistence in the definition, but found that data were not systematically
available concerning persistence.  That is,  systematic data were not found
concerning the cumulative effect of relevant fate and effect processes for the
full range of specific pollutants in the Great Lakes System under field
conditions, or under laboratory conditions which had been field correlated and
verified.  Upon review  of comments received, and after reevaluation  of
available data, EPA believes that it is possible, though unlikely, that very
specific data might become available showing that a toxic, bioaccumulative
pollutant is very short-lived in  the aquatic environment.  For example,  the

-------
                           Section II: Regulatory Requirements                        65
pollutant may be highly volatile and escape to the atmosphere.  In this
situation, it would not be necessary to have a full range of data on the
cumulative effect of relevant fate and effect pathways for comparison with
other pollutants.  Rather, the specific data could be used to determine that
the pollutant is not persistent.  For this reason, EPA has amended the
definition of BCC in the final Guidance to include the qualification that
"chemicals with half-lives of less than eight weeks in the water column,
sediment, and biota are not BCCs."  The definition conforms with the Great
Lakes Water Quality Agreement, which defines persistent toxic substances as
any substance with a half life in water greater than eight weeks.

      EPA does not agree with comments that highly persistent pollutants
should be subject to the special provisions developed for BCCs even if they
are not highly bioaccumulative.  First, as discussed above, the special
provisions for BCCs in the final Guidance are designed to ensure that exposure
to humans and wildlife from BCCs is minimized.  The most likely fate for many
persistent but not highly bioaccumulative pollutants is to be deposited in
sediments, where they will likely remain for long periods.  The potential for
exposure to humans and wildlife while possible is diminished because the
contaminants do not bioaccumulate and in many cases are buried in the
sediments.

      Second, the threat these pollutants pose to benthic and other aquatic
organisms that come in direct contact with the sediment is being addressed
through other approaches.  Benthic organisms are represented in the
methodology for development of criteria to protect aquatic life.  In addition,
the potential toxicity to benthic organisms from desorption of pollutants from
sediment is addressed in existing State programs on a case-by-case basis
through implementation of narrative criteria.  EPA has also developed a
methodology which, when finalized, will be available to assist States and
Tribes in addressing the potential toxicity to benthic organisms more
systematically.  In January 1994 EPA published a notice announcing the
availability of proposed national sediment quality criteria for the protection
of benthic organisms, guidelines for deriving these criteria on a site-
specific basis, and the technical basis for deriving the criteria (59 FR 2652,
January 18, 1994).  EPA is analyzing the comments received in response to the
notice, and will be developing final sediment quality criteria based on the
analysis.  When the.methodology is finalized, it will be available for EPA,
States, and Tribes to develop sediment quality criteria for the protection of
benthic organisms.  This approach is scientifically more appropriate for the
control of persistent but not highly bioaccumulative pollutants than the
special provisions developed for BCCs in the final Guidance.  The BCC
provisions were designed to reduce loadings, not to specifically achieve
protective levels of contaminants in sediments.

      Third, it is reasonable to limit application of the special BCC
provisions to highly bioaccumulative pollutants.  The special provisions for
BCCs and the methodology for defining these pollutants were developed by the
senior water program managers in the eight Great Lakes States and three EPA
Regional Offices.  These managers selected this approach based on their many
years of regulating pollutants, including direct experience in the Great Lakes
basin.

      EPA agrees with comments that recommended considering the sediment route
of exposure in predicting BAFs, and partially agrees with comments that
recommended that only field-measured BAFs, not predicted BAFs, should be used
in determining BCCs.  As a result, EPA has added the field-measured biota-
sediment accumulation factor  (BSAF), which considers the sediment route of
exposure, to the hierarchy of methods for deriving BAFs; has selected a new
BAF model that better accounts for chemical uptake through sediments; and has
modified the Tier I minimum data requirements for human health and wildlife
criteria to specify minimum bioaccumulation data.  The definition of BCCs has
also been revised to be consistent with the above changes.  The final Guidance
now specifies that the minimum BAF information needed to define an organic

-------
66     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

chemical as a BCC is either a field-measured BAF or a BAF derived using the
BSAF methodology, and that the minimum BAF information needed to define an
inorganic chemical, including an organometal, as a BCC is either a field-
measured BAF or a laboratory-measured BCF.  The reasons for these changes as
well as other changes affecting the derivation of BAFs are discussed in
section IV.B.2 of this document.

      In response to a number of comments on the BAF cutoff level for defining
a BCC, EPA has reviewed all of the information and policy .considerations in
selecting the cutoff level.  As a result, EPA has made the risk management
decision to retain the proposed BAF cutoff level of 1000 for defining BCCs.

      EPA does not agree with the comments that the selection of a BAF  cutoff
level of 1000 for defining BCCs is arbitrary.  EPA believes that this comment
may have resulted from a confusion about the nature of risk management
decisions.  As EPA explained in the preamble to the proposal (58 FR 20844),
the selection of a BAF cutoff level is a risk management decision that
involves weighing information and policy considerations, rather than a  risk
assessment assumption that results solely from a scientific analysis.   It is
not possible, therefore, to specify a mathematical formula or systematic
algorithm employing environmental data to select a cutoff level.

      EPA weighed a wide range of information and policy considerations in
this decision.  These include the following considerations:

      --  The cutoff level for a BAF should include pollutants that are
"highly bioaccumulative" based on exercise of reasonable best professional
scientific judgment.  EPA believes that a BAF of over two or three orders of
magnitude  (100 or 1000) would meet this definition.

          The cutoff level should be high enough that it includes most
pollutants for which the fish consumption pathway is the most important route
of exposure for humans and wildlife.  To select a lower level would cause a
scientist using reasonable best professional judgment to question whether the
pollutant was "highly bioaccumulative."

          The cutoff level should be low enough to include those persistent,
bioaccumulative pollutants already found to be causing significant
contamination, including 2,3,7,8-TCDD (dioxin), mercury, and PCBs.
                   •
          The cutoff level should be low enough to provide adequate assurance
that other chemicals that could potentially contaminate the food web of the
Great Lakes ecosystem in the future are subject to the special provisions for
BCCs.

      --  The cutoff level should not be set so low that the regulatory and
administrative structure in place in the Great Lakes States and Tribes  for
control of discharges to waters would be overwhelmed.

          The cutoff level should be developed after consideration of the
scientific, policy,'administrative, and technical input from stakeholders in
the Great Lakes basin--State regulators, regulated community, and public
interest groups.

      EPA has determined that the cutoff level of 1000 initially selected by
the GLI Steering Committee meets all of the above considerations.  As
explained in the preamble to the proposal, a pollutant with a BAF greater than
1000 was believed by the Steering Committee to have a high potential to be
found in aquatic organisms of the Great Lakes System and therefore to have the
potential to cause a significant risk to the health of the aquatic life and
consumers of the aquatic life such as wildlife and humans inhabiting the Great
Lakes basin.  The Steering Committee made its recommendation on  the basis of
information available to them as managers of water quality programs.

-------
                           Section II: Regulatory Requirements                        67
      EPA recognizes that other numbers meeting the above considerations could
have been selected as a cutoff.  For example, several commenters suggested
lowering the cutoff to 250.  Others suggested a cutoff of 100, and others
suggested including all pollutants in Table 6 regardless of BAF.  None of
these commenters, however,' provided reasons other than increasing the number
of pollutants to be treated as BCCs as high as possible to avoid as much risk
as possible.  In other words, none of these commenters provided a rationale
that would prefer one "low" cutoff over any other.

      On the other hand, a few commenters suggested raising the cutoff to
10,000 or 100,000.  They appear to believe the cutoff should be as high as
possible without exceeding the BAF of the least bioaccumulative pollutant
currently known to cause problems in the basin.  EPA believes there are
currently known "problem" BCCs, such as lindane, with BAFs lower than 10,000.
EPA believes that it is reasonable and appropriate to retain the proposed
cutoff of 1000 not only to avoid excluding such pollutants, but also to
prevent adverse inputs from additional bioaccumulative chemicals in the
future.  Past discharges of highly bioaccumulative pollutants have resulted in
contamination of the Great Lakes System that is taking decades to subside.
EPA believes it is reasonable and appropriate to prevent this from happening
with other chemicals in the future.

      EPA does not agree with comments that EPA should solicit formal public
comment before States or Tribes treat any additional chemicals as BCCs in the
future.  EPA believes that the States and Tribes should have the ability to
designate additional chemicals for BCC controls based on information available
to them without waiting for EPA to act.  As discussed above, EPA will operate
the GLI Clearinghouse as a means to share pollutant information, including
BAFs, as quickly as possible.  If new information becomes available showing an
organic chemical to have a field-measured BAF of over 1000, for example, this
information would be reviewed by EPA and other Clearinghouse participants and
placed in the Clearinghouse, where States and Tribes would be alerted.  States
and Tribes would be able to apply the special BCC provisions to the pollutant
after following their applicable State or Tribal public review procedures for
revisions to water quality standards or for permit development.  For example,
the State or Tribe could include a description of the special BCC provisions
in the public notice for a NPDES permit.  EPA believes this would be a more
efficient approach than relying in all cases on EPA to sponsor a public review
and comment process, which has often taken several years for similar types of
actions.

      EPA has some concern that inconsistencies could arise among States and
Tribes concerning future identification of BCCs under the above approach.  EPA
believes operation of the Clearinghouse will minimize this possibility.
Nevertheless, if serious inconsistencies arise, EPA may from time to time
publish available BAF data for a pollutant and solicit public comments.  EPA
could then issue final technical assistance and recommendations concerning the
pollutant to assist State and Tribal revisions to water programs.

      c.    Final Guidance.  EPA has revised the definition of a BCC in
§ 132.2 of the final Guidance to exclude chemicals that do not have the
potential to cause adverse effects.  EPA has also modified the proposed
definition to exclude chemicals with half-lives of less than eight weeks in
the water column, sediment, and biota.  Finally, the definition specifies that
the human health BAF for non-metals must be derived from a field-measured BAF
or a field-measured BSAF.  EPA has retained all other features of the proposed
definition,  including the BAF cutoff level of 1000.  The revised definition
reads:

      BCC is any chemical that has the potential to cause adverse effects
      which, upon entering the surface waters, by itself or as its toxic
      transformation product, accumulates in aquatic organisms by a human
      health bioaccumulation factor greater than 1000, after considering
      metabolism and other physicochemical properties that might enhance or

-------
68     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      inhibit bioaccumulation, in accordance with the methodology in appendix
      B of this part.  Chemicals with half-lives of less than eight weeks in
      the water column, sediment, and biota are not BCCs.  The minimum BAF
      information needed to define an organic chemical as a BCC is either a
      field-measured BAF or a BAF derived using the BSAF methodology.  The
      minimum BAF information needed to define an inorganic chemical,
      including an -organometal,  as a BCC is either a field-measured BAF or a
      laboratory-measured BCF.  BCCs include, but are not limited to, the
      pollutants identified as BCCs in section A of Table ,6 of this part.

      EPA has applied the above changes in the definition to calculate BAFs
for individual pollutants listed in Table 6.  The results are included in the
Great Lakes Water Quality Initiative Bioaccumulation Factor Technical Support
Document, which is available in the docket for this rulemaking.  These
calculations have resulted in the deletion of six chemicals from the proposed
list of pollutants in Table 6A:  aldrin, 4-bromophenyl phenyl ether, endrin,
heptachlor, heptachlor epoxide,  and methoxychlor.  For these six chemicals
there were insufficient data to develop a field-measured BAF or a BAF
predicted from a field-measured BSAF.  BAFs based on laboratory-measured BCFs
are available for endrin, heptachlor, heptachlor epoxide, and methoxychlor;
BAFs based on a laboratory-predicted BCF are available for aldrin and 4-
bromophenyl phenyl ether.

      As a result, the final Guidance lists 22 BCCs in Table 6A to part 132.
They represent EPA's best scientific judgment at this time concerning which
pollutants on Table 6 meet the final definition of BCCs.  States and Tribes
may determine, however, that additional pollutants meet the definition or
should be subject to the special provisions for BCCs.  If so, they should take
appropriate regulatory action as discussed above to treat the chemicals as
BCCs in their water quality standards and NPDES permit programs, including the
special provisions for BCCs contained in the final Guidance.

      EPA has underway field testing programs to develop field-measured BAFs
and field-measured BSAFs for additional pollutants, including the six
chemicals deleted from the proposed list of BCCs.  It is possible that some of
this testing may result in identification of additional pollutants that meet
the definition of BCC.  When this testing is complete and BAFs have been
calculated, EPA will place the results in the GLI Clearinghouse where they
will be available for States to use in making their determinations.

9.    Potential Bioaccumulative Chemicals of Concern

      a.    Proposal.  The proposal listed  ten chemicals as potential BCCs in
section B of the proposed Table 6.  Although the ten chemicals had BAFs of
greater than 1000 based on predicted BCFs  (eight chemicals) and on laboratory-
measured BCFs  (two chemicals), other available information described in the
proposal to the preamble raised serious doubts that the actual BAFs for these
chemicals exceeded 1000.  For this reason,  EPA proposed that the special
provisions for BCCs would not apply to these ten pollutants.  EPA invited
comment on whether any or all of the potential BCCs should be listed as BCCs
and any additional data relevant to these determinations.

      b.    Comments.  Many  comments recommended that EPA delete the list  of
potential BCCs and not apply the special BCC provisions to  these chemicals.
These commenters asserted that the ten potential BCCs are readily metabolized
and do not bioaccumulate at  high rates.

      Some comments recommended  that several polynuclear aromatic hydrocarbons
 (PAHs) on the  list of potential BCCs should be treated  as BCCs, asserting  that
these chemicals  have relatively high BAFs when measured in  lower trophic level
species, and therefore pose  a risk to  wildlife or people eating those  species,
even though they are metabolized by higher  trophic  level species and therefore
have relatively  low' human health BAFs.

-------
                           Section II: Regulatory Requirements                        69
      EPA agrees that the potential BCCs should not be treated as BCCs, and
should be deleted from the final Guidance.  For the reasons discussed above  in
section II.C.8 of this document, EPA has modified the definition of BCC in the
final Guidance to use only the two methods of developing BAFs for non-metals
that take into account metabolism.  None of the ten potential BCCs have BAFs
available using these two methods, and therefore they are not listed as BCCs
in Table 6.  Furthermore, as discussed in the preamble to the proposed
Guidance, there are indications that the ten chemicals would likely not have
BAFs exceeding 1000 if metabolism were taken into account,.  If reliable data
were to become available in the future showing a BAF of over 1000 for any of
these ten chemicals or other chemicals based on a field-measured BAF or BSAF,
then States and Tribes would need to apply the special BCC provisions to these
chemicals.

      EPA does not agree that the PAHs with high BAFs at low trophic levels
should be treated as BCCs at this time.  The special provisions for BCCs are
designed to address pollutants that accumulate in the food web of the Great
Lakes ecosystem.  Pollutants that accumulate in lower trophic levels but which
are metabolized at higher levels are not as likely to affect the food web as a
whole as those that continue to accumulate at higher trophic levels.  The
definition of a BCC in the final Guidance therefore does not include any
special procedures that would classify such a pollutant as a BCC.  Such a
pollutant would, however, become a BCC if data were to become available in the
future resulting in a human health BAF of over 1000.

      EPA established the category of "potential BCC" in the proposed Guidance
primarily to obtain comment on whether such pollutants met the definition of
BCCs and should, accordingly, be subject to the special provisions for that
class of pollutants.  Since EPA has now completed its analysis of comments and
determined that potential BCCs do not appear to satisfy the final definition
of BCC, and since no special provisions in the Guidance apply to potential
BCCs that do not apply to other pollutants, there is no further purpose for
retaining the list of potential BCCs in the final Guidance.  Therefore, the
proposed list of potential BCCs has been deleted.

      c.    Final Guidance.  For the reasons discussed above, the proposed
list of potential BCCs in section B of Table 6 of part 132 has been deleted.
Section C of Table 6 has been redesignated as section B, Pollutants That Are
Not BCCs.

10.   Pollutants of Initial Focus

      a.    Proposal.  As described in the preamble to the proposed Guidance
(58 FR 20843-44), 138 pollutants were identified by the Great Lakes Initiative
Steering Committee and listed in Table 6 of the proposed Guidance as the
pollutants of initial focus in the Great Lakes Water Quality Initiative.  The
138 pollutants in proposed Table 6 included: (a) the 126 pollutants identified
by EPA as priority toxic pollutants in appendix A of 40 CFR 423; (b) selected
pollutants listed in the Great Lakes Water Quality Agreement of 1978 (as
amended by the Protocol signed November 18, 1987) ; (c) certain pollutants
categorized under the Lake Ontario Toxics Management Plan and the Niagara
River Toxics Management Plan; and (d) three pollutants included on a case-by-
case basis.

      The primary purpose of the Initiative Committees in specifying
pollutants in Table 6 was to provide an initial focus for criteria development
and the calculation'of bioaccumulation factors in the Great Lakes System.  The
pollutants included in Table 6 were not intended to be a comprehensive
inventory of all pollutants present, used,  manufactured, or stored in the
Great Lakes System, but were thought to represent pollutants which may be of
concern in the Great Lakes, and for which adequate effluent, ambient, and
toxicity data would be available to assess the impact of the various options
considered in developing the Guidance.

-------
70     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      The proposed Guidance provided an initial focus on the Table 6
pollutants in the following three ways:  First, the pollutants for which EPA
and the States applied the proposed criteria methodologies to derive numeric
water quality criteria--that is, the pollutants in Tables 1 through 4--were
selected from the list of pollutants in Table 6.  Second, EPA and the Great
Lakes States calculated bioaccumulation factors (BAFs) for the Table 6
pollutants to.assist EPA and the States in developing criteria and values to
protect human health and wildlife.  Third, the Table 6 list of pollutants is
one factor used in determining when States, Tribes, and/or permittees need to
generate data necessary to calculate Tier II values used in developing water
quality-based effluent limits.  Comments on the use of Table & in the data
generation provisions of procedure 5 of appendix F are discussed in section
VIII.E of this document.

      b.    Comments.  Some comments suggested additional pollutants to be
added to Table 6, including: pollutants listed in the Great Lakes Water
Quality Agreement, Annex 10, Appendix l (Hazardous Polluting Substances) or
Appendix 2  (Potentially Hazardous Polluting Substances) ,- high-use pesticides;
persistent pollutants that do not bioaccumulate; all drinking water
contaminants as listed in the 1986 amendments to the Safe Drinking Water Act;
pollutants categorized as ID or IE in the Categorization of Toxics in Lake
Ontario or in the Categorization of Toxic Substances in the Niagara River; all
the dioxins, furans and PCBs that are known to operate via the Ah receptor in
any animal species,- as well as a number of specific pollutants listed by
commenters.

      Some comments suggested that Table 6 should be limited to chemicals
which are causing demonstrated impacts to water quality in the Great Lakes
System.  Other comments suggest limiting Table 6 to pollutants which pose
water quality concerns unique to the Great Lake system, such as the list of
critical pollutants being developed by Lakewide Management Plans.  Other
comments recommended that in the interest of establishing reasonable
priorities for the Great Lakes States, EPA should eliminate non-BCCs from
Table 6.  Other comments recommended removing specific pollutants from Table
6, such as asbestos, fluoride, methylene chloride, phenol, phthalate esters,
and silver.

      Some comments expressed concern that the proposed Guidance did not
include clear procedures on how additional toxic pollutants that are
introduced or discovered in the Great Lakes Ecosystem will be added to Table
6, or be regulated prior to formal revision of the Guidance.

      EPA does not agree that additional pollutants should be added to Table
6.  EPA believes it would be counterproductive to expand the pollutants of
initial focus to a significantly larger set of pollutants.  Table 6 has
already been used successfully during the development of the proposed and
final Guidance to focus development of data needed to implement the final
Guidance.  As a result of EPA and State efforts, many of the data gaps that
existed for these pollutants at the start of the Great Lakes Water Quality
Initiative have been filled.  Applying limited EPA, State, Tribal, and
discharger resources at this time to address a broader list of pollutants in
the ways described under the Proposal section above would divert resources
away from other important actions, such as developing a richer data base
concerning effects of pollutants  in the proposed Table 6 on endangered or
threatened species.

      EPA also does*not agree with comments  that the list of 138 should be
reduced.  All of the 138 pollutants have been identified as either priority
pollutants under the CWA or as pollutants of specific concern in the Great
Lakes basin.  EPA also does not agree that Table 6 should be limited to BCCs.
BCCs are not the only types of pollutants currently or potentially affecting
the Great Lakes ecosystem.  For the same reasons, EPA does not agree with
comments that specific pollutants be removed from the  list.

-------
                           Section II: Regulatory Requirements                        71
      EPA recognizes that the final Guidance does not include procedures for
adding additional pollutants to Table 6.  EPA believes that the GLI
Clearinghouse can be used as a forum for determining additional needs for BAF
calculations, and for sharing of BAF results.  If it should become apparent in
the future that adding pollutants to Table 6 would assist in reducing
disparities between' data generation approaches of the States and Tribes under
procedure 5.C of appendix F, EPA would consider recommending that States and
Tribes expand the Table 6 lists that they have adopted into their programs to
be consistent with the final Guidance.  This could be done during the
triennial review of water quality standards programs under section 303 of the
CWA.

      c.    Final Guidance.  The final Guidance retains Table 6 as proposed.

      EPA would like to clarify that the methodologies and procedures in the
final Guidance generally apply to all pollutants, except for the pollutants in
Table 5.  The Table 6 list of pollutants is one factor used in determining
when States, Tribes, and/or permittees need to generate data necessary to
calculate Tier II values used in developing water quality-based effluent
limits.

D.    Procedures for Adoption and EPA Review

1.    Adoption Procedures

      a.    Proposal.  Proposed § 132.5(a) required the Great Lakes States and
Tribes to adopt and-submit for EPA review and approval the criteria,
methodologies, policies and procedures developed pursuant to part 132 by a
date no later than 18 months from the date of final publication of the part
132 requirements.  EPA proposed the 18-month deadline in order to allow the
full time available under the statute for EPA review and approval of
submissions and for States and Tribes to correct any identified deficiencies,
and still allow EPA to meet the section 118(c)(2)(C) requirement for review,
approval or disapproval and promulgation by EPA, if necessary, within two
years after the final publication of the Guidance.  Proposed § 132.5(d) also
provided a 30-day public comment period on the submissions.

      If a Great Lakes State or Tribe failed to submit criteria,
methodologies, policies, and procedures to EPA for review, or if EPA
disapproved portions of all of a State or Tribal submission because it was
inconsistent with part 132, proposed § 132.5 provided that the requirements of
part 132 would apply to discharges within the State or Federal Indian
Reservation upon EPA's publication of a final rule in the Federal Register
indicating the effective date of the part 132 requirements in the identified
jurisdictions.

      b.    Comments.  A few commenters recommended that the Guidance be
adopted as soon as possible because they believe it is a major step forward in
efforts to protect the Great Lakes and fulfill promises of the U.S.-Canada
Great Lakes Water Quality Agreement and the Great Lakes Governors' Toxic
Substances Control Agreement.

      Some commenters indicated that the two-year adoption period was too
short.  Other commenters recommended that EPA eliminate the 18-month deadline
for State submittals, or allow extensions to the deadline on a case-by-case
basis as long as the State and EPA agree on a schedule that will meet the
statutory deadline.  Several commenters stated that the public comment
requirements during EPA's review of State submittals may not be necessary, and
are inconsistent with the current national program.

      EPA agrees that States and Tribes should adopt provisions consistent
with the Guidance as soon as possible.  EPA has reviewed the steps States,
Tribes,  and EPA must take to develop, adopt,  approve, and if necessary
disapprove and promulgate the Guidance, however, and has concluded that it

-------
72     Water Quality Guidance for the Great Lakes System - Supplementary Information Document

would not be possible nor practical to complete these steps in less time than
the two years specified in section 118(c) (2) (C) of the CWA.  EPA has therefore
not shortened the deadline in the final § 132.5(a).  EPA encourages States and
Tribes to accelerate their adoption and submission processes as much as
possible, and as discussed further below, EPA Regional Offices will work with
States and Tribes to facilitate early adoption and approval where possible.

      EPA recognizes that the two-year deadline specified by the Congress may
be extremely ambitious.  EPA's experience in reviewing State water quality
standards adopted under section 303(c)  indicates that many of the Great Lakes
States have had difficulties adopting even routine water quality standards
revisions within three years as specified for triennial reviews in the CWA.
Nevertheless, EPA believes the States,  Tribes, and EPA Regional Offices can
work together to meet the deadline, especially since the majority of
provisions of the final Guidance were developed by the States themselves as
part of the Great Lakes Water Quality Initiative and many are currently in
effect in one or more of the States.  Furthermore, the two-year deadline is
established explicitly by section 118(c)(2)(C) of the CWA.  Therefore, the
final Guidance retains the two-year deadline for final approval or
promulgation.

      EPA agrees with comments, however, that flexibility may be appropriate
in some cases concerning the eighteen-month deadline for State and Tribal
submissions to EPA under § 132.5.  There may be situations where the EPA
Regional Office has worked closely with the State or Tribe during their
development and adoption of provisions consistent with the Guidance and
believes that the full six month period for EPA review of the submission will
be unnecessary.  For example, a State may decide to adopt the provisions of
the Guidance with only minor modifications, and may have early indications
that the public supports this approach.  In this situation, EPA believes that
an extension to the eighteen-month deadline would be reasonable because EPA
would have a high degree of assurance that it would still be able to review
and approve the submission within the two-year statutory deadline.  Therefore,
EPA has added § 132.5(c) that allows the EPA Regional Administrator to extend
the deadline for a State or Tribal submission beyond 18 months if the Regional
Administrator believes that the submission will be consistent with the
requirements of the.final Guidance and can be reviewed and approved within the
two-year deadline.  In these cases, EPA expects that the Regional
Administrator will need to reach early agreement with the State or Tribal
director on a joint schedule for specific steps that the State or Tribe and
EPA will take over the two-year period to meet the requirements of § 132.5.
To ensure success, the EPA Regional Offices will likely need to provide
technical guidance and assistance to the States or Tribes early in the
adoption process, and will need to begin analyzing drafts of the State or
Tribal submission even before it is formally  submitted to determine whether
provisions are consistent with the final Guidance.

      EPA also agrees in part with comments  that the proposed requirement  for
a 30-day comment period was unnecessary and  should be revised.  The proposed
provision for public comment was patterned  in general on existing minimum
public comment requirements for proposed modifications to State NPDES programs
under 40 CFR 123.62(b) and State water quality standards programs under part
131.  Section 123.62(b) requires a public comment period on State NPDES
submissions only if the proposed program modification is substantial.
Furthermore, section 303 of the CWA and part  131 do not require EPA to provide
public review and comment on EPA  approvals  of State water quality standards
submissions.  Citv of Albuquerque v. Browner.  38 ERG 2062  (DNM  1993) .  EPA
believes that it is.not necessary or appropriate to impose more extensive
public notice-and-comment requirements  for  submissions under part 132 than for
the submissions  in the underlying NPDES  and water  quality  standards programs
affected by part 132.  EPA also believes  that EPA  Regional Offices will be
able to determine whether a  specific provision in  the submission constitutes a
substantial modification to  the underlying  State program without great
difficulty  because of their  familiarity with the existing  State programs.

-------
                           Section II: Regulatory Requirements                        73
Therefore, EPA has modified proposed § 132.5(d), now redesignated  §  132.5(e),
to conform with the minimum public notice and comment requirements of parts
123 and 131.

      One commenter urged that phase-in periods for implementing certain
provisions  (e.g., the ban on mixing zones for BCCs) be dated from  the time of
final federal publication of the Guidance, not the date of State and Tribal
final adoptions.  EPA does not agree with this comment.  Section 118(c) (2) (C)
requires States to adopt provisions consistent with the final Guidance by no
later than two years from date of publication, or be subject to EPA
promulgation within that two-year period.  Because the provisions  will not be
effective until they are promulgated by a State, Tribe, or Federal agency, EPA
believes that it is reasonable to delay the computation of any phase-in period
until the statutory deadline for such promulgation.  Accordingly,  all phase-in
periods and similar provisions that distinguish between activities occurring
before or after a specified date  (e.g., the definition of "new Great Lakes
discharger") are calculated from the date two years after publication of this
final Guidance.

      c.    Final Guidance.  Section 132.5(a) of the final Guidance  retains
the general deadline of eighteen months after publication of the final
Guidance, or September 23, 1996, for States and Tribes to submit criteria,
methodologies, policies, and procedures developed pursuant to the  Guidance.

      The final Guidance has been modified to add new § 132.5(c) which allows
the Regional Administrator to extend the deadline for the submission required
in § 132.5(a) if the Regional Administrator believes that the submission will
be consistent with the requirements of the final Guidance and can  be reviewed
and approved no later than March 23, 1997.  As discussed under the Comments
section above, EPA expects that before granting an extension the Regional
Administrator will reach agreement with the State or Tribal director on a
joint schedule for specific steps to meet the requirements of § 132.5.  The
joint schedule should include opportunities for EPA to review drafts of State
or Tribal provisions in order to facilitate EPA review and approval.  In these
cases, EPA intends to provide technical guidance and assistance to the State
or Tribe throughout the adoption process, beginning as soon as possible.
Conforming provisions have been added to § 132.5(a), and proposed  §  132.5(c)
has been redesignated as § 132.5(d).

      The provisions for public notice and comment in proposed § 132.5(d) have
been redesignated as § 132.5 (e) and modified to provide for public notice and
at least 30 days' public comment only in situations where the State  submits
substantial modifications to its NPDES program.

      The provisions for EPA approval and disapproval in proposed  §  132.5(d)
have been moved to new § 132.5 (f) .

      In reviewing proposed § 132.5, EPA noted that it would be
administratively impossible for EPA to provide the 30-day comment  period on
State submissions and still issue a notice of approval of a State  or Tribe's
submission within 60 days of receipt of the submission as proposed in
§ 132.5(d)(1).  The proposed 60-day deadline was based on the similar deadline
in section 303(c) of the CWA for approving State water quality standards.  The
section 303(c) approach does not include a requirement for public  notice and
comment,  however.  Since the final Guidance includes a minimum of  30 days for
such a requirement in cases where a State submits a substantial modification
to its NPDES program, EPA has modified the final Guidance to add 30  days to
the deadline for EPA approval.  Final § 132.5(f)(1) now requires EPA to issue
a notice of approval of a State or Tribe's submission within 90 days of
receipt of the submission.

      Proposed § 132.5(e) has been redesignated as § 132.5(g).

-------
74     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      A new § 132.5(h) has been added concerning results of consultation under
the Endangered Species Act.  It is discussed further in section II.G of this
document.

      Proposed § 132.5 (f) has been redesignated as § 132.5 (i).

2.     Interpretation of "Consistent With"

      a.    Proposal.  Section 118(c)(2)(C) of the CWA requires Great Lakes
States to adopt water quality standards, antidegradation policies, and
implementation procedures consistent with the final Guidance, or EPA is
required to promulgate for the States.   Section 132.5(e) of the proposed
Guidance specified the conditions under a State or Tribe submission is
consistent with the requirements of part 132.  Generally, the proposed
Guidance provided that submitted criteria, methodologies, policies and
procedures would be consistent with part 132 if they are "equal to or more
restrictive than" the provisions in the final Guidance.  Proposed
§ 132.5(e)(3) clarified EPA's intention to evaluate the State and Tribal
submissions on a provision-by-provision basis by providing that if States or
Tribes adopt provisions more restrictive than the final Guidance, the more
restrictive provision may not be offset by relaxation of other specific
elements of the final Guidance.

      b.    Comments.  Many comments urged EPA to strictly interpret the
"consistent with" requirements, and to require States and Tribes to justify
deviations from the Guidance, in order to achieve the desired standardization
of programs basin-wide.  Several comments urged EPA to require explicitly that
all State and Tribal procedures and criteria be consistent with and no less
stringent than Guidance procedures and criteria.  A few comments said States
should be directed to retain existing numeric water quality criteria and
procedures where they are more stringent than the final Guidance, in order to
maintain continual progress toward zero discharge as outlined in the Great
Lakes Water Quality Agreement and the CWA.

      Some commenters interpreted the CPA as directing EPA to publish guidance
to the states, rather than a fixed rule, and therefore urge EPA to give the
States a measure of  latitude in implementing the Guidance.  Some commenters
argued that EPA should not reduce State flexibility currently available in
States' implementation of section 303(c) of  the CWA.

      Some commenters believed that State flexibility is needed to enable
cost-effective implementation of the Guidance, to maintain State primacy as
provided in the CWA,  to minimize the administrative burden on the States, to
reflect the innovative nature of the Guidance, and because States are more
familiar with the environmental conditions  in local areas  and are better able
to work effectively  with dischargers to address problems.  Some commenters
urged additional state flexibility to ensure that future improvements in
science are readily  incorporated into the Guidance.  Some  comments stated that
more flexibility is  needed to reflect ecological differences between areas in
the Great Lakes basin which  require  different regulatory responses.  Some
commenters recommended that  since many  real-world permitting situations
probably exist which preclude a straightforward application of the proposed
implementation procedures, EPA should to  the maximum extent possible structure
the final Guidance to function truly as  "guidance" for  the States.

      Some comments  recommended that where  the overall  outcome is  the same,
EPA should allow a state to  be less  stringent in one area  of the methodology
if it is more stringent  in another.  Some  commenters suggested that  EPA should
interpret  "consistent with"  as meaning  "substantially  equivalent  to"  so that
States  and Tribes may adopt  criteria and procedures tailored to  their own
circumstances so long as the overall impact of the program is to treat
substantially similar- sources  in a substantially equivalent  manner.

-------
                           Section II: Regulatory Requirements                        75
      EPA does not agree that the amendments to section 118(c)(2) of the CWA
directed EPA simply to publish non-binding guidance.  This section not only
directs EPA to provide guidance to the Great Lakes States on minimum water
quality standards, antidegradation policies, and implementation procedures for
the Great Lakes System, but also requires the States to adopt water quality
standards, antidegradation policies, and implementation procedures for waters
within.the Great Lakes System which are consistent with such guidance or'be
subject to EPA promulgation  (section 118(c)(2)(C)) .  EPA believes that whether
States and Tribes adopt minimum standards, policies, and procedures consistent
with the final Guidance, or whether EPA promulgates them, the Congress
intended that the final Guidance would establish minimum, and ultimately,
enforceable requirements for the Great Lakes System.

      This interpretation of section 118(c)is supported by the primary authors
of the Critical Programs Act.  In a June 9, 1994, letter to Governor Cuomo
discussing this issue, Senators Levin, Glenn, and Kohl emphasized the need to
provide enforceable requirements through a federal regulation in order to
improve consistency in State water quality programs in the Great Lakes System.

      ...While flexibility is built into the law, the Great Lakes Guidance is
      nevertheless intended to be an enforceable federal
      regulation....Guidelines without enforceability cannot achieve the
      overarching goal of the Great Lakes Initiative to ensure consistent
      Great Lakes water quality protections throughout the basin.

            In light of the differences among our state water quality
      programs, the Great Lakes Guidance will no doubt require adjustments in
      each of our states.  But our states knew that when they launched the
      Initiative and committed themselves to developing a single set of rules
      for everyone to live by in the Great Lakes Basin.  In fact, reducing
      state disparities was a driving force behind the Initiative....

      This interpretation of section 118(c) is also supported by the extremely
short time frame specified in the statute for EPA promulgation of the Guidance
in the absence of State adoption.  As discussed above, provisions consistent
with the final Guidance must ultimately be adopted by Great Lakes States and
Tribes or be promulgated by EPA within two years.  This statutory deadline for
State adoption and EPA approval or promulgation of the Guidance does not allow
sufficient time for EPA to take additional steps, such as publishing the
proposed and final Guidance first, followed by proposing and finalizing a
separate rule governing EPA approval and promulgation.  Rather, the rule that
was proposed and is' now being finalized contains both the substance of the
Guidance,  including the criteria in Tables 1 through 4 and the methodologies,
policies,  and procedures in appendixes A through F, as well as the approval
and promulgation procedures that EPA will use for review of State and Tribal
submissions.

      EPA also does not agree that the CPA intended to afford States the same
flexibility currently available under section 303(c) of the CWA.  Section
303(c)  requires that States adopt and implement water quality standards to
help restore and maintain the chemical, physical, and biological integrity of
the Nation's waters.   EPA has periodically provided guidance to States in
implementing these requirements, including water quality criteria published
pursuant to section 304(a) of the CWA.  States have the flexibility to use
section 304(a)  guidance or other scientifically defensible approaches in
adopting and implementing water quality standards.  As discussed in the
preamble to the proposed Guidance (58 FR 20835), however, the Initiative
Committees believed this level of flexibility over the years had resulted in
significant differences in State adopted water quality standards,
antidegradation policies, and implementation procedures, as well as
inconsistencies in regulatory approaches and individual permit decisions in
the Great Lakes basin.  The Congress was aware of these inconsistencies and of
the efforts of the Initiative Committees to develop the Guidance when it
enacted the CPA.  The requirements in section 118(c)(2) described above

-------
76     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

represent the direction of the Congress to adopt or promulgate minimum  water
quality standards, antidegradation policies, and implementation procedures for
the entire" Great Lakes System.  In order to have any meaning, implementation
of these specific requirements for the Great Lakes System will necessarily
result in some reduction in the flexibility of the Great Lakes States.

      EPA agrees with comments that the final Guidance should require State
and Tribal submissions to be at least as protective as the final Guidance in
order to achieve the increased consistency and minimum threshold levels of
control intended by the Congress.  For this reason, EPA has retained the
proposed § 132.5 (e) with clarifying amendments, now redesignated as §
132.5(g), under which submitted criteria, methodologies, policies and
procedures will be considered consistent with part 132 if they are "as
protective as" the provisions in the final Guidance.  EPA believes that
specifying "as protective as" will ensure that provisions adopted by the
States and Tribes will be equivalent to or more protective than the final
Guidance.  Further, EPA believes that this is a reasonable and'appropriate
mechanism for implementing EPA's duty to define the "minimum" requirements for
water programs in the basin.

      EPA also agrees with comments, however, that it is appropriate to
provide reasonable flexibility to States and Tribes, to the extent that this
can be done and still meet the requirements and purpose of the CWA.  In
overseeing States' implementation of the CWA, EPA has found that reasonable
flexibility is not only necessary to accommodate unforeseen circumstances, but
is also appropriate to enable innovation and progress as new approaches and
information become available.  To address the need for flexibility, EPA made
several changes to the proposed Guidance.

      First, EPA reviewed all sections of the proposed Guidance and all
comments to determine the appropriate level of flexibility.  Based on this
review, the final Guidance provides increased flexibility for State and Tribal
adoption and implementation of these provisions in many areas, including
antidegradation, TMDLs, intake credits, site-specific modifications,
variances, compliance schedules, elimination of mixing zones for BCCs,  and the
scientific defensibility exclusion.  The final Guidance also provides reduced
detail of provisions in many areas, and provisions for the exercise of  best
professional judgment by the Great Lakes States and Tribes when implementing
many individual provisions.  This increased flexibility is discussed further
in section I of this document.

      Second, EPA clarified how  "offsets" will be considered between elements
of State and Tribal programs.  See discussion below.

      Third, EPA added  §132.1(b)  in the final Guidance to clarify that
verbatim adoption of the Guidance is not required.  EPA believes that State
and Tribal programs do not need to be identical to the Guidance  to result in
equally protective criteria, methodologies, policies, and procedures.   States
and Tribes instead have flexibility to adopt alternative provisions as  long as
they are as protective  of human health, wildlife, and aquatic life in the
Great Lakes System as the Guidance.

      EPA believes that the flexibility available in the final Guidance will
satisfy many of the specific concerns of commenters.  For example, based  on
their familiarity with  environmental  conditions in local areas,  States  and
Tribes may use the site-specific  modification procedures to adjust criteria
and values in order-to  reflect ecological differences from site  to site.
States and Tribes also  have flexibility  to  adjust criteria, methodologies, or
procedures to reflect future improvements in scientific understanding.  For
example, States and Tribes may use the scientific defensibility  exclusion in
§132.4(h) to apply a new, alternative methodology or implementation procedure
 (see section II.C.6 of  this document), or they may adjust elements within a
Guidance provision as long as the provision as a whole  is consistent with the
final Guidance  (see discussion of "offsets" below in this section).

-------
                           Section II: Regulatory Requirements                        77
Additionally, as discussed in section II.C.I of this document, there are
several steps within the criteria methodologies where some flexibility  is
available to reflect new findings and data.

      EPA understands that the science of risk assessment utilized  in
environmental protection, including criteria methodologies, is rapidly
evolving and improving.  Therefore, to ensure that the scientific basis for
the methodologies in appendices A through D is always current and peer
reviewed, EPA will review the methodologies and revise them as appropriate
every three years.

      The final Guidance is not, by itself, enforceable.  Provisions
consistent with the Guidance will become enforceable only when adopted  by a
State or Tribe as part of its NPDES or water quality standards programs,
promulgated by EPA in the absence of State or Tribal action, or when included
in a NPDES permit.

      EPA does not agree with comments that States should be directed to
retain existing numeric water quality criteria that are more stringent  than
the final Guidance.  Although under § 132.4(i) States may choose to retain
more stringent criteria, the final Guidance does not require them to do so.
The CPA requires the Guidance criteria to protect human health, aquatic life,
and wildlife.  EPA and the Initiative Committees have designed the  criteria
methodologies to meet this requirement.  Therefore, while the development of
more stringent provisions may be necessary because of site-specific conditions
within their jurisdiction and is also available as an option for States and
Tribes under any circumstances, automatic retention of existing more stringent
criteria is not required by the CPA.  For this reason, the final Guidance
generally requires only that the States and Tribes adopt and use criteria and
values that are as protective as those produced by the Guidance methodologies.

      EPA also does not agree with recommendations that the decision whether
State or Tribal submissions are "consistent with" the final Guidance should be
based only on the overall outcome of applying all of the Guidance provisions.
Such an approach, which would allow "offsets" between separate provisions,
would be technically and administratively unworkable for the following
reasons.  Because of the differing nature of each methodology and procedure
and differences in the site-specific characteristics of each discharge  and
discharge location, it would be administratively difficult, if not  impossible,
to quantify and compare numerically the effects of allowing more stringent
changes in one methodology or procedure, for example, to "offset" less
stringent changes in different methodologies or procedures.  The only
potential way to make a valid comparison would be to apply the full set of
Guidance methodologies and procedures to specific cases, such as developing
water quality-based effluent limits for a range of NPDES discharges to a range
of receiving waters, and comparing those limits with limits derived using
alternative State or Tribal procedures for the same dischargers.  Based on its
experience, EPA does not believe such analyses can produce useful results
within a reasonable time or with available resources.  Additionally, EPA does
not believe that it could undertake this lengthy and complex analysis within
the short time period specified by Congress in section 118(c)(2)(C) for review
and approval or disapproval and promulgation of the Guidance  provisions in
the eight Great Lakes States.  Finally, allowing offsets between provisions
(e.g.,  between the human health criteria methodology and the aquatic life
criteria methodology, or between the TMDL procedure and the mixing  zones
provisions),  would not ensure appropriate levels of protection for  human
health, wildlife, and aquatic life in the Great Lakes System.  For  these
reasons, EPA has retained the prohibition against offsetting changes between
different provisions, but has made editorial changes to clarify that
variations or offsets of elements within a particular provision are acceptable
as long as the submitted provision is consistent with the Guidance.

      c.    Final Guidance.  The final Guidance generally retains the proposed
conditions in § 132.5 (e), redes'ignated as § 132.5 (g), under which EPA will

-------
78     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

determine that the criteria, methodologies, policies, and procedures  in a
State or Tribal submission are consistent with the requirements  of  the  final
Guidance.

      EPA corrected.the header for §  132.5(g)(2) to refer to  "pollutants other
than those listed in Tables 1, 2, 3,  4, and 5."  In the proposal, reference to
Table 5 was omitted by error.

      EPA modified § 132.5(g)(2)(ii)(A) to refer to "numeric  criteria adopted
by the State into State water quality standards and approved  by  EPA prior  to
March 23, 1997."  EPA added the reference to EPA approval, and changed  the
date to reflect the two years allowed for State adoption of the  final
Guidance.  EPA found that these technical changes were needed to clarify the
intent of this section.

      For the reasons discussed in the Comments section above, EPA  made
editorial changes to clarify the general prohibition in § 132.5(g)(3) against
offsets between provisions.  The final Guidance states that "adoption of a.
more protective element in one provision is not justification for adoption of
a less protective element in another  provision of this part."  States and
Tribes are not precluded, however, from offsetting elements within  a
particular provision as long as the adopted provision is consistent with the
final Guidance.  For example, a State could not use a more protective element
in the human health cancer methodology to offset a less protective  element in
either the human health non-cancer methodology or the aquatic life
methodology.  In contrast, a State could use a more protective element  within
the human health cancer methodology   (e.g., cancer risk level assumption)  to
offset a less stringent element within the same methodology (e.g.,  fish
consumption value) as long as the resulting methodology and criteria  were  as
protective as the final Guidance.

      When determining whether criteria adopted by the State  or  Tribe comply
with § 132.5(g)(1) , EPA will consider not only whether the State or Tribal
criteria are numerically as protective as the criteria in Tables 1  through 4,
but also whether all other provisions of the Guidance pertaining to the
criteria have been fully incorporated.  For example, EPA will consider  under
procedure 1 of appendix F whether more stringent criteria are needed  to
protect a threatened or endangered species.  Additionally, if a  State or Tribe
submits site-specific criteria less stringent than the criteria  in  Tables  1
through 4, then the State or Tribe should submit documentation as required by
§ 132.5(b)(4) showing how procedure 1 has been used to develop the  less
stringent site-specific modifications.  If a State or Tribe uses the
scientific defensibility exclusion in § 132.4(h) to develop less stringent
criteria, then the State or Tribe should likewise submit full documentation of
its demonstration that the corresponding Guidance criteria methodology  is
scientifically indefensible for the particular situation.  It is expected  that
§ 132.4(h) will only be invoked for data which become available  after this
publication of the final Guidance.  That is, § 132.4(h) will  generally  not
serve as the basis to reconsider data already available and considered  by  EPA
prior to this publication.

      Similarly, in reviewing criteria and values for pollutants other  than
those in Tables 1 through 5, EPA will consider whether all relevant
requirements of the Guidance have been fully incorporated into  State  or Tribal
programs.  For example, States and Tribes  should provide documentation  on
their use of § 132.5(g)(2)(i) or  § 132.5(g)(2)(ii) to develop numeric water
quality criteria; all data supporting any  site-specific modifications;  and
whether provisions regarding endangered or threatened species have  been
fulfilled.   Similarly,  if the State or Tribe has adopted § 132.5(g)(2)(ii),
they should  include documentation supporting adoption and implementation of
the required procedure  for developing water quality-based effluent  limits  and
total maximum daily loads.

-------
                           Section II: Regulatory Requirements                        79
      Finally, supporting documentation will also be necessary to assist EPA
in determining whether adopted methodologies, policies, and procedures comply
with § 132.5(g)(3).  That section describes how EPA will evaluate whether
State or Tribal methodologies, policies, and procedures are consistent with
the Guidance.  For example, the documentation might simply state that the
State or Tribe has adopted provisions verbatim from the Guidance, or with only
conforming changes.  If provisions were adopted with modifications, the
supporting documentation should identify the modifications, and should
demonstrate that the resulting methodology, policy, or procedure is expected
to produce results that are as protective as the corresponding Guidance
methodology, policy, or procedure.

      EPA would like to clarify that EPA intends to apply the general factors
in § 132.5(g) when approving or disapproving State or Tribal water quality
standards under § 131.21(b) and modifications to NPDES programs under part
123.  That is, future submissions or revisions to water quality standards and
NPDES programs for the Great Lakes System must continue to be consistent with
the Guidance even after EPA has approved or promulgated standards under
§ 132.5.

3.    Indian Tribes

      a.    Proposal.  The CPA requires Great Lakes States to adopt, or EPA to
promulgate, water quality standards, antidegradation policies, and
implementation procedures for waters within the Great Lakes System which are
consistent with the final Guidance.  In the proposal, EPA expanded the
requirement to "Great Lakes States and Great Lakes Tribes."  "Great Lakes
Tribes" are defined as any Indian Tribe whose reservation lies in whole or in
part within the drainage basin of the Great Lakes, and for which EPA has
approved water quality standards under section 303 of the CWA or which EPA has
authorized to administer an NPDES program under section 402 of the CWA.
"Indian Tribe" is further defined as any Indian Tribe, band, group, or
community recognized by the Secretary of the Interior and exercising
governmental authority over a Federal Indian reservation.  EPA believes that
inclusion of Great Lakes Tribes in this way is necessary and appropriate to be
consistent with section 518 of the CWA.  The reasons for EPA's proposal are
discussed further in the preamble to the proposed Guidance  (58 FR 20834) .

      b.    Comments.  Several comments supported the inclusion of Great Lakes
Tribes in the proposal, asserting that all land that is part of the Great
Lakes ecosystem should be subject to the Guidance.

      No comments were received rejecting the proposed approach.  Some
comments urged EPA to clarify that time frames for Tribes to adopt the final
Guidance should not start until the date a Tribe is authorized to administer
the program.  Another urged EPA to clarify that the proposed 18-month adoption
deadline "in no way limits any Tribe's rights or time limits to seek
qualification" for treatment as a State.

      EPA agrees that Great Lakes Tribes should be subject to the final
Guidance for reasons described in the preamble to the proposed Guidance (58 FR
20834).  EPA has therefore retained the proposed approach that generally
includes Great Lakes Tribes in the provisions of part 132 that otherwise apply
to Great Lakes States.

      EPA also agrees that any time deadlines for Tribes to adopt the final
Guidance should not start until the date a Tribe has a water quality standards
or NPDES program in place.  For water quality standards, this means the date
on which EPA has approved water quality standards for the Tribe, including
such basic elements as designated uses, narrative criteria, numeric criteria
for at least some pollutants necessary to protect designated uses, and legal
authorities.  Without-such basic elements, the water quality standards
provisions in the final Guidance could not be implemented.  For NPDES

-------
80     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

programs, this means the date on which EPA has authorized the Tribe to
administer an NPDES program under section 402 of the CWA.

      If any Tribe had received either water quality standards approval or
NPDES program authorization prior to this publication of the final Guidance
(that is, March 23, 1995), then the Tribe would be subject to all provisions
of the final Guidance, including the submission requirements contained  in
§ 132.5.  No Tribes-in the Great Lakes basin have yet received such approval
or authorization.

      If any Great Lakes Tribe seeks to receive such approval or authorization
after March 23, 1995, then the Tribes will be subject to all provisions of the
Guidance except the submission requirements in § 132.5.  For example, in order
for a Tribe to receive approval of its water quality standards, assuming it
has been approved to administer a water quality standards program under
section 303, the Tribe will need to adopt and submit to EPA water quality
standards that meet all requirements of part 131, including § 131.5(a)(5),
which requires water quality standards to be consistent with the Guidance in
part 132.  The requirements for submission and EPA review for such standards
are specified in part 131.  Additionally, as discussed further in section
II.D.2 of this document, EPA will apply the general factors in § 132.5(g) for
determining consistency with the final Guidance, as well as all other
applicable requirements of part 131, when approving or disapproving Tribal
water quality standards under § 131.21(b).

      EPA also agrees with comments that the provisions of part 132, including
the submission deadlines in § 132.5, in no way limit any Tribe's rights or
time limits to seek qualification to administer water quality standards or
NPDES programs.

      Upon request, EPA will provide technical guidance and assistance
concerning both basic water quality provisions, and provisions consistent with
the Guidance, to Tribes that have applied or wish to apply for approval to
implement water quality standards or NPDES permit programs.

      c.    Final Guidance.  The final Guidance retains provisions for  Great
Lakes Tribes to adopt water quality standards, antidegradation policies, and
implementation procedures for waters within the Great Lakes System which are
consistent with the final Guidance.  The definition of Great Lakes Tribe has
been modified to clarify that this includes only those Indian Tribes for which
EPA has approved water quality standards under section 303 of the CWA or which
EPA has authorized to administer an NPDES program under section 402 of  the
CWA.

E.    Amendments to NPDES and Water Quality Standards Program Regulations

      1.    Proposal.  The proposal included conforming amendments to the
NPDES and water quality standards regulations in 40 CFR parts 122, 123, and
131 to insert references to the provisions of part 132 applicable to Great
Lakes States and Tribes.  These amendments ensure that future actions under
NPDES and water quality standards programs in the Great Lakes basin will be
consistent with the Guidance.

      2.    Final Guidance.  No specific comments were received concerning
these amendments.  The  final Guidance  retains the proposed amendments to  §§
122.4, 123.25, 123.44,  123.62, 123.63, 131.1, 131.5, and 131.21.  The
amendments to  §§ 122.4, 123.44, and 123.63 were modified to clarify that  they
apply only after provisions consistent with the final Guidance have been
promulgated by the State, Tribe,  or EPA.  The amendment to § 123.63 was
further modified to  clarify that  the criteria for withdrawal of a Great Lakes
State or Tribal NPDES program include  failure to adequately incorporate the
NPDES implementation procedures in 40  CFR part  132 into individual permits.

-------
                           Section II: Regulatory Requirements                        81
      F.    Precedential Effects of Elements of the Guidance

      1.    Proposal.  The requirements in the proposed Guidance were
expressly applicable only to the waters of the Great Lakes System.  EPA
requested comments on whether EPA should issue National guidance or propose
any modifications to 40 CFR parts 122  - 124, 130 and 131 in the future to
correspond with specific elements of the proposed Guidance.

      2.    Comments.  Some comments favored broader application, recommending
that either the entire final Guidance  or all non-Great Lakes-specific aspects
of the Guidance be applied nationally  to help enhance national consistency.

      Many commenters opposed broader  application of the Guidance because they
believe the Guidance was developed for the unique characteristics of the Great
Lakes basin, or because they believe the scientific basis of the Guidance is
unproven or its implementation untried.  Some commenters recommended
developing a national regulation instead of the Great Lakes-specific Guidance,
either because they believe the Guidance will reduce the competitiveness of
the Great Lakes region, or because it  will be counterproductive to have two
different NPDES programs within Great  Lakes States having waters both inside
and outside of the basin.  Other commenters recommended having States outside
the basin adopt existing nationwide EPA guidance and policy to help alleviate
some of the concern that the GLI will  cause a competitive disadvantage for the
Great Lakes basin.

      There are many provisions of the Guidance that might be beneficially
applied in other jurisdictions to improve the national program and foster
consistency.  For example, EPA believes many of the concepts in the
methodology for development of bioaccumulation factors and its use in
developing criteria to protect human health and wildlife could be applied
elsewhere.  At the same time, EPA would not consider applying specific
Guidance elements outside the basin in situations where they are not
scientifically or technically defensible.  For example, the special provisions
for BCCs may not be appropriate in systems not having the long retention times
and other chemical, biological, and physical characteristics of the Great
Lakes System described in sections I.A and I.B of this document.

      During the normal course of developing and improving national water
quality programs, EPA will consider incorporating any elements of the final
Guidance that appear to be scientifically and technically appropriate.  Such
changes could be implemented, for example, as internal guidelines for EPA
staff in developing criteria guidance  under section 304(a), as guidance to
States and Tribes in adopting criteria or implementation procedures into their
programs, or as Federal rulemaking.  Any significant proposed changes
affecting national programs would be announced in advance in order to solicit
comment from the scientific and technical community, as well as the public at
large.  In addition, EPA will seek peer review in accordance with EPA policy,
which states that major scientifically and technically based work products
related to EPA decisions normally should be peer reviewed.  EPA will solicit
external peer review for those work products that are intended to support the
most important decisions or that have  special importance in their own right.

      EPA is currently considering broader application of concepts in the
Guidance methodologies for development of bioaccumulation factors and wildlife
criteria.  In April 1994, EPA staff presented some of these concepts to the
EPA Science Advisory Board for review.  In addition, EPA is in the process of
reviewing and revising the 1980 National Guidelines which would apply to
development of EPA National human health water quality criteria under section
304(a) of the CWA.  If EPA decides to proceed with any of these concepts
nationally,  it will announce its proposal in the Federal Register.

      EPA does not agree that the scientific basis for the final Guidance is
unproven.  The scientific basis for the final Guidance is discussed further in
sections I and III through VIII of this document.  EPA agrees, however, that

-------
82     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

the experience of EPA, Great Lakes States, and Great Lakes Tribes in
implementing the final Guidance will be useful in deciding how elements of the
Guidance can be appropriately applied in other jurisdictions.

      EPA does not agree that the Guidance would lead to two different NPDES
programs within a State.  States are already accustomed to developing
different permit conditions and limitations depending on water quality
conditions in different parts of the State.  The Guidance would simply ensure
that consistent approaches are used for waters of the Great Lakes System.

      EPA also does.not agree with comments recommending a national regulation
in place of the proposed Guidance.  First, the CPA required EPA to issue water
quality guidance for the Great Lakes System, not for the nation.  Therefore,
EPA did not generally provide opportunity for States, Tribes, or members of
the public and regulated community outside the Great Lakes basin to
participate in developing the Guidance.  Second, the short statutory and
judicial deadlines for producing and implementing the Guidance did not allow
time for considering broader application beyond the Great Lakes System or
obtaining appropriate public comment.  Third, EPA does not believe there will
be significant detrimental effects to the economy of the Great Lakes region
that would place the region at a competitive disadvantage to other parts of
the country.  As discussed further in section IX.1 of this document, a study
conducted for the Council of Great Lakes Governors showed that such effects
are expected to be minimal.

      EPA agrees that it would be useful to increase the consistency of water
quality programs applied nationally.  EPA will continue to work toward this
goal in implementing water quality standards and NPDES permit programs
throughout the country.

      3.    Final Guidance.  The final Guidance contains no mandatory
requirements for discharges outside the Great Lakes System.

      EPA would like to reemphasize that the provisions in the proposed and
final Guidance are expressly applicable only to the waters of the Great Lakes
System.  EPA has initiated no rulemaking action to extend any Guidance
provisions beyond'the Great Lakes System.  EPA's request for comments in the
proposal was soliciting views only on whether any future national guidance or
rulemaking affecting water programs beyond the Great Lakes System should
include any concepts contained in the final Guidance.

      States or Tribes with waters outside the Great Lakes System, in whole or
in part, are encouraged to implement any of the Guidance methodologies or
procedures that are scientifically and technically appropriate for their
situations.  This would include any Great Lakes States or Tribes that may
choose to apply some or all of the Guidance in portions of their jurisdictions
outside the Great Lakes System.   Some Great Lakes States indicated their
intention to do so  in their written comments on the proposed Guidance.

G.    Implementation of Endangered Species Act

      1.    Proposal.  The proposed Guidance contained no specific provisions
concerning the protection  of endangered or  threatened species under  the
Endangered Species Act  (ESA) .  EPA invited  comment on several possible
approaches that EPA was considering including in the Guidance to reflect EPA's
responsibilities under  the ESA  (58 FR 20848-20849).  Under section  7(a)(2)  of
the ESA, each Federal agency shall, in consultation with the U.S. Fish  and
Wildlife Service  (FWS), or the National Marine  Fisheries Service for species
under its jurisdiction, ensure that actions authorized, funded  or carried  out
by the Federal agency are  not likely to jeopardize the  continued existence of
any endangered or threatened species listed under  section 4  of  the  ESA,  or
result in the destruction  or adverse modification  of  such species'  critical
habitat  (i.e., are  not  likely to  "cause jeopardy"  to  "listed species").

-------
                           Section II: Regulatory Requirements                        83
 (There are no species tinder the jurisdiction of the National Marine Fisheries
Service in the Great Lakes basin.)

      2.    Consultation with the U.S. Fish and Wildlife Service.  Procedural
requirements for the consultation envisioned in section 7(a)(2) of the ESA  are
found in 50 CFR part 402.  These regulations provide for two types of
consultation: formal and informal.  Formal consultation procedures are
required in situations where a Federal action may affect listed species or
critical habitat.  The formal procedures include requirements for the Federal
agency to provide available information to the FWS, and for the FWS to develop
a Biological Opinion discussing the effects of the Federal action on the
listed species.  Informal consultation is an optional procedure that includes
discussions and correspondence between the FWS and the Federal agency to
determine whether an action is likely to adversely affect a listed species.
If the Federal agency finds, and the FWS concurs in writing, that the action
is not likely to adversely affect listed species, consultation is concluded
and formal consultation is not necessary.

      In January 1993 EPA and the FWS initiated informal consultation under
section 7 of the ESA concerning the issuance of the final Guidance.  The two
agencies identified several specific issues for detailed consultation.  In
September 1994, EPA initiated formal consultation with the FWS on two of the
issues: adequacy of the Guidance's aquatic life criteria methodologies and
related implementation procedures to protect endangered mussel species in the
Great Lakes basin, and the appropriateness of the Guidance's methodology to
develop wildlife criteria.  On February 21, 1995, the FWS provided EPA with a
written Biological Opinion.  The Opinion is available in the docket for this
rulemaking.  The results of the formal and informal consultation are as
follows:

      a.    Protection of Endangered Mussel Species

      EPA and the FWS were concerned about the effects of water quality
resulting from implementation of the final Guidance on three listed endangered
freshwater mussel species in the Great Lakes System: Pleurobema clava
 (Clubshell), Epioblasma torulosa ranqiana  (Northern riffleshell), and E.
obliouata peroblioua (White cat's paw pearly mussel).  The agencies were
concerned that since there is very limited aquatic toxicity information for
these species or their surrogates, it is difficult to assess whether the
aquatic life criteria methodology in the final Guidance produces criteria that
are stringent enough to protect the endangered species.

      EPA and the FWS undertook a formal consultation to address this issue.
EPA provided information about the proposed Guidance and its expected effects
on mussel species to the FWS, including the best scientific and commercial
data available.  The FWS then carried out a review and evaluation that
resulted in the February 21, 1995, Biological Opinion.

      The Biological Opinion concluded that the water quality resulting from
implementation of the final Guidance is not likely to jeopardize the continued
existence of the mussel species.  The FWS identified possible concerns,
however,  that the aquatic life methodologies in the final Guidance may not  be
stringent enough to avoid adverse effects to listed mussels in all cases.   The
FWS conducted risk assessment analyses using existing toxicity studies on
freshwater mussels.  Although such studies are limited in number, and were  not
used by EPA either because they were not available in time for inclusion or
because they did not meet EPA data requirements, the FWS believes that
freshwater mussel data would provide a better indication of the sensitivity of
the listed mussels than data based on less closely related test organisms.  In
situations such as this, where information on listed species is limited, the
policy of the FWS is to provide the "benefit of the doubt" to the species
concerned (51 FR 19952, June 3, 1986; H.R. GOnf. Rep. No. 697, 96th Cong.,  2d
Sess. at 12 (1979)).  The FWS assessment resulted in effects concentrations
lower than the Guidance aquatic life criteria for some contaminants.  Based on

-------
84     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

these results, the FWS concluded that the water quality resulting from
implementation of the aquatic life criteria and values in the final Guidance
may result in "incidental take" of an unquantified number of endangered
freshwater mussels in the Great Lakes basin.

      Section 9 of the ESA prohibits  "take" of listed species of fish or
wildlife--that is, .harassing, harming, pursuing, hunting, shooting, wounding,
killing, trapping, capturing or collecting such species, or attempting to
engage in any such conduct--unless the take is exempted from this provision as
discussed below.  FWS policies further define "harm" to include significant
habitat modification or degradation that results in death or injury to listed
species by significantly impairing behavioral patterns such as breeding,
feeding, or sheltering  (50 CFR 17.3).  Under the terms of section 7(b)(4) and
7(o) (2) of the ESA, taking that is "incidental" to and not intended as part of
the Federal agency'p action is not considered a prohibited taking provided
that such taking is in compliance with requirements in the Biological Opinion
specifying reasonable and prudent measures necessary and appropriate to
minimize the incidental take, and specifying mandatory terms and conditions to
be followed.

      To minimize the amount or extent of any incidental take resulting from
implementation of the final Guidance, the FWS consulted closely with EPA to
develop a coordinated approach.  The final Biological Opinion specified
reasonable and prudent measures that the FWS considers necessary or
appropriate to minimize such impact as the aquatic life criteria,
methodologies, and related implementation procedures are implemented in the
Great Lakes System as specified in the final Guidance.  These measures are as
follows:

      --  During the next two years, EPA will undertake a screening analysis
consisting of acute toxicity testing of freshwater mussels for a limited
number of pollutants.  During the same period, to enable successful completion
of the measures, FWS will develop and analyze life history information on the
endangered mussels.  The life history information is particularly important in
identifying surrogate species, developing test protocols, and culturing test
organisms, and further determining what other environmental parameters may be
contributing to the'decline of the species.

          EPA and FWS will review the results from the above efforts.  The
purpose of the review is to compare the sensitivity of the surrogate mussel
species with the sensitivity of other species used in developing acute
criteria under the final Guidance.  The review will also help determine the
types of testing needed to conduct a  screening analysis for chronic effects.

          During a subsequent period  of approximately four years, EPA and FWS
will conduct a screening analysis consisting of selected chronic toxicity
testing of freshwater mussels for a limited number of pollutants.  The scope
and design of this testing will depend heavily on the results of the efforts
of the first two years described above.

          Based on the combined results of the acute and chronic screening
analyses, EPA and FWS will assess whether there is evidence that the three
endangered mussels or their surrogate species are more sensitive to aquatic
pollutants than other species assessed as part of EPA and State criteria
development efforts under the final Guidance.  If so, then the agencies will
develop a joint approach for developing the necessary information to adjust
criteria as necessary to protect the  species.
                    *
          EPA and the FWS will develop a protocol that can be used  to monitor
any populations of  listed mussels downstream from new discharges permitted
under the final Guidance.  EPA will then seek through its oversight of State
water quality programs  to ensure  that monitoring of such populations takes
place in accordance with the protocol to identify whether any incidental  take
occurs  that is  likely to cause jeopardy to  the  species.

-------
                           Section II: Regulatory Requirements                        85
      EPA believes that the methodologies and procedures  in the  final Guidance
will be protective of endangered or threatened aquatic species in  the Great
Lakes System.  EPA acknowledges that it  is possible that  data collected  in the
future, however, could identify an unusual sensitivity of mussels  to a
particular pollutant that is not reflected in the methodologies  and
procedures.  Based on experience in implementing water quality criteria  under
the CWA, EPA believes that this possibility is small, and that the degree of
any adverse effects would likely be limited.  Nevertheless, EPA  agrees that it
would be desirable to obtain additional  data on freshwater mussels to confirm
its belief that mussels will not be adversely affected by water  quality
resulting from implementation of the final Guidance.  For this reason, EPA has
agreed to implement the reasonable and prudent measures described  above.

      b.    Wildlife Criteria Methodology

      EPA and the FWS were concerned about the effects of water  quality
resulting from implementation of the final Guidance on five wildlife species
listed as endangered or threatened in the Great Lakes basin: Canis lupus (Gray
wolf), Mvotis sodalis  (Indiana bat), Haliaeetus leucocephalus  (Bald eagle),
Palco pereorinus  (Peregrine falcon), and Charadrius melodus  (Piping plover).
Of these, the primary concern was the bald eagle because  of its  relatively
greater dependence on food sources found at higher levels of the aquatic food
chain.

      EPA and the FWS undertook a formal consultation to  address this issue.
EPA provided information to the FWS about the wildlife methodology and its
predicted effects on bald eagles.  The FWS then carried out a review and
evaluation that was included in the February 21, 1995, Biological  Opinion.

      The Biological Opinion concluded that the final Guidance is  not likely
to jeopardize the continued existence of the bald eagle,  peregrine falcon, or
piping plover.  The FWS identified possible concerns, however, that the
wildlife methodology in the final Guidance may not be stringent  enough to
protect endangered or threatened species in all cases.  In their risk
assessment, the FWS used assumptions and assessment approaches that differ
from those developed by the Initiative Committees, whose  approach  was reviewed
by the EPA Science Advisory Board and used in the proposed and final Guidance.
The model developed by the FWS is based  on the dose to the target  tissue (the
egg), rather than the dose received in the diet.  The FWS assessment also
considers the protection of individuals  as opposed to populations, and
emphasizes the need for toxic equivalency factors to account for additive or
synergistic effects of complex mixtures  of bioaccumulative compounds to
wildlife.  Based on the results of their assessment, which gives the "benefit
of the doubt" to the species in accordance with FWS policy as discussed  above,
the FWS concluded that the water quality resulting from implementation of the
final Guidance may result in "incidental take" of an unquantified  number of
bald eagles, peregrine falcons, and piping plovers in the Great  Lakes basin
due to detrimental effects resulting from chronic toxicity.

      To minimize the amount or extent of any incidental  take resulting  from
implementation of the final Guidance, the FWS consulted closely  with EPA to
develop a coordinated approach.  The Biological Opinion specified  reasonable
and prudent measures that the FWS considers necessary or  appropriate to
minimize such impact as the wildlife criteria, methodologies, and  related
implementation procedures are implemented in the Great Lakes basin as
specified in the final Guidance.  These measures are as follows:

          During thfe nine years following publication of  the final Guidance,
EPA will gather information concerning the impacts on endangered or threatened
species in the Great Lakes basin from the four chemicals  with numeric wildlife
criteria in the final Guidance.  If at any time new information  indicates that
water quality after full implementation of the wildlife components of the
Guidance may adversely affect endangered or threatened species,  EPA will
immediately reinitiate consultation.

-------
86     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

          EPA will establish and begin implementing a data-gathering plan, in
cooperation with FWS and the States or Tribes, to determine progress in
reducing the concentrations of the four chemicals in the Great Lakes aquatic
ecosystem, and to help improve understanding of the sensitivities of the
listed species to the chemicals if possible.

          EPA will during the next five years reevaluate and better define
loadings of BCCs to the Great Lakes System that might affect listed wildlife
species, including a report describing ongoing activities (e.g., Lakewide
Management Plans and Remedial Action Plans) that are designed to reduce
current levels of contamination in the basin.

          EPA and FWS will cooperatively plan and host one or more workshops
to explore the utility of additivity models in developing wildlife criteria,
to consider the scientific basis for use of toxicity equivalency factors
(TEFs) when establishing total maximum daily loads or water quality-based
effluent limits to implement wildlife criteria, and to discuss research and
data needs.  The -findings of these workshops will be used to develop
approaches for utilizing TEFs if appropriate in the development and
implementation of wildlife criteria.  These approaches will be presented to
the EPA SAB for review and comment.

      EPA believes that the wildlife criteria methodology in the final
Guidance, including the procedures for site-specific criteria modifications,
will be protective of endangered or threatened wildlife species in the Great
Lakes basin.  Nevertheless, EPA believes it will be useful to conduct the type
of review described above to confirm that water quality resulting from
implementation of the final Guidance is not likely to adversely affect
endangered or threatened wildlife species.  For these reasons, EPA has agreed
to implement the reasonable and prudent measures described above.

      c.    Other Issues

      As a result of informal consultation, and after considering comments
(discussed below), EPA decided to include certain additional provisions in the
final Guidance to ensure protection of endangered or threatened species
consistent with EPA's responsibilities under section 7.  These provisions are
discussed below.  They generally reflect the types of.provisions discussed in
the preamble to the proposed Guidance  (58 PR 20848-49) .

      EPA and the FWS consulted informally on specific regulatory provisions
to ensure that aquatic life and wildlife criteria will be developed or
modified on a site-specific basis to protect endangered or threatened species
in the Great Lakes System.  The two agencies agreed that procedure 1 of
appendix F should be modified to provide that Great Lakes States and Tribes
must develop more stringent site-specific modifications of criteria and values
to protect listed and proposed species, where such modifications are necessary
to ensure that water quality is not likely to cause jeopardy to the species.
Because efforts to protect and promote recovery of listed and proposed species
generally focus on management of specific ecosystems, it is appropriate that
States, Tribes, and EPA utilize the procedure for site-specific modifications
of criteria and values to address conditions in those ecosystems.

      In the above provisions, EPA has included proposed as well as listed
species.  EPA believes that prevention of water quality conditions that would
be likely to jeopardize the continued existence of any species  is inherent in
the CWA's requirement for EPA to provide guidance on minimum water quality
standards to protect aquatic life and wildlife in the Great Lakes basin  (CWA
section 118(c) (2)), as well as the fundamental principle that water quality
criteria must protect designated uses  (CWA  section 303(c) (2) (A); 40 CFR
131.11(a) (2)) .  Obviously, if impaired water quality will likely cause the
extinction of a species, such water quality would not meet the Act's
requirements.  In a sense, then, mandating that States adopt site-specific

-------
                           Section II: Regulatory Requirements                        87
criteria to avoid causing jeopardy to any species is simply an explicit
statement of a principle that EPA believes is inherent in the CWA.

      The two agencies also agreed that States and Tribes should be encouraged
to adopt more stringent site-specific modifications of criteria or values
where necessary to protect candidate "Cl" species being considered by the FWS
for listing under section 4 of the ESA.  The "Cl" category refers to species
for which the FWS has enough substantial information to support proposals to
list as endangered or threatened under section 4 of the ESA.  The new
provisions of procedure 1 of appendix F to part 132 are discussed under the
Final Guidance section below.

      EPA and the FWS also consulted informally on concerns about whether the
final Guidance should apply to ammonia and chlorine.  The two agencies agreed
that EPA will undertake a review of water quality standards and implementation
of those standards for ammonia and chlorine in the Great Lakes basin as part
of EPA's responsibilities under section 303(c)  of the CWA.  This review is
discussed further in section II.C.B.c of this document.

      The two agencies also consulted informally concerning the adequacy of
the Guidance methodologies for developing criteria and values to protect
wildlife from certain pollutants that are metabolized by higher trophic level
aquatic organisms but not by lower trophic aquatic organisms.  Such pollutants
include, for example, polynuclear aromatic hydrocarbons (PAHs) which, because
of metabolism by fishes, have relatively high BAFs when measured in lower
trophic level species, even though they have relatively low BAFs in higher
trophic level fish.  The methodology for development of BAFs for organic
chemicals in appendix B includes procedures for calculating field-measured
and/or predicted BAFs for all trophic levels, and allows,  on a case-specific
basis, wildlife BAFs to be weighted to reflect the proportion of plants,
invertebrates, and fish in the diet of the species to be protected.

      If a listed wildlife species in the Great Lakes basin were found to be
more highly exposed to a bioaccumulative pollutant because of its diet than
the representative species used in the wildlife criteria methodology, then
procedure l.A of appendix F would specify that a site-specific modification is
needed to protect such species.  The site-specific modification could utilize
the method described above to weight the BAFs based on the diet of the listed
species.

      The two agencies agreed that the final methodology for the development
of bioaccumulation factors in appendix B, and procedure 1 of appendix F for
site-specific modifications, provide adequate protection for listed wildlife
species that may consume highly contaminated prey at lower trophic levels.

      The two agencies also agreed that the final Guidance should clarify
EPA's responsibilities for protecting endangered or threatened species when
reviewing and approving State and Tribal submissions under § 132.5.  Section
132.5(h) has been added to the final Guidance to require States and Tribes to
include in their part 132 submissions any provisions that EPA determines,
based on EPA's authorities under the CWA and the results of consultation under
section 7 of the ESA, are necessary to ensure that water quality is not likely
to cause jeopardy to listed species.  This provision is discussed under the
Final Guidance section below.

      Finally, the two agencies discussed and have agreed upon the procedures
the two agencies will use to conduct section 7 consultations where necessary
on future EPA actions.  These actions include EPA approval or disapproval of
submissions by Great Lakes States and Tribes under § 132.5, and EPA approval
actions for State and Tribal water quality standards, criteria modifications,
and variances under 40 CFR part 131.  The procedures are summarized under the
Final Guidance section below.

-------
88     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      With the issuance of the FWS1 Biological Opinion, formal consultation on
the issues concerning freshwater mussels and wildlife criteria has been
completed.  For the remaining issues, as a result of the informal consultation
described 'above, the FWS concurs with EPA that the final Guidance is not
likely to adversely affect endangered or threatened species in the Great Lakes
basin ecosystem.  Therefore, consultation concerning EPA's publication of the
final Guidance has been completed.

      3.    Comments.   Some comments asserted that specific ESA provisions in
the final Guidance would be an unnecessary additional burden on the States,
since most States have a state program for the protection of endangered or
threatened species which deals directly with the FWS.  Several commenters
supported using site-specific modifications to provide protection for
endangered or threatened species.  Many of these commenters preferred that
such a provision be guidance.  Other comments asserted that full consultation
by EPA with the FWS on administration of all elements of the final Guidance is
necessary, and that consultation procedures should be clarified.

      EPA does not agree that regulatory provisions regarding ESA are
unnecessary in the final Guidance.  The ESA imposes specific responsibilities
on Federal agencies to protect endangered or threatened species in actions the
agencies carry out.  EPA believes that the final Guidance should articulate
the conditions that.EPA,  based on its consultation with the Service, believes
are necessary to include in State and Tribal submissions to provide the
protection for endangered or threatened species envisioned in section 7(a)(2)
of the ESA.  For example, procedure l.A of appendix F clarifies the role of
site-specific modifications to water quality criteria in protecting endangered
or threatened species, and § 132.5(h) of the final Guidance clarifies that EPA
will make a determination based on EPA's authorities under the CWA and the
results of section 7 consultation concerning the adequacy of State or Tribal
provisions for protecting listed species contained in submissions under part
132.  EPA believes these clarifications will be helpful in ensuring an orderly
approach to protect listed species.  Furthermore, although individual States
may have State-imposed responsibilities to protect endangered or threatened
species, these responsibilities may not coincide in detail with applicable
Federal standards.  For example, a State's endangered species program may
address a list of species that could be different from the species listed
under section 4 of the ESA.  The clarifications in part 132 concerning the
Federal requirements therefore will be helpful in reducing confusion on such
matters.

      EPA agrees for the reasons discussed under section II.G.2 above that
site-specific modifications to criteria and values should be required to
protect endangered or threatened species.  EPA also agrees that the final
Guidance should provide States and Tribes flexibility in achieving such
protection, and therefore provides a choice of methods for adopting site-
specific modifications.  This flexibility in adopting site-specific
modifications to protect endangered or threatened aquatic and wildlife species
is discussed further in section VIII.A of this document.

      4.    Final Guidance.  As a result of consultation with the FWS, and in
response to comments, EPA has made the following changes in the final
Guidance:

      EPA has added.§ 132.5(h) to the final Guidance which provides that Great
Lakes States and Tribes will need to include provisions that EPA determines,
based on the results of consultation under  section 7 of the ESA, are necessary
to ensure that water quality is not  likely  to jeopardize the continued
existence of any endangered or threatened species listed under  section 4 of
the ESA or result in the destruction or adverse modification of such species'
critical habitat.

      To carry out  the ESA, relevant EPA Regional Offices will  initiate
consultation with the FWS as soon as possible concerning EPA's  approval

-------
                           Section II: Regulatory Requirements                        89
actions under part 132 for each of the Great Lakes States and Tribes.  Such
consultations will be facilitated if States and Tribes have already consulted
with the FWS under State and Tribal arrangements with FWS where such
arrangements exist.  Initiating consultations early in the process of the
States' and Tribes' development of their part 132 submissions is appropriate
and necessary because under section 118 (c) (2) (C) of the CWA and § 132.5, the
States .and Tribes must complete their part 132 submissions to EPA generally by
September 23, 1996, and EPA must approve such submissions, or promulgate EPA
requirements, by March 23, 1997.  EPA believes the consultations can and
should occur simultaneously with the States' and Tribes' efforts to develop
their part 132 submissions.

      As a result of the consultation that has taken place with the FWS
concerning the final Guidance, EPA believes that consultations can be
concluded rapidly and routinely concerning part 132 submissions from States or
Tribes that are consistent with the final Guidance.  EPA will make every
effort to expedite exchange of information and consultation between FWS staff
members, State or Tribal representatives, and EPA Regional Office staff
members to help determine whether any additional modifications to a particular
submission are necessary, such as site-specific modifications.  If such
situations occur, EPA will notify the State or Tribe as soon as possible.

      EPA has consulted with the FWS regarding the methodologies for
development of criteria and values in the final Guidance.  Based upon
available data and information, EPA and the FWS have concluded that criteria
and values consistent with the final Guidance generally will result in water
quality that is not likely to jeopardize the continued existence of listed
species or destroy or adversely modify critical habitat.  Thus, in the absence
of a need for a site-specific modification, discussed below, States and Tribes
adopting criteria and values, methodologies, policies, and implementation
procedures consistent with the final Guidance will be approvable by EPA as
consistent with the ESA.

      EPA recognizes that information may become available--either prior to or
subsequent to approval of State and Tribal submissions under part 132--
indicating that more stringent criteria/values may be necessary to protect
certain species.  Therefore, EPA has modified procedure l.A of appendix F of
the final Guidance to provide that Great Lakes States and Tribes need to
develop more stringent site-specific modifications of criteria and values to
protect listed or proposed species, where such modifications are necessary to
ensure that water quality is not likely to cause jeopardy to listed species.
As discussed above, this provision is necessary to clarify the role of site-
specific modifications in protecting endangered or threatened species listed
or proposed under the ESA,  Procedure l.A also encourages States and Tribes to
adopt more stringent site-specific modifications of criteria or values where
necessary to protect candidate (Cl) species.

      Procedure l.A of appendix F provides that such site-specific
modifications be made where they are necessary to ensure water quality is not
likely to cause jeopardy to listed or proposed species.  EPA interprets the
phrase "where necessary" to include geographic sites where the species may be
affected, and where critical habitat has been designated.  Procedure 1 of
appendix F allows flexibility in interpreting the geographic extent of
"sites."  Therefore, States, Tribes, and EPA in consultation with the FWS,
should establish the geographic extent of the site-specific modifications to
be large enough to provide adequate protection to the species, including
survival and recovery.  For example, a site-specific modification could be for
an area as small as a local stream reach where a listed species currently
resides, or as large as the entire portion of the State located within the
Great Lakes basin, depending on the needs of the species.

      EPA also interprets the phrase "where necessary" to mean where available
data indicate that listed or proposed species are especially vulnerable to
specific pollutants, such as being more sensitive than the species used to

-------
90     Water Quality Guidance for the Great Lakes System — Supplementary Information Document


develop existing criteria, or being more exposed than assumed in the Guidance
methodologies.

      The GLI Clearinghouse, discussed in section II.C.I of this document,
will facilitate the incorporation of new data on listed, proposed, or
candidate  (Cl)  species under the ESA, or surrogates for such species, into
State and Tribal prpgrams.  Any such data will be placed in the EPA-operated
Clearinghouse,- where they will be available to States, Tribes, and other
interested persons.  EPA will place special emphasis in the Clearinghouse on
data relevant to protection of such species.  EPA will alert States and Tribes
to data that indicate unusual sensitivity of such species in the Great Lakes
basin.  EPA and the FWS will also work with States and Tribes to identify
sites where special protection may be recommended.

      EPA will also work with States and Tribes to disseminate any new data
that become available in future years showing that a site-specific
modification to criteria is appropriate because a listed or proposed species
is especially sensitive to aquatic pollutants.  For example, if data relevant
to a listed aquatic species become available concerning a SMAV that is less
than the calculated FAV, or that would yield a lower criterion or value using
the EPA recalculation procedure or resident species procedure, a site-specific
modification would be necessary for waters where the species is found.  If the
State or Tribe already had numeric water quality criteria for the pollutant,
the State or Tribe would need to adopt a site-specific modification and submit
it to EPA for approval as currently specified in § 131.20.  EPA would then
consult with the FWS concerning the approval of this submitted revision to the
water quality criteria.  If the State or Tribe did not submit such a
modification, or did not adopt criteria protective of the listed species, and
did not demonstrate to EPA why such modifications are not necessary, EPA would
have the authority under section 303(c) of the CWA to disapprove the State
criteria for relevant areas and take necessary actions.  Under section 303(c),
EPA would then be required to propose Federal criteria in place of the State
or Tribal criteria.

      If the State or Tribe had not adopted criteria for the pollutant of
concern, it would nevertheless need to utilize the new data in developing
criteria or values to be used as appropriate in establishing water quality-
based effluent limits for the pollutant, in accordance with § 122.44(d),
§ 132.4, and adopted State or Tribal provisions consistent with the
implementation procedures in appendix F of the final Guidance.

      EPA has added provisions to final procedures I.A.I and 1.A.2 of appendix
F to clarify ways that the site-specific modifications required in procedure
l.A may be accomplished.  For aquatic life, procedure l.A.l(c) describes two
ways criteria may be modified to reflect endangered or threatened species that
are particularly sensitive to a pollutant.  For example, the Species Mean
Acute Value  (SMAV) for the listed or  surrogate species may be found to be
lower than the calculated Final Acute Value  (FAV) developed using the Tier I
methodology in appendix A.  In this situation, the lower SMAV would be
substituted for the' calculated FAV in developing the aquatic life acute
criterion.  Alternatively, the State  or Tribe may lower the criterion using
EPA's recalculation procedure or resident species procedure as provided in
Chapter 3 of the U.S. EPA Water Quality Standards Handbook, Second Edition
 (August 1994).  Procedures I.A.I and  l.A.2 are discussed further in section
VIII.A of this document.  During the  consultation process EPA will solicit the
views of the FWS concerning the choice of method for criteria modification.

      For wildlife, procedure l.A.2 describes a recommended approach for
modifying wildlife criteria to protect endangered or threatened  species.
Procedure l.A.2 is discussed further  in section VIII.A of this document.

      EPA has added provisions to appendix E and to procedures 2 and 3 of
appendix F of the final Guidance that restrict certain actions States and
Tribes may take to allow  lowering of  water quality in high quality waters, to

-------
                           Section II: Regulatory Requirements                        91
grant variances, and to allow mixing zones.  Section I.A of appendix E
specifies that any lowering of water quality in high quality waters shall be
prohibited if such lowering of water quality will be likely to cause jeopardy
to listed species.  It is discussed further in section VII of this document.
Procedure 2.A of appendix F specifies that a variance to a water quality
standard may not be granted that would result in water quality likely to cause
jeopardy to listed species.  It is discussed further in section VIII.B of this
document.  Procedure 3 of appendix F includes provisions that mixing zones in
or outside the context of a TMDL may be allowed, at a minimum, only to the
extent that water quality would not likely cause jeopardy to listed species.
these provisions are discussed further in section VIII.C of this document.

      EPA has determined that each of these restrictions is necessary to
clarify how the requirements of the ESA affect decisions regarding
implementation of water quality standards.  EPA believes including these
provisions in the final Guidance will be helpful to States, Tribes, and
dischargers to understand the provisions that EPA believes will need to be
included in part 132 submissions to meet ESA requirements.

      Under section 7 of the ESA, EPA is obligated to consult with the FWS
concerning actions carried out, authorized, or funded by EPA that may affect
listed species.  As discussed above, EPA will consult with the FWS concerning
EPA's actions to approve State and Tribal submissions under § 132.5.  EPA also
intends to consult on approval actions under 40 CFR part 131 for water quality
standards in the Great Lakes basin, including criteria modifications and water
quality standards variances, that may affect listed species, submitted as
revisions to a State's or Tribe's water quality standards.  The FWS and EPA
recognize that to accomplish timely implementation of standards that may
affect listed species, early involvement and technical assistance by the FWS
is needed.  For example, EPA and the FWS will meet during the triennial period
for reviews of water quality standards under section 303(c) of the CWA,
preferably during the period when EPA and a State or Tribe discuss the extent
of an upcoming review.  EPA and FWS agree that meetings, discussions, and
exchanges of information throughout the process--from State and Tribal
development, adoption, submission of standards, through EPA final action--will
facilitate the consultations required by the ESA.  EPA will also invite and
encourage the States and Tribes to participate actively in the consultations.

-------
                               Section ID: Aquatic Life                             93
                           .  AQUATIC LIFE
A.    Summary of Final Rule

      The final Guidance is identical to the Guidance proposed on April 16,
1993, except for the following changes:

      -- States and Tribes are permitted to use either the averaging period
and frequency of exceedance specified for the proposed Tier I criteria
concentrations, or some other scientifically acceptable averaging period
and/or frequency of.exceedance.   If EPA determines that it is necessary to
promulgate criteria for a State or Tribe, it will promulgate averaging periods
of one hour  (Criterion Maximum Concentration) and four days (Criterion
Continuous Concentration) and a frequency not to exceed the criteria of more
than once-in-three-years (both CMC and CCC).

      -- Aquatic life criteria for the nine metals proposed in Tables 1 and 2
of part 132 are expressed as dissolved concentrations and total recoverable
concentrations.  Conversion factors for the proposed total recoverable metals
criteria were derived and publicly noticed on August 30,  1994 (59 FR 44678) .
EPA adjusted these conversion factors and used them in the calculations to
convert the proposed criteria from total recoverable concentrations to
dissolved concentrations for the final Guidance.  If a State or Tribe fails to
adopt aquatic life criteria for metals, EPA will promulgate dissolved
criteria.

      - - The final Guidance clarifies and elaborates on the option of using
indicator parameter limits under 40 CFR 122.44(d)(1)(vi)(C).   A full
discussion of the use of the indicator parameter option and its role in the
final Guidance appears in section VIII.E.2.f of this document (Reasonable
Potential Procedure).  As discussed in that section of this document the
States or Tribes are still required to adopt provisions consistent with the
Tier II methodology for aquatic life consistent with the methodology in
appendix A.  As described in procedure 5 of appendix F to part 132, States and
Tribes will then be required under the final Guidance to implement the Tier II
values through water quality-based effluent limitations (WQBEL)  based on such
values when, under procedure 5 of appendix F to part 132,  such limits are
determined to be required.   When deriving limits to meet Tier II values,
States and Tribes have the option of using an indicator parameter limit in
lieu of a Tier II-based limit.  If an indicator parameter is used, the State
or Tribe must ensure that the indicator parameter will attain the "applicable
water quality standard"  (as described in 40 CFR 122.44(d) (1) (vi) (C) ) .  The
"applicable water quality standard" in this instance would be the State's or
Tribe's narrative water quality criteria that protects aquatic life,  as
interpreted using the Tier II methodology.

B.    Final Tier I Methodology

      The final Guidance includes methodologies for deriving Tier I criteria
and Tier II values for the protection of aquatic life in the Great Lakes
System.  The Aquatic Life Tier I methodology is an adaptation of the
"Guidelines for Deriving Numerical National Water Quality Criteria for the
Protection of Aquatic Organisms and Their Uses"   (U.S.  EPA,  1985)   (1985
National Guidelines), developed under section 304(a) of the Clean Water Act
(CWA).   A copy of the 1985 National Guidelines is available in the
administrative record for this rulemaking.  The 1985 National Guidelines can
also be obtained through the National Technical Information Service (PB 85-
227049) .   Provisions consistent with the final Tier I methodology must be used
when deriving Great Lakes aquatic life criteria for use in State and Tribal
water quality standards.  The final Guidance also contains Tier I criteria for

-------
94     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

16 pollutants that are based on the latest scientific knowledge and derived
using an extensive aquatic toxicological data base.

      As stated above, the proposed Tier I methodology was an adaptation of
the 1985 National Guidelines which has previously undergone scientific  peer
review and public review and comment, and had been revised where appropriate.
EPA stated in the proposal that issues already addressed by EPA in response to
previous comments on the 1985 National Guidelines would not be addressed in
this proceeding  (58FR20852).  Therefore, the following discussion focuses on
those portions of the Tier I methodology which differ from the 1985 National
Guidelines.

1.    Required Data

      a.    Proposal:  In the proposed Guidance, two ambient criteria
concentrations to protect aquatic life for any given pollutant are derived
(the Criterion Maximum Concentration  (CMC) and the Criterion Continuous
Concentration  (CCC)).  In order to derive a CMC for a pollutant, it is
necessary that acceptable acute toxicity tests exist for aquatic animals in at
least eight families which represent differing habitats and taxonomic groups.
The CCC is the lowest of the Final Chronic Value  (FCV) or the Final Plant
Value  (FPV)  (see appendix A for a complete description of how to derive a CCC
or CMC).  The aquatic life methodologies to derive both Tier I criteria and
Tier II values did allow the use of some data for freshwater species that do
not reside in the Great Lakes System.

      b.    Comments:  Several commenters recommended that the proposed
Guidance derive different criteria for coldwater versus warmwater assemblages
of aquatic species, and possibly for other receiving water categories  (based
on inclusion or exclusion of different species in the underlying data set).
Several commenters also recommended limiting the toxicity data set to species
residing in the Great Lakes basin.

      EPA recognizes the potential differences between coldwater versus
warmwater aquatic species and that there are varied aquatic species
assemblages across the Great Lakes System.  Because of this, EPA has included
provisions in the final Guidance to calculate site-specific modifications to
criteria to adjust for  site-specific differences in water quality
characteristics, species, etc.  EPA believes it is better to derive site-
specific criteria where necessary than to try to define and protect generic
assemblages of aquatic  species  (such as warmwater, coldwater, etc.).  Site-
specific criteria can be developed to protect the types of species at each
site and, therefore, are more precise.  If a State or Tribe wishes to modify
the criteria on a site-specific basis to account for a different species
assemblage at the site, the recalculation procedure as described in appendix L
of the U.S. EPA Water Quality Standards Handbook, Second Edition - Revised
 (U.S. EPA, 1994) may be used.  The 1994 Water Quality Standards Handbook is
available in the administrative record for this rulemaking and can also be
obtained through the Water Resource Center, 401 M. Street, SW  (RC4100),
Washington, DC 20460 or by calling  (202) 260-7786.

      EPA disagrees that toxicity data should be  limited to those species
residing in the Great Lakes System.  EPA believes that species nonresident to
the Great Lakes  System  can represent untested resident species and may  used as
surrogates for those resident species not yet tested.  This approach is
consistent with  the 1985 National Guidelines.

       c.     Final Guidance:  The data requirements in the final Guidance for
deriving Tier  I  criteria are the  same as the proposed Guidance.  EPA would
like to clarify  that all data used in the derivation of Tier I criteria or
Tier II values must be  from  species which are resident to North America.

-------
                               Section ffl: Aquatic Life                            95
2 .    Commercially and Recreationallv Important Species

      a.    Proposal :  The 1985 National Guidelines stated that the Final
Acute Value  (FAV) must be set equal to the lower 95th percentile value, or the
Species Mean Acute Value for a species of commercial or recreational
importance.  The proposed Tier I methodology differed from the 1985 National
Guidelines by specifying that the FAV should be lowered only for a species
that is recreationally or commercially important to the Great Lakes System.
For example, the proposed CMC for cadmium was lowered from 2.3 M9/L to 2.1
     to protect the Chinook salmon.
      b.    Comments :  Several commenters noted that the commercially and
recreationally important species used to lower some of the criteria do not
reside in all receiving waters of the Great Lakes System.  Those commenters
suggested that the criteria should be lowered to protect such species where
they reside, and should not be lowered where they do not reside.

      EPA agrees that some of the recreational and commercially important
species used to lower criteria do not reside in all receiving waters of the
Great Lakes Basin.  However, many of these species are located over a large
area of the Basin and migrate throughout the lakes and tributaries.  EPA
believes it is reasonable to adjust the criteria for these species System-wide
instead of site-specifically as advocated by the commenters.  The current
National program and the final Guidance allow States and Tribes to modify the
aquatic life criteria on a site -specific basis to account for the situation
where the species does not reside at a particular site.  To accomplish this, a
State or Tribe may choose to modify the criteria using the recalculation
procedure described, above for those waters where the commercially or
recreationally important species does not "occur at the site."

      c.    Final Guidance :   The final Guidance requires the lowering of the
FAV for commercially or recreationally important species to the Great Lakes
System for the reasons stated above .

3 .    Ecologically Important Species

      a.    Proposal :  In the proposal, EPA invited comment on whether the
Tier I methodology should define "ecologically important" species and include
provisions to lower' the FAV for such species.  EPA also invited comment on
whether "ecologically important" species were adequately protected by the
proposed Guidance without adding additional provisions.

      b.    Comments :  Many commenters expressed a preference not to include a
provision to lower the FAV to protect "ecologically important" species because
they believed that a reasonable definition could not be developed.  A few
commenters favored lowering the FAV to protect "ecologically important"
species.  A few commenters favored treating all species as "ecologically
important," but no commenter offered any other definition of "ecologically
important" species..  Some commenters also recommended adding the concept of
"culturally important."  Many of these commenters named certain species as
"culturally" or "ecologically important;" however, these suggestions did not
identify any new species meriting protection since the named species are also
recognized as commercially or recreationally important species.  Several
commenters suggested that a provision for "ecologically important" species
could be used to provide protection for threatened and endangered species .

      EPA agrees with commenters that it is not necessary to include a
provision for lowering the FAV for "ecologically important" species.  EPA
believes that the Tier I methodology as proposed, utilizing all available
data,  adequately protects "ecologically important" and "culturally important"
species.  Further, EPA is unable to develop meaningful and workable
definitions for these- terms that would distinguish ecologically or culturally
important species from other species .

-------
96     Water Quality Guidance for the Great Lakes System - Supplementary Information Document

      EPA agrees that threatened and endangered species should be protected,
but does not believe the use of "ecologically important" species is the best
mechanism.  Instead, EPA has added a provision to procedure 1 of appendix F to
part 132 specifying that States and Tribes must adopt site-specific criteria
modifications to protect threatened and endangered species that are listed or
proposed under the Federal Endangered Species Act.  This provision is
discussed further in sections II.G. and VIII.A. of this document.  EPA has
determined that site-specific modifications to ensure protection to threatened
or endangered species are more appropriate than modifying .the State-wide or
Tribal aquatic life criteria or values as done for commercially or
recreationally important species, since threatened or endangered aquatic
species tend to be isolated to specific sites.  States, Tribes, and EPA in
consultation with the FWS, should establish the geographic extent of the site-
specific modifications to be large enough to provide adequate protection to
the listed species.  If a threatened or endangered species listed under the
Federal Endangered Species Act is found throughout the State's or Tribe's
waters, then it may be more prudent to adopt a State-wide or Tribal criteria
or value rather than develop numerous site-specific modifications.

      c.    Final Guidance:  For the reasons stated above, the final Guidance
does not include a provision for lowering the FAV for "ecologically important"
species nor provides a definition for "ecologically important."  However,
under CWA section 132.4 (i) which allows States and Tribes to adopt more
stringent provisions, a State or Tribe may independently choose to lower the
FAV to protect "ecologically important," "culturally important," or any other
group of species that it defines.

4.    Elimination of Final Residue Value

      a.    Proposal:   For chronic exposures, EPA proposed to modify the 1985
National Guidelines approach by deleting the option of using the Final  Residue
Value  (FRV) in deriving the Criterion Continuous Concentration  (CCC).   The CCC
was proposed as the lower of the Final Chronic Value  (FCV) or the Final Plant
Value  (FPV).

      b.    Comments:  Many commenters supported eliminating the FRV.  A few
commenters advocated retaining the FRV because it yields lower criteria.

      EPA agrees with the  comments that the FRV should not be utilized  in the
calculation of the CCC.  As explained in the proposal  (58FR20851), the  FRV is
intended to prevent concentrations of pollutants in aquatic species from
affecting the marketability of those  species or affecting the wildlife  that
consume aquatic  life.   Consequently,  EPA believes use of the FRV is redundant
with the methodologies  for human health and wildlife criteria contained within
the final Guidance.  EPA believes that those methodologies will yield more
appropriate criteria to protect humans and wildlife.  Furthermore, the  human
health and wildlife criteria derived  for dieldrin, endrin, and mercury  are
more stringent than the National aquatic life  criteria found in  Table III-4.

       c.    Final Guidance:  The  final Guidance does not require the use of
the FRV in the derivation  of the CCC  for the reasons stated above.

5.    Acute-Chronic Ratios

       a.    Proposal:   The Acute-Chronic Ratio (ACR) is a method of relating
acute  and  chronic toxicities.  To derive an ACR, comparable acute and chronic
toxicity  studies which  have  been conducted  under similar conditions for a
given  species are used.   From  comparable measurements of acute  and  chronic
values, an ACR is calculated by  dividing the measured acute value by the
measured  chronic value.   EPA proposed to follow the  1985 National Guidelines
by requiring ACRs for at  least three  families  of aquatic animals.  The  Final
Acute-Chronic Ratio (FACR) must  be either the  geometric mean of some or all  of
the species ACRs or another  value  appropriate  for  sensitive  species.

-------
                               Section HI: Aquatic Life                            97
      The proposal expressed a. preference for the use of freshwater ACRs when
deriving a Final Chronic Value  (FCV),  but allowed the use of ACRs for
saltwater species in the derivation of the FCV when the data set for deriving
a FCV does not contain three ACRs for freshwater species.  EPA invited comment
on the preference for freshwater ACRs in calculating a FCV to protect species
in the Great Lakes System.

      b.    Comments:   Some commenters preferred using saltwater ACRs only
when there were not-an adequate number of freshwater ACRs, while other
commenters wanted to use fresh and saltwater ACRs interchangeably, and others
suggested not using any saltwater ACRs.

       EPA agrees with those comments suggesting that saltwater ACRs should
only be used when there are an inadequate number of freshwater ACRs.  As
stated in the proposal, because the Great Lakes are freshwater lakes, the
preference is to use freshwater ACRs when deriving FCV for freshwater animals.
However, there may be situations where the freshwater data needed to derive a
freshwater ACR is limited.  In these situations, EPA believes it is
appropriate to use saltwater ACRs in place of the freshwater ACRs (see 1985
National Guidelines',  pgs.  14 and 15).

      c.    Final Guidance:  The final Guidance retains the provision that
saltwater ACRs may be used to derive a FCV when the data set does not contain
three ACRs for freshwater species as specified and if the freshwater and
saltwater ACRs are within a factor of 10 (as described in the aquatic life
Tier I criteria methodology, appendix A, VI.K.2 and the 1985 National
Guidelines).

6.    Bioavailability

      a.    Proposal:   The criteria for metals in Tables 1 and 2 of part 132
were expressed as total recoverable concentrations.  EPA invited comment on
whether the bioavailability of contaminants was adequately addressed using
site-specific modification approaches, as well as alternatives to address the
issue of expressing toxicity of both bioavailable and total contaminant
concentrations.

      Subsequent to the proposal, EPA issued a memorandum to all EPA Regional
Water Management Division Directors providing policy and guidance on the
interpretation and implementation of aquatic life criteria for the management
of metals  (The Office of Water Policy and Technical Guidance on Interpretation
and Implementation of Aquatic Life Metals Criteria (Prothro, 1993) .   The
memorandum covered a number'of areas including the expression of aquatic life
criteria, total maximum daily loads, permits, effluent monitoring, compliance,
and ambient monitoring.  With regard to expression of aquatic life criteria,
the memorandum recommended that State water quality standards be based on
dissolved metals because dissolved metal concentrations more closely
approximate the bioavailable fraction of metal in the water column than does
total recoverable metal concentrations.  However, because the present National
aquatic life criteria were developed using total recoverable measurements, it
is necessary to use, a conversion factor to convert the total recoverable metal
concentrations to equivalent dissolved metal concentrations.  The Prothro
(1993)  memorandum suggested conversion factors for 10 metals.

      Given EPA1s policy and guidance and the comments received on this issue,
EPA made available data to convert the proposed total recoverable metals
criteria to the comparable dissolved metal criteria on August 30, 1994 (59 FR
44678) .   EPA invited comment on the procedure used to derive conversion
factors for the nine proposed metals criteria in the "Draft Report:   Results
of Simulation Tests Concerning the Percent Dissolved Metal in Freshwater
Toxicity Tests  (Stephan, 8/1994) .  EPA also invited comment on the data
collected utilizing-the procedures as well as1the calculated conversion
factors for seven of the nine proposed metals.  However, this draft report did

-------
98     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

not contain data or conversion factors for cadmium or mercury(II) because the
testing vas not yet complete for those metals at the time of the notice.

      b.    Comments:  Numerous commenters stated that the proposed criteria
for metals overestimate the bioavailability of metals and that recent
technical and policy developments of EPA suggest the criteria are
overprotective.  These same commenters urged EPA to modify the proposed
Guidance to base the numeric metals criteria on the dissolved metal fraction
instead of the total recoverable metals approach used in the proposed
Guidance.  Some commenters, however, stated a preference for expressing the
criteria as total recoverable concentrations.  Many commenters asked EPA to
maintain the flexibility for the State or Tribe to choose the form of
pollutant used to develop criteria.
      Comments received from the August 30, 1994, Federal Register notice
generally favored using dissolved metals criteria for the final Guidance.
Many commenters agreed that repeating all of the toxicity tests to generate
new dissolved metals criteria was too resource-intensive and that the
simulation tests contained in the report were technically preferable.   Some
commenters also stated that dissolved metals criteria may allow potential
increases in metals loadings and potential impairment to the aquatic ecosystem
by ignoring fate and transport of particulate bound metals.

      Some commenters suggested that, except for chromium(III) and lead, EPA
should treat the current total recoverable metal criteria as dissolved
concentrations  (i.e., make the conversion factor 1.0).  Many commenters
requested the opportunity to comment on any new data used to derive the
conversion factors for cadmium and mercury(II).

      Although the draft report (Stephan, 1994) did not discuss implementation
of the dissolved metals criteria,  EPA received considerable comment regarding
implementation.  Generally commenters felt that it was important for EPA to
specify in the final Guidance how the dissolved criteria should be
implemented.  Commenters cautioned EPA against adoption of the dissolved
criteria without guidance on Water Effect Ratios  (WERs), modelling,
translation for limits, clean analytical and sampling methods and use of
appropriate historical data.  Some commenters suggested that EPA adopt  the
Prothro  (1993) memorandum into the final Guidance and others suggested  that
existing guidance was insufficient and EPA should update and/or revise
existing implementation guidance.  Comments were received regarding use of the
acidification method in place of the total recoverable method in the
simulation tests and the implications to analytical testing for compliance
purposes.

      EPA agrees with comments that, in general, the dissolved metal fraction
more closely approximates the bioavailable fraction of metal in the water
column than does total recoverable metal.  Aquatic life criteria are designed
to protect aquatic organisms from water column toxicity.  The primary
mechanism for water column toxicity is adsorption at the gill surface which
requires metals to be in the dissolved form.  The use of the dissolved  form of
the metal will, therefore, better approximate the toxicity to the aquatic
organism.

      This does not suggest, however, that the expression of metals criteria
as total recoverable is not scientifically defensible, nor does  this imply
that State and Tribes are required to adopt  the dissolved metals criteria.
EPA agrees with commenters that States and Tribes should be allowed to  choose
the form of pollutant for which to develop criteria.  EPA, while stating a
preference for dissolved metals criteria in  the final Guidance,  realizes that
there may be situations, such as  consideration of sediments or food chain
effects, where a State  or Tribe believes the expression of metals criteria as
total recoverable is preferable.  EPA will allow  the States and  Tribes  the
flexibility to adopt total recoverable criteria for metals as stated in the
Prothro  (1993) memorandum.

-------
                               Section ID: Aquatic Life                             99
      EPA recognizes that many of tne conversion factors published in the
draft report are very close to 1.0.  EPA chose to create experimentally
derived conversion factors rather than to assume that the existing total
recoverable criteria to be 100 percent dissolved.  EPA believes that this
approach is more technically sound.

      EPA recommends use of "The Office of Water Policy and Technical Guidance
on Interpretation and Implementation of Aquatic Life Metals Criteria"  (U.S.
EPA, 1993)  and the report titled "Interim Guidance on Determination and Use of
Water Effect Ratios for Metals" (U.S.EPA, 1994) for guidance on implementation
of dissolved metals criteria.   U.S.EPA (1993) contains guidance on dynamic
modelling and translators (Attachment #3),  and monitoring (Attachment #4).
U.S. EPA (1994) presents an effluent-specific approach for calculating a total
recoverable permit limit from a dissolved criterion (see pages 116 and 128-130
of U.S. EPA, 1994).  This approach is based on the percent of the total
recoverable metal in the effluent experimentally determined as described on
pages 112 and 125 (U.S. EPA, 1994).  A similar approach can be used to
calculate a permit limit for criterion expressed as free cyanide; in this case
the calculation is based on the percent of the total cyanide in the effluent
that becomes free cyanide in the downstream water.  EPA will continue to
update implementation guidance as needed in the future.

      For purposes of this rulemaking, Attachment #2 of U.S. EPA  (1993) is
superseded by the conversion factors and data in the report titled "Derivation
of Conversion Factors for the Calculation of Dissolved Freshwater Aquatic Life
Criteria for Metals" (U.S. EPA, 1995).  These conversion factors are also
found in Table III-l.

      EPA used the acidification method in the simulation tests, in place of
the total recoverable method,  because the metals potentially bound within the
test water would have been released by acidifying the samples.  The
acidification method is generally more accurate because it produces smaller
coefficients of variation than the total recoverable method.  In whole
effluent, metals may form stronger bonds to other components of the effluent.
Hence, for measuring total metals in effluent the total recoverable method is
the appropriate method.  The steps of heating and digesting the sample in the
total recoverable analytical method were not deemed necessary for purposes of
the simulation tests.  Moreover, the fraction of metal in the dissolved phase
may be different between effluent and surface waters.   In order to account for
all the metal being discharged into surface waters, and then calculate the
fraction that is dissolved,  the permitting authority must know the total
amount of metal in effluent.  Use of the acidification method was appropriate
for the simulation tests, but this method is not being considered for
regulatory use.

      Due to resource constraints,  EPA did not complete additional testing
required to update the conversion factors for mercury given in earlier EPA
guidance (Prothro, 1993) .  "The Office of Water Policy and Technical Guidance
on Interpretation and Implementation of Aquatic Life Metals Criteria"
(Prothro, 1993) did not contain conversion factors for the CMC for selenium or
for the for selenium and mercury.   Prothro (1993) , however,  did contain a
conversion factor of 0.85 for the National mercury CMC.   EPA made available
data to convert the proposed total recoverable metals criteria to the
comparable dissolved metal criteria on August 30, 1994 (59 FR 44678) for
selenium and mercury.  The conversion factor for the CMC for mercury came from
Prothro  (1993)  and this value was proposed in the August 30, 1994, notice for
the mercury CCC.

      Conversion factors were derived for selenium, even though Prothro  (1993)
stated that it is not appropriate to adjust the CMC and CCC for selenium or
the CCC for mercury because these are bioaccumulative chemicals.  Regardless
of whether these metals bioaccumulate, the important consideration is the
exposure that relates to the effect on which the CMC or CCC is based.  The CMC
is based on acute toxicity and so a relevant consideration is the

-------
100    Water Quality Guidance for the Great Lakes System ~ Supplementary Information Document

bioavailability of the metal in the water column; therefore, a total
recoverable CMC may be converted to a dissolved CMC if an appropriate
conversion factor is used and if there are no unacceptable risk management
considerations.  For a CCC, the exposure can be from the water column and from
the food, with the food chain consisting of some organisms whose primary
exposure is to pollutants in the water column and other organisms whose
primary exposure is to pollutants in the sediment.  It appears that exposure
to the sediment contributes substantially to the concentration of mercury in
the food chain, but does not contribute substantially to the concentration  of
selenium in the food chain.  Therefore, it is as acceptable to convert  the
total recoverable CCC for selenium to a dissolved CCC as it is to convert,  for
example, the total recoverable CMC for copper to a dissolved CMC.  In
contrast, it is not acceptable to convert a total recoverable CCC for mercury
to a dissolved CCC if the CCC for mercury is based on mercury residues  in
aquatic organisms.  It can, however, be acceptable to convert a CCC for
mercury to a dissolved CCC if the CCC is based on the toxicity of mercury to
aquatic organisms because the important exposure is through the water column;
as before, a dissolved criterion may be derived if an appropriate conversion
factor is used and if there are no unacceptable risk management
considerations.  Although a conversion factor cannot be used to derive  a
dissolved CCC for mercury if the CCC is based on mercury residues in aquatic
organisms, this does not necessarily mean that losses due to fate and
transport processes, such as volatility, cannot be taken into account in the
derivation of a permit limit.

      EPA believes that use of the CMC conversion factor to convert the total
recoverable mercury* CMC and CCC is technically preferable to having no
conversion factor for the mercury(II) CCC or requiring the CCC for mercury  be
expressed as total recoverable.  EPA will make available any new information
regarding alternative conversion factors for the mercury(II) CCC.

      EPA has completed testing to determine acceptable conversion factors  for
cadmium.  Because these conversion factors are substantially different  from
those in the Prothro  (1993) guidance, EPA will request public comment on the
data used to derive these factors.  EPA intends to amend the final Guidance
for the aquatic life cadmium criteria after comment is received on the  data
for the cadmium conversion factors.

      One commenter suggested that EPA consider the relationship of the
dissolved concentration of lead to hardness.  In the simulation tests it was
found that the dissolved concentration consistently fluctuated with variations
in hardness.  Therefore, EPA has amended the conversion factors for lead to
account for this hardness relationship.  Although Tier I criteria are not
provided for lead, EPA has included a conversion factor for lead in the final
report  (Stephan, 1995) for computing dissolved Tier II values or Tier I
criteria.

-------
                               Section ffl: Aquatic Life
101
                                  Table  III-l
1
Metal
Arsenic (III)
Cadmium
Chromium (III)
Chromium (VI )
Copper
Mercury (II)
Nickel
Selenium (IV)
Zinc
Recommended Conversion Factors
CMC
1.000
0.850
0.316
0.982
0.960
0.850
0.998
0.922
0.978
CCC
1.000
0.850
0.860
0.962
0.960
0.850
0.997
0.922
0.986
These recommended conversion factors are given to three decimal places  and are
not rounded off because  they are intermediate values in the calculation of
dissolved criteria.'

-------
102    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      c.    Final Guidance:  For the reasons stated above, the final Guidance
contains metals criteria expressed as dissolved concentrations and as total
recoverable concentrations.  EPA converted the proposed criteria for total
recoverable metals to their comparable dissolved criteria using the conversion
factors found in Table III-l.  The final conversion factors and supporting
data are contained in the Final Report: "Derivation of Conversion Factors for
the Calculation of Dissolved Freshwater Aquatic Life Criteria for Metals"
(U.S. EPA, 1995) .  The conversion factors for the final GLI mercury(II) CMC
and CCC are 0.85 from "The Office of Water Policy and Technical Guidance on
Interpretation and Implementation of Aquatic Life Metals Criteria" (Prothro,
1993) .   EPA has also included the conversion factors from the Prothro  (1993)
guidance for cadmium in the final Guidance.  EPA intends to amend the final
Guidance for the aquatic life cadmium criteria after comment is received on
the data for the cadmium conversion factors.  States and Tribes may adopt
criteria as total recoverable metals.  The total recoverable analytical
methods found in 40 CFR § 136 must be used for purposes of compliance
monitoring in NPDES permit limits.

7.    Averaging Period/Frequency of Exceedance

      a.    Proposal:  The proposed Guidance, consistent with the current
National guidance, required that, except possibly where a locally important
species is very sensitive, aquatic organisms and their uses should not be
affected unacceptably if the following conditions are met:  for chronic
criteria, the four-day average concentration of a chemical does not exceed the
CCC or Secondary Continuous Concentration  (SCO more than once every three
years on the average; for acute criteria, the one-hour average concentration
of a chemical does not exceed the CMC or Secondary Maximum Concentration  (SMC)
more than once every three years on the average.

      Averaging periods are time periods over which ambient concentrations are
to be averaged to determine whether criteria are exceeded.  If the mean
ambient concentration of a pollutant exceeds the criteria over the averaging
period, adverse impacts on the resident aquatic life could occur.  Averaging
periods are one means of accounting for the exposure time required to elicit
toxic effects.

      An allowable frequency for exceeding  the criteria is incorporated into
the  criteria because it is not necessary for concentrations to be below
criteria at all times in order to adequately protect aquatic ecosystems.
Also, it is not generally possible to ensure that criteria are never exceeded.
Frequently, concentrations above criteria may occur without corresponding
impacts on  the aquatic biota if the duration is less than the averaging
period.  This is dependent on the magnitude and duration of the exceedance.

      b.    Comments:  A few commenters questioned the validity of the one-
hour and four-day averaging periods, and the once-in-three-year frequency
stating that these components may be overly stringent and inappropriate for
chemicals  such  as bioaccumulative chemicals of concern.  Commenters stated
that a  four-day averaging  period  is  too short for bioaccumulative chemicals,

      EPA agrees that the  averaging periods proposed in the Tier I and II
methodologies will protect ecosystems  from  the effects of toxicants to which
they are  applied but that  other averaging periods may also be appropriate for
certain pollutants.  The one-hour averaging period proposed for the CMC and
SMC  and the four-day averaging period  or duration for the CCC and SCC  were
based on  fast acting toxicants.  Because different pollutants cause effects to
aquatic organisms at different rates,  these assumptions might be overly
protective for  some pollutants in some situations.

      EPA's rationale for  the once-in-three-year frequency proposed for acute
and  chronic criteria and values  is presented in the 1985 National Guidelines
and  is  also explained in  appendix D  of the  "Technical Support Document for

-------
                               Section HI: Aquatic Life                            103
Water Quality-Based Toxics Control" (U.S. EPA, 1991) which is included in the
docket for this rulemaking.

      c.    Final Guidance:  EPA believes that attainment of the one-hour or
four-day averaging periods and the once-in-three-year frequency will protect
ecosystems from adverse effects of toxicants to which they are applied.
Consistent with the available data, however, EPA agrees with commenters that
alternate averaging periods and frequencies may also be appropriate for
certain pollutants.  Consequently, the final Guidance allows the use of
alternative averaging periods and frequencies if States or Tribes can
demonstrate to EPA that they are scientifically defensible.

      In evaluating whether an alternative averaging period is scientifically
defensible, EPA will pay particular attention to those toxicity tests which
account for the time needed to elicit effects.  Greatest emphasis will be
placed on results for sensitive species  (e.g., those affected at relatively
low concentrations).  In assessing lethality from acute or chronic toxicity
tests, EPA will pay particular attention to those relationships (a) between
effect concentrations and time of exposure, or (b) between time to death and
concentration.  In assessing sublethal (e.g., growth and reproductive)
effects, EPA will consider data from growth or reproductive tests comparing
time-varying exposure with steady exposure.  In such tests, evidence would be
considered supportive of a longer-term averaging period where time-variable
and steady concentrations yield the same growth or reproductive effects.

      In considering alternative frequencies of exceedance, EPA recognizes
that a substantial excursion above a criterion is likely to affect more taxa
more severely than a marginal excursion,  and that the setting of the allowable
frequency for marginal excursions implicitly sets even lower frequencies for
more severe excursions.  Consequently, in evaluating whether an alternative is
scientifically defensible, EPA will consider assessments that either
explicitly or implicitly account for  (a)  the probable frequency of lethal
events for an assemblage of taxa covering a range of sensitivities to
pollutants; (b) the probable frequency of sublethal effects for such taxa;  (c)
the differing effects of lethal and sublethal events in reducing populations
of such taxa; and  (d) the time needed to replace organisms lost as a result of
toxicity.  For an assemblage of taxa for which toxicity data are available,
such an assessment should yield an overall measure of toxic impacts that would
result from the alternative excursion frequency being proposed.  One such
measure of toxic impacts could be the projected aggregate population deficit,
compared to a system unaffected by pollutant toxicity.

      Alternatively, EPA will consider information from well-designed field
biological surveys indicating the relationship between biological quality and
chemical criteria excursion frequency, provided that confounding factors such
as bioavailability are appropriately taken into account and providing that the
field surveys are adequately sensitive.

8.     Final Tier I Criteria

      a.    Proposal:  The proposal provided a detailed discussion of how EPA
selected the 16 pollutants for which Tier I criteria were calculated.  EPA
chose not to propose specific numeric criteria for ten pollutants  (aldrin,
aluminum, chlordane, chlorpyrifos, DDT, endosulfan, heptachlor, lead, PCBs,
and toxaphene) for which EPA has National aquatic life criteria for the
reasons stated in the proposal (See 58FR20852).   EPA requested comment on the
proposed alternative of requiring States and Tribes to adopt the current
National criteria for those 10 pollutants.

      EPA proposed acute criteria for 16 pollutants and chronic criteria for
15 pollutants  (See Tables 1 and 2 of proposed part 132) .  Footnotes within the
tables delineated which proposed criteria were hardness or pH dependent and
those criteria were listed at a hardness of 50 mg CaCO3/L  or a  pH  of  6.5.

-------
104    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      b.    Comments:  Several coiranenters asked EPA to perform the testing
needed to develop, or to develop, Tier I criteria for the ten remaining
pollutants.  Numerous commenters argued against using National Criteria where
the data set for the pollutant did not meet the proposed Tier I aquatic life
data requirements.  Some commenters favored using Tier II methods for these
pollutants.  One commenter thought that EPA was obligated to require adoption
of National aquatic' life criteria  (where Tier I criteria do not exist) by
States and Tribes to meet the "no less restrictive" clause of the Critical
Programs Act.

      EPA disagrees that it should be solely responsible for developing the
additional data to derive Tier I criteria for the 10 pollutants with National
criteria.  As discussed in section II.C. of this document, EPA in cooperation
with States and Tribes will establish a Clearinghouse to assist States and
Tribes in developing numeric Tier I water quality criteria and Tier II water
quality values.

      EPA agrees with commenters that States and Tribes should not be required
to adopt the current National criteria for the 10 pollutants that do not meet
the minimum data requirements for a Tier I criteria.

      In the preamble to the proposed Guidance, EPA explained that the
proposed Great Lakes criteria for aquatic life would not be less restrictive
than the National criteria because States and Tribes would be required  to
generate Tier I criteria or Tier II values to regulate these 10 pollutants.
EPA's position has not changed even though, as explained below, the final
Guidance provides States and Tribes the option of regulating pollutants
lacking a sufficient data set to derive Tier I criteria by either using Tier
II values or establishing limits based on an indicator parameter.  In the
National program, where a State or Tribe must regulate a pollutant that lacks
a promulgated numeric criterion, the State or Tribe may base a permit limit on
either a numeric value or an indicator parameter.  Consequently, EPA does not
consider its decision to allow States and Tribes to use indicator parameters
in place of Tier II values to make the final Guidance less restrictive  than
the National program for these 10 pollutants.

      EPA also notes that the final Guidance contains numeric Tier I criteria
for human health or. wildlife for chlordane, DDT, PCBs, and toxaphene that are
more stringent than the current National aquatic life criteria for the  same
pollutants.  These criteria will help ensure that the final Guidance will
protect aquatic life species as effectively as the National program for these
four pollutants.

      Several commenters requested that EPA provide the final acute and
chronic equations within the final rule.  EPA agrees that these equations
should be part of Tables 1 and 2.  EPA historically has provided criteria for
pollutants that are hardness dependent in the form of an equation that
accounts for hardness as part of the National criteria statement.

      c.    Final Guidance:  For the  reasons stated above, EPA has determined
that it is appropriate  to require States and Tribes to regulate these 10
pollutants by developing Tier II values.  The final Guidance includes the
acute and chronic hardness-dependent  criteria for cadmium, chromium(III),
copper, nickel, and  zinc.  The hardness and pH equations are found in Table
III-2 as well as each individual criteria document for the above chemicals.

      Table III-3 presents the CMCs calculated using the Tier I methodology
for aquatic life.  CMCs for metals are expressed as dissolved concentrations
and total recoverable concentrations.  Conversion factors utilized to convert
the metals criteria* from total recoverable  concentrations to dissolved
concentrations are found in Table  III-l.  For comparison the CMCs of existing
National criteria are also included.  Hardness-dependent CMCs for cadmium,
chromium(III), copper,  nickel, and zinc are found in Table III-2.  A pH-
dependent CMC for pentachlorophenol is also found in Table III-2.

-------
                               Section HI: Aquatic Life                           105
      A technical support document, Great Lakes Water Quality Initiative
Criteria Document for the Protection of Aquatic Life in Ambient Water  (EPA
820-B-95-004) (Final Criteria Documents), (available in the administrative
record to this rulemaking) presents the derivation of each of the final Tier I
CMCs and the toxicity studies from which the criteria were derived.  The final
Guidance requires that the numeric criteria in Table 1 to part 132 modified as
appropriate to reflect site-specific conditions  (or more stringent criteria)
be adopted by the Great Lakes States and Tribes and incorporated into their
ambient water quality standards.  The specific requirements on how these
criteria are to be incorporated into State and Tribal water quality standards
are discussed in section II  (Regulatory Requirements) of this document.

      Table III-4 presents final CCCs calculated using the final Tier I
methodology for aquatic life.  Final CCCs for metals are expressed as
dissolved concentrations and total recoverable concentrations.  Conversion
factors utilized to convert the metals criteria from total recoverable
concentrations to dissolved concentrations are found in Table III-l.  For
comparison the CCCs'in existing National criteria are also included.
Differences between proposed and final Great Lakes Tier I CMCs and CCCs are
described later in this section.

      The derivation of each of these final CCCs, and the toxicity studies
upon which they are based, are also discussed in Final Criteria Documents.
This document is available in the administrative record for this rulemaking.
Hardness-dependent CCCs for cadmium, chromium(III), copper, nickel, and zinc
are found in Table III-2.  A pH-dependent CCC for pentachlorophenol is also
found in Table III-2.  Criteria may be calculated at different concentrations
of hardness  (measured as mg/L of CaCO3  or pH.   For example,  using the
equations given in Table III-2, the total recoverable CMCs for zinc at a
hardness of 10,  50, 100, or 200 mg/L CaC03 are  17,  67,  120,  or 220  ptg/L
respectively.  The hardness equations used to calculate hardness dependent
freshwater criteria for metals  (i.e., cadmium,  chromium(III), copper, nickel,
and zinc)  can be used at any hardness.  Most of the data used to develop these
hardness formulas were in the range of 25 mg/L to 400 mg/L CaCO3  and the
formulas are therefore most accurate in this range.  Irrespective of this data
set, for waters with a hardness less than 25 mg/L CaCO3 criteria  should be
calculated using the actual ambient hardness of the surface water.  Limiting
use of the equation only to hardness above 25 could result in underprotective
criteria where the actual ambient hardness is below 25 mg/L CaC03.   The
majority of waters nationwide have a hardness of less that 400 mg/L CaCO3.   If
however, the hardness is over 400 mg/L CaCO3, two options  are  available for
the State or Tribe:  1)  use 400 mg/L CaC03 for  the criteria calculation or 2)
require an analysis to calculate a water-effect ratio for the site with
hardness above 400 mg/L CaCO3 and modify the  final criteria concentration
using the calculated water-effect ratio.  Use of a water-effect ratio in this
instance is recommended at a hardness above 400 mg/L CaCO3 because  other
confounding factors,  which may cause this hardness, can also affect the
toxicity of the metal.

      The final Guidance requires that the numeric criteria in Table 2 of part
132 (or more stringent criteria) be adopted by the Great Lakes States and
Tribes and incorporated into their ambient water quality standards.  The
specific requirements on how these criteria are to be incorporated into State
and Tribal water quality standards are discussed in section II of this
document.

-------
106   Water Quality Guidance for the Great Lakes System — Supplementary Information Document
                              Table III-2

                        Hardness and pH Equations'
Chemical
Cadmium
Chromium (III)
Copper
Nickel
Pentachlorophenol
Zinc
Final Acute Equation
gl-128 (tahMdncM) -3.6867
gO.819 (to budncM) + 3.7256
gO.9422 On huJnow) - 1.700
gO.846 On Imrinew) + 2.255
e 1. 005 (J>H)- 4.869
gO.8473 (In hMdnew) + 0.884
Final Chronic Equation
e0.7852 (ta budneu) - 2.715
pO-819 (In hudKM) -1-0.6848
gO.8545 (IntanJneM) - 1.702
gO.S4« (tn hardneu) -f 0.0584
gl.005 (pH) -5.134
gO.8473 (In hudncw) + 0.884
           These  equations  are  for  criteria expressed  as  total
           recoverable concentrations.• Conversion factors should be
           applied to the resulting value if  criteria expressed as
           dissolved concentrations are desired.

-------
                         Section HI: Aquatic Life
107
                            TABLE III-3
       Acute Ambient Water Quality Criteria for Aquatic Life
Chemical
Arsenic (III)
Cadmium b
Chromium (II I) b
Chromium (VI )
Copper b
Cyanide , free
Dieldrin
Endrin
Lindane
Mercury (II)
Nickel b
Parathion
Pentachlorophenol °
Selenium
Zinc b
Great Lakes
Final CMC*
(dissolved)
340
1.8
320
16
7.0
n/a
n/a
n/a
n/a
1.4
260
n/a
n/a
18
65
Great Lakes
CMCs"
(total)
339.8
2.1
1000
16.02
7.3
22
0.24
0.086
0.95
1.694
260
0.065
5.3
19.34
67
National
CMC"
360
1.8
980
16
9.2
22
2.5 «
0.18 "
2.0 d
2.4
790
0.065
5.5
20
65
      All values are in /ig/L.

      The toxicity of this chemical is hardness-related;  the criterion
      expressed is at a hardness of 50 mg/L.

      The criterion for this chemical is pH dependent;  the criterion
      expressed is at a pH of 6.5.

      This value is an FAV that was calculated according  to the 1980
      guidelines.  Although the CMC = FAV/2 in the  1985 National
      Guidelines, there is no CMC in the 1980 guidelines  and the
      procedure used to derive the FAV is different from  that used in
      the 1985 National Guidelines.
Note: The term "n/a" means not applicable.

-------
108
Water Quality Guidance for the Great Lakes System — Supplementary Information Document
                                   Table III-4
            Chronic Ambient Water Quality Criteria for  Aquatic Life
Chemical
Arsenic (III)
Cadmium b
Chromium (II I) "
Chromium (VI)
Copper b
Cyanide, free
Dieldrin
Endrin
Mercury (II)
Nickel b
Parathion
Pentachlorophenol c
Selenium
Zinc b
Great Lakes
Final CCC"
(dissolved)
150
1.2
42
11
5.0
n/a
n/a
n/a
0.77
29
n/a
n/a
4.6
66
Great Lakes
CMCs*
(total)
147.9
1.4
49
10.98
5.2
5.2
0.056
0.036
0.9081
29
0.013
4.05
5
67
National
CCC*
190
0.66
120
11
6.5
5.2
0.0019 d
0.0023 d
0.012 d
88
0.013
3.5
5.0
59
      '     All values in

      b     The toxicity of this chemical is hardness-related;  the criterion
            expressed is at a hardness  of 50 mg/L.

      c     The toxj.city of this chemical is pH related; the  criterion
            expressed is at a pH of  6.5.

      d     Based upon Final Residue Value.

      Note:   The term "n/a" means  not applicable.

-------
                               Section ffl: Aquatic Life                           109
9 .     Tier I Criteria

      a.    Proposal:   The proposal contained CMCs  (i.e., acute criteria) for
sixteen pollutants calculated using the proposed Tier I methodology for
aquatic life.  CMCs were proposed for arsenic(II), cadmium, chromium(III),
chromium(VI), copper,  cyanide, dieldrin, endrin, lindane, mercury, nickel,
parathion, pentachlorophenol,  phenol, selenium, and zinc.  CCCs  (i.e., chronic
criteria)  were proposed for fifteen of the sixteen chemicals listed above.
EPA did not propose a CCC for lindane because the minimum data requirements
were not met.

      b.    Comments:   Several commenters noted the lack of information or
lack of explanation in the proposed aquatic life criteria documents.  For
example, it was pointed out that EPA did not present the ACRs in the
arsenic(III) document.  In addition, commenters noted several small errors
within the documents,  such as mathematical calculation errors.

      Several commenters questioned the data used in the derivation of
criteria.   For example, for mercury EPA chose a 96-hour I,CX for Cranqonvx
which was two to three orders of magnitude lower than the 48-hour IiCK.
Several commenters indicated that some criteria derived were unrealistically
stringent, but generally did not present data to support this opinion.

      c.    Final Guidance:  EPA agrees with comments that the criteria
documents could have been more descriptive in the derivation of intermediate
values such as the FACR and in how the final CMCs and CCCs were calculated.
In response to these comments EPA has attempted to better explain how the
criteria were derived.  EPA notes that the most recent National aquatic life
criteria document and the final GLI document together contain the complete
data set for the chemical.  In addition, the final criteria for endrin,
cadmium, mercury, arsenic(III), lindane, nickel, zinc, copper, dieldrin, and
parathion were changed from the proposal for the reasons cited below.  The
proposed criteria concentrations for chromium(III), chromium(VI), cyanide, and
selenium are the same as the proposal.  Some explanatory language was added to
all the documents to better explain how the criteria were derived and the
documents were slightly reformatted.

      Some data were removed in response to comments.  Chemicals for which
data were deleted are cadmium, endrin, and mercury.  The proposed endrin
document stated that some acute toxicity data from the EPA endrin criteria
document were not used in development of the GLI criteria because the test
protocols did not meet current acceptable toxicity testing procedures.  Some
of these data, however, were not deleted and was used to calculate the
proposed endrin criteria.  EPA has deleted this data from the criteria
derivation for the final GLI endrin criteria.  This resulted in a slight
change from the proposed criteria.

      In reviewing the cadmium document EPA ascertained that the
concentrations of cadmium were not measured in a chronic test with Moina.
This data was deleted along with several acute and chronic tests conducted in
river water (Spehar'and Carlson, 1984).  The Spehar and Carlson  (1984) tests
were deleted because variables within the water which might affect the
toxicity of cadmium are unknown.  These changes resulted in raising the CCC by
nearly a factor of two.  In the proposed cadmium document, the range of
Species Mean Acute Value  (SMAVs) was greater than a factor of five for the
genus Morone.  Because of this wide range, EPA set the Genus Mean Acute Value
(GMAV) for Morone equal to the lowest SMAV for that genus.

      In the final cadmium document, EPA also set that the GMAV for Daphnia
equal to the lowest SMAV because the range of SMAVs was greater than a factor
of five.  EPA also has made the Genus Mean Chronic Value (GMCV) for Daphnia
equal to the lowest SMCV for this genus.  These changes also contributed to
the1 the higher CCC for cadmium.  The CMC did not change as a result of these
corrections.  EPA believes the approach used for selecting the GMAV is more

-------
110    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

appropriate than the approach used for selecting the GMCV when there  is a wide
range in the SMAVs or SMCVs.

      In the mercury document, the 96-hour LCjo for Cranqonvx was deleted and
the 48-hour LCy, was not used because  these values were from the same  test.
EPA determined that the concentration measured was too great a change from 48
to 96 hours, making, the test or measurements suspect.  This change resulted in
an increase in both the CMC and CCC.

      Some of the data which was listed in the references and used to derive
the criteria did not appear in the data tables.  For example, the ACRs used in
the derivation of the arsenic(III) CCC did not appear in Table 3 as
referenced.  EPA did include all of this data in the references and most of
the information was located in the National arsenic criteria document.  EPA
has amended the final arsenic(III) GLI criteria document to include all data
referenced.  References and data within the tables were examined and  similar
corrections were made as necessary to the documents for arsenic(III), copper,
dieldrin, and endrin.

      Minor mathematical and rounding errors were found in the proposed
criteria arsenic(III), cadmium, endrin, lindane  (acute), nickel, and  zinc.
Simple mathematical errors in addition, subtraction, multiplication and
division were corrected where found.  In addition, all the documents  were
checked to ensure that intermediate values were not rounded.  The criteria for
lindane and nickel did not change as  a result of these corrections.

      A new FACR was calculated for zinc.  The proposed CCC was derived with a
FACR which contained a saltwater ACR.  Upon additional review of the  data, EPA
determined that this saltwater ACR was not needed.  The FACR for zinc was
recalculated based only on freshwater ACRs.  This changed the CCC some what
and made it equal to the CMC.

      EPA found similar data trends  (wide ranges in SMAVs) in the copper,
dieldrin, and parathion data sets.  EPA consistently applied the procedure of
making the GMAV or GMCV equal to the  lowest SMAV or SMCV  (respectively) when
the range of SMAVs or SMCVs is greater than five.  EPA believes  that  this is
consistent with  section XI.B. of appendix A to part 132 which asserts that
appropriate modifications of the methodology shall be used consistent with
sound scientific evidence where warranted.  EPA believes that this
modification is warranted to ensure protection to the more sensitive  species
within a genus.  In this situation, the mean of the data could  result in
underprotection  for these species if  the genus is very sensitive.  EPA,
therefore,  consistently applied this  procedure.  Applying this procedure to
copper, dieldrin, and parathion did not result in changes to the proposed
criterion concentrations.

      After the Tier I criteria for phenol were proposed, EPA reevaluated the
data used to derive the criteria.  EPA has determined that at this time there
is insufficient  data of adequate quality to finalize the CCC for phenol.  EPA
also realized that not all  of the data which could be used to derive  the CMC
and CCC for phenol were made available for public comment at the time of the
proposal.   Because of the withdrawal  of some of data used to derive the
proposed criteria and the  inability for public comment on other data  available
prior to proposal, EPA will not finalize the proposed criteria  for phenol.
EPA will review  the- additional data which was not used in the proposal  and
make it available through  the clearing house once established.

10.   Potential  Changes to  National Guidelines

      a.    Proposal:  In  the proposal, EPA noted that the 1985 National
Guidelines  were  being revised.  To  date, no revisions have been proposed  for
the  1985 National Guidelines.

-------
                               Section ffl: Aquatic Life                            111
      b.    Comments:   Several commenters asked for clarification about what
will happen when the 1985 National Guidelines are revised.

      c.    Final Guidance:  Due to current resource constraints on EPA,
revisions to the 1985 National Guidelines are not anticipated in the near
future.  When revised National Guidelines are proposed, EPA will announce
whether the revisions should replace or supersede any portion of the final
Guidance.  Until such time, Great Lakes States and Tribes are required to use
the final Tier I methodology herein.

C.    Final Tier II Methodolocrv

      The proposed Guidance required the use of a Tier II value where a State
or Tribe lacked the-full eight families of toxicity data needed to set a
chemical-specific Tier I criterion for a pollutant.  As proposed, the Tier II
methodology would allow derivation of a chemical-specific value using toxicity
data for as few as one species.  It contained adjustment factors to compensate
for the missing data.   EPA proposed to require use of methodologies consistent
with the Tier II methodology to interpret narrative criteria (such as the
prohibition on the discharge of no toxic chemicals in toxic amounts) in all
cases where a State or Tribe determined using proposed procedure 5 of appendix
F to part 132, that an effluent discharging into the Great Lakes system
contained a pollutant in an amount that causes, has the reasonable potential
to cause, or contributes to an excursion above any water quality standard,
including the narrative criterion for water quality.  EPA also proposed to
require the use of methodologies consistent with the Tier II values as the
applicable water quality standard for use in "reasonable potential"
determinations under procedure 5 of appendix F to part 132 where there is no
Tier I criterion nor sufficient data to calculate Tier I criteria.

1.    Requirement for use in Interpreting the Narrative Toxics Criterion

      Under the current National permitting requirements, where a State or
Tribe determines that the discharge of a pollutant causes, has reasonable
potential to cause,, or contributes to an excursion above a State's or Tribe's
narrative criterion, the State or Tribe must, in the absence of a numeric
criterion for that pollutant, interpret the narrative criterion and include
water quality-based effluent limitations (WQBELs)  in the permit that are
derived from and comply with the narrative as interpreted.  Depending on the
case-specific circumstances, such limits may currently be required for the
specific pollutant of concern and/or for whole effluent toxicity  (WET)  (see 40
CFR section 122.44(d)(1)(vi)).

      a.    Proposal:   The Tier II methodology developed for the proposed
Guidance provided a consistent means for interpreting narrative criteria that
protect aquatic life for all toxic pollutants with a very small number of
data.  EPA proposed to require States and Tribes to use the Tier II
methodology to interpret their narrative criteria and where necessary to
establish WQBELs.  Under procedure 6 of appendix F to part 132, EPA also
proposed the specific circumstances under which WET limits would be required
in a permit.  Both EPA and the Initiative Committees thought that requiring
permit limits based on Tier II values would increase consistency among Great
Lakes States and Tribes.  The Steering Committee also hoped that the
relatively stringent Tier II values would motivate some permittees to develop
the toxicity data needed to derive a Tier I criteria.

      At the same time,  EPA recognized that a facility with a WET limit could
argue that it had already reduced each of the individual pollutants in that
effluent sufficiently to protect the tested aquatic species in the receiving
water body (which often would be identical to the single species required as
the minimum data for a Tier II value),  and that further reductions in the
amounts of individual  pollutants based on a chemical-specific Tier II goal for
the water body were unnecessary.  Consequently, EPA asked for comment on the

-------
112    Water Quality Guidance for the Great Lakes System -- Supplementary Information Document


need for requiring both permit limits based on Tier II values and permit
limits for WET, as well as other options for harmonizing the two requirements.

      b.    Comments:  Many commenters suggested that the Whole Effluent
Toxicity (WET) criteria be used in place of the aquatic life Tier II values
and furthermore, that the Tier II methodology be published as merely guidance
to the States and Tribes.  Some commenters suggested that the Tier II values
and WET protocols overlap substantially.  These commenters recommended  that
both the Tier II methodology and guidance require that WET protocols be
utilized to interpret the narrative criteria, in lieu of the proposed Tier II
values.  Some commenters thought that values derived using the Tier II  method
would actually create less consistency among States and Tribes.  Other
commenters preferred the proposed option of requiring use of both the Tier II
values and WET.  ,

      EPA disagrees that WET be used in place of the Tier II aquatic life
methodology and that the Tier II methodology be published as guidance.  EPA
notes that the Tier II aquatic life methodology offers some significant
practical benefits for both the regulated community and the States and  Tribes.
For the regulated community, the chemical-specific Tier II approach offers the
advantage of allowing the permittee to focus immediately on a single
contaminant for the purposes of designing effluent treatment.  In contrast,
WET often leads to a facility conducting fairly extensive investigations to
identify the cause of adverse effects on the tested organisms and to develop
an effective approach to reducing the effects.

      EPA notes that an individual discharger will not always need to have
both Tier II and WET limits in its permit in order to protect the narrative
water quality standards.  However, EPA maintains as it did in its rulemaking
promulgated on June 2, 1989 at 40 CFR 122.44(d)(1)(vi)(C), that once a  finding
is made that the discharge of a pollutant causes, has the reasonable potential
to cause, or contributes to the excursion above the narrative criterion,
reliance on WET alone, in lieu of chemical-specific limits must only be done
where the discharger can demonstrate that WET sufficiently guards against
excursions above the applicable water quality standard  (i.e., the Tier  II
interpretation of the narrative standard).  EPA continues to believe that the
use of the Tier II methodology ie important for deriving Tier II values to
determine whether a pollutant has the reasonable potential to exceed a  Water
Quality-Based Effluent Limit  (WQBEL).

      c.    Final Guidance:  The final Guidance specifies the Tier II
methodology as the methodology States and Tribes will have to use in
interpreting their narrative water quality criteria for the protection  of
aquatic life.  Procedures 5 and 6 of appendix F to part 132 specify the
conditions under which chemical-specific and/or WET limits will be required
for an individual discharger.  Both kinds of limits will not always be
necessary or required for a discharger.  As described in section VIII.E of
this document, and consistent with 40 CFR 122.44(d)(1)(vi)(C), limits on
indicator parameters may be used in lieu of limits on the pollutant of  concern
when implementing the narrative criteria, including, a WET limit in lieu of a
chemical numeric limit.  However, when an indicator parameter is used,  the
State or Tribe must ensure that the indicator parameter will attain the Tier
II interpretation of the "applicable water quality standard"  (as described in
40 CFR 122.44(d)(1)(vi)(C)).

      It should be noted that a WET limit can not be used in lieu of a
chemical-specific limit where a total maximum daily load  (TMDL), developed
pursuant to 40 CFR 130.7 and approved by EPA, and corresponding wasteload
allocation specifies the acceptable level for that chemical in a discharger's
effluent.  A State, Tribe  or permittee in the Great Lakes System that wished
to base permit limits on a TMDL, would need to  generate a Tier I criterion or
a Tier II value  that could then be apportioned  among contributing sources(40
CFR 122.44(d)(1)(vii)(B) ).

-------
                               Section HI: Aquatic Life                            113
      Finally, it should be noted that Tier II values derived using the Tier
II methodology are the appropriate means of determining whether a pollutant is
or may be discharged into the Great Lakes System at a level which causes, has
the reasonable potential to cause, or contributes to an excursion above a
narrative criterion.  WET testing is not an appropriate substitute for this
chemical-specific determination.  EPA does not believe that the Tier II method
by itself would cause less consistency among the Great Lakes States and
Tribes.  Currently, States have different mechanisms to translate narrative
criteria.  EPA believes that the Tier II methodology enables States and Tribes
to translate their narrative criteria with the same mechanism.

2.    Data Requirements

      a.    Proposal:  The proposed Tier II methodology utilized all available
data and provided for derivation of a Tier II value when data sufficient to
derive a Tier I criterion are not available.

      b.    Comments:  Several commenters were concerned that the data
requirements for the Tier II methodology were not adequate.  Some commenters
recommended that EPA require a minimum of three to seven of the eight
taxonomic families specified.  A few commenters further suggested that EPA, at
a minimum, require data for a daphnid, rainbow trout and fathead minnow.

      c.    Final Guidance:  EPA believes that the Tier II minimum data
requirements are sufficient given the purpose of the Tier II methodology.  As
described in the proposal (58FR20835), the Initiative Committees sought to
ensure consistency among States in the Great Lakes System as to how limited
toxicity data are used to interpret narrative standards.   The proposed aquatic
life Tier II methodology fulfilled this goal.  Section 303(c)(2)(B) of the CWA
specifies that States and Tribes shall adopt criteria for all toxic pollutants
for which presence in the affected waters could reasonably be expected to
interfere with designated uses adopted by the State or Tribe.  The minimum
data required for the Tier II methodology promotes consistency in how this
requirement is implemented.  The proposal provided a standardized process for
utilizing available data to derive values for purposes of interpreting
narrative standards, thereby achieving greater consistency among the States
and Tribes in this activity.

3.    Other Methods.for Tier II Values

      a.    Proposal:  EPA's Science Advisory Board (SAB), in its report,
"Evaluation of the Guidance for the Great Lakes Water Quality Initiative,"
suggested the use of short-term and short-cut chronic toxicity tests to derive
a Tier II value.  The SAB believed this could overcome the cost of completing
standard chronic toxicity tests.  EPA invited comment on the appropriateness
of short-term and short-cut chronic tests for the derivation of Tier II
values.

      b.    Comments:  Many commenters did not feel short-term or short-cut
chronic tests were appropriate for criteria or value derivation.  Other
commenters supported the use of short-cut chronic tests,  but did not provide
any examples of acceptable short-cut tests.  One commenter did recommend a
short-term chronic test (7-day fathead minnow test, EPA/600/4-89/001).

      At this time, EPA does not consider short-cut chronic tests appropriate
for the derivation of a Tier I criterion or Tier II values.  EPA has serious
concerns over whether short-cut chronic tests can accurately predict effects
from long-term exposures.   A short-cut chronic test is a short-term exposure
to an aquatic organism during a sensitive life stage(s) which might yield
toxicity results similar to those obtained from complete life-cycle, partial
life-cycle, or early life-stage tests.  Although this type of test might yield
results that are similar to those from a complete chronic test for some
chemicals, significant exceptions can occur.  No commenter pointed to any
studies in which a short-cut chronic method was compared to a complete life-

-------
114    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


cycle toxicity test for the same species.  Without such validation or
verification of a short-cut chronic test, EPA is hesitant to allow its  use as
a replacement for a life-cycle, partial life-cycle or early life-stage  test.

      EPA does, however, think that the 7-day Ceriodaphnid test, which
measures survival, growth and reproduction,  is acceptable for use in deriving
Tier I criteria and Tier II values, because  this is a life-cycle test
according to the American Society for Testing and Materials (See ASTM
Definition in the Administrative Record).  EPA does not consider this test a
short-cut test because it is a life-cycle test.  The 7-day duration is
appropriate for a life-cycle test with species in the genus Ceriodaphnia,
whereas a 21-day duration is appropriate for species in the genus Daohnia.

      c.    Final Guidance:  The final Guidance does not allow data from
short-cut tests to be used in deriving a Tier I criterion or Tier II value.
However, the final Guidance does allow use of the 7-day Ceriodaphnid life-
cycle test for derivation of both Tier I criteria and Tier II values.

4.    Adjustment Factors

      a.    Proposal:  The proposed Tier II methodology uses adjustment
factors obtained in the statistical analysis described by Host,  et al.  (1991)
in the draft paper, "Analysis of Acute and Chronic Data for Aquatic Life,"
(which may be found in the administrative record for this rulemaking) to
derive Tier II values from data for one to seven of the eight taxonomic
families required for Tier I calculations.   Depending upon the number of Tier
I minimum data requirements satisfied by the data base, different adjustment
factors are applied to the lowest Genus Mean Acute Value to arrive at the
Secondary Acute Value (SAV).  These adjustment factors are intended to  relate
the results of one to seven toxicity tests to a FAV.  EPA invited comment on
the selection of an 80th percentile in establishing the proposed adjustment
factors.  EPA also invited comment on the use of a set of lower  adjustment
factors to be used only where daphnid data were available as opposed to the
higher adjustment factors that would be necessary if data for the specified
daphnids are not required.

      b.    Comments:  Several commenters also suggested that EPA set up a
hierarchy of toxicity testing requirements for Tier II.  This would establish
the order in which data are added to the Tier II data set.  Several of  these
commenters suggested that EPA establish the  hierarchy where more sensitive
species would be tested first.

      It would be very difficult to set up a hierarchy of toxicity testing
requirements for Tier II because there is a  range of sensitivities among
species that satisfy any one minimum data requirement and because the species
sensitivity varies with the pollutant under  consideration.  EPA  believes it
would be speculative to create such a hierarchy and impose it on diverse
pollutants which elicit varying toxic responses.

      Many commenters stated that the 80th percentile was too conservative and
recommended that EPA establish median or 50th percentile adjustment factors.
Some commenters  supported  the use of  the 80th percentile adjustment factors.
A few commenters recommended that the level  of protection for adjustment
factors be set at the 95th percentile to provide the same level  of protection
as a Tier I criterion.  Many of the comments supported the use of adjustment
factors with daphnid data  rather  than those  without.

      c.    Final Guidance;  When deriving Tier II values for chemicals for
which limited  data are  available  for  establishing permit limits, EPA desires
to be highly certain that  adequate protection is afforded.  EPA  must be highly
certain that Tier II values provide a level  of protection greater than  or
equal to a Tier  I criterion.  EPA believes  that use of a procedure that would
result  in less protection  than use of the  80th percentile would  cause Tier II
values  to be significantly less protective  than Tier I criteria. Use of the

-------
                               Section HI: Aquatic Life                           115
50th percentile would allow less protection than a Tier I criterion in 50
percent of the cases.  EPA does not believe that 50th percentile factors would
ensure that adequate protection is afforded in most cases.  If the Tier II
value is thought to be unnecessarily low, the State, Tribe, or discharger can
generate the data necessary to allow derivation of a different (usually
higher)  Tier II acute value or a Tier I FAV.

      Possible factors were calculated in a number of ways in the draft report
on which the proposed adjustment factors were based.  The proposed factors
were calculated as "overall 80th percentiles," i.e., the percentiles were
calculated as percent of the simulations rather than as percent of the
chemicals.  After further consideration, EPA has decided that the most
appropriate adjustment factors to use are those that were calculated as
"medians of the 95th percentiles" on page 62 of the draft report because these
percentiles were calculated as percent of the chemicals (Host, et al., 1991) .
These final adjustment factors (with daphnid required) are given in Table III-
5.  In addition, some minor anomalies occur in the overall 80th percentile
values,  but no such anomaly occurs in the medians of the 95th percentiles.
Further, the calculation of the assumed ACR is based on the percent of the
chemicals.

      It might seem that the adjustment factors and the assumed ACR should be
based on the same percentile.  Because the assumed ACR will often be used with
a Tier II acute value, it does not seem reasonable to increase the level of
protection by having both be at the 95th percentile; in this case it might be
appropriate to have the ACR at a lower percentile.  In some cases, however, an
assumed ACR will be used with a Tier I FAV; in these cases it might be
appropriate to have the assumed ACR be the 95th percentile.  It does not seem
prudent to use two different values for the assumed ACR, and so EPA has
retained the value of 18, which is based on the 80th percentile (see the next
section).  The final Guidance also retains the use of adjustment factors "with
daphnid data."

-------
116    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


                                    Table III-5

                 Adjustment Factors (with Daphnid Data Required)


                             Sample Size
                     (Number  of  Data Requirements Fulfilled)

Percentile     1234567


   95          21.9  •  13.0    8.0   7.0    6.1     5.2     4.3

-------
                               Section HI: Aquatic Life                           117
5.    Assumed ACRs

      a.    Proposal:  In the proposal, EPA requested comment on the use of
assumed ACRs in place of experimentally derived ACRs, and particularly on the
use of 18 as the assumed ACR.  When there are less than three experimentally
derived ACRs, the proposal required the use of enough assumed ACRs of 18
 (along.with experimentally derived ACRs) to bring the total number of measured
and assumed ACRs to three  (See appendix A.XIII. of proposed part 132).

      b.    Comments:  Several commenters supported the use of assumed ACRs
and the proposed assumed ACR of 18.  One of those commenters suggested that
the assumed ACR of 18 be used as a cap.  No comments were received which
questioned the use of assumed ACRs.  Commenters suggested using alternate
assumed ACRs developed by the European Centre for Ecotoxicology and
Toxicology.  These ACRs are median values generated for various chemical
classes  (e.g., pesticides, metal/organometals, etc.).

      EPA agrees with commenters that use of assumed ACRs are appropriate when
experimentally derived ACRs are not available.  EPA disagrees with that ACRs
should be developed using median values.  EPA does not believe it is
appropriate to base ACRs on median values whether they are generated across
chemical classes or for a single class of chemicals, because the median will
be too low for half of the chemicals.

      c.    Final Guidance:  The final Guidance retains the use of assumed
ACRs when insufficient experimentally derived ACRs exist.  A final assumed ACR
of 18, based on an 80th percentile shall be used when there are insufficient
measured ACRs.  EPA believes that it is inappropriate to alter experimentally
derived ACRs capping them at 18.  The final Guidance requires experimentally
derived ACRs when available, but assumed ACRs shall be used where data are
lacking.  Since the value of 18 is the 80th percentile, twenty percent of the
chemicals are expected to have larger ACRs.  The final Guidance does not
preclude a State or Tribe from choosing an assumed ACR which is more stringent
 (or greater than 18).

D.    Comparison with the CWA and EPA's National Guidance

      Section 118(c)(2)(A) of the CWA requires the water quality criteria for
the Great Lakes to be "no less restrictive" than the National criteria and
guidance.  As EPA explained in the proposal, it will not promulgate Tier I
criteria for several pollutants for which National criteria exist.  EPA
continues to believe that this decision does not make the Great Lakes criteria
"less restrictive" than National criteria.  Moreover, EPA notes that its
decision to change the adjustment factors used to compute Tier II values makes
the Tier II values more closely resemble Tier I criteria.  Tier II values will
now be as or more restrictive as Tier I criteria in 95 percent of all cases,
rather than 80 percent.

      In the proposal EPA also explained that the Great Lakes Tier I criteria
for four pollutants that appeared to be less conservative than the National
criteria for the same pollutants were not actually less restrictive.  EPA
continues to rely on that analysis.  Similarly, EPA continues to rely on the
proposal's conclusion that the Tier II methodology is not less stringent than
the more general narrative standards allowed under the National program.

      In response to comments on the proposal EPA has decided to express all
Tier I criteria for metals in terms of dissolved rather than total recoverable
concentrations.  This change results in criteria that are more appropriate
but,  arguably, less-stringent than the National criteria (which are expressed
as total recoverable concentrations).  As explained above in the discussion of
this issue, EPA has already begun to allow the use of dissolved values for
metals under the National program.  EPA intends to continue revising the
National program to be more consistent with the Great Lakes program on this
issue.

-------
118    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


E.    Conformance with the Great Lakes Water Quality Agreement

1.    Tier I Aquatic Life Criteria and Methodology

      The Great Lakes Water Quality Agreement  (GLWQA)sets out both general and
specific "objectives" for water quality in the Great Lakes.  The specific
objectives include a list of concentration levels for individual pollutants
set out in Annex 1 to the GLWQA.  The Annex indicates that some of these
levels were chosen to protect aquatic life.   (Other levels protect either
human health or wildlife.)  Some of the Annex 1 concentrations for the
protection of aquatic life, however, appear to be more conservative than the
corresponding final Tier I criteria for aquatic life.  EPA nevertheless
believes that the final criteria, as well as the methodology from which they
were derived, conform with the provisions and objectives of the GLWQA.

      In the first place, the language of the GLWQA itself shows that neither
the general not the specific objectives are legally binding, precise
standards.  Article III  (general objectives), Article IV  (specific objectives)
and Annex 1  (specific objectives) all describe their provisions as
"objectives," or goals, rather than requirements.  Further, both Article II
and Annex I use the permissive term "should" rather than the mandatory
"shall."  Article IV contains some provisions that appear more mandatory, but
they do not relate to the achievement of specific water quality requirements.
Moreover, it describes the Annex 1 numbers as  "desired" levels rather than
requirements.  Consequently, EPA thinks it is reasonable to believe that the
framers of the GLWQA did not intend the Annex 1 numbers to become enforceable,
mandatory requirements under U.S. or Canadian law.   (The U.S. Department of
State reached the same conclusion where it analyzed the GLWQA for the Office
of Management and Budget in  1978.)

      EPA also believes that the Great Lakes Critical Programs Acto of  1990
did not change the legal status of the GLWQA's objectives or provisions.
Section 118(c)(2)(A) of the CWA directs EPA to adopt guidance that  "conforms"
with the objectives of the GLWQA.  "Conform" means to "make similar" or "bring
into harmony," not to  "duplicate" or "follow precisely."  The very  same
sentence in  section 118 of the CWA shows that Congress knew how  to  require a
closer relationship-when it wanted to impose one:  it requires EPA's guidance
to be  "no less restrictive than the provisions of  [the CWA] and National water
quality criteria and guidance."   Clearly, Congress did not impose the same
standard with respect  to the general and specific objectives of  the GLWQA.

      EPA also finds it  significant that section 118(c)(2)(A) of the CWA
directs EPA to develop numeric  limits for the Great Lakes waters rather than
simply to incorporate  the  GLWQA numeric values.  EPA believes that  Congress
would have  been very explicit if  it had intended to deprive EPA  of  the
authority to exercise  its  own judgement on  the technical and scientific issues
involved.   Moreover, the  legislative history shows that Congress knew and
approved of the ongoing  work of  the Great Lakes Initiative Committees.  S.
Rep. 101-339.  101st Cong., 2d Sess. at 18  (June 27., 1990); 136  Cong. Rec.
S15616  (Oct. 17,  1990).   Since  the legislative history  so prominently
acknowledges the  committees' work, it is reasonable  to  assume that  Congress
expected EPA to develop  its  own  criteria.   Consequently, EPA does not believe
that "conformance" with  the  GLWQA requires  the numeric  criteria  proposed to be
 identical to or no less  restrictive  than the GLWQA objectives, including
 individual  Annex  1 values.   Rather,  EPA's  guidance as a whole needs to  further
 the objectives of the  GLWQA.  In EPA's judgement, the final Tier I  aquatic
 life criteria  are consistent with the general objective that Great  Lakes
waters  should  be  "free from  materials...that...will  produce conditions  that
 are toxic or harmful  to  ...aquatic  life."   Great Lakes  Water Quality Agreement
 of 1978, Art.  Ill, para,  (d).

       In  the preamble  to the proposed rule,  EPA stated  that  it would seek  to
 revise some of the  Specific  Objectives of  the Agreement.   EPA, however, has
 reconsidered this approach.   In light of  all the  strong evidence that  the

-------
                               Section ffl: Aquatic Life                           119
Specific Objectives were meant to be goals rather than requirements, EPA
believes revisions are unnecessary.  It does not intend to pursue them at this
time.

2.     Tier II Values and Methodology

      Tier II is a conservative methodology designed to establish
environmentally protective limits on the discharge of pollutants into the
Great Lakes System.  The methodology will regulate the discharge of certain
pollutants which, in certain Great Lakes States, may have been regulated by
narrative criteria rather than specific numeric criteria.  The Tier II
methodology is consistent with the general objective of the GLWQA cited above.
Moreover, it serves as a translator mechanism for that "narrative" objective.
The Tier II methodology will enhance protection of aquatic life in the Great
Lakes basin and will serve the GLWQA's purpose of promoting consistency in the
regulation of toxics in the Great Lakes basin.  Therefore, it also "conforms"
to the GLWQA.

-------
                          Section IV: Bioaccumulation Factors                       121
          IV.   BIOACCUMULATION FACTORS
A.    Summary of Final Rule

      The final Guidance is similar in substance to the Guidance proposed on
April 16, 1993, except for the following changes:

      - - Taking into account the freely dissolved  concentration of a chemical
in the derivation of bioaccumulation factors  (BAFs) for organic chemicals.

      -- Use of the equation,  baseline BCF  =  KOW in place  of the equation
originally proposed, when predicting a bioconcentration factor (BCF) from a
chemical's octanol-water partition coefficient  (KQW) for organic chemicals.

      -- Use of a model adapted from Gobas  (1993)  rather  than the model
discussed in the proposal to determine food-chain  multipliers (FCMs) for
organic chemicals,  which provides FCMs for  the  entire range of KOWS,  rather
than defaulting to one for log KQWS greater than six.

      -- Addition of an option for predicting a BAF based on the biota-
sediment accumulation factor (BSAF)  methodology as the second most preferred
method in the hierarchy for derivation of BAFs  for organic chemicals.

      -- The use of standard lipid values of  3.10  percent in edible tissue of
trophic level 4 fish and 1.82  percent in edible tissue for trophic level 3
fish for use in determining human health BAFs for  organic chemicals in place
of the lipid value originally proposed.   The  use of standard lipid values of
10.31 percent in whole body of trophic level  4  fish and 6.46 percent in whole
body for trophic level 3 fish for use in determining wildlife BAFs for organic
chemicals in place of the value originally  proposed.

      -- BAFs for individual PCB congeners, weighted by their relative
concentrations in salmonids, which is the predominant route of exposure,
rather than the mean of the nine most common  congeners or calculation from
concentrations of total PCBs in fish and ambient water.

B.    Explanation of Final Provisions

1.    BAFs

      Aquatic organisms are exposed to chemicals through  the water that they
live in, the food that they eat,  and contact  with  sediment.  Chemicals enter
these organisms via gills, epidermis,  or the  gastrointestinal tract; this
uptake of chemicals process is called bioaccumulation.  For certain chemicals,
uptake through the food chain is the most important route of exposure.  As
lower trophic level organisms are consumed  by higher trophic level organisms,
the tissue concentrations of some chemicals increase through one or more
trophic levels so that residues in organisms  at trophic level 3 and 4 might be
many orders of magnitude greater than the concentration of the chemical in the
ambient water.  While the concentration in  the  ambient water may be too low to
affect the lowest level organisms,  this biomagnification process can result in
exposure concentrations for the consumers of  top trophic  level aquatic
organisms that are above the fish tissue concentrations that correspond to
current Clean Water Act (CWA)  section 304(a)  water quality criteria by several
orders of magnitude (58 FR 20816).   Table 1-1 in the proposed Guidance  (58 FR
20816)  compares the measured concentrations of  PCBs and three pesticides found
in fish tissue (De Vault 1993a)  with the fish tissue concentrations
corresponding to current CWA section 304 (a) water  quality criteria at a 10"5
risk level.  Consumers of these fish would  be consuming fish that contain up

-------
122    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

to 200 times the levels of PCBs calculated in the 304(a) criteria  to
correspond to a 10~5 risk.

      In the final Guidance, as in the proposal, EPA relies on BAFs to  reflect
the propensity of a-chemical to accumulate in the tissues of aquatic
organisms, accounting for exposure from all sources of a chemical.  In  order
to properly account for exposure to a chemical, both the wildlife  criteria and
the human health criteria and values have been developed to incorporate
appropriate BAFs.  In addition, the human health BAFs are used to  identify
Bioaccumulative Chemicals of Concern  (BCCs) which warrant increased attention,
and more stringent controls, within the basin.  See discussion of  BCCs  in
section II.C.8 of this document.

      As discussed in the proposal, bioaccumulation refers to the  net
accumulation of a substance by an aquatic organism from its ambient water,
sediment and food.  A BAF represents the ratio  (in L/kg) of a substance's
concentration in the tissue of aquatic organisms to its concentration in the
ambient water in situations where both the organism and its food are exposed
and the ratio does not change substantially over time.  Measured BAFs are
based on field data.

      A BCF  (in L/kg) is the net accumulation of a substance by an aquatic
organism from the ambient water only through gill membranes or other external
body surfaces.  BCFs are determined either by measuring bioconcentration in
laboratory tests  (comparing fish tissue residues to chemical concentrations in
test waters), or by.predicting the BCF from the octanol-water partition
coefficient  (Kow or p)  of a chemical.   The equation,  log BCF = 0.79 log KOW -
0.40, was used in the proposal to relate the BCF and the octanol-water
partition coefficient for a chemical  (see 58 FR 20859) .

      EPA's 1991 National guidance documents, the "Technical Support Document
for Water Quality-based Toxics Control" and draft "Assessment and  Control of
Bioconcentratable Contaminants in Surface Waters," recommend a methodology for
estimating the BAF where a field-measured BAF is not available.  This
methodology predicts the BAF by multiplying the BCF by a factor which accounts
for the biomagnification of a chemical through trophic levels in a food chain.
The numerical factor which represents the magnitude of this biomagnification
through the food chain is called the food-chain multiplier  (FCM) in these 1991
National guidance documents.  In the 1991 documents and the proposed Guidance,
EPA calculated the FCMs using a model of the step-wise increase  in the
concentration of an organic chemical from phytoplankton  (trophic level  1)
through the top predatory fish  (trophic level 4) of a food chain  (Thomann,
1989)  (see 58 FR 20859 for a discussion of the Thomann model).

2.    Measured and Predicted BAFs

a.    Hierarchy of Methods.

      i.    Proposal:  The proposed Guidance listed three methods  for deriving
BAFs for organic chemicals, described below  in order of decreasing preference:

       (1)   A BAF measured in the field, preferably in fish collected in the
Great Lakes which are at the top of the food chain;

       (2)   A BAF predicted by multiplying a BCF measured in the laboratory,
preferably on a  fish species indigenous to the Great Lakes, by the food-chain
multiplier;  and,

       (3)   A BAF predicted by multiplying a BCF calculated from the log KOW
 (using the equation, log BCF =  0.79 log KOW  - 0.40 (Veith and Kosian, 1983))
by the food-chain multiplier.

      Subsequent  to the proposed  Guidance, EPA requested comment in  a Notice
of Data Availability  (59 FR 44678) on incorporation of a BAF derived from  the

-------
                           Section IV: Bioaccumulation Factors                       123
BSAF methodology as the second method in the hierarchy for deriving BAFs for
organics.

      ii.  • Comments: Many commenters stated that field-measured BAFs and
predicted BAFs for a chemical do not correlate well and in most cases the
predicted BAF for a.chemical overestimates the field-measured BAF.  Because of
this the commenters'advocated using only field-measured BAFs when deriving
human health and wildlife criteria.

      Several commenters argued that BCFs should be used instead of BAFs
because of the high degree of variability and site-specificity in field-
measured BAFs; the current state of the science does not support the
transition from BCF to BAF; and biomagnification might be due to data
anomalies and oversimplification of the food chain rather than trophic
transfer of a chemical.

      Many commenters wanted EPA to discuss the uncertainty associated with
the BAFs because they were concerned that all the uncertainties inherent in
each portion of the methodology could produce an aggregate uncertainty larger
than the BAF.

      Several commenters questioned the use of a combination of the Veith and
Kosian  (1983) regression equation with the Thomann model (1989), and suggested
using either the whole Thomann approach or testing the validity of the
combination approach.

      Many commenters suggested using the BSAF or Bioavailability Index  (BI)
for derivation of criteria in place of the BAF.  Along the same lines, many
commenters advocated using the concentration in sediments instead of ambient
water when deriving criteria because sediments are the primary repository of
chemicals with log KOWS greater than four and using minute,  trace water
concentrations of chemicals with log KOWS greater than four in the derivation
of BAFs will result in inappropriately high values.

      Finally many commenters wanted the flexibility to revise the BAF when
new data become available, while others stressed the need for communication
between EPA and the regulatory agencies to share data on BAFs.

      EPA does not agree that the field-measured BAFs and predicted BAFs for a
chemical do not correlate well.  The adaptation of the Gobas model for
estimating FCMs (see discussion in section IV.B.4 below on the Gobas model)
reduces much of the uncertainty and variability associated with comparing
field-measured BAFs and predicted BAFs.  A comparison of the BAFs predicted by
the Gobas model (1993) against the field-measured BAFs from Oliver and Niimi
(1988)  for the 52 chemicals which have field-measured BAFs for at least three
fish shows that differences between the mean BAFs are less than a two-fold for
46 of the 52 chemicals, and less than a three-fold for 51 of the 52 chemicals
(Zipf,  1995) .  EPA concludes that when field-measured BAFs are not available,
the model used in the final Guidance acceptably predicts BAFs for the Great
Lakes System.

      EPA partially agrees with commenters who advocate using field-measured
BAFs when deriving criteria.  In the proposal, Tier I criteria and Tier II
values for human health and wildlife were differentiated based on the quantity
and quality of toxicological data only.  After reconsideration, EPA has
decided to differentiate the Tier I criteria and Tier II values for human
health and Tier I criteria for wildlife based on the quantity and quality of
both the toxicological and bioaccumulation data.  The minimum toxicological
data for human health is discussed in section V and for wildlife in section
VI.   The new minimum BAF data required to derive Tier I human health criteria
for organic chemicals include either:  (a) a field-measured BAF;  (b) a BAF
derived from the BSAF methodology, or  (c) a chemical with a BAF less than 125
regardless of how the BAF was derived.  The new minimum BAF data required to
derive Tier I wildlife criteria for organic chemicals include either:  (a) a

-------
124    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


field-measured BAF, or (b) a BAF derived from the BSAF methodology.  For all
inorganic chemicals, including organometallies such as mercury, the minimum
BAF data required to derive a Tier I human health and wildlife criteria
include either:  (a) a field-measured BAF or  (b) a laboratory-measured BCF.
For the majority of inorganic chemicals, the BAF is equal to the BCF (i.e.,
FCM = 1) because there is no apparent biomagnification or metabolism.  The
basis for these new requirements is explained below.

      Requiring the use of field-measured BAFs or field-measured BSAFs when
deriving Tier I criteria for organic chemicals with predicted BAFs greater
than 125 eliminates for these Tier I criteria concerns about the effect of
metabolism on the BAF.  This is the case because field studies measure
chemical concentrations in the tissues of the fish that are exposed to the
chemical from food, ambient water and sediment.  The measured concentrations
in the fish inherently account for the effect of metabolism from all sources
of exposure.  On the other hand, the concentration of the chemical in fish
tissue from laboratory-measured BCF accounts for exposure from ambient water
only.  Consequently, BCFs do not account for the effect of metabolism of
chemicals accumulated from exposure through the diet of the aquatic organism.
Metabolism may either increase or decrease the concentration of a chemical and
its by-products in the tissue of an aquatic organism.  EPA's BAF methodology
uses a model  (Gobas, 1993) to predict the accumulation of a chemical from food
sources.  The model of Gobas (1993), however, does not account for the effects
of metabolism; in other words,  the entire concentration of the chemical input
to the model plus the concentration biomagnified through the food chain is
predicted to accumulate in the fish tissue.  Consequently, BAFs based on
multiplying a laboratory-measured BCF times a FCM (predicted from Gobas, 1993)
may either under- or overestimate the amount of a chemical a fish will
bioaccumulate.  In addition, predicted BAFs that are obtained by multiplying a
predicted BCF by a FCM make no allowance for metabolism.

      EPA has decided that Tier I criteria, which must be adopted into State
regulations and become the goal for permit limits throughout an entire State,
should accurately account for the effects of metabolism.  Accordingly, EPA is
requiring the use of field studies for determining a BAF to be used in the
derivation of a Tier I criterion for human health.

      Tier I human health criteria may also be derived for chemicals with BAFs
less than 125 regardless of which of the four methods specified in the final
Guidance is used to derive the BAF.  For chemicals with a BAF less than 125,
exposure from consumption of fish is less than or equal to the exposure from
consumption of drinking water.  This assumes a fish consumption rate of 15
grams per day and drinking water consumption of two liters per day.  Thus for
these chemicals the effects of metabolism by aquatic organisms are not as
significant a determinant of ultimate human exposure as for chemicals with
larger BAFs.  Therefore, for organic chemicals with a BAF less than 125, all
four methods specified in the final Guidance can be used to obtain a BAF used
to derive Tier I criteria.

      EPA has decided, notwithstanding the fact that predicted BAFs will at
best only partially account for the effects of metabolism, to allow derivation
of Tier II human health values using BAFs based on BCFs and the FCM.  EPA's
decision is based on several factors.  First, as described elsewhere in this
document, available information indicates a very good correlation between
predicted BAFs based on the methodology in the final Guidance and field-
measured BAFs.  Therefore, it appears that the error introduced by using
predicted BAFs rather than field-measured BAFs may be relatively small for
many chemicals.  Second, Tier II values by definition are based on a less than
ideal data base.  Acknowledging that  the data available is not perfect for
Tier II chemicals, EPA believes that  it is nevertheless important to protect
human health  from potential adverse effects  resulting from their discharge to
surface waters.  The Tier II methodology, including use of predicted BAFs,
specifies the best available protocols for assessing protective ambient levels
for these chemicals for which Tier  I  data are not available  This will provide

-------
                           Section IV: Bioaccumulation Factors                       125
consistency among Great Lakes States and Tribes in the interpretation of
narrative human health criteria when establishing water quality-based effluent
limitations  (WQBELs).  To the extent that a discharger of a chemical for which
a Tier II value is established on the basis of a predicted BAF questions
whether a field-measured BAF would not provide a more accurate measure of
bioaccumulation, that discharger is free to conduct an adequate field study
and request that it-be utilized by the appropriate regulatory authority in
establishing a-Tier II value or Tier I criterion.  Finally, for those
chemicals where available data indicates that metabolism is likely to
considerably increase or decrease a predicted BAF, appendix B section VIII of
the final Guidance allows States and Tribes to modify the predicted BAFs
accordingly.  EPA believes that derivation of an "effective FCM" for certain
chemicals, as described in the proposed Guidance, may in many cases reasonably
account for the metabolism.  However, States and Tribes are not limited to the
use of an "effective FCM" in correcting a predicted BAF for metabolism.

      EPA disagrees with commenters that BCFs should be used instead of BAFs.
Bioaccumulation is what occurs in nature, and is what determines the total
concentration of chemicals in aquatic organisms that are consumed by humans
and wildlife.  For some chemicals the biomagnification of a chemical through
the food chain can be substantial.  Using BCFs, which only account for
exposure from the ambient water, could substantially underestimate the
potential exposure to humans and wildlife for some of these chemicals and
result in criteria or values which are underprotective.  The use of BAFs,
which account for uptake from all sources, will ensure that the potential
exposure from these chemicals is adequately accounted for in the derivation of
human health and wildlife criteria.  Using BAFs is the most comprehensive and
scientifically valid approach.  As mentioned in the proposal (58 FR 20858),
BAFs have been used in deriving human health criteria development since 1980.

      EPA recognizes that field-measured BAFs will have some variability from
site to site.  In recognition of this, EPA allows the derivation of site-
specific BAFs as discussed in procedure 1 of appendix F to part 132.  Although
there might be some variability in field-measured BAFs, it does not invalidate
their usefulness in estimating the potential exposure to humans and wildlife,
nor does it imply that BAFs are less accurate than BCFs in predicting that
exposure.

      In addition, EPA does not agree with commenters that biomagnification
might be due to data anomalies and oversimplification of the food chain.
There is ample evidence of biomagnification occurring because of trophic
transfer of chemicals.  The importance of uptake of chemicals through the diet
and the potential for a stepwise increase in bioaccumulation from one trophic
level to the next in natural systems has been recognized for many years
(Hamelink, et.  al.,  1971) .   Many researchers have noted that the BAFs of some
chemicals in nature exceed the BCFs measured in the laboratory or estimated by
log KOW models  (e.g.,  Oliver and Niimi 1983,  Oliver and Niimi  1988,  Niimi
1985, Swackhammer and Hites 1988).

      Further,  EPA believes that the state of the science supports the use of
BAFs.  EPA's Science Advisory Board  (SAB) in its December 16,  1992, report on
the Evaluation of the Guidance for the Great Lakes Water Quality Initiative
stated that the BAF procedure is more advanced and scientifically credible
than existing BCF procedures and that the use of the BCF, FCM,  and BAF
approach appears to be fundamentally sound (EPA-SAB-EPEC/DWC-93-005).  In a
subsequent SAB report on August 12, 1993 on the ongoing revision of the
methodology for deriving National Ambient Water Quality Criteria for the
protection of human health  (EPA-SAB-DWC-93-016), the Drinking Water Committee
reported on a similar BAF methodology.  Although cautioning that its
"criticisms should not be taken as a recommendation to relax standards or to
ignore the potential for bioaccumulation where it is known to play an
important role," the Drinking Water Committee'also stated that "for the time
being, the Agency should focus attention on BCFs rather than BAFs, because of
the higher likelihood of collecting an adequate BCF data base."  In evaluating

-------
126    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

the two SAB committee reports, it is  important to keep in mind that the first
committee was reviewing the proposed  BAF methodology for the Great Lakes
Guidance, while the second committee  was reviewing a similar proposed
methodology that would be applicable  nation-wide.  Thus, the second
committee's recommendation that a sufficient BAF data base may not be
available at the present time to derive BAFs for chemicals for nation-wide
criteria guidance does not imply that sufficient information is not available
to develop BAFs for regional water quality standards in the Great Lakes'.
Indeed, to rely on BCFs in the Great  Lakes System would be directly contrary
to the Drinking Water Committee's exhortation that their criticism not be
taken as a recommendation "to ignore  the potential for bioaccumulation where
it is known to play an important role."  EPA has revised the BAF methodology
where possible at this time to take into account the concerns raised by both
SAB Committees, and. believes after careful review of SAB and public comments
that use of BAFs in the final Guidance represents the most scientifically
defensible approach for accounting for chemical uptake by aquatic biota in the
Great Lakes System.

      EPA also recognizes there were  some uncertainties in application of the
proposed BAF methodology and has addressed these in the final Guidance.  For
example, to reduce the uncertainty in predicting the biomagnification of
chemicals, EPA is using a model in the final Guidance that uses Great Lakes
specific parameters and includes a benthic food chain component to estimate
FCMs.  In addition, the final Guidance takes into account the freely dissolved
concentration of a chemical in the derivation of BAFs for organic chemicals.
Taking the freely dissolved concentration into account will eliminate much of
the variability associated with specific waterbodies because most of the site-
specific differences in bioaccumulation arises from the partitioning of the
chemical to the particulate organic carbon and dissolved organic carbon of the
water column.  However, professional  judgement is still required throughout
the derivation of BAFs and a degree of uncertainty is still associated with
the determination of any BAF, BSAF, BCF or KOW.   Despite this uncertainty,
EPA maintains that BAFs are the most  useful measure of the exposure of an
aquatic organism to all chemicals.

      EPA agrees with those commenters who suggested the use of the BSAF
approach for deriving BAFs and has modified the proposed hierarchy of methods
for deriving BAFs to include a BAF derived from the BSAF methodology as the
second preferred method after field-measured BAFs.  The BSAF provides a method
by which the concentrations of chemicals in the sediment are related to the
concentrations in fish tissue.  The concentrations of chemicals with log KQWS
greater than 6.5 are greater in the sediment than in the water column and more
readily measured; therefore use of the BSAF reduces the uncertainty associated
with relating concentration in fish tissue to the concentration  in the water
column for these chemicals.  This is  particularly true for chemicals with
higher KOWS since these generally show a greater affinity for sediments.  The
BI is the same method as the BSAF and the terms can be interchanged.  For
further details on deriving BAFs using the BSAF methodology, and the data
supporting the approach, see the final Great Lakes Water Quality Initiative
Technical Support Document for the Procedure to Determine Bioaccumulation
Factors  (EPA 820-B-95-005)  (BAF TSD)  for BAFs which is available in the public
docket for this rulemaking.

      In response to commenters who wanted EPA to develop sediment criteria
instead of water criteria, EPA asserts that ambient water quality criteria
provide a measure of acceptable chemical levels in the medium in which  non-
benthic aquatic organisms primarily exist, and are therefore an  important
measure of acceptable environmental conditions for these organisms.  EPA
agrees that sediment criteria may also provide useful indices of acceptable
chemical levels, especially when the  greatest risks are known to be associated
with toxic effects  to benthic organisms.

      EPA has  changed the final Guidance in response to commenters' concerns
with the use of the Veith and Kosian  regression equation in  combination with

-------
                           Section IV: Bioaccumulation Factors                       127
the Thomann model  (1989).  In the final Guidance, the equation baseline BCF =
KQW that is used to predict BCFs is equivalent to the equation BCF" = KQW for
log KQWS less than three that is used in the Gobas 1993 model to predict FCMs.
Thus there is no longer a need to validate the use of the regression equation
by Veith and Kosian  (1983) in combination with the Thomann model  (1989).

      Finally, EPA agrees with commenters that it is important to revise BAFs
when new data become available and anticipates that the Clearinghouse
discussed in section II.C.I will provide the mechanism through which new data
are evaluated and disseminated.  In addition, the final hierarchy of data
preference allows for the incorporation of new data.  For example, if a field-
measured BAF is calculated for a chemical for which only a predicted BAF was
previously available, preference would be given to the field-measured BAF.

      iii.  Final Guidance:  EPA revised the proposed hierarchy of methods for
deriving BAFs based on public comments.  The final Guidance lists four methods
for deriving BAFs for organic chemicals, listed below in order of decreasing
preference: a BAF measured in the field, in fish collected from the Great
Lakes which are at or near the top of the food chain; a BAF derived using the
BSAF methodology; a BAF predicted by multiplying a BCF measured in the
laboratory, preferably on a fish species indigenous to the Great Lakes, by the
FCM; and a BAF predicted by multiplying a BCF calculated from the KOW by the
FCM.

b.    Field-Measured BAFs

      i.    Proposal:  As discussed above, the proposal stated a preference
for using field-measured BAFs over predicted BAFs.  This preference stems from
the fact that field-measured BAFs automatically account for any
biomagnification and metabolism that might occur.  The proposal also stated
that field-measured BAFs should be based on fish species, preferably living in
the Great Lakes at or near the top of the aquatic food chain.  In its December
16, 1992 report, the EPA's SAB stated that field-measured BAFs must be
interpreted very carefully, and it should be recognized that they might
contain substantial errors and variability due to several factors (see 58 FR
20860).

      ii.   Comments:  Many commenters stated that the field-measured BAFs
might contain substantial errors and expressed a preference for the more
established BCF.  Other commenters were concerned that the field-measured BAFs
were based on a single data source, that limiting the data base to the Great
Lakes for calculation of field-measured BAFs ignores a wealth of information,
and that it was important to show that the BAFs measured in one lake
accurately predict fish tissue levels in other Great Lakes.  Several
commenters suggested developing guidelines for measuring and/or evaluating
field-measured BAFs. which would include provisions for the simultaneous
collection of water and fish tissue and accounting for fish mobility, as well
as discussing the importance of the sampling location and temporal change.

      EPA acknowledges that there can be errors in determining field-measured
BAFs,  as with any field-measurements, and has attempted to minimize these
potential errors when deriving BAFs for the final Guidance by carefully
screening the data used to calculate the BAFs.  EPA continues to contend that
a field-measured BAF is a more accurate gauge of what is occurring in nature
than a laboratory-measured or -predicted BCF because the BAF measures the
actual impacts of biomagnification, bioavailability, concentration in the
sediment,  growth dilution, and metabolism rather than predicting them through
use of a model.

      EPA also agrees with the SAB's comments and with commenters concerned
about the difficulty of collecting and interpreting field-measured BAFs.  EPA,
however, thinks that States and Tribes can adequately use and interpret field
studies.  To assist them in this task, EPA plans to provide guidance
concerning the determination and interpretation of field-measured BAFs before

-------
128    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

the States and Tribes are required to adopt water quality standards  consistent
with the final Guidance.  This will provide interested parties with  a  set  of
procedures that will assist them in collecting and interpreting  the  field-
measured BAFs.

      The majority of data used to calculate the field-measured  BAFs in the
final Guidance came from the data of Oliver and Niimi  (1988).  This  data set
is generally recognized as being the most complete set of data available in
the Great Lakes for estimating field-measured BAFs.  EPA acknowledges  that the
data from Oliver and Niimi come from Lake Ontario, but believes  that the data
can be used to predict BAFs in other Great Lakes because the values  take into
account the percent lipid and are based on the freely dissolved  concentration
of the chemical in the ambient water.  Taking the lipid content  into account
allows the data to be applied to other fish species.  Derivation of  the BAFs
on a freely dissolved basis from field data eliminates the site-specific
nature of the BAFs caused by the amounts of dissolved and particulate  organic
carbon present at the field site and therefore, allows the use of  the  derived
BAFs in the other Great Lakes.

      Using data from the Great Lakes is preferable to using information from
other bodies of water because it better represents the physical, chemical, and
hydrological conditions present within the Great Lakes.

      iii.  Final Guidance:  The final Guidance requires that field-measured
BAFs be the preferred method for deriving BAFs because of their  ability to
account for biomagnification, growth, metabolism, concentration  in the
sediment, and bioavailability.

c.    Field-Measured BSAFs

      i.    Proposal:  As discussed in the August 30, 1994 Notice  of Data
Availability (59 FR 44678), BSAFs can be used for measuring bioaccumulation
directly from concentrations of chemicals in surface sediments or  to estimate
BAFs.  Because BSAFs are based on field data and incorporate the effects of
metabolism, biomagnification, growth, concentration in the sediment, and
bioavailability, BAFs derived from the BSAF methodology will incorporate the
net effect of these factors.  The BSAF approach is particularly  beneficial for
developing water quality criteria for chemicals such as polychlorinated
dibenzo-p-dioxins  (PCDDs) and polychlorinated dibenzofurans  (PCDFs)  and
certain biphenyl congeners which are detectable in fish tissues  and  sediments
but are difficult to measure in the ambient water.

      BSAFs are measured by relating  lipid-normalized  concentrations of
chemicals in an organism to organic carbon-normalized  concentrations of the
chemicals in surface sediment samples associated with  the average  exposure
environment of the organism.  The BSAF is defined  as:

                                           C/
                                 BSAF = —-L


      where:

            C,    =  the lipid-normalized concentration of the chemical in
                     tissues of the biota  (/zg/g lipid) .

            Csoc   =  the organic carbon-normalized concentration of  the
                     chemical in the surface sediment  (jig/g  sediment organic
                     carbon).

For further explanation,  see  the  BAF  TSD.

       ii.   Comments:   Commenters generally supported the  use  of the BSAF
methodology in deriving BAFs.   Some  commenters  stated that while chemicals in
the sediment may contribute  to  bioaccumulation,  the  proposed BSAF model is

-------
                           Section IV: Bioaccumulation Factors                       129
only valid if limited to a chemical- and site-specific context.  Some
commenters suggested that EPA more fully address the issue of relative
concentrations of a chemical in the sediment and water column.

      Other commenters questioned the validity of data for reference chemicals
used when deriving BAFs from the BSAF methodology since the data are from a
single unpublished study.  Others raised questions about the availability of
the study.

      EPA acknowledges that BAFs derived from field-measured BSAFs require
chemical-, species- and site-specific data.  EPA is not intending to place an
undue burden on the States, Tribes or dischargers to generate data.  If
pertinent data are available, EPA has set forth a methodology in which they
can be used.  Basing the BSAF on lipid-normalized concentrations and organic
carbon-normalized concentrations eliminates many of the species- and site-
specific characteristics, and therefore the BSAF is applicable to a wider area
than that at which it was measured.

      In response to commenters'  concerns about the relative concentrations of
a chemical in the sediment and the ambient water, EPA believes that fish are
exposed to organic chemicals through contact with water, food and, and to some
extent,  sediment.  At steady-state, the concentrations of these chemicals in
water or surface sediment, although numerically quite different, are equally
useful for prediction of bioaccumulation.in fish.  When concentrations of some
chemicals are temporally variable and/or nondetectable in water, BSAFs can
provide the most reliable field-measurement of bioaccumulation.  BAFs are
needed,  however, to calculate water quality criteria.  Fortunately, the BSAF
methodology inherently includes a measure of the disequilibrium that usually
occurs between the sediment-water distribution of the chemicals by using
reference chemicals for which a field-measured BAF is available.  The relative
concentrations of the chemical in the sediment and water are therefore
accounted for in the BSAF.  The BSAF method translates the bioaccumulation and
disequilibrium information presented by the BSAF into a BAF through comparison
to reference chemicals with similar sediment-water disequilibrium at the same
site.  In this method the reference chemicals provide key relationships
between measured BSAFs and BAFs for the ecosystem.

      EPA understands the commenters' concerns with the availability of the
data used for deriving BAFs from the BSAF methodology.  All the BSAF, BAF and
KQW data used to derive the BAFs  based on the  BSAF methodology were presented
in the August 30, 1994 Notice of Data Availability (59 FR 44678).  EPA
acknowledges that some of the fish and sediment analytical data have not been
published. However, the use of the Oliver and Niimi data (1988), which have
been published,  to demonstrate correlations and numerical similarity between
BSAFs and BAFs calculated from the two independent data sets should provide
additional assurance of the applicability of the unpublished data for
demonstration of the BSAF method for deriving BAFs from BSAFs.

      iii.  Final Guidance:  In the final Guidance, a BAF derived from the
BSAF methodology is added as the second most preferred method in the hierarchy
for derivation of BAFs for organic chemicals.

d.    Measured and Predicted BCFs

      i.    Proposal:   The proposed Guidance preamble (58 FR 20859) discussed
three analytical techniques that can be used to measure organic chemicals in
tissue and water for the purposes of establishing a laboratory-measured BCFs:
gas chromatography, high pressure liquid chromatography or radio-labeled
organic chemicals.

      The proposed Guidance predicted BCFs from the octanol-water partition
coefficient (KQW or P)  of a chemical using the equation:

            log BCF =0.79 log KQW - 0.40 (Veith and Kosian,  1983)

-------
130    Water Quality Guidance for the Great Lakes System ~ Supplementary Information Document


Subsequent to the proposed Guidance, EPA requested comment in the August 30,
1994, Notice of Data Availability  (59 FR 44678) on the use of an alternative
equation from the proposal to predict BCFs.  The equation was:

                                 BCFf  = Kow

where the BCF,W is reported on a lipid-normalized basis using the freely •
dissolved concentration of the chemical in the water.

      ii.   Comments:   Some commenters indicated that use of radio-labeled
organic compounds would overestimate the laboratory-measured BCF.  Other
commenters stated that difficulties with radio-labeled materials are easily
avoided in actual practice.

      Several commenters stated that predicted BCFs should not be used  for
chemicals that are suspected of metabolism.  Many commenters supported  the
change from the Veith and Kosian  (1983) equation to the equation BCF" = KOW/
but recommended that EPA specify the limitations of its use;  commenters
wanted EPA to be explicit about the assumed percent lipid in derivation of the
BCF.  Several commenters requested development of guidelines for measuring
KOW-

      EPA partially agrees with the commenters who recommended against  the use
of radio-labeled organic chemicals for measuring BCFs because the organism may
also accumulate a metabolite of the parent compound, thereby overstating the
actual BAF.  Attempts to measure the amount of radio-labeled compound obtained
through tissue analysis, instead of measuring the radioactivity of the  fish,
have not been definitive.  There is also the possibility of contamination of
the labeled compound.   Because of these concerns, EPA has decided that  BCFs
for organic chemicals may be based on measurement of radioactivity only when
the BCF is intended to include metabolites or when there is confidence  that
there is no interference due to metabolites.

      EPA partially agrees with commenters who suggest that predicted BCFs
should not be used for chemicals that are suspected of metabolism.  EPA stated
in the Technical Support Document for the Procedure to Determine
Bioaccumulation Factors  (EPA-822-R-94-002) that the relationship BCF" = KQW
is applicable to organic chemicals which are either slowly or not metabolized
by aquatic organisms.   Since predicted BCFs do not account for metabolism,
they will not be used in the derivation of Tier I human health and wildlife
criteria unless the predicted BAF is  less than 125.  Predicted BCFs, however,
can be used in the derivation of Tier II human health values if no laboratory-
measured BCF data are available.  For a more detailed discussion on Tier II
values, see section IV.B.2.a.ii of this document.

      In response to commenters' questions regarding the assumed percent
lipid, the 1983 average percent lipid associated with the Veith and Kosian
equation was 7.6.  The BCF in the equation BCF" = KOW is reported on a  lipid
normalized basis, and therefore by definition assumes 100 percent lipid.  This
BCF is then adjusted to the percent lipid appropriate for a given trophic
level.  See section III.B.3 of this document for further explanation.

      EPA is in the process of developing guidelines for measurement of KOWS.
In  the interim period,  section III.F  of appendix B to part 132 lists the
analytical technique priorities for deriving  KOWS.  KOWS are an integral
factor in developing BAFs used in derivation of human health criteria and
values and wildlife criteria; therefore, EPA believes that providing guidance
on  the acceptability of KOWS will result in more consistent criteria.

-------
                           Section IV: Bioaccumulation Factors                       131
      iii.  Final Guidance:  In the final Guidance, predicted BCFs are derived
from a chemical's KQW using the following equation:

                             Baseline BCF = Kow

See the final BAF TSD for details.

      The BCF .based on this equation provides a more consistent and
scientifically defensible basis for predicting BAFs than the equation  (Veith
and Kosian, 1983) used in the proposal.  The theoretical basis presented by
Mackay (1982) and the experimental data referenced in the August 30, 1994,
Notice of Data Availability  (58 FR 44678), suggest that octanol is a
reasonable surrogate for lipids.

      For laboratory-measured BCFs, EPA continues to strongly recommend the
use of the procedural and quality assurance requirements specified in the
American Society for Testing Materials (ASTM) (1990) "Standard Practice for
Conducting Bioconcentration Tests with Fishes and Saltwater Bivalve Molluscs,"
and in the EPA guidance contained in Stephan et al. (1985), "Guidelines for
Deriving Numerical National Water Quality Criteria for the Protection of
Aquatic Organisms and Their Uses."

e.    Inorganic Chemicals

      i.     Proposal:  The proposed Guidance limited the methods for obtaining
a BAF for inorganic chemicals to a field-measured BAF or a laboratory-measured
BCF.  This is because no method is available for reliably predicting BCFs or
BAFs for inorganic chemicals.  BCFs and BAFs for some inorganic chemicals vary
between species, and from one tissue to another within a species.  The
proposal included BAFs for 17 inorganic chemicals, including mercury.

      ii.   Comments:  Several commenters indicated that there was a wide
range of variability in laboratory-measured BCFs for inorganic chemicals.
Other comments indicated that in order to check the reliability of measured
data, modelling should be allowed for inorganic chemicals.  Other commenters
advocated site-specific modifications for metals based on each Lake's
characteristics.

      EPA acknowledges that there is some variability in laboratory-measured
BCFs for inorganic chemicals, and that there are real differences between
species.   For example, saltwater molluscs have substantially higher BCFs for
many metals than do freshwater and other saltwater fishes and invertebrates.
Variability is not a major 'concern for the majority of inorganic chemicals,
however,  because their BCFs are low for species consumed from the Great Lakes.
For example, the BAF for ten of the 17 inorganic chemicals in the list of
chemicals in Table 6 of part 132 is below ten.  This indicates that even
though there may be some variability in laboratory-measured BCFs, the impact
on the final criteria will be small.

      EPA agrees with commenters that modelling should be allowed if the model
has been shown to be appropriate.   EPA advocates but does not require the use
of modelling to verify the reliability of the measured data for inorganic
chemicals. In addition,  EPA agrees that site-specific modifications should be
allowed for BAFs and provides for this in procedure 1 of appendix F to part
132.

      iii.  Final Guidance:  After review of the comments, EPA  decided that
the proposed Guidance for derivation of BAFs for inorganic chemicals will not
be amended.  The most accurate measurement of bioaccumulation for inorganic
chemicals are field-measured BAFs  and laboratory-measured BCFs.

-------
132    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

3.     Lipid Values

      Consistent with the existing National guidance, the proposed Guidance
relied on the assumption that an organism's ability to bioaccumulate organic
chemicals is related to it's lipid content.

      Therefore to determine a BAF for organic chemicals for use in deriving
wildlife Tier I criteria and human health Tier I criteria and Tier II values
it is necessary to know the percent lipid content of the organisms being
consumed.  Humans typically eat fish  fillets which usually have  lower lipid
content than the whole fish consumed  by wildlife.  EPA's current recommended
National percent lipid values for calculation of human health criteria  is 3
percent  (U.S. EPA, 1980}.  There is currently no National guidance for
wildlife and thus there is no recommended percent lipid value.

      In the proposal, EPA requested  comments on a variety of issues including
what solvent should be used in the measurement of percent lipids.  Some
commenters advocated the use of a standardized extraction method and a
consistent system to measure lipid content, while a few suggested use of
methylene chloride as the extraction  solvent.  No rationale for  selecting
between the solvents which have been  proposed was presented by commenters.
EPA believes that the data is inconclusive  as to which extraction method and
solvent to use and has therefore has  not made a recommendation on which method
and solvent to use in the measurement of percent lipids.  EPA will be
providing additional guidance on which extraction method(s) and  solvent(s) to
use in the guidance on the determination and interpretation of field-measured
BAFs.

a.    Lipid Value for Human Health BAFs

      i.    Proposal:  The proposed Guidance used a lipid value  of 5.0  percent
in edible tissue for use in determining human health BAFs for organic
chemicals.  Percent lipid data for edible tissue  (mostly skin-on fillets) were
gathered from the fish contaminant monitoring programs in Michigan, Wisconsin,
Ohio, Indiana, New York and Minnesota.  The use of skin-on data  to determine
the lipid values provided an extra margin of safety to the many  anglers who
remove the  skin  from the fillet.

      In selecting the lipid value for human health BAFs, lipid  data for the
following fish groups were considered: salmonids only; salmonids and non-
salmonid game fish; and all fish  (game and  nongame species).

      The mean lipid value for salmonid and non-salmonid game fish of 5.02
percent  lipid was proposed because this option best represented  the range of
species  typically cpnsumed by people  in the Great Lakes basin.   Also
considered  was the mean lipid values  weighted by human consumption patterns.
The resulting consumption-weighted mean for all sport-caught game fish  was
4.72 +  2.42 percent  lipid.  Because these results were not  statistically
different from the unweighted mean, use of  the unweighted mean of 5.02  percent
 (rounded to 5.0  percent) was proposed for the human health  BAFs.

      ii.    Comments:  A few commenters did not agree with  the use of skin-on
filets,  and suggested using  skin-off  filets for lipid measurement with  a
corresponding lipid  value  of four percent.   Other commenters stated that  the
five percent lipid value did not  provide  an adequate  margin of safety nor
protect  high-risk subpopulations  (e.g., Tribes and subsistence fisherpersons)
and  suggested using  an 11  percent lipid value.  Other commenters suggested
using three percent  for  the  tributaries of  the Great  Lakes, which is
 consistent  with  the  National guidelines dated November  28,  1980  (45 FR  79347) .
 Some commenters  stated that  the  data  did  not exist to determine  which fish
 species were consumed and  the  corresponding percentage  of  lipid  for high-risk
 groups.

-------
                           Section IV: Bioaccumulation Factors                       133
      In order to further examine whether the five percent lipid value was
appropriate, EPA conducted additional analysis of the data from a second fish
consumption survey conducted by West, et al.  (1993)  (see section V of this
document, Human Health, for a complete discussion of this study).  EPA
requested comments on the appropriateness of the data presented in the study
in a Federal Register notice on August 30, 1994  (59 FR 44678).  The results
from this analysis indicate that the consumption-weighted mean percent lipid
value for trophic level 4 fish is 3.10 and 1.82 for trophic level 3.  EPA
believes that the use of the West et al.  (1993) survey to estimate the percent
lipid used for deriving BAFs is an improvement on the methods utilized in the
proposal because the West survey allows a determination of the actual fish
species consumed and the rate of consumption.  When this information is
coupled with the information on percent lipid values for these fish, it is
possible to derive a more accurate reflection of the grams of lipid from fish
that are consumed by humans from each trophic level.  EPA acknowledges that
the West study only covered anglers in the State of Michigan, but concludes it
represents the best study to use for deriving consumption-weighted mean
percent lipid values for trophic levels 3 and 4.   States and Tribes can derive
alternative percent lipid values to be used in the derivation of site-specific
BAFs whenever they have the information needed to revise the derivation.

      EPA does not agree with those commenters advocating the use of skin-off
fillets for deriving percent lipid values.  Although many people remove the
skin and other fatty tissue when they prepare their fish for cooking, the
study by West et. al.  (1993) indicates that about 37 percent of anglers in
Michigan continue to prepare fish with the skin on even though they are aware
of the State fish advisories recommendation to trim fat and/or skin from the
fish.

      EPA also disagrees with those commenters advocating the use of a three
percent lipid value for tributaries.  Due to the mobility of the prey and its
host, it is difficult to characterize their territorial range.  The
consumption weighted mean percent lipid for the respective trophic levels
represents an overall average for the Great Lakes System.  The fish lipid data
used to determine the percent lipid values were gathered from fish contaminant
monitoring programs in the Great Lakes System  (including its tributaries)  and
represent the species consumed by people in the West et al. survey.  EPA also
does not agree that the lipid values should be increased to 11 percent
representative of lake trout, a species with the highest lipid value.  In the
majority of the cases people consume a variety of species and not simply lake
trout,  as evidenced by the West survey.  The lipid values selected for use in
deriving BAFs represent the wide variety of fish consumed by sport anglers in
the Great Lakes System.  In cases where it can be documented that a
subpopulation consumes fish with an average lipid content higher than those
prescribed in the final Guidance, then it may be appropriate for a permitting
authority to increase the lipid value in deriving a site-specific criterion
for waters fished by the subpopulation.  The permitting authority should
evaluate all aspects of exposure, including amount consumed, before altering
just one factor such as percent lipid, since the values for these variables
are interrelated.

      iii.  Final Guidance:   EPA has specified the use of a consumption-
weighted mean percent lipid value for trophic level 4 fish of 3.10 and 1.82
for trophic level 3 in edible tissue for use in determining human health BAFs
for organic chemicals in the final Guidance.

b.    Lipid Value for Wildlife BAFs

      i.    Proposal:   A lipid value of 7.9 percent for wildlife BAFs, based
on consumption of whole fish, was included in the proposal.  The lipid value
for the wildlife BAFs was determined using whole fish lipid data from the U.S.
Fish and Wildlife Service National Contaminant Biomonitoring Program and the
Canadian Department of Fisheries and Oceans.  The 7.9 percent lipid value was
the mean of lipid values for all fish, game and nongame, in the entire Great

-------
134    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

Lakes System.  Data for all fish were used because wildlife typically are
nondiscriminatory consumers of fish.

      ii.   Comments:  A few comments indicated that the percent  lipid  value
should be based on fattier fish.  Other commenters felt that 7.9  percent lipid
was based on a reasonable analysis and assumptions, but would like more
information on. prey preferences and whether high lipid organs are
preferentially eaten by wildlife.

      iii.  Final Guidance:  In the final Guidance, the percent lipid for the
actual prey species consumed by the representative wildlife species is  used to
estimate the BAF for the trophic levels at which wildlife consume.  The
percent lipid is based on the consumption patterns of wildlife and cross-
referenced with fish weight and size and appropriate percent lipid  (see final
TSD for BAFs).  This approach is a more accurate reflection of the lipid
content of the fish consumed by wildlife species than the approach used in the
proposal.

      EPA has required use of a percent lipid value for trophic level 4 fish
of 10.31 and 6.46 for trophic level 3 in whole fish for use in determining
wildlife BAFs for organic chemicals in the final Guidance.

4.    FCMs

      a.    Proposal:  As discussed in the proposed Guidance, when a field-
measured BAF is not available, a predicted BAF can be calculated  by
multiplying a laboratory-measured or predicted BCF by a food-chain multiplier.
The FCM accounts for the biomagnification of a chemical through trophic levels
in the food chain.  The FCMs in the proposal were based on a model by Thomann
(1989)  (see 58 FR 20859 and 20861 for a more complete description of the model
and its uses).

      In the August 30, 1994 Notice of Data Availability  (59 FR 44678), EPA
requested comment on use of a food-chain model by Gobas  (1993) which, unlike
the Thomann 1989 model, includes both benthic and pelagic food chains,  thereby
estimating exposure of organisms to chemicals from both the sediment and the
water column.

      b.    Comments:  Many commenters cited the 1992 SAB report  contention
that the FCM model  (Thomann, 1989) has not been adequately peer reviewed or
sufficiently validated to be used in a regulatory framework.  Commenters also
questioned the applicability of one model to a diverse ecosystem  such as the
Great Lakes and encouraged EPA to discuss alternate models.

      Many commenters criticized the use of the Thomann model for not
adequately accounting for metabolism, biotransformation, degradation,
persistence, or seasonal or temporal variability.  In addition, commenters
argued that the model is extremely sensitive to certain input parameters such
as the lipid content and that the input parameters for the FCMs that EPA
calculated were taken directly from Thomann's paper instead of using Great
Lakes specific assumptions.  Other commenters questioned the model assumption
that the system is at steady state, while others supported this assumption due
to the difficulty in describing the ecosystem parameters that would affect the
likelihood of reaching equilibrium.

      Many comments stated that the Thomann model had little application for
chemicals with log KQWS greater than 6.5 because it did not consider sediment
as a route of exposure.  Commenters suggested using alternative food-chain
models such as Thomann  (1992) which incorporates sediment as a route of
exposure; or Thomann and Connolly  (1984) which has Great Lakes specific input
parameters.

      Some commenters on the August 30, 1994 Notice of Data Availability  (59
FR 44678) supported the use of FCMs derived from the Gobas model, assuming

-------
                           Section IV: Bioaccumulation Factors                       135
that no metabolism occurs, because it incorporates the concentration of the
chemical in both sediment and the water column.  Other commenters on the
Notice requested further explanation as to why some FCMs for trophic level 3
are greater than trophic level 4.  Other commenters stated that additional
field validation and documentation of both the Thomann 1989 and the Gobas 1993
model is needed before they can be used in a regulation.

      EPA acknowledges that the 1989 Thomann model does not account for
certain important processes such as metabolism.  Similarly, although the 1993
Gobas model includes a metabolic rate constant, it was set equal to zero due
to a scarcity of data for individual chemicals.  Based on EPA's review of the
commenters1 concerns regarding metabolism, EPA has decided in the final
Guidance to differentiate which BAF data are required to derive Tier I human
health and wildlife criteria for organic chemicals based on whether or not
metabolism is taken into account.

      EPA agrees with commenters that it is important to use Great Lakes-
specific parameters whenever possible and that there should be an attempt to
account for the most sensitive input parameters to the model.  In light of
these concerns, EPA has used Great Lakes-specific input parameters in the
Gobas model that is used to derive FCMs for the final Guidance.  In addition,
EPA selected the model of Gobas (1993) to derive FCMs in part because this
model, in contrast to the model of Thomann (1992), required fewer input
parameters and had input parameters which could be more easily specified.

      EPA agrees that assuming a system is at steady state might be a
simplifying assumption in some cases, but as noted by commenters it is very
difficult to determine the parameters affecting a system reaching equilibrium.
This is especially true in a system as large as the Great Lakes.  The model of
Gobas (1993)  when used with conditions that are not at steady-state will
predict BAFs which are very similar to those obtained from steady-state
conditions.  The concentrations of the chemicals in the water and sediment in
the Gobas model (1993),  are used as input parameters and therefore, the
disequilibrium between the water column and the sediments are included in the
model calculations.   The differences between measured and predicted BAFs when
conditions are changing depends upon the rate of change.  For the Great Lakes,
the rate of change for PCBs and other bioaccumulative chemicals is quite slow
because burial in sediments and volatilization into the atmosphere are the
major routes of removal for the chemicals from the ecosystem.  For all these
reasons, EPA concludes that it is reasonable to continue to use a model for
estimating the FCMs which assumes that the system is at steady state.

      EPA agrees that sediment should be considered as a route of exposure in
the model, especially for chemicals with log KOWS  greater than 6.5.   EPA
considers the model by Gobas (1993) an improvement on the 1989 Thomann model
because it incorporates the exposure of organisms to chemicals from the
sediment by including a benthic food-chain component.

      The FCMs for trophic level 3 in some cases are greater than those for
trophic level 4.  Potential causes of the higher concentrations (on a lipid
basis) in the trophic level 3 fish include: 1) growth rates which are much
slower than the predator fishes; 2) differing rates of depuration and
elimination of the chemical by the predator fishes.  Field-measured BAFs
derived from the data of Oliver and Niimi  (1988)  are, in general,  consistent
with the model results for smelt.  For example, DDT, PCB 66, PCB 70+76, PCB
56+60+81, and PCB 49 have field-measured log BAF™s for large smelt  (trophic
level 3) of 7.93,  7.88,  7.71, 8.12, and 7.66 and for piscivorous fishes
(trophic level 4)  of 7.78, 7.79, 7.56, 7.96,  and 7.13, respectively  (Oliver
and Niimi, 1988).

      EPA in developing the final Guidance FCMs (Table B.I) calculated the
trophic level 3 values by averaging  (geometric mean) the individual FCMs for
sculpin and alewife.  The FCMs for smelt were not included in the calculation
of the trophic level 3 values because these organisms are at a trophic level

-------
136    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

higher than 3 but less than 4.  The FCMs for trophic levels 3 and 4  in  the
final Guidance are composed of fish with diets consisting of solely  trophic
level 2 and 3 organisms,  respectively.

      EPA, in part, agrees with the comments that using one set  of modelling
conditions might not be totally representative of the entire Great Lakes
System.  However, numerous similarities do exist among the food  webs in the
five Great Lakes.  First, all of the Great Lakes have both benthic and  pelagic
food web components.  Second, all of the Great Lakes except for  Lake Erie have
salmonids as their piscivorous fish.  Third, all of the Great Lakes  have their
piscivorous fish occupying the fourth trophic level.  Fourth, all of the Great
Lakes have forage fishes occupying the third trophic level.  The food web used
in the development of the FCMs was based upon a four trophic level food web
with both benthic and pelagic food web components taken from Lake Ontario.
EPA has determined that enough similarities exist among the five Great  Lakes
to derive FCMs using one set of modelling conditions taken from  Lake Ontario.

      In selecting a model to use in developing FCMs, EPA did consider
alternative models,'ie.,  Thomann et al.  (1992), before selecting the model of
Gobas  (1993).  The model of Gobas (1993) required the specification  of  fewer
input parameters for benthic food-web components in comparison to the model of
Thomann et al. (1992).  The parameters required by the model of  Thomann et al.
(1992) for the benthic food web are not readily available and would  have
required assumptions or guesses for the appropriate values.  In  contrast, the
model of Gobas required no assumptions or guesses for the input  parameters
used with the benthic food-web components of the model.

      EPA does not agree with commenters suggesting that additional  validation
of the models is needed before use in the final Guidance.  EPA does
acknowledge that a model is not a perfect simulation of what is  occurring in
an aquatic ecosystem.  However, based on the comparison of field-measured BAFs
(from Oliver and Niimi, 1988) to predicted BAFs, the 1993 Gobas  model
acceptably predicts BAFs for the Great Lakes System.

      c.    Final Guidance:  For the reasons cited above, EPA decided to use
the 1993 Gobas model in the development of FCMs to be used in the final
Guidance.

      For chemicals with log KOWS greater than 6.5,  the proposed Guidance
recommended that a FCM of one should be used when no chemical-specific  data
was available.  In the final Guidance, this is no longer necessary because the
Gobas model allows the derivation of  FCMs for the entire range of K<,ws.

      The resulting FCMs for trophic  levels 2, 3, and 4 along with the  input
parameters for the model, are included in the final TSD for BAFs, and in
appendix B of part 132.

5.    Accounting for the Effect of Metabolism in Predicted BAFs

      a.    Proposal:  The proposed Guidance acknowledged that many  organic
chemicals that are taken up by aquatic organisms are transformed to  some
extent by the organism's metabolic processes, and that the rate  of metabolism
varies widely from one chemical to another.  For most, but not all,  organic
chemicals, metabolism  increases the depuration rate, decreases the BAF, and
reduces  the harmful  effects to the organism.

      Because accounting for metabolism  is  difficult, the proposed BAF
methodology included a provision  that allowed for predicted BAFs to  be
modified if justified by the data  (e.g., if  information  showed bioaccumulation
was reduced by metabolism).

       EPA requested  comments on suggested methods to adjust predicted BAFs  for
chemicals that are metabolized; the types of chemicals or chemical groups  for
which  the BAF might  be affected by metabolism; and  an approach to account  for

-------
                           Section IV: Bioaccumulation Factors                       137
metabolism by using an "effective FCM" when a field-measured BAF is not
available but a laboratory-measured BCF is available.

      b.    Comments:  Some commenters argued that the "effective FCM" should
not be used because it did not encourage data generation.  According to the
commenters, adopting a lower "effective KQW"  to back calculate an "effective
FCM" implies that elimination at all trophic levels would be enhanced by '
metabolism.  However, this may not be the case given that lower trophic levels
may lack metabolic capabilities possessed by higher trophic level animals.
Other commenters stated the "effective FCM" was appropriate for conservative
protection of fish consumers.  Several commenters stated that the GLI does not
address the effects of metabolism of pollutants, such as PAHs.  However,
according to the commenters, no adjustment of BAFs for metabolism should be
included in the GLI because no reliable methods are available to predict
potential decreases and increases in toxicity due to metabolism of pollutants.
Other comments on metabolism have been incorporated in section IV.B.2b.

      EPA acknowledges that metabolism is not incorporated in the FCMs or
predicted BCFs and that use of the "effective FCM" has limited applicability
and may generalize the effects of metabolism.  However, by including a BAF
predicted from the BSAF methodology as the second data preference, EPA is
including an additional method for calculating BAFs that accounts for
metabolism.  In addition, since only field-measured BAFs, BAFs derived from
the BSAF methodology, BAFs less than 125 can be used to derive Tier I criteria
for human health and wildlife,  metabolism is either accounted for in these
measurements or cannot substantially reduce the criterion.  Finally, EPA notes
that for a chemical such as aldrin, transformation to dieldrin, does not
reduce the risk of adverse impacts.

      c.    Final Guidance:   Because of the comments and these modifications
to Tier I data requirements, EPA has not required the use of the an "effective
FCM, " but recognizes that it is a valid method that could be used by States or
Tribes to account for metabolism.

6.     Bioavailability

      a.    Proposal:  In the proposed Guidance, the predicted human health
and wildlife BAFs for organic chemicals were based on the total concentration
of the chemical in water and bioavailability was not taken into account.  EPA
acknowledged in the proposal that for chemicals with log KQWS greater than
6.5, a substantial percentage of the total concentration can be associated
with particulate and dissolved organic matter in water and therefore be
unavailable for accumulation in the water column.  EPA requested comment on
deriving BAFs in terms of "freely dissolved"  chemical, (i.e., that which is
dissolved and not associated with other organic matter) to adjust for the
difference in bioavailability between the site water and the water on which
the predicted BAFs were based.

      In a subsequent Notice dated August 30, 1994 (59 FR 44678), EPA
requested comment on an equation which defines the relationship of a BAF
reported on the basis of the total concentration of the chemical in the water
to a BAF reported on the basis of the freely dissolved concentration of the
chemical in the water:

                             BAF; =  (ffd) (BAF/d)

where:

      BAF)   =     BAF (L/kg of lipid) reported on the basis of the lipid-
                  normalized concentration of chemical in the biota (kg/kg
                  lipid)  divided by the total concentration of the chemical in
                  the ambient water  (kg/L);

-------
138    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      BAF,™ =      BAF  (L/kg of lipid) reported on the basis of the  lipid-
                  normalized concentration of chemical in the biota (kg/kg
                  lipid)  divided by the freely dissolved concentration  of the
                  chemical in the ambient water  (kg/L); and

      ffd    =     fraction of the total chemical in the ambient water that  is
                  freely dissolved.

      The fraction of the chemical in the ambient water that is freely
dissolved, ffd,  can be calculated using the KQW for the chemical and the
concentration of DOC and POC in the ambient
                    f   = 	±_
                    ,fd   .    ( DOC ) ( Kow )
water:                             10
                                               (POC) (Kow)
where:

      POC   =     concentration of particulate organic  carbon,  kg of  organic
                  carbon/L of ambient water;

      DOC   =     concentration of dissolved organic  carbon,  kg of organic
                  carbon/L of ambient water; and

      KOW   =     octanol-water partition  coefficient.

      b.    Comments;..  Some  commenters  stated that  using the  bioavailable
fraction of the chemical in  the ambient water would more accurately reflect
the fraction of the total chemical available to bioaccumulate in the  biota.
Other commenters  stated that it was not possible  to accurately measure the
concentration of  the chemical freely dissolved in water.   Commenters  also
stated that due to a lack of data, the  bioavailable fraction  cannot be
predicted and it  is necessary to use total concentration for  derivation of
BAFs.  Other commenters felt that EPA had  not been  explicit in the definition
of  "concentration in water"  and requested  a more  clearly defined statement.

      Several commenters wanted clarification on  the  suggested POC/DOC values
from Lake Superior and application of the  methodology and an  explanation of
why the baseline  BAFs  were converted to total concentration for the derivation
of criteria.  Other  commenters stated that EPA did  not  evaluate the potential
errors associated with the conversion equations,  variations over time or
throughout a waterbody.

      EPA agrees  with  commenters that taking into account the bioavailable
fraction of the chemical in  the ambient water would more accurately reflect
the fraction of the  total chemical available to bioaccumulate in the biota.
In  the Notice dated August 30, 1994  (59 FR 44678),  EPA  set forth the equation,
fw = 1/(1 + POC • KOW + DOC • Kow/10) , from which  the  fraction of the chemical
that is freely dissolved in  the water can  be calculated using the KQW for the
chemical and the  DOC and POC in  the  ambient water.  EPA acknowledges that the
freely dissolved  concentration of  a  chemical is difficult to  measure, however,
the KOW, DOC and  POC can be  measured or estimated and used to calculate the
freely dissolved  concentration.

      The baseline BAF is based  on the  freely  dissolved concentration of a
chemical, while the  BAF used in  the  derivation of the human health and
wildlife Tier I criteria will  reflect the  total  concentration of the chemical.
In  order to implement  the criteria,  the BAFs need to  be based on a total
concentration of  the chemical  in  the water column because analytical methods
in  40 CFR part  136  that are  used for compliance monitoring determine the total
amount of chemical  in  the water.   The  concentration of  POC and DOC estimated
from Lake Superior  from Eadie  et  al.  (1990) will  be used to calculate BAFs

-------
                           Section IV: Bioaccumulation Factors                       139
based on total concentration for derivation of human health Tier I criteria
and Tier II values and wildlife Tier I 'criteria.  EPA believes that the values
for POC and DOC are.protective of the entire Great Lakes System.  Other values
for POC and DOC may be used to derive site-specific criteria if scientifically
justified.

      Concentrations of POC and DOC are those discussed in the August 30, 1994
Notice and are specified  (see appendix B to part 132) and are accounted for in
the criteria.  The BAFs based on the concentration of the freely dissolved
chemical in the water when used properly will provide the same predicted
residue in aquatic organisms as the BAFs based on the concentration of the
total chemical in water.

      The calculation of the freely dissolved concentration of a chemical in
the water column assumes equilibrium conditions, and therefore variations over
time and throughout a waterbody are negligible.  In EPA's judgement, the
errors associated with the conversion equations are minimal in comparison to
the benefit of normalizing the site-specific parameters of POC and DOC in
calculation of the BAF.

      c.    Final Guidance:  EPA has decided to use the freely dissolved
concentration of organic chemicals in the derivation of baseline BAFs and the
total concentration of the chemical for derivation of Tier I human health and
wildlife criteria. tThe fraction of the chemical in the ambient water that is
freely dissolved, fH,  will be calculated using the KOW for the  chemical  and
the concentration of DOC and POC in the ambient water.  For further details
concerning this equation, see the final TSD for BAFs which is available in the
public docket for this rulemaking.  Basing the measured and predicted baseline
BAFs on the concentration of the freely dissolved chemical in water permits
the derivation of generic BAFs devoid of site-specific influences and
considerations, such as varying concentrations of POC and DOC and allows
consistent usage and derivation of the BAFs throughout the final Guidance.

7.    Calculation of Baseline BAFs

      a.    Proposal:   In the Technical Support Document for the Procedure to
Determine Bioaccumulation Factors - July 1994  (EPA-822-R-94-002) which
accompanied the August 30, 1994, Notice of Data Availability (58 FR 44678),
equations were set forth for calculating baseline BAFs.  Slight modifications
were made to the equations in order to correct for the errors in the original
equations.

      b.    Final Guidance:  In the final Guidance, a baseline BAF shall be
calculated from a field-measured BAF of acceptable quality using the following
equation:
                Baseline BAF =
                                   Measured BAFj
                                         ffd
- l
      where:

          BAF|   =  BAF based on total concentration in tissue and water.

          f,     =  fraction of the tissue that is lipid.

          ffd     =  fraction of the total chemical that is freely dissolved in
                   the ambient water.

      The trophic level to which the baseline BAF applies is the same as the
trophic level of the organisms used in the determination of the field-measured
BAF.

-------
140    Water Quality Guidance for the Great Lakes System — Supplementary Information Document
      A baseline BAF for organic chemical "i" shall be calculated from a
field-measured BSAF. of acceptable quality using the following equation:

              (BaSeUneBM,,.(BAF,,.
      where :
        ( BAFffd )
      (BSAF)i

      (BSAF)
                           BAF based on the measurement of freely dissolved

                           reference chemical in the water column.

                           BSAF for chemical "i".

                           BSAF for the reference chemical "r" .

                           octanol -water partition coefficient for chemical
      (KoW)r
                           octanol -water partition coefficient for the
                           reference chemical "r".
      The trophic level to which the baseline BAF applies is the same as the
trophic level of the organisms used in the determination of the BSAF.

      A baseline BAF for trophic level 3 and a baseline BAF for trophic level
4 shall be calculated from a laboratory-measured BCF of acceptable quality and
a FCM using the following equation:
             Baseline BAF  =  (FCM)
                                      Measured
                                                - 1
                                             -fd
where :

    BCF!

    f,

    fa


    FCM
                   BCF based on total concentration in tissue and water.

                   fraction of the tissue that is lipid.

                   fraction of the total chemical that is freely dissolved in
                   the ambient water.

                   the food- chain multiplier obtained from Table B-l by
                   linear interpolation for trophic level 3 or 4, as
                   necessary .
      A baseline BAF for trophic level 3 and a baseline BAF for trophic level
4 shall be calculated from a KQW of acceptable quality and a FCM using the
following equation:

          Baseline BAF =  (FCM)  (predicted BCF)  =  (FCM) (Kow)

      where :

            FCM    =  the food-chain multiplier obtained from Table B-l by
                      linear interpolation for trophic level 3 or 4,  as
                      necessary.
                   =  octanol-water partition coefficient.

-------
                           Section IV: Bioaccumulation Factors                       141
8.    Other Uses of BAFs

      In the final Guidance, human health BAFs are used to identify chemicals
of greatest concern within the Great Lakes System.  Chemicals identified as
BCCs are those for which extra controls are required as specified  in the final
implementation procedures and under the antidegradation procedures in the
final Guidance.  See discussion of BCCs in section II.G of this document.

9.    Individual BAFs

      a.    2.3.7.8-TCDD

      i.    Proposal:  In the proposed Guidance, the BAF^, for TCDD for
derivation of wildlife criteria was 79,000 for trophic level 3 and trophic
level 4.  The BAFf,*, in the August 30, 1994 Notice of Data Availability  (59
FR 44678) was 320,000 for trophic level 3 and 160,000 for trophic level 4.
The BAFy0%/ for TCDD for derivation of human health criteria in the proposed
Guidance was 50,000 for trophic level 4.  The BAFj0%, in the August 30, 1994
Notice of Data Availability  (59 FR 44678) was 101,000.  Differences between
the proposed values and those in the Notice were based on differences in the
type of data used. The proposed value is based on a laboratory-measured BCF,
while the value in the Notice is based on the BSAF methodology.

      ii.   Comments:  Several commenters stated that the documentation for
the proposed BAF for 2,3,7,8-TCDD was inadequate and that all relevant studies
must be cited and available for review.  Many commenters stated that the
proposal was inconsistent with the Interim Report on Data and Methods for
Assessment of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Risks to Aquatic Life and
Associated Wildlife (EPA, 1993) and argued that the inconsistency was due to
the fact the EPA was not being forthright with all the available information.

      Commenters stated that the basis for the trophic level 3 BAF for
2,3,7,8-TCDD is unexplained and that the scientific basis for what has been
done should be made, available prior to publication of the final Guidance.

      EPA has cited and made available all studies used in the derivation of
the BAF for 2,3,7,8-TCDD.  The data for the basis of the trophic level 3 BAF
was made available in the BSAF section of the Technical Support Document (see
EPA-822-R-94-002).  Any perception of inconsistency between the Interim Report
and the proposal may be due to the way the data are used and the evolution of
the BSAF method after the Interim Report was completed.  The BSAF method does
not require estimation of the concentration of 2,3,7,8-TCDD in water to
determine the BAF as was done in the Interim Report on Data and Methods for
Assessment of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Risks to Aquatic Life and
Associated Wildlife- (EPA, 1993) .

      The BAF for 2,3,7,8-TCDD is greater for trophic level 3 fish than
trophic level 4 fish because the Lake Ontario BSAF data for smelt and trout
indicated this difference.  This may be attributable to growth dilution at
trophic level 4  (i.e., with trout).

      iii.  Final Guidance:  For 2,3,7,8-TCDD, the baseline BAF will be based
on a predicted BAF derived from the BSAF methodology; the baseline BAF for
trophic level 3 is calculated using the ratios of FCMs for log KQW = 7.02.
Both baseline BAFs pake into account the bioavailability of 2,3,7,8-TCDD in
the water column and is lipid normalized.  The baseline BAFs for 2,3,7,8-TCDD
for trophic level 3 and trophic level 4 are 9,360,000 and 9,000,000
respectively.  The BAFs for derivation of Tier I human health criteria for
2,3,7,8-TCDD are 48,490 for trophic level 3 and 79,420 for trophic level 4,
based on the total concentration of the chemical in the water column and
assuming 1.82 and 3.10 percent lipid, respectively.  The BAFs for derivation
of Tier I wildlife criteria for 2,3,7,8-TCDD are 172,100 for trophic level 3
and 264,100 for trophic level 4,  based on the total concentration of the

-------
142    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

chemical in the water column and assuming 6.46 and 10.31 percent  lipid,
respectively.

      b.    PCBs

      i.    Proposal:  The BAFj.,,, for PCB in the proposed Guidance was
1,000,000 for trophic level 3 and 2,800,000 for trophic level 4.  The
in the August 30, 1994 Notice of Data Availability  (59 FR 44678)  was 658,000
for trophic level 3 and 687,000 for trophic level 4.  The BAF15.0%,  for PCB  for
derivation of human health criteria in the proposed Guidance was  1,776,860  for
trophic level 4.  The BAFj0%< in the August 30, 1994 Notice of Data
Availability  (59 FR 44678) was 435,000 for trophic level 4.

      ii.   Comments:  Commenters were concerned that the apparent
disequilibrium between the sediment, water column and fish tissue
concentration makes predictions of BAFs difficult.  In place of the BAF,
commenters suggested use of the BSAF.

      Other commenters stated that the one BAF for all PCBs was not
appropriate but instead BAFs should be congener specific because  of varying
bioavailability and rates of uptake and metabolism.  One commenter suggested
developing BAFs for lower chlorinated and higher chlorinated PCBs.  Commenters
felt that development of a BAF based on the "most prevalent" congeners  did  not
satisfactorily address the problem that there is no single BAF for all  PCBs.
According to the commenters, data show that BAFs for different congeners  vary
widely.  Other commenters were concerned that the BAF in the August 30, 1994
Notice of Data Availability  (59 FR 44678) for trophic level 4 would decrease
the BAF four-fold and therefore result in a more lenient water quality
criteria and believe that EPA has not justified this proposed relaxation  in
PCB BAFs.

      Commenters criticized the Oliver and Niimi  (1988) data set  for PCBs
because of temporal and spatial variability in the data, use of unadjusted
whole fish values, and use of centrifuged water concentrations.

      EPA recommends the use of the BSAF, as detailed in section  B.2.c  above,
if a field-measured BAF is not available.  For PCBs, however, field-measured
BAFs are available for most congeners and therefore will be used  in the
derivation of the baseline BAF.  Field-measured BAFs inherently account for
the effects caused by a) metabolism, b) the disequilibrium of the chemical
between the sediment and the water column, and c) all the other naturally
occurring environmental processes.

      EPA has revised the calculation of the BAF for PCBs based on comments
that use of the most prevalent congeners in Schultz et al.  (1989) is
inappropriate.  In the  calculation of the revised BAF for PCBs, the BAFs  for
the individual congeners were assigned weights based on their concentration in
salmonids as reported by Oliver and Niimi  (1988), as suggested by a commenter.
The revised BAF is based on the concentrations of PCBs in salmonids in  the
Great Lakes and not  the concentration of PCBs in the water column since fish
are the predominant  route of exposure to humans and wildlife.  The BAF  is
based on field-measurements which inherently account for metabolism,
bioavailability, growth, concentration in the sediment and rate of uptake.  It
is necessary to calculate a single BAF in order to derive Tier I  human  health
and wildlife criteria because the corresponding toxicological data are  only
available for Aroclor 1260, a mixture of congeners.  Therefore development  of
a BAF for lower- anfl higher-chlorinated PCBs is not appropriate.

      The BAF for PCBs  was derived using field-measured data for  the congeners
based on their weighted concentration in salmonids.  The BAFs take into
account the bioavailability of the chemical in the water column and are lipid
normalized.   In derivation of the final BAF, the baseline BAF was converted to
a BAF based on the total  concentration of PCBs in the water column.

-------
                           Section IV: Bioaccumulation Factors                       143
      EPA acknowledges the commenters' concerns that the trophic level 4 BAF
decreased four-fold in the August 30, 1994 Notice of Data Availability  (58 FR
44678) as compared to the proposed value.  EPA has recalculated the BAF, as
described in the above paragraph, which resulted in a baseline BAF value of
116,600,000 for trophic level 4.

      EPA lipid normalized the fish tissue concentrations given by Oliver and
Niimi (1988) in order to avoid use of unadjusted whole fish values, and
calculated the freely dissolved concentration of a chemical in the water
column to account for the centrifuged water concentrations.  EPA has
determined that if these parameters are accounted for, much of the variability
is eliminated.

      iii.  Final Guidance:  The baseline BAF for PCBs is based on field-
measured BAFs for the congeners of PCBs based on their concentration in fish
and thus takes into account the freely dissolved concentration of PCBs in the
water column and lipid normalization.  The baseline BAFs for trophic level 3
and trophic level 4 are 55,280,000 and 116,600,000, respectively.  The BAFs
for derivation of Tier I human health criteria for PCBs are 520,900 for
trophic level 3 and 1,871,000 for trophic level 4, based on the total
concentration of the chemical in the water column and assuming 1.82 and 3.10
percent lipid, respectively.  The BAFs for derivation of Tier I wildlife
criteria for PCBs are 1,850,000 for trophic level 3 and 6,224,000 for trophic
level 4, based on the total concentration of the chemical in the water column
and assuming 6.46 and 10.31 percent lipid, respectively.

      c.    Mercury

      i.    Proposal:  The proposed mercury BAF was 60,000 for trophic level 3
and 130,000 for trophic level 4.  Subsequent to the proposal, EPA asked for
comment on a recalculated mercury BAF in the August 30, 1994 Notice of Data
Availability  (59 FR 44678) .  The recalculated BAF was 28,000 for trophic level
3 and 140,000 for trophic level 4.  The derivation of the recalculated BAF
differed from the proposed BAF for mercury in the following ways (see EPA-822-
R-94-002 for discussion of the data used):

      1)    The estimated percent of total mercury in water that is
methylmercury was changed from 25 percent to 17 percent.

      2)     The estimated biomagnification factors (BMFs) were changed as
follows:

      Trophic Level           New BMF           Old BMF
      2 to 3                  1.26              2.154
      3 to 4                  5.0               2.154

      3)     The estimated percent of the total mercury in fish that is
methylmercury was changed from 85.3 percent to 97.5 percent.

      ii.    Comments:  Many commenters stated that the derivation of the BAF
for mercury is not scientifically valid because the assumptions regarding the
bioavailability and fate of mercury in the water column were conservative and
EPA had not accurately assessed the sequestration of mercury in the
environment, especially in sediment.  Other commenters advocated site-specific
modifications for mercury, stating that a single BAF is oversimplistic and
ignores variations.

      Commenters stated that the mercury BAF in the August 30, 1994 Notice of
Data Availability (59 FR 44678)  did not utilize available field data
submitted.

      In development of a scientifically-defensible mercury criteria,
commenters stated that EPA must consider that fish accumulate the majority of
mercury via ingestion of contaminated sediments and food, not by gill uptake

-------
144    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


of dissolved mercury.  Commenters recommended using the BSAF method to
recalculate BAF for mercury.

      EPA does not agree that the BAF for mercury is not scientifically valid.
The assumptions used by EPA regarding the bioavailability and fate, with one
exception described below, were based on the best available data including  •
that submitted by commenters.  After review of the data, EPA revised the-
percent of total mercury which is found as methylmercury from 25 percent to 17
percent based on data of Gill and Bruland  (1990).

      EPA also does not agree that it did not accurately assess the
sequestration of mercury in the environment.  Sequestration of an inorganic
chemical in sediment is generally not important because BAFs are derived based
on the relationship between fish tissue concentrations and the water column.
For chemicals like mercury, methylation appears to occur in the sediment which
releases methylmercury into the water column.  This is different from the
transfer through the benthic food-chain component that is part of the Gobas
model used for organic chemicals.  If partitioning to the water column  occurs,
BAFs based on the bioavailable portion would be higher than the BAFs in the
final Guidance which are based on total.

      EPA agrees with the commenters that exposure of aquatic organisms to
mercury is primarily through food and sediment and has developed a
biomagnification factor for mercury based on field data.  This was necessary
because the 1993 Gobas model is only applicable to organic chemicals.   The
BSAF method as described in the final Guidance is also applicable to only
organic chemicals, and therefore cannot be used in the derivation of the BAF
for mercury.  Mercury's BAF is derived based on the percentage of methyl
mercury and amount of trophic uptake as explained in appendix D of the  July
1994 TSD for BAFs  (EPA-822-R-94-002).  EPA has provided the rationale for the
derivation of the BAF in the final TSD for BAFs which is available in the
public docket for this rulemaking.

      EPA has completed a more comprehensive national analysis of the data
concerning the bioaccumulation of mercury by fish, which is being peer
reviewed at this time.  EPA had intended to use in the final Guidance the
baseline BAFs contained in the initial draft of the report but EPA decided to
wait until the report has been peer-reviewed and completed.  EPA is planning
to issue a proposal* to revise the mercury criterion in the final Guidance to
reflect this new information.

      iii.  Final Guidance:  The baseline BAF for mercury is based on a
laboratory-measured BCF and a biomagnification factor.  The baseline BAFs for
both wildlife criteria and human health criteria and values for trophic level
3 are 27,900 and for trophic level 4 are 140,000.

C.    Conformance to the CWA. Great Lakes Water Quality Agreement and Great
      Lakes Critical Programs Act of  1990

      Section 118(c) of the CWA requires EPA to develop, "inter alia,"
guidance on minimum water quality limits to protect human health, aquatic life
and wildlife in  the Great Lakes System.  The Great Lakes Critical Programs Act
of 1990  (CPA) states that the final Guidance shall be no less restrictive than
the provisions of the CWA, National water quality criteria and National
guidance, and shall conform with  the  objectives and provisions of the Great
Lakes Water Quality Agreement  (GLWQA).  For reasons set out in the preamble to
the final Guidance  (see 58 FR 20858), EPA has determined that the final
Guidance meets these requirements.

D.    Adoption of Water Quality Standards Consistent with the Final Guidance

      The final  Guidance  for deriving BAFs  is included in appendix B to part
132.  The BAF TSD, which  discusses the basis for the final methodology  and
which sets  forth the data and considerations upon which the individual  BAFs

-------
                           Section IV: Bioaccumulation Factors                       145
are based, is available in the docket for this rulemaking.  Copies are also
available upon written request as described in section XIII of this document.

E.    References

      ASTM. 1990. Standard Practice for Conducting Bioconcentration Tests with
Fishes and Saltwater Bivalve Molluscs.  Designation E 1022 - 84.  Pages 606-
622.  In Annual Book of ASTM Standards.  Section 11, Water and Environmental
Technology, Volume 11.04.  American Society for Testing and Materials.  1916
Race Street, Phila., PA 19103.

      De Vault, D.S. 1993. Data on Contaminant Trends in Lake Trout
(Unpublished). Contains data for 1984-1990.

      Eadie, B.J., N.R. Morehead, and P.P. Landrum.  1990. "Three-phase
partitioning of hydrophobic organic compounds in Great Lakes waters."
Chemosphere, 20, 161-178.

      Gill, G.A., and K.W. Bruland.  1990.  Mercury Speciation in Surface
Freshwater Systems in California and Other Areas.  Environ. Sci. Technol.
24:1392-1400.

      Gobas, F.A.P.C.  1993, "A model for predicting the bioaccumulation of
hydrophobic organic chemicals in aquatic food-webs: application to Lake
Ontario."  Ecological Modelling, 69, 1-17.

      Hamelink, J.L., R.C. Waybrant and R.C. Ball. 1971.  A proposal: exchange
equilibria control the degree chlorinated hydrocarbons are biologically
magnified in lentic environments. Trans. Amer. Fish. Soc. 100: 207-214.

      Mackay, D. 1982.  Correlation of bioconcentration factors.  Environ.
Sci. Technol. 16:274-278.

      Niimi, A.J. 1985.  Use of laboratory studies in assessing the behavior
of contaminants in fish inhabiting natural ecosystems.  Water Poll. Res. J.
Canada 20:79-88.

      Oliver, B.C. and A.J. Niimi. 1983.  Bioconcentration of chlorobenzenes
from water by rainbow trout: correlations with partition coefficients and
environmental residues.  Environ. Sci. Technol. 17:287-291.

      Oliver, E.G. and A.J. Niimi. 1988.  Trophodynamic analysis of
polychlorinated biphenyl congeners and other chlorinated hydrocarbons in the
lake Ontario ecosystem.  Environ. Sci. Technol. 22:388-397.

      Schultz, D.E., G. Petrick, and J.C. Duinker, 1989.  Complete
Characterization of Polychlorinated Biphenyl Congeners in Commercial Aroclor
and Clophen Mixtures by Multidimensional Gas Chromotography-Electron Capture
Detection.  Environ. Sci. Technol. 23: 853-859.

      Swackhammer, p.L. and R.A. Kites. 1988.  Occurrence and bioaccumulation
of organochlorine compounds in fishes from Siskwit Lake, Isle Royale, Lake
Superior.  Environ. Sci. Technol. 22:543-548.

      Thomann, R.V. 1989.  Bioaccumulation Model of Organic Chemical
Distribution in Aquatic Food Chains.  Environ. Sci. Technol. 23:699-707.

      Thomann, R.V. and J.P. Connolly. 1984. Model of PCB in the Lake Michigan
lake trout food chain. Environ. Sci. Technol. 18: 65-71.

      Thomann, R.V., J.P. Connolly, T.F. Parkerton. 1992. An Equilibrium Model
of Organic Chemical-Accumulation in Aquatic Foodwebs with Sediment
Interaction.  Environ. Toxicol. Chem. 11: 615-629.

-------
146    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      U.S. EPA. 1980. Seafood consumption data analysis.  Stanford Research
Institute International, Menlo Park.  California.  Final Report.  Task 11,
Contract No. 68-01-3887.

      U.S. EPA guidance contained  in  Stephan et al.  (1985)  "Guidelines for
Deriving Numerical National Water  Quality Criteria for the  Protection of
Aquatic Organisms and Their Uses."  NTIS  # PB85-227049.  U.S.  Department of
Commerce, 5285 Port Royal Road,  Springfield,  VA 22161.

      U.S. EPA. 1993. Interim Report  on  Data and Methods for Assessment of
2,3,7,8-Tetrachlordibenzo-p-dioxin Risks to  Aquatic Life and Associated
Wildlife. EPA/600/R-93/055.

      Veith, G.D. and P. Kosian. 1983.   Estimating Bioconcentration Potential
from Octanol/Water Partition  Coefficients.   Chapter 15 in PCBs  in the Great
Lakes.  Mackay, D., R.  Patterson,  S.  Eisenreich,  and M.  Simmons (eds.) Ann
Arbor Science.

      West, P.C. et al. 1989.   Michigan  Sport Angler Fish Consumption Survey:
A Report to the Michigan Toxic Substance Control Committee, University of
Michigan Natural resources Sociology  Research Lab.  Technical Report #1, Ann
Arbor, Michigan, MDMB Contract # 87-20141.

      Zipf, G.W..  Memorandum to Michael Cox.  1995.

-------
                              Section V: Human Health                           147
                      V.   HUMAN  HEALTH
A.    Summary of Final Rule

      The methodology for the development of human health criteria and values,
as outlined in appendix C of part 132,  is identical to the Guidance proposed
on April 16, 1993 except for the following changes:

      -- When deviations from values presented in the Integrated Risk
Information System (IRIS)  are anticipated or considered necessary,  it is
strongly recommended that such actions be communicated to the EPA Reference
Dose  (RjD) and/or Cancer Risk Assessment Verification Endeavor (CRAVE) work
group immediately.

      - - To determine the weight of evidence of carcinogenicity  for a
chemical, and to determine (on a case-by-case basis)  whether a Group C
chemical should be a Tier I criterion or Tier II value the following data (if
available)  shall be considered:  mutagenicity/genotoxicity (determinations  of
whether the chemical interacts directly with DNA); structure activity;
metabolism and mode of action.

      -- Clarification has been added to the discussion on additional
uncertainty factors which may be applied to short-term (28-day)  study results
used in the development of Tier II values.  In some cases, an uncertainty
factor, from 1-10,  may be needed to extrapolate from 28-day study results to
subchronic  (90-day) results.   This decision must be made on a case-by-case
basis.

      - - If the duration of a cancer bioassay is significantly less than  the
natural lifespan of the test animal, the slope may be adjusted on a
case-by-case basis to compensate for latent tumors which were not expressed
during the cancer bioassay.  This is a change from the proposal  which required
an adjustment to compensate for latent tumors if the cancer bioassays were
shorter than 78 weeks for mice and 90 weeks for rats.

      -- A default relative source contribution factor (RSC)  of  0.8 will  be
applied to all noncarcinogenic chemicals.  In the proposed Guidance,  an RSC of
0.8 was applied to noncarcinogenic bioaccumulative chemicals of  concern (BCCs)
and an RSC of 100 percent was applied to noncarcinogenic nonbioaccumulatives.
If actual exposure data exists, States and Tribes may use this data,  following
procedures outlined in the 1980 National Guidelines to calculate an RSC.

      -- A fish consumption rate of 15 grams will be applied in  the derivation
of criteria (3.6 grams for trophic level 3 fish and 11.4 grams for trophic
level 4 fish).
      - - The minimum data requirements for deriving a Tier I criterion have
been changed to include minimum data requirements for bioaccumulation factors
as well as minimum toxicological data.   For organic chemicals, field-measured
bioaccumulation factors (BAFs), predicted BAFs based on the biota-sediment
accumulation factor (BSAF), and BAFs less than 125 will be used  in calculating
Tier I criteria.  For inorganic chemicals, either a field-measured BAF or a
laboratory-measured BCF can be used to calculate Tier I criteria.

B.    Explanation of Final Provisions

      The final Guidance for human health is described below. As with the
final Guidance for aquatic life discussed earlier, EPA is proposing a two-tier
approach for human health.  Minimum data requirements for Tier I criteria and
Tier II values are discussed later in this section.

-------
148    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Sample human health criteria have been calculated using the proposed
methodology for 18 chemicals.  The 18 chemicals chosen for criteria
development were selected from the Great Lakes Water Quality Initiative
(GLWQI) group of chemicals of concern listed in proposed 40 CFR part 132,
Table 6, to represent a broad cross- section of the types of chemicals  found
in the Great Lakes basin.  The intent of the chemical selection was to  test
the final methodology against a broad range of chemicals and to demonstrate
how the criteria development process will be carried out.  Selection from
among the chemicals of concern was made from the perspective of demonstrating
the final methodologies' applicability to all types of chemicals, rather than
on the basis of health risk priorities.

C.    Criteria Methodologies

      The  final Guidance establishes methodologies to derive human health
criteria which will not result in zero risk, but will provide a level of
protection likely to be without appreciable risk.

1.    Endpoints Addressed

      a.    Proposal:  The proposed Guidance addressed noncancer and cancer
effects only.  Organoleptic effects  (taste and odor), while of concern  from an
aesthetic standpoint, were not considered a significant health concern  and
therefore were not included in the proposal.  EPA asked for comment on  whether
EPA should require the Great Lakes States and Tribes to adopt Tier I criteria
identical to the existing National guidance for organoleptic substances
developed under section 304(a) of the Clean Water Act  (CWA) .

      b.    Comments:  Several commenters were in favor of the EPA proposal to
focus only on cancer and noncancer criteria development.  Several other
commenters disagreed, stating that existing National guidance for organoleptic
criteria should be required to be adopted by the States.  If organoleptic
criteria are not in place, these commenters argued, people may turn to  other
sources of drinking water  (e.g., bottled) or may turn away from Great Lakes
fish  (if they are tainted) in favor of other protein sources.  The commenters
also  stated such alternatives may have negative health effects: the quality of
bottled drinking water may be unknown and the health implications of eating
other  sources of pr&tein  (or sources of food) such as beef, chicken, pork etc.
may reduce one's exposure to bioaccumulated chemicals but may increase  one's
exposure to higher cholesterol.

       c.    Final Guidance:  The final Guidance will continue to focus  on
noncancer and cancer effects and will not require the adoption of Tier  I
organoleptic criteria.  While it is conceivable that some people may seek
alternative water and food sources due to organoleptic properties of chemicals
and such choices may themselves entail a risk of incurring adverse health
effects, EPA believes that the human health protection goals reflected  in the
CWA are best served.by  focusing on actual, health-related effects  (cancer and
noncancer effects) due  to the exposure to the aquatic resource itself.
Moreover, the current National criteria guidance developed for organoleptic
effects are available for use by Great Lakes States and Tribes in developing
criteria.   If States or Tribes want to set  criteria based on more stringent
organoleptic criteria,  as set forth in the  1980 Federal Register Notice of
Water Quality Criteria Documents  (U.S. EPA,1980), they are free  to do so..

2.     Mechanism  of Action

       a.    Cancer.

       i.    Proposal:   In the proposal, EPA regarded   carcinogenicity
generally as a non-threshold adverse  health effect.  Given  that  assumption,
"no effect" levels for  carcinogens  (other than  zero) were not established,
because even extremely  small doses are assumed  to potentially elicit a  finite
increase in the  incidence of cancer.  The proposed Guidance would have

-------
                              Section V: Human Health                           149
provided that a non-threshold mechanism be assumed for carcinogens unless data
exist that demonstrate a threshold mechanism.  Such a finding would have to
weigh all the studies for a particular chemical to determine whether a true
threshold effect is occurring across all sensitive test species.  The proposal
also recommended that States and Tribes confer closely with EPA prior to
submitting for EPA approval any criterion for a carcinogen that is based on
the assumption that a "safe" threshold dose exists for the chemical.  The EPA
proposal also stated that the Linearized Multistage Model  ("LMS") would be
used to extrapolate from actual animal bioassay data to the dose/response
relationship expected at low doses, unless it can be established on a
case-by-case basis that another model such as the "time-to-tumor" model or
ones based on modifications of the LMS model, are more appropriate.

      ii.   Comments:  In response to the EPA-held assumption that
carcinogenisis is a nonthreshold event, several commenters stated that EPA
relies too heavily on the nonthreshold model of carcinogenicity and rarely
factors into the weight of evidence such information as pharmacokinetic data,
genotoxicity and mutagenicity data, structure activity and mode of action.
With regard to the use of the LMS as a default cancer model, several
commenters supported the continued use of the LMS as a default cancer potency
model.  Other commenters stated that the use of the LMS model is outdated and
does not consider all the available mechanistic, pharmacokinetic and other
relevant data in assessing cancer potency.

      iii.  Final Guidance:  Based on the rationale below, EPA will continue
to include the use of the LMS model as a default model in the final Guidance,
unless it can be established on a case-by-case basis that another model is
more appropriate.  As the 1986 Guidelines for Carcinogen Risk Assessment
states:  When pharmacokinetic or metabolic data are available, or when other
substantial evidence on the mechanistic aspects of the carcinogenesis process
exist, a low dose extrapolation model other than the linearized multistage
procedure might be considered more appropriate on biological grounds.  When a
different model is chosen, the risk assessment should clearly discuss the
nature and weight of evidence that lead to the choice (U.S. EPA, 1986).

      EPA believes the nonthreshold assumption of carcinogenesis is a valid
assumption in the absence of data which indicate otherwise.  EPA also
reiterates that the methodology provides an opportunity to demonstrate a
threshold mechanism for a chemical and that on a case-by-case basis another
model (threshold or'non-threshold) can be applied to the cancer data if it can
be established that the other model is more appropriate.

      EPA disagrees with the comment that it does not consider all the data in
making judgements on potential carcinogens.  EPA considers all the available
data when conducting a cancer assessment for a chemical, including
pharmacokinetic, metabolic, mutagenicity/genotoxic and structure activity
data.  To ensure that this occurs with regard to development of criteria in
the future,  EPA has revised the human health methodology in appendix C to part
132 to provide for consideration of each of these specific pieces of evidence
when evaluating the-potential carcinogenic risk of a chemical.

      with regard to the choice of a cancer model, and specifically the
appropriateness of continuing reliance on the LMS, EPA believes that, in the
absence of adequate information to the contrary (such as information on the
mechanism of carcinogenic action), the LMS is the best of the mathematical
extrapolation models used for extrapolating from high dose to low dose.  As
stated in the 1986 Guidelines for Carcinogenic Risk Assessment:  When data and
information are limited,  and when much uncertainty exists regarding the
mechanism of carcinogenic action,  models or procedures which incorporate low
dose linearity are preferred when compatible with the limited information.  In
the absence of adequate information to the contrary, the linearized multistage
procedure will be employed (U.S. EPA, 1986).  Because of the uncertainties
associated with dose response, animal to human extrapolation, and the serious
public health consequences that could result if risk were under-estimated, EPA

-------
150    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


believes that it is prudent and consistent with public health goals of the CWA
to use the LMS to estimate cancer risk for ambient water quality criteria.
The LMS has been endorsed by four agencies in the Inter-Agency Regulatory
Liaison Group and was characterized as less likely to under-estimate risk at
the low doses typical of environmental exposure than other models that could
be used (Inter-Agency Regulatory Liaison Group, 1979) .

      b.    Noncancer.

      i.    Proposal:  The proposed Guidance would have provided that a
threshold mechanism of action be assumed in deriving criteria for protection
against noncancer effects, unless it is demonstrated on a case-by-case basis
that there is no threshold with respect to a given chemical's toxicity
effect(s).  This means there is a dose below which no-adverse effects should
be observed, or if an adverse effect is observed, the risk of deleterious
effect over the span of a lifetime is not appreciable.  EPA also recognized
that there may be exceptions to this principle:  for some non-carcinogenic
effects, no identifiable threshold of effects has been demonstrated.
Chemicals which may exert non-threshold non-cancer effects include genotoxic
teratogens and germline mutagens.

      ii.    Comments:  Several commenters agreed that EPA should allow for a
showing that a chemical has no threshold; however these same commenters did
not express an opinion regarding the default assumption of a threshold for
noncarcinogens.   One commenter believed all noncancer effects should be
considered nonthreshold events unless proven otherwise.  Other commenters
believed making such a showing would be difficult and possibly very costly.
Some commenters suggested that EPA apply a "weight of evidence" approach to
developing an RfD; that we not ignore studies showing no adverse effects while
focussing only on studies indicating adverse effects.

      iii.  Final Guidance:  For the reasons stated below, the final Guidance
will continue to assume that noncancer endpoints of toxicity exhibit a
threshold unless data indicate otherwise.  This is consistent with the 1980
National Guidelines which concluded for noncarcinogens that there is a dose
below which no adverse effects should be observed.  It is also EPA's
experience  (in developing RfDs) that noncancer effects are regarded almost
without exception as threshold events and that nonthreshold noncancer effects
are extremely rare and in some cases may be perceived as such due to a lack of
toxicological information  (e.g., extremely low doses have not been tested or
cannot be measured).  The possible exceptions to this rule, as stated in the
proposal,  are genotoxic teratogens and germline mutagens.  EPA reiterates that
in the rare instances that this type of chemical is encountered, it is
recommended that States and Tribes confer closely with EPA prior to
establishing a noncancer criterion on the basis of a non-threshold effect.
EPA believes the proposed language is adequate in allowing a demonstration of
nonthreshold noncarcinogenicity and is adopting this language in the final
methodology.  EPA has provided additional guidance in "Great Lakes Water
Quality Initiative Technical Support Document for Human Health Criteria and
Values" (EPA 820-B-95-007)  (Human Health TSD) on determining whether a
chemical demonstrates nonthreshold characteristics.

      With regard to the comment that EPA treat all chemicals as if they are
nonthreshold chemicals, EPA believes the overwhelming scientific evidence
supports the assumption that noncancer endpoints are threshold events.  EPA
most often assumes that noncarcinogenic and/or nonmutagenic changes have a
threshold, that  is, a. dose level below which a response is unlikely, because
homeostatic, compensating and adaptive mechanisms in the cell protect against
toxic effects at levels below this threshold.  Therefore, in EPA's judgement
it would be inappropriate to assume that noncarcinogens act by nonthreshold
mechanisms, when the overwhelming evidence suggests otherwise  (U.S. EPA,
1991) .

-------
                              Section V: Human Health                            151
      With regard to the comment that EPA should apply a weight-of-evidence
approach in developing noncancer criteria and values, EPA believes  it  does
employ a weight-of-evidence approach in that it evaluates all the data before
choosing a specific study upon which quantification  is based.  EPA  believes
its process for evaluating noncancer effects is similar to the cancer  weight -
of-evidence approach.  For example, in developing a  noncancer criterion, EPA
reviews all the toxicological effects data and determines which  study  appears
to reflect the most critical endpoint.  Negative studies  (i.e.,  those  that do
not elicit a responee) as well as positive studies  (i.e.,.those  that do elicit
a response) are considered.  IRIS coversheets describe the critical study
which serves as the basis for an RfD but also lists  all the critical and
supporting studies that were considered in the development of the RfD.  EPA
agrees that the main focus when evaluating the potential adverse effects of a
chemical are on studies that show adverse effects.   However, EPA believes this
is reasonable to ensure that humans are protected against potential adverse
effects.  EPA will continue to focus  on studies which are most  relevant to
the consideration of human risk assessment and with  the development of more
pharmacokinetic data, which can clearly identify metabolic/toxicokinetic
differences between species, human risk assessments  will become  easier.

3.    Choice of Risk Level

      a.    Proposal:  The proposal derived criteria which correspond  to a
plausible upper bound increased incremental risk of  developing cancer  of one
in 100,000 (10"5)  over a lifetime of exposure.  The choice of 10'5 risk  level
was recommended by the Initiative Committees and is  within a range  of  risk
levels  (i.e., 10"4 to 10"*)  that EPA considers to be adequately protective and
which EPA has historically considered acceptable in  making regulatory
decisions.  The majority of the Great Lakes States traditionally have  used a
10"5 risk level in setting their water quality criteria.  EPA asked  for comment
on the choice of risk level, and on alternate risk levels, such  as  10"* and
10"",  which could be adopted in the final Great Lakes human health criteria
methodology.

      b.    Comments:  Many commenters were in favor of the proposed risk
level of one in 100,000, while  other commenters were in favor of higher and
lower risk levels (one in 10,000 and one in a million).  Those who  favored one
in a million risk believed one in 100,000 was not stringent enough, and that
many of the exposure parameters in the criterion equation were not  protective
of high-end exposed individuals (e.g., high-end fish consumers). That  is, many
people who eat more*than 15 grams of fish per day would be protected at less
than 1 in 100,000 risk from cancer (that is, a higher risk).  Those that
favored a higher risk level, at one in 10,000, believed that EPA's  cancer
model, and other exposure assumptions are already so conservative that there
is no need to set criteria at the one in 100,000 risk level.  Other commenters
referred to EPA's Superfund program,  which uses a risk level of  one in 10,000,
and cited inconsistency among EPA programs.

      c.    Final Guidance:  The final Guidance continues to derive criteria
and values based on a 10'5 risk level.  EPA believes  that this is a  reasonable
decision in light of the fact that it reflects the policy preferences  of most
of the Great Lakes States.  EPA notes, however, that selection of this risk
level does not reflect a judgment that this is the only level of acceptable
risk that would achieve the human health protection  goals of the CWA.  Rather,
as noted above, EPA believes that ensuring protection in the range  of  10"* to
10"6 is acceptable and consistent with the CWA's objectives.  Commenters were
correct in noting that some identifiable subpopulations or groups consume more
fish than 15  grams/day, as assumed in the methodology, and thus  such consumers
may face a risk that is higher than 10"5.  However,   EPA believes these
consumers will nonetheless be adequately protected.   (See section V.B.S.e of
this document for detailed information on fish consumption in the Great Lakes
basin.)

-------
152    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      EPA disagrees with those commenters advocating the use of  a  one  in
10,000 risk level.  While the model used to estimate cancer risk has many
conservative assumptions that are likely to overestimate actual  risks,  EPA
believes this conservatism is appropriate given the severity of  the potential
outcome  (cancer) and the uncertainties inherent in extrapolating from  the  high
doses administered to laboratory animals to the low doses experienced  by
humans.  Moreover, setting the target risk level at 10"4 for average consumers
would not include any margin of safety that might be appropriate to ensure the
protection of high-end consumers.  With regard to the commenter's  concern  with
the conservative nature of the exposure assumptions, EPA attempts  to select
reasonable assumptions given the need to provide adequate protection to
humans.

       EPA disagrees with the commenter who asserted that EPA requires the use
of a 10"4 risk level in the Superfund program, and that the risk  level  used in
the proposed Guidance's human health methodology is therefore inconsistent
with the Superfund program.  Under EPA's Superfund program, acceptable
exposure levels are generally those falling within the  10"4 to 10"* risk range.
See 40 C.F.R. 330.430  (2) (i) (A) (2) .

      Finally, EPA disagrees that the final Guidance is inconsistent with
other EPA programs regarding the use of a 10"5 risk level.  As stated above,
EPA believes the 10"4 to 10"* risk level will provide adequate protection of
public health.  In any particular regulatory action under its various
authorities, EPA targets a level of public health protection that  effectively
implements EPA's statutory duties and which is tailored to the specific
circumstances being addressed by the regulatory action.  As stated previously,
EPA believes the approach in the final Guidance is consistent with EPA's
historical practices in its public health protection programs.

      As noted in section II of this document, levels of some pollutants fish
tissue in the Great Lakes basin currently demonstrate that applicable  criteria
are being exceeded, and baseline risks associated with these exposures in  some
cases exceed the 10"4 to 10"* risk range that EPA considers adequately
protective.  In particular, groups that generally consume more fish on average
than the general population  (e.g.. Native American subsistence fishers and low
income minority sport anglers) are at greatest risk, which EPA has estimated
may be as high as 3.7 x 10"2 for Native Americans in Lake Michigan  due
primarily to PCBs.  The purpose of the human health criteria and the criteria
methodology contained in the final Guidance is to ensure that the  adoption of
water quality standards that, where they are attained would provide adequate
public health protection.  Obviously, where standards are not yet  attained,
actual risk levels will be higher and steps will have to be taken  to meet
applicable standards.  Nonetheless, EPA evaluates the protectiveness of the
criteria and the methodology assuming that criteria are met in order to ensure
that protective water quality standards will be established in the Great Lakes
basin.

      With regard to comments that EPA should adopt a lower risk level (i.e.,
10-6), EPA notes that States and Tribes are free to adopt a more stringent
approach than that contained in the final Guidance.  Given that  a  plurality of
States have utilized the 10-5 risk level, EPA believes that the  choice of  risk
level is a reasonable one.

      With regard to criteria for protection against non-cancer  effects, these
criteria are derived so as to prevent hypothetically exposed individuals
 (i.e., those consuming pollutant-bearing fish and drinking water at the rate
assumed  in the criteria-derivation formulas explained below) from  receiving a
dose of  the chemical above that which is calculated to correspond  to no
appreciable risk of adverse effect, based on a threshold model of  chemical
activity.  The issue of risk level is therefore not relevant to  such threshold
effects  (except  in these rare instances where a State or Tribe may find, and
EPA would agree, that a noncancer effect exhibits no threshold).

-------
                              Section V: Human Health                           153
      For the reasons stated above, the final Guidance will continue to derive
criteria and values based on a 10'5 risk level.

4.    Acceptable Dose

      The proposed Guidance used the term risk associated dose  (RAD) to
represent the dose associated with a one in 100,000 plausible upper bound risk
of developing cancer from lifetime exposure to a carcinogen.  The term
Acceptable Daily Exposure  (ADE) -was used to represent the dose of a
noncarcinogen expected to result in no appreciable risk of adverse health
effects from lifetime exposure.

      a.    RAD.

      The proposed Guidance discussed the steps for determining a Risk
Associated Dose  (RAD)  (see 58 FR 20865) .  EPA also pointed out that many of
the steps in developing a RAD may have already been conducted by EPA for a
particular chemical and the results made available to the public through EPA's
Integrated Risk Information System (IRIS). It was recommended that IRIS be
consulted when developing a RAD.   (The reader is referred to section V.C.4.C
for a description of IRIS and its utility in developing RADs.)

      i•    Biologically Relevant versus Sensitive Species

      (A).  Proposal:  EPA pointed out that when animal studies are used to
estimate effects on humans, data from species most biologically relevant to
humans are generally preferred  (i.e., a species in which pharmacokinetics
and/or toxic mechanisms of action appear closely related to humans).  In the
absence of data to distinguish the most relevant species, data from the most
sensitive animal species tested, i.e., the species exhibiting a carcinogenic
response at the lowest administered dose, given a relevant route of exposure,
should be used.

      (B).  Comments:  The majority of commenters argued that the most
biologically relevant species rather than the most sensitive species should be
used. Commenters stated that use of the most sensitive species must be weighed
against the assumption that humans are as or more sensitive than the most
sensitive species. A minority of commenters agreed with EPA's proposed
language to use the most sensitive species as a default, in the absence of
data.

      EPA agrees that the most biologically relevant species to humans should
be used when data is available for the chemical on the mechanism of the
carcinogenesis as well as the pharmacokinetics of the test species relative to
humans.   However, in many cases there will be limited data to assess how well
an animal model reflects human toxicological response for a chemical. In these
situations,  EPA believes it is prudent to use data from the most sensitive
species.  This is supported by the 1986 Guidelines for Carcinogen Risk
Assessment which states: In the absence of appropriate human studies, data
from a species that responds most like humans should be used, if information
to this effect exists...because it is possible that human sensitivity is as
high as the most sensitive responding animal species, in the absence of
evidence to the contrary,  the biologically acceptable data set from long-term
animal studies showing the greatest sensitivity should generally be given the
greatest emphasis, again with due regard to biological and statistical
consideration (U.S.'EPA, 1986).

      (C).   Final Guidance:  For the reasons stated above, the final Guidance
is retaining the proposed language that data from species most biologically
relevant to humans is preferred.  However,  in the absence of data to
distinguish the most relevant species, data from the most sensitive species
will be used.   This provision applies to both cancer and noncancer effects.

-------
154    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      ii.    Less than Lifetime Adjustment Factor.

      (A).   Proposal:  In the proposed Guidance,  EPA included   a. factor
(L/Le)3, where  L is the natural  lifespan of  the test species and Le is the
duration of the study.  This adjustment factor is necessitated when an  animal
study being used to derive a dose response relationship is not of sufficient
duration to measure cancer development over the natural lifespan of the test
species.  The rationale for such a factor is described in detail in the 1980
National Guidelines (45 PR 79352).  Under the proposal, the slope factor
adjustment was applied to mouse and rat data if the study duration  (Le) was
less than 78 weeks for mice or 90 weeks for rats, by multiplying the  cancer
slope factor by the factor (L/Le)3.   EPA requested comment on whether the use
of this adjustment factor for studies with less than lifetime duration  was
appropriate.

      (B).   Comments:  Several commenters agreed that an adjustment factor
could be used but disagreed with the choice of the factor proposed and  the
cutoff  (78 and 90 weeks)  chosen in the proposed Guidance.  These commenters
suggested that the adjustment factor should be applied on a case-by-case
basis, considering such factors as mechanism of action, and type of tumor and
the organ affected.

      EPA agrees with the commenters that the best way to evaluate this
adjustment factor  is on a case-by-case basis since not all carcinogens  behave
mechanistically the same way  (as pointed out by the 1986 Guidelines for
Carcinogen Risk Assessment).  For some carcinogens, exposure early on (during
childhood or adolescence, or the equivalent in test animals) may be critical
to tumor expression.  For others, exposure of a test animal over a year may
result in no more  tumors than exposure over a lifetime  (two years).

      (C).  Final Guidance:  To allow flexibility in making these judgments,
EPA has changed  the final Guidance to allow for less than lifetime adjustments
on a case-by-case basis.  Less than lifetime adjustments to the  cancer  slope
factor are now optipnal, not required, and should be made on the basis  of
existing mechanistic data.  However, in the absence of data on mechanisms and
time to tumor, States and Tribes may continue  to use the adjustment factor and
the duration cutoffs prescribed  in the proposal as a default approach.

      iii.  Species Scaling Factor.

      (A).  Proposal:  In the proposed Guidance,  EPA would have required the
use of a "surface  area species scaling factor" in deriving a dose  response
relationship for humans that  is based on animal data. The proposed Guidance
assumed that exposures in milligrams per kilogram of body weight per  day
raised to the  2/3  power  (also referred to as the two-thirds exponent) would
yield equivalent cancer responses in test animals and humans.   EPA requested
comment on  the proposed use of a  two-thirds  exponent for  surface area,  and the
possible use of  a  three-fourths  exponent for body weight, for  scaling animal
doses to equivalent human doses.  EPA also  asked for comment on whether use  of
some other  scaling factor  should be used in the final Guidance.

       (B).  Comments:  Several  commenters were in favor of  the  3/4 body weight
scaling factor while  others  were in favor  of  the 2/3  surface  area scaling
factor.  Those  in  favor of the  3/4 body weight scaling  factor  cited the draft
consensus by the Inter-Agency group  (U.S. EPA,  1992) and  empirical  data which
supports the scaling  on the  basis of body weight  (Allen et  al.,  1987) .   Others
suggested flexibility and  allowing adjustments on a case-by-case basis,
determined  by  the  pharmacokinetics of the chemical.

      As cited in  the proposed  Guidance, an Inter-Agency  group comprised of
EPA, U.S. Food and Drug Administration  and  CPSC has  deliberated on the  issue
of  appropriate,  consistent scaling factors  for use  by  all agencies in
developing  risk assessments.  While  the  draft  recommendation from this group
was  to  apply a 3/4 exponent  scaling  factor,  this  undertaking has not been

-------
                              Section V: Human Health                           155
finalized nor adopted as EPA policy to date.  The current policy is stated in
the 1986 Guidelines for Carcinogen Risk Assessment: In the absence of
comparative toxicological, physiological, metabolic and pharmacokinetics data
for a given suspect carcinogen, EPA takes the position that extrapolation on
the basis of surface area is considered to be appropriate because certain
pharmacological effects commonly scale according to surface area  (Dedrick
1973, Freireich, et al., 1966; Pinkel;, 1958).  As discussed in the preamble
to the proposal, there are divergent views among the scientific community
regarding whether a body weight or surface area scaling factor provides the
best means of extrapolating results from animal studies to humans, and as
reflected by the work of the Inter-Agency Pharmacokinetics Group, a consensus
among relevant governmental agencies may be reached in the future that the
body weight approach is preferable.  However, no final consensus has yet been
reached.  Because EPA's current, longstanding policy calls for use of the
surface area scaling factor, EPA is hesitant to change its approach in the
absence of a final determination that adoption of the body weight scaling
factor is appropriate.

      (C).  Final Guidance:  Consistent with the existing EPA policy, the
final Guidance calls for the use of the 2/3 exponent scaling factor  (surface
area) , in the development of cancer slope factor development.

      b.    APE.

      For non-carcinogens, the proposed Guidance called for a data hierarchy
for calculating the ADE.  The process outlined is the same one used by EPA's
RfD development process but differs in the amount of data required to develop
a criterion or value, especially with regard to Tier II ADEs.  In some cases,
an ADE may be identical to an EPA RfD if the same data and judgments are used.
However, since States and Tribes may derive values which may vary based on
different interpretations of the supporting data and/or the existence of new
data, EPA proposed the term ADE to distinguish the RfD from the ADE.

      The proposed Guidance indicated that calculating an ADE for a chemical
involves the following steps: 1) determining whether there is evidence from
epidemiologic or animal studies that exposure to a chemical may result in
adverse noncancer health effects; 2)  using available data to determine a
threshold dose value that is likely to be without appreciable risk of adverse
effect;  and 3)  reducing this threshold dose value to account for
uncertainties inherent in the risk assessment to yield an ADE for humans.

i.    List of Deleterious Effects.

      (A).  Proposal:  The proposed Guidance presented a list of adverse
effects which the noncancer criteria are protective of, including adverse
acute, subchronic and chronic effects and reproductive and developmental
effects.  EPA solicited comment on whether it should specify in the
methodology a longer list of deleterious effects against which noncancer
criteria should protect.

      (B).  Comments:  Several commenters agreed with the list of deleterious
effects while others requested that EPA expand the list to include such
effects as immunotoxicity.  Other commenters believed a list was not entirely
necessary since the assessment of adverse effects was mostly based on
professional judgment.

      (C).  Final Guidance:  EPA believes the list as stated encompasses any
effect which can be deemed adverse including immunotoxicity, based on the
State or Tribe's professional judgment.  Thus, EPA is retaining the proposed
list in the final Guidance.

-------
156    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

ii.   Uncertainty Factors.

       (A).  Proposal:  The proposal provided guidance on the use of
uncertainty factors to account for uncertainties in predicting acceptable
exposure levels for the general human population.  The proposal  indicated  that
the size of the uncertainty factor varies depending on the data  available  for
ADE calculation, including whether the data are from a study on  humans  or  test
animals, and on whether the study demonstrates a no observed adverse effect
level  (NOAEL) or a lowest observed adverse effect level  (LOAEL).  Under the
proposed Guidance a composite uncertainty factor of 30,000 was described as
the maximum uncertainty allowed when deriving a Tier I criterion or Tier II
value.  EPA solicited comments on the uncertainty factors included in the
proposal, and particularly on whether other uncertainty factors might offer
better assessment of risk, and be more appropriate in deriving water quality
values and criteria.

       (B).  Comments:  Many commenters stated that the use of default
uncertainty factors of 10  (each) to account for intra-, interspecies
variability, and subchronic to chronic adjustments are too rigidly applied.
Commenters felt uncertainty factors of less than 10 should be allowed if there
is a scientifically sound rationale.  Other commenters suggested that EPA  add
an additional uncertainty factor to the calculation of an ADE to account for
the sensitivity of children or other sensitive subgroups.

      With regard to the use of uncertainty factors less than 10, EPA agrees
that uncertainty factors of less than 10 can be used to account  for intra-,
interspecies variability, subchronic to chronic adjustments and  LOAEL to NOAEL
adjustments as long as data exists to justify a lower uncertainty factor.  As
stated  in the proposed Guidance, EPA generally applies uncertainty factors of
10, but is not advocating the rigid application of uncertainty factors  of  ten
in all  instances.  This is only meant to be a default position when data is
lacking to justify a lower uncertainty factor.  For example, if
pharmacokinetics data shows that a test animal metabolizes a chemical in a
fashion identical to humans, a lower  interspecies uncertainty factor is
probably justified.  For information on selecting uncertainty factors less
than 10, see appendix A of the Human Health TSD.

      With regard to the comment suggesting EPA apply an additional
uncertainty factor to protect children and sensitive subgroups,  EPA believes
an RfD  (ADE) should,be protective  (by definition) of children and sensitive
subgroups.  In assessing the data base, consideration should be  made of these
segments of the population.  If the overall data base does not yield
information on the toxicity of test organisms during all segments of their
lifetime  (such as infancy, childhood, adolescence), or does not  consider
sensitive individuals,  an additional uncertainty factor may be applied  to
account for such uncertainty in an incomplete data base.  It is  important  to
note that the intraspecies uncertainty factor of 10, which is commonly
applied, is designed to account for sensitive individuals in the population.
Therefore,  the application of another uncertainty factor would be redundant.

       (C).  Final Guidance:  As stated above, the proposed Guidance  stated
that a maximum uncertainty factor of  30,000  could be applied in  deriving a
Tier I  criterion or  Tier  II value.  In the  final Guidance, EPA is clarifying
the use of  uncertainty  factors  to include a maximum of 10,000 for Tier  I
criteria and a maximum  of  30,000  for  Tier II value development.  When deriving
a Tier I criterion,  the likely maximum composite uncertainty factor  applied to
a 90-day NOAEL may be 3000.  The  total of 3000  is based  on four  separate
uncertainty factors: 10 to account for intraspecies variability; a factor  of
10 to  account for the uncertainty in  extrapolating animal data to the  case of
humans;  a factor of  10  to account for subchronic to chronic variability;  and a
factor of three  to  account for  an incomplete data base.  However,  in rare
cases  where an  extra uncertainty  factor  is  required' to compensate for an
incomplete  data  set  or  to compensate  for  the uncertainty associated  with  a
LOAEL  (rather than  a NOAEL), a  composite  uncertainty  factor  of  10,000 may be

-------
                              Section V: Human Health                           157
used.  Under this assumption an uncertainty factor of 10 would be applied to
account for an incomplete data base or the use of a LOAEL in place of a NOAEL
instead of using an uncertainty factor of three.  This decision requires the
use of professional judgment in determining the adequacy of the data base.
For Tier II values, the maximum allowable composite uncertainty factor may be
30,000 which would be applied to a studies with durations greater than a 28-
day minimal LOAEL  (e.g., 30 days) .

      The total of 30,000 is based on the three factors of generally 10 each
for intraspecies variability, interspecies extrapolation, subchronic to
chronic extrapolation, a combined factor of 10 to account for incompleteness
of data bases and the difference between a minimal LOAEL and NOAEL (which
together result in an overall uncertainty factor of 10,000), and an additional
uncertainty factor of 3 to account for the uncertainty in extrapolating from a
study greater than 28 days, but sufficiently less than 90 days.
      i .    Proposal :  The proposed Guidance indicated that EPA has a process
to develop consensus on cancer slope factors (ql*s) and RfDs  (referred to as
ADEs for the Great Lakes) .  These values are derived by two EPA work groups
called the RfD/RfC and CRAVE work groups and made available as guidance to EPA
program offices and' the public via a data base called the Integrated Risk
Information System  (IRIS) which is accessible through the National Library of
Medicine's Hazardous Substance Data Base (HSDB) .  The values  (RfDs and ql*s)
listed on IRIS are guidance; they are not regulatory in nature.

      The proposed Guidance recommended that verified IRIS values  (RfDs and
cancer slope factors) be considered as a first step in deriving the Great
Lakes Human Health criteria.  It is also important to note that many chemicals
have been reviewed by the RfD or CRAVE Work Groups, have been verified, but
have not yet been entered  into the IRIS system.  A verified RfD or ql* is a
value which the RfD. or CRAVE Work group has reviewed and officially verified
during an RfD or CRAVE meeting.  EPA encourages States to call EPA's RfD and
CRAVE Work Groups if they are not sure of the current status of a specific
chemical.  A verified RfD or ql* can be utilized by States and Tribes  even
though it has not been officially placed into the IRIS system.  The proposal
also allowed deviation from these values when new data are available or if a
different interpretation of the data is made .  EPA requested comments on
deviating from IRIS values in deriving Great Lakes criteria and values for the
reasons highlighted above.

      ii.   Comments :  Many commenters wanted the flexibility to deviate from
IRIS while others argued that deviation from IRIS values should not be
allowed.

      EPA has decided that deviations from IRIS should be allowed since IRIS
is only guidance and that other interpretations of the data may be valid.
However, to foster consistency between EPA and the States, EPA strongly urges
the States/Tribes to communicate any anticipated or necessary deviation from
IRIS to the EPA RfD and/or the CRAVE work groups as soon as possible .
Following this recommended course of action will allow EPA to discuss the
potential deviations with the State or Tribe and could lead to an expedited
review of the chemical and data by the EPA work group of concern.  In
addition, as noted earlier, States may use verified RfDs or cancer potency
factors, which have not been entered onto IRIS, for development of Tier I
criteria.  Finally, when deviating from IRIS, States/Tribes are encouraged to
work with the Clearinghouse described in section II, to ensure other
States/Tribes are aware of the deviations .

      iii.  Final Guidance :  For the reasons stated above, States and Tribes
may deviate from IRIS- provided the approach is scientifically defensible.

-------
158    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


S.     Exposure Assumptions

      The proposed Guidance identified seven factors which affect an
individual's oral exposure to a chemical: body weight; duration  of exposure;
recreational exposure; drinking water consumption; fish consumption;
bioaccumulation factor and relative source contribution.

      a.    Body Weight.

      i.    Proposal:  The proposed criteria methodology assumed a mean adult
human body weight of 70 kg.  This value is consistent with that  recommended in
the EPA's Exposure Factors Handbook (EPA 600/8-89/043, July 1989).  EPA
requested comments on the use of the  70 kg body weight assumption and also
asked for comments on the issue of using body weights of sensitive
subpopulations  (such as children) when a chemical's toxicity indicates  a
specific subpopulation is most sensitive to exposures.

      ii.   Comments:  Several commenters advocated the continued use of  the
70 kg body weight assumption.  Several other commenters stated that the body
weight used in calculating criteria and values should be lower than 70  kg to
protect women of childbearing age and children and fetuses.  Several of these
commenters recommended 55 kg as the body weight.  No commenters  suggested an
actual child weight, which should be used nor did any commenter advocate a
higher body weight than 70 kg.

      EPA believes 70 kg is an appropriate body weight because it represents a
reasonable measurement for the entire population.  As stated in  the EPA
Exposure Factors Handbook  (1989), the mean body weights for women and men
nationally are 65 kg and 78 kg, respectively, with an overall mean adult  body
weight of 71.8 kg.  The Handbook also indicates that the mean body weight for
child bearing women ages 18-45 is 63  kg.)  If a State believes that use of a
lower body weight is appropriate  (which yields a more stringent  criterion),
the State or Tribe may adopt such an  assumption in calculating their criteria
and values under their authority to establish more stringent requirements
pursuant to section 510 of the Act.

      As to whether lower body weights should be used to protect women  of
childbearing age, children and fetuses, EPA believes that categorically
adopting more conservative body weight assumptions may not be appropriate.
Each chemical must be addressed  separately since some chemicals  may be
generically toxic to both  sexes, while others may be  specifically toxic to one
sex more than the other.   It therefore would not be appropriate  to require
generally that all criteria be based  on conservative body weight assumptions.
In the  case of mercury, however, a fetotoxic chemical, to be protective of
women of child bearing age, EPA has assumed a body weight of 65  kg  (as  opposed
to 70 kg) which  results in a Tier I mercury criterion of 1.8 ng/L, which  is
slightly less than the proposed  criterion of 2 ng/L.  EPA has set a. final Tier
I criterion for  mercury at 1.8 ng/L.

      iii.  Final Guidance:  Based on the reasons above, the final Guidance
retains the 70  kg body weight assumption.   States or Tribes may  use the female
body weight  (55-65 kg) or  a  child's body weight  (10-30 kg) or any other more
 "stringent" assumption on  a  chemical-by-chemical basis as deemed appropriate
based on the properties of a particular chemical.  If a  child's  body weight  is
assumed, the water  consumption  rate  should  also be adjusted for  a child to  l
liter/day.

      b.    Duration  of Exposure.

       i.    Proposal:  The proposed  Guidance assumed  that oral exposure
 remains constant for  a  lifetime  and  assumed a  70 year exposure period in
 developing  criteria  and values.   EPA requested comments  on  the use  of  longer
 lifetime exposure periods,  such as 75 years instead of  the  currently proposed
 70 years,  since recent  census  data indicates that  the average  lifespan for

-------
                              Section V: Human Health                           159
Americans is 74.7 years.  EPA also requested comments on whether the use of
shorter exposure periods  (i.e., less than 70 years) would be more appropriate
to account for mobility of individuals in and out of the Great Lakes basin.

      ii.   Comments:   Several commenters suggested that EPA increase the
lifespan to 75 years or longer to account for women and people who live past
70 years.  Other commenters suggested  retaining the proposed 70 year value.
Other commenters cited the EPA Exposure Factors handbook which identifies nine
and 30 years as representative of the average and reasonable upper bound,
respectively, of length of time residing in the same house.  Commenters also
suggested that mobility in the area is high and that individuals in the Great
Lakes will not be exposed.to contaminants uniformly over their lifetimes.

      EPA believes the 70 year lifespan assumption is appropriate because it
conservatively assumes a uniform exposure to contaminants over a lifetime.
Since consumption patterns may change from infancy to death, EPA is seeking to
protect those individuals who may live their entire lives in the Region and
those who may encounter similar exposures while outside the region.  Without
exhaustive analysis, it would be difficult to determine that people who
migrate in and out of the region are actually experiencing different exposure
patterns.  With regard to increasing the lifetime exposure to 75 years or
more, EPA believes 70 years is an appropriate assumption as it is still a
generally accepted value used by most risk assessors and regulatory agencies.

      iii.  Final Guidance:  For the reasons cited above, the final Guidance
retains the 70-year exposure period for developing criteria and values.  A
State or Tribe may chose to be more stringent and adopt a longer exposure
period in its criteria derivation.  It should be noted that this assumption
does not influence cancer or noncancer criteria and values which are based on
animal data since all lifetime animal data is assumed equivalent to a human
lifetime, but may have a bearing on criteria and values based on epidemiology
studies, depending on other numerical considerations in the calculation.  If
an epidemiology study is less than lifetime in duration, the dose must be
translated into a lifetime human dose based on a specific lifetime duration
assumption, either numerically or through the use of an uncertainty factor.
In such a case,  the assumption regarding the duration of a lifetime would be
relevant to determining the human health criterion.  In fact, it is relatively
rare that epidemiological data is used to derive human health criteria due to
limitations in the study which preclude a conclusion of a cause and effect.
In the final Guidance there are two cases where epidemiological data were
used; in deriving the HCV for benzene and the human noncancer value (HNV) for
mercury.

      c.    Incidental Exposure.

      i.    Proposal:   The proposed Guidance included an incidental ingestion
exposure factor which accounted for oral exposures which might occur through
recreational activities in or on the water.  The factor  (0.01 liters/day) was
included in the derivation of the GLWQI criteria for those surface waters that
are not used as a drinking water source (Nondrinking HNVs and HCVs).  This
factor was not used in those cases where the waterbody was designated as a
drinking water source.  EPA requested comments on whether the factor was
justified and also requested comments on whether a factor should be included
for incidental dermal exposure which occurs through recreational activities.
EPA also requested submission of any data that could be used to derive such a
factor.

      ii.   Comments:   EPA received many comments stating that the extra
factor for incidental exposure was unnecessary since the drinking water
assumption of 2 L/day appears overly conservative.  EPA also received comment
that attempting to account for dermal uptake would be very difficult,  if not
impossible.  No data were received on dermal uptake.

-------
160    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      EPA agrees that the extra factor for incidental exposure is not
necessary to include in the calculation of criteria and values which are to be
used in protecting waters designated as drinking water supplies  because such
an exposure assumption is of negligible significance when viewed in light of
the 2 L/day assumption used in deriving criteria for these waters.
(Numerically, the 2.0 liter value in the numerator is so large in comparison
to the 0.01 liter value,  that it has no numerical effect on the  calculated
criterion.)  However, EPA continues to believe that the inclusion of the
incidental exposure factor is appropriate in developing criteria and values
for waters designated as non-drinking water  (i.e., recreational  only).

      The effect of using the incidental exposure factor when deriving
criteria and values for nonbioaccumulative chemicals can be numerically
significant.  For example, for nondrinking water sources, when the incidental
exposure factor is included, the HNV for methylene chloride is 90 mg/L.  When
the incidental exposure factor is not included, the HNV is 120 mg/L.

      In addition, EPA also believes it is appropriate to account for
incidental ingestion in a recreational setting since on an individual or
subpopulation basis this exposure may be significant.  For example, those
people who spend more than an average amount of time recreating  in water may
exceed the 0.01 liters/day incidental ingestion rate.   (The 0.01 value  is
based on a number of average assumptions for the Great Lakes basin population
as a whole.)  If States or Tribes want to establish a larger incidental
exposure factor for recreational  subpopulations  (swimmers, waterskiers,
rafters, kayakers), they may do so.  This factor may become especially
important in the future,  when criteria for microbiological agents are
developed, since very minute amounts of microbially contaminated water  can
cause adverse health effects.

      iii.  Final Guidance:  The  final Guidance retains the incidental
exposure factor of 0.01 liters/day for those surface waters that are
designated as nondrinking water sources.

      d.    Drinking Water Consumption.

      i.    Proposal:  EPA proposed that the criteria and values be derived
using an assumption,of 2 L/day drinking water consumption from untreated
surface waters for waters designated as drinking water sources.  This
represents approximately the 90th percentile ingestion value for the  general
population.  The Exposure Factors Handbook,  1989, states:  For the reasonable
worst-case value, the 90th percentile rate  reported by Gillies and Paulin
 (1983), 1.90 L/day, suggests that a rate of  2.0 L/day may be a reasonable
approximation.

      The  90th percentile value suggested by Cantor et al.  (1987) is  also
approximately 2  L/day.  This is also  supported by a more recent  study
specifically characterizing  tap water intake by Ershow and Cantor  (1989) in
which 2 L/day represents approximately the  85th percentile value of drinking
water consumption.  This value, 2 L/day, is  recommended as the reasonable
worst-case  consumption rate.  EPA requested comment on  whether  selection  of
another value such as 1.4  liters  per  day, the national average,  would be more
appropriate.  EPA also noted that, since the 2.0  liters value  is a
conservative assumption  (only 10  percent of  the population drinks two liters
of water a day and considerably less  can be  expected to drink  two liters of
untreated  surface water) and the  0.01 liter associated with incidental
exposure is  relatively insignificant  in comparison, EPA presumed that 2 L/day
 is protective of both drinking water  and incidental ingestion  exposures for
waters  which may be^both a  drinking water source  and used for  recreation.  EPA
requested  comments on whether such an assumption  was justified.  EPA  also
requested  comments on whether surface water criteria for waters  designated for
drinking water uses  should assume consumption of  untreated water, as  was
proposed.

-------
                              Section V: Human Health                           161
      ii.   Comments:  Several commenters suggested 2 L/day was too
conservative.  Others stated that the level of conservatism was appropriate.
Those that felt it was overconservative believed that the national average is
1.4 liters/day and that people do not generally drink untreated surface water
(in any amount).  No commenters suggested that 2 L/day was underprotective.
Other commenters stated that some cultural activities conducted by Native
Americans, involve the direct consumption of untreated surface water, although
the specific nature of these practices was not described.

      EPA believes that the consumption rate of 2 L/day, even though
conservative, is reasonable given that it will provide a high level of
protection for the people in the Great Lakes consistent with the protective
goals of the Act.  EPA, in determining a water consumption rate, considered
adopting the average consumption rate (1.4 liters/day), as discussed in the
preamble to the proposal and suggested by commenters.  However, in choosing
among the available'policy options, EPA chose 2 L/day to provide an extra
degree of protection for those individuals (which may be a substantial number
of persons) who consume more than the population on average.  EPA believes
that this assumption is reasonable in light of the public health protection
goals of the CWA.  EPA also believes that it would not be appropriate to
dismiss drinking water usages, because as asserted by some commenters, there
are expressed cultural practices which include consumption of untreated
drinking water.  EPA also believes that it would not be appropriate to assume
that untreated water is never consumed by users since the CWA requires
protection of designated uses.  To protect waters designated as drinking water
sources,  it is impprtant that public drinking water systems not be the only
point of clean up and that surface water discharges also share the burden of
maintaining the use designation of the water.

      iii.  Final Guidance:  The final Guidance retains the assumption that
individuals consume 2 L/day of untreated surface waters when developing
criteria/values for waters designated as drinking water sources.

      e.    Fish Consumption.

      i.    Proposal:  The proposed Guidance included a fish consumption rate
of 15 grams per day-,  which represents the mean exposure level for regionally
caught fish for the regional sportfishing population.  This value also
approximates the 90th percentile for the entire regional population.  Thus, a
more conservative target population was chosen than is used in the National
criteria methodology and the proposed fish consumption value was based on
Great Lakes-specific statistical data.  The actual value of 15 grams per day
was derived from review of several regional studies in Michigan (West, et al.,
1989),  Wisconsin (Fiore et al.,  1989) and New York (Connelly, et al., 1990).
Since the proposal, several new pieces of information  have emerged that were
used in EPA's consideration of the appropriate fish consumption rate: First,
an Executive Order on environmental justice was issued February 11, 1994.  It
requires Federal agencies (to the greatest extent practicable and permitted by
law)  to "identify and address, as appropriate, disproportionately high and
adverse human health or environmental effects of its programs, policies, and
activities on minority population and low income populations."  Federal
agencies are also required,  whenever practical and appropriate to "collect,
maintain, and analyze information on the consumption patterns of populations
who principally rely on fish and/or wildlife for subsistence."

      Second, a study conducted by West et al. (1993) for the State of
Michigan was provided by the author to EPA during the public comment period.
This study is a full year (February 1991 to February 1992)  fish consumption
survey of 7000 licensed Michigan anglers.  The survey found that the average
sport fish consumption rate, adjusted for non-response bias, was 14.5
grams/day.  The average total fish (all fish, not just Great Lakes sport fish)
consumption rate, adjusted for non-response bias, was 24.4 grams/day.  This
study indicated that fish consumption rates may differ according to race and
income  level.  The lowest income group (< $i4,999/year)  averaged 21 grams/day

-------
162    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


sport fish consumption as compared to 14.7 grams/day for those making  $40,000
or more/year.  The average sport fish consumption rate for minorities  was  23.2
grams/day as compared with 16.3 grams/day for non-minority individuals.  Lower
income ($24,999 or less) minorities averaged the highest consumption rate  of
all groups in the survey: 43.1 grams/day sport caught fish and 57.9 grams/day
total fish;  Non-minority individuals of lower income averaged 18.6 grams/day
sport fish and. 25.8 grams/day total fish.  The study also indicated that
minorities eat less.fish from the Great Lakes and more fish from the inland
tributaries than non-minority individuals.

      In the proposal, EPA also requested comments how to protect against
acute and subchronic effects, the possibility of adverse effects due to binge
eating.  Specifically, EPA invited comment on the use of 454 grams  (one pound)
as a reasonable worst-case, one-day fish consumption estimate and 2,240 grams
as a reasonable worst case, 10-day fish consumption estimate based on  10
consecutive days of consumption of approximately one-half pounds of fish per
day.  These values could be used in deriving one-day and 10-day
criteria/values protective of acute and subchronic effects.  In a Notice of
Availability dated August 30, 1994  (U.S. EPA, 1994a)  EPA asked for comments
on the fish consumption rates identified in the West  et al.  (1993) study  and
on the quality of the study itself.

      ii.   Comments:  Many commenters believed EPA's proposed rate of 15
grams/day would not be protective of recreational and subsistence anglers  such
as the Native American anglers and minority anglers, or women of childbearing
age and children.  An alternative rate of 50 grams/day was suggested.   Others
believed the rate of 15 grams/day was too conservative since it is protective
of at least 90 percent of the overall population and these commenters
suggested a lower rate.  With regard to the use of one-day and ten-day worst
case rates, comments suggested that this is a worthwhile attempt at dealing
with a potential short term risk, but it was not clear how this could  be
implemented in a water quality standards context.  Other commenters stated
that one or ten day exposures are not long enough for biouptake and
distribution to reach a steady state of chemical concentration in a human  high
enough to result in a toxic effect.

      With regard to comments on the August 30, 1994 Notice of Data
Availability  (U.S. EPA, 1994a), many commenters were in favor of retaining the
proposed fish consumption rate of 15 grams/day, citing the relative
conservatism of the-entire criterion methodology.  Many of these same
commenters questioned the validity of the West et al.  (1993) study findings.
Specifically, they questioned the lower income minority average of 43.1
grams/day since this number was based on a very small sample size and  was
associated with a large standard deviation.  These same commenters
acknowledged that there are subpopulations who consume more than 15 grams/day
and that the best way to protect these people is to set site-specific  fish
consumption rates.  Several other commenters were in favor of a higher fish
consumption rate ranging from 30-60 grams/day.  These  same commenters
supported the findings of the West et al.  (1993) study, claiming that  the
study does show that lower income and minority subpopulations eat more sport
caught fish on average, but also pointed out that while the data was presented
correctly, many of the opinions presented in the study appeared overly
subjective.

      One group of commenters suggested that EPA establish in the final
Guidance a new approach to establishing fish consumption rates based on
default values derived  from national data.  These commenters argued that
regional fish consumption  survey data are time-consuming and costly to
collect, and  consequently are not often available.  The commenters recommended
that States and Tribes  instead utilize default fish consumption  rates  based
upon the particular-fish consumption use of the waterbody  (i.e., 20 grams/day
for average fishing, 45-140 grams/day for sport fishing and 90-165 grams/day
for subsistence  fishing, with special consideration for Native American
subsistence uses).  These  commenters  also argued that use of the proposed fish

-------
                              Section V: Human Health                           163
consumption rate would not be in accordance with Executive Order 12898
addressing Environmental Justice, and would be discriminatory under the Civil
Rights laws.

      After consideration of the public comments, EPA has decided to retain
the 15 grams/day fish consumption rate.  EPA does not agree with commenters
who argued that this assumption would not provide adequate protection of '
public health.  As stated in the proposal, this value approximates the 90th
percentile for the overall population in the Great Lakes basin and the mean
value for sport anglers in the basin.  EPA believes the results from the West
et al. (1993) study  (a mean consumption rate of 14.5 grams/day for Michigan
sport anglers), along with the studies cited in the proposal  (see 58 Fed. Reg.
20870)  supports the reasonableness of the fish consumption rate used in the
final Guidance.

      EPA also believes that attainment of criteria in ambient waters that are
derived using the assumed fish consumption rate will result in adequate
protection for highly exposed populations.  In carrying out regulatory actions
under its statutory authorities, including the CWA, EPA generally views an
upper bound incremental cancer risk in the range of 10"4 to 10"* as adequately
protective of public health.  As discussed above, the human health criteria
methodology is based on a target risk level of 10"5.  If fish are contaminated
at the level permitted by criteria derived under the final Guidance,
individuals eating up to ten times the assumed fish consumption rate  (i.e.,
150 grams/day) would still be protected at a 10"4 incremental risk level,
within the range that EPA believes is protective of public health.  Thus,
while EPA acknowledges that some portion of the population will consume
greater than 15 grams/day, the fish consumption rate and the target risk level
chosen in the final Guidance combine to ensure that the population as a whole
will be adequately protected by human health criteria when the criteria are
met in the ambient water.

      Available data support the protectiveness of this approach.  An analysis
of the West study indicates that approximately 99 percent of all persons
surveyed in that study consumed less than 150 grams/day and that, even among
low-income minorities who as a group consume more fish than the population on
average,  approximately 95 percent consumed less than this amount.  Thus, EPA
concludes that full implementation of the final Guidance will result in a high
degree of protection for the population as a whole, including those
populations that consume greater amounts of fish than the population on
average.

      The final Guidance requires, moreover, that States and Tribes adopt an
additional degree of protection where it is determined to be appropriate for
highly exposed populations.  Section A.4.a. of procedure 1 in appendix F to
part 132 states that "Human health criteria or values shall be modified on a
site-specific basis to provide additional protection appropriate for highly
exposed populations."  Thus, if a State or Tribe determines that a highly
exposed subpopulation (e.g., a Tribe of subsistence fishers) would not be
adequately protected by criteria or values derived using the final Guidance
methodology, the State or Tribe would be required to adopt a more stringent
site-specific modification to criteria to provide appropriate additional
protection (e.g., to ensure protection at least at the 10"4 risk level, or
other more stringent degree of protection determined by the State or Tribe to
be appropriate).  Finally, the final Guidance requires States and Tribes to
adopt provisions to ensure that human health is protected from the adverse
effects of mixtures of pollutants in the Great Lakes System, including
mixtures of carcinogens.  See procedure 4 of appendix F to part 132; section
VIII.D of this document.  Thus, taking into account all relevant portions of
the final Guidance - - e.g., the target risk level, site-specific modification
procedures,  additivity provisions, as well as the many conservative
assumptions in the human health methodology discussed elsewhere in this
document --  EPA believes that waterbodies meeting criteria developed in

-------
164    Water Quality Guidance for the Great Lakes System -- Supplementary Information Document


accordance with the final Guidance will be protective of all segments of the
public, including any more highly exposed populations.

      EPA acknowledges the problem identified by commenters that high quality
survey data describing fish consumption patterns are often not available.  EPA
believes, however, that States and Tribes retain substantial flexibility in
determining how to address the fish consumption issue should the State  or
Tribe decide that a more stringent approach than EPA's is appropriate.  Where
there is an absence of survey data for an area, but evidence of uses that the
State or Tribe believe should be specifically reflected in a more stringent
fish consumption rate, the State and Tribe could reasonably decide to adopt
"default" values to provide a level of protection determined by the State or
Tribe to be appropriate for the particular fishing use made of the waterbody.
(EPA does not express a view as to the appropriateness of the specific  default
values suggested by commenters and noted above; such a determination can
appropriately be made by the State or Tribe based on its best professional
judgment.).

      Finally, EPA believes that its action here is fully consistent with the
Executive Order on Environmental Justice and civil rights laws because  the
final Guidance ensures that waterbodies meeting criteria developed using
methodologies consistent with the final Guidance will protect the health of
all of the public, including more highly exposed populations.  Section  1-101
of Executive Order 12898 provides that "[t]o the greatest extent practicable
and permitted by law  . . . each federal agency shall make achieving
environmental justice part of its mission by identifying and addressing, as
appropriate, disproportionately high and adverse human health or environmental
effects of its programs, policies, and activities on minority populations and
low-income populations . . . ."  Section 2-2 directs federal agencies to
conduct their activities in a manner that ensures that the "activities  do not
have the effect of ... subjecting persons  (including populations) to
discrimination under such programs, policies and activities because of  their
race, color, or national origin."  Finally, section 4-4 of the Order
specifically discusses subsistence consumption of fish and wildlife, and
provides that "[f]ederal agencies, whenever practicable and appropriate, shall
collect, maintain and analyze information on the consumption patterns of
populations who principally rely on fish and/or wildlife for subsistence," and
that "[f]ederal agencies shall communicate to the public the risks of those
consumption patterns."

      EPA's action here is consistent with the terms of this Executive  Order.
In developing the final Guidance, the EPA made a special effort to identify
and address the potential risks to highly exposed low-income and minority
populations, to maintain and analyze information on these risks, and to
communicate risks to the public, as reflected in the August 30, 1994, Notice
of Data Availability soliciting public comment on the fish consumption  issue
and the provisions in the final Guidance specifically targeted to protecting
highly exposed populations discussed above.  EPA also flatly rejects the
notion, advanced by-some commenters, that the final Guidance is in any  respect
discriminatory against persons or populations because of their race, color, or
national origin.  The final Guidance establishes criteria methodologies and
implementation procedures that are expressly designed to ensure full
protection of the public, including highly exposed populations.  While  some
groups and individuals, including some low income and minority persons  and
populations, may face a greater risk of adverse health effects than the
general population due to their particular fish consumption patterns, EPA has
included in the final Guidance mechanisms that, upon full implementation by
States and Tribes, will ensure that these groups will nonetheless receive a
level of public health protection within the range that the EPA has long
considered to be appropriate in its environmental programs  (i.e., 10"4 to 10"6
incremental cancer risk).  Obviously, as long as there is variability in fish
consumption patterns  among various segments of the population, it would be
impossible for EPA to ensure that all groups would face identical risk  from
consuming fish.  Therefore, EPA has sought to ensure that, after attainment of

-------
                              Section V: Human Health                           165
water quality criteria in ambient waters, no group is subject to increased
cancer risks greater than the risk range that the EPA has long considered
protective.

      As noted above, EPA solicited comment on whether EPA should develop
short-term criteria to protect against possible effects of worst-case short
term exposures.  No commenter supported the development of such criteria.
Commenters generally did not understand how such short-term criteria could be
implemented in the water quality criterion program as now constituted.
Commenters did generally support further research into the unanswered
questions associated with high, short-term exposure effects, including
fetotoxic effects.  Several commenters felt that fish advisories were the best
way to deal with this issue.

      EPA agrees with the commenters that this is an area that needs more
research before short-term water quality standards could be set.  Research in
the area of bio-uptake and whether a steady state can be achieved in a short
time period is needed before EPA could determine whether to develop short-term
criteria and values that would be sufficiently supported by the available
science to serve as sources of regulatory controls.  EPA acknowledges that
fish intake rates d& vary over the course of a lifetime and while EPA believes
the 15 grams/day assumption is adequately protective of the Great Lakes
population over a lifetime, States have the flexibility to establish criteria
specifically targeted to provide additional protection to sensitive
subpopulations (e.g., pregnant/nursing women, infants, children) or highly
exposed subpopulation (e.g.. Native Americans) using adjusted values for
exposure parameters for fish consumption, body weight, and duration of
exposure.  EPA believes the values (short-term, worst-case fish consumption
exposure assumptions of 448 grams as a worst-case, one-day fish consumption
estimate and 2,240 grams as a worst-case, 10-day fish consumption estimate)
provided may be used in setting fish advisories, especially for those
chemicals for which acute noncancer effects are most notable.

      iii.  Final Guidance: For the reasons cited above, the final Guidance
uses a fish consumption rate of 15 grams per day when deriving human health
criteria.  The 15 grams per day is divided into the grams of trophic level 3
fish consumed (3.6 grams) and the grams of trophic level 4 fish consumed  (11.4
grams).  The grams of fish consumed at each trophic level were estimated using
the data in the West et al. (1993) survey.  The results of this analysis are
discussed in section IV.B.3 of this document.  EPA believes separating the
quantity of trophic level 3 and trophic level 4 fish consumed better
represents the potential exposure to consumers.

      f.     BAFs.

      The proposal included a methodology for deriving BAFs, and technical
support documents describing the methodology and BAF derivation for those
chemicals for which human health and wildlife criteria are proposed.  The BAF
methodology in the final Guidance has been modified from the proposal.
However,  BAFs are still required to be utilized in the derivation of human
health criteria.

      The new minimum BAF data required to derive Tier 1 human health criteria
for organic chemicals include either: (a) a field-measured BAF; (b) a BAF
derived from the BSAF methodology, or (c) a chemical with a BAF less than 125
regardless of how the BAF was derived.  For all inorganic chemicals, including
organometals such as mercury,  the minimum BAF data required to derive a Tier I
human health and wildlife criteria include either: (a) a field-measured BAF or
(b) a laboratory-measured BCF.  For the majority of inorganic chemicals, the
BAF is equal to the BCF  (i.e., FCM = 1)  because there is no apparent
biomagnification or metabolism.  The basis for these new requirements are
discussed in section IV of this document.

-------
166    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      g.    Relative Source Contribution.

      i.    Proposal:   In the proposed Guidance, EPA assumed an  80 percent
relative source contribution (RSC) from surface water pathways  (water and
fish) for BCCs, and 100 percent RSC for non-BCCs, in deriving noncancer
criteria/values.  A 100 percent RSC was assumed for all chemicals in deriving
cancer criteria/values.  The proposal explained the reasoning of the
Initiative Committees for choosing each RSC default percentage  (see 58 FR
20870).

      EPA requested comments on the proposed RSCs and on possible alternatives
for derivation of noncancer criteria and values  (see 58 FR 20870).  EPA also
requested public comment on whether any of the options described in the
proposal for use of an RSC in deriving noncancer criteria and values should be
considered in calculating Great Lakes cancer criteria and values (HCVs).

      ii.   Comments:   Many commenters suggested that EPA not use a default of
100 or 80 percent, but to only derive an RSC when exposure data exists.  Other
commenters stated that the reason for a different RSC for BCCs and non-BCCs
for noncarcinogens was not clear.  One commenter suggested we use an RSC of 40
percent for all chemicals, but also acknowledged that this value was chosen
arbitrarily, not based on any specific data set.  Many other commenters
believed that no RSC should be derived at all; that the RSC should be 100
percent for BCCs and non-BCCs.  Several commenters stated that EPA should not
apply an RSC to the development of cancer criteria and values but did not
provide reasons for this opinion.

      EPA agrees with the commenters who suggested that actual data should be
used in developing an RSC when available.  As stated in the 1980 National
Guidelines, to account for exposures from other sources, actual  exposure data
can be subtracted from the RfD  (ADI, as it was called in 1980) to account for
contributions of the pollutant from diet and air  (ADI -  (DT + IN) where DT is
the estimated non-fish dietary intake and IN is the estimated daily intake by
inhalation  (U.S. EPA,   1980).  Therefore, where data are available, if States
or Tribes want to use actual data in developing their RSC, they  may do  so,
following the procedure outlined  in the 1980 National Guidelines.       It is
important to note, however, that EPA's policy on how to use exposure data in
developing an RSC is now under review.  Once EPA has finalized its policy
review on the RSC, EPA will address the application of the RSC during the
triennial review of Water Quality Standards under section 303 of the CWA.
Until such time, the EPA has decided  to apply an RSC of 80 percent to all
noncarcinogenic chemicals  (both BCCs  and non-BCCs).

      After further consideration and review of public comments, EPA does not
support the reasoning  of the Initiative Committees that there is a clear
difference in RSC development for BCCs as opposed to non-BCCs.   While it may
be true that surface water may be the major route of exposure for
bioaccumulatives  (through fish consumption),  even though a pollutant is not
bioaccumulative, it does not preclude the possibility that there may be other
significant sources of exposure.

      With regard to the use of a 80  percent default value, EPA believes that
the assumption helps to provide some  measure of protection against the
possibility that exposures from other sources may contribute to  the overall
exposure of the public to a particular contaminant.  Available  data for
indicate that non-water sources contribute  varying amounts to overall exposure
to a particular  chemical  (U.S. EPA  1982; U.S. EPA 1983).  Such  exposures can
occur  through air and  the  diet.   Since available  data indicate  that such
exposures can and do occur,  but these data  are  often limited  in their ability
to predict with  precision  the relative source contribution, EPA believes it is
prudent not to assume  that all exposure  to  a. pollutant  occurs from one  medium.
The  80 percent default was chosen because  it 'reflects the approximate
contribution from surface water pathways  (fish  consumption) to  the overall

-------
                              Section V: Human Health                           167
exposure to BCCs such as PCBs in the basin  (see section III., Relative Source
Contribution, in the Human Health TSD).

      For nonbioaccumulatives, 80 percent was also chosen as a default value
to account for the other possible non-water sources which may contribute to
the overall exposure of the chemical.  However, actual exposure data may also
be used in the. final Guidance by States and Tribes to calculate a relative
source contribution.  EPA recognizes that the choice of a default value of 80
percent in these cases is fundamentally a policy judgment that criteria
development should reflect the fact that exposures to a pollutant occur
through other media', rather than an empirically-based calculation of the
precise proportion of exposure via water versus non-water sources, since such
values vary on a case-by-case basis.  EPA also acknowledges that use of a 80
percent default for non-BCCs is a conservative measure, however, if other
significant exposures are not accounted for, the criteria could underestimate
overall exposure to the chemical and thus could underestimate the risk of
adverse health effects.  In addition, in the absence of data, it is prudent
and consistent with the health protection goals of the CWA to include a margin
of safety in the event that there are exposures from other sources.  The
important fact, EPA believes, is to take some accounting of other possible
exposure pathways. •

      With regard to the concern raised by some commenters that point sources
should not be expected to compensate for the failure to address other
pollutant sources, EPA does not believe that the relative source contribution
factor in the final methodology unduly burdens point source dischargers.  It
is common practice in EPA programs  (e.g., in establishing maximum contaminant
level goals under the Safe Drinking Water Act) to take into account other
routes of exposure to a chemical when establishing health-based standards for
a particular route of exposure.  If this step is not taken, and EPA were
always to assume that no exposures occurred through other media (in spite of
evidence to the contrary), then the totality of exposures could obviously
result in adverse health effects, contrary to EPA's goal of establishing
standards that insure that such effects do not occur.  EPA agrees, however,   "
that it is important to take steps to address all routes of exposure to
pollutants in order to achieve the greatest overall public health protection
at the least cost.

      iii.  Final Guidance:  The final Guidance provides for the application
of a 80 percent RSC for all noncarcinogenic chemicals.  This is a change from
the proposal, which provided for the application of the 80 percent RSC to
noncarcinogenic BCCs only.  The rationale for the change is explained above.

6.    Minimum Data Requirements/Tier I and Tier II

      In the proposal, EPA established a two tier system, each tier with a
different set of minimum data requirements.  This was established to
facilitate interpretation of State narrative standards and to lead to
development of as many criteria and values as possible.  EPA stated that the
data base necessary to derive Tier I criteria was fairly extensive and could
preclude the development of a permit limit for a chemical of concern, if the
minimum data was not available. In the absence of Tier I criterion, the
permitting authorities needed some mechanism with which to interpret and
ensure that the narrative prohibition against the discharge of toxic
substances in toxic amounts is reflected in water quality-based effluent
limitations.  To address this issue, a Tier II methodology, which requires a
less extensive data base (similar to the Tier II methodology for the
development of aquatic life values), was proposed for the development of human
health values.

      In the proposal, Tier I criteria and Tier II values for human health
were differentiated based on the quantity and quality of toxicological data
only.  There was no. differentiation based on the quantity or quality of BAFs.
After reconsideration, EPA has decided to differentiate the Tier I criteria

-------
168    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

and Tier II values for humans based on both the quantity and quality of
toxicological and bioaccumulation data."  The minimum BAF data required  to
derive a Tier I human health criterion for non-polar organics include either a
field-measured BAF, a BAF predicted from a field-measured BSAF, or a chemical
with a BAF less than 125 regardless of how the BAF was derived.  For inorganic
chemicals including mercury, BAFs derived based on field-measured BAFs  or
laboratory-measured BCFs are considered adequate for derivation of Tier I
criteria because there is little concern with metabolism for the majority of
inorganics.  The rationale for establishing the minimum data requirements for
BAFs is included in section IV.2.b of this document.  Based on this change,
the minimum data to derive Tier I criteria must include both the minimum
toxicological data, discussed below, and the minimum BAF data.  If these data
are not available, then a Tier II human health value must be developed.

      a.    Carcinogens.

      i.    Proposal:  Under the proposal, the methodology for deriving Tier I
criteria and Tier II values for carcinogens  {the human cancer values or HCVs)
was identical.  The only difference between Tier I and II was the weight of
evidence, and the amount and quality of data that was required for use  in
deriving the criteria or values.  The major issue regarding the two Tiers was
how to address Group C chemicals.  The proposal would have provided that Tier
I criteria be set for those types of Group C chemicals which are well
characterized and supported by a well-conducted study.  For those Group C
chemicals in which the cancer study  (or studies) indicate(s) a significant
increase of cancer in test animals but are limited by either: a marginal
statistical correlation between chemical and tumors due to high control tumor
incidence; a weak dose-response relationship; or an incidence of benign tumors
rather than malignant tumors, Tier II cancer values were to be derived  (if
enough data is available to conduct a quantification).  If a cancer
quantification could not be conducted due to lack of data  (number of test
animals, and/or only one dose group of animals has responded, making it
impossible to determine a slope factor) then the chemical was to be assessed
on a noncancer basis and a Tier I or Tier II human noncancer criteria or value
 (HNV) should be developed, if available data exists.

      EPA requested comments on: the procedures proposed for derivation of
Tier I criteria and Tier II values for possible carcinogens  (Group C);  the
alternative of using an additional uncertainty factor  (up to 10) on a
noncancer endpoint for Group C chemicals to provide protection from possible
carcinogenicity; and the alternative of deriving criteria and values for Group
C only through noncancer assessments without an added uncertainty factor for
possible carcinogenicity.

      Another option, which EPA requested comments on, was whether a Tier II
value could be set for an unquantifiable Group C chemical, whether endpoints
other than malignant tumors such as benign tumors or other precursors to
malignant tumors  (such as hyperplastic nodules  or peroxisome proliferation)
could be used to set a value.

      ii.   Comments:  With regard to how Group C chemicals should be
addressed, several commenters believed EPA should set criteria and/or values
for Group C chemicals on a case-by-case basis since the variability in  quality
of the data base which is used in establishing Group C status can be so wide.
Others were split on whether Group C  chemicals should be regulated at all
under Tier II.   Several commenters believed  the data supporting  Group C
chemicals was much too minimal to support Tier II value development.  Other
commenters supported the development  of Tier II values for Group C chemicals
citing the need  to protect the public from potential carcinogens even when the
supporting data  base is minimal.

      With regard  to the use of uncertainty  factors to account for potential
carcinogenesis,  commenters were split on  this issue.  Many found the practice
of using an extra uncertainty factor  indefensible from a  scientific viewpoint;

-------
                              Section V: Human Health                           169
others felt it was a sound policy which should be continued in the Drinking
Water Program and also used in the final Guidance.

      In response to the use of precursors, such as hyperplastic nodules or
peroxisome proliferation, to set a value, some commenters stated that their
use in setting criteria or values was controversial and could be difficult to
defend.  They cited examples where precursors such as hyperplastic nodules or
peroxisome proliferation were not associated with the onset of cancer.

      With regard to how Group C chemicals should be addressed, EPA agrees
with commenters that Group C chemicals should be dealt with on a case-by-case
basis and has changed the final Guidance to reflect this.  As the final
Guidance is written, States and Tribes have the discretion to develop criteria
or values for Group C chemicals based on the overall toxicological data base.
The final Guidance directs that this case-by-case determination be made taking
into account information including data on mutagenicity, genotoxicity,
structure activity, and mode of action.  EPA believes that those Group C
chemicals which act via a genotoxic mechanism (that is through direct
interaction with DNA),  may be most appropriately dealt with through use of a
linearized multistage model (IiMS) or other models which appropriately reflect
this type of mechanism of action (nonthreshold).  If the chemicals does not
interact with DNA, it may be best dealt with as a noncarcinogen and an RfD
should be developed.  (See the updated Human Health TSD, section II - Tier
designations - for guidance on determining whether an agent interacts with DNA
directly.  Several assays which are key to making such a determination are
listed.)

      With regard to the use of uncertainty factors to account for potential
carcinogenesis,  EPA believes the use of an extra uncertainty factor of up to
10 can be justified if there is concern of potential carcinogenesis (i.e.,
equivocal bioassay and genotoxicity results) and that the State or Tribe
should make this determination on a case-by-case basis.  However, as stated
above, a clear determination should be made whether the chemical interacts
directly with DNA.  If this is a clear cut decision (i.e., the genotoxicity
data is not equivocal),  then the use of an extra uncertainty factor may not be
necessary: either the chemical can be addressed as a carcinogen and quantified
using an LMS or other appropriate model or it can be addressed as a noncancer
agent and an ADE is set.  The determination whether to use an extra
uncertainty factor can also be based on which data set is more reliable or
convincing.  If the cancer data is marginal in terms of testing protocol and
statistics, but the noncancer data is well-conducted and unambiguous, it may
be preferable to use the noncancer data in setting a criterion with an extra
uncertainty factor of up to 10 to account for possible carcinogenicity.  EPA
stresses that the entire data base should be used in developing an overall
weight of evidence for human carcinogenicity before choosing a course of
action with regard to selecting a Tier or a risk assessment approach  (cancer
or noncancer).

      With regard to the use of precursors in developing Tier II values, EPA
agrees that the use of precursors as an endpoint may be controversial and
difficult to defend unless a clear determination of mechanism of action is
made implicating a precursor to an eventual tumor incidence.  However, if a
chemical is well studied and the mechanism of carcinogenesis is well
established indicating a clear procession from precursor to malignant tumor,
such endpoints can be used by States/Tribes in developing a Tier I criterion
or Tier II value.

      iii.  Final Guidance:  The final Guidance provides for a case-by-case
determination of carcinogenicity for Group C chemicals with an emphasis on
evaluating the entire data base.

-------
170    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      b.    Non-carcinogens.

      i.    Proposal:  The proposed Guidance also made a clear distinction
between Tier I and Tier II for noncarcinogens. Human Noncancer Values  (HNVs)
for the Tiers I and II are again distinguished on the basis of the available
data base.   The minimum acceptable data base for derivation of a Tier  I
criterion is at least one well-conducted subchronic mammalian study.   The
duration of the study must be at least 90 days in rodents, or 10 percent of
the lifespan of other appropriate species.  The preferred,minimum data point
for decision making is a NOAEL; however, the proposal also allowed the use of
a LOAEL involving mild, reversible effects, which may be considered acceptable
from longer term studies where a NOAEL may not be available. In developing
Tier II values, the proposal stated that the minimum acceptable data base was
a well conducted repeated dose mammalian study of at least 28 days.  However,
the proposal also maintained that the minimum acceptable data point for
decision making on such short term exposure data was a NOAEL.  In addition,
the proposal stated that the study ideally should be designed to observe all
possible systemic effects and include examinations for histopathology.  Data
from studies of longer duration (greater than 28 days) and LOAELs from such
studies were also allowed in some cases for derivation of Tier II values.  EPA
did not want to preclude the use of LOAELs from studies slightly longer than
the required 28-day studies  (such as 30 day tests) if the LOAEL from such a
study represents an effect which is mild, reversible, close to a probable or
actual NOAEL, and is  representative of effects observed over chronic
exposures.

      EPA also recognized in the proposal that when the Tier II methodology is
used to derive a HNV, an additional uncertainty factor of up to 10 may be
applied in deriving the ADE to account for the difficulties in extrapolating
from a short term NOAEL to a long-term NOAEL.  Structure activity
relationships  (SARs), and all other available data on the chemical should be
used to determine the appropriate additional uncertainty factor.

      EPA requested comments on several issues including: the use of a 28-day,
or other subacute study with the use of additional uncertainty factors;
whether the use of the Tier II human health methodology was appropriate;
whether or not even shorter term studies, such as 14 day studies, might
effectively be used in a Tier II HNV approach; and the appropriateness of
using surrogate chemicals employing an SAR evaluation to develop Tier  II
values and the use of SAR as a screen for requiring additional toxicity
information.

      ii.   Comments:  The majority of commenters opposed the use of 14 day
studies, while they were divided on the use of 28-day data for developing Tier
II values.   Many commenters stated the 28-day minimum was scientifically
indefensible since a 28-day study is not designed to characterize subchronic
or chronic risk  (it is less than 10 percent of most test mammals' lifespan)
and that EPA should only develop Tier I criteria using the data base specified
for Tier I.  Other commenters believed the 28-day study was the minimally
acceptable test but they also cautioned that the values derived with 28-day
study data be used only on an interim basis in order to encourage the
development of longer term data.  Other commenters stated that the use of a
LOAEL should not be allowed.  With regard to comments on the overall Tier II
methodology, several commenters supported the Tier II process but indicated
that resulting values should only serve as an interim risk assessment  on a
chemical.  Other commenters opposed the Tier II process stating that if good
data does not exist for a chemical, it is unreasonable to regulate it  on such
a basis.

      With regard to the use of SAR in the development of Tier II values, many
commenters were  in favor of using SAR to evaluate Tier II chemicals but not to
develop  surrogate chemical values.  Commenters believed that SARs are  very
theoretical in nature, without solid data foundation, and could not form the
basis for a permit limit.

-------
                              Section V: Human Health                           171
      EPA believes the use of 28-day data is supportable in the context of
Tier II development.  While such short 'term studies typically do not reveal
evidence of possible adverse effects resulting from longer term exposures,
there is some evidence that  the data can be correlated to longer term studies
(Wiel and McCollister, 1963) .

      EPA acknowledges that the certainty in the overall risk assessment is
not as high as- with 90-day study results since it is necessary to extrapolate
from a short term study to lifetime exposures, and the correlatability of
short term study results to long term study results is not definitive and
quantitative.  However, EPA believes that the number derived using the 28-day
data and an extra uncertainty factor will be adequately protective  (since the
uncertainty factor will further reduce the final value) and will serve as
motivation to obtain longer term data to develop Tier I criteria.  For these
same reasons, EPA believes the Tier II process is a sound approach to
developing interim regulatory values for pollutants with minimal data bases.
      In addition, EPA believes that for the majority of chemicals that will
be found in discharge effluents, there is already adequate data to develop
Tier I criteria.  This is based on a. review of the toxicological data bases
for the 138 pollutants of initial focus.  These are the chemicals identified
by the Steering Committee to be known or suspected of being of primary concern
in the Great Lakes basin (see 58 FR 20843).  Of the 138, EPA estimates that
about 120 have enough data to derive Tier I criteria.  This leaves about 20
chemicals with either insufficient data to calculate a Tier I criterion or no
data to calculate either a Tier I criterion or Tier II value.  Thus, EPA
believes the number of chemicals with Tier II values will be minimal.  (See
section II.C.5 for a more complete discussion.)

      With regard to the use of SAR, EPA agrees that SAR may be useful in
assessing the potential effects of a chemical and may be valuable in selecting
the uncertainty factor for a Tier II value.  However, it is a process which
requires a great deal of scientific judgement and can be open to differing
interpretations.  Because of these concerns, EPA has decided to not require
the use of SAR as the basis for II values.   EPA believes the expertise or the
resources may not exist in many States to use SAR for routine regulatory
purposes.  In addition, the interpretation of SAR data may lead to
inconsistent values, among the Great Lake States. With further research and a
greater confidence in the process, SAR may be used in the future to derive
surrogate chemical Tier II values.

      iii.  Final Guidance:   For the reasons stated above, EPA is maintaining
the methodology for developing Tier II values, which relies upon the use of
28-day study results as the'basis of value development and the use of SAR for
the selection of uncertainty factors.

D.    Criteria Derivation

      The final Tier I human cancer criteria or Tier II value are calculated
as follows:

      HCV  =      RAD x BW	
                  WC + [(FCnj x  BAFTU)  +  (PC™ x BAFTL4) ]

      Where:

      HCV = Human Cancer Value in micrograms per liter  (/jg/L) .

      RAD = RAD in milligrams toxicant per kilogram body weight per day
(mg/kg/day)  that is associated with a lifetime incremental cancer risk equal
to 1 in 100,000.

      BW = Body weight of an average human (BW = 70kg).

-------
172    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      WC = average per capita water consumption  (both drinking and incidental
exposure) for surface waters classified as public water supplies  (WCd = 2
L/day) and average per capita incidental daily water exposure for  surface
waters not used as public water supplies  (WCr =  0.01 liters/day)

      FCnj = mean consumption of trophic level 3 fish by regional  sport
fishers = 0.0036 kg/day

      FCtL4 = mean consumption of trophic level 4 fish by regional  sport
fishers = 0.0114 kg/day

      BAFTL3 = BAF for trophic level 3 fish

      BAFTL4 = BAF for trophic level 4 fish

      The Tier I human noncancer criteria or Tier II value  is calculated as
follows:

     HNV  =  APE x BW x RSC	
             WC +  [(FCTL3 x  BAFTL3)  + (FCru x BAFTL4) ]

      Where:

      HNV = HNV in micrograms per  liter  (jjg/L) .

      ADE = ADE in milligrams toxicant per kilogram  body weight  per day
 (mg/kg/day).

      RSC = RCS factor of 0.8 for  all chemicals  of concern.   This is used to
allow for potential exposure via sources  other than  consumption  of
contaminated water and fish recreational  exposure.   States may develop an RSC
using actual exposure data following the  procedures  specified in the 1980
National Guidelines.

1.    Proposed Criteria and Values

      40 CFR part  132, Table 3, sets forth HCVs  and  HNVs for 18  chemicals
which have been derived using the  final human health methodology.  Note that
for each HCV and HNV, two criteria are provided.   The first  is that which
applies when exposure is from recreational activities and consumption of
aquatic organisms. .The second  is  that which applies when exposure is from
consumption of aquatic organisms,  drinking water and recreational activities.
EPA requested comments on the proposed HCVs and  HNCs in Table 3  of proposed 40
CFR part 132.  The docket for the  final rulemaking contains  technical support
documents describing the details of derivation of each criterion and value.

      The criteria for the chemicals in Table 3  of part 132  have been modified
from the proposal  to reflect changes in the BAFs and the use of  a default RSC
of 80 percent for  all noncarcinogens not  just those  noncarcinogens that are
bioaccumulative chemicals of concern  (BCC).  The rationale  for the changes in
the BAFs are discussed in section  IV of this document.  The  change in the RSC
is discussed in section V.C.S.g above.

      The criterion for heptachlor and pentachlorophenol are not included in
Table 3 of part 132 because  the BAFs for  the chemicals were  estimated using a
laboratory predicted BCF multiplied by a  food chain  multiplier and their BAFs
were greater than  125.  Consequently, they did not meet the  minimum BAF data
requirements for deriving a  Tier  I human  health  criterion as discussed in
 section V.C.6 above.

      The noncancer criterion  for  trichloroethylene  and toxaphene are not
 included in Table  3 of part  132 because  the RfDs needed to derive a human
noncancer  criterion'are currently under  review by EPA.  In the proposal, the
noncancer  criterion were derived  using RfDs that had not been verified by EPA.

-------
                              Section V: Human Health                           173
EPA has decided that it would not be appropriate to include criterion  in Table
of part 132 that do not have a RfD verified by EPA.  For this reason,  the
final Guidance does not include a human noncancer criterion for  toxaphene or
trichloroethylene.  However, even if the noncancer criterion used in the
proposal were retained, the cancer criterion for the two chemicals would be
lower  (toxaphene - cancer criterion is 6.8 x 10"5 A*g/L and noncancer criterion
would be 2.1 x 10~2 pig/L; trichloroethylene - cancer criterion is 29 /xg/L and
noncancer criterion would be 470 /KJ/L) .

      The ADE  (RfD) for chlordane was modified to correct an error in  the
uncertainty factors used in the proposal.  The ADE in the proposal was 5.50 x
10"4 mg/kg/day.  This was based on a NOAEL of 5.5 x 10'2 mg/kg/day and an
uncertainty factor of 100 to account for interspecies variability and
intraspecies variability.  An additional uncertainty factor of 10 should have
been included to account for the lack of an adequate reproduction study and
adequate chronic study in a second mammalian species and the generally
inadequate sensitive endpoints studied in existing studies.  The resulting
uncertainty factor of 1000 is recommended by EPA in its IRIS data base.  The
resulting ADE is 5.50 x 10"5 mg/kg/day and the noncancer criterion for
chlordane is 1.4 x 10"3 /ig/L.  The cancer criterion for chlordane is 2.5 x 10"*
      The NOAEL for hexachloroethane in the proposal was adjusted from 1.0
mg/kg/day to 0.71 mg/kg/day to account for the fact that the test animals were
fed only 5 days a week.  Upon further review of the study used to estimate the
NOAEL, it was determined that the test animals were fed every day of the week
and not 5 days per week as assumed in the proposal.  An adjustment of the
NOAEL in the proposal to account for the 5 day feeding week was therefore not
justified.  Thus, the final Guidance uses a NOAEL of 1.0 mg/kg/day.

      The NOAEL used for deriving noncancer criterion for methylene chloride
in the proposal was 6.47 mg/kg/day.  This was based on a study in which the
NOAELs were 5.85 and 6.47 mg/kg/day for male and female rats, respectively.
The NOAEL for the female rat was used in the proposal as the basis for
deriving a noncancer criterion.  Based on further review, EPA believes that
the NOAEL of the male rats should be used in deriving a noncancer criterion.
EPA decided to use the male NOAEL because it will provide slightly greater
protection and is consistent with the NOAEL used in IRIS.  In either case, the
cancer criterion for methylene chloride is more stringent than the resulting
noncancer criterion,.  The noncancer criterion using the female rat study would
be 1.8 x 103 £jg/L,  the  criterion using the  male  rat study would be 1.6  x 103,
and the cancer criterion is 47
      a.    PCBs  (Human Cancer Value)

      i -    Proposal :  The proposed criterion document for PCBs set a HCV of 3
pg/L for both drinking water sources and non-drinking water sources.  These
criteria were based on a slope factor of 7.7  (mg/kg/day) -1 derived from the
rat bioassay of Norback and Weltman  (1985) .  This study utilized large numbers
of Sprague Dawley rats  (70/sex/dose) in a  chronic Aroclor 1260 feeding study.
Animals were administered 100 ppm for 16 months, followed by a 50 ppm diet for
an additional eight months, then a basal diet for five months. Among the test
animals that survived for at least 18 months, females exhibited a 91 percent
incidence of malignant liver tumors.  In males, corresponding incidences were
four percent for liver tumor and 11 percent for neoplastic nodules.

      ii .   Comments:  Several commenters  were critical of the PCB criterion
and made the following comments : there is  no evidence that congeners other
than those found in Aroclor 1260 are carcinogenic; EPA should pool the data
from all Aroclor studies and develop a geometric mean from these studies; the
re-evaluation of tumors in the Norback and Wellman (1985) study by IEHR
results in a substantially lower cancer potency; and the epidemiology data on
PCBs indicates that the current animal based cancer potency is overly
conservative .

-------
174    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      With regard to the comment that congeners other than those found  in
Aroclor 1260 are not carcinogenic, as stated on IRIS:  "Although animal
feeding studies demonstrate the carcinogenicity of commercial PCB
preparations, it is not known which of the PCB congeners in such preparations
are responsible for these effects, or if decomposition products, contaminants
or metabolites are involved in the toxic response."  Nominally, it appears
that animal studies with Aroclor 1260 are the only studies indicating
carcinogenic response.  However, there are indications that Aroclor 1254 may
also be carcinogenic.  A study by NCI (1978) reported carcinomas of the
gastrointestinal tract in Fischer rats treated with Aroclor 1254, however the
incidence was not statistically significant.  While it is not statistically
significant, the presence of such a response still raises the concern
regarding the carcinogenicity of PCB mixtures other than Aroclor 1260.   The
EPA believes it is not reasonable to develop a criterion for each PCB Aroclor
mixture.  PCBs are mixtures of chlorinated biphenyls.  Each mixture may
contain up to 209 possible individual compounds.  Each of the mixtures would
be expected to contain all combinations of chlorinated compounds even though
some of them only in small or trace amounts.  Since it is not known which
compound of the 209 is clearly responsible for eliciting a cancer response, to
conservatively protect against the potential for carcinogenicity associated
with all PCB mixtures, EPA has chosen to base criteria on study results using
Aroclor 1260.

      EPA disagrees*with the commenters who suggest EPA should pool all the
cancer data from all the available congener studies in developing a criterion
for PCBs for following reasons:  The Norback and Wellman  (1985) study is
judged by EPA as acceptable in design and conduct compared to other available
studies.  As stated on IRIS, "The estimate  (slope factor) based on the data of
Norback and Weltman  (1985) is preferred because Sprague Dawley rats are known
to have low incidence of spontaneous hepatocellular neoplasms.  Moreover, the
latter study spanned the natural life of the animal, and concurrent
morphological liver studies showed the sequential progression of liver  lesions
to hepatocellular carcinomas."   Normally EPA will pool study data only if
available studies are considered of marginal quality.  The Norback and Wellman
study is considered a good study and therefore the pooling of studies is not
needed.

      In response to the reevaluation of the tumor data by IEHR, the IEHR
potency factor was 5.1  (mg/kg/day)-1 as compared to the EPA derived cancer
potency factor of 7.7  (mg/kg/day)-1.  The IEHR reevaluation was a reread of
the histopathology slides from Norback and Wellman study, in which several
lesions which were originally interpreted as malignant lesions were
reinterpreted as non-malignant lesions  (neoplastic nodules).  The EPA is
currently reviewing the IEHR data for PCBs.  Until the EPA can fully assess
the validity of the'conclusions drawn in the IEHR reevaluation, it would be
premature to make the suggested changes in the criterion.

      In response to the epidemiology data indicating that humans are less
sensitive than test species to PCBs,  EPA does not believe the epidemiological
results are as conclusive as the animal results from the Norback and Wellman
study.  As presented on IRIS, the human carcinogenicity data is considered
inadequate.  Although there are many epidemiological studies, the data  are
inadequate due to confounding factors.  The factors noted in IRIS are:
population differences  in alcohol consumption, dietary habits, ethnic
composition, contamination of PCBs by dibenzofurans, and exposure of workers
to other known carcinogens.  It is EPA's longstanding practice to rely  upon
studies in animals for  risk assessment in the absence of adequate human data.

      iii.  Final Guidance:  For the  reasons stated above, EPA continues to
believe that the study  by Norback and Weltman is the best available study to
use for deriving a human health criterion for PCBs.

-------
                              Section V: Human Health                           175
      b.    TCDD.

      In the proposal, EPA stated that it is currently conducting a major new
dioxin research and analysis effort, the results of which could not be
reflected in the proposed Guidance.   EPA also stated that if the results of
this study became available prior to the finalization of the rule, EPA
expected to publish-a notice of availability and solicit comment on whether
the rule should be modified to reflect this new information.   EPA invited
comment on the approach it should take to establishing dioxin criteria pending
completion of its ongoing dioxin studies.

      Many commenters stated that EPA should not develop criteria for dioxin
until the dioxin reassessment is final.  Other commenters stated that EPA
should not delay the publication of dioxin criteria until the reassessment is
finalized.

      EPA agrees with commenters that it is important that EPA develop a
dioxin criterion regardless of the status of the dioxin reassessment.  Dioxin,
from all indications, is one of the most potent carcinogens and must be
regulated with the most recent available data at hand.  Once the dioxin
reassessment is final, EPA will revisit the dioxin criterion, and make changes
if needed.  EPA's proposed dioxin reassessment was made public on September
13, 1994.  The final dioxin reassessment is anticipated sometime in late 1995.

      i.    TCDD - Noncancer Criterion

       (A).  Proposal:  In the proposal, EPA developed a Tier I criterion of
0.1 pg/L for drinking water and nondrinking water sources based on a
reproductive study by Bowman et al.  (1989)  on rhesus monkeys which indicates a
LOAEL based on behavioral effects at 25 ppt (0.67 ng/kg/day) and a NOAEL at 5
ppt (0.13 ng/kg/day).  The ADE was developed by dividing the NOAEL of 0.13
ng/kg/day by an uncertainty factor of 100 (10X for intraspecies variability
and 10X for interspecies extrapolation).

       (B).  Comments:  Several commenters believed the intra- and interspecies
adjustment of 10 was unjustified since the rhesus monkey data was used.  They
argued that the rhesus monkey is very close to humans in terms of
pharmacokinetics and therefore an interspecies uncertainty factor of three is
warranted instead of the usual uncertainty factor of 10 which is applied for
interspecies differences between humans and less similar mammals such as
rodents.

       (C).  Final Guidance:   With regard to the noncancer criterion, it is
EPA's judgement to apply an intraspecies uncertainty factor of 10 to account
for variability and sensitivity within the human population.   EPA also
believes that the interspecies uncertainty factor of 10 is justified because
the study groups were very limited in size and the statistical and biological
significance of the findings are unclear.   In addition, metabolic and
pharmacokinetics parameters for humans and rhesus monkeys may be sufficiently
different.  For these reasons, EPA believes an interspecies uncertainty factor
of 10 is justified.

      ii.   TCDD - Cancer Criterion

       (A)   Proposal:  The proposed criterion document for dioxin presented a
human cancer value of 0.01 pg/L for drinking and nondrinking water sources.
These criteria were based on a slope factor of 7.5 x 104  (mg/kg/day)-1  based
on the pooled significant tumors in female rats of Kociba, et al. (1978) with
the liver tumor reevaluation of the Pathology Working Group  (Sauer,  1990) .

       (B).  Comments:  Commenters stated that the dioxin criteria should
reflect the 1986 cancer guidelines  (U.S. EPA,  1986), not the draft EPA cancer
guidelines which EPA is in the process of revising.

-------
176    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      (C).   Final Guidance:  EPA agrees that the 1986 cancer guidelines are
appropriate for use in developing the dioxin criterion, and believes this is
appropriate until the revised cancer guidelines are peer reviewed, publicly
reviewed and finalized.

      c.    Mercury.

      i.    Proposal:  EPA proposed a criterion of 2 ng/L for mercury for both
drinking water and nondrinking water sources.   This criterion was based on a
LOAEL of 3 ug/kg/d and an uncertainty factor of 50.  The LOAEL was based on
several studies that have shown neurological symptoms of mercury toxicity in
adults at blood levels of mercury in the range of 200 to 500 ng/ml.  The 200
ng/ml mercury levels in blood have been associated with an oral intake  in
adults of 3 ug/kg/d of mercury and with adult hair concentrations of 50 ug/g.
The uncertainty factor of 50 is composed of a 10-fold factor to adjust  the
LOAEL to a NOAEL and an additional 5-fold factor to ensure the criterion will
provide protection from the potential fetal effects of mercury exposure via
maternal ingestion of mercury contaminated fish.

      ii.   Comments:  Several commenters agreed with the EPA criterion and
the choice of the study and uncertainty factors. Other commenters agreed with
the mercury criterion but disagreed with the basis for developing the
criterion.   These commenters stated that the LOAEL should be based on studies
indicating fetal effects on the central nervous system occur at a LOAEL of 10
ug/g in maternal hair.  They argue that if the 10 ug/g LOAEL is used, the 5-
fold uncertainty factor to ensure the criterion will provide protection from
potential fetal effects would not be needed.

      iii.  Final Guidance:  EPA continues to believe the adult LOAEL of 3
ug/kg/d should be used in the derivation of criterion instead of the 10 ug/g
maternal hair concentrations suggested by the commenter.  EPA believes  the
adult effects are more clearly delineated from the available data than  the
fetal effects and thus the use of a LOAEL of 3 ug/kg/d  (50 ug/g adult hair
concentrations) is appropriate.  The LOAEL of 10 ug/g maternal hair
concentrations is predicted and therefore can be viewed as a somewhat less
reliable endpoint upon which to base a criterion than the adult endpoints.
EPA continues to believe the 5-fold uncertainty factor is justified to protect
central nervous systems development during the sensitive fetal life stages.
In addition, as discussed in section 5.a. above, EPA has assumed a body weight
of 65 kg (as opposed to 70 kg) for mercury.  The resulting Tier I mercury
criterion is therefore 1.9 ng/L, which is slightly less than the proposed
criterion of 2 ng/L.

      Since the proposal, EPA's Rf>  work  group has  recently revised the  RfD,
using an effect level of 1 ug/kg/d and using an uncertainty factor of 10 to
account for within-human variability and for an insufficient data base.  The
resulting RfD is 0.1 ug/kg/d which is higher than the proposed RfD  (ADE) of
0.06 ug/kg/d.  However, because the new RfD of 0.1 ug/kg/d was not verified
until early February 1995, it was not possible to publish the data, request
comment, and revise the final Guidance, if needed, prior to promulgation of
the final Guidance. Consequently, EPA plans to publish a Notice of Data
Availability after the publication of the final Guidance with the new mercury
assessment for human health and will change  the final mercury criteria  for
human health if appropriate.

E.    Relationship of the  Great Lakes Initiative Guidelines to National
      Guidelines Revisions

      1.    Proposal:  As  stated  in the proposal, much of the Great Lakes
methodology for deriving human health criteria was based on the  1980
methodology and advances in  the science since 1980.  Concurrent with the
development of the fi-nal Guidance, EPA is also in  the process of reviewing and
revising the 1980 National Guidelines which would  apply to development  of EPA
National water quality criteria under section 304(a) of the CWA.    It is

-------
                              Section V: Human Health                           177
expected that a proposed revision to the National guidelines will be published
in the Federal Register in  1995 and that there will be a separate opportunity
for public comment on that proposal.  In the proposal, EPA discussed the
possibility of making no change from the current 1980 methodology.
Accordingly, EPA requested comments on the possibility of retaining the
approach set forth in the 1980 National Guidelines with respect to each
individual component of the proposal that differs from the current Natiorial
guidelines.

      2.    Comments:  A few commenters stated that EPA should retain the 1980
methodology but these commenters did not  discuss why EPA should take such an
approach.  The majority of commenters were in favor of EPA using the best
science in conducting any risk assessment which becomes the basis for water
quality criteria.  Other commenters stressed that EPA must strive for
consistency between the methodology presented in the proposal and the revised
national methodology.

      3.    Final Guidance:  EPA believes it has presented a methodology which
reflects the best science to date.  EPA  also believes it is essential to
revise the 1980 National Methodology to reflect the latest science and policy
of the EPA and the scientific community.  Prior to revising the Guidelines EPA
will request comments on the latest developments in draft EPA policy such as
the draft revised Cancer Guidelines, the proposed policy on body weight to
surface area scaling, and the proposed RSC policy.   Dntil EPA's revised
policies are finally adopted, after consideration of public comments, EPA
believes it would be premature to adopt such revisions as part of the final
Guidance.

      With regard to fish consumption, EPA has not yet revised its national
policy on developing a fish consumption rates  However, in the final Guidance,
EPA believed it was appropriate to develop a regional fish consumption rate
for the protection of the population of the Great Lakes basin and therefore
adjusted the national fish consumption rate accordingly with regional
consideration in mind.

F.    Comparison with the CWA and Great Lakes Water Quality Agreement

      The Great Lakes Critical Programs Act of 1990 (CPA) states that the
proposed Guidance shall be no less restrictive than the provisions of the CWA
and national water quality criteria and guidance.  The CPA also specifies that
the final Guidance is to conform with the objectives and provisions of the
Great Lakes Water Quality Agreement (GLWQA).  The discussion below addresses
conformance of the final human health methodologies and criteria with these
requirements.

      1.    Tier I Criteria/Methodology

      a.    Comparison with the CWA.

      Under the authority of section 304(a)(1) of the CWA, EPA established the
1980 National Guidelines, to be used in deriving National human health
criteria.  EPA believes that although the final Tier I human health criteria
methodology and the criteria are not identical to the 1980 National Guidelines
and individual National criteria in all details, they are generally no less
restrictive.

      First, as discussed above in this section of the document, EPA finalized
Tier I human health criteria for 18 pollutants for which National criteria
exist.  These pollutants include a broad selection of pollutants of initial
focus proposed by the Initiative Committees to test the proposed methodology.
Although the final Guidance includes only these 18 pollutants while National
human health criteria are currently available for 91 pollutants, EPA believes
that this approach will not result in less stringent levels of control.  This
is because under the implementation scheme presented today, Great Lakes States

-------
178    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


would be required to derive criteria and values for these pollutants and  for
all other pollutants except those listed in Table 5 of part 132  whenever
sufficient data exist to meet Tier I or Tier II minimum data requirements and
the State determines that it is necessary to control these pollutants.  Thus,
the scope of the final Guidance in terms of pollutants covered is actually
broader than the current National guidance.

      Furthermore, because the Tier I criteria for human health  assume  a
higher fish consumption rate than the national criteria, use BAFs rather  than
BCFs to calculate fish tissue residues, and include relative source
contributions for all noncarcinogens the final numeric criteria  are equivalent
to, or more restrictive than, the current national criteria.

      b.    Conformance with the GLWOA.  For the reasons stated  in section
III.D  (Aquatic Life) of this document, EPA believes that the final Guidance
conforms to the General Objectives of the Agreement regarding the elimination
or reduction of discharges into the Great Lakes System.  For the 18 pollutants
for which Tier I human health criteria have been derived, the final Guidance
criteria are more stringent than the water quality criteria presented in  the
Agreement, except for lindane.  The GLWQI criterion for lindane  is based  on
noncancer effects. EPA is currently reviewing the carcinogenicity of lindane.
EPA believes that a.cancer criterion for lindane would be lower  than the  GLWQA
for lindane.

2.    Tier II Criteria/Methodology

      a.    Comparison with the CWA.  EPA's current guidance and regulations
for water quality standards contain nothing directly analogous to the two-tier
approach proposed today for human health.  States currently have very broad
discretion when regulating pollutants that are subject only to narrative
criteria.  EPA believes that the final Guidance is more rigorous than the
current National requirements in this area because the Tier II methodology
derives generally more conservative values for non-cancer criteria to
compensate for greater uncertainty in the data base.   Based on  studies done
to date, EPA expects that Tier II values will be more stringent  than existing
standards for these pollutants in most cases.  Further, this approach imposes
a structure to the process of translating narrative criteria into numeric
values.  Finally, the final Guidance will result in more uniform control  of
pollutants lacking National standards in the Great Lakes States.

      b.    Conformance with the GLWOA.  EPA believes that the Tier II
methodology is consistent with the General Objectives of the Agreement.
Moreover, it serves.as a translator mechanism of the States' narrative  water
quality standards.  The Tier II methodology will enhance regulatory efforts in
the Great Lakes basin, will serve its purpose of promoting consistency  in the
regulation of toxics in the Great Lakes basin, and is therefore  also in
Conformance with the Agreement.

G.    References

      Bowman, R.E., et al. 1989.  Chronic dietary intake of 2,3,7,8-
tetrachlorodibenzo-p-dioxin  (TCDD) at 5 or 25 parts per trillion in the
monkey: TCDD kinetics and dose-effect estimate of reproductive toxicity.
Chemosphere. 18(1-6); 243-252.

      Cantor, K.P., R.  Hoover, P. Hartage.  1987.  Bladder cancer, drinking
water source, and Tap water  consumption: A  case control study.   J. National
Cancer  Institute. 79(6): 1269-1279.

      Connelly,  N.A., T.L. Brown  and  B.A. Smith.  1990.  New York Statewide
Angler  Survey, 1988.  New York State  Department of Envirnomental Conservation,
Albany, NY.

-------
                              Section V: Human Health                           179
      Dedrick, R.L/1973.  Animal Scale Up.  J. Pharamcokin. Biopharm. 1:435-
461.

      Ershow, A.G. and K.P. Cantor, 1989.  Total Water and Tapwater Intake' in
the United States: Population-Based Estimates of Quantities and Sources,
National Cancer Institute, Bethesda, MD.

      Fiore, B.J. et al., 1989.  Sport fish consumption and body burden levels
of chlorinated hydrocarbons: A study of Wisconsin anglers, Archives of
Environmental Health, 44:82-88.

      Freireich, E.J., E.A. Gehan, D.P. Rail, L.H. Schmidt, and H.E. Skipper.
1966.  Quantitative comparison of toxicity of anticancer agents in mouse, rat,
hamster, dog, monkey and man.  Cancer Chemother.  Rep. 50:219-244.

      Interagency Regulatory Liaison Group. 1979.  scientific Basis for
Identification of Potential Carcinogens and Estimate of Risks.  Journal of
National Cancer Institute. Vol. 63. pp. 241-268.

      Kociba, R.J. et al.  1978.  Results of a two-year chronic toxicity and
oncogenicity study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rats.  Toxicol.
Applied Pharmacol. '46:279-303.

      NCI  (national Cancer Institute). 1978. bioassay of Aroclor 1254 for
possible carcinogenicity. CAS No. 27323-18-8.  NCI Carcinogenesis Tech. Rep.
Ser. no. 38.

      Norback, D. and R.H. Weltman.  1985.  Polychlorinated biphenyl induction
of hepatocellular carcinomas in the Sprague-Dawley rat.  Environ. Health
Perspectives.  60:97-105.

      Pinkel, D.. 1958.  The use of body surface area as a criterion of drug
dosage in cancer chemotherapy.  Cancer Res. 18:853-856.

      Sauer, R.M. 1990.  Pathology Working Group: 2,3,7,8-Tetrachlorodibenzo-
p-dioxin in Sprague-Dawley rats.  Pathco, Inc.  Submitted to the Maine
Scientific Advisory Panel.

      U.S. Environmental Protection Agency.  1980.  Water Quality criteria
Availability, Appendix C - Guidelines and methodology used in the preparation
of health effects assessment Chapters of the Consent Decree Water Quality
Criteria Documents.  Federal Register. Vol. 45, November 29, 1980, 79347-
79357.

     U.S. Environmental Protection Agency. 1982.  Dichloromethane  (Methylene
Chloride).  Occurrence in Drinking Water, Food and Air.  Prepared by JRB
Associates Under EPA Contract No. 68-01-6388.  Office of Drinking Water,
USEPA.  July 11, 1982.

     U.S. Environmental Protection Agency. 1983.  Benzene.  Occurrence in
Drinking Water, Food and Air.  Prepared by JRB Associates Under EPA Contract
No. 68-01-6388.  Office of Drinking Water, USEPA.  November 23, 1983.

      U.S. Environmental Protection Agency. 1986.  Guidelines for Carcinogenic
Risk Assessment.  Federal Register, vol. 51, No. 185, September 24, 1986,
33992-34002.

      U.S. Environmental Protection Agency. 1989.  Exposure Factors Handbook,
Washington, DC, Office of Health and Environmental Assessment.  EPA/600/8-
89/043.

      U.S. Environmental Protection Agency. 1991.  General Quantitative Risk
Assessment Guidelines for Noncancer Health Effects.  External Review Draft.
EPA Document No. ECAO-CIN-538.  Feb. 1991. pp. 1-3.

-------
180    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      U.S. Environmental Protection Agency-  1992.  Draft Report:  A Cross-
species 'Scaling Factor  for Carcinogenic  Risk Assessment based on Equivalence
of mg/kg3'4 per  day.   Federal  Register.  June 5, 1992, 57 FR 24152.

      U.S. Environmental Protection Agency.  1994a.  Notice of data
availability and request for  comments. Federal  Register.  Vol. 59,  No. 167,
August 30, 1994.  44678-44684.

      U.S. Environmental Protection Agency.  1995.   Fish Consumption Estimates
Based on  the 1991-1992  Michigan  Sport  Anglers Fish Consumption Survey. Final
Report. EPA Contract No. 68-C4-0046, SAIC Project  No.  01-0813-07-1676-040.
February  21, 1995.

      West, P., M. Fly, R. Marans, F.  Larkin and D.  Rosenblatt.  1993.  1991-
1992 Michigan Sport-Anglers Fish Consumption Study.   Final report to the
Michigan  Great Lakes Protection  Fund,  Michigan  Dept.  of Natural  Resources.
University of Michigan, School of Natural Resources.  Natural Resources
Sociology Research Lab.  Technical Report #6.   May 1993.

      West, P.C. et al. 1989.  Michigan  Sport Angler Fish Consumption Survey:
A Report  to the Michigan Toxic Substance Control Committee,  University of
Michigan  Natural resources Sociology Research Lab.  Technical Report #1, Ann
Arbor, Michigan, MDMB Contract # 87-20141.

-------
                                Section VI: Wildlife                             181
                           VI.   WILDLIFE
A.    Introduction
      For the purposes of the final Guidance,  "wildlife" is defined as all
non-domesticated species in the taxonomic classes Aves and Mammalia (birds and
mammals).   The final Guidance for deriving wildlife criteria is provided in
appendix D to part 132.   The final Great Lakes Water Quality Initiative
Technical Support Document for Wildlife Criteria (EPA 820-B-95-009)  (Wildlife
TSD) and the final Great Lakes Water Quality Initiative Criteria Documents for
the Protection of Wildlife:  DDT and Metabolites; Mercury; 2,3,7,8 TCDD; and
PCBs (EPA 820-B-95-008)  (Wildlife Criteria Documents),  which provide the data
and the derivation of each individual criterion, are available in the docket
for this rulemaking.

      As stated in the proposal and in section I (Background)  of the final
Guidance,  wildlife in the Great Lakes are at risk from contaminants in Great
Lakes water.  To address this problem, the proposal presented a methodology
for deriving criteria and values that would likely be protective of wildlife
within the Great Lakes basin.  EPA requested comments and asked for additional
information on several topics relating to the methodology.  Described below is
a summary of the proposal for that topic, significant comments received (with
EPA's response to each), and the requirements of the final Guidance.
Responses to all comments received are provided in the Response to Comments
Document,  which is included as part of the docket for the final Guidance.

      Currently, there is no National methodology for the development of water
quality criteria for the protection of wildlife comparable to the proposed
Guidance;  however, there is a mechanism for consideration of wildlife impacts
within the 1985 National aquatic life criteria guidelines (Stephan et al.,
1985).   While the aquatic life methodology provides a mechanism to protect
against bioaccumulation of a compound within a food web, it has several
limitations.  The current approach in the national aquatic life guidelines to
provide protection to wildlife is the use of the final residue value (FRV),
which represents the highest prey tissue concentration of a toxicant which
will not produce an adverse effect in the consumer organism.  The FRV is
derived through one of three approaches:  either from a Food and Drug
Administration  (FDA) action level; a long-term wildlife field study; or a
chronic wildlife feeding study.  Because an FDA action level is intended to
protect humans, and not wildlife, an FRV derived from an FDA action level for
a specific contaminant does not ensure protection of wildlife species which
may consume contaminated aquatic organisms as a larger portion of their diet
or exhibit a greater sensitivity than humans.   While an FRV may be more
adequate based on a*long-term wildlife field study or a chronic wildlife
feeding study, there are very few chemicals for which these studies are
available.  In addition, such studies are frequently only conducted after the
chemical has been theorized to be responsible for a problem in the
environment.

      In cases where no FRV is available, biomagnification of a chemical into
the higher trophic levels of a food web, and potential impacts on these
wildlife species, is not considered in the derivation of the aquatic life
criterion.

      EPA has begun a separate effort to derive National wildlife criteria.
Following the release of the 1987 General Accounting Office (GAO) report
entitled "National Refuge Contamination is Difficult to Confirm and Clean Up,"
(GAO,  1987), EPA began to work cooperatively with the U.S. Fish and Wildlife
Service to develop methods for deriving National wildlife criteria.   The
wildlife criteria efforts carried by the Initiative Committees have been

-------
182    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

coordinated with the on-going National efforts. Because the effects on
wildlife may not be accounted for in other methodologies for deriving water
quality criteria, the proposed methodology was developed to address this
deficiency-  The proposed methodology contained both an effect and an exposure
component to account for varying species susceptibility to chemical
intoxication, and dissimilar feeding and drinking behaviors and natural
histories.

B.    Scope of Methodology

      1.    Proposal:  The proposed wildlife methodology was based largely on
the noncancer human health paradigm described in appendix C to part 132.  The
methodology, like the aquatic life and human health methodologies, included  a
two-tiered approach for deriving wildlife criteria and values, as described  in
the proposal  (58 FR 20878).  The main difference between the two tiers  in the
proposed Wildlife methodology was the extent of the data base needed to
generate a final wildlife criterion or value.  In order to generate a Tier I
criterion, data frofh both mammals and birds were required.  Separate values
for mammals and birds were developed with the lower of the two values being
the final Tier I criterion.  In order to generate Tier II values, data  from
either mammals or birds were required.  This value was modified by an
interclass uncertainty factor  (UF) to account for sensitivities between
taxonomic classes, and was then applied as the Tier II wildlife value.
Because the rationale behind Tier II values was that they were to be typically
more stringent than Tier I criteria, due to the smaller data set available,  it
was expected that the magnitude of some of the UFs applied in the derivation
would be larger.

      2.    Discussion of Comments

      Comment:  Some commenters believed that the focus of the wildlife
methodology is misdirected because other factors, such as habitat destruction
and exotic species, are more important than toxic contaminants in affecting
species' viability in the Great Lakes System.

      Response:  EPA disagrees that the wildlife methodology focus is
misdirected.  As discussed in section I of this document, research on wildlife
species resident in the Great Lakes indicates that some wildlife populations
remain threatened in areas of high contamination by toxic chemicals.  In  the
Great Lakes, reproductive  impairment  of numerous wildlife species has been
correlated with  the presence of polychlorinated biphenyls  (PCBs), p,p'-
dichlorodiphenyltrichloroethane  (DDT) and its metabolites, and other
contaminants  (58 FR 20806).

      Because many wildlife  species are at  the top of the aquatic food  web,
current water quality criteria derived to protect fish may be inadequate  to
protect wildlife who consume contaminated fish.  Wildlife are especially  at
risk from chemicals which  biomagnify  because they are frequently exposed  to
very high levels of,these  contaminants because many species feed primarily
from aquatic food webs.  For this reason, emphasis was placed on selecting
piscivorous wildlife species  (i.e., those which eat fish) for the derivation
of wildlife  criteria as to represent  species likely to experience significant
contamination  through an aquatic  food web.  Wildlife species may also be
uniquely  susceptible to some chemicals, as  compared to aquatic species.

      EPA acknowledges that  other factors,  such as habitat destruction  and
introduction of  exotic species are  impacting species viability in the Great
Lakes System.  EPA has programs  in place that are attempting to  address many
of these  problems; however,  this  does not mean that toxic  chemical discharges
into the  Great Lakes basin are not  also adversely affecting species  viability.
As discussed in  section I, and in U.S. EPA  (1995b), effects of toxic chemicals
on wildlife  species  continue to  be  a  problem.

-------
                                Section VI: Wildlife                             183
      CommentB:  Many commenters expressed general concerns over the
scientific soundness of the proposed methodology.  In particular, some were
critical of the use of the noncancer human health paradigm as a model for
ecological risk assessment (which might include an assessment of all
ecological stressors on all components of the ecosystem).  Those commenters
were also concerned-that the effect component of the noncancer human health
paradigm methodology is based on individual-level, rather than population-
level measurement endpoints.   Some commenters recommended either delaying
implementation of the methodology, or making the methodology just guidance.

      Responses:  EPA agrees that the use of the noncancer human health
paradigm does not consider all potential stressors on an ecosystem.  However,
the final wildlife methodology is not designed to be a comprehensive model for
assessing all ecological risk to the entire Great Lakes ecosystem.  Instead,
the intent of the methodology is to initially focus attention on those avian
and mammalian species in the System which are likely to experience significant
exposure to contaminants through aquatic food chains.  The methodology
specifically excludes reptiles and amphibians because there is currently
insufficient toxicity data for these species and insufficient understanding of
exposure routes to estimate risks from toxic contaminants to include these
groups in the methodology.  While it would be better to provide a
comprehensive ecological risk assessment approach for chemical and non-
chemical stressors to the System, it is currently not possible given the many
data gaps and enormous resources required to develop such an approach.  As
explained below, EPA finds the noncancer human health paradigm to be
appropriate for the avian and mammalian species on which the method focuses.

      EPA considered not including the methodology in the final Guidance until
a more comprehensive multi-stressor risk assessment approach could be
designed.  However, this option was rejected because such a program will take
many years to develop and a sound wildlife methodology was available at this
time to provide protection to those species at greatest risk from persistent,
bioaccumulative pollutants.  EPA selected a reasonable approach to address the
adverse ecological effects from toxic contaminants in the Great Lakes System.
In addition, based on the results from the two National meetings and the April
1994 EPA Science Advisory Board  (SAB) (U.S. EPA, 1994a) commentary (discussed
below),  EPA concludes that the paradigm is a scientifically reasonable
approach to address'impacts from bioaccumulative compounds on avian and
mammalian species in the Great Lakes at this time.

      During the development of the wildlife methodology, EPA hosted two
public meetings held in December 1989 and April 1992,  (U.S. EPA, 1989, 1994b)
as part of a national effort to develop methodologies to protect wildlife
criteria.  During both of these meetings, there was general consensus that the
proposed Guidance methodology was fundamentally sound from bioaccumulative
contaminants.  In addition, other concepts developed for the national program
were extensively used in the development of the wildlife portion of the final
Guidance.

      Finally, EPA discussed the use of the proposed paradigm for developing
water quality criteria to protect wildlife with EPA's SAB in February 1992 and
April 1994  (U.S. EPA, 1992, 1994a).  The report from the February 1992 SAB
(U.S. EPA, 1992) meeting indicated concern with the wildlife criteria concepts
being formulated around the perceived requirements of the noncancer human
health paradigm, which might be inadequate for wildlife.  In response to the
SAB's commentary, EPA made several changes that were discussed in the proposed
Guidance  (see 58 FR 20882) .  The April 1994 SAB commentary (U.S. EPA, 1994a)
stated that, while the use of the noncancer human health paradigm for the
development of wildlife criteria is in the early stages of development, it
promises to be an innovative and valuable new method for understanding the
fate and effects of contaminants in the environment.  Based on the changes
made to the methodology in response to the February 1992 SAB commentary (U.S.
EPA, 1992) and the support for the use of the methodology expressed by the
report from the April 1994 SAB commentary  (U.S. EPA, 1994a),  the paradigm

-------
184    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

being pursued is appropriate for the  species and stressors EPA is currently
addressing.

      EPA also does not agree that the wildlife methodology focuses too
extensively on the protection of individuals.  The methodology focuses  on
population-level impacts by restricting the toxicological measurement
endpoints on which .a criterion is based to those that are likely to adversely
affect populations.  If these toxic responses were observed in wildlife
populations in the Great Lakes System, the continuation of breeding
populations of the wildlife species could be jeopardized.  The SAB  (U.S. EPA,
1994a) endorsed the basis of the wildlife criteria on the protection of
wildlife populations from the direct  effects of chemical stressors.

      There are two distinctions between the noncancer human health paradigm
and the wildlife approach used in the proposal.  Because the wildlife approach
is designed to protect populations and not individuals, the wildlife paradigm
does not include an intraspecies UF (although exceptions can be made in cases
where toxicological or exposure data  suggest that species listed pursuant to
section 4 of the Endangered Species Act will not be protected by system-wide
criteria) to ensure.better protection of toxicologically sensitive members of
a given population.  Further, the selection of toxicological endpoints  in the
wildlife methodology is restricted to gross endpoints likely to adversely
affect population dynamics  (i.e., reproductive or developmental effects).
This approach is consistent with the  recommendation from the SAB  (U.S.  EPA,
1994a), but is different from the human health methodology which focuses on
the effects on individuals.  This is  illustrated by comparing the Human Health
criteria documents with the DDT section in the final Wildlife Criteria
Documents.  Both documents reference  the same study, but the same dose  level
considered a No Observed Adverse Effect Level  (NOAEL) for wildlife criteria
derivation, based on reproductive endpoints, is cited as a Lowest Observed
Adverse Effect Level  (LOAEL) in the noncancer human health criteria document,
based on liver lesions as an endpoint.  Examples of acceptable endpoints are
made available in the final Wildlife  Criteria Documents, as well as in  the
proposal (58 FR 20882).

      Comments:  The use of Tier II wildlife values was generally criticized
by many commenters' who believe there  were technical weaknesses with the
methodology and because of insufficient data being used to derive these
values.  Some commenters were concerned with the resource requirements  to
derive wildlife criteria or values for the entire universe of pollutants.
Some commenters supported the use of  Tier II values for wildlife.

      Responses:  The methodology for deriving Tier II values is sound.  EPA
agrees, however, with commenters that the data for deriving wildlife values is
currently limited.  In addition, EPA  agrees with those commenters who
cautioned against advancing too rapidly with a new methodology before
additional field validation can be made.  Therefore, during review of comments
received on the proposal, EPA reconsidered the scope of application of  the
wildlife methodology.  EPA decided to limit the methodology to bioaccumulative
chemicals for which the determining route of exposure is through the diet.
EPA still agrees, however, that the methodology can be modified to derive
reasonable wildlife-values where other exposures become significant.  For non-
bioaccumulative chemicals, it may be  more appropriate to select different
representative species which are better examples of wildlife species with the
greater exposure for a given chemical.

      In addition, EPA decided to limit the methodology to require developing
only Tier I criteria for several reasons, including concerns that a Tier II
value, which is based on toxicity data from only one taxonomic class, could
not be protective of wildlife species in other taxonomic classes when there is
evidence of wide differences in sensitivities across classes.  Further, a Tier
II value is based on an interclass UF that  is not used in deriving a Tier I
criterion, making the uncertainty inherent  in the value potentially
unreasonably greater.   Finally, the proposed wildlife methodology was a new

-------
                                Section VI: Wildlife                             185
approach to ecological risk assessment for wildlife.  EPA believes, however,
that the taxonoznic class-specific wildlife values provide adequate protection
for species of that-class, even where toxicity data from other classes are
missing.

      3.     Final Guidance:  Based on the above discussion, the use of the
proposed methodology is appropriate and the methodology does not emphasize on
individual effects.

      However, EPA has decided to modify the scope of the final wildlife
methodology.  The final Guidance requires the States or Tribes to use the
methodology to derive Tier I numeric wildlife criteria for only the
bioaccumulative chemicals of concern  (BCCs)  listed in Table 6-A of part 132.
EPA considered making the methodology optional for all chemicals, as advocated
by some commenters,  but decided that the methodology was advanced enough to
use for those chemicals of greatest concern (BCCs) to the higher trophic level
wildlife species feeding from the aquatic food webs in the Great Lakes basin.

      For the development of Tier II wildlife values for all pollutants, use
of the proposed Tier II methodology is encouraged, but not required.  To
derive Tier I numeric wildlife criteria for chemicals not listed in Table 6-A
of part 132, the methodology contained in the final Guidance is also
encouraged, but not required.  While States or Tribes may develop and
implement additional Tier I criteria or Tier II values as deemed necessary,
any derived Tier I criterion remains subject to EPA review and approval at 40
CFR part 131.  In the event that the methodology is used to derive Tier I
criteria for pollutants not listed in Table 6-A of part 132 or to derive Tier
II values, States and Tribes are also encouraged to use the methodology for
deriving bioaccumulation factors (BAFs), described in appendix B to part 132.

C.    Effect Component

      As with the noncancer human health methodology, the wildlife methodology
consists of both an effect and an exposure component.  The effect component  is
determined by the toxicity data and the UFs used to account for uncertainties
in predicting an appropriate test dose (TD)  for wildlife species.

1.    Minimum Data Recruirements

      a.     Proposal:  The effect component of the proposed methodology was
defined by the KOAEL, which is the maximum concentration of the toxicant in
the food of the test species which did not cause adverse effects to those test
organisms.  A NOAEL is derived from published studies from which dose-response
curves can be developed.  EPA proposed that the NOAEL selected must be based
on studies of adequate length (i.e., a minimum of 90 days for mammals, and 28
days for birds) so that chronic effects may be reasonably expected to be
expressed.  The use of a LOAEL,  adjusted by an UF, could be used to derive
criteria where an acceptable NOAEL was not available.  A LOAEL is defined as
the lowest concentration of the toxicant in the diet of the test species which
produces an adverse effect on the test organisms.

      The acceptable observed endpoints in these toxicity studies were those
that are directly or indirectly related to maintaining viable wildlife
populations.  Examples of endpoints which would reasonably be expected to be
related to the reproductive or developmental success of the species included
the number of viable young per female, or hatching or whelping success.

      The proposal also established an order of preference for selecting the
appropriate NOAEL or LOAEL to be used to calculate individual wildlife values
for each taxon.  Field study data were preferred over laboratory data, but the
latter could be used in place of field study data if best professional
judgement deemed it to be of better quality. • Examples of circumstances where
laboratory data may be more appropriate than field study data include cases
where:  dose-response curves or cause-effect relationships cannot be

-------
186    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

established for the field study because of the design of the field  study;
because the effects of other stressors cannot be identified or controlled;
where laboratory data are determined  to be more consistent with  other
published data than are field data; or, where there are quality  control
concerns.

      b.    Comments:  Many commenters argued that the minimum exposure
durations required for toxicity studies were too short to adequately evaluate
the potential reproductive effects of chemicals.  Other commenters  either
supported or challenged the use of field studies in the development of
wildlife criteria.  Commenters in support argued that field studies provide a
reasonable estimate of the impact of  chemicals in a natural setting,
incorporating actual exposures and metabolic impacts of the chemicals.
Commenters opposed to the use of field studies believed that current studies
were not constructed to provide quality data or adequate controls to satisfy
the intent of the final Guidance.  In addition, EPA received comments
recommending that existing field studies be used to validate criteria as they
are derived.

      Responses:  The endpoints of concern, as defined in the proposed
methodology, were those expected to impact adversely the reproductive or
developmental success of a species.  The intent of establishing  the minimum
study duration was to limit the use of short-term acute toxicity information
because it may not fully reflect potential impacts on the endpoints of  concern
(i.e., reproduction and development), and, therefore, could result  in an
under-protective criterion.  EPA agrees that longer-term studies, including
multi-generational studies, are desirable and should be used where  available.
It is also important not to make the  study duration requirements so long that
most available data, which could be used to derive wildlife criteria or
values, was eliminated.

      The study duration of 90 days for mammals is consistent with  the minimum
requirements established in the 1980  Human Health National Guidelines  (45 PR
79347) and in the final Guidance for  developing noncancer human  health
criteria  (Appendix C.II.l to part 132).  In that guidance, "subchronic"
toxicity tests are defined as continuous or repeated exposures for  a period of
90 days, or approximately 10 percent  of a rat's lifespan.  EPA acknowledges
that the test species used for development of wildlife criteria  could have
significantly different life-spans than a rat, and therefore permit some
flexibility in the selection of an appropriate study.  However,  rat data are
allowed to be used in deriving wildlife criteria with the proper use of UFs;
therefore, it is reasonable to use a  minimum 90-day study duration  for
wildlife.

      The minimum study duration for  bird taxa was changed from  28  days to 70
days.  The 70-day period was selected to conform with established EPA test
protocols for reproductive effects on avian species, described in U.S.  EPA
(1986).  If a study evaluates impacts on growth or mortality of  chicks, post-
hatching, 28 days may be an adequate  exposure duration.

      EPA continues to support the use of field studies in the development of
wildlife criteria because such studies can be used to predict the impacts of
chemicals in the environment, by integrating food web, dietary preferences,
and metabolic considerations.  EPA cautions that both field study data  and
laboratory study data must be carefully reviewed and evaluated for  their
usefulness and adequacy in deriving wildlife criteria.  Where appropriate, and
based on best professional judgement, EPA supports the use of laboratory data
over field study data when it has been determined that the laboratory data are
of better quality and are more likely to predict impacts on wildlife species.
EPA also supports the use of any other appropriate data to validate not only
the derived criterion, but any and all UFs and exposure parameters  used in the
derivation of that criterion.

-------
                                Section VI: Wildlife                             187
      The final Wildlife TSD provides further guidance on the appropriate
selection of toxicity data, including proper evaluation of studies.

      c.    Final Guidance:  A Tier I criterion is based on toxicological data
of sufficient duration  (generally 90 days or more for mammals and 70 days or
more for birds) from which a dose-response curve may be developed.  Assuming a
dose-response or a cause-effect relationship can be established, field studies
remain the preferred type of data for deriving a TD, although laboratory data
continues to be allowed if it is determined to be of better quality and more
appropriate, or where field data are lacking.

2.    UFs

      a.    Proposal:  EPA proposed to allow the use of UFs to address
uncertainty in any extrapolation of toxicity data to an appropriate endpoint.
For chemicals lacking an acceptable NOAEL, the LOAEL could be substituted,
with the application of an UF (ranging from 1 through 10) to extrapolate to an
estimated NOAEL.  The minimum test durations for the use of LOAELs were the
same as for NOAELs  (28 days for birds, and 90 days for mammals); these differ
from the noncancer Human Health methodology for Tier I criteria, which
requires a LOAEL to be based on a one year or longer rodent study.  The one-
year requirement provides a level of  conservatism which is not needed for the
protection of wildlife populations, where species typically have shorter life-
spans and reproductive cycles (see 58 FR 20879).

      In cases where a chronic endpoint was not observed, but subchronic
effects were expressed which could reasonably be expected to lead to chronic
effects in a longer study, these data could be used with the application of a
subchronic-to-chronic UF.  The range proposed for this factor was 1 through 10
(see 58 FR 20879).

      Because the species tested may not necessarily reflect the sensitivities
of a representative wildlife species in the Great Lakes System, an additional
UF, the species sensitivity factor, could be applied to extrapolate to protect
species of greater sensitivity than the test species.  In the proposal, this
factor ranged from 0.01 through 1 and was applied as a multiplier, rather than
a divisor, as were the other UFs described above (see 58 FR 20880) .

      Finally, in the proposal (58 FR 20881) EPA requested comment on an
alternate formula.  This formula was functionally the same as the proposed
formula, except that each of the UFS was explicitly included in it.

      b.    Comments:  Several commenters considered the use of UFs
inappropriate because it could result in overly conservative criteria because
the factors are multiplicative in nature, thereby resulting in large
uncertainties being'applied.  Commenters also suggested that there is an
insufficient data set available to evaluate the appropriateness and range of
these factors.  Other commenters complained that there was insufficient
guidance to choose appropriate UFs in the ranges provided,  A number of
commenters expressed a preference for using the alternate equation described
in the proposal.  They believed this equation more clearly described the
application of the UFs in the methodology.

      Responses:  EPA believes that the use of UFs is appropriate.  Wildlife
UFs are needed for several reasons, including the limited amount of available
toxicological data,, test duration,  observed endpoints, and differences in the
sensitivities among wildlife taxa.   All of these factors introduce uncertainty
into the criterion derivation process, making the precise determinations of
the effect of ambient concentrations of specific chemicals on wildlife
populations difficult.

      EPA agrees that any UF must be applied >with careful consideration of the
magnitude of the factor; however, the ranges for the UFs proposed for the
wildlife methodology are reasonable.  In the proposal, the data to support the

-------
188    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


appropriateness and range of specific UFs was, in most cases, from work that
had been completed for human health.

      Subsequent to the proposal, EPA performed work that confirms the  ranges
of the UFs used for wildlife.  The analysis to support the range of
interspecies UFs of 1 to 100 involved examinations of acute and chronic
toxicity data..  The analysis compared median lethal doses for numerous  species
for a variety of chemicals.  The results indicate that approximately 90
percent of the median lethal doses among the species for the same chemical
tested were within a factor of 20.  The chronic toxicity analysis involved a
review of data for four chemicals  (including DDT and mercury).  It was
determined that 90 percent of the chronic endpoints for each species were
within two orders of magnitude of the corresponding NOAELs for the other
species tested.

      Regarding the basis of the LOAEL-to-NOAEL UF, results indicate that 95
percent of the ratios of LOAELs to NOAELs for birds and mammals are less than
10.  Finally, the. recommended range of the subchronic-to-chronic UF is
supported by previous reviews on the toxicity of chemicals to laboratory
mammals, and a new analysis of toxicity to birds.  In two separate reviews
more than 95 percent of the ratios of the NOAELs for subchronic exposures
(approximately three months) to NOAELs for chronic exposures (approximately
two years) were less than 10 (Weil and McCollister, 1963; McNamara, 1976).
More detailed information is provided in the final Wildlife TSD and in  U.S.
EPA  (1995b).

      Although a cumulative UF of 10,000 is possible, the range of the
combined UFs for the four chemicals listed in Table 6-A of part 132 range from
six to 10.  In comparison, the Human Health methodology, section V.C.4.b.ii,
estimates a likely maximum composite UF of 3,000.  Therefore, while it  is
possible to use a large cumulative UF for a chemical, in practice the
magnitude of the UFs in the final criteria are likely to be small.  EPA
considers it unlikely that a State or Tribe will derive cumulative UFs  that
are significantly higher.  To ensure that UFs remain reasonable, EPA
recommends that States or Tribes consider a cap on the maximum composite UF of
1,000 because composite UFs greater than 1,000 may indicate a level of
uncertainty that is unacceptable.

      EPA believes it has provided sufficient guidance on the selection of
UFs.  Even so, it is important to note that the selection of UFs will,  in many
situations, be based on best professional judgement.  EPA anticipates that the
Clearinghouse described in section II of this document will provide a forum
for assisting in selecting appropriate UFs for chemicals.

      EPA agrees that the alternate equation is preferred because  it
eliminates the potential confusion of having to multiply the TD by the  species
sensitivity factor, while at the same time dividing the subchronic-to-chronic
and LOAEL-to-NOAEL factors into the TD.  Instead, all UFs are divided directly
into the TD.  In addition, the term "species sensitivity factor" has been
changed to  "interspecies UF" to reduce potential confusion regarding its use.

      Several changes were made to the alternate equation as proposed.  The
intraspecies UF was removed, which was to be applied when additional
protection of the individual or more sensitive members of a species was deemed
appropriate.  Guidance for modifying the criteria to provide for this added
protection is provided in procedure 1 of appendix F to part 132, and in the
final Wildlife TSD.  In addition, the subscripts for the three remaining UFs
were changed to be consistent with U.S. EPA  (1991) .  Finally, several of the
representative species feed at two or more trophic levels, which was not
readily apparent in the proposed equation.  To more clearly characterize
uptake through food, the food ingestion rate for each trophic component has
been separately calculated, and these are presented in Table 1  (discussed
below).  Contaminant uptake through the food is calculated by summing all the

-------
                                Section VI: Wildlife                              189
products of the trophic level -specific food  ingestion  rates and  the
appropriate RAF.

      During its review of the UFs, EPA considered  including  in  the  equation
 (described in the final Guidance section, below) an allometric scaling factor
to derive doses which would be more toxicologically equivalent when
extrapolating from test animals to the representative  species, based on
differences between the species' body weights and metabolic rates.   This  is
because the current state of the science indicates  that  some  of  the
variability in sensitivity across species can be related to general
physiological and anatomical differences observed across organisms within the
same taxonomic class  (e.g., mammals) .  The rates of such processes,  such  as
basal metabolic rates, cardiac output, renal clearance,  oxygen consumption,
food consumption, and water consumption, tend to vary  across  species according
to allometric scaling factors that can be expressed as a non- linear  function
of body weight.  The relationship of these and other physiological processes
to toxicokinetics has led to the explicit use of allometric scaling  for
estimating more toxicologically equivalent doses in EPA's human  health cancer
methodology and its implicit use in the human health noncancer methodology.
EPA recommends that in the determination of  an interspecies UP,  States or
Tribes apply the equation below to assess the allometric scaling factor for
each representative species and to consider  that assessment as one component
in the determination of an appropriate interspecies UF.   This equation was
endorsed by EPA in 1992 (57 FR 24152) .  In the derivation of  wildlife
criteria, allometric scaling is useful in adjusting for  some  of  the
toxicokinetic differences across species.  However, it may not accurately
reflect the toxicokinetics of all chemicals nor encompass all the
toxicodynamic differences among species.  Therefore, in  determining  an
interspecies UF, allometrically- derived TDs  should  be  considered in
conjunction with chemical class -specific information on  sensitivity,
toxicokinetics, and toxicodynamics across species.  This is consistent with
the guidance provided in the SAB commentary  (U.S. EPA, 1994a) which  stated
that allometric relationships should not be  the sole basis for selecting  an
interspecies UF.


                                           fft  1/4
where :  TDR =  Test Dose scaled for the given representative species in
question  (mg/kg-d) . '
               TD for the test species (mg/kg-d) .

        WtT =  Body Weight of the test species (kg) .

        WtR =  Body Weight of the given representative species  (kg; presented
in Table 1) .

      c.    Final Guidance:  The use of the LQAEL-to-NQAEL, subchronic-to-
chronic, and interspecies UFs described in the proposal  (58 FR  20879)  remain
in the final Guidance, although they are applied as divisors in the equation,
consistent with the preferred alternate equation,  as well as for clarity as to
their application.  Their ranges remain as in the  proposal, 1 through  10,
except for the interspecies UF which ranges from 1 through 100  (the
mathematical inverse of its counterpart, the species sensitivity factor,
contained in the proposal) .

      In addition, the alternate equation, as modified in accordance with the
above discussion, has 'been selected for the final  Guidance:

-------
190    Water Quality Guidance for the Great Lakes System — Supplementary Information Document
                                UFA x UFS x UF,
                                  A     s     L
                                ** + £
Where:

      WV = Wildlife' Value in milligrams of substance per liter  (mg/L) .

      TO = TD in milligrams of substance per kilograms per day  (mg/kg-d)  for
the test species.  This shall be either a NOAEL or a LOAEL.

      UFA = UF for extrapolating toxicity data across species (unitless) .  A
species -specific DF shall be selected and applied to each representative
species, consistent with the equation.

      UFS = UF for extrapolating from subchronic to chronic exposures
(unitless) .

      UFL = UF for LQAEL to NOAEL extrapolations (unitless) .

      Wt = Average weight in kilograms  (kg) for the representative  species.

      W = Average daily volume of water consumed in liters per  day  (L/d)  by
the representative species .

      FTU = Average daily amount of food consumed from trophic level i in
kilograms per day  (kg/d) by the representative species .
            = RAF for wildlife food in trophic level i in liters per kilogram
 (L/kg) , developed using guidelines for wildlife presented in appendix  B  to
part 132, Methodology for Development of Bioaccumulation Factors.   For
consumption of piscivorous birds by other birds  (e.g., herring  gulls by
eagles) , the BAF is derived by multiplying the trophic level 3  BAF  for fish by
a biomagnification factor  (BMP) for biomagnification from fish  to birds.

D.    Exposure Component

      The proposed exposure component of the wildlife methodology consisted of
three general areas :  selection of the representative species ;  exposure
parameters for those representative species; and use of BAFs specific  for
wildlife diet.  The first two areas will be discussed below.  For a discussion
on the BAFs refer to section IV of this document.

1.    Representative Species

      a.    Proposal:  During the development of the proposed methodology, EPA
considered using a hypothetical model wildlife species on which to  base  the
derivation of wildlife criteria or values.  EPA decided, however, to use
actual Great Lakes System species representing various foraging behaviors at
upper trophic levels and, therefore, greater exposure than  other wildlife
species inhabiting the Great Lakes System.  In addition, the use of
representative species allowed a basis for deriving an appropriate
interspecies UF in cases where it is known that  there are more  sensitive
species than the species from which the NOAEL was  derived.  This would have
been difficult to determine if a hypothetical model wildlife  species was used.

      Five species were proposed as the representative species: two mammals,
the mink  (Mustela vison) and river otter  (Lutra  canadensis) ;  and three birds,
the belted kingfisher  (Cervle alcvon) , osprey  (Pandion haliatus) ,  and  bald
eagle  (Haliaeetus leucocephalus) .  These species were proposed based on

-------
                                Section VI: Wildlife                             191
geographic distribution, dietary habits, and the trophic level of their prey.
EPA did not consider routes of exposure other than through the aquatic food
web in selecting representative species because of the focus on
bioaccumulative pollutants.

      b.    Comments:  While some commenters expressed support for the
selection of the five representative species, other commenters argued that
these species are not ecologically representative of the Great Lakes wildlife
species because they may not be widespread throughout all parts of the Great
Lakes System, nor are they resident in all ecological habitats.  Other
commenters recommended replacing or adding species, such as gulls, cormorants,
or the raccoon.

      Responses;  EPA agrees that the proposed species selected may not be
ecologically representative of all the possible wildlife species in the Great
Lakes System, but they were not selected for this reason.  Rather, EPA's
intent was to select disparate species most likely to be exposed to
environmental contaminants from aquatic ecosystems to serve as surrogates of
wildlife species that are highly exposed to toxicants from the aquatic food
chain.  The representative species are not necessarily the most
toxicologically sensitive species, nor do they represent species most likely
to be exposed to bioaccumulative environmental contaminants from terrestrial
ecosystems, or through other routes of exposure.  It was not EPA's intent to
select species to account for every available niche.

      After reviewing comments which suggested the use of different
representative species, EPA decided to replace the osprey with the herring
gull  (Larus arcrentatus) based on a re-evaluation of the exposure parameters of
Great Lakes wildlife species, including species identified by commenters for
inclusion on the list  (e.g., raccoons, herring gulls, common terns, and
double-crested cormorants).  EPA first re-evaluated all five of the proposed
species to determine if they are truly representative of the species that are
the most likely to be exposed to environmental contamination through the
aquatic ecosystem.  EPA showed that, although the osprey is one of the more
highly exposed piscivorous birds, it is not one of the most exposed and that
the gull is potentially more exposed.  In addition, the osprey's foraging
behavior is similar to the bald eagle's and a bird with foraging behaviors
different from the kingfisher and eagle would be preferred.  The herring gull
was selected because its body weight is roughly in between the eagle and the
kingfisher, its food ingestion rate is greater than the osprey, and the
trophic levels at which it feeds are higher  (72 percent at trophic level 3, 18
percent at trophic level 4 and 10 percent terrestrial prey, compared to 100
percent at trophic level 3 for the osprey).

      EPA also considered replacing the kingfisher with the common tern;
however, the estimated exposure parameters for the two species were
essentially the same,  and either species could be used.  Distribution
attributes of the common tern that may make it appear a more representative of
the Great Lakes ecosystem do not affect the derivation of the a wildlife
criterion by the methodology in the proposed or final Guidance.  EPA,
therefore, retained the kingfisher as a representative of a small piscivorous
bird feeding entirely on trophic level 3 fish.

      No changes were made to the mammalian representative species because all
other appropriate species in the Great Lakes System have much lower rates of
consumption of aquatic organisms  (e.g., raccoon) than either the mink or
otter.  (See U.S. EPA, 1995a.)

      c.    Final Guidance:  Elements of the exposure component remain largely
the same as the proposal:  the use of five representative species, as
described in the proposal, with the replacement of the osprey by the herring
gull.  The focus on the food and water uptake routes of exposure remains as in
the proposal.

-------
192    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

2.    Exposure Parameters

      a.    Proposal:  The routes of  exposure to wildlife  considered in the
proposal were uptake through drinking the ambient water and uptake  through
dietary consumption of prey in the aquatic food web. EPA presumed that other
routes of exposure  (e.g., inhalation  or trans-dermal uptake)  were
insignificant for these two taxonomic classes where bioaccumulative chemicals
are concerned 'as in the case for the  final Guidance.   Exposure parameters were
derived for each representative species, consisting of body weight,  and food
and water ingestion rates.  In each case, the "average" individual  was
assumed.  Where the ingestion rates could not be determined from published
literature because of the lack of available data, use  of allometric equations
from Nagy (1987) were proposed.  Additional information on the food and water
ingestion rates selected may be found at 58 PR 21005.

      The degree of accumulation of a contaminant at different steps in the
aquatic food chain was determined through the application  of  BAF calculated
pursuant to the proposed BAF methodology  (58  PR 21022) .

      b.    Comments:  Several commenters criticized the use  of  the allometric
equations to determine food and water ingestion rates  because rates change
throughout the life of the organism;  others criticized the values of the rates
presented in the proposal.  Some commenters stated that the bald eagle diet
consists in part of herring gulls, and that the methodology should  be modified
to reflect that route of exposure, taking into account the proper trophic
level of the gull.

      Responses:  Where available, EPA continues to support the  use of
empirically-derived ingestion rates.  EPA conducted a  review  of  the available
literature to re-evaluate ingestion rates used for the representative species.
In this re-evaluation, EPA used empirical data  (where  available)  or allometric
equations to determine basal metabolic rates  for free-living  animals.  Food
ingestion rates were determined from  the basal metabolic rates so different
caloric contents of wildlife food could be considered. The revised exposure
parameters are presented in Table 1,  below.

-------
 g
•H
JJ
 U
 0)
 a
•0
 0)
•H
 4J
 C
 m
 d)
-H
 u
 01

 &
-H
4J
td
4J
 C
 0)
CO
0)
^
 ft
 d)

 u
 O
14-1

 m
 n
 0)
4J
 0)
 n)
 (X

 a)
 m
 O
1
rH >,
0) 0)

d) PI
i4
U O
•H
.c
g

fr1




0) £ iH
4J U 0)
rt ea >
Oi U 0)
C jjj
0 O U
-H (M -H
• I .-I
•*-' f*H
DQ ^^ QJ

0) O 4J
4J-rH «J
* 4J oi
S CO
Oi
Oi
c
H
3*4
*O CQ "rl
•a; CD
m
0)
-H
o
0)
a
CO





w
0)
-H
•o

o
JJ
c
0)
u
0)




1
Oi









a
•0
l_3




Ol

m
-•H
g







**

O O



m M
H) 0)
4J
O




a\ t*
in i-i
H H
h^ 0)
^J
O


rH
00
O
0




O
00 .
O
C
-H
s
^






0*0 <*>

O O
00 (N


ro ^j*
HH






P- 'J'
(Tt CN
O O


ro *»•
H EH




o
o
vo
0




"!
p-
Otter







*•

o
o
rH

PO
hi






OJ
p-
VO
o
o


m
H




P-
o
0




in
H
0
n
5
10
-H
IM
Oi
C
-H
^


*»*•* *

O O O O
CM 00 M H


m ^<
HJt3
•• EH H
£ M
01 d)
-H £
fc 4J



O P-
(N 00 VO
H O 0
0 O O


m if to

4J
o


m
vo
o
o




H

rH
rH
01
C
•rl
^
^4
0)
X
*>*•*> **.*.

MOO 00 O O
erv oo IN P- ro


n TC CO U
^ J O.-H
£ m id
m -a 3
-H W O1
fc -H BJ
ffl •
0
C
(T\ CO rH
rH CN 00 CN
P- O> tN H
ro o o o
O O O O


r*i 4< CQ M
r3 rl a (I)
H H .C

O


o
vo
H
O




VO

0)
rH
S1
0)

•o
rH
Iti
PQ


 OQ  CO     OQ
-rt-H    'O
IM IH     ^
       tn-H
m ^ TJ J3
       ^
rH rH -H  U
 0)  (IJ ja-H
 >  >     W
 v  a) m  nt
                                                                                                    U U  H  10
                                                                                                    •H -H  O   '
                                                                                                    J   -  -   c
                                                                                                    H  ^4  CO
                                                                                                    4J 4-> -H  H
                                                                                                           Oi
                                                                                                     II  II     to
                                                                                                           II  0)
                                                                                                    ro ^*    jr«

                                                                                                    H H So
                                                                                                    0)
                                                                                                    JJ

-------
194    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Because ingestion rate data, however, are very limited, EPA considers
the use of the allometric equations presented in the methodology to be
adequate when more specific measured values or more appropriate  allometric
equations are not available.  In addition, EPA also supports the use of the
estimation methods described in U.S. EPA  (1993a) or the information contained
in U.S. EPA  (1988) .  The States or Tribes can at their discretion use any of
the estimation methpds or the allometric equations to derive ingestion rate
data.

      In the sensitivity analysis sections for each pollutant in the proposed
Wildlife Criteria Documents, EPA considered the ingestion of mammals, non-fish
eating birds, and fish-eating birds by the eagle.  To define better the
magnitude of the ingestion of these prey, EPA conducted an analysis described
in the final Wildlife TSD and in U.S. EPA  (1995a), which characterized the
diet of all five representative species, including the eagle and other birds
in the Great Lakes basin.  Based on this analysis, the composition of the
eagle's diet for the Great Lakes basin was modified.  The method also
incorporates the use of chemical-specific biomagnification factors to account
for the accumulation of contaminants in piscivorous birds which  serve as prey
for the eagle.  The final Wildlife TSD and the final Wildlife Criteria
Documents further discuss this method for incorporating the biomagnification
factor into the criteria.

      c.    Final Guidance:  For use where ingestion data are lacking, the
final methodology retains the use of the allometric equations from Nagy  (1987)
or the estimation methods contained in U.S. EPA  (1993a).

      EPA also made.modifications to the body weight, and food and water
ingestion rate exposure parameters for each of the five representative
species.  These modifications were based on the continued review of the
technical literature, and as requested by commenters  (see Table VT-1) .

      EPA also reviewed the dietary composition of each of the five
representative species.  EPA found that aquatic organisms comprise virtually
all of the prey for two of the species:  the river otter and kingfisher;
however, the mink, herring gull, and eagle obtain a significant  portion of
their diets from mammals and birds.  The contribution of these aquatic
organisms to the uptake of the contaminant from the aquatic system by these
three species is reflected through an appropriate weighting of the energy
intake through the aquatic and terrestrial components of the diet, and by
assuming zero uptake of the contaminant through the terrestrial  component of
the food chain.

      To properly account for the exposure through the food chain, the portion
of fish-eating prey in the diet must be quantified, and an appropriate
adjustment factor  (a BMP) for the fish-eating birds  (prey) must  be determined.
The derivation and application of the BAF and BMP, and examples  of how these
calculations can be made are described in the final Wildlife TSD and  in  the
final Wildlife Criteria Documents.

E.    Protection of Individual Members of a Population

      1.    Proposal:  The proposed methodology also made allowance for  the
protection of individuals in cases where  decreasing population size or density
threatened the continuing existence of the  species in the Great  Lakes System.
An intraspecies UF, ranging from  1 through  10 could be applied in the effect
component of the methodology, in  a manner  similar to the LOAEL-to-NOAEL,
interspecies, or  subchronic-to-chronic UFs.

      2.    Comments:  Several commenters were  concerned that the wildlife
criteria did not allow enough flexibility in making site-specific
modifications to existing criteria.  Some  commenters argued that the  criteria
should be waterbody-specific, and other  commenters believed that site-specific
modifications less restrictive than the  System-wide criteria  should be

-------
                                Section VI: Wildlife                             195
allowed.  EPA's response to this issue is addressed in section VIII.A. of this
document.

F.    Wildlife Criteria

      1.    Proposal:  The proposed Guidance presented Tier I wildlife
criteria for four pollutants:  mercury (including methylmercury); PCBs;
2,3,7,8-tetrachlorodibenzo-p-dioxin  (TCDD); and DDT and its metabolites.  A
discussion of the TD, UFs, and other assumptions were fully described in the
Wildlife Criteria Documents.  No Tier II values were derived in  the proposal.

      2.    Discussion:  The four Tier I criteria listed in the  proposal have
been modified based on reconsideration of the applicable UFs, on revised
exposure parameters for the final five representative species, and
recalculation of the BAFs pursuant appendix B to part 132 of the final
Guidance.  Additional changes from the proposal, including changes to the UFs
and the BAF used to account for uptake from piscivorous birds fco the eagle are
discussed below..

      In general, values of one, three, or 10 were used as the UFs, to reflect
relative degrees of conservatism.  Theoretically, the UFA could range to 100,
but that did not occur in the derivation of these four criteria.  A value of
one for any of these UFs indicates that there is reasonable certainty that the
TD portrays the expected toxicological condition (i.e., the result is a true
chronic condition with an adequate duration, the test species is reasonably
expected to be among the most sensitive,  and the endpoint reflects a
reasonable NQAEL); a value of 10 indicates that there is significant
uncertainty that the TD portrays the expected toxicological condition; and a
value of three indicates an intermediate level of certainty.

      Additional justification for the toxicological and exposure data used  in
the derivation of the four wildlife criteria may be found in the final
Wildlife Criteria Documents.

      a.    Mercury.

      Comments:  Several commenters stated that the proposed mercury criterion
of 180 pg/L is significantly below natural background, and therefore, is
erroneous.  These commenters cite natural background levels averaging 1 ng/L,
with upper bounds of near 7 ng/L.  Other commenters stated that  the proposed
mercury criterion is too stringent and that it is orders of magnitude below
the level of detection.  These commenters believed compliance with the
criterion would be impossible.  Some commenters believed that the avian LOAEL
for mercury should be based on the ingestion rates of the test animals, not
the ingestion rates of the controls.

      Responses:  EPA revised the numeric criterion for mercury  to reflect
adjustments made to the exposure parameters for the five final representative
species and modifications to the BAFs  (particularly the trophic  level 3 BAF).
A re-evaluation of the interspecies UF for the avian class was also made.  The
resultant value is 1300 pg/L, approximately the same as the natural background
concentrations cited by commenters.

      EPA conducted a study to determine the natural background  levels of
mercury (U.S. EPA,  1993b).  In general, there is an unfortunate  lack of
reliable data because the global transport of mercury from anthropogenic
sources has created concern over whether pristine areas receive  significant
loads of mercury from atmospheric deposition into freshwater ecosystems.

      Data contained in Noreheim and Forslie (1978), Wren (1983), and Vermeer
et al.  (1973) indicate a BMP range of three to 12 between fish and piscivorous
birds (i.e.,  bird prey species for the eagle).   The value 10 was selected as
being a reasonable BMP.

-------
196    Water Quality Gujdance for the Great Lakes System - Supplementary Information Document

      The dose-response curve from the studies from which the avian TD  was
based (Heinz, 1974, 1975, 1976a, 1976b, and 1979) indicate that  the TD  level
was close to the NOAEL.  In addition, the UFL selected in the proposal was
two.  For these reasons the UFL remains at two.   The UFA  was  changed for all
three representative species from 10 to three because reconsideration of the
ranges of sensitivities of various avian species to wildlife indicates  that
the mallard is relatively sensitive, but not the most sensitive  species. -The
avian LOAEL was modified from 64 /zg/kg-day to 78 /zg/kg-day to accurately
reflect the ingestipn rate of the test animals, rather than the  controls as
had been done in the proposal.

      The mammalian DFs were not changed from the proposal.  EPA notes  that
the UFS  was set  to 10  because of the test length;  in the  study selected
(Wobeser et al.f 1976) there was histopathological effects observed at  the  low
doses and the authors indicated that if the exposure period was  longer  death
may have resulted.

      EPA considered the concern on the level of detection in ambient systems,
and the impact on compliance monitoring by discharging facilities.  The final
Guidance sets forth-an appropriate mechanism to describe reasonable compliance
goals in cases where the criterion is below the level of detection.  Section
303(c)(2)(A) of the Clean Water Act states that water quality standards are to
be adopted which are protective of the uses designated for the waters
affected.  In compliance with this provision, aquatic life criteria for metals
have been promulgated in recent years by several states  in the Great Lakes
System that are more restrictive than the level of detection.  Implementation
of the water quality standards, however, does take into  account  the ability to
detect the pollutant in the waste stream.  Procedure 8 of appendix F to part
132 provides that the water quality-based effluent limit must be derived from
the water quality criterion; compliance with that limit, however, may be based
(at the State's discretion) on the level of quantification, defined in
procedure 8 of part 132.

      EPA agrees that the ingestion rates of the test animals, not the
controls should be used in deriving a LOAEL for the avian value.  The LOAEL
was revised to account for the different ingestion rate.

      b.    DDT.

      Comments:  Several commenters argued against the use of the Anderson  et
al.  (1975)  study from which the avian wildlife value for DDT was derived.
They argued that a dose-response curve could not be generated from the  data
presented in that study.  Another commenter argued that  the different
toxicities of the various DDT congeners should be accounted for  in the  Tier I
criterion.

      Responses:  The Anderson et al.  (1975) study used  to derive a NOAEL
represents an adequate basis from which to derive an effect level for use in
deriving a wildlife water quality criterion for DDT and  its metabolites.  The
study analyzed DDT concentrations in the food items of West Coast pelicans
over several years and  correlated those concentrations to reproductive
effects, including hatching success, egg shell thinning, and fledgling
success.  From these data a dose-response curve can be developed, with  the
appropriate application of a UFL to account for uncertainties in extrapolating
to a NOAEL and for uncertainties in time lags for eliminating the toxicant
from the parents.

      The Anderson et al.  (1975) study did not differentiate the effects  of
DDT and those of its metabolites, and therefore, the avian value is based on
total DDT  (DDT, DDE, and DDD) at a near  steady-state condition  in an  aquatic
system.  It is important to account for  the  differences  in bioaccumulation  of
each metabolite through the food web  and establish appropriate  BAFs  for the
DDT mixture in the Great Lakes.  Based on data from Oliver and  Niimi  (1988),
EPA derived composite DDT and metabolite BAF values for  aquatic trophic levels

-------
                                Section VI: Wildlife                             197
3 and 4, based on weighting the BAF for DDT, DDE, and ODD  (derived for the
dissolved fraction), in accordance with the fraction of each compound in the
Great Lakes fish species  (see section IV).  The values for the two BAFs for
trophic levels 3 and 4, respectively, are 1,687,000 and 9,357,000 L/kg.

      Data contained in Braune and Norstrom (1989) indicate a value of 85 for
a BMP for DDE for trophic level 3 fish to herring gulls is a reasonable, "yet
conservative value.  The BMP used in the calculation for DDT and metabolites
was 63  (see appendix K of the GLWQI TSD for the Procedure.to Determine
Bioaccumulation Factors for a discussion on how the BMP of 63 was derived).
This value was obtained from weighting the proportions of DDT, DDE, and ODD in
the tissues of the organisms, and was used in assessing that portion of the
eagle's diet that is comprised of piscivorous birds.  Additional information
may be found in the final Wildlife Criteria Documents.

      The avian UFL was changed from 10 to three because of the relatively low
level of response noted in the study as the LOAEL  (Anderson et al., 1975).  A
value of one is inappropriate because there is a clear indication of an
adverse impact at the TD.  The UFA was also changed from 10 to one for each of
the representative species because published data indicate that the test
species used, the pelican, is among the most sensitive birds in terms of
available LOAELs, particularly when compared to the three avian representative
species.

      The mammalian test dose value was changed from 0.5 mg/kg/day to 0.8
mg/kg/day, based on a recalculation of the food ingestion rate of the test
animals.  Further the contaminant used in the mammalian study was DDT, not a
mixture of DDT and metabolites.  Therefore, the BAFs used to derive the
mammalian wildlife value were based on DDT only, not the mixture weighted BAF
used for birds.  The mammalian UFs were not changed from the proposal.

      The criterion'is reported for DDT and its metabolites because the
criterion is based on the avian wildlife value, which used a test dose that
included DDT, DDD, and DDE.

      c.    PCBs.  Data contained in Braune and Norstrom  (1989)  indicate a
value of 90 for a BMP between fish and piscivorous birds  (i.e., bird prey
species for the eagle)  is reasonable, yet conservative.

      The UFL for birds was changed from 10 to three because a re-evaluation
of the dose-response relationship suggested that a factor of 10 was overly
conservative.

      The UFS for mammals  was adjusted from 10 to one because the duration for
the study selected (Aulerich and Ringer, 1977) was 300 days which was
considered of sufficient length for manifestation of chronic effects.

      d.    2.3.7.8 TCDD.  Data contained in Braune and Norstrom (1989)
indicate a value of 30 for a BMP between fish and piscivorous birds (i.e.,
bird prey species for the eagle) is reasonable, yet conservative.

      The avian UFS was adjusted from one to 10 because the test  duration of
10 weeks (Nosek et al., 1992) was only a small fraction of the reported half-
life for TCDD elimination in non-egg laying adult pheasants  (U.S. EPA, 1993c).
The UFA was  adjusted from 10 to one for each representative species because
other species showed only slightly greater sensitivities than the test
organism,  particularly when compared to the three avian representative
species.

      No changes were made to the values selected for the UFs for mammals.
EPA notes that the UFA  value for each representative species was  selected
based on a comparison' of single-dose lethality data for the rat and mink.

-------
198    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      3.     Final Guidance:  The final values for each of the four Tier I
criteria, modified as discussed above,  'are presented in Table 2.  States and
Tribes must adopt criteria for these four pollutants that are no less
stringent than EPA;s final criteria.  Additional information may be found in
the final Wildlife Criteria Documents.

G.    Comparison of Wildlife Criteria and Methods to National Program and to
      Great Lakes Water Quality Agreement

      The proposal at 58 FR 20884 contained a discussion of the relationship
of this methodology to both the National Program and to the Great Lakes Water
Quality Agreement (GLWQA).  The final Guidance largely supports that
discussion, including the statement that the four wildlife criteria continue
to be more restrictive than existing aquatic life values.

      The proposal described that the wildlife criterion for DDT and
metabolites at 0.87 pg/L was more stringent than the corresponding GLWQA Annex
1 value of 3.0 pg/L.  The final criterion is now 11 pg/L, which is slightly
less stringent than'the Annex 1 value.   For the reasons described in the
aquatic life section of this document,  EPA continues to believe that the DDT
criterion adequately conforms to the GLWQA.

H.    References

      Anderson, D.W., J.R. Jehl, R.W. Risebrough, L.A. Woods, L.R. Deweese,
and W.G.  Edgecombe.  1975.  Brown pelicans:  improved reproduction off the
southern California coast.  Science 190:  806-808.

      Aulerich, R.J. and R.K. Ringer.  1977.  Current status of PCB toxicity
to mink,  and effect on the reproduction.  Arch. Environ. Contam. Toxicol.  6:
279-292.

      Braune, B.M. and R.J. Norstrom.  1989.  Dynamics of organochlorine
compounds in herring gulls:  III.  Tissue distribution and bioaccumulation in
Lake Ontario gulls.  Environ. Toxicol.  Chem.  8:  957-968.

      Heinz, G.H. ' 1974.  Effects of low dietary levels of methyl mercury on
mallard reproduction.  Bull. Environ. Contam. Toxicol.  11:  386-392.3

      Heinz, G.H.  1975.  Effects of methylmercury on approach and avoidance
behavior of mallard ducklings.  Bull. Environ. Contam. Toxicol.  13:  554-564.

      Heinz, G.H.  1976a.  Methylmercury:   second-year feeding effects on
mallard reproduction and duckling behavior.  J. Wildl. Manage.  40(1):  82-90.

      Heinz, G.H.  1976b.  Methylmercury:   second-year reproductive and
behavioral effects on mallard ducks.  J. Wildl. Manage.  40(4):  710-715.

      Heinz, G.H.  1979.  Methylmercury:  reproductive and behavioral effects
on three generations of mallard ducks.  J.  Wildl. Manage.  43:  394-401.

      McNamara, B.P.  1976.  Concepts in Health Evaluation of Commercial and
Industrial Chemicals.  In;  Mehlman, M.A. et al.,  (eds).  Advances in Modern
Toxicology, Vol.1, Part  1.  New Concepts in Safety Evaluation.  John Wiley and
Sons, New York.

      Nagy, K. A.  1987.   Field Metabolic Rate and Food Requirement Scaling in
Mammals and Birds.  Ecological Monographs.   57(2):   111-128.

      Noreheim, G. and A.  Forslie.   1978.   The degree of methylation and
organic distribution in  some birds  of prey.  Acad. Pharmacol. Toxicol.  43:
196-204.

-------
                                Section VI: Wildlife
                       199
      Nosek, J.A., S.R. Craven, J.R. Sullivan, J.R. Olson, and R.E. Peterson.
1992.  Metabolism and disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin in
ring-necked pheasant hens, chicks, and eggs.  J. Toxicol. Environ. Health.
35:  153-164.

      Oliver, E.G. and A.J. Niimi.  1988.  Trophodynamic analysis of
polychlorinated biphenyl congeners and other chlorinated hydrocarbons in the
Lake Ontario ecosystem:  Environ. Sci. Technol.  22:  388-397.

      Stephan C.E., D.I. Mount, D.J. Hansen, J.H. Gentile, G.A. Chapman, and
W.A. Brungs.  1985.  Guidelines for deriving numerical National water quality
criteria for the protection of aquatic organisms and their uses.  Office of
Research and Development.

      U.S. Environmental Protection Agency.  1986.  Hazard Evaluation
Division:  Standard Evaluation Procedure -- Avian Reproduction Test.  EPA
540/9-86-139.

      U.S. Environmental Protection Agency.  1988.  Recommendations for, and
documentation of biological values for use in risk assessment.  NTIS-PB88-
179874.
      U.S. Environmental Protection Agency.  1989.
Protect Wildlife Resources.  EPA/600/3-89/067.
Water Quality Criteria to
      U.S. Environmental Protection Agency.  1991.  General Quantitative Risk
Assessment Guidelines for Noncancer Health Effects.  EPA Document No. ECAO-
CIN-538.

      U.S. Environmental Protection Agency.  1992.  An SAB Report:  Evaluation
of the Guidance for the Great Lakes Water Quality Initiative.  EPA-SAB-
EPEC/DWC-93-005.

      U.S. Environmental Protection Agency.  1993a.  Wildlife Exposure Factors
Handbook, Volumes I^and II.  EPA/600/R-93/187a and b.

      U.S. Environmental Protection Agency.  1993b.  Assessment of Mercury
Occurrence in Pristine Freshwater Ecosystems.  Draft.

      U.S. Environmental Protection Agency.  1993c.  Interim Report on Data
and Methods for Assessment of 2,3,7,8-tetrachlorodibenzo-p-dioxin Risks to
Aquatic Life and Associated Wildlife.  Office of Research and Development.
EPA/600/R-93/055.

      U.S. Environmental Protection Agency.  1994a.  Advisory on the
Development of a National Wildlife Criteria Program.  EPA-SAB-EPEC-ADV-94-001.

      U.S. Environmental Protection Agency.  1994b.  Draft Proceedings of the
National Wildlife Criteria Methodologies Meeting.

      U.S. Environmental Protection Agency.  1995a.  Trophic Level and
Exposure Analyses for Selected Piscivorous Birds and Mammals.  Volumes 1 and
3.  Draft.

      U.S. Environmental Protection Agency.  1995b.  Technical Basis for
Recommended Ranges of Uncertainty Factors used in Deriving Widlife Criteria
for the Great Lakes"Water Quality Initiative.  Draft.

      U.S. General Accounting Office.  1987.  Wildlife Management:  National
Refuge Contamination is Difficult to Confirm and Clean Up.  Gaithersburg, MD.
GAO/RCED-87-128.

      Vermeer, K.,  F.A.J. Armstrong, and D.R.M. Hatch.  1973.  Mercury in
aquatic birds at Clay Lake, Western Ontario.  J. Wildl. Manage.  37:  58-61.

-------
200    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Weil, C.S. and D.D.  McCollister.   1963.  Relationship Between Short- and
Long-term Feeding  Studies  in Designing'an Effective Toxicity Test.
Agricultural and Food Chemistry.   11(6):  486-491.

      Wobeser, G.f N.D.  Nielsen,  and B. Schiefer.  1976.  Mercury and mink:
II.  Experimental  methyl mercury intoxication.  Can. J. Comp. Med.   40:   34-
45.

      Wren, C.D.,  H.R. MacCrimmon,  and B.R. Loescher.  1983.  Examination of
bioaccumulation and  biomagnification of metals in a precambrian  shield lake.
Water, Air, and Soil Pollut.   19:  277-291.  1983.

-------
                                 Section VI: Wildlife
201
Table  2.   Final Tier  I  Wildlife Criteria.
Pollutant
mercury (including methylmercury)
PCBs
2,3,7,8-TCDD
DDT and metabolites
Criteria (/ig/L)
1.3 x Id'3
7.4 x 1CT5
3.1 x 1CT9
1.1 x ICr5

-------
202     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
                             Section VII: Antidegradation                         203
                  VH.  ANTIDEGRADATION


A.    General Discussion/Background

      This document explains the intent of the final Guidance on
antidegradation and.EPA's expectations of the Tribes and States in the
development and implementation of antidegradation policies and procedures.
This document also includes guidance to Tribes and States in the
implementation of the agreement entitled "A Bi-National Program to Restore  and
Protect the Lake Superior Basin." (September, 1991)   Tribes and States may
extend additional protection to the waters of the Lake Superior Basin through
the use of the special designations contained in the final Guidance if they so
choose.

1.    History of the Great Lakes Antidegradation Guidance

      The history of the Federal policy on antidegradation is discussed in
detail in the preamble of the proposed Guidance; readers are referred there
for a complete discussion of the origins and evolution of the Federal
antidegradation policy.  EPA's Water Quality Standards Handbook (September,
1993) also contains useful background information.

      Current national policy is found in the Federal regulations  at 40 CFR
131.6 and 40 CFR 131.12.   Federal regulations at 40  CFR 131.6 specify that
Tribal or State water quality standards must include an antidegradation
policy.  Federal Regulations at 40 CFR 131.12 identify the elements of an
acceptable antidegradation policy.  The Federal antidegradation policy is
composed of three levels of protection commonly referred to as tiers.  The
first element identified at 40 CFR 131.12(a)(1)  protects the minimum level  of
water quality necessary to support existing uses and applies to all waters.
This element establishes the ultimate limit on the extent to which water
quality can be lowered in a water body.  Lowering of water quality to the
point where existing uses are impaired is prohibited.  The second  element is
found at 40 CFR 131.12(a)(2), and protects water quality where water quality
is better than that needed to support fish and aquatic life and recreation  in
and on the water.  Where these conditions exist, the water body is considered
high quality and water quality must be maintained and protected unless
lowering water quality is necessary to support important social and economic
development.  The third element at 40 CFR 131.12(a)(3)  involves the protection
of water quality in water bodies that are of exceptional ecological,  aesthetic
or recreational significance.  Water quality in such water bodies,  identified
as Outstanding National Resource Waters (ONRW),  must be maintained and
protected.

      The protection of high quality waters under antidegradation  causes
considerable confusion and controversy.  It is often interpreted incorrectly
as an absolute prohibition on lowering of water quality in high quality
waters.  Such a prohibition would amount to a "no growth" policy which is not
consistent with EPA"1 s position as expressed in the regulations. Neither the
existing Federal regulations nor the final Guidance  prohibit activities that
would lower water quality in high quality waters.  The antidegradation
provisions contained in the final Guidance provide a structure for the
systematic evaluation of activities that are expected to lower water quality.
Implementation of antidegradation allows Tribes and  States to arrive at a
decision that considers all the available information regarding the social,
economic and environmental impacts of lowering water quality.   Review of such
activities under a Tribe's or State's antidegradation policy is intended to
ensure that any lowering of water quality is necessary,  that the lowering of
water quality is minimized and that desirable economic and social  benefits

-------
204    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

accrue to the area affected by the lowered water quality as a result of the
lowering of water quality.

      The applicability of antidegradation to nonpoint sources is also a
source of confusion.  EPA policy is that water quality standards, including
antidegradation, are applicable to any activity that might affect water
quality (see "Interpretation of Federal Antidegradation Regulatory
Requirement," memorandum from Tudor Davies, Director, Office of Science and
Technology, to Water Management Division Directors, dated February 22, 1994) .
However, despite the broad applicability of water quality standards,
mechanisms to implement water quality standards may not exist in all
circumstances.  Neither the antidegradation provisions contained in the final
Guidance,  nor existing regulations, confer any additional authority upon
States, Tribes or EPA to regulate nonpoint sources of pollution.  However,
where independent regulatory authority over nonpoint sources exists requiring
compliance with water quality standards, compliance with the antidegradation
provisions of the final Guidance is required.

      From the inception of the joint State-EPA effort to develop consistent
water quality standards for the Great Lakes, the participants recognized
antidegradation guidance was an essential element of such an effort.  The
Great Lakes Critical Programs Act of 1990 also recognized this by requiring
that the final Guidance published by EPA include antidegradation provisions at
least as stringent as existing Federal requirements.

2.    Summary of the Proposed Guidance

      The proposed Guidance consisted of four components each of which were to
be adopted by Tribes and States:

      a.    An Antidegradation Standard;
      b.    Antidegradation Implementation Procedures;
      c.    An Antidegradation Demonstration; and
      d.    An Antidegradation Decision.

Each of the components is discussed in detail below.

a.    The Antidegradation Standard

      As proposed, the Great Lakes antidegradation standard was derived from
the existing Federal antidegradation policy at 40 CFR 131.12.  The proposed
Guidance provided additional detail on antidegradation to assist Tribes and
States in implementing the standard and to  encourage consistency across the
basin.

      The protection of existing uses contained in the proposed Guidance
refers to the definition found at 40 CFR 131.3.  The intent of this reference
was to make clear that water quality necessary to support existing uses,  as
well as designated uses, is protected under antidegradation.  The proposed
Guidance prohibited lowering water quality  if, as a result, either existing or
designated uses wouj.d be impaired.

      The proposed Guidance and existing regulations at 40 CFR 131.12 provide
similar protection for high quality waters.  The proposed Guidance expanded
upon Federal regulations, however, in that  it specified that high quality
waters should be identified on a parameter-by-parameter basis, whereas
existing Federal regulations do not specify a particular method  for
identifying high quality waters.  The parameter-by-parameter approach was
selected for the proposed Guidance because  it was  considered the most
reasonable and workable way of identifying  high quality waters and because it
ensured that water quality improvements would be protected.  Also,  a
parameter-by-parameter  approach allows a water body to be considered high
quality for one pollutant even if  the criteria for another is exceeded,  as is
the case in many water  bodies within the Great Lakes basin.  Finally, this

-------
                             Section VII: Antidegradation                          205
approach is consistent with how waters are assessed and regulated in other
aspects of States' and Tribes' water quality programs.

      The final major element of the standard is the protection of Outstanding
National Resource Waters  (ONRWs).  The standard contained in the proposed
Guidance is identical to existing Federal regulations.

b.    Antidegradation -Implementation Procedures

      The implementation procedures contained in the proposed Guidance were
intended to provide direction to States and Tribes on how the antidegradation
standard should be applied.  The implementation procedures included
definitions of key terms and descriptions of how the Tribes and States were to
carry out the requirements of the Great Lakes antidegradation standard.  The
proposed Guidance stated explicitly that designated uses must reflect existing
uses and that where designated uses are impaired, water quality may not be
lowered at all with respect to the pollutant or pollutants causing the
impairment.

      For high quality waters, the proposed Guidance defined the concepts of
de minimis lowering of water quality and significant lowering of water
quality.  The implementation procedures recognized that considerable variation
was possible in the effects of different activities on water quality.  The
proposed Guidance included a mechanism for distinguishing between activities
based on the extent to which water quality was anticipated to be lowered.
Small reductions in water quality were identified as de minimis and not
subject to antidegradation review, whereas larger reductions were identified
as significant and subject to antidegradation review.  Both concepts were
intended as mechanisms to allow Tribes and States to differentiate between
activities that are likely to have an inconsequential effect on water quality
and those that are likely to have significant effects and to focus their
efforts on those that are of the most consequence to water quality.  The
definition of de minimis in the proposed Guidance provided criteria to be used
by Tribes and States to identify when a lowering of water quality could be
considered de minimis.  The proposed Guidance also provided criteria for
identifying when an activity would result in a significant lowering of water
quality subject to' antidegradation review.

      The proposed Guidance recognized that the significance of a potential
lowering of water quality depended not just on the magnitude of the lowering
of water quality, but also on the types of pollutants involved.  In keeping
with the emphasis of the Guidance on BCCs, the proposed Guidance established a
more restrictive threshold of significance for BCCs than for non-BCCs.  The
concept of existing effluent quality (EEQ) was specified in the proposed
Guidance as the means to implement antidegradation for BCCs.  EEQ served as
the basis for either permit limits or notification requirements which, if
exceeded, triggered antidegradation review.  Each of these concepts is
discussed in greater detail below.

      It should also be noted that for high quality waters and ONRWs, the
implementation procedures in the proposed Guidance allowed short-term and
temporary lowering of water quality without review under the antidegradation
provisions of the Guidance.  This is consistent with existing national policy
as described in USEPA's Water Quality Standards Handbook (August, 1994) and
allows Tribes and States greater flexibility in how they use the
antidegradation procedures as regulatory tools.

c.    Antidegradation Demonstration

      When an action is considered that could lower water quality in a high
quality water,  Federal regulations require that a determination be made that
the action is necessary to support important Social and economic development
{40 CFR 131.12(a) (2)}.  The proposed Guidance explained the process for
evaluating such activities and the information a discharger considering such

-------
206    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

an action must provide to the regulatory agency making the determination.  The
antidegradation demonstration process consisted of two sub-demonstrations:

1.    a demonstration that the significant lowering of water quality is
      necessary and,

2.    a demonstration that the significant lowering of water quality will
      support -important social and economic development.

Under the proposal, the two demonstrations were intended to be performed
sequentially, with performance of the social and economic development
demonstration contingent on a successful demonstration that the significant
lowering of water quality was necessary.  To assess the need for a significant
lowering of water quality, a person proposing an action that would lower water
quality would first determine whether or not existing treatment, pollution
prevention, additional treatment or some combination within a defined cost
range could avoid the need to lower water quality.  If this was the case, the
significant lowering of water quality would not be permitted.  If not, the
discharger would be required to demonstrate that important social benefits,
economic benefits or both would accrue to the community as a result of the
activity responsible for the significant lowering of water quality.

      The proposal also included special requirements applicable to RCRA and
Superfund actions.  In addition, requirements for implementing optional
special designations for Lake Superior were included.

d.    Antidecrradation Decision

      The proposed Guidance described how Tribes and States should evaluate
the data provided by a discharger in making a decision concerning whether or
not to allow a significant lowering of water quality.  The specifics of this
section are discussed in greater detail below.

B.    Overview of the Final Guidance

      The final Guidance includes Great Lakes-specific antidegradation
requirements for BCCs only.  A lack of specific requirements in the final
Guidance for non-BCCs should not be construed as relieving States and Tribes
of their responsibilities under the CWA and Federal regulations to adopt water
quality standards consistent with the CWA and Federal regulations.      As
stated at 40 CFR 131.6, antidegradation provisions consistent with 40 CFR
131.12 are compulsory elements of any State's or Tribe's water quality
standards; States and Tribes must adopt antidegradation policies and
implementation procedures just as they must adopt designated uses and
criteria.  For non-BCCs, the antidegradation policy and implementation
procedures developed by States and Tribes must define key terms,, specify what
types of information will be required to demonstrate that a significant
lowering of water quality is both necessary and will support important social
and economic development, how that information will be used to arrive at a
final decision regarding a request to lower water quality, and how public
participation will be factored into the final decision consistent with the
antidegradation standard and 40 CFR 25.  This SID includes EPA's
recommendations as to how antidegradation should be implemented for non-BCCs.

      For BCCs, States and Tribes must adopt specific provisions consistent
with the antidegradation standard and implementation elements contained  in
appendix E.  Special requirements for BCCs are necessary because of the
demonstrated sensitivity of the Great Lakes  System to such pollutants.
Consequently, imposing these requirements on Great Lakes States and Tribes is
warranted to protect the shared resources of the Great Lakes System.

      A second major change to the final Guidance is that the final Guidance
does not require the use of existing effluent quality  (EEQ)-based limits to
implement antidegradation for BCCs.  EEQ-based limits are not included in  the

-------
                             Section VII: Antidegradation                          207
final Guidance because EPA believes that it is possible that the use of EEQ-
based limits to implement antidegradation could have unintended effects
contrary to the goals of the CWA and the intent of antidegradation.

      The final Guidance retains the overall structure of the proposed
Guidance, that is, it contains an antidegradation standard, implementation
procedures, demonstration requirements, and a final decision.  Very few
changes were made tb the antidegradation standard between the proposal and the
final Guidance.  An implementation framework, consisting of implementation
procedures, demonstration requirements and the requirement for a final
decision is provided in the final Guidance.  However, many of the details
contained in the proposed Guidance, such as specific benchmarks for assessing
affordability of alternative treatment technologies, are included elsewhere in
this document as guidelines only.

C.    Detailed Discussion of the Final Regulation

1.    The Antidegradation Standard

      The antidegradation standard contained in the final Guidance differs
little from either the proposed Guidance or existing regulations on
antidegradation at 40 CFR 131.12.  The section pertaining to the protection of
high quality waters was reworded to cross-reference the definition of high
quality waters at 40 CFR 131.3 (See discussion below).  States and Tribes are
required to adopt an antidegradation standard consistent with the Standard
contained in appendix E applicable to increased loadings of BCCs to the Great
Lakes System.  States and Tribes may adopt the antidegradation standard as
applicable to all waters and pollutants.  As required by 40 CFR 131.6, water
quality standards adopted by a State or Tribe must include an antidegradation
policy consistent with 40 CFR 131.12.

2.    Antidsgradation Implementation Procedures

      The antidegradation implementation procedures were the second major
component of the proposed Guidance.  The implementation procedures consisted
of definitions of key terms and the processes to be used by Tribes and States
in applying antidegradation.  The proposed Guidance provided procedures for
protecting existing uses, high quality waters and ONRWs.  In addition, the
implementation procedures included methods for applying optional special
protection designations to waters in the Lake Superior Basin.

a.    De Minimis Lowering of Water Quality

i.    Background

      The proposed Guidance made provision for identifying certain small
increases in loading as de minimis and not subject to the requirements for
antidegradation review.  The "de minimis test" identified three criteria to be
used in classifying an increased loading as de minimis.  These were:

1.    only non-BCCs will be released as a result of the proposed activity
      responsible for the anticipated lowering of water quality;
2.    the proposed lowering of water quality uses less than 10 percent of the
      available assimilative capacity; and
3.    for pollutants included in 40 CFR 132.2, Table 5, at least ten percent
      of the total assimilative capacity remains unused following the lowering
      of water quality.

The proposal also required that any decision to allow a de minimis lowering of
water quality comply with the requirements in the TMDL section of the
implementation procedures to maintain a margin of safety (MOS).  Finally, the
proposal specified how and when the total and unused assimilative capacity of
a water body should be calculated.

-------
208    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      EPA's intent in including the de minimis test in the proposed Guidance
was to recognize that certain activities, although they may result in some
lowering of water quality, will not lower water quality to such an extent as
to result in a significant lowering of water quality.  EPA's goal in allowing
States and Tribes to identify certain increases as de minimis was to provide a
means of reducing the administrative burden on all parties associated with
activities of little or no consequence to the environment.  De minimis
determinations were made at the discretion of the State or Tribal Director;
the Director could choose to treat as significant any lowering of water
quality that would otherwise be de minimis where the circumstances of a
particular discharge warrant more thorough review.

ii.   Discussion of Significant Comments

      EPA received a number of comments on the proposed de minimis provisions.
Significant comments are discussed below.  For responses to specific comments,
see the comment response document in the docket of this rulemaking.

      Comment:  Some commenters stated that the de minimis provision is too
broad and should be"narrowed to include fewer activities.

      Response:  De minimis provisions are not part of existing regulatory
requirements, nor are they authorized for BCCs under the final Guidance.  For
non-BCCs, States and Tribes may include de minimis provisions in their
antidegradation policy at their discretion.  Although de minimis provisions do
involve non-conservative assumptions, the de minimis provisions included in
the proposed Guidance are not likely to seriously undermine the protection
afforded a high quality water body through antidegradation.De minimis
provisions provide a means for States and Tribes to differentiate between
actions that will result in an increased loading of a pollutant to a receiving
water that is likely to have a significant impact on water quality and those
that are unlikely to do so and focus review efforts on actions that will
degrade water quality.  It is reasonable to assume that loading increases of
non-BCCs that will use less than  ten percent of the remaining assimilative
capacity in a water body will have a negligible effect on ambient water
quality.

      Comment:  Some commenters felt that the de minimis provisions are too
narrow and should be broadened to reduce the number of activities subject to
antidegradation review.

      Response;  Given that existing Federal regulations are silent on the
concept of de minimis and that allowing any de minimis provision involves non-
conservative assumptions about the effects of increased loadings on water
quality, an expanded de minimis provision would be incompatible with the
antidegradation standard that States and Tribes are required to adopt under 40
CFR 131.6.  De minimis provisions must be narrow in order for antidegradation
to function as it was intended.   Some activities or actions have a negligible
impact on water quality and may be exempted appropriately from antidegradation
review through a de minimis provision; however, allowing loading increases
that use more than ten percent of the remaining assimilative capacity of a
water body to be considered de minimis strains the credibility of a State's or
Tribe's antidegradation implementation procedures and increases the chance
that a significant lowering of water quality could occur without
antidegradation review.

      Comment:  The de minimis provisions should be extended to include BCCs.

      Response;  EPA does not agree that even small increases in the loadings
of BCCs to the Great Lakes Basin  can be considered de minimis.  Low levels  of
BCCs in the Great Lakes have adverse impacts on the organisms that inhabit
them.  Further, because BCCs are  both resistant to degradation  and
hydrophobic, they tend  to accumulate in sediments and biota, amplifying  their

-------
                             Section VH: Antidegradation                          209
effects.  For these reasons, even small increases in loadings of this type of
pollutant must be considered significant.

      Comment:  There is no need to reserve a portion of the unused
assimilative capacity as a MOS for either the pollutants included in the final
Guidance or those listed on Table 5 and excluded from the Guidance.

      Response:  EPA does not agree with this comment.  A MOS is essential to
preserve high quality waters in their high quality state.  It is inconsistent
with the intent of antidegradation to allow such waters to be degraded without
scrutiny to the point where criteria are only just achieved.  Consequently, a
de minimis provision must include a MOS to ensure that such a situation does
not occur.  When a State or Tribe chooses to define de minimis as a percentage
of the remaining assimilative capacity, there will always be an opportunity
for a de minimis increase, albeit a smaller and smaller one.  Absent a MOS
requirement, this could lead to a situation where the water quality in a water
body is reduced to the point where the water body is effectively no longer
high quality, without antidegradation review.  To prevent this, a MOS similar
to the requirements contained in the final Guidance should be a part of a
State's or Tribe's antidegradation implementation procedures if the procedures
include a de minimis provision.

      Comment:  Many commenters felt that EPA should either do more to account
for the cumulative effects of multiple de minimis lowerings of water quality
or, conversely, that special protection against such cumulative effects is not
necessary.

      Response:  EPA agrees with those commenters concerned over the
cumulative impacts of a State's or Tribe's de minimis provisions.  MOS
requirements ensure that in no case will the water quality of a high quality
water body be degraded to the point where the water body is no longer high
quality without review and should be a part of a State's or Tribe's
antidegradation implementation procedures.

      MOS requirements, however, address only the more extreme cumulative
effects of de minimis increases.  Several commenters expressed concern that a
single discharger could request numerous de minimis increases resulting in a
significant impact on water quality that would not be subjected to
antidegradation review.  Similarly, several dischargers to the same water body
could receive permission for de minimis increases leading to a significant
lowering of water quality.  Such concerns are inherent in any decision to
allow for de minimis provisions.  However, the benefits of the de minimis
provisions in allowing States and Tribes to reduce administrative burdens and
prioritize their efforts outweigh the possibility for abuse, especially given
the MOS requirements and Tribal, State and EPA oversight of water quality
program implementation.  Abuse of the de minimis provisions can also be
prevented if States and Tribes adopt a definition of significant lowering of
water quality that allows the Director to identify any lowering of water
quality as significant on a case-by-case basis.

iii.  The Final Guidance

      Because the final Guidance does not address non-BCCs, it also does not
make provision for de minimis lowering of water quality.  Where a State or
Tribe wishes to include a de minimis provision in the State's or Tribe's
antidegradation policy and procedures for non-BCCs, EPA recommends adopting an
approach based on the proposed Guidance.  Any de minimis provision should be
based on a percentage of the unused assimilative capacity to protect against
over-allocation of the water body. In addition, the de minimis provision
should include a MOS based upon a percentage of the total assimilative
capacity such that once more than a certain percentage of the total
assimilative capacity is used,  any further lowering of water quality is
subject to antidegradation review.  If a TMDL is in place, total assimilative
capacity and loading capacity are functionally equivalent; it is not necessary

-------
210    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

to recalculate the total assimilative capacity of a water body where loading
capacity has already been calculated.  'Unused assimilative capacity must
account for background loading and should be determined based on permitted
rather than actual discharge levels at point and nonpoint sources.  As written
in the proposal, unused assimilative capacity could have been overestimated if
background loadings were not considered or if dischargers to the water body
were discharging below permitted levels.   Finally,the unused assimilative
capacity must be recalculated prior to each request for a lowering of water
quality to ensure that it reflects current conditions in the water body.

b.     High Quality Waters

i.     Background

      As explained above, existing Federal regulations at 40 CFR 131.12
establish a framework for antidegradation consisting of three levels of
protection.  The second level creates the category known as high quality
waters and defines high quality waters as those water bodies with water
quality better than minimum levels necessary to support fish and other aquatic
life and recreational uses.  Federal regulations provide that where water
quality is better than that needed to support propagation of fish, shellfish,
and wildlife and recreation in and on the waters, that water quality shall be
protected unless lowering of water quality is necessary to support important
social and economic development.

      The proposed Guidance followed closely the provisions contained in the
Federal regulations.  In addition, the proposed Guidance specified that the
level of protection afforded a water body under antidegradation be determined
on a parameter-by- parameter basis, considering each individual pollutant
separately from the others present in a water body.  Under the proposed
Guidance, a discharger contemplating an action that would result in an
increased loading wpuld identify the constituents of its effluent that would
increase as a result of the action.  Then, the ambient level of the pollutants
of interest would be determined and compared to the applicable criteria.
Where ambient concentrations of the pollutants in question are less than
criteria concentrations, the water body would be considered high quality for
those pollutants and increases in those pollutants would be subject to the
requirements applicable to high quality waters.

      EPA selected this method of identifying high quality waters for the
proposed Guidance for several reasons.  First, it establishes a clear, easily
evaluated criterion for determining whether or not a water body is high
quality.  The proposed method is also consistent with existing EPA guidance on
the subject of identifying high quality waters for purposes of antidegradation
(see "Application of Antidegradation Policy to the Niagara River," memorandum
from Martha G. Prothro, Director, Office of Water Regulations and Standards,
to Richard L. Caspe, Director, Water Management Division, Region II, dated
August 4, 1989, available in the docket for this rulemaking) and the current
approach taken by many States in the Great Lakes basin to implementing
antidegradation.  Also, the parameter-by parameter approach ensures internal
consistency between the mechanisms used in the final Guidance to identify
impaired waters and those used to identify high quality waters.  Finally,
identifying high quality waters on a parameter-by-parameter basis protects
high quality water wherever  it exists in the basin, even when individual
criteria may not be met and ensures that the water quality of the Great Lakes
System will not degrade below that necessary to support fish and other aquatic
life and recreation in and on the waters.

ii.   Discussion of Significant Comments

      Comment;  A number of  commenters stated that factors other  than just the
ambient pollutant concentration should be considered in determining whether  or
not a water body is high quality.

-------
                             Section VII: Antidegradation                          211
      Response:  EPA does not agree with this comment as a general matter.
The language of the,final Guidance conforms with existing Federal regulations
that refer to ambient water quality as the measure of whether or not a water
body is high quality.  In addition, ambient water quality provides a
quantitative measure for identifying high quality waters with regard to a
specific pollutant that is compatible with implementation through a permit to
discharge or other control document.  Finally, in general, the water bodies
found in the Great Lakes basin are, or should be, able to support fish and
aquatic life and recreational uses.  Therefore, where water quality is better
than necessary to support fish and aquatic life and recreation, that water
quality should be protected.

      Comment:  Some commenters stated that the parameter-by-parameter
approach to identifying high quality waters should be required for BCCs only.

      Response:  The commenter does not provide any compelling reasons for
employing a different approach to identify high quality waters for non-BCCs
than is used for BCCs.  On the contrary, the rationale for using a parameter-
by-parameter approach is equally valid for both BCCs and non-BCCs.  While the
final Guidance requiring the pollutant-by-pollutant approach pertains only to
BCCs, EPA believes that consistent with current EPA policy, States and Tribes
should implement antidegradation on a parameter-by-parameter basis for non-
BCCs, and identify high quality waters on this basis.

      Comment:  The identification of high quality waters for purposes of
antidegradation should consider the attainability of fishable/swimmable uses
as well as ambient water quality.

      Response;  Given that final Guidance applies only to increased
discharges of BCCs, and given the properties of BCCs, it is not reasonable to
assume that there exists anywhere in the Great Lakes System a water body where
an increased loading of BCCs does not have the potential to ultimately affect
water quality in the a water body where fish and aquatic life uses and
recreation in and on the water occur.  Consequently, there is no basis for
distinguishing between different water bodies within the Great Lakes System
for purposes of antidegradation.  In order the protect the Great Lakes System
as a whole, any water body where water quality is better than the minimum
necessary to support fish and aquatic life and recreation in and on the water
is considered high quality and subject to full protection under
antidegradation.

      Attainability of a use should not be confused with impairment of a use.
In some instances, a designated use may be impaired due to pollution resulting
from historic activities, such as fish consumption advisories due to PCB
contamination,  and yet the use may be considered attainable with control of
the contamination.  Identifying high quality waters on a parameter-by-
parameter basis provides Tribes and States the flexibility to both encourage
restoration by prohibiting increased loadings of the pollutant causing the
impairment, and at the same time provide antidegradation protection for
otherwise high quality waters.  As a result, water quality problems will be
kept from worsening through the protection of existing and designated uses
while new water quality problems will be prevented through the protection of
high quality waters.

      Comment:  The. definition of high quality waters should recognize that
some waters that meet the requirements on a parameter-by-parameter basis may
not be high quality resources.  The final Guidance should allow for less
rigorous review for activities affecting such waters.

      Response:  As stated above, any release of BCCs to the Great Lakes
System has the potential to affect water quality throughout the system,
because of their persistence and tendency to'bioaccumulate.  In order to
maintain the Great Lakes System as a whole in a high quality condition, all
water bodies within the system must be treated as high quality.  However, EPA

-------
212    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

agrees with the concerns raised by this comment with respect to non-BCCs.  EPA
may approve implementation procedures for non-BCCs which include an exemption
from the requirements for identifying high quality waters on a parameter-by-
parameter basis for non-BCCs for water bodies that are of limited ecological,
recreational and aesthetic significance.

iii.  The Final Guidance

      Much of what was included in the proposed Guidance is incorporated in
the final Guidance.  For BCCs, the final Guidance requires States and Tribes
to identify high quality waters on a parameter-by-parameter basis.  For
pollutants other than BCCs, EPA recommends that, in general, high quality
waters be identified on a parameter-by-parameter basis in the same manner as
for BCCs.

      The final Guidance includes one additional change to improve its
consistency and clarity.  A detailed definition of high quality water is
included in the definition section.  In the proposed Guidance, the definition
was found only in the antidegradation standard.

c.    Lake Superior Basin - Outstanding International Resource Waters

      The final Guidance includes a definition of the term "Lake Superior
Basin - Outstanding International Resource Waters."  This term, which was not
defined in the proposed Guidance, is defined in the document, "A Bi-National
Program to Restore and Protect the Lake Superior Basin,"  (September, 1991)
available in the docket to the rule.  A definition is included in the final
Guidance in response to comments requesting the inclusion of the definition.

d.    Outstanding National Resource Water

i.    Background

      The proposed Guidance included a definition of Outstanding National
Resource Water (ONRW).  The definition was derived from existing Federal
regulations at 40 CFR 131.12(a)(3).  The definition specified that designation
of ONRWs is at the discretion of States and Tribes.  The definition also
included examples of the types of waters that might be afforded ONRW status by
States and Tribes.

ii.   Discussion of Significant Comments

      Comment:  Some commenters stated that the proposed Guidance broadened
the definition of ONRW beyond existing Federal regulations.  The commenters
stated that the examples of types of waters that could be considered for ONRW
designation could encompass any water body within the Great Lakes basin and
that this could lead to delays in the permitting process as disagreements
arose over whether or not a water body was or should be an ONRW.

      Response:  EPA does not agree with this comment.  The final Guidance
does not contemplate any additional requirements beyond existing Federal
regulations for ONRWs.  The types of waters subject to listing were included
in the proposal as examples of what States and Tribes should consider  in
designating ONRWs.  Identification of an ONRW is inherently site- and  case-
specific and depends not only on  the water body but also on its  importance as
a resource to the State or Tribe.  Any cold water stream may be  an ONRW where
such resources are rare; elsewhere, where they may be more common, the
presence of a cold water community may not  in itself be sufficient to  warrant
designation as an ONRW.  Designation  of ONRWs occurs through the normal water
quality standards review process  and  should not delay permit issuance.

-------
                             Section VII: Antidegradation                          213
iii.  The Final Guidance

      The content of the final Guidance was not changed from the proposal.
The definition included in the final Guidance represents accurately what  is
intended by an ONRW designation.

e.    Significant Lowering of Water Quality

i.    Background

      The proposed Guidance identified criteria for determining when a
lowering of water quality is significant.  For all regulated discharges of
BCCs, any increase in loading to a high quality water, measured as a change  in
EEQ, was considered significant and subject to antidegradation review.  For
point source discharges of non-BCCs to high quality waters, a significant
lowering of water quality was defined generally as an increase in applicable
permit limits greater than a de minimis increase.  For nonpoint sources of
non-BCCs, a significant lowering of water quality was defined generally as an
increase in the rate of mass loading authorized by the governing nonpoint
source program.  Increased mass loadings of non-BCCs that would not change the
concentration of the pollutant outside of a designated mixing zone were not
required to undergo antidegradation review.  Finally, there was a "safety
provision" that allowed the Director to consider any action as significant on
a case-by-case basis.

ii.   Discussion of Significant Comments

      Comment:  Several comments stated that there is a need for a succinct
definition of an action in the definition of a significant lowering of water
quality.

      Response:  EPA takes the term "action" or "activity" to have a broad
meaning with respect to determining what may or may not be subject to
antidegradation.  Although the final Guidance includes examples of actions and
activities, it is not possible to anticipate each activity or action that
might result in an increased loading of pollutants or otherwise lead to a
deterioration of water quality.  Anything more than a broad definition could
constrain States and Tribes from requiring antidegradation reviews where
appropriate.  Therefore, EPA does not agree that including a definition of the
terms "action" or "activity" would be appropriate; States and Tribes need to
retain a measure of flexibility to adapt their antidegradation policies and
implementation procedures to the circumstances encountered in the day-to-day
operation of a water quality management program.  However, based on comments
received, EPA has determined that some commenters believe certain activities
to be potentially considered an "action" which EPA believes do not constitute
an action, accordingly, EPA has listed these activities in the final rule, and
noted that they are not subject to antidegradation review.

      Comment:  A number of commenters stated that it is inaccurate to assume
that an increase in.the loading of a pollutant to a water body will result in
a lowering of water quality in that water body.

      Response;  An antidegradation review is required whenever a
"significant" lowering of water quality is considered.  For BCCs, because of
their persistence and ability to exert profound impacts through accumulation
in the food chain, the final Guidance defines any increased loading of such
pollutants as a significant lowering of water quality.

      Comment;  A number of commenters stated that the "catch-all" provision
allowing any action to be considered significant is too broad.

      Response;  As stated above in the discussion of the term "action,"
States and Tribes need a certain amount of flexibility to be able to respond
to all the various types of situations that may arise as they attempt to

-------
214    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

implement this Guidance.  The "catch-all" provision gives them the ability to
consider whether individual actions that might lower water quality and are not
covered explicitly by the final Guidance are of concern, and to require an.
antidegradation review, as appropriate.  This flexibility is important and is
therefore retained in the final Guidance.  However, EPA has revised this
provision to clarify that it applies to deliberate activities.

      Comment r  A number of commenters stated that the term "significarit" is
ambiguous and should be deleted.

      Response:   The term "significant" does not stand alone, but is rather a
part of the term "significant lowering of water quality."  EPA has provided,
in the final Guidance, additional language to the definition of the term so
that the perceived ambiguity is minimized.

iii.  The Final Guidance

      The final Guidance differs from the proposed Guidance in several
important ways.   First, the EEQ provisions contained in the proposed guidance
for identifying significant lowering of water quality for BCCs were dropped in
favor of requirements that facilities provide notification and an
antidegradation demonstration in support of actions that are expected to
result in any increase in loading of BCCs prior to taking the action.
Although the underlying concept that any increase in loadings of BCCs to the
Great Lakes System warrants review under antidegradation remains intact, the
mechanism for triggering reviews was modified.  This change addresses comments
received and is discussed in greater detail in the section of this document
addressing antidegradation implementation.

      With respect to non-BCCs, the most significant change is that States and
Tribes are no longer required to adopt a definition of significant lowering of
water quality that is consistent with the definition contained in the final
Guidance  States and Tribes are only required to adopt definitions as needed
to support their implementation procedures, consistent with their
antidegradation standard and implementation procedures.  At a minimum, EPA
recommends that any activity that is expected to increase loadings of
pollutants such that relaxed permit limits are required, or which will result
in a new discharge of pollutants, should receive an antidegradation review.

      If a State or Tribe chooses to link antidegradation review to changes in
permit limits for pollutants other than BCCs, the State or Tribe must also
ensure that when an action at a point source increases the loading of a
pollutant that was not subject to a previous permit limit, that  increase is
subject to antidegradation review.  Examples of activities that  could receive
an antidegradation review include installation of a new source, a change in
process at an existing source that results in the addition of new pollutants
into the waste stream or the tie-in of a new industrial user to a municipal
treatment plant.  Similarly, for loadings of non-BCCs from nonpoint sources,
antidegradation review could be required when a new project seeks
authorization from the State or Tribe.  The revisions discussed above make the
implementation of antidegradation more equitable by subjecting both new and
existing dischargers to the same requirements.

      Antidegradation review is not required when a new limit is imposed as a
result of new monitoring data, improved monitoring techniques or a change in a
wasteload allocation or TMDL that results in no increase in loading or a de
minimis increase in loading.  Under the final Guidance, antidegradation review
for new limits is only necessary when there  is or will be a significant
lowering of water quality.

f.    Deleted Definitions

      Two definitions were included in the antidegradation section of the
proposed Guidance that are not included in the final Guidance.   These were  the

-------
                             Section VH: Antidegradation                          215
definitions of  "bioaccumulative chemical of concern" and "pollutant."
Definitions for these terms are included in the general definition section of
the final Guidance at section II B, making it unnecessary to include them in
appendix B.

g.    Implementation Procedures

i.    Background

      The proposed Guidance required that, for all waters, the level of water
quality necessary to protect existing uses be maintained.  The proposed
Guidance also specified that the uses designated by Tribes and States must
include existing uses.  Where water quality criteria for a particular
parameter are not attained, the proposed Guidance prohibited further
degradation of water quality for the parameters in question.

      For high quality waters, the proposed Guidance restricted when
significant lowering of water quality could occur to those instances where the
discharger demonstrates that the significant lowering of water quality is both
necessary and will support important social and economic development.
Different measures of significance were proposed for BCCs and non-BCCs.

      For BCCs, the proposed Guidance defined any increase in loading as
significant.  EEQ was selected to serve as the baseline against which
increased loadings would be measured.  EEQ was structured so as to track
effluent quality and could be reduced in subsequent permits to reflect load
reductions or increased to reflect loading increases approved under
antidegradation.  Tribes and States were given the option of either imposing
EEQ-based effluent limits in discharge permits or including provisions in
discharge permits requiring an antidegradation review if an EEQ value were
exceeded.  Tribes and States were also required to prohibit actions by
dischargers that would increase loadings of BCCs without prior approval of the
Director pursuant to section IV of the proposed Guidance.

      Where the proposed increase involved non-BCCs, a significant lowering of
water quality was defined as an increase in permit limits greater than de
minimis.  If a discharger was able to operate below permit limits such that an
increased loading from the discharger would not exceed existing permit limits,
no antidegradation review would be required.  Similarly, if the proposed
increase in permit limits was less than a de minimis amount, no
antidegradation review would be required.

      For ONRWs, the proposed Guidance stated that water quality must be
maintained and protected.  The proposed Guidance did not permit permanent
degradation of water quality in ONRWs.  Short-term and temporary degradation,
replacement of a failing septic system for example, could be permitted at the
discretion of the Director.

      The proposed Guidance also included provisions for Lake Superior if
Tribes and States wished to adopt special protection designations for the
lake.  Two special protection designations were included in the proposed
Guidance, Lake Superior Basin - Outstanding National Resource Waters (LS-ONRW)
and Lake Superior Basin - Outstanding International Resource Waters  (LS-OIRW).
The LS-ONRW designation could be used by Tribes and States to prohibit new or
increased discharges of certain listed pollutants, known as Lake Superior
bioaccumulative substances of immediate concern (BSICs), to portions of the
Lake Superior basin so designated.  The LS-OIRW designation could be used by
Tribes and States to require more stringent treatment for new or expanding
sources of BSICs.

      In addition, the implementation section included a description of the
circumstances under which the proposed implementation procedures would not
apply.  These included short-term and temporary lowering of water quality,

-------
216    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

bypasses not prohibited at 40 CFR 122.41(m) and certain actions under the
Comprehensive Environmental Response, Compensation and Liability Act.

ii.   Discussion of Significant Comments

      The majority of the comments received on the antidegradation
implementation, procedures addressed EEQ.  The comments received are summarized
below.

      Comment:  Use of EEQ is a disincentive to optimal waste water treatment
beyond the minimum necessary to achieve permit limits.  Dischargers that do so
will be "penalized" by receiving more stringent permit limits, whereas
dischargers that do the bare minimum receive less stringent effluent limits.

      Response;  EPA modified the final Guidance to address this concern.
Effluent limits are no longer required as the mechanism for implementing
antidegradation.  Instead, the final Guidance builds off of existing reporting
requirements and -requires an antidegradation review prior to commencement of
any activity that has the potential to result in an increased loading of BCCs
to waters within the Great Lakes System.  There are a number of advantages to
the mechanism included in the final Guidance for identifying when a
significant lowering of water quality will occur over what was in the
proposal.  First, the mechanism included in the final Guidance requires
consideration of antidegradation at the appropriate point in the process,
before the lowering-of water quality occurs and while the project is still in
the planning stages.  EEQ, on the other hand, serves a mechanism for
identifying failure to consider antidegradation and thus has a. punitive rather
than preventative function.  Second, unlike EEQ, the mechanism contained in
the final Guidance does not expose dischargers to enforcement actions solely
as a result of unusual effluent variability.  Also, by linking antidegradation
requirements to actions taken by a discharger, there is no danger of a
discharger being forced to undergo spurious antidegradation reviews to justify
apparent increases in loadings.  Finally, since antidegradation is independent
of effluent limits, there is no disincentive to efforts by dischargers to
optimize waste water treatment.

      Comment:  A significant number of commenters stated that use of EEQ
deprives dischargers of compliance benefits of improved waste water treatment.
Dischargers frequently achieve better effluent than is required by their
permit limits in order to ensure compliance with effluent limits and account
for process variability.  The EEQ provisions in the proposed Guidance
eliminated this option and exposed dischargers to greater enforcement
liability.

      Responses  Revisions to the implementation procedures in the final
Guidance address this concern.  Under the final Guidance, antidegradation
considerations for BCCs are uncoupled from  effluent limits.  As a result, if
dischargers wish to reduce loadings to improve compliance, they may do so
without a corresponding change in applicable effluent limits.  Consequently,
dischargers will be able to build as large  a margin between the limits
contained in their control documents and their actual effluent quality as they
choose.

      Comment:  EEQ is burdensome to dischargers and regulators.  The EEQ
requirements in the proposed Guidance will  necessitate  increased monitoring by
dischargers.  In addition, the need to analyze data and recalculate EEQ  at
each permit reissuahce will be difficult for regulators and slow the permit
issuance process.

      Response;  The final Guidance does not require calculation of permit
limits to implement antidegradation for BCCs.  Thus, no additional monitoring
or reporting is  likely to be necessitated  by the final  Guidance.  Although  the
final Guidance does specify monitoring requirements for discharges that  are
known or suspected  containing  BCCs,  this requirement should not be onerous  as

-------
                             Section VII: Antidegradation                          217
in most case, regardless of the antidegradation provisions, monitoring would
be necessary to assess compliance and provide data for States' and Tribes'
reasonable potential procedures.

      Comment;  Many commenters felt that the EEQ provisions create
uncertainty, making it difficult for dischargers to anticipate regulatory
requirements.

      Response:  The changes in the antidegradation implementation procedures
incorporated in the final Guidance should reduce the uncertainty of
dischargers regarding future discharge limits.  Dischargers will receive
limits for BCCs based on the reasonable potential and limit calculation
procedures adopted by States and Tribes consistent with the final Guidance.
For BCCs, antidegradation will be administered separately from numeric limits
on BCCs.  Further, the procedures contained in the final Guidance remove the
uncertainty associated with the proposed approach that resulted from limits
that became tighter over time as a result of antidegradation.  As a result,
dischargers will have greater certainty about what future control document
requirements will be.  For a discussion of the anticipated costs of
implementing the final Guidance, see the discussion of Costs and Benefits,
section IX of this document.

      Comment:  Commenters believed that monitoring associated with an EEQ-
based approach will be excessive.

      Response:  While it could be argued that the proposed approach would
have led to increased monitoring, under the final Guidance monitoring
requirements for dischargers should not be increased significantly as a result
of the antidegradation provisions.  In most cases, the requirement to monitor
for BCCs that is included in the final Guidance will be met through normal
monitoring to measure compliance with control document requirements and
support control document reissuance.  No unusual or more sensitive monitoring
is envisioned as a result of antidegradation.

      Comment:  A number of commenters stated that the EEQ provisions are
inconsistent with the TMDL process and the watershed approach.  Commenters
stated that the antidegradation procedures undermined the TMDL process and
EPA1s watershed protection goals by allowing limits more stringent than those
derived through the TMDL process, making the TMDLs irrelevant.

      Response:  EPA does not agree that antidegradation is incompatible with
TMDLs.  First, TMDLs and antidegradation address different and complementary
components of- the water quality program and both are necessary and required
for the protection of water quality in the Great Lakes System.  TMDLs, in the
simplest sense, establish the maximum amount of a pollutant that may be
discharged into a water body without exceeding water quality criteria
necessary to protect designated uses.  Antidegradation, on the other hand,
protects existing water quality where water quality is better than the
criteria.  Thus, conceptually, protection of high quality waters under
antidegradation is concerned with water quality that is better than the
minimum required by a TMDL.  Second, TMDLs and antidegradation are likely to
address different subsets of water bodies.  At present, TMDLs are only
required to be calculated for threatened or impaired water bodies and result
in imposition of water quality-based effluent limits sufficient to ensure
compliance with water quality standards.  In contrast, the effects of the
final antidegradation Guidance will be felt most strongly by dischargers to
waters that attain water quality standards and whose effluent limits are
technology-based.  Finally, unlike the proposed Guidance, in the final
Guidance, antidegradation is not implemented through effluent limitations for
BCCs.  In practice, antidegradation and TMDLs each reinforce the other, and
either one or the other of the two processes will be the important controlling
mechanism in a given water body for a given pollutant.

-------
218    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

iii.  The Final Guidance

      The final Guidance incorporates a number of changes in response to
comments received during the public comment period on the proposed Guidance.
These include revisions to the treatment of BCCs in high quality waters.

      EPA received a significant number of comments on the use of EEQ in the
proposed Guidance as a means of controlling loadings of BCCs.  Numerous
comments opposed the use of EEQ in the proposal on the grounds that it would
be a disincentive to optimal waste water treatment and would deprive
dischargers of any "cushion" they create between their effluent  limits and
actual effluent quality to ensure compliance.  However, other commenters
supported the reasoning in the proposed Guidance that led to the proposed use
of EEQ, namely that the environmental risk from BCCs is so great that any
increased loading should be considered significant.

      To reconcile these two positions, the final Guidance employs an approach
that, while not linked to specific, numeric limits, requires an
antidegradation review any time a discharger undertakes an activity that could
result in an increased loading of BCCs.  The new approach will be implemented
through notification requirements in a discharger's control document.
Antidegradation review will occur when an action is proposed that is likely to
result in an increased loading of BCCs.  No regulated facility would be
permitted to take an action that would result in an increased loading of BCCs
without completing an antidegradation demonstration and receiving approval
from the State or Tribe with jurisdiction over the affected waters.  Taking an
action without receiving approval would be a violation of the applicable
control document and would be subject to enforcement.   EPA selected this
approach because it eliminates perceived problems with EEQ and yet is
consistent with defining any increase in loadings of BCCs as a significant
lowering of water quality.  Also, it places greater emphasis on  the prevention
and minimization of any significant lowering of water quality and less on use
of antidegradation as punitive measure against dischargers.  As  envisioned by
the final Guidance, antidegradation is a tool to be used by States and Tribes
to make rational, well-supported decisions regarding activities  that affect
water quality.  For dischargers, antidegradation should be viewed as a reality
check on a proposed project, that all possible opportunities to minimize
impacts on water quality have been considered and implemented, as appropriate.
States and Tribes may employ EEQ as a means of implementing antidegradation if
they so choose.

      A further benefit of the approach taken in the final Guidance is that it
does not require quantifiable effluent concentrations for implementation.
Where the available data are insufficient to calculate a loading level but
BCCs are known to be discharged, as in the case where discharge  concentrations
are below levels of quantification, increased loadings of BCCs are controlled
by a requirement in the control document prohibiting a discharger from taking
any deliberate action that would increase loadings of BCCs without first
providing an antidegradation demonstration and receiving approval for the
increase from the Director.  Thus, where loadings of BCCs are occurring, but
are not quantifiable, increases are still subject to antidegradation.

      It should be noted that requiring a discharger to receive  approval prior
to taking an action that will result  in a lowering of water quality is not a
new requirement.  The intent of the antidegradation policy contained in the
Federal regulations at 40 CFR 131.12  is that antidegradation review of  an
activity that will lower water quality occur before the activity takes place,
when the antidegradation review will be most effective in identifying ways
that degradation of water quality may be minimized.

      For non-BCCs, EPA recommends that, at a minimum antidegradation review
be required any time a permit limit is made less stringent.  States and Tribes
must also ensure that the antidegradation policy and procedures  they adopt
address new sources and increasing sources where no permit limit currently

-------
                             Section VII: Antidegradation                          219
exists for a pollutant.  States and Tribes may also include a provision in
their procedures that allows them to review any action that is not
specifically addressed to provide the flexibility to respond to unforeseen
circumstances.

3.    Antideqradation Demonstration

a.    Background

      Existing Federal regulations at 40 CFR 131.12 require that water quality
in a high quality water be maintained and protected unless a lowering of water
quality is necessary to support important social and economic development.
The phrase "antidegradation demonstration" is used as shorthand for the
information gathered for review by the permitting agency to show whether or
not an action proposed by a discharger that will result in a significant
lowering of water quality both is necessary and will support important social
and economic development.

      The proposed Guidance required an antidegradation demonstration be
performed prior to any significant lowering of water quality.  The proposed
Guidance interpreted the Federal regulations at 40 CFR 131.12 as requiring two
distinct demonstrations; first, that the significant lowering of water quality
was "necessary," meaning the proposed activity could not occur without a
concomitant significant lowering of water quality, and second, that the
proposed activity would result in social and economic development.  Consistent
with Federal regulations, the proposed Guidance organized the antidegradation
demonstration into a hierarchy of three separate demonstrations to be made by
the discharger.  The discharger was required to demonstrate that the
significant lowering of water quality could not be reduced or prevented
through the application of prudent and feasible pollution prevention
alternatives, that the significant lowering of water quality could not be
prevented through the application of alternative or enhanced treatment within
a defined cost range and finally, if, as a result of the preceding
demonstrations, a significant lowering of water quality was shown to be
necessary, that the proposed activity would result in social and economic
development.  Each of the demonstrations included certain required elements to
be considered by dischargers conducting an antidegradation demonstration.

b.    Discussion of Significant Comments

      Comment:  Tribes and States lack the technical expertise required to
review antidegradation demonstrations.

      Response:  The final Guidance identifies the broad areas that should be
a part of any antidegradation demonstration.  This document identifies key
areas that should be considered in evaluating an antidegradation
demonstration, which will assist Tribes and States in their reviews.  Using
the final Guidance, Tribal and State water quality personnel should be capable
of determining whether or not an antidegradation demonstration prepared by a
discharger considers all of the essential elements.

      It should also be noted that preparing a demonstration may in and of
itself be useful to the proponent of an activity that would significantly
lower water quality.  In order to request a significant lowering of water
quality to accommodate a particular activity, a discharger must investigate
alternatives to lowering water quality.  This investigation may lead the
discharger to cost-effective alternatives to lowering water quality.

      Comment:  More detailed guidance is needed on all aspects of the
demonstration.

      Response;  In reviewing the comments received, EPA is convinced that it
is not possible to write guidance that would cover adequately every possible
situation.  In addition, including great detail would hinder efforts by Tribes

-------
220    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

and States to adapt the final Guidance to existing regulatory structures and
thereby slow its ability to respond to requests to lower water quality.
Consequently, the final Guidance provides a framework within which Tribal and
State programs must, operate with respect to BCCs.  This general  approach will
also prove more versatile as new circumstances arise that were not foreseen
when the Guidance was developed.  Further, general guidance allows Tribes and
States to tailor antidegradation reviews, including level of detail and
documentation,- to the specific circumstances encountered.

      Comment:  A number of commenters noted that the costs of pollution
prevention need to be considered.

      Response;  The final Guidance has been revised to require  that a
discharger identify "cost-effective" pollution prevention alternatives that
will eliminate or greatly reduce the extent of the significant lowering of
water quality.  Alternatives that would cause the proponent of the activity to
incur excessive costs would not be considered cost-effective; however,
pollution prevention can be the most cost-effective and environmentally benign
approach to protecting the environment.

      EPA's intent in having pollution prevention figure so prominently in the
antidegradation demonstration is to focus attention on alternatives that will
not lead to the release of pollutants to the environment rather  than on those
that depend upon treating the pollution after it is generated.   Such an
emphasis is consistent with the hierarchy outlined in EPA's National Pollution
Prevention Policy (see discussion in section I of this document).

      Comment;  Several commenters stated that the ten percent mandatory
expenditure requirement for enhanced and/or alternate treatment  is arbitrary.

      Response:  EPA agrees that the ten percent mandatory expenditure
requirement may limit the ability of States and Tribes to consider individual
circumstances in reviewing antidegradation demonstrations.  As a result, the
ten percent additional cost benchmark is included in this document as guidance
only.  EPA realizes that the determination of what represents affordable
treatment options is specific to the case in question.  Therefore, a strict
cut-off at ten percent additional costs is not realistic.  Greater costs may
be affordable in some cases; in others, ten percent may be too expensive.  In
the final Guidance, the determination of what treatment alternatives are
practicable is left to Tribes and States.  EPA is developing additional
National guidance on a variety of issues related to economic considerations in
water quality standards that will provide direction to Tribes and States as
they implement their antidegradation policies.  It is important  to note that
the affordability measures discussed above are separate and distinct from any
determination of penalties or ability to pay within the context  of an
enforcement action.

      Comment;_  Some commenters suggested that the cost-effectiveness
alternative described in the preamble to the proposal is preferable to the
approach taken in the proposed Guidance.

      Response:  Although EPA considers the ten percent additional cost
benchmark to be the easiest measure of affordability to implement, the final
Guidance does not mandate any one approach to identifying available treatment
options.  Tribes and States are free  to use any approach they choose that is
appropriate to the specific situation under consideration.  The  increased
flexibility of the antidegradation requirements of the final Guidance will
make it more sensitive to the circumstances of each individual  situation.  Any
benchmark chosen by a State or Tribe  to determine whether or not additional or
alternative treatment is affordable is relevant only within the  context of an
antidegradation demonstration for purposes of evaluating information provided
in support of a request to lower water quality in a high quality water.

-------
                             Section VH: Antidegradation                          221
      Comment:  Some commenters believed that expenditures should be required
even if improvements are only incremental.

      Response:  The final Guidance does not preclude Tribes or States from
granting only partial approval of requests to lower water quality.  Where a
Tribe or State believes that a project may occur with a less extensive
lowering of water quality than was requested, control documents may be issued
with limits that reflect the lowering of water quality necessary to
accommodate the activity.

      Comment:  Several commenters stated that the antidegradation
demonstration should establish a direct link between the degradation of water
quality and the social and economic development.

      Response:  EPA agrees with this comment in principle.  In the final
Guidance, a demonstration of the important social and economic development
that will result fr6m the proposed significant lowering of water quality is a
precondition to approving a significant lowering of water quality.  The other
components of the antidegradation demonstration, the cost-effective pollution
prevention analysis and the enhanced and alternate treatment analysis are
intended to ensure that social and economic development takes place with a
minimum of environmental impact.  However, it is not always possible to
quantify exactly how much degradation is associated with a certain activity
that will produce social and economic development.

c.    The Final Guidance

      The final Guidance retains the framework contained in the proposed
Guidance, but with less detail.  The demonstration is divided into pollution
prevention, alternative and enhanced treatment and social and economic
development components that are performed in series.  States and Tribes are
required to adopt the demonstration elements contained in the final Guidance
for purposes of regulating new and increasing discharges of BCCs.  For non-
BCCs, States and Tribes may either adopt the procedures contained in the final
Guidance, or develop their own antidegradation demonstration requirements
provided they are consistent with the requirements of 40 CPR 131.12.

      The first two- components of the demonstration, the pollution prevention
alternatives and the enhanced and alternative treatment demonstrations,
address the question of whether or not water quality must be lowered to
accommodate the proposed activity (i.e., is the significant lowering of water
quality necessary).  EPA's intent in mandating these tests is to ensure that,
when they are available, feasible alternatives that would allow an activity to
occur with little or no degradation of water quality are identified and
implemented and the lowering of water quality is minimized.  In keeping with
EPA's preference for source reduction over waste treatment, dischargers are
first directed to identify any and all cost-effective pollution prevention
alternatives that might eliminate or reduce the proposed significant lowering
of water quality,  if the significant lowering of water quality cannot be
eliminated entirely through pollution prevention, then alternative or enhanced
waste water treatment should be considered.  Only after pollution prevention
and alternative and enhanced treatment are examined and no feasible
alternative to the significant lowering of water quality is found can it be
considered necessary.  A more detailed discussion of the components follows.

i.    Identification of Cost-Effective Pollution Prevention Alternatives to
Prevent or Reduce the Significant Lowering of Water Quality

      This is the starting point for the antidegradation demonstration.  The
proponent of the activity should consider five broad categories of pollution
prevention activities in determining whether or not alternatives exist that
would reduce or eliminate the anticipated significant lowering of water
quality.  These include:

-------
222    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      -- Substitution of non-bioaccumulative or non-toxic chemicals for BCCs.
The primary objective of this evaluation is to determine if the source of a
BCC, which would otherwise cause or contribute to a significant lowering of
water quality, can be eliminated in favor of a less environmentally
problematic substance, especially one that is not a BCC.

      -- Application of water conservation methods.  The objective of this
portion of the pollution prevention analysis is to determine whether or not
reductions in the overall volume of waste water are possible and would reduce
pollutant loadings so that the proposed activity could occur without a
significant lowering of water quality.

      -- Waste source reduction within process streams.  The objective is to
evaluate all waste streams involved in the process affected by the proposed
activity.  Opportunities to control more carefully the use of raw materials
and reduce waste should be identified and implemented where feasible.

      -- Recycle or reuse of waste byproducts, either liquid, solid or gas.
The objective is to identify ways in which recycling and reuse of internal
waste streams can be employed to reduce the loadings of pollutants to the
environment.  This is a common practice in industry and can reduce energy, raw
material and waste disposal costs.  The proponent of the proposed activity
should investigate the process involved to determine whether or not
opportunities exist to implement such changes where they would alleviate the
need for a significant lowering of water quality.

      -- Manufacturing Process Operational Changes.  The focus of this part of
the investigation should be to identify different means of achieving the
desired end that will produce either smaller quantities of toxic waste
products or waste products that are less toxic.  All of the processes that
will contribute to the significant lowering of water quality should be
examined and alternatives that would reduce or eliminate the need to lower
water quality significantly should be identified.

      The pollution prevention possibilities discussed above are not intended
to be all inclusive.  Dischargers seeking approval for an action that will
result in a significant lowering of water quality should evaluate all aspects
of the proposed activity for opportunities to reduce pollutant loadings.  The
categories of pollution prevention discussed above should be viewed as
guidance to those performing an antidegradation demonstration and those
evaluating it.

      Pollution prevention is applicable to municipal as well as industrial
dischargers.  Where a significant lowering of water quality will result from
increased industrial use of a municipal treatment plant, the municipality
should ensure that the loadings from the industrial users are minimized
through pollution prevention to the extent possible.  Tools available to
municipalities to encourage pollution prevention include local limits and fees
levied on industrial users.  Municipalities should also consider public
education and bans on certain substances to reduce loadings from nonindustrial
sources.

      The determination that an alternative or combination of alternatives is
cost-effective is the decision of the Director, and not the entity that is
seeking to significantly lower water quality.  States and Tribes are
encouraged to develop their own guidelines  to follow when evaluating pollution
prevention alternatives identified by the entity to select those that are
cost-effective.

ii.   Alternative or Enhanced Treatment to  Eliminate the Significant Lowering
      of Water Quality

      The second part of determining whether  or not a significant lowering of
water quality is necessary is the alternative or enhanced treatment analysis.

-------
                             Section VII: Antidegradation                          223
This analysis should be undertaken after the pollution prevention analysis is
completed and should focus on removing the remaining incremental increase in
pollutant loadings after cost-effective pollution prevention measures are
taken.  If application of pollution prevention techniques alone are sufficient
to eliminate the significant lowering of water quality, the alternative or
enhanced treatment analysis need not be performed.

      The objective-of the alternative or enhanced treatment analysis is to
ensure that the actual degradation of the high quality water is reduced to the
greatest extent practicable.  The analysis proceeds by identifying the least
costly options for additional treatment under which the proposed activity
could occur without resulting in a significant lowering of water quality.  The
costs of the different treatment options are determined and compared to the
costs of the treatment needed to achieve all applicable standards, including
Federal effluent guidelines, water quality-based effluent limits and all other
applicable Federal and State or Tribal requirements.  Where treatment options
are identified that are comparable in cost to baseline treatment costs and
allow the proposed activity to occur without leading to a significant lowering
of water quality, those treatment options should be implemented in lieu of
lowering water quality.

      In the proposed Guidance, EPA considered using a ten percent increase in
treatment costs as a benchmark for determining whether or not alternative or
enhanced treatment options identified through this analysis were affordable.
Where treatment options with costs up to ten percent more than the cost of
meeting all applicable discharge requirements were identified, the discharger
would have been required to implement the treatment option identified and the
request to lower water quality would be denied.  In the final Guidance, the
ten percent cost benchmark is provided as guidance only.  Tribes and States
may use the ten percent value to assist them in evaluating antidegradation
demonstrations.  Tribes and States may also use other appropriate means of
identifying practicable treatment options such as the cost-effectiveness
approach described in the preamble to the proposed Guidance (see 58 FR 20910).
The final Guidance does not provide specific direction to Tribes and States on
how to determine the affordability of a treatment option under consideration
because to do so would limit their ability to respond to the unique
circumstances of each antidegradation demonstration.  Tribes and States retain
the ability to require treatment beyond the benchmarks discussed in the
preamble to the proposal, this document (SID), proposed or final Guidance, or
to require treatment that will only reduce the magnitude of the significant
lowering of water quality.  Tribes and States should also see EPA's draft,
Economic Guidance for Water Quality Standards Workbook. (November 1993) for
additional guidance on how to evaluate the affordability of treatment options.

iii.  Important Social and Economic Development

      In the final part of the antidegradation demonstration,  the proponent of
the activity that will result in a significant lowering of water quality must
show that the significant lowering of water quality proposed will support
social and economic, development.  This part of the demonstration should occur
only if no pollution prevention or alternative treatment options are
identified that will eliminate the need for a significant lowering of water
quality.  In determining whether or not a proposed activity will support
important social and economic development, Tribes and States should consider
the geographic area in which the significant lowering of water quality will
occur, the current or baseline economic condition of that area, the net
positive impacts that will result from the proposed activity and the
possibility of other development occurring in the area that will result in
similar economic and social benefits but will not cause a significant lowering
of water quality.

-------
224    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      -- Identification of the Affected Area

      The area in which the economic benefits occur should correspond with the
area in which water quality is lowered.  Determining the area is a case-by-
case decision, made taking into account the pollutants involved as well as the
location of the discharge.

      - - Baseline Economic Condition

      Once the affected area is identified, the baseline economic condition of
the area should be assessed.  Factors that may be useful include unemployment
rates, percentage of the population living below poverty levels, percentage of
the population that*are elderly and average household income relative to State
and National averages.

      -- Net Positive Impact

      In determining net positive impact the Tribe or State should attempt to
assess the benefits of the proposed activity corrected for any negative
economic impacts of the activity.  The types of benefits from the activity to
be considered include an increase in the number of jobs, an increase in
personal income and/or wages, reduction in unemployment rates or social
service expenses, increased tax revenues and provision of necessary social
services.  Other measures may be relevant on a case-by-case basis.

      Adverse economic impacts may also result from an activity that supports
social and economic development.  For example, a new industrial facility may
provide additional jobs in a community; however it may also make the receiving
water less attractive for recreation and cause a loss in tourism dollars.
Such impacts should be considered in determining whether or not a project or
activity that will result in a significant lowering of water quality will also
support important social and economic development.

      - - Other Developments

      Tribes and States should also consider whether a proposed activity will
preclude another activity that may not affect water quality yet yield
comparable social and economic benefits.  In the example above, the siting of
an industrial plant may preclude water front development or building of a
marina that would provide comparable social and economic development at less
cost to the environment.

4.    Antidegradation Decision

a.    Background

      The final section of the proposed Guidance established the process to be
used by the Director for arriving at a final decision regarding whether or not
to allow a significant lowering of water quality.  The proposed Guidance
specified how the information obtained through each of the antidegradation
demonstration components was to be evaluated and presented two alternatives
for factoring public comment into the final decision.  Under the proposed
Guidance, public comment could be sought following a tentative analysis of the
available data, or the data analysis could be deferred and public comment
sought on a tentative decision to deny the request to lower water quality.

b.    Discussion of Significant Comments

      Comment:  Many commenters stated that Tribes and States should be
allowed greater flexibility in how data are used and public comment is sought
in arriving at a final decision regarding an antidegradation demonstration.

      Response:  EPA agrees with  the views expressed by the commenters.  The
final Guidance no longer specifies how States and Tribes  should consider the

-------
                             Section VII: Antidegradation                          225
data and public input they obtain in making a final decision.  Also, States
and Tribes are required to adopt a procedure consistent with the final
Guidance for BCCs only; States and Tribes may develop their own decision-
making procedures for non-BCCs consistent with the requirements 40 CFR 131.12.

c.    The Final Guidance

      The final Guidance is less prescriptive than the proposed Guidance in
that the process for arriving at a decision is left to individual Tribes and
States.   States and Tribes are only required to adopt procedures consistent
with the Guidance for BCCs.  However, the essential requirements remain the
same and the options presented in the proposed Guidance may be used by Tribes
and States.  The Director must examine each of the demonstrations submitted by
the proponent of the activity that will lower water quality.  Also, an
opportunity for public comment must be provided and public comment must be
factored into the final decision, consistent with the antidegradation
standard.  EPA expects that the process of making a decision will parallel the
antidegradation demonstration process.  The Director should first evaluate the
analysis of cost-effective pollution prevention alternatives and determine the
extent to which pollution prevention can reduce or eliminate the significant
lowering of water quality.  The Director should then review the information
generated through the alternative and enhanced treatment analysis.  If the
analysis identifies affordable treatment options that, combined with the
pollution prevention alternatives, will eliminate the need to lower water
quality significantly, the Director should deny the request to lower water
quality.  If, however, the pollution prevention and alternative treatment
analyses are unable to eliminate the need to lower water quality, the Director
should weigh the social and economic development that will result from the
significant lowering of water quality in the affected area.  If the proposed
activity is found tp support important social and economic development, the
Director may decide to grant all or part of the requested significant lowering
of water quality, provided water quality sufficient to protect existing and
designated uses is maintained and provided the decision is subject to public
comment.

      Opportunity for public comment is an essential element of the
antidegradation decision making process and is required under Federal
regulations at 40 CFR 131.12.  If the tentative decision relates to an
activity subject to a NPDES permit,  the public participation requirements may
be fulfilled by the public notice of the draft permit and fact sheet.  In any
event,  the public notice of the tentative decision must either set forth the
extent to which water quality will be significantly lowered and the basis for
the tentative decision to allow the lowering,  or, if analysis of the
demonstration has been deferred, a tentative decision to deny the request to
lower water quality pending public comment and analysis of the information
obtained through the antidegradation demonstration.

-------
226     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
                        Section VELA: Site-Specific Modifications                     227
             .  IMPLEMENTATION PROCEDURES


A.    Site-Specific Modifications to Criteria and Values

1.    General

      As discussed in the proposed Guidance,  there  currently exists National
guidance to modify aquatic life 'criteria on a site-specific basis, but there
is no such guidance for modifying human health or wildlife criteria on a site-
specific basis (see 58 PR 20918).  The proposed Guidance continued to allow
more stringent site-specific modifications  to the wildlife and human health
criteria or values and to bioaccumulation factors  (BAFs).  Sections 1 through
4 below discuss the changes to the site-specific modifications for aquatic
life, wildlife, BAFs, and human health.

2.    Aquatic Life

      a.    Proposal:  The proposed Guidance,  consistent with the current
national Guidance, provided that  Great Lakes States and Tribes may adopt site-
specific modifications allowing more stringent or less stringent aquatic life
criteria or values when local water quality characteristics such as pH,
hardness, temperature, color, etc.,  alter the biological availability or
toxicity of a pollutant; or criteria when the sensitivity of the local aquatic
organisms (i.e.,  those that would live in the water absent human-induced
pollution) differs significantly  from the species actually tested in
developing the criteria.  The proposal suggested use of the existing national
guidance provided in the U.S. EPA Water Quality Standards Handbook (1983)
(1983 Handbook) when modifying the criteria.   (Since the proposal, the 1983
Handbook has been reformatted and republished as the U.S. EPA Water Quality
Standards Handbook, Second Edition - Revised,  1994)  (Revised Handbook) The
Revised Handbook is available in  the docket for this rulemaking.

      The proposed Guidance went  beyond existing national guidance by also
allowing the Great Lakes States and Tribes  to develop site-specific
modifications to chronic aquatic  life criteria for  the Great Lakes System to
reflect local physical and hydrologic conditions. Such conditions might
include natural features of a water body, such as lack of proper substrate,
cover, flow, depth, .pools, and riffles, unrelated to ambient water quality.
Specifically, such conditions would include any local physical or hydrological
condition which precluded aquatic life from remaining at the site for 96 hours
or more.  The proposal stated that sites where conditions precluded all but a
few forms of aquatic life from living may be protected by less stringent
chronic criteria.  EPA expected that this provision would typically be used
for waters where a full aquatic life use is unattainable.  The proposal did
not include such a provision for  acute aquatic life criteria.

      b.    Comments:  Many commenters supported the proposal for allowing
more or less stringent site-specific modifications  to aquatic life
criteria/values when local water  quality parameters alter the biological
availability 'or toxicity of a pollutant; or criteria when the sensitivity of
the local aquatic organisms differs significantly from the species actually
tested in developing the criteria.  A few commenters requested that EPA not
allow any less stringent site-specific modifications to aquatic life criteria.
These commenters were concerned that less stringent criteria in localized
areas in the Great Lakes System would cause declines in downstream water
quality.

      Several commenters suggested that EPA also allow less stringent site-
specific modifications for acute  aquatic life criteria to reflect local
physical and hydrologic conditions.   They reasoned  that when physical or

-------
228    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

hydrological conditions preclude aquatic life from remaining at a site  for a
period of time in which acute effects may occur, less stringent site-specific
aquatic life criteria should be allowed.  Further, commenters pointed out that
EPA offered no support for why this exception was allowed for chronic
criteria, but not acute criteria.

      EPA agrees with the commenters suggesting that there may be instances
where aquatic life might not inhabit a site due to the site's physical  or
hydrological characteristics (e.g., natural features of a water body, such as
lack of proper substrate, cover, flow, depth, pools, and riffles, unrelated to
ambient water quality).  To make such modifications, EPA will expect the State
or Tribe to demonstrate that aquatic organisms do not inhabit a site or do not
spend sufficient time at the site to experience acute effects.  The final
Guidance provides that the acute criteria may be modified to be less stringent
to reflect local physical and hydrological conditions.  Such modification may
be made to the criterion using the recalculation procedure provided in  Chapter
3 of the Revised Handbook.

      Like the National program, State and Tribal programs must ensure  that
less stringent site-specific modifications do not impair the water quality of
downstream waters. ,In addition, allowance for less-stringent site-specific
criteria must still'ensure protection of species which do "occur at the site"
(as defined in appendix F.A.l.a.ii of this rulemaking); in other words, the
absence of acute and chronic criteria is unacceptable.

      c.    Final Guidance:  The final Guidance maintains the requirements in
the proposal with the following changes:  States or Tribes are allowed  to
develop site-specific modifications to both acute and chronic criteria  to
reflect local physical and hydrological conditions.  This is a change from the
proposal which only allowed modifications for physical or hydrological
conditions to chronic criteria.  The final Guidance still allows modifications
to both acute and chronic criteria/values to reflect water chemistry and to
criteria to reflect species sensitivity differences.  The final Guidance
provides the State or Tribe with some options for achieving the requirement to
protect endangered or threatened species, such as:  (i) if the Species  Mean
Acute Value (SMAV) for a listed species, or for a surrogate of a listed
species, is lower than the calculated Final Acute Value  (FAV), such lower SMAV
is used instead of the calculated FAV in developing site-specific modified
criteria; or  (ii) the site-specific criteria may be calculated using the
recalculation procedure described in Chapter 3 of the Revised Handbook.  In
option  (ii) , acceptable toxicity data  (as defined in appendix A to part 132
for the aquatic life Tier I criteria methodology) for the listed species or a
surrogate for the listed species must be included in the data set in which the
criterion is recalculated.

      In defining a site the State or Tribe should consider the known range of
the threatened or endangered species.  Either option will ensure protection
for species listed under the Federal Endangered Species Act when adequate data
are available.  If there is a critical food source which is also an aquatic
animal, option  (ii) might provide a greater level of protection for that
listed species if that food organism were more sensitive than the listed
species itself.  By including data for the organism which is a critical food
for the listed species, option  (ii) will ensure protection of that food source
from water quality effects.  Option  (i) only considers the  sensitivity  of the
listed species itself.  EPA wishes to clarify that the recalculation procedure
contains a provision that the recalculated FAV, Criterion Maximum
Concentration  (CMC), and/or Criterion Continuous Concentration  (CCC) should be
lowered if necessary to ensure that the criterion is not likely  to jeopardize
the continued existence of any endangered or threatened species  listed  under
section 4 of the Endangered Species Act or result in  the destruction or
adverse modification of such species' critical habitat.

      For metals criteria, a State or Tribe may wish  to use total recoverable
metals criteria rather  than dissolved criteria to provide a greater  level of

-------
                        Section Vm.A: Site-Specific Modifications                     229

protection for such species.  EPA recommends use of total recoverable metals
criteria when it is known that an endangered or threatened species is
chronically sensitive to metals and lives in or on sediments and whose diet is
substantially comprised of benthic aquatic organisms.

      If the SMAV for a listed species, or for a surrogate of a listed
species, is not lower than the calculated FAV or if the recalculated criteria
(with data for the listed species included) is less stringent than the
established Tier I criteria or Tier II value, then the existing Tier I
criteria or Tier II value provides adequate protection for the listed species.

3.    Wildlife

      Many comments were received on the proposed use of site-specific
modifications to wildlife criteria.  The significant comments, including EPA
responses and changes to the proposal, are set out below.  Responses to all
comments are contained in the Response to Comments Document, which is part of
the docket for this action.

a.    site-specific Modifications

      i.    Proposal:  The proposed procedure 1 allowed wildlife criteria and
values to be modified on a site-specific basis to provide an additional level
of protection.  This additional protection could be provided through the use
of an intraspecies uncertainty factor  (UF), which was applied in addition to
the other UFs in the wildlife methodology (58 FR 20880).  The proposal did not
allow for less-stringent modification to wildlife criteria or values.

      ii.   Comments:  Many commenters agreed with the need for site-specific
modifications for wildlife, citing such factors as waterbody-specific
characteristics driving the bioavailability of the chemical in question, the
geographical representation of the representative species, and the dietary
habits of the representative species.  Many of the commenters criticized the
proposal for allowing only those modifications which would be more stringent
than the Great Lakes System-wide criteria, stating that it is not technically
defensible because certain taxa may be missing from specific locations, the
assumptions of the environmental fate of the chemical of concern may not be
valid for all sites in the system, and wildlife mobility may affect exposure
assumptions.

      EPA recognizes that there may exist situations in which a site-specific
modification, resulting in a less stringent criterion or value, may be
appropriate; therefore, the final Guidance allows for the development of site-
specific modifications to wildlife criteria that may be more or less stringent
than the Great Lakes System-wide criteria.  The following discussion provides
additional guidance for the development and application of site-specific
modifications that are more or less stringent than system-wide wildlife water
quality criteria.

      (A).  Less Stringent Site-specific Modifications

      Modifications to wildlife criteria that result in less stringent site-
specific criteria may be allowed when a site-specific BAF is derived, which is
lower than the System-wide BAF derived under appendix B to part 132.  At the
same time, it is important that a site-specific relaxation of system-wide
criteria not produce off-site impairments of designated uses due to wildlife
and prey mobility,  pollutant transport, and interbreeding of populations of
varying sensitivities.  To safeguard against that, the final Guidance requires
that a showing be made before approval of any site-specific modification is
granted that:  any increased uptake of the toxicant by prey species utilizing
the site will not cause adverse effects in wildlife populations in the Great
Lakes System; wildlife populations utilizing.the site, or boundary or
interconnecting waters will continue to be fully protected; and, the mobility

-------
230    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

of all prey organisms and wildlife populations have been adequately described
in justifying the site.

      EPA considered allowing less stringent modifications to the wildlife
criteria based on site-specific distributions of wildlife species, and their
associated dietary habits.  EPA believes that in most cases a less stringent
site-specific modification based on these factors will be difficult because
the representative species listed in the final Guidance were not selected to
be the sole species targeted for protection; rather, they were selected to
exemplify the highly exposed wildlife species resident in the Great Lakes
basin.  Hence, even if a representative species is not found at a specific
geographic site, or has a diet different than that listed in appendix D to
part 132.  other highly exposed wildlife species would still need the
protection granted by the criteria derived under appendix D to part 132, or
those criteria listed in Table 4 to part 132.  In addition, because in many
cases it is difficult to determine the relative sensitivities of the species
within a specific area, the use of the highly exposed species to represent all
species ensures that the majority of species will be protected if they are
more sensitive.  Although, EPA believes it would be difficult to demonstrate
that there are no species within an area that are not as exposed or not as
sensitive as the representative species, EPA believes it should not preclude
calculating wildlife criteria using a different set of species than those used
in deriving the four wildlife criteria.

      An important component of a site-specific modification is the definition
of the site to which the modification is applicable.  A site may range from
being a portion of a watershed to the entire part of the Great Lakes System
under the jurisdiction of the State or Tribe proposing the modification.

      (B).  More Stringent Site-specific Modifications

      States retain the authority through section 510 of the Clean Water Act
(CWA) to develop site-specific modifications that are more stringent than
system-wide criteria.  Use of the wildlife methodology described in appendix D
to part 132 is recommended, with the appropriate changes to the toxicological
or exposure parameters that are contained in that methodology.  The allometric
equations provided in appendix D to part 132, section III.E may be used if
feeding or drinking rates for the species are not available.  Because the
site-specific modifications subject to this subsection will be more stringent
than the corresponding system-wide value or criterion, the concerns behind the
demonstration described above for less stringent modifications are not
applicable.

      iii.  Final Guidance:  A Tier I wildlife criterion may be modified to be
either more or less stringent than the corresponding system-wide criterion or
value.  Where a site-specific modification  is proposed that is less stringent
than the system-wide criterion, a demonstration is needed that the site-
specific modification will not only protect on-site wildlife species but also
protect off-site wildlife species and their prey.  The demonstration must
consider the mobility of both the wildlife  species and the wildlife prey which
utilize the site.

      b.    Protection of Endangered or Threatened Species

      i.    Proposal:  The proposed Guidance did not contain any specific
provisions regarding protection of endangered or threatened wildlife species.

      ii.   Comments:  Some commenters stated that protection of the
individual, rather than the population, is  warranted in cases of endangered or
threatened species, but were concerned that the proposed Guidance was too
broad in allowing the application of  site-specific modifications for
protecting individuals beyond those listed  under the ESA.

-------
                        Section VELA: Site-Specific Modifications                     231

      Procedure 1.A of the final Guidance provides that States or Tribes must
develop site-specific modifications to protect endangered or threatened
species listed under section 4 of the ESA; it also describes a recommended
methodology to accomplish such a modification  (procedure 1.A.2.C).  The final
Guidance recommends the use of the methodology contained in appendix D to part
132, modified as described below, to provide a greater degree of protection
for endangered or threatened species.  Where available, toxicological or
exposure data for the species in question shall be used.  Where ingestion
rates are not available, the allometric equations contained in appendix D to
part 132 may be used.  Also, EPA strongly encourages the use of the
methodology for determining BAFs (Appendix B to part 132) to derive
appropriate BAFs for the site-specific modification.

      The equation from appendix D to part 132 is modified by the inclusion of
an intraspecies UF to provide additional protection for the benefit of
individual members of populations.  The recommended range of this UF is 1 to
10 and additional discussion is provided in the final Great Lakes Technical
Support Document for Wildlife Criteria.  In addition, the selection of
endpoints for protection of listed species may be broader than those described
in appendix D to part 132; however, the selection of an endpoint must be
defended by establishing its relevancy to the viability of an individual or
its ability to reproduce.

      In developing'the site-specific modification, it is important to
delineate the geographic area to which the modification shall apply.  In
designating the appropriate site, it is important to consider fully not only
the current and potential ranges of the wildlife species in question, but also
the range and mobility of the prey organisms.  Therefore, a site could, by
necessity, be defined as the entire Great Lakes System within the jurisdiction
of the State or Tribe.

      iii.  Final Guidance;  Because the equation contained in appendix D to
part 132 has been revised in accordance with comments received (see section VI
of this document), the following equation is recommended in place of the
equation presented in the proposal.  In addition, where available and
appropriate, toxicological, epidemiological, or exposure information for the
species in question is recommended to be used in the methodology.


                                       10
                             UFA x UFL x UFsx  UF,
Where the terms are defined in section II, appendix D to part 132, and also
include:

      UF, = Intraspecies Uncertainty  Factor for  extrapolating  to the  most
sensitive individuals within a population (unitless).

      The lowest of the wildlife values calculated from the above methodology
contained in appendix F.I.2 to part 132, and the two class-specific wildlife
values derived for £he system-wide wildlife water quality criterion should be
selected as the site-specific modification.

4.     Bioaccumulation Factors

      a.     Proposal:  The proposed Guidance allowed only more  stringent site-
specific modification for BAFs pursuant to authority reserved to the States
and Tribes under CWA section 510.  BAFs could be modified on a  site-specific
basis where reliable data showed that local bioaccumulation was greater than
the system-wide value.

-------
232    Water Qualify Guidance for the Great Lakes System — Supplementary Information Document

      b.    Comments:  Many commenters stated that it was not scientifically
justifiable to allow only more stringent criteria  (a higher BAF) and that EPA
should also allow less stringent criteria  (a lower BAF).

      Many commenters stated that less stringent modifications to the BAFs
should be allowed based on both chemical- and site-specific characteristics,
such as particulate organic matter and dissolved organic matter which may
modify the bioavailability of an organic chemical at a specific location; and
the partitioning of the chemical between the water column and sediment since
this may be a significant factor affecting the bioavailability of the
contaminant.

      Several commenters stated that they wanted the flexibility to adjust the
percent lipid value.to reflect that of fish consumed from a particular
location.  Other commenters were concerned that the fish species used in
deriving the BAF may not be present at a particular site, which may lead to a
more stringent value than needed at that site.

      Many commenters stated that the current BAF methodology precludes using
site-specific information.  The commenters wanted EPA to instead allow for
calculation of a BAF using field data from the site of discharge or other
modifications to reflect the characteristics of the site.  In particular,
commenters suggested modifications to the food-chain model based on chemical-,
site- and species-specific data.   Some commenters wanted EPA to develop both
the methodology and'the data for deriving site-specific BAF values.

      EPA agrees that both more and less stringent modifications of the BAFs
should be allowed on a site-specific basis if there is scientific
justification.

      EPA agrees with commenters that modifications to the BAFs should include
consideration of the bioavailability of the chemical. EPA does not agree that
the partitioning of the chemical between the water column and sediment, and
other fate processes affect the BAF.  However, modifications based on
chemical-specific characteristics are not specific to a particular site, and
therefore not appropriate for procedure 1 of appendix F to part 132.
Chemical-specific considerations are addressed in the derivation of the
system-wide BAF for the chemical.  Because the final criteria are based on the
total concentration of the chemical in the water,  the final BAF)  used in the
calculation of the criteria may be modified for site-specific particulate and
dissolved organic carbon concentrations.

      EPA also agrees that site-specific modifications should be allowed if it
can be demonstrated that the percent lipid of aquatic organisms is different
than the percent lipid values used in the derivation of BAFs.  The percent
lipid of 1.82 for trophic level three and 3.10 for trophic level four in
edible tissue for use in determining human health BAFs and the percent lipid
of 6.46 for trophic level three and 10.31 for trophic level four in whole fish
for use in determining wildlife BAFs for an organic chemical are protective of
most sites in the Great Lakes.  In cases where it  can be documented that the
percent lipid for the fish species consumed at a site differs from these
values, modifications can be made to the BAF.

      In the final Guidance, derivation of Food Chain Multipliers  (FCMs) based
upon the model of Gobas  (1993) is used because the input parameters are easily
defined and measured, the calculated BAFs are in better agreement with
measured BAFs for chemicals with very high K^s than  the  Thomann model  (1989)
used in the proposal, and the model uses equilibrium partitioning theory to
predict chemical residues in benthic organisms.  The model of Gobas  (1993) can
also be adjusted for site-specific considerations.  The FCMs in Table 1 of
appendix B to part 132 were calculated using Great Lakes-specific data.  If  it
can be demonstrated that the values for input parameters used by EPA are not
appropriate for a given site, use of other values  is permitted.  In
particular. States and Tribes can use site-specific data for the food web and

-------
                        Section VELA: Site-Specific Modifications                     233


accompanying lipid content of the aquatic species, and concentrations of the
dissolved organic carbon and particulate organic carbon.  If one input
parameter is modified, site-specific values must be used for all input
parameters.  Selective modification of the FCM is not allowed because it will
not accurately represent the characteristics of a specific site.

      For the final Guidance, a distinct single BAF for each trophic level of
concern for derivation of human health criteria and a separate distinct single
BAP for each trophic level of concern for derivation of wildlife criteria were
calculated using the concentrations of POC and DOC in Lake Superior as
reasonable worst-case conditions.  As noted in the above responses and in the
methodology set forth in procedure 1 of appendix F to part 132, and
subsequently in appendix B to part 132,  EPA is allowing for modifications to
the BAF based on site-specific characteristics.

      c.    Final Guidance:  EPA is allowing site-specific modifications to
the BAF based on the procedure set forth in procedure 1 of appendix F to part
132.

5.    Human Health

      a.    Proposal:  The proposed Guidance stated that there is no specific
EPA guidance regarding site-specific modifications to human health water
quality criteria.  (The 1983 Handbook does not offer any guidance in the area
of human health site-specific criteria development.)  Consistent with the
Steering Committee proposal, the proposed Guidance restricted site-specific
modification to human health criteria/values to only those which would
increase the level of protection for humans. The proposed Guidance also stated
that human health criteria or values shall be modified on a site-specific
basis to provide additional protection appropriate for highly exposed
subpopulations.

        EPA invited comments on whether the proposed approach for humans and
wildlife was reasonable or whether less stringent site-specific modifications
could be allowed under certain circumstances.

      b.    Comments;  EPA received numerous comments that States should be
allowed to make a demonstration for less stringent human health criteria based
on site-specific  conditions, if scientifically supported.  The majority of
commenters suggested that two exposure components, fish consumption and fish
lipid content should be evaluated on a site-specific basis.  These same
commenters believed that there could be parts of the Great Lakes System
(especially tributaries) where one could demonstrate that the fish consumption
rate and/or fish lipid concentration are significantly different (presumably
lower) from those in the proposed Guidance.   Other commenters argued that the
body weight of sensitive subpopulations (women of childbearing age and
children)  should be accounted for in developing human health criteria on a
site-specific basis.   Commenters also believed that site-specific criteria
should be generated only if the resulting criterion were protective of
potential uses at the site and all downstream uses.  Commenters stated that
EPA did not provide scientific justification for allowing only more stringent
site-specific human health criteria.

      A few commenters were opposed to the idea of less stringent site-
specific human health criteria.  Commenters stated that less stringent
criteria should not be allowed for the open waters of the Great Lakes, since
this would result in inconsistency throughout the basin, but was in favor of
allowing less stringent modifications for tributaries (waters other than the
open waters of the Great lakes).  Other commenters believed that EPA should
not allow site-specific modifications for fish consumption of subsistence and
recreational angler;  but should select a fish consumption rate protective of
sport anglers and other highly exposed subgroups.  Still other commenters
pointed to the impractical nature of conducting meaningful site-specific
studies because the Great Lakes Basin is large with free movements of

-------
234    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

fisheries and distribution of pollutant loadings, believing the practice would
only result in States imposing different health advisories and ultimately non-
attainment of water uses.

      EPA agrees with many of the comments, and has changed the final Guidance
to allow for demonstrations for less stringent site-specific human health
criteria as related,to fish consumption rate and BAFs.  Other exposure
parameters such as drinking water consumption rate, body weight and incidental
ingestion rate can fluctuate on a population basis, but are not considered
parameters likely to change on a site-specific basis.  For example, the mean
body weight of a State may be greater or less than the 70 kilograms assumed in
the criteria methodology,  but making a determination that a specific site has
a higher or lower mean body weight may be implausible and impractical.  To
make a showing for a less stringent criterion, a State or Tribe would have to
demonstrate that there was a site-related population of people weighing more
than 70 kilograms.  In addition, EPA has provided guidance on choosing
protective body weights for those parts of the population considered more
vulnerable to the toxicological effects of environmental contaminants.  With
regard to toxicological assessments, EPA does not believe there are likely
conditions under which a site-specific toxicological assessment can be made.
For example, to make a site-specific toxicological assessment, it would have
to be shown that a particular human population at a specific site was more or
less sensitive to an environmental contaminant due to genetic predisposition
or site-related conditions which mitigate or enhance the toxicological effects
of a specific chemical contaminant.   Again, EPA believes making such a
showing is highly unlikely and probably implausible.

      i.    Fish Consumption Rate:  EPA believes, on the other hand, that it
is conceivable that there might be isolated tributaries and populations of
people who do not consume as much fish as the rate presented in the proposal.
Such a demonstration would likely be difficult to make, due to the transience
of people in the Great Lake area, but if a State or Tribe can demonstrate
based on data that a group of people who are the exclusive users of a
waterbody has a significantly lower fish consumption rate than the rest of the
Great Lakes population, they may apply that lower rate in developing their
human health criteria for that waterbody.  The States and Tribes must ensure
that fish migration from the waterbody in question will not lead to increased
exposure to other human populations.  The State or Tribe must also demonstrate
that the specified waterbody is not associated with a known or anticipated
group of individuals who may consume more fish, such as a sport or subsistence
angler population.  To ascertain such information, the State or Tribe must
conduct a site-specific fish consumption survey.  When determining a site-
specific fish consumption rate, a site-specific fish lipid percentage must
also be determined.

      In response to commenters who stated  EPA should not allow for site-
specific criteria adjustments due to the difficulty in conducting meaningful
site-specific surveys,  EPA does not believe it is fair to disallow such a
demonstration if a State or locality believes there is a basis for such a
showing.  While it is true that a well conducted fish consumption survey is
costly and sometimes open to various interpretations, the data from such a
survey may be more useful in determining subpopulation consumption trends than
a default value or a generalized  assumption for an entire population of
people.

      ii.   Percent Fish Lipid:   EPA has concluded that the percentage  of fish
lipid can be adjusted on a site-specific basis.  This can be  done  in
conjunction with a fish consumption survey  or localized monitoring data on
fish species of the area.  From a site-specific  fish  consumption  survey, the
predominant fish  species consumed,  the percentage  of  each  species  consumed,
and the lipid concentration  of  those species can be  determined.   If the
recalculated weighted mean percent  lipid is significantly  different than  that
in the final Guidance,  it may be  used  in calculating site-specific human
health criteria.

-------
                        Section Vffl.A: Site-Specific Modifications                     235

      iii.  BAF:  EPA has concluded that the BAF can be adjusted on a site-
specific basis.   (See section A.4 of this document on BAFs to determine how a
site-specific BAF can be derived.)

      In response to the commenters who stated site-specific criteria should
be limited to tributaries only, EPA concludes that a showing can be made for
any part of the Great Lakes System as long as it is scientifically justified.
Site-specific conditions must be the basis for the criteria adjustment '
regardless of geographical location in the Great Lakes System.

      With regard to the Guidance which states that human health criteria or
values shall be modified on a site-specific basis to provide additional
protection appropriate for highly exposed subpopulations, EPA believes the
best way to determine whether a highly exposed subpopulation exists is through
the use of fish consumption surveys targeted toward waterbodies and sites
which are known (qualitatively, if not quantitatively) for high levels of
sport and subsistence angler usage.  Once a site/waterbody is established as a
high-use site, a fish consumption survey should be conducted to determine
average and high level (90th or 95th percentile consumption rates) consumption
rates for the sport and subsistence angler subpopulations.  Once it is
established that a highly exposed subpopulation does exist at a distinct site,
the criteria or values must be modified using the fish consumption rate, the
percent lipid, and the resulting BAFs associated with the higher consumption
rates attributed to the subpopulation in question.   For information on how to
conduct a fish consumption survey and how to analyze the results of such a
survey, refer to the following EPA document: Consumption Surveys for Fish and
Shellfish.  A Review and Analysis of Survey Methods.  Feb. 1992.  EPA 822/R-
92-001.

      An alternative to conducting site-specific fish consumption surveys is
to use default assumptions for fish consumption by recreational and
subsistence anglers.  Several commenters recommended default assumptions for
recreational and subsistence fish consumption.  For recreational fish
consumption, the commenters recommended using a range of 45-150 grams/day in
developing water quality criteria protective of sport anglers.  For
subsistence fishing, the commenters recommended using a range of 90-165
grams/day in developing water quality criteria protective of subsistence
anglers.  EPA believes the use of default assumptions are acceptable.  The
actual default values used, however, will depend on the circumstances in that
State or Tribal area.  For more details on the commenters1 recommendations
refer to the Environmental Defense Fund/ Penobscot Indian Nation/NAACP Legal
Department document entitled: The Protection of Sport and Subsistence Fishing
Populations in the United States: Recommendations to the Administrator, EPA,
For Implementation of the President's Executive Order on Environmental Justice
and the Subsistence Consumption of Fish and Wildlife, June 1994 which is
available in the docket for this rulemaking.

      c.    Final Guidance;  The final Guidance allows human health criteria
or values to be modified on a site-specific basis based on differences in fish
consumption, or BAF where the resulting criteria may be less stringent than
the final criteria listed in Tables 3, for the reasons stated above.

-------
                              Section Vffl.B: Variances                           237
B.    Variances from Water Quality Standards for Point Sources

      The final Great Lakes Water Quality Guidance allows Great Lakes States
and Tribes to include water quality standards variance provisions  in their
water quality standards, and grant variances based on those provisions,  as
long as they are consistent with procedure 2 of appendix F.  These water
quality standards variance procedures provide a mechanism for States and
Tribes to maintain the basic standards  (uses and criteria) as goals and  assure
compliance with sections 301(b)(1)(C) and 402(a)(1) of the CWA that require
NPDES permits meet applicable water quality standards, while granting
temporary relief to point source dischargers under appropriate circumstances.
This Guidance does not require the States or Tribes to include a variance
provision as part of their standards program.

      The intent of the variance provision is to: provide a mechanism by which
permits can be written to meet a modified standard where compliance with the
underlying water quality standard is demonstrated to be infeasible; encourage
maintenance of original standards as goals rather than removing uses that may
be ultimately attainable; identify conditions under which variances may  be
granted; identify the requirements for variance applications; and  ensure the
highest level of water quality achievable while the variance is in effect.

      The final variance procedures included in procedure 2 of appendix  F
differ little from the proposal and provide for consistent application of
water quality standards variances for Great Lakes States and Tribes.
Variances can be requested for any of the same reasons which justify removing
designated uses as described at 40 CFR 131.10(g).

      In the final Guidance, States and Tribes retain the discretion to  define
what specific information they will require in a permittee's variance
demonstration and application.  States and Tribes also have the discretion to
define the decision criteria to use when approving or disapproving a variance,
as long as they are at least as stringent as the requirements in procedure 2
of appendix F and subject to EPA review and approval.

      A State or Tribe choosing to adopt variance procedures as part of  its
part 132 submission*will provide information on the requirements for the
variance demonstration and application as well as the evaluation criteria that
the State or Tribe will use to approve or disapprove specific variances.  A
variance procedure should assure that: the public has sufficient information
to comment on the appropriateness of a State's or Tribe's WQS variance
process; EPA has sufficient details to determine if the State or Tribe
procedures comply with the CWA and are approvable; and both EPA and the  public
have adequate information on which to judge State or Tribal compliance with
its own procedures when making individual variance decisions.  The final
guidance does not require States or Tribes to grant variances in any specific
circumstance.

      In addition to reviewing individual variances granted by a State or
Tribe,  EPA will review the State or Tribal variance procedure itself as  part
of its review of the State or Tribe's GLI submission.

1.    Appljcabi1itv

      a.    Proposal:  The proposed Guidance limited the availability of a
variance to the permittee requesting the variance and only for the
pollutant(s)  specified.  The water quality standards for the affected water
body would not otherwise be changed by a variance.  The proposal also did not
allow variances for new or recommencing dischargers as those terms are defined
at 40 CFR 122.2.

      In procedure 2.C of the proposal, EPA also did not allow a variance to
be granted if standards would be attained by implementing effluent limitations
required under sections 301(b)  and 306 of the CWA and by the permittee

-------
238    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

implementing cost-effective and reasonable best management practices.  (In the
final Guidance, this requirement has been moved to the Applicability section
for clarity.)

      b.    Discussion of Significant Comments

      Comment:  Several commenters suggested that EPA also allow basin-wide or
waterbody variances often citing the need for relief from "ubiquitous"
pollutants.  Other commenters objected to allowing waterbody variances.

      Response:  EPA considered these comments and has decided not to
incorporate waterbody variances into the final guidance for the following
reasons:  (i) EPA has not developed methodologies for applying waterbody
variances and has not allowed this practice nationally,  (ii) EPA is concerned
that waterbody variances could provide inappropriate relief to nonpoint
sources and cause ari additional regulatory burden to point source dischargers,
and  (iii) EPA believes that the appropriate long term solution to the problem
of "ubiquitous" pollutants is the total maximum daily load  (TMDL) process
which can account for all sources of pollutant loading both point and
nonpoint, and that the appropriate short term relief consists of discharger
specific variances and determinations made pursuant to the Reasonable
Potential portion of the final guidance.  See sections VIII.C. and VIII.E. of
this document for a discussion of TMDLs, Reasonable Potential and "ubiquitous"
pollutants.  In order to provide the most efficient short term relief to
"ubiquitous" pollutants, EPA encourages States to consider multiple discharger
specific variance requests on a watershed basis where appropriate.    Multiple
discharger variances are similar to, but not the same as a waterbody variance.
Where all point source discharges on a waterbody request a variance, both
approaches will have the same effect on the point sources.  The difference is
that a waterbody variance applies to all nonpoint sources of the pollutant as
well, whereas a multiple discharger variance does not.

      Comment:  Some commenters stated that variances should be available to
dischargers who applied for NPDES permits on or before the effective date of
the rule.

      Response;  EPA agrees.  The new definition of "new discharger" at §132.2
developed for purposes of compliance schedules and mixing zones will
accomplish this.

      Comment:  Several commenters objected to excluding new or recommencing
dischargers for consideration for WQS variances, citing  the presence of
"ubiquitous" pollutants in the Great Lakes  System.  Other commenters agreed
with the proposal to exclude new and recommencing discharges from variances.

      Response:  As was stated in the preamble to the proposed Guidance, and
reiterated above, WQS variances are used to allow existing dischargers to
comply with a  modified standard when compliance with the underlying  standard
is infeasible.  Variances are not intended  to allow water quality that is
already below  standards to be further degraded, which would be the case if new
or recommencing dischargers  add increased loads of a pollutant to a  waterbody.
This concept is reinforced in the proposed  and final Guidance at procedure
2.F.I which requires that the NPDES permit  limitation be no less stringent
than that achieved under the previous permit.  Since new or recommencing
dischargers have no previous load, they could not add  any load under the
variance procedure's terms.  If new, recommencing or existing dischargers
desires  relief from  requirements to control "ubiquitous" pollutants and will
not add  increased loads of the pollutant to the waterbody,  the appropriate
mechanisms  are found in the  TMDL  procedure  3 and the Reasonable  Potential
procedure  5  in appendix F.

      Comment:   Some commenters requested that EPA more clearly define
"recommencing  discharger," and  specify  "...that facilities  that  cease

-------
                              Section Vm.B:  Variances                           239
discharging temporarily, but retain their NPDES permits, are not recommencing
dischargers."

      Response:  EPA disagrees that further definition of "recommencing
discharger" is needed.  40 CFR 122.2 defines the term as "...a source that
recommences discharge after terminating operations."  EPA interprets
terminating operations to mean eligible for NPDES permit revocation.  Hence,
sources that retain'their NPDES permits would not be considered "recommencing
dischargers."

      Comment;  Several commenters suggested that EPA should eliminate the
requirement of Best Management Practices as a condition for obtaining a
variance.  Other commenters stated that the BMP requirement should be
clarified, or that BMPs should be limited to those that may be implemented by
a particular discharger on a reasonable and cost-effective basis.

      Response:  EPA disagrees that the BMP requirement should be eliminated.
EPA agrees, however, that the BMPs that must be implemented before a variance
may be granted should be limited to those that may be implemented by a
particular discharger.  WQS variances are not intended to allow water quality
that is already below standards to be further degraded.  In addition, as
stated in procedure 2.F.I, the purpose of variances are to improve water
quality as much as possible by requiring effluent limitations that represent
the level of water quality achievable by the permittee.  If the permittee can
implement cost effective and reasonable BMPs for nonpoint sources, over which
it has control, that will attain water quality standards, the permittee should
implement those BMPs rather than requesting a variance for its point source
discharge.  If implementing such BMPs will improve water quality but not meet
the standards, implementation by the permittee will result in a reduced
variance request and an overall improvement in water quality.

      c.    Final Guidance:  For the reasons stated above, EPA has retained
paragraph A, the Applicability section of procedure F.2, in the final variance
procedures except:  (i) the term "new discharger" has been changed to "new
Great Lakes discharger" and defined at §132.2, (ii) a new paragraph 2.A.2 has
been added to assure compliance with section 7 of the Endangered Species Act,
and (iii) the final sentence from procedure 2.C of the proposal has been moved
and re-numbered as procedure 2.A.3.

2.    Maximum Timeframe for Variances

      a.    Proposal:  Procedure 2.B of the proposal, specified a variance
term not to exceed three years.  EPA proposed the three-year term to reinforce
the triennial review required of all water quality standards in accordance
with section 303(c) of the Clean Water Act.

      b.    Discussion of Significant Comments

      Comment:  This section received a large number of comments.  Some
commenters favored retaining the three-year requirement but the majority
favored extending the life of a variance to five years or the term of the
NPDES permit for which it is granted.  Several commenters stated that a three
year variance term would present a substantial administrative burden on both
the discharger and the permitting authority,  especially for situations where
a variance renewal is appropriate.

      Response:  EPA's 1979 Guidance (see the preamble to the proposed
Guidance (58 FR 20921) for discussion of EPA's 1979 Guidance) clearly
indicated that variances are granted for a specific time period and must be
reviewed every three years, but does not suggest that variances must expire at
the same three-year interval.

      c.    Final Guidance:  Based on the comments and the rationale above,
EPA has extended the allowable term for a water quality standards variance to

-------
240    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

five years or the term of the NPDES permit, whichever is less.  This reduces
the administrative burden of a facility in preparing a variance request since
the facility can prepare the request at the same time it prepares the NPDES
permit application.  To ensure that variances are indeed reviewed trienriially
with the rest of the State or Tribal standards, and modified as appropriate,
EPA has added that requirement to procedure 2.B.  In addition, a new
requirement has been added at procedure 2.F.4 to provide for reopening and
modifying the NPDES'permit as appropriate based on changes made during the
water quality standards review, e.g. if the variance is terminated or
modified.

3.     Conditions to Grant a Variance

      a.    Proposal:  Variances under the proposed Great Lakes Guidance were
applicable if any of five specified types of waterbody conditions exist and/or
the affected community would encounter substantial and widespread economic and
social impacts as a result of the point source having to install controls,
beyond technology-based requirements, necessary to meet the WQS.

      The permittee was required to make two other demonstrations.  The first
demonstration was that the requested variance is consistent with State or
Tribal antidegradation procedures.  In the second demonstration, the applicant
was required to characterize the extent of any increased risk to human health
and the environment associated with granting the variance compared to the
original water quality standards, and the State or Tribe was required to find
that any such increased risk is consistent with the protection of the public
health, safety and welfare before granting the variance.  Because variances
are from water quality standards that meet the goals and requirements of the
Clean Water Act, this language was intended to ensure that the general
requirement of section 303(c)(2)(A) of the CWA be met even though specific
protective criteria may be temporarily exceeded.

      Consistent with other approaches for regulatory relief, the permittee
was responsible for providing sufficient relevant information, pursuant to
State or Tribal requirements, to make a variance demonstration for the
pollutant(s) in question.  Failure of the permittee to make an adequate
demonstration or to provide sufficient information was sufficient for a State
or tribal denial of the variance.

      b.    Discussion of Significant Comments

      Comment:  Some commenters suggested adding additional reasons for
granting a variance to those proposed at procedure 2.C including technological
infeasibility and background pollutants.

      Response;  Based on current EPA guidance  (discussed in the preamble to
the proposed rule at 58 FR 20921}, the justifications for a WQS variance are
the six reasons for removing a designated use found at 40 CFR 131.10(g).
Because the final Guidance is to be no less  restrictive than national guidance
 (CWA §118(c) (2) (A))> EPA has decided that it would be inappropriate to provide
additional justifications for variances at  this time.  However, it is EPA's
intent to provide the Great Lakes States  and Tribes reasonable discretion in
defining, characterizing and developing decision criteria for the conditions
for granting a variance, subject  to EPA review and approval.  It is EPA's
position that by maintaining this discretion, States and Tribes will be able
to address specific concerns in  the State/Tribal procedures developed pursuant
to procedure 2.  Flexibility for  addressing background pollutants is provided
for in the Reasonable Potential  section of  the Guidance.

      Comment:  Some commenters were confused about the last  two requirements
in proposed procedure 2.C  regarding:  (i)  the demonstration that
antidegradation requirements had been met,  and  (ii) the demonstration  of  the
extent of increased risk to human health  and the environment.

-------
                              Section Vm.B:  Variances                           241
      Response:  EPA agrees that the proposal was confusing and has clarified
this in the final Guidance.  The term "demonstrates" as used in the proposal
was not intended to be synonymous with the "antidegradation demonstration"
required in appendix E.  The "demonstration" in procedure 2 was intended to be
a showing that either water quality would not be lowered in the waterbody  (the
requirement at procedure 2.F.I requiring dischargers to maintain the level of
treatment achieved under the previous permit would almost always prevent a
discharger from being granted a variances that would result in an actual
lowering of water quality) or, if water quality would be lowered, that the
antidegradation requirements of appendix E were met.  For these reasons, EPA
anticipates that an "antidegradation demonstration" pursuant to appendix E
would rarely be required when granting a variance.  To avoid further confusion
in this section, the term  "demonstrates" has been changed to "shows" in the
final guidance.

      In addition, the two requirements have been re-formatted at procedure
2.C.2 to indicate that they are indeed separate requirements, as was the
intent of the proposal, and are not associated with the social/economic
justification at procedure 2.C.l.f.

      Comment:  Some commenters thought it important to integrate the variance
procedures with the antidegradation procedures while others objected to this
requirement.  Many of those objecting were concerned with the amount of effort
required to conduct an antidegradation demonstration for every variance
request.  Commenters also stated that this language was necessary to ensure
that granting a variance does not have a downstream impact on high quality
waters.

      Response:  EPA believes that many of the objections to the
antidegradation requirement stem from the confusing terminology discussed in
the previous comment/response.  The terminology change in the final guidance,
as discussed above, addresses these objections.

      This antidegradation requirement in the variance procedures was intended
to prevent a variance that would result in a lowering of actual water quality
for any pollutant where water quality for that pollutant does not support
either the designated or existing uses or in any water constituting an
outstanding national resource (ONRW) at part 132, appendix E, section I.C as
well as to prevent dischargers from avoiding the proposed requirements of part
132, appendix E, section I.B in high quality waters by requesting a variance
rather than conducting an antidegradation demonstration.  The requirement at
procedure 2.F.I requiring dischargers to maintain the level of treatment
achieved under the previous permit is expected to prevent a discharger from
being granted a variance that would result in a significant lowering of water
quality, and an appendix E antidegradation demonstration would not normally be
necessary.  An appendix E antidegradation demonstration would only be
necessary if required by the language in appendix E.  The antidegradation
showing here would most often demonstrate to the State and public that a
concurrent antidegradation question is not at issue or that, if one is, the
regulatory provisions for antidegradation are being met.

      Comment:  Comments were also received on several of the specific
conditions for granting a variance at procedure 2.C.  Some of these comments
requested more specific guidance on some of the conditions.

      Response;  It is EPA's intent to provide the Great Lakes States and
Tribes reasonable discretion in defining, characterizing and developing
decision criteria for the conditions for granting a variance, subject to EPA
review and approval.  It is EPA's position that by maintaining this
discretion,  States and Tribes will be able to address specific concerns in the
State or Tribal procedures developed pursuant to procedure 2.

      c.    Final Guidance:  EPA has maintained procedure 2.C, including the
requirements on antidegradation and on demonstrating that there will be no

-------
242    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

increased risk to human health and the environment from granting the variance,
essentially unchanged in the final Guidance.  EPA has made a terminology
change, discussed above,  to clarify the intent that an antidegradation
demonstration pursuant to appendix E will apply only if the variance results
in significant lowering of water quality in a high quality waterbody, a
circumstance which is not expected to occur under procedure 2.  In addition,
the two "showing" requirements have been re-formatted at procedure 2.C.2 to
indicate that they are separate requirements, and not associated with the
social/economic justification at procedure 2.C.l.f.

4.     Submittal of Variance Application

      a.    Proposal:  The proposed Guidance required permittees to submit a
variance application to the State or Tribe no later than 60 days after the
regulatory authority reissued or modified the permit.  The application was
required to provide information necessary to evaluate whether  the conditions
for granting a variance,  as set forth in procedure 2.C of the proposal, were
met.

      b.    Discussion of Significant Comments

      Comment:  Some commenters stated that the 60 day time frame was too
short and/or that the variance application should be allowed at other times
 (e.g., during the permit application process or any time during the permit
cycle).  Several commenters pointed out that this language would require all
permits for which a variance is granted to be reopened and modified and
suggested that processing a variance request during the permit application
process would substantially reduce the administrative burden on the discharger
as well as the permitting authority.  Many similar comments were received
regarding the timeframe for application for a variance renewal at procedure
2.H.

      Response:  EPA agrees and has modified procedure 2.D to allow States and
Tribes to determine the most appropriate variance application schedule to suit
their administrative procedures for both first-time variances at procedure 2.D
and for variance renewal at procedure 2.H.

      c.    Final Guidance;  In the final guidance EPA has removed both the 60
day limit and the requirement that a variance be applied for after a final
NPDES permit is issued.  EPA expects States and Tribes to specify in their
variance procedures the timing that best suits their own regulations and
permitting procedures.



      a.    Proposal:  The proposed Guidance provided for public notice and
opportunity to comment on a variance request  (procedure 2.E) and on the draft
modified NPDES permit  (procedure  2.G.).  In addition, the requirement at
procedure 2.J that variances be appended to State water quality standards
rules ensured that the public is made aware of which variances have been
granted.

      b.    Discussion of Significant Comments

      Comment;  Commenter stated  the proposed public notice requirements  for
variance procedures^were not adequate to allow full public involvement.

      Response;  EPA  does not agree.  Procedure 2  does not modify the  normal
public participation  requirements for adopting WQS, but rather adds  a
requirement that the  notice also  present the State or Tribe's preliminary
decision.

      The following is a summary  of the elements that EPA expects to be made
available to the public in order  to meet the public notification requirements

-------
                              Section Vm.B:  Variances                           243
of the water quality standards regulation  (40 CFR I3l.20(b)).   if these items
are not included in detail in the public notice, they must be in the public
record and the public must be made aware of their existence and of how and
where they may be obtained.

      (i)   A statement that the action complies with the State's or Tribe's
variance procedures- and description of those procedures;

      (ii)  The permittee's demonstration, including the rationale for the
requested variance, and the extent of any increased risk to human health and
the environment associated with granting the variance; and

      (iii) The public notice for any draft NPDES permit, the public comments
and public hearing records pursuant to procedure 2.E, and the State approval.
This public notice can be combined with the public notice for a draft NPDES
permit,  as long as the variance is identified and all the necessary
information pertaining to the variance is included.

      Comment;  Some commenters stated that a timeframe should be required for
the public notice and/or comment period.

      Response:  EPA does not agree that timeframes for these activities need
to be specified in the Guidance because the various State and Tribal
administrative procedures are different.  EPA does, however, expect the States
and Tribes to include appropriate timeframes in their variance procedures,
pursuant to their administrative procedures, to provide meaningful opportunity
for public comment as well as orderly and timely State or Tribal action.

      c.    Final Guidance:  For the reasons stated above, EPA has retained
the language contained in the proposal, with the clarification that the public
notice can be combined with the public notice for a draft NPDES permit, as
long as the variance is identified and all the necessary information
pertaining to the variance is included.

6.    Final Decision on Variance Request

      a.    Proposal:  The proposed Guidance required the State or Tribe to
issue a final decision on a variance request within 90 days of the expiration
of the public comment period.  The proposed Guidance also required that this
decision specify all NPDES permit conditions needed to implement the variance.
These conditions were to assure that:  the permittee minimizes the water
quality standards exceedance by implementing the level of treatment currently
achievable (i.e., conditions requiring effluent limitations at least as
stringent as those achieved under the previous permit); the permittee makes
reasonable progress toward attaining the water quality standards; and effluent
limits sufficient to protect water quality standards are in effect upon
expiration of the variance.  States/Tribes were to deny a requested variance
if the permittee failed to make the demonstrations required under section C of
the proposed procedure.

      b.    Discussion of Significant Comments

      Comment;  Some commenters objected to the permit condition requiring
"reasonable progress" be made toward attaining water quality standards,
stating for example, that in most cases, the permittee receiving a variance
will not contribute to the conditions that prevent the receiving waters from
attaining water quality standards and therefore will be powerless to make
"reasonable progress" toward attaining water quality standards for the
waterbody.

      Response:  EPA does not agree that a discharger is always "powerless" to
affect pollution sources outside its immediate contribution.  Dischargers can,
for example,  participate in total maximum daily load  (TMDL) development and/or
engage in pollution trading schemes to reduce load from outside its facility.

-------
244    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Comment:  Some commenters were unclear about what EPA intends by
"reasonable progress."

      EPA agrees that "reasonable progress" is imprecise, but believes this is
necessary to provide flexibility to the State or Tribe to require activities
appropriate to the State or Tribal programmatic objectives and the individual
situation subject to EPA review and approval.  In addition to progress
reducing loads outside its immediate contribution discussed above, reasonable
progress can be made in improving the quality of the discharge through
conditions such as: (i)  the establishment of a capital improvements fund; and
(ii)  continued investigations of treatment technologies, process changes,
pollution prevention,  wastewater reuse and/or other techniques that will
reduce the level of the pollutant or result in compliance by the permittee
with the WQS and submission of reports on the investigations at such time
specified by the State.   It would be difficult for EPA to define what
"reasonable progress" is without considering site-specific information on the
facility, pollutant and receiving water for a given situation.  In instances
where a discharger is indeed "powerless" to affect any change in conditions,
the State or Tribe may interpret existing practices as "reasonable."

      Comment;  Procedure 2.F.3 proposed requiring, upon expiration of a
variance, compliance with the effluent limitation in effect prior to the
granting of the variance.  Commenters thought this requirement confusing.

      Response:  Because the final Guidance no longer requires that a final
NPDES permit be issued prior to a variance application, there may well be no
pre-existing effluent limitation implementing the underlying WQS with which to
comply.  EPA has, therefore, modified this requirement to provide that
effluent limitations be in compliance with applicable water quality standards
upon expiration of the variance.

      c.    Final Guidance:  For the reasons stated above, EPA has retained
procedure 2.F essentially unchanged.  Because of the changes to procedure 2.B
which extend the allowable term for a variance beyond the 3 years in the
proposal  (see discussion at B.2.c, above) new language has been added at
procedure 2.F.4 requiring a permit condition that allows the permitting
authority to reopen and modify a permit which is based on a WQS variance if
that variance has been rescinded or modified pursuant to a State or Tribal
triennial standards revision.

7.    Incorporating'Variances into NPDES Permits

      a.    Proposal:  Once a variance is granted, the proposed Guidance
required the State or Tribe to modify the NPDES permit to incorporate all
NPDES permit conditions determined to be necessary to implement the variance.

      b.    Discussion of Significant Comments

      Comment:  Some commenters stated that the Guidance should not routinely
require that a permit be modified in order to implement a variance.

      Response:  EPA agrees.  EPA has removed the requirement in procedure  2.D
that a variance may be applied for only after a final NPDES permit is issued.
By allowing variance applications any time during the permit renewal process,
EPA anticipates that most permits that receive variances will have the
provisions of the variance incorporated into the permit when it is reissued.

      c.    Final Guidance:  EPA has removed the requirement in procedure 2.D
that a variance may be  applied for only after a final NPDES permit is issued.
The final Guidance allows  the permitting authority to incorporate the
provisions of  the variance into the permit at a time, and in a manner which is
appropriate, given the  circumstances  of the variance.

-------
                              Section Vm.B: Variances                           245
8.    Renewal of Variances

      a.    Proposal:  The proposed Guidance required a permittee  to  re-apply
for a variance no later than the submission of a permit application or 60  days
prior to the expiration of the variance.  The permittee was required  to make a
new showing of justification; variances would not be automatically renewed.
As part of the renewal application, the permittee would be required to
demonstrate that it has met the NPDES permit conditions implementing  the
existing variance.  The same public notice requirements for the  initial
issuance of a variance applied to the renewal.  Permittees not demonstrating
compliance with these conditions would not be eligible for a variance renewal.

      b.    Discussion of Significant Comments

      Comment^  Several commenters stated that an application for  variance
renewal should not be required until the permit renewal becomes  final.  Others
stated that the variance reapplication process should occur in conjunction
with the permit renewal process.

      Response:  EPA agrees that such flexibility is reasonable  and has
modified procedure 2.H to allow States and Tribes to determine the most
appropriate variance re-application schedule to suit their administrative
procedures.

      Comment;  Some commenters stated that discharges that violate conditions
of a variance should not necessarily be ineligible for a renewed variance
pointing out that EPA and the states already have ample authority  to  impose
enforcement sanctions on dischargers violating their permit conditions,
including variance conditions, and that EPA need not require states to impose
a second penalty on those dischargers.  Other commenters also stated  that EPA
proposes to treat all violations of variance conditions the same,  and
therefore, even de minimis violations would result in the inability to renew a
variance.

      Response;    The permit conditions are intended to implement the basic
principles of water.quality standards variances that (i) the non-attainment of
standards be minimized, (ii) progress toward attaining standards be achieved
where possible, and  (iii)  the discharger must meet the standard upon
expiration of the variance  (see discussion of current EPA policy in the
preamble to the proposed guidance at 58 PR 20921.)  Dischargers  that  fail to
meet permit conditions designed to achieve these basic principles  should not
be allowed to continue the practice through a variance renewal.  EPA  intends,
however, that States and Tribes should have discretion in determining whether
a particular non-compliance with variance conditions in NPDES permit  would
warrant denial of a variance renewal request, subject to EPA review and
approval.

      c.    Final Guidance;  Based on the rationale above, EPA has retained
procedure 2.H but has modified it to allow States and Tribes to determine the
most appropriate variance re-application schedule to suit their administrative
procedures.   EPA has also modified the final Guidance consistent with 40 CFR
122.64 to allow the permitting authority to deny a variance renewal where the
permittee has not fully complied with the permit conditions applicable to the
variance.

9.     EPA Approval
                   *
      a.    Proposal;  The proposed guidance listed the information to be
submitted to EPA for review and approval, including the permit application,
public comments.  State or Tribal approval, and NPDES permit; and gave
timeframes for those submittals.

-------
246    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

      b.    Discussion of Significant Comments

      Commentj  Some commenters stated that a time requirement  for EPA review
of WQS variances was needed.

      Response:  EPA does not agree that it is necessary  for  this  Guidance to
specify an EPA deadline for review of variances.  Variances are modifications
of State or Tribal water quality standards and are, therefore,  subject to EPA
review and approval.  Like other water quality standards  changes,  variances
are effective when adopted  (under the terms of the adoption), whether or not
EPA review is complete.  In addition, there are already statutory  and
regulatory timeframes, at §303(c)(2)(B) of the CWA and  40 CFR 131.21(a)
respectively, that govern EPA approval/disapproval decisions  on all State or
Tribal water quality standards revisions including variances.

      c.    Final Guidance:  Based on the rationale above, EPA  has retained
procedure 2.1 as proposed.  Editorial changes have been made  for clarity.

10.   State or Tribal Water Quality Standards Revisions

      a.    Proposal;  The proposed Guidance  required the State or Tribe to
append the State- or Tribal-adopted variances to the State or Tribe's water
quality standards.

      b.    Discussion of Significant Comments

      Comment:  EPA requested comment on whether a timeframe  was necessary for
the State or Tribe to append variances to their standards.  Commenters stated
that a timeframe is not necessary.

      Response;  EPA agrees and expects States and Tribes to  append variances
to their standards in as timely a fashion as  possible,  consistent  with their
administrative procedures,  so that the public is fully  informed on the
variance that have been adopted.  The appended information should  include at a
minimum:  the discharger receiving the variance, the term (beginning  and
ending dates) of the variance, the waterbody  or waterbodies affected, the
pollutant(s) affected by the variance, and the modified allowable  ambient
concentration value(s) for  those pollutants.

      c.    Final Guidance  EPA has retained  procedure  2.J as proposed.

-------
                               Section Vm.C: TMDLs                           247
C.    Total Maximum Daily Loads

1.    Background

      Section 303(d) of the Clean Water Act requires the establishment of
total maximum daily loads  (TMDLs), in accordance with priority rankings, for
waters, that are failing to meet or not expected to meet applicable water*
quality standards despite implementation of technology-based and other
existing controls.  See 40 CFR 130.7 and existing EPA guidance including
"Guidance for Water Quality-based Decisions: The TMDL Process," EPA 440/1-91-
001, April 1991.

      TMDLs quantify the maximum allowable loading of a pollutant to a water
body, and allocate this loading capacity to contributing point and nonpoint
sources (including natural background, in-place contaminants, direct wet and
dry deposition, groundwater inflow, and overland runoff) such that water
quality standards will be attained.  A TMDL must incorporate a margin of
safety  (MOS) that accounts for uncertainty about the relationship between
pollutant loads and water quality.  TMDLs may involve a single pollutant
source or multiple sources (e.g., both point sources and nonpoint sources).
Current regulations specify that TMDLs need to take into account critical
conditions for stream flow, loading, and water quality parameters  (see 40  CFR
130.7(c)(1)).  Site-specific factors are thus to be reflected in the TMDL  even
though the TMDL process may be used to ensure that water quality goals are
achieved for a waterbody segment, whole waterbody or watershed.

      Under the CWA, States and Tribes are primarily responsible for
developing TMDLs.  EPA is required to review and approve or disapprove TMDLs
developed and submitted by States and Tribes.  If EPA disapproves a State  or
Tribal TMDL, EPA must establish such TMDL {CWA section 303(d)(2)and 40 CFR
130.7(d)}.

      When applicable water quality standards cannot be attained through the
implementation of controls on point sources, within the time period specified
in the applicable standards or implementing regulations, States and Tribes may
choose to develop TMDLs using a phased approach.  The phased approach to TMDL
development is intended to achieve load reductions capable of ensuring the
attainment and maintenance of water quality standards.  EPA expects the
allocations within phased TMDLs to be based on a reasonable expectation that
water quality standards will be met in a reasonable period of time.

      The phased approach to TMDL development is an iterative process that
provides for pollution reduction while the regulatory agency collects and  uses
new monitoring data and the demonstrated performance of existing controls  to
evaluate the TMDL and revise it as necessary.  TMDLs established using the
phased approach are based on best available information, sound professional
judgment,  and a margin of safety to account for uncertainty in available data
and the anticipated relationship between controls, loading reductions and
predicted changes in water quality.  Such TMDLs require a monitoring plan, a
schedule for installation of controls, collection of monitoring data to verify
point and nonpoint source load reductions, assessment of water quality
standards attainment and additional modelling, where appropriate.  If
standards are not attained after implementation of controls recommended by the
TMDL, the data obtained through the monitoring program should be used to
revise the TMDL.
                   *
      The phased approach to TMDL development recognizes that water quality
standards cannot be attained immediately, but TMDLs developed on this basis
nevertheless must reflect reasonable assurances that water quality standards
will be attained in a reasonable period of time.  When developing a TMDL using
the phased approach,  all known sources of pollution are considered, although
specific controls on those sources may be implemented in stages.  The time
period associated with these stages of implementation ultimately determines
when water quality standards will be met for a particular waterbody.  The

-------
248    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

phased approach may provide a scheduled time frame in which to implement
controls' recommended by the TMDL and achieve water quality standards and may
be particularly appropriate when addressing difficult water quality problems
in cases when data,-models and predictive tools are generally less well-
developed than for water quality problems associated primarily with the
discharge of a few point source pollutants into small watersheds.  Determining
the reasonable period of time in which water quality standards will be met is
a case-specific determination.  This determination depends upon a number of
factors, including, but not limited to, receiving water characteristics,
persistence, behavior and ubiquity of pollutants of concern, type of
remediation activities necessary, available regulatory and non-regulatory
controls and individual State requirements for the implementation of water
quality standards.

      TMDLs established using the phased approach are the preferred approach
for developing schedules of how and when water quality standards will be met
in cases when data, models, and predictive tools are not yet adequate to
address complex water quality situations characterized by persistent,
ubiquitous pollutants and water quality impacts resulting from nonpoint
sources of pollution.  EPA believes that it is reasonable and appropriate in
these circumstances to establish TMDLs which schedule implementation
activities over a period of time.  This would result in some sources achieving
load allocations prior to other sources, provided that progress is being made
in achieving water quality standards in accordance with the schedule
established by the TMDL.  Thus, for example, EPA believes it is reasonable to
consider expected nonpoint source load reductions if they will result from the
implementation of specific voluntary or non-voluntary controls, are specific
to the pollutant of concern and the waterbody for which the TMDL is being
developed.  In some cases, for example, water quality standards may reasonably
be expected to be met within one NPDES five-year permit cycle.  In other cases
the reasonable expectation of meeting water quality standards could be  twenty
years, following the implementation of controls on nonpoint sources such as
sediment.  In still other cases, the reasonable expectation of meeting water
quality standards could be keyed to the implementation of other controls,
(e.g. air quality standards.)
                   *
      The final Guidance is not intended to comprehensively address all
aspects of TMDL development and implementation of CWA section 303(d).   Rather,
for specific matters not addressed by the final Guidance, national regulations
and guidance for the TMDL program will continue to apply to States and  Tribes
in the Great Lakes System  (see 40 CFR 130.7 and existing EPA guidance
documents such as the Technical Support Document for Water Quality-based Toxic
Control,  (TSD) EPA 505/2-90'-001, March 1991, and Guidance for Water Quality-
based Decisions:  The TMDL Process, EPA 440/1-91-001, April 1991, both
available in the docket).

      The final Guidance does not include specific provisions for deriving
nonpoint source load allocations and implementing nonpoint source controls.
While general guidance on how TMDLs should consider nonpoint source loadings
is provided, EPA regulations and technical guidance should be consulted for
more specific information.   (See, e.g., Guidance Specifying Management
Measures for Sources of Nonpoint Pollution in Coastal Waters, EPA 840-B-92-
002, January 1992, for a discussion of best management practices for nonpoint
sources; and Technical Guidance for Estimating Total Maximum Daily Loads
(TMDLs): Integrating Steady-State and Episodic Point and Nonpoint Sources,
draft June, 1994, both available in the docket).

      Procedure 3 specifies procedures for establishing total maximum daily
loads, wasteload allocations, and load allocations.  Portions of this
procedure also apply to wasteload allocations calculated in the  absence of a
TMDLs, and to preliminary wasteload allocations for the purpose  of determining
the need for Water Quality Based Effluent Limits  (WQBELs) under  procedure  5 of
appendix F.  See procedure 5.A.2 and  5.F.2 of appendix F and corresponding

-------
                               Section Vm.C: TMDLs                           249
discussion at section VIII.E.2.A and VIII.E.2.H of this document for further
information.

2.    Overview of Proposed Procedures 3A and 3B

      a.    Proposal:  The proposed guidance included two distinct approaches
for developing TMDLs: procedure 3A  (Option A) and procedure 3B  (Option B).
Both options contained the same eleven general conditions applicable to TMDL
development and special conditions regarding control of bioaccumulative
chemicals of concern  (BCCs).  Options A and B were also essentially the same
with respect to the development of TMDLs for open waters and connecting
channels of the Great Lakes System as defined at section 132.2 of the proposed
Guidance.

      The main differences between the two options existed in the development
of TMDLs for discharges to tributaries. These differences reflected the
process by which such TMDLs were developed and the degree of specificity
contained in the particular procedure. A TMDL developed for discharges to
tributaries under Option A was to be based on evaluation of the basin as a
whole, followed by site-by-site adjustments.  In contrast, Option B focused
initially on evaluating limits needed for individual point sources, with
supplemental emphasis on basin-wide considerations as necessary.  Specific
components of Options A and B are discussed in greater detail below, and in
the preamble to the proposed guidance.  Readers are encouraged to review the
preamble to the proposed guidance for more detailed information on the
proposed Options A and B  (58 PR 20928) .

      EPA sought comments on all aspects of both options, including the
overall technical and programmatic approaches set out in each option, the
consistency of each option with regard to existing national policy and program
approaches, and the degree to which each option allows for integrated
development of effective point and nonpoint source controls.  EPA requested
comments on how the options should be incorporated into the final
implementation procedure and specifically asked whether all States and Tribes
in the Great Lakes System should be required to adopt either Option A or B, or
whether States and Tribes should be allowed to choose an approach that is
consistent with one of the proposed options depending on the situation at
hand.  EPA also solicited comments on the option of not providing specific
TMDL provisions in the final Guidance and instead relying on existing TMDL
regulations and guidance.

      b.    Comments: Several commenters claimed that the proposed TMDL
procedures were confusing, fragmented and provided insufficient guidance on
how water quality-based permit limits would be calculated.  For example, many
commenters found the formulas specified in Option B confusing and some
suggested that certain components of the formulas were inaccurate or
inappropriate.

       Most commenters expressed no clear preference for either option.  Many
commenters advocated that the final Guidance allow States to choose either
procedure 3A or 3B.  Some expressed preferences for particular elements of an
option.  For example, one commenter suggested that the State or Tribe should
be given discretion to deviate from Option A's basin-wide approach and use an
area-specific approach if appropriate in a particular circumstance.

      A number of commenters suggested that, although both options had merit,
and/or limitations,, only one should be adopted to ensure consistency
throughout the Great Lakes System.  Many commenters,  including a number of
States, preferred Option B and maintained that if a single procedure is
adopted in the final Guidance,  it should be Option B.  These commenters
believed that Option B would provide greater consistency among the States than
Option A. Several commenters preferred Option B but suggested that stronger
elements from Option A should be incorporated into a revised Option B.  A
number of commenters suggested that Option A was too burdensome.

-------
250    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Several commenters supported the watershed-based approach reflected in
Option A.  Other commenters preferred Option A but recommended specific
modifications.  Among the recommended changes to Option A were establishing
specific requirements for mixing zones for non-BCCs.  Some commenters
suggested including,specific formulas for calculating nonpoint source
loadings.

      c.    Final Guidance:  In response to these comments, EPA simplified the
TMDL procedure in the final Guidance and clarified a number of provisions.
EPA includes only one TMDL procedure in the final Guidance in response to
concerns that the TMDL procedure promote consistency throughout the Great
Lakes System.  The final procedure 3 combines aspects of both Options A and B,
and, in response to comments, includes some of the more specific provisions of
both options A and B.  For example, in order to promote consistency among the
Great Lakes States and Tribes, EPA is retaining, with some modifications,
certain mixing zone'provisions for non-BCCs from option B.  EPA eliminated
some of the more burdensome and confusing aspects of both the proposed
options.  For example, in the final Guidance, the formulas in Option B are no
longer included.

      The final Guidance provides a greater degree of flexibility than
afforded by either proposed procedure 3A or 3B by allowing States and Tribes
to choose different implementation approaches while at the same time ensuring
a level of consistency by requiring implementation of specific components of
the procedure.  For example, the final Guidance does not specify whether a
State must adopt a basin-wide approach such as that in proposed Option A, or
an approach like proposed Option B, which would focus initially on evaluating
limits needed for individual point sources.

      The final Guidance also retains the flexibility provided in the
proposal.  For example, although the final Guidance specifies that States and
Tribes consider nonpoint source loadings, EPA has not adopted the commenters'
suggestion to specify a formula to calculate nonpoint source contributions.
Rather, States and Tribes are provided flexibility to evaluate such
contributions and to address nonpoint source contributions through existing
programs.

      The final Guidance retains the eleven general conditions and the
separate provisions for open waters of the Great Lakes System and tributaries,
with certain modifications.  Like both the proposed options 3A and 3B, the
final Guidance requires the elimination of mixing zones for BCCs; however, the
final Guidance adds a procedure to grant an exception for existing discharges
of BCCs in limited circumstances.  The general conditions of application and
specific provisions of the final procedure 3 are discussed in detail below.

      In addition, procedure 3 has been revised to include new language
(section A), which authorizes the use of certain assessment and remediation
plans in lieu of TMDLs whenever, in the final Guidance, a TMDL would be used
as the basis for a wasteload allocation.  Specifically, these assessment and
remediation plans could be used in lieu of TMDLs when deriving wasteload
allocations under procedures other than the  "baseline" procedures in procedure
5.F.2 of appendix F, when establishing mixing zones for existing discharges of
BCCs in waters not attaining water quality standards under procedure 3.C.6 of
appendix F, or as an alternative to TMDLs and the intake pollutant procedures
in procedures 5.D-E of appendix F when adjusting point source controls to
account for intake pollutants as provided in procedure 5.D.1.C of appendix F.
Thus, for example, when developing a WLA for a particular pollutant and point
source, a State or Tribe would rely upon the applicable WLA established  in an
approved TMDL or assessment and remediation plan.  If no such TMDL or
assessment and remediation plan exists, the WLA would be derived using
procedure 5.E.2.a or 5.F.2 of appendix F as appropriate.

      Under procedure 3.A of appendix F, assessment and remediation plans may
be used in lieu of TMDLs if they meet all the requirements of procedure  3,

-------
                               Section Vm.C: TMDLs                           251
satisfy the public participation requirements applicable to TMDLs, and are
approved by EPA under 40 C.F.R. § 130.6 as meeting these requirements.  Once
approved by EPA, the assessment and remediation plans will function as updates
to State or Tribal continuing planning processes, which may include, among
other things, TMDLs and areawide waste management plans under section 208.
When seeking EPA approval of these assessment and remediation plans, States
and Tribes must certify that the requirements of procedure 3 are met.
Procedure 3.A also authorizes the use of qualifying assessment and remediation
plans, such as Remedial Action Plans  (RAPs) and Lakewide Management Plans
(LaMPs), under section 118(c)(3) &  (4) of the CWA.

      The TMDL process is an important planning tool that helps identify water
quality problems and recommends solutions that link the development and
implementation of control actions to the attainment of water quality
standards.  The objective of a TMDL is to allocate allowable loads of a
particular pollutant among difference sources of that pollutant so that the
appropriate control actions can be taken and water quality standards achieved.
As discussed in section VIII.C.I above, when water quality standards cannot be
attained immediately, TMDLs may be developed under a phased approach if
appropriate.  While TMDLs are the preferred mechanism for addressing water
quality impairments, particularly where nonpoint source contributions are
significant, EPA recognizes that other mechanisms can employ the same type of
analysis and obtain the same results as formal TMDLs.  EPA also acknowledges
the comments, particularly of States, that identify comparable planning tools.
In particular, as described in section I.D.4 of this document, the States and
EPA Regional offices in the Great Lakes basin have undertaken significant
assessment and remediation planning efforts through the development of RAPs
and LaMPs.  Some States may undertake similar efforts through water quality
management plans under sections 208 of the CWA.  Accordingly, the final
Guidance specifically recognizes that assessment and remediation plans other
than TMDLs can be used with comparable water quality effect, provided that
they contain certain basic elements.  In other words, EPA expects that
assessment and remediation plans developed and approved under procedure 3.A of
appendix F can function in lieu of a TMDL for water quality decisionmaking in
the Great Lakes System because such plans, at a minimum, will assess the
sources causing or contributing to a particular water quality impairment,
identify remediation activities that are reasonably expected to result in
nonpoint source load reductions as necessary, in combination with point source
controls, to achieve water quality standards within a reasonable period of
time, incorporate a'margin of safety, and establish wasteload allocations for
point sources that are consistent with these water quality objectives.
Procedure 3.A also provides that any part of an assessment and remediation
plan that also satisfies one or more requirements under CWA section 303(d) or
implementing regulations may be incorporated by reference into a TMDL as
appropriate.  If a State or Tribe submits for EPA approval an assessment and
remediation plan under procedure 3.A that fully satisfies the requirements for
a TMDL, EPA may also approve that plan under section 303(d).

3.    General Conditions of Application

      As proposed,  Options A and B both contained the same eleven general
conditions of application for every TMDL established under the GLWQI to assure
that TMDLs employed consistent methodologies, analytical approaches and
assumptions.  Commenters overwhelmingly supported the proposal to include a
set of general conditions applicable to all aspects of TMDL development.

      Language is added in the final Guidance to clarify that the general
conditions also apply,  where indicated, to wasteload allocations (WLAs)
calculated in the absence of TMDLs and preliminary WLAs for purposes of
determining the need for WQBELs under procedure 5 of the final Guidance.

-------
252    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

a.    General Condition 1 - TMDLs Required

      i.    Proposal:  General condition 1 described the circumstances under
which a TMDL would be required upon State or Tribal adoption or EPA
promulgation of the Guidance.  In the proposal, general condition 1 specified
that, at a minimum, TMDLs were to be established for each pollutant for which
it was.determined that there is reasonable potential that a discharge will
cause or contribute to an exceedance of water quality standards as determined
pursuant to proposed procedure 5.  As proposed, such TMDLs would need to be
established in advance of the issuance of any new or revised permit for the
discharge of the pollutant, unless it was determined pursuant to the proposed
procedures that a TMDL is not needed.

      Proposed procedure 5 specified that the State or Tribe was to include a
water quality-based effluent limit in an NPDES permit whenever a pollutant is
or may be discharged into the Great Lakes System at a level that will cause,
have the reasonable,potential to cause, or contribute to an excursion above
any Tier I criterion or Tier II value.  Dnder procedure 5, as proposed. States
or Tribes would have been required to develop preliminary effluent limitations
to determine if all Tier I criteria and Tier II values would be met after
discharge, where there was data to develop such criteria or values.
Preliminary effluent limitations were to be derived from preliminary wasteload
allocations, which in turn were to be based upon and consistent with the
wasteload allocation procedures defined in the proposed procedure 3.  As
proposed, the procedure 3 requirement that a TMDL be developed whenever
reasonable potential to cause or contribute to an exceedance of water quality
standards was found, and the requirement that preliminary effluent limits
based on preliminary WLAs developed under procedure 3 be used to determine if
there is reasonable potential, were confusing and implied a circular logic.
The proposal provided, in effect, that TMDLs be developed when reasonable
potential was demonstrated, and that reasonable potential be demonstrated on
the basis of preliminary effluent limits, normally derived from TMDLs.

      ii.   Comments;  EPA received numerous comments on general condition 1.
A number of commenters were concerned with the perceived burden associated
with developing TMDLs under the proposal and, in particular, many were
concerned with the' burden associated with requiring a TMDL for a waterbody in
advance of issuing any new or revised permits.  These commenters asserted that
general condition l'would essentially prohibit any point source discharges of
a particular pollutant in the absence of a TMDL for that pollutant.  The
commenters contended that the effect of the prohibition would be to require
TMDLs on the basis of a single discharger's  "reasonable potential" to exceed
standards even in waters where TMDLs would have minimal environmental benefit,
and thus would be inefficient.  Several commenters claimed that the existing
national TMDL program requirements for identifying waters not meeting
standards, and for setting priorities to develop TMDLs, are sufficient  to
ensure  that TMDLs are developed for those waterbodies most in need.

      Numerous commenters pointed out the ambiguity between proposed procedure
3 and proposed procedure 5 relating to determination of reasonable potential.
Under proposed procedure 3, a TMDL was required when there was a finding of
reasonable potential.  However, under proposed procedure 5, a finding of
reasonable potential would be based on a  "preliminary wasteload allocation"
prepared using the procedures set forth in  the TMDL procedure.  The proposal
did not define "preliminary WLA."

      A number of  commenters suggested other circumstances that should  trigger
the requirement  to establish a TMDL.  Several  commenters  suggested that TMDLs
be required  for  pollutants that have  caused fish  consumption advisories on  the
premise that waters' subject  to fish  consumption advisories are  exceeding
narrative water  quality  criteria, if  not numeric  criteria.

      iii.   Final  Guidance;  General  condition 1  in  the final Guidance  no
longer  specifies that a  TMDL would need  to  be  developed for each pollutant  for

-------
                               Section Vm.C: TMDLs                           253
which reasonable potential is found.  Instead, TMDLs shall be established in
accordance with the waterbody listing and prioritization process outlined in
CWA section 303(d), 40 CFR 130.7 and existing EPA guidance.  Under existing
law, if existing required controls are not sufficient to attain and maintain
applicable water quality standards, the waterbody must be included on the
303(d) list, which, under the regulations, is to be submitted to EPA for
review.and approval or disapproval.  The list must include a priority ranking
of listed waters and must identify those waters targeted for TMDL development
as required by CWA section 303(d)(1)(A) and 40 CFR 130.7(b)(4).  EPA makes
this change in response to commenters1 concerns about the proposal.  First,
the final Guidance refers back to the national TMDL program rather than
creating a new trigger for TMDL development based on a finding of reasonable
potential under procedure 5.  This should minimize the confusion created by
the proposal.  The final Guidance,  by referring to existing TMDL regulations,
should also minimize concerns about the additional burden that might have
occurred under the proposal  {e.g.,  permitting subject to TMDL development).
Changes to the proposal were also made to address concerns about the use of
limited resources to develop TMDLs in waters presently attaining water quality
standards but where1 the discharge of a particular pollutant has the reasonable
potential to cause or contribute to an excursion above those water quality
standards.  While TMDLs for waters currently attaining water quality standards
are important tools to ensure that such standards are maintained, EPA
recognizes that many States and Tribes may choose to place a higher priority
on restoring impaired or threatened waters and will choose to use their
limited resources for that purpose.  This change is intended to preserve State
and Tribal discretion in establishing priorities for TMDL development and
implementation.

      In response tp comments advocating that fish consumption advisories
should trigger TMDL development EPA is developing guidance to clarify the
relationship between fish advisories and section 303(d) lists.  (Draft memo
dated July, 1994, available in the docket). EPA believes that, absent
information to the contrary, it should be presumed that fish consumption
advisories demonstrate use impairments for waters designated for the uses
specified in section 101(a) of the Clean Water Act, when defined by a State or
Tribe to include fishing.  The listing of such waterbodies on section 303(d)
lists is consistent with the purpose and intent of the Clean Water Act.

      General condition 1 also provides that, when water quality standards
cannot be attained immediately,  the TMDL must reflect reasonable assurances
that they will be achieved in a reasonable period of time.  For a more
thorough discussion of this concept, see section VIII.C.I above.

b.    General Condition 2 - Attainment of Water Quality Standards

      i.    Proposal:  In the proposal, general condition 2 discussed the load
reductions that should be achieved through TMDLs.  The first sentence of
general condition 2 supplemented the provisions of proposed general condition
1 by specifying that TMDLs would also need to be developed whenever the sum of
existing point source and nonpoint source (including natural background)
loadings of a particular pollutant exceeds the loading capacity of the water
for that particular pollutant, minus any margin of safety and minus any
capacity reserved for future growth.  As proposed, general condition 2 also
established that a TMDL for a given pollutant must implement all criteria for
that pollutant that are applicable to the waterbody in question.

      ii.   Final Guidance:  EPA did not receive significant comments on
general condition 2 as proposed.  However, EPA has reorganized proposed
general conditions 2 (Load Reductions), 3 (WLA Values) and 9  (TMDL
Allocations) in order to present the same material in a sequence that more
closely tracks the requirements of CWA section 303(d).  General condition 2 is
now entitled "Attainment of Water Quality Standards" and consists of a single
sentence on that subject drawn from proposed general condition two.  The final
guidance specifies that a TMDL must ensure attainment of water quality

-------
254    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

standards, including all numeric and narrative criteria, Tier I criteria, and
Tier II values where applicable for each pollutant or pollutants for which a
TMDL is established.  By including a specific reference to water quality
standards in addition to criteria, the final Guidance clarifies that, under
section 303(d), TMDLs must provide for the attainment of water quality
standards in their entirety, and not just their criteria components.

      The third sentence of proposed general condition 2 has been incorporated
into general condition 3 (now entitled "TMDL Allocations"), of the final
Guidance.  The final Guidance does not include the proposed language
specifying that TMDLs be prepared if the sum of existing point source and
nonpoint source loadings exceeds the loading capacity minus any specified
margin of safety for a substance.  This sentence was intended merely to
restate existing requirements under section 303(d) of the Clean Water Act and
TMDL regulations at 40 CFR 130.7, and therefore is unnecessary.  In EPA's
view, these provisions and other applicable requirements of the Guidance are
sufficient to ensure that TMDLs developed under this final Guidance will
provide for attainment of water quality standards.

c.    General Condition 3 - TMDL Allocations

      i.    Proposal:  This general condition was numbered as general
condition 9 in the proposal.  As proposed, this condition provided that
nonpoint source load allocations must be based on existing loading rates or on
anticipated increased loading rates unless a lower loading rate is expected to
occur within a reasonable period of time as a result of implementation of best
management practices or other control measures.  It also provided that the
portion of the loading capacity not assigned to nonpoint sources, or to an
MOS, or reserved for future growth is allocated to point sources.  Finally, it
stated that, upon reissuance, NPDES permits for these point sources must
include limitations' consistent with the WLAs in EPA-approved or EPA-
established TMDLs.

      ii.   Comments:  Some commenters advocated that the final Guidance only
allow the incorporation of nonpoint source reductions where such reductions
are required by legally enforceable mechanisms to ensure that reductions from
nonpoint sources are "reasonably expected to occur" within relevant time
frames.  Furthermore, the commenters suggested that a reasonable period for
such reductions would be eight years.  Another commenter supported the phased
approach for load allocations because it allowed an iterative process for
implementing nonpoint and point source controls.

      iii.  Final Guidance:   As part of its reorganization of the general
conditions in the final Guidance, EPA has renumbered proposed general
condition 9 to become general condition 3 in the final Guidance.  As part of
that reorganization, EPA has also incorporated into new general condition 3
the language proposed under the heading "Load Reductions" that defines the
elements of a TMDL.  EPA has also established subparagraphs within general
condition 3 of the final Guidance to correspond to the discussion in general
condition 3 of the elements of a TMDL, nonpoint source load allocations, point
source wasteload allocations, and monitoring.

      Specifically, EPA has added as subparagraph  (a) the statement from
proposed general condition 2 that TMDLs shall include wasteload allocations
and load allocations for nonpoint sources,  including natural background, such
that the sum of these allocations is not greater  than the loading capacity of
the water for the pollutant addressed by the TMDL, minus the sum of a
specified margin of safety and any capacity reserved for future growth.  EPA
has made only minor changes to the proposed language to clarify that the
nonpoint source load allocations  include natural  background conditions and to
link loading capacity to the pollutant for  which  the TMDL is being  developed.

      Subparagraph  (b)  comprises  the portions of  proposed general condition  9
pertaining  to nonpoint  sources.   These provisions were modified  in  general

-------
                               Section Vm.C: TMDLs                           255
condition 3 only to make consistent use of the term loadings and to clarify
that expectations regarding decreased loadings from nonpoint sources must be
based on a reasonableness standard.  The only significant comments EPA
received on proposed general condition 9 addressed nonpoint source issues.
EPA disagrees with the commenter's suggestion that nonpoint source reductions
be considered only when such controls are required by legally enforceable
mechanisms.  EPA suggests that means other than legally enforceable mechanisms
are available to ensure that nonpoint source reductions that are "reasonably
expected to occur" within a specified time frame actually do occur.  For
example, funding nonpoint source controls and using the monitoring component
of the phased approach to TMDL development, as described earlier in this
document, are means to assure that anticipated load reductions are actually
occurring.  Although EPA supports the use of a phased approach to TMDL
development where appropriate, EPA stresses that smaller load allocations to
nonpoint sources can be used to justify larger WLAs to point sources only when
the anticipated reductions in nonpoint source loadings are reasonably expected
to occur.

      EPA agrees with the comment that a TMDL can consider anticipated
nonpoint source loading reductions.  TMDLs developed using the phased approach
are based on the reasonable expectation that water quality standards will be
met in a reasonable period of time and that specific controls may be
implemented in stages.  What constitutes a reasonable period of time will vary
depending upon the situation.  Therefore, EPA will not specify any particular
period, such as eight years. The time period associated with these stages of
implementation will ultimately determine when water quality standards will be
met for a particular waterbody.  To the extent consistent with other
applicable law concerning schedules of compliance, permits issued after the
completion of a TMDL should be consistent with implementation schedules
established by the TMDL.

      Placed within'new subparagraph (c) are the provisions in proposed
general condition 9 pertaining to point source wasteload allocations and their
effect on NPDES permits.  Apart from including a reference to natural
background in connection with nonpoint sources, these provisions are unchanged
from the proposal.

      In the final Guidance, EPA added subparagraph (d) to address the
monitoring issues encompassed within the proposal's discussion of anticipated
decreases in pollutant loadings from nonpoint sources.  Subparagraph (d)
provides that, for load allocations established on the basis of (a) (iii) of
general condition 3,  monitoring data shall be collected and analyzed in order
to validate the TMDL's assumptions, to verify the anticipated load reductions,
to evaluate the effectiveness of controls being used to implement the TMDL,
and to revise the WLAs and load allocations as necessary to ensure that water
quality standards will be achieved within the time period established in the
TMDL.  This monitoring can be performed as part of the water monitoring
program established by the State (or at its election by the Tribe)  under 40
CFR 130.4, which specifies development and review of TMDLs, wasteload
allocations and load allocations as among the uses for such monitoring data.

d.    General Condition 4 - WLA Values

      i.    Proposal:  This general condition was numbered as general
condition 3 in the proposal.  As proposed, this condition specified that point
sources be regulated to ensure attainment of all downstream water quality
standards.  Proposed general condition 3 also recognized that TMDLs developed
for a particular waterbody may include WLAs for sources already covered by a
TMDL of a different geographic scope.  For example, a source-specific TMDL may
already be in place when a basin-wide TMDL is developed.  General condition 3,
as proposed,  provided that water quality-based effluent limits (WQBELs) in
NPDES permits for a particular pollutant be consistent with the most stringent
of the WLAs for that pollutant and point source included in any EPA-approved
or EPA-established TMDLs.  This provision was intended to assure that water

-------
256    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

quality standards will be met throughout a drainage basin, including in
downstream waters. .

      ii.   Final Guidance:  EPA did not receive significant comments on
proposed general condition 3.  Thus, the final Guidance retains  the substance
of the general condition with slight modifications, but EPA has  renumbered it
as general condition 4 in the final Guidance to reflect EPA's decision to move
proposed general condition 9 (TMDL Allocations) up to become new general
condition 3 in the final Guidance.

      This provision in the final Guidance, like the proposal, directs permit
writers to apply the most stringent of the WLAs included in any  EPA-approved
or EPA-established TMDL.  The final Guidance clarifies that this provision
applies only when more than one approved TMDL establishes a different WLA for
the same pollutant discharged by the same point source.  In addition to
renumbering this as general condition 4, EPA made one other change. The
proposed language stating that "point sources must be regulated  so as to
ensure attainment of all downstream water quality standards" has been deleted
in the final Guidance because it merely restated current law.  Specifically,
under existing CWA section 402 and 301(b)(1)(C), WQBELs in NPDES permits must
ensure attainment of all applicable water quality standards, including
downstream water quality standards.  Under 40 CFR 122.44(d)(1)(vii), such
WQBELs must be consistent with any available WLAs developed and  approved
pursuant to 40 CFR 130.7.

e.    General Condition 5  - Margin of Safety

      i.    Proposal;  This general condition was numbered as general
condition 4 in the proposal.  As proposed, this condition reiterated the
requirement in CWA section 303(d) that each TMDL include a margin of safety
(MOS) and described the manner in which the MOS is provided.  It also
reiterated EPA guidance that the MOS may be established either by setting
aside a portion of the loading capacity or by using conservative modelling
assumptions in deriving the TMDL.

      ii.   Comments:  Several commenters were concerned that it would be
inappropriate to leave determination of an MOS to the discretion of the permit
writer.  One commenter recommended that in order to facilitate basin-wide
consistency and maximum environmental protection, the Guidance should
implement an explicit MOS  factor equal to the Criterion Maximum  Concentration
(CMC) value  (which equals  one-half of the Final Acute Value  (FAV)) .  Other
commenters advocated specifying a specific  confidence level to use in modeling
a MOS.

      Several commenters believe that the MOS requirement is redundant  given
the number of conservative assumptions built into the criteria development
process and into the assumptions on fate and transport.

       Several commenters  were concerned that including uncertainties
regarding controlling pollutants from nonpoint sources into the  margin  of
safety merely shifted the  control burden to point sources without requiring
EPA, States or Tribes to regulate other sources of pollution.  They were
concerned that a larger MOS would result in a smaller WLA,  thus  requiring a
facility to discharge less and treat more while nonpoint  sources would  not be
controlled.

      iii.  Final Guidance:  Apart  from minor changes to  improve clarity and
renumbering to reflect  the overall  reorganization of procedure 3.B of appendix
F, the final Guidance is unchanged  from the proposal.  General condition 5
maintains flexibility for  the State or Tribe to consider  a  number of  factors,
including case-specific conditions  (e.g., availability and  quality of data)  in
establishing a margin of safety.  As  indicated  in 40 CFR 130.7(c)(1), the
margin of safety is intended to  account for uncertainty  in  the available  data
or in the actual effect controls  will have  on  loading  reductions and receiving

-------
                               Section Vm.C: TMDLs                           257
water quality.  EPA has determined that because of the need to reflect local
conditions and case-specific technical 'considerations, it is inappropriate to
specify a universal.MOS factor.  Although EPA recognizes the flexibility of
the State or Tribe to assess available information, EPA retains the authority
to disapprove a TMDL if EPA finds that a MOS is inadequate.

      In response to comments that the MOS has the effect of shifting the
burden of load reductions to point sources, EPA notes that the MOS requirement
does not compensate for failure to consider some sources  (e.g., nonpoint
sources as suggested by commenters) but rather is intended to account for any
technical uncertainty regarding both point and nonpoint source loading data
and the effectiveness of controls.  EPA acknowledges that the technical
uncertainty related to nonpoint sources may in fact be greater than
uncertainty regarding the effects of point sources.  EPA believes that the
phased approach to TMDL development provides, over time, an effective
mechanism for reducing technical uncertainty related to nonpoint sources.
This reduction in uncertainty will, over time, quantify and consider relative
contributions and water quality impacts and lead to appropriate levels of
control for both point and nonpoint sources.

      EPA disagrees with the commenters' suggestion that the MOS is redundant
given the conservative assumptions built into the criteria development and
into assumptions on fate and transport.  The MOS, as required by CWA section
303(d), is intended.to account for technical uncertainties regarding the
relationship between pollutant loads and water quality.  These factors are not
considered in the development of criteria and thus are not duplicative of
assumptions used in developing criteria.  Conservative assumptions in criteria
development are designed to address specific uncertainties and concerns
regarding extrapolations of toxicity data to individual or population
endpoints.  EPA also suggests that there should not be an issue of redundancy
regarding the fate and transport assumptions and the MOS.  The assumption of
no pollutant degradation for purposes of TMDL development is rebuttable when
scientifically valid field studies or other relevant information demonstrate
that degradation of the pollutant is expected to occur.

f.    General Condition 6 - More Stringent Requirements

      This general condition was numbered as general condition 4 in the
proposal.  As proposed, this condition provided that States may employ section
510 of the CWA to establish TMDLs more stringent than those developed pursuant
to procedure 3. The condition reiterated the reserved right of States to
require more stringent controls than those required under the CWA.

      EPA received no significant comments on this provision.  The proposed
language is modified slightly in the final Guidance to clarify that both
States and Tribes may employ section 510 and to correct a typographical error.
It has also been renumbered as general condition 6 to reflect the overall
reorganization of procedure 3.B of appendix F.

g.    General Condition 7 - Accumulation in Sediments

      i.    Proposal:  This general condition was numbered as general
condition 6 in the proposal.  As proposed, this condition specified that TMDLs
must be stringent enough to prevent accumulation of the pollutant of concern
in sediments to levels injurious to designated or existing uses, human health,
wildlife and aquatic life.  It also specified that TMDLs consider
contributions to the water column from sediments inside and outside applicable
mixing zones.  Although TMDLs are calculated on the basis of pollutants in the
water column, the preamble to the proposal indicated that all sources of
pollution, including sediment re-release of pollutants to the water column,
would need to be considered in establishing TMDLs.

      ii.    Comments:  EPA received numerous comments on this condition.  A
number of commenters disagreed with the proposal. Several suggested that

-------
258    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

proposed general condition 6 be deleted until EPA finalizes and  implements a
national sediment strategy.  One commenter suggested that proposed general
condition 6 be optional depending on the availability of information or, if
not, that general condition 6 should be removed entirely.

      Several commenters stressed the importance of considering  the release of
toxics from contaminated sediments, which in many instances may  result in a
failure to meet water quality standards.  Several commenters, while agreeing
with the need to consider all sources of pollutants, including sediment
release or resuspension of pollutants,  believe that methodologies do not
currently exist to accurately reflect the sediment re-release process.  One
commenter suggested that sediments should only be accounted for  by
concentrations measured in the water column and that any additional factors
would be duplicative.  Commenters recommended that EPA continue  to work on
National guidance fbr such methods and suggested that any process for
developing sediment criteria should be subject to a peer review process.

      iii.  Final Guidance:  The final Guidance retains the requirement that
TMDLs reflect processes such as re-release of pollutants from sediments,
because, as noted by many commenters, contaminated sediments are often a
source of pollutant loading to the water column and thus may cause or
contribute to an exceedance of water quality standards.  However, EPA has
modified this provision to clarify that such contributions should be
considered only where appropriate and where sufficient data are  available.
EPA has renumbered this provision as general condition 7 to reflect the
overall reorganization of procedure 3.B of appendix F.

      EPA agrees with commenters that existing methodologies may not fully
reflect all aspects of the sediment re-release process.  However, EPA recently
proposed its Contaminated Sediment Management Strategy  (EPA 823-R-94-001) for
public comment, 59 FR 44880, (August 30, 1994, available in the  docket), and
is continuing to develop methodologies to evaluate the sediment  re-release
process.  The strategy proposes establishing standardized test methods to
assess whether sediments are contaminated and proposes to continue supporting
research on the re-release of pollutants from contaminated sediment.  Under
the strategy, EPA  would develop new biological methods to assess the
ecological and human health effects of sediment contaminants, sediment
wasteload allocation models, and technologies for remediation of contaminated
sediment.  EPA is also working to develop chemical-specific sediment quality
criteria.  This process will involve review from outside parties.  See 59 FR
2652, January 18, 1994 for further information.

      EPA is moving forward with many of the activities described in the draft
Contaminated Sediment Management Strategy and expects many of these activities
to be completed in time to support State and Tribal procedures under part 132.
The final Contaminated Sediment Management Strategy and associated outreach
efforts will support States and Tribes in implementing general condition 6.
Therefore, EPA disagrees with the comment that this condition needs to be
deleted until EPA finalizes the Strategy.

      Several commenters suggested that situations may exist where information
is not available to determine the nature and extent of contaminated sediments'
contributions of pollutants to the water column.  EPA has modified the final
Guidance to specify that contributions to the water column from  contaminated
sediments be included where appropriate.  It may be considered appropriate to
reflect contributions of pollutants from contaminated sediment only where data
exist regarding sediment re-release of the pollutant(s) of concern.  Where
such information does exist, however, the TMDL must account for  contributions
from contaminated sediments.

      In the final Guidance, EPA has reversed the order of the two sentences
appearing in proposed general condition 6 in order to emphasize  that
contaminated sediments can be sources of pollutants to the water column  and
that TMDLs need to account for contributions from that source.   As in the

-------
                               Section Vm.C: TMDLs                           259
proposal, in addition to specifying that sediment re-release of pollutants
shall be considered where appropriate, the final Guidance provides that TMDLs
must be sufficiently stringent so as to prevent injurious accumulation of the
pollutant of concern in sediments, because such injurious accumulations would
represent exceedances of Water quality standards  (at a minimum by impairing a
designated aquatic life use).

h.    General Condition 8 - Wet Weather Events

      i.    Proposal:  This general condition was numbered as general
condition 7 in the proposal.  As proposed, this condition recognized that some
of the TMDL development procedures may be appropriate for wet weather events
 (e.g., nonpoint sources, storm water discharges, and combined sewer
overflows).  However, the proposed TMDL implementation procedures did not
include explicit procedures detailing how to develop TMDLs to reflect wet
weather events; rather it left maximum flexibility to the States and Tribes on
how best to accomplish this.  The preamble discussion of proposed general
condition 7 interpreted that condition as providing that loadings from wet
weather events be included in establishing TMDLs, but the proposal itself was
silent on this point.

      ii.   Comments;  Several commenters suggested that proposed general
condition 7 needed to clarify that all TMDLs must include consideration of
necessary waste load allocation and load allocations for wet-weather pollutant
contributions.  Another commenter pointed out that certain POTWs face
compliance difficulties as a result of wet weather flows.  The commenter
suggested that these factors, which are beyond the control of the POTW, be
considered in developing permit limits. Several commenters asserted that wet
weather contributions cannot be accurately estimated and therefore suggested
this general condition be removed all together.

      iii.  Final Guidance:  The final Guidance retains the proposed language
on wet weather flows with minor modifications and an additional sentence for
clarification purposes.  This provision has been renumbered in the final
Guidance as general.condition 8 to reflect the overall reorganization of
procedure 3.B of appendix F.  EPA agrees with the commenter's suggestion that
this general condition should be clarified to state specifically that TMDLs
must consider pollutant loadings resulting from wet weather events, where
appropriate and where sufficient data are available.  EPA believes TMDLs
reflecting wet weather events would be appropriate where such events
contribute the pollutant(s) during the flow conditions for which the TMDL is
being developed.  For example, the TMDL for a pollutant that has an annual
averaging period (e.g., dioxin) would need to consider loadings from wet
weather events because such events can occur during the yearly averaging
period. However, a TMDL  based on a 7-day critical low flow (e.g., lead) for a
pollutant that has a 4-day averaging period would not directly consider
loadings from wet weather events because such events are unlikely to occur
during critical 7-day low flows. Contributions from previous wet weather
events would be considered through load allocations to the sediment.  In
addition,  a TMDL based on dynamic or stochastic water quality model would
include all dry and wet weather loadings from all sources.  In any case, where
the TMDL for the receiving water accounts for loadings that occur from wet
weather events, the resulting WLAs, including those for POTWs, must be
consistent with the TMDL and WLAs.  The only exception to this is where the
POTW discharge meets the definition of wet weather point source under 132.5.
The final Guidance does not regulate wet weather point sources.

      Many nonpoint sources and wet weather point sources as defined at
section 132.2 of this Guidance typically have their greatest impacts following
storm events and the influx of pollutants from these events needs to be
factored in when establishing a TMDL to ensure attainment of water quality
standards.  Accordingly, EPA has inserted language to clarify this point in
the final Guidance and has amended 132.4 (e) (1) to provide specifically that
procedure 3 applies to wet weather events, as appropriate.  Like the proposal,

-------
260    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

the final Guidance does not require a specific procedure to address wet
weather flows, but rather leaves it to the discretion of the State or Tribe to
choose the most appropriate procedure, considering ail relevant  facility
specific, .pollutant specific, and receiving water specific factors.

      In EPA's view, this clarification will not subject POTWs to any
additional burden.  Any adjustments to a POTW's permit conditions to account
for wet-weather flows should be addressed through the NPDES permitting  and
enforcement policies and procedures.  Finally, EPA disagrees with the comments
asserting that wet weather contributions cannot be accurately estimated.  A
number of models currently exist to generate loadings estimates  from a  range
of wet weather events.  EPA is working on additional guidance on assessing
pollutant loadings associated with CSOs and nonpoint sources (see "Technical
Guidance for Estimating Total Maximum Daily Loads (TMDLs):  Integrating
Steady-State Episodic Point and Nonpoint Sources, draft, June 1994, available
in the docket).

i.    General Condition 9 - Background Concentrations of Pollutants

      This general condition was numbered as general condition 8 in the
proposal.  As proposed, this condition established procedures for determining
representative background concentrations of pollutants to assure that
background concentrations are consistently considered in TMDL development
among the Great Lakes States.  The proposal included provisions  for
calculating background.  The proposal defined background, described the choice
of data set, the use of the geometric mean, and the treatment of data sets
with data points abpve and below detection.  EPA received no significant
comments on the definition of background and the proposed language is retained
in the final Guidance with only minor changes to account for the use of the
term in procedure 5.  The proposal, comments and the final Guidance for each
provision are discussed below.  EPA renumbered this provision as general
condition 9 in the final Guidance to reflect the overall reorganization of
procedure 3.B of appendix F.

i.    Choice of Data Set

      (A)   Proposal:  The proposal provided that the representative
background concentration for a pollutant shall be established as the geometric
mean of one of three possible data sets: available ambient water column data
(e.g., ambient monitoring data), representative caged fish tissue data, or
representative pollutant loading data.  When more than one data  set exists,
best professional judgment  (BPJ) would be used to determine which data  set
most accurately estimated background concentrations.  The preamble to the
proposal stated that, in general, ambient monitoring data are preferred over
other sources of data.  The preamble also recognized that there  may be
instances where other data sets may be more appropriate, such as where  ambient
data are not available, or where ambient data are not as informative or
reliable as either caged fish tissue data or pollutant loading data because of
limits in analytical detection methods.

      (B)   Comments:  Several commenters supported EPA's proposal to allow
States and Tribes to choose among data sources. Others suggested that,  by
allowing a choice of data sets, there was too much discretion allowed to the
State or Tribe in establishing background levels and suggested that EPA
provide more specific guidance on the choice of data sets.

      One commenter suggested that States and Tribes should be required, where
possible, to eliminate unrepresentative data from the data set using factual
information and statistical methods.  Commenters suggested that  more recent
data should take precedence over older data even when the more recent data  set
is smaller.  Furthermore, they believe that data more than five  years old
should not be considered.  One commenter suggested that  fish tissue and
pollutant loading calculations should be rejected as acceptable  data sets when
those calculations predict background concentrations above the criteria for

-------
                               Section vm.C: TMDLs                           261
ambient monitoring data and such concentrations were not detected by ambient
monitoring.

      Several commenters advocated that only ambient data be used to estimate
background concentrations. Other commenters wanted the Guidance to require
regulators to use ambient monitoring data to calculate background
concentrations of pollutants when such data is available.

      A number of commenters disagreed with the requirement to consider caged
fish tissue data in calculating background concentration because procedures
for the use of caged fish analysis have not been thoroughly evaluated,
validated, or standardized.  Several commenters believe that the quality data
necessary to provide accurate background data using the caged fish approach is
not available.  Commenters suggested that using  resident fish tissue as a
basis for deriving background would be more accurate.  A commenter further
suggested that EPA attempt to calibrate the fish tissue and pollutant loading
models with real data.  Commenters also requested more specific procedural and
technical information relating to use of caged fish data.

      (C)   Final Guidance: In response to comments and concerns, EPA has
added resident fish tissue data as a fourth specified data set available for
calculating background.  Apart from that, EPA retains the proposed language
with only minor modifications to ensure clarity and avoid redundancy.

      In the final Guidance, EPA has consolidated into a single section the
list of available data sets and the basis for determining what available data
is acceptable for use in calculating background.  These provisions are now
included in the subparagraph specifying calculation requirements.  The final
Guidance retains flexibility for States and Tribes to choose from among a
number of data sets, including fish tissue data, in calculating background
concentrations.  EPA concludes that because of wide variability in the
suitability of available data for a particular situation and because of site-
specific considerations, use of BPJ is appropriate to make case-by-case
determinations.  EPA recognizes that more recent data, with improved detection
or quantification levels may be more appropriate, while some older data with
poorer detection or quantification levels may be less acceptable.  However,
EPA recognizes that, in some instances, the older data may be the only data
available may be the only representative data of sufficient quality from which
to make decisions and thus is not establishing a prohibition on the use of
older data.  In the*final Guidance, the State or Tribe retains the flexibility
to use BPJ to eliminate unrepresentative data or to give greater weight to the
most recent data as suggested by commenters.  States and Tribes may also use
statistical techniques to identify and eliminate unrepresentative data.

      The final Guidance thus does not include more specific direction to
limit the use of any particular data set.  Although EPA agrees with the
commenters1 suggestion that ambient monitoring data are generally preferred
over other data sources, there may be situations where ambient data are not
available, or are not as informative or reliable as either fish tissue or
pollutant loading data because of limits in analytical detection methods.
Because of limits in existing technologies, ambient data may still yield non-
detects above criteria levels.  Fish tissue data and pollutant loading data
may be particularly useful alternatives for these situations.

      EPA recognizes that caged fish tissue studies may have limitations in
that such studies may not fully account for duration of exposure and food
chain magnification.  However, EPA has determined that such studies should be
considered with other data sources in choosing among data sets to calculate
background concentration.  Aquatic organisms can serve as valuable indicators
of whether water quality standards are being attained.  The final Guidance
also authorizes the'use of resident fish tissue data, as suggested by
commenters, because of concerns regarding food chain effects and in response
to concerns about the lack of caged fish tissue data.  Use of resident as well

-------
262    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

as caged fish tissue data is intended to provide more latitude in selecting
the appropriate data set.

      Like the proposal, the final Guidance does not provide a methodology to
use in translating fish tissue concentrations to a water column concentration,
or for evaluating their validity.  EPA agrees that care should be exercised in
determining what fish tissue data are representative of background pollutant
concentrations and encourages permitting authorities to consult EPA guidance
on this topic.  For,example, EPA recommends that when fish tissue data are
available from resident fish the geometric mean is divided by the
bioaccumulation factor pursuant to the methodology in appendix B of this final
Guidance, to yield estimated ambient concentrations.  See Assessing Human
Health Risks from Chemically Contaminated Fish and Shellfish: A Guidance
Manual (USEPA, September, 1989, EPA-503/8-89-002, available in the docket).

       EPA believes that best professional judgment should be used to
determine if caged fish tissue data is appropriate for calculating background
concentration in -a given situation.  Furthermore, the use of caged fish tissue
data is not required unless no other data exist to calculate background
pollutant concentrations.  Even in a situation where only caged fish tissue
data exists, a facility always has the option to collect alternative data that
more accurately reflects background concentrations  (e.g., ambient monitoring
data).  In addition, the final Guidance does not require that new fish tissue
"studies" be conducted in the absence of existing fish tissue data.

ii.   Geometric Mean

      (A).  Proposal;  The proposal specified that the representative
background concentration for a pollutant shall be established as the geometric
mean of one of the selected data sets described in that paragraph and the
preamble offered guidance for performing the calculations.  EPA is retaining
the proposed language in the final Guidance.  The Agency received no
significant comments on this provision.

      As the preamble to the proposal explained, a geometric mean is
calculated for the set of data chosen to represent background conditions.  The
geometric mean calculated is based on both measured concentrations and an
appropriate methodology for treating measurements below quantification levels.
For pollutant loading data, the geometric mean should be taken of pollutant
loading data from individual sources.  The individual means of each of the
individual sources should then be added to estimate total loading to the
receiving water.  Background concentration is calculated by dividing total
loadings by the volume of water available at the appropriate design flow.
Design flow will vary depending on the criterion being implemented at the
point immediately upstream of the watershed, water body or water body segment
for which the TMDL, WLA in the absence of a TMDL, or preliminary WLA for the
purpose of determining reasonable potential under procedure 5 of this Guidance
is being established.  For further discussion, see the preamble to the
proposal at 58 FR 20929.

      (B).  Final Guidance:  EPA received no significant comments on this
provision.  EPA believes that the use of the geometric mean is the best
approach for calculating a median concentration from data chosen to represent
background conditions.  An arithmetic mean would be one appropriate method for
calculating median values when a sample concentration is as likely to be above
the true average concentration as it is to be below the true average
concentration.  However, concentration measurements in fish tissue and water
are more likely to be below the true average concentration.  Under these
conditions, the geometric mean is an appropriate estimator for the median
while the arithmetic average will generally produce a value that is higher
than the median.  More explicitly, fish tissue and water concentration
measurements generally follow positively skewed probability distributions
where the median is appropriately estimated by the  geometric mean.

-------
                               Section Vm.C: TMDLs                           263
iii.  Data Points Above and Below Detection

       (A)   Proposal:  The proposal allowed the use of best professional
judgment to determine which data points are acceptable.  However, within a
given data set, some data points may indicate that the pollutant was not
present at levels capable of being detected by the analytical method used.
For these data points, the true concentration of the pollutant can be zero or
is somewhere between zero and the detection level of the analytical method.
Other data points may indicate that the pollutant was detected, but at levels
below which the analytical method is capable of reliable quantification.  For
these data points, the true concentration will be between the detection level
and the quantification level of the analytical method.  Finally, there may be
data points showing reliably quantified levels of the pollutant.        The
proposed Guidance specified that the following assumptions be used in
calculating background when, within a data set, some data points are
determined to be above and others are below the detection level.  The proposal
included the following assumptions: data points reported at levels below
detection shall be set equal to one half of the detection level; and data
points reported at levels greater than the detection level but less than the
quantification level, shall be set equal to the midpoint between the detection
level and the quantification level.  If all acceptable available data points
in a data set are reported as below the detection level for a specific
pollutant, then all the data for that data set are assumed to be zero.

      Section 132.2 of the proposed Guidance included a definition of
detection level that is identical in substance to the definition at 40 CFR
136.2 (f).  There is no similar long-established definition of the term
quantification level.  However, the proposal defined the quantification level
as the concentration at which a particular substance can be quantitatively
measured using a specified laboratory procedure.  EPA solicited comment on the
definition and on the issue of whether a particular degree of confidence
should be specified.

       (B)   Comments:  EPA received a number of comments on the proposal to
assign values equal to one-half of the detection level to data points reported
below the detection level when other data points in the data set were reported
above the detection level.

      Several commenters supported the use of one-half of the detection level
as a default.  Another commenter suggested that if 25% or more of the data
points are quantifiable, the remaining values reported as less than the
detection limit should be zero. One commenter advocated that the requirement
to use one-half of the detection level as a background concentration be
deleted and the evaluation left to best professional judgment.  Another
commenter recommended that if a large proportion of the data is reported as
non-detect,  assumptions regarding what value to assign should be left up to
the permitting agency and that such determinations need to be made on a ca.se-
by-case basis rather than through the application of a general rule.  Several
commenters wanted EPA to allow the use of appropriate statistical methods for
data sets that include a large number of values below the detection limit and
further advocated that the final Guidance cite examples of such statistical
methods.  One commenter suggested that a statistically valid sliding scale be
used to assign concentration values to any non-detect measurements.  Another
commenter expressed concern that the proposed approach will result in
unrealistically high background concentrations for data sets with a large
share of measurements below the detection level and suggested that the final
Guidance include methods presented in "NCASI Technical Bulletin No. 621."
Several commenters supported the use of one-half the detection level when
calculating means or averages from data sets that include non-detect values.

      EPA also received comments addressing situations where some of the data
is between the detection level and the quantification level.  One commenter
suggested that the final Guidance require the quantification level be used as
the default value in determining the mean for pollutants that have caused or

-------
264    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

contributed to fish advisories downstream.  Another commenter suggested that
when data points are below the detection level or quantification level, zero
or a default percentage of the criteria value, when the criteria value  is also
below the level of detection, should be assumed.

      In situations where all the data points in a particular data set  are
below detection, several commenters agreed with the proposal that these data
points should be assumed to be zero.  One commenter suggested that for  a data
set of more than ten data points, the proposal should apply, but that if there
are fewer  than ten data points and all the data points are below detection,
background shall be assumed to be one-half the detection level.    Several
commenters supported EPA's definition of quantification level.  A few
commenters did not support including the quantification level definition.
Another commenter suggested that the definition of quantification level should
be the same as that used in setting the Compliance Evaluation Level  (CEL) for
determining permit compliance in proposed procedure 8.  The CEL was defined in
the proposal, as the level at which compliance with an effluent limit is
assessed.  Some commenters advocated that the term "detection level" be
changed to "method detection level" since the proposal defined detection level
the same as method detection level is defined in 40 CFR 136.

      (C)   Final Guidance;  The final Guidance recognizes the need for
flexibility when calculating background using a data set containing data
points both above and below the detection level or quantification level.  EPA
has concluded that, for these data sets, although default values of one-half
of the reported detection level for data points reported as below detection,
and the mid-point between the detection level and quantification level  for
data points reported below the quantification level and above the detection
level, are a reasonable and appropriate estimate for purposes of calculating
background concentration, they are not the only reasonable and appropriate
approach.  As many of the commenters pointed out, there are a number of
commonly accepted statistical approaches to evaluating mixed data sets  (also
known as censored data sets).  Therefore, in the final Guidance, States and
Tribes are required to use commonly accepted statistical techniques to
evaluate data sets containing values both above and below the detection level.
Commonly accepted statistical techniques can include a variety of approaches,
including the use 'of default values as proposed.  Some commonly accepted
statistical techniques are outlined in Chapter 14 of Statistical Methods for
Environmental Pollution Monitoring  (Richard 0. Gilbert; published by Van
Nostrand Reinhold) and Truncated and Censored Samples  (A.Clifford Cohen;
published by Marcel Dekker).

      Because there is no universal method to reliably quantify pollutant
concentrations below the detection level, EPA believes that using a default
value of one-half of the reported detection level is a reasonable balance of a
State and Tribes' obligation to provide dischargers with an appropriately
stringent WLA and the statutory requirement that TMDLs ensure the attainment
of water quality standards.  The same reasoning applies when calculating WLAs
in the absence of a TMDL.  Likewise, EPA has concluded that the  reasoning
above also supports-using the mid-point between the detection level and
quantification level as an acceptable, reasonable approach for dealing  with
data points above the detection level but below the quantification level.  In
this situation, EPA does not endorse using the detection level as a default
value.  Using the detection level as a default to calculate background  could
result in WLAs that would not provide the necessary assurances,  as required by
the CWA, that water quality standards will be attained.  Again,  EPA believes
this is of particular concern for pollutants with criteria values below the
level of detection.

      EPA retains the approach in the proposal that assigns zero values to
data points when all the data in the data set are below the level of  detection
for the particular pollutant.  When all analytical tests for a chemical result
in determinations that fall below the detection level, one would have,  in
effect, a finding that the target analyte cannot be known with confidence to

-------
                               Section Vffl.C: TMDLs                           265
be present in any of the samples.  Where this is the case, and no other
analytical results are available to indicate that the chemical may be present
in any sample, EPA believes the appropriate finding is that the chemical is
not present.  In contrast, as described above, where analytical tests show the
chemical to be present in some samples, EPA believes that an appropriate and
reasonable approach is to assume that the chemical may be present even in
those samples in which the chemical is not detected, and therefore assign a
value to the iron-detect measurement of one-half the value of the detection
level.  Although EPA recognizes that this could potentially result in an
underestimate of background concentration for a given pollutant, it could also
result in an overestimate of background concentration for a given pollutant.
EPA believes that this approach is reasonable because it strikes a balance
between the desire to accommodate dischargers with a reasonable WLA and the
CWA requirement that TMDLs ensure the attainment of water quality standards.
In addition, as discussed in the preamble to the proposal, there is no
universal method to reliably quantify pollutant concentrations below the
detection level.  States and Tribes may want to consider a more stringent
approach, whether as a general matter or in establishing individual TMDLs, as
authorized by section 510 of the Clean Water Act and general condition 6 of
this procedure.

      EPA is retaining the proposed definition of "quantification level" for
purposes of this procedure.  EPA has concluded that a standard definition of
quantification will improve consistency among States and Tribes in the Great
Lakes System when calculating the background concentration of pollutants.
Consistency among Great Lakes States and Tribes is one of the major objectives
of the final Guidance, although the definition is broad enough to allow
consideration of other factors as appropriate.  For example, a State or Tribe
may consider the nature of the pollutant, the method being used, and the past
performance of the testing facility or laboratory.  In addition, EPA agrees
with commenters that asserted the proposed definition of "detection level" is
confusing since it is substantively identical to the existing 40 CFR 136
definition of "method detection level."  EPA has therefore, renamed "detection
level" to "method detection level" to avoid confusion and maintain
consistency.  Substantively, the text of the definition was not changed.

      A State or Tribe's use of procedures for estimating representative
background concentrations of pollutants will also be reviewed by EPA on a
case-by-case basis when it approves or disapproves State or Tribal TMDLs
submitted under section 303(d).  A State or Tribes's approach will be reviewed
as part of the program submission and adoption process set forth at section
132.5 of this Guidance.  EPA also retains the authority to object to an NPDES
permit containing a WQBEL derived from a WLA in the absence of a TMDL if EPA
determines that the estimates of representative background concentrations were
unreasonable and that the permit would therefore not implement water quality
standards as required by section 301(b)(1)(C) of the CWA.

      The only substantive change to the proposal is the addition of language
authorizing the use* of commonly accepted statistical techniques in evaluating
data sets consisting of values both above and below the method detection
level.  EPA added this additional flexibility in response to a number of
comments supporting the use of such approaches. EPA encourages the use of
commonly accepted techniques.  Such statistical approaches can be a useful
tool when dealing with sparse data sets.  In all other respects, general
condition 9 is substantively the same as the proposal, except that it states
explicitly that it applies to data sets having values both above and below the
method detection level.  The final Guidance has also been modified to ensure
that the term "reported" is used consistently throughout this condition.

j.     General Condition 10 - Effluent Flow

      General condition 10 in the proposal provided that, if WLAs are
expressed as a concentration of a pollutant in a discharge, the TMDL must also
specify the point source effluent flow assumed in deriving the WLA.  Since

-------
266    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

TMDLs are based on mass loadings to a system, the assumed flows used to derive
the mass loadings need to be specified.  This provision also facilitates the
establishment of mass loading limitations in NPDES permits as required by
procedure 7 of appendix F.  Substantive comments on establishing an effluent
flow are addressed in the loading limits section of this document  (section
VIII.G).  The final Guidance retains the proposed language with minor changes
to improve clarity.  This should assure that common assumptions are used in
establishing TMDLs and corresponding NPDES permit limits.

k.    General Condition 11 - Reserved Allocations

      i.    Proposal:  General condition 11, as proposed, provided that once a
TMDL for a particular pollutant is in place for a waterbody, a new source or
new discharger can discharge that pollutant to the waterbody only if its
loadings are consistent with the existing TMDL.  The existing TMDL must
include a reserved allocation for future growth or the TMDL must be revised to
include an allocation for the new discharge.

      ii.   Comments:  Many commenters suggested that the provision related to
the use of "reserved allocations" for future growth should be strengthened to
require that a specific share be set aside.  One commenter suggested that EPA
should describe the procedure to determine a reasonable reserve capacity for
future growth while allowing the State the discretion to make this
determination.

      iii.  Final Guidance:  The final Guidance makes only minor modifications
to change the title from "New Source or Discharger" to "Reserved Allocations"
and to clarify that the general condition applies only to new discharges of
the particular pollutant for which the TMDL was developed. The purpose of
general condition 11 is to assure that the impacts of new pollutant sources
will be considered.. Without such a condition, a TMDL might fail to take into
account new discharges of the pollutant of concern with the result that the
TMDL would need to be revised in order to allow the new discharge.  While EPA
appreciates the comments urging that this provision be strengthened by
establishing a specific procedure for reserving capacity for future growth,
EPA believes that States and Tribes are in the best position to determine a
reasonable allocation for future growth and thus the final Guidance provides
them the flexibility to make the determination.  States and Tribes will need
to make the determination by balancing local and economic development with
water quality requirements.

4.    Special Provisions for BCCs

      a.    Proposal;  The proposed Guidance recommended restrictions on the
introduction of bioaccumulative chemicals of concern  (BCCs) in the Great Lakes
System by specifying, in general, that mixing zones for existing dischargers
of BCCs be eliminated within 10 years of the effective date of this final
Guidance, and for new dischargers or new sources, that no mixing zone for BCCs
be provided.  The proposal also specified that mixing zones calculated during
the ten year phase-out period prior to elimination of mixing zones for BCCs
would be established using the mixing zone provisions for non-BCCs, set forth
in sections C and D^ of proposed options A and B.  The proposal allowed a
limited exception to the elimination of mixing zones for BCCs when water
conservation measures result in an increased concentration but lead to an
overall reduction in load.

      b.    Comments:  EPA received numerous comments both supporting and
opposing the provision to eliminate mixing  zones for BCCs.  Many commenters
supported the phase-out of mixing zones for all discharges of BCCs within the
Great Lakes System.  Several of these comments pointed out that the proposed
elimination of mixing zones is consistent with the Great Lakes Water Quality
Agreement's emphasis on limiting any future introduction of persistent toxics
into the Great Lakes System.

-------
                               Section Vm.C: TMDLs                           267
      A number of coiranenters urged that the elimination of mixing zones be
broadened to include all persistent toxic chemicals, not just BCCs. Several
commenters specifically mentioned the need to address lead and cadmium.  One
commenter suggested that in order to ban the discharge of toxic substances
into the Great Lakes Ecosystem, EPA needs to ensure that all sources of
pollution, including air, contaminated sediments and runoff, are controlled
and that EPA should require comprehensive pollution prevention programs
throughout the basin.  One commenter suggested that while mixing zones for
BCCs should, in general, be eliminated, mixing zones should be allowed under
strict conditions, such as when pollution prevention measures are implemented
and have resulted in reduced loadings.

      Many commenters opposed the elimination of mixing zones for existing
dischargers of BCCs and believe that the mixing zone prohibition is
unattainable and inefficient.  Many commenters mentioned that there would be
high costs associated with elimination of mixing zones in return for limited
environmental benefits.  Commenters claimed that the elimination of mixing
zones requiring dischargers to meet criteria end of pipe would, in effect,
result in a zero discharge requirement.

        Several municipalities mentioned that they would be unable to impose
additional requirements on their industrial dischargers that would allow them
to meet water quality goals without mixing zones.  They also felt the phase-
out of mixing zones'for BCCs would provide a disincentive for them to take on
new industrial dischargers.

      Many commenters suggested that if mixing zones are phased out,
reductions must be limited to levels that are economically and technically
feasible.  Several commenters advocated that additional pollution prevention
measures also be required to help minimize the release of BCCs into the Great
Lakes.

      Commenters also suggested that eliminating mixing zones for BCCs may not
be the most cost-effective means of reducing certain BCC loadings  (e.g.,
mercury) and that reductions need to come from other sources, such as
atmospheric deposition. Commenters suggested that greater load reductions
would occur if nonpoint sources were targeted for controls.   Commenters
asserted that extraordinary controls on point sources of BCCs will have little
impact on water quality because point sources only contribute a small
percentage of the total load of BCCs to the basin and that the major loading
of BCCs is from nonpoint sources.  Several commenters claimed that the
increased stringency in permits would not lead to an overall improvement in
ambient water quality and that limits without mixing zones would be unduly
restrictive.

      Numerous commenters stated that the elimination of mixing zones has no
scientific merit and is merely a policy decision. Many commenters pointed out
that existing EPA technical guidance, such as the Technical Support Document
for Water Quality-based Toxics Control (TSD), does not disallow mixing zones.
Commenters suggested that existing EPA and State policy should determine when
mixing zones are appropriate.  One commenter advocated that methods
recommended in the TSD be used to predict the fate and transport of pollutants
such as BCCs and that these approaches be used to develop TMDLs for the BCCs
rather than disallowing mixing zones.

      A number of commenters indicated that the proposed time frame for the
phase-out is reasonable. One commenter suggested that the final Guidance
should make it clear that the mixing zone phase-out for existing discharges
will be effective ten years after the Guidance is incorporated into state
rules rather than ten years after publication of the final Guidance.  Numerous
environmental groups suggested that the implementation period is too long and
recommended an accelerated phase-out of mixing zones for BCCs.  Many supported
a 5-year phase-out rather than 10 years.   Commenters specifically suggested
partial reductions of mixing zones, in terms of the available dilution ratio,

-------
268    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

be used at the time of the first NPDES permit reissued after the final
Guidance is published.

      One commenter advocated that EPA establish a mass loading-based limit on
the proposed water conservation exemption by placing a cap on the increased
concentration allowed in exchange for water conservation measures.  Commenters
supported the proposed restriction that the mixing zone granted under this
provision be consistent with the mixing zone provisions of sections C
(deriving TMDLs for discharges to Lakes) and D  (deriving TMDLs for discharges
to Tributaries) of proposed procedure 3.

      c.    Final Guidance:  The final Guidance retains the ten-year phase-out
of mixing zones for'BCCS and the immediate elimination of mixing zones for new
discharges, which are defined for the purpose of procedure 3.C as  (i)
discharges from new Great Lakes dischargers; or (ii) a new or expanded
discharge from an existing Great Lakes discharger.  All other discharges of
BCCs are defined as existing discharges.  The final Guidance is consistent
with the Steering Committee's policy that every reasonable effort be made to
reduce all loadings of BCCs to the Great Lakes System.  The Steering Committee
recommended that mixing zones be eliminated for BCCs as a way to reduce mass
loadings to the Great Lakes.   However, in response to numerous comments that
the proposed phase-out may be technically or economically infeasible, the
Guidance does provide a limited exception to the elimination of mixing zones
for existing discharges of BCCs to the Great Lakes System.  This exception is
provided only in limited circumstances when the State or Tribe finds that the
discharger seeking the exception is implementing controls to reduce the BCCs
for which a mixing zone is sought to the maximum extent possible yet still
cannot meet a WQBEL based on no mixing zone.    EPA has concluded, after
considering all the comments, that elimination of mixing zones for BCCs may
not be reasonable in all circumstances, and thus has provided for a limited
exception  (described below) in the final Guidance.

      The final Guidance uses the terms "new Great Lakes discharger" and
"existing" Great Lakes discharger as discussed in section II.B of this
document.  In the final Guidance, the time deadline has been clarified to
provide that mixing zones for existing Great Lakes dischargers will be phased-
out within twelve years from the date of publication of the final Guidance.
The proposal set the phase-out at ten years, but this has been modified in the
final Guidance to reflect explicitly the two years allowed for State and
Tribal adoption of implementation procedures for the final Guidance.  The
phase-out deadline for new Great Lakes dischargers is stated in the final
Guidance as two years after publication of the final Guidance.

      The phase-out of the elimination of mixing is consistent with existing
EPA regulations and'guidance, and the Great Lakes Water Quality Agreement.
EPA regulations provide that States and Tribes may, at their discretion,
provide for mixing zones as part of their State and Tribal water quality
standards  (40 CFR 131.13).  However, the Technical Support Document for Water
Quality-based Toxics Control recommends that States and Tribes provide a
definitive statement in their water quality standards as to whether or not
mixing zones are allowed and suggests that:  "As our understanding of
pollutant impacts on ecological systems evolves, there may be cases identified
where no mixing zone is appropriate."   For example, EPA's Water Quality
Standards Handbook  (EPA-823-B-93-002) states that "Careful consideration must
be given to the appropriateness of a mixing zone where a substance discharged
is bioaccumulative, persistent, carcinogenic, mutagenic, or teratogenic."  The
Handbook recommends that "denial  (of mixing zones) should be considered when
bioaccumulative pollutants are in the discharge."

      A general principle  of the Great  Lakes Water Quality Agreement  (see
Annex 2 Paragraph 2.(d)) supports the elimination of point source  impact zones
(i.e., mixing zones) for toxic substances as consistent with the overall
policy of the virtual elimination of persistent toxic substances.  According
to the Agreement, pending  the achievement of the virtual elimination of

-------
                               Section Vm.C: TMDLs                           269
persistent toxic substances, the size of such zones shall be reduced to the
maximum extent possible by the best available technology so as to limit the
effects of toxic substances in the vicinity of these discharges.

      Although levels of certain bioaccumulative chemicals of concern  (BCCs)
have significantly declined in the Great Lakes System in recent years, EPA
estimates that under current loadings it will take years, perhaps decades, for
fish tissue concentrations of certain BCCs to decline to levels that would
allow unrestricted consumption of fish in the Great Lakes.  Due to the unique
characteristics of the Great Lakes, special limitations are necessary to
reduce loadings of BCCs to assure that similar problems do not occur in the
future for other BCCs.  For a more thorough discussion of ambient
concentrations of BCCs, see sections I and II.C.8 of this document.

      A number of commenters mentioned that there would be significant costs
associated with complying with the mixing zone ban for existing discharges and
that EPA should not mandate reductions that are technically and economically
infeasible.  Mixing zones allow facilities to exceed applicable water quality
criteria in a portion of the stream segment or lake close to the discharge
point.  EPA recognizes that, in certain limited situations, the elimination of
mixing zones for BCCs for existing discharges may be technically or
economically infeasible, and in limited circumstances, may not be a reasonable
approach  despite the ten-year phase-out period. Therefore, the final Guidance
provides a process whereby a State or Tribe may grant a mixing zone for
existing discharges of BCCs in limited circumstances.  EPA emphasizes that no
such exception to the mixing zone prohibition is authorized for new Great
Lakes dischargers of new or expanded discharges from an existing Great Lakes
discharger because EPA has determined that facilities contemplating such
discharges have more flexibility in designing and constructing their processes
and treatment technologies to meet applicable water quality criteria at the
point of discharge.  In addition, EPA notes that States and Tribes are not
required to grant mixing zones in any instance.

      The final Guidance authorizes the granting of a mixing zone for BCCs for
existing discharges, after the phase-out period, only upon finding that:   (1)
the facility is in compliance with and will continue to implement all
applicable treatment: and pretreatment requirements of Clean Water Act sections
301, 302, 304, 306, 307, 401, and 402, including existing NPDES water-quality
based effluent limitations; and  (2) the discharger has reduced its discharge
of the BCC for which a mixing zone is requested, and will continue to
implement controls to further reduce such discharge, to the maximum extent
possible.  Because of concerns about the impacts of BCCs to the Great Lakes
System and the significant public support for the elimination of mixing zones
for BCCs, EPA intends that this exception only be granted in limited
situations.

      In making a finding that a discharger has reduced the discharge of BCCs
for which the mixing zone is sought to the maximum extent possible, the State
or Tribe should consider the availability and feasibility of additional
controls for that discharger to reduce and ultimately eliminate BCCs,
including those controls and strategies used by similar dischargers.  For
purposes of this subparagraph, "similar dischargers" is to be interpreted
broadly to include, at a minimum, facilities with similar industrial or
treatment processes, similar pollutants, and similar products or similar by-
products .

      For purposes of determining whether to grant a mixing zone for an
existing discharges of BCCs after the phase-out period, the State or Tribe
should also consider whether the discharger, or affected community or
communities,  will suffer severe economic hardship if the mixing zone is
eliminated.  In evaluating economic impacts, State or Tribe should consider
costs of all pollution reduction options including available treatment
technologies and control strategies beyond those already being implemented.
Costs should reflect design and current operating flow.  States or Tribes

-------
270    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

should also evaluate the influent water quality, type of BCC, volume of
effluent and concentration of the BCCs for which the mixing zone is being
sought present in the effluent, and ambient receiving water quality.  Finally,
the State or Tribe should evaluate information on the facility's current
financial health including, where appropriate, existing municipal and
pretreatment user charges and existing profitability.  Where appropriate, the
State or Tribe may also want to consider information on the current
profitability and overall financial health of the facility1s'parent
corporation, where such information is available.  EPA expects that factors to
be considered in assessing economic impacts will vary on a facility-by-
facility basis.  (See Economic Guidance for Water Quality Standards -
Workbook, Draft, November 1993, available  in the docket for this rulemaking.)
The State or Tribe should also evaluate potential effects on employment rates,
tax revenues, and where appropriate, on user fees from increased costs
associated with meeting water quality criteria in the absence of a mixing
zone.

      Under the final guidance, a mixing zone for a BCC may be granted only if
the permitting authority determines, inter alia, that the discharger has
reduced its loadings of that BCC to maximum extent possible.  Therefore, an
exception to the mixing zone elimination provision may not be granted if
pollution prevention and/or control and treatment strategies exist that make
it technically possible for the discharger to achieve the applicable water
quality criteria at^the point of discharge, and if the discharger, or affected
community or communities, will not suffer severe economic hardship in
implementing such strategies.  For example, in assessing whether the
discharger has reduced its discharge of the BCC for which a mixing zone is
requested to the maximum extent possible, the State or Tribe should consider
the availability and feasibility of alternate treatment technologies and
control strategies including pollution prevention measures that reduce and
eliminate BCCs, and whether or not these technologies and strategies are
currently being implemented by the facility.  Relevant strategies include
those that would apply both to the facility and upstream sources  (e.g., a
municipalities's industrial users).  After evaluating alternate technologies
and strategies, the-permitting authority should consider the technical reasons
that implementation of some or all of them cannot reasonably be expected to
eliminate the discharger's need for a mixing zone.  EPA emphasizes that this
exception to the elimination of mixing zones for existing discharges of BCCs
is intended to be very limited and only granted in exceptional circumstances.
In addition, if a mixing zone for existing discharges of BCCs is proven
necessary, the State or Tribe should only grant the amount of mixing needed to
address the remaining technical and economic limitations.  In no circumstance
should the amount of mixing allowed exceed the maximum mixing zones specified
for non-BCCs in sections D  (deriving TMDLS for discharges to Lakes) and E
(deriving TMDLS for discharges to tributaries) in procedure 3 of appendix F.

      The State or Tribe should also consider whether or not the discharger
agrees to develop and implement an ambient monitoring plan.   Monitoring data
compiled by dischargers could be used to supplement State or Tribal monitoring
data and provide additional information on receiving water assimilative
capacity and on the extent of impacts, if any, associated with the mixing
zones.  Ambient monitoring data would be used, in attained waters, to ensure
compliance with water quality criteria at the edge of any mixing zone, and in
non-attained waters to ensure that the projected improvement in water quality
under the TMDL or comparable assessment and remediation plan is occurring.
Ambient monitoring data can also be used to provide the basis for future
decisions on the granting of mixing zones for BCCs.  The State or Tribe is
encouraged to seek additional information, as necessary, to determine whether
a mixing zone for BCCs is warranted for an existing discharge.

      The final Guidance incorporates a number of limitations on any mixing
zones for existing discharges of BCCs granted after March 23, 2007.
Specifically, under the final Guidance, no mixing zone for existing discharges
of BCCs shall result in any less stringent limitations than those existing

-------
                               Section vm.C: TMDLs                           271
prior to March 23, 1997.  Furthermore, the mixing zone shall be limited to one
permit term.  Mixing zones may not be granted thereafter unless the State or
Tribe makes the necessary findings discussed above for each successive permit
application in which a mixing zone for BCCs is sought.  The size of the mixing
zone shall also be evaluated and shall reflect all new information obtained by
the State or Tribe in considering mixing zones for BCCs after the phase-out.
In addition, any mixing zone for BCCs granted under this exception for
attained waters must protect all designated and existing uses of the receiving
water and must ensure the attainment of applicable aquatic, life, wildlife, and
human health criteria.  In non-attained waters any mixing zone granted for
BCCs under the exception must be consistent with the TMDL or comparable
assessment and remediation plan under procedure 3 .A of appendix F.

      EPA recognizes that pollution prevention approaches are an effective
means of reducing loadings to the environment and are usually less costly than
treatment.  Thus, the final Guidance provides that, in granting any exception
to the mixing zone elimination provision for existing discharges of BCCs, the
State or Tribes needs to ensure that the discharger has developed and
conducted a pollutant minimization program for that pollutant consistent with
procedure 8 of the Guidance, where applicable.  Procedure 8 of the final
Guidance provides that when a water quality-based effluent limitation for a
pollutant is determined to be less than the quantification level, the
permitting authority shall include a condition in the permit requiring the
permittee to develop and conduct a pollutant minimization program.  The goal
of the pollutant minimization program is to reduce all potential sources of
the pollutant and thus to maintain the effluent at or below the WQBEL.  Based
on current detection levels for the twenty-eight BCCs that are included in
Table 6 of the final Guidance as pollutants of Initial Focus in the Great
Lakes Water Quality Initiative, it is estimated that 22 of the BCCs will have
criteria established at levels below what the most sensitive analytical
techniques can currently quantify, and will also likely result in WQBELs less
than their quantification levels.  Therefore, EPA believes that in most
instances, a facility will already be required to develop pollutant
minimization programs for most BCCs.  It is possible that in some situations,
addition of a mixing zone may result in an increased limit that will then
cause the WQBEL to be greater than the quantification level; procedure 8 would
no longer apply and a pollutant minimization program would no longer be
required.  In those instances. States and Tribes should consider requiring the
permittee to develop and conduct a pollutant minimization program as a
condition of receiving the mixing zone for BCCs.

      Finally, the final Guidance provides that no mixing zone for a BCC shall
be granted unless alternative means for reducing BCCs elsewhere in the
watershed are evaluated.  This limitation reflects concerns raised by many
commenters that nonpoint source contributions of BCCs might be more
significant than point source contributions and therefore nonpoint sources
should be taken into account when determining the availability of mixing zones
for existing point source discharges of BCCs.  This evaluation can be
conducted either by the State or Tribe or by the discharger seeking the mixing
zone for BCCs.  EPA expects that this evaluation may identify opportunities to
reduce BCC loadings within the watershed from other sources and may facilitate
a more effective and less costly strategy for point sources to achieve overall
reductions in BCCs.  EPA expects controls necessary to obtain additional
reductions in BCCs will be implemented under existing State, Tribal, federal
and local authorities and believes that this provision will provide additional
incentives for dischargers to assist States and Tribes in identifying other
sources of BCCs.  As suggested by some commenters,  reductions of some of these
nonpoint source loadings may prove to be more cost-effective and may result in
greater environmental benefits than would be achieved by increasing controls
on point sources.

      The final Guidance provides that exceptions to the mixing zone
elimination provision will be granted solely at the discretion of the State or
Tribe on a case-by-case basis.  States or Tribes may also choose not to

-------
272    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

authorize such exceptions as part of their part 132 adoption, and thus could
simply require the elimination of mixing zones for existing discharges of BCCs
no later than March.23, 2007.

      Because of the importance of controlling BCCs in the Great Lakes System,
it is critical that the public have an opportunity to comment on permit-
specific exceptions to the general policy of eliminating mixing zones for
existing dischargers of BCCs.  The final Guidance provides that each draft
permit that includes a mixing zone for one or more BCCs after the phase-out
period must specify, either in the fact sheet or in the statement of basis for
the draft permit, the mixing provisions used in calculating the permit limits,
and must identify each BCC for which a mixing zone is proposed.  The draft
permit, including the fact sheet or statement of basis, is required to be
publicly noticed and made available for public comment under 40 CFR 124.6(e).
The final Guidance also specifies that any mixing zone for existing BCC
dischargers authorized under procedure 3.C.6 of appendix F must also be
consistent with procedure 3.D and 3.E of appendix F.

      Under the final Guidance, the elimination of mixing zones will continue
to be limited to BCCs.  BCCs are the pollutants of primary concern in the
Great Lakes System.  Documented widespread impacts warrant the special
emphasis on controlling BCCs (see section I of this document, and the preamble
to the proposal at 58 FR 20806).  In addition, States already have the
discretion under current EPA regulations to eliminate mixing zones for other
persistent chemicals such as lead and cadmium.

      The final Guidance retains the ten year phase-out period for existing
discharges but clarifies that this begins after States and Tribes adopt the
part 132 implementation procedures.  As authorized by section 132.5, States
may be granted up to two years in which to adopt and submit for EPA approval
criteria, methodologies and policies and procedures consistent with the final
Guidance.  The ten year time period corresponds to two five-year NPDES permit
terms.  EPA has determined that it represents a reasonable period for
implementing the mixing zone phase-out and that this period is consistent with
the Great Lakes Water Quality Agreement goal of virtual elimination of
persistent toxic substances.

      EPA has concluded that a shorter time period for existing Great Lakes
discharges, such as a phase-out within five years as suggested by some
commenters, may not afford facilities with existing discharges sufficient time
to retrofit existing treatment technologies or to adopt new pollution
prevention or alternative control strategies as necessary to achieve the
applicable water quality criteria at the point of discharge.  Therefore, EPA
is retaining the proposed ten year phase-out period.  EPA notes, however, that
States and Tribes may choose to establish a shorter phase-out time when they
adopt the final Guidance.

      The proposal also included a provision that WLAs be set at a more
stringent level than the most stringent water quality criteria or values if
necessary due to background concentrations to meet criteria and values at the
point of discharge.  This clause has been omitted from the final Guidance.
The final Guidance provides simply that the WLA for new and existing
discharges of BCCs shall be set equal to the most stringent applicable water
quality criteria or values for the BCC in question.  This would also be the
case for a BCC for which the water body is  in non-attainment.  See section
VIII.E.2.h of this document for a discussion of the rationale.  Section
301(b)(1)(C) and 402 of the Clean Water Act and implementing regulations
address discharges to non-attained waters and ensure that limitations more
stringent than criteria will be imposed where appropriate; thus EPA determined
that the omitted clause was unnecessary.

      EPA has made other modifications to the mixing zone section.  The order
of this section has been rearranged to correspond to the chronological
sequence of events.  Also, the final Guidance clarifies that specific

-------
                               Section VHI.C: TMDLs                           273
provisions in this section apply to WLAs calculated in the absence of TMDLs
and preliminary WLAs developed for purposes of determining reasonable
potential under procedure 5 of appendix F, as well as to the development of
TMDLs.  This change, reflects the modification to General Condition 1,
discussed above, which no longer specifies that TMDLs must be developed prior
to the issuance of a new or revised NPDES permit upon a finding of reasonable
potential.  WLAs and corresponding WQBELs may be calculated in the absence of
a TMDL.  The new reference in this section is intended to clarify that these
mixing zone provisions apply even in those situations when no TMDL has been
established.

      The final Guidance retains the exception to the mixing zone elimination
for BCCs for existing discharges from a facility implementing water
conservation measures.  EPA recognizes that, as a result of water conservation
measures, concentrations of a BCC in an effluent may increase, while the mass
of the BCC being discharged does not.  EPA concludes that because water
conservation is desirable, an exception may be appropriate in certain
circumstances.  The primary concern for BCCs is the mass of the pollutant
entering the Great Lakes System.  EPA agrees with commenter's concerns
regarding allowable increases above criteria and has retained the provision
that restricts mixing zones under the water conservation provision to those
allowed for non-BCCs (i.e., a 10:1 dilution ratio for lakes and 25 percent of
design flow for tributaries).

5.    TMDLs for Open Waters of the Great Lakes (OWGLs)

      Both options A and B described the process for developing TMDLs for open
waters of the Great Lakes  (OWGLs), inland lakes and other waters of the Great
Lakes System that exhibit lentic conditions {see proposed sections 3A.C  (58 FR
21036) and 3B.C (58 FR 21039)}.  Both options provided general guidance for
development of TMDLs on a lake-wide basis, including specifications for mixing
zones for non-BCCs, calculation of load allocations, protection from acute
effects, procedures when high background concentrations are present, and a
provision for a margin of safety for chronic and acute effects.

        In the final Guidance, language has been added to state explicitly
that TMDLs developed under this section must comply with General Conditions 1
through 11 and requirements of section 303(d)  of the CWA and 40 CFR 130.7.
(see citations under general condition 1 in procedure 3 of appendix F).  The
final Guidance also identifies the provisions of this section that apply for
purposes of calculating WLAs in the absence of TMDLs and preliminary WLAs for
purposes of determining reasonable potential under procedure 5 of appendix F.
Aspects of both procedures 3A and 3B have been retained in the final Guidance
and modifications to specific components of the proposal are described in more
detail in the following sections.  It should be noted that nothing in this
section should be construed as authorizing mixing zones for BCCs that are
prohibited under procedure 3.C of appendix F.   These procedures are to be
used, however, when establishing a mixing zone allowed under procedure 3.C of
appendix F.

a.    Mixing Zones for non-BCCs

      i.    Proposal:  Both options provided that, absent a mixing zone study,
individual wasteload allocations for point sources shall not be based on a
mixing zone larger than is provided by mixing one part effluent with ten parts
lake water, including background concentrations of pollutants.  Option A
described the 10:1 mixing zone in a narrative format, while Option B embodied
the concept in a formula.  Option B included language providing that in no
case shall the permitting authority grant a mixing zone that exceeds the area
where discharge-induced mixing, i.e., the area in which the momentum from the
discharge pipe ceases to have a major impact,  occurs.

      Under proposed Option B, for non-BCCs, when a facility believes the
actual area of discharge-induced mixing is greater than 10:1, a larger mixing

-------
274    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


zone could be allowed if a mixing zone demonstration is successfully completed
in accordance with proposed section 3B:'E.  Under Option A, the mixing zone
available is not necessarily constrained by the area of discharge-induced
mixing if a facility demonstrates that an alternative mixing zone is
appropriate for protection of designated and existing uses and implementation
of all criteria and. values.

      ii.   Comments:  Several commenters disagreed with the provision
limiting allowable mixing zones to the area of discharge-induced mixing.
Several commenters advocated that credit be given for the use of diffusers and
other forms of enhanced mixing to increase discharge-induced dilution.

      Several commenters suggested that there is not sufficient justification
for a maximum dilution factor and therefore disagreed with the 10:1 specified
in the proposal.  One commenter stated that the studies cited in the proposal
support setting the 10:1 factor as a default value but do not provide a
scientific basis to*establish the 10:1 as a maximum.  Several commenters
mentioned that the proposal is inconsistent with existing State mixing zone
policies and recommended that the final Guidance be modified to allow each
State to use its existing mixing zone provisions, which have already been
approved by EPA.

      One commenter advocated that mixing zones be prohibited for new source
discharges of non-BCCs to lakes unless a mixing zone demonstration was
conducted by a discharger.  One commenter suggested that, for new sources, a
dilution factor of up to 75% should be allowed without a mixing zone
demonstration.

      iii.  Final Guidance;  The final Guidance consolidates aspects of both
options A and B into one provision.  Like both options, the final Guidance
specifies that WLAs calculated in the absence of a TMDL and preliminary WLAs
for purpose of determining the need for WQBELs under procedure 5 of appendix F
shall assume no greater dilution rate than one part effluent to 10 parts
receiving water.  The final Guidance clarifies that this dilution factor
applies to both new and existing dischargers.  Language appearing in both
proposed options was modified to clarify that the provision applies to WLAs
developed both for numeric and narrative criteria.  The final Guidance retains
the provision in Option B that limits the area of the mixing zone to the area
of discharge-induced mixing.  Consistent with both proposed options, a larger
mixing zone is allowed if a facility successfully completes a mixing zone
demonstration pursuant to procedure 3.F of appendix F.  As discussed below,
the final Guidance adopts the mixing zone demonstration provisions proposed as
part of Option B.

      As described in the preamble to the proposal  (58 FR 20932), the 10:1
mixing factor was derived from mixing zone studies conducted for the Milwaukee
Metropolitan South Shore wastewater treatment plant and for the Green Bay
Metropolitan wastewater treatment plant.  For these cases, it was shown that
the 10:1 mixing factor represented an area of mixing where the velocity and
momentum associated with an effluent being discharged from the end of a pipe
was dissipated and any further dilution or mixing that then occurred was
associated only with the typically slower natural process of diffusion, wind,
temperature or current induced dispersion.  While recognizing that mixing zone
allocations are largely a policy decision, EPA believes that these studies
provide a scientific basis for default mixing zone assumptions for discharges
to open waters of the Great Lakes.  The final Guidance does allow for
recognition of site-specific conditions by allowing alternative mixing zones
subject to the mixing zone demonstration requirements set forth in procedure
3.F of appendix F.  -EPA recognizes that mixing zone demonstrations are subject
to resource and timing constraints.

      EPA acknowledges that different situations, such as the use of diffusers
and other technologies to enhance mixing, may increase the area of discharge-
induced mixing, thereby warranting a  larger  dilution factor; and the final

-------
                               Section Vm.C: TMDLs                           275
Guidance authorizes States and Tribes to afford dischargers the opportunity to
demonstrate that an alternative mixing''zone is appropriate.  However, in the
interest of ensuring consistency throughout the Great Lakes System, in the
absence of site-specific data from a mixing zone demonstration, EPA has
determined that a maximum default mixing factor of 10:1 will be retained in
the final Guidance.

b.    Calculating Load Allocations

      Under both proposed Options A and B, State law formed the basis for
determining appropriate dilution assumptions to be used on a case-by-case
basis when establishing load allocations for nonpoint sources for OWGLs,
inland lakes and other waters of the Great Lakes System with no appreciable
flow relative to their volumes. This is consistent with the general approach
in the Guidance which generally allows States and Tribes flexibility to use
their own procedures to address nonpoint source contributions to these water
bodies.

      EPA received general comments regarding the need to give States and
Tribes flexibility to consider site-specific factors in addressing point and
nonpoint source pollutants in developing TMDLs.  The final Guidance retains
the proposed language allowing States and Tribes the flexibility to consider
appropriate mixing zone assumptions for nonpoint sources, consistent with
applicable State and Tribal requirements.

c.    Protection from Acute Effects

      i.    Proposal:  Both options included provisions to assure attainment
of acute criteria and values within the allowable acute mixing zones for
discharges to the OWGLs and other waters described in paragraph B.  Option A
did not include a specific cap, but instead relied on site-specific analyses
of limits necessary to assure attainment of acute criteria and values within
the applicable acute mixing zone.  Option B specified that effluent
limitations for point sources may not exceed a final acute value (FAV).  The
preamble to the proposal noted that, in some circumstances, however, an
effluent limit based on an acute mixing zone may need to be more stringent
than the FAV to protect against acute effects within the mixing zone.  The FAV
is defined as twice the Criterion Maximum Concentration  (CMC)  (see section
132.2) of this final Guidance. Therefore, if the effluent is at twice the
maximum concentration for protection against acute effects, acute toxicity may
occur near the point of discharge depending on site-specific conditions.

      ii.   Comments:  Most commenters opposed the use of acute mixing zones.
Several advocated eliminating mixing factors altogether, at least in sensitive
and/or impaired areas.  Several commenters suggested that acute mixing zones
for non-BCCs be developed on a case-by-case basis without an automatic FAV
limit (Option A).  Other commenters recommended the use of best professional
judgment instead of a specified cap.

      A number of commenters preferred the Option A acute mixing zone
provisions to Option B because they suggested that the mixing zone limit in
Option B is inconsistent with existing State policies.  Other commenters
argued that Option B sets an arbitrary constraint on mixing zones.

      Several commenters preferred Option B because it is numeric and thus
provides a well-defined benchmark for more consistent application in the Great
Lakes System.  Commenters argued that Option B should be mandatory, not
discretionary.  Several commenters were concerned that mixing zones under
Option A could be substantially larger than under Option B and would not
promote consistency in permit limits among States and Tribes.  Many commenters
were concerned that'Option A provides too much discretion for establishing
mixing zones and dilution flows, and that Option B, which delineates a
calculation method, is needed to promote uniformity across the Great Lakes
System.

-------
276    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Several commenters expressed confusion because the proposed Guidance
specifically listed Criterion Maximum Concentration  (CMC) values, thereby
implying that such values should be used in establishing permit limits, while
the TMDL implementation procedure allowed permit limits up to the FAV  (twice
the CMC value).

      iii.  Final Guidance;  The final Guidance provides that WLAs based on
acute aquatic life criteria or values for discharges to the OWGLs and other
waters described in'paragraph B must not exceed the Final Acute Value  (FAV).
As proposed in Option A, the final Guidance also requires that a WLA based on
such criteria and values be reviewed to assure that it prevents acute effects
at the boundary of any acute mixing zone allowed under State law.

      In the final Guidance, EPA combines the two proposed approaches into a
single provision.  EPA acknowledges the concerns raised by commenters
regarding acute mixing zones and has retained language from Option B
specifying a cap based on the FAV for acute mixing zones in order to promote
consistency in developing permit limits within the Great Lakes System, while
also minimizing areas of acute toxicity.  EPA agrees with commenters that a
numeric benchmark should ensure consistency better than narrative
considerations.  In response to comments, the final Guidance also provides
that if mixing zones from two or more proximate sources interact or overlap,
the combined effect must be evaluated to assure that criteria and values will
be met in the area where any applicable acute mixing zones overlap.  In
addition, EPA agrees with commenters that site-specific considerations might
authorize a larger mixing zone than otherwise authorized by the FAV cap.
Accordingly, the final Guidance allows the use of a mixing zone demonstration
to exceed the FAV if the demonstration is conducted and approved pursuant to
procedure 3.F of appendix F.

      EPA recognizes that some commenters, including some States, support
eliminating acute mixing zones but notes that States and Tribes retain the
authority to adopt provisions more stringent than those in the final Guidance
consistent with CWA section 510.  Accordingly, States and Tribes may eliminate
mixing zones altogether or in selected locations such as sensitive and/or
impaired areas.  EPA is retaining the FAV cap for acute effects because it
more accurately reflects discharge specific scenarios such as cases where
there is rapid mixing  (e.g., where high rate diffusers are used).

6.    TMDLs for Discharges to Tributaries

      The principal differences between options A and B in the proposal
related to TMDL development for tributaries.  The initial focus of Option A
was on attainment of water quality standards throughout a basin, followed up
with site-specific cross checks at discharge points  throughout the basin.   The
site-specific cross checks would  assure that standards are being attained
around individual discharge points.  Option A did not specify the size of
mixing zones.  Rather,  it  left such considerations to existing State
requirements.  Option B focused initially on evaluating limits needed  for
individual point sources, with supplemental  emphasis on basin-wide
considerations as necessary.  Option B also  included more detailed procedures
including specific mixing  zone requirements.

       As discussed earlier  in this document, EPA has decided that one
procedure will apply for development of TMDLs for tributaries to the Great
Lakes in order to ensure that some level of  consistency  applies  throughout  the
Great Lakes  System.  The procedure specified in  the  final Guidance includes
elements of  both proposed  Options A and B but has eliminated some of  the more
burdensome  and confusing aspects  of the proposed Guidance.   The  final  Guidance
provides a  greater  degree  of flexibility than afforded by either proposed
procedure,  by allowing  States and Tribes to adopt different  implementation
approaches  while at'the same time ensuring  consistency by requiring  States  and
Tribes to implement specific components of  the procedure.  Nothing  in this
section  should be construed as authorizing  mixing zones  for  BCCs that  are

-------
                               Section Vm.C: TMDLs                           277
prohibited under procedure 3.C of appendix F.  These procedures are to be
used, however, to determine the magnitude of any mixing zone allowed under
procedure 3.C of appendix F.

      Specific components of the proposal, comments on those specific
components, and modifications in the final Guidance are discussed below.

a.    Steady State vs. Dynamic Modeling

      i.    Proposal:  In the proposal, both options envisioned that, in most
instances, a simple, steady-state mass balance approach would be used to
develop TMDLs, WLAs in the absence of a TMDL or preliminary wasteload
allocation for the purpose of determining the need for WQBELs reasonable
potential under procedure 5 of appendix F.  A mass balance approach is a
method used to approximate the mass of pollutants within a water body.  It is
based on the physical law of conservation of mass which dictates that mass
cannot be created or destroyed but only transformed.  This approach assumes
that the input of mass into a system  (e.g., through point and nonpoint source
loadings, atmospheric deposition, groundwater seepage) equals the loss of mass
from a system plus any losses due to transformation of mass within the system.

      The proposal provided that the results of dynamic modeling be used only
where the results can be shown to be more restrictive than the results due to
the steady-state assumptions of both options A and B.  EPA requested comments
on whether the States should be allowed to use dynamic modeling regardless of
whether the results are more or less stringent than results from using a
steady-state approach.

      ii.   Comments:  In general, commenters supported the use of dynamic
modeling without the limitation that the results must be more restrictive than
the results using steady-state assumptions recommended in both options A and
B.  Commenters pointed out that existing EPA guidance promotes the use of
dynamic modeling and that the final Guidance should not contradict existing
guidance by imposing new restrictions on the use of dynamic modeling.

      iii.  Final Guidance:  EPA agrees with commenters and the final Guidance
allows the use of both steady-state and dynamic models to support
establishment of TMDLs. The final Guidance therefore retains provisions for
using a steady-state, mass balance approach, but also allows the use of
dynamic modeling regardless of whether the results are more or less
restrictive than would be generated under steady-state assumptions.   For an
in-depth discussion of available models, see EPA's Technical Support Document
for Water Quality-based Toxics Control  (TSD), EPA/505/2-90-001, 1991,
available in the docket.  EPA recommends that a model be selected based on its
adequacy for the particular application.  For example, adequacy of a model may
depend on the type of pollutant  (e.g., BOD/DO, toxics, etc.) or the type of
waterbody (e.g., river or lake).  Steady-state models compute average spatial
profiles of constituents within a waterbody assuming that loadings, upstream
water quality, stream flow rates, and meteorological conditions remain
constant over time.  Dynamic models predict both temporal and spatial
variations in water'quality due to varied loadings, flow conditions and
meteorological conditions.  Dynamic models are thus particularly useful for
analyzing impacts that vary over time, such as loadings resulting from storm
events and long term seasonal cycles. In determining whether to use a steady
state or dynamic model, the cost of application, data requirements, the
availability of historical data, and the availability of the particular model
and model support need to be considered.

b.    Stream Design Flows

      i.    Proposal:  In the proposal, both options A and B specified the
stream design flow under which criteria and1 values are to be implemented.
Although most point sources discharge to continuously flowing streams, the
amount of water available to dilute the discharge typically varies with the

-------
278    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

season and with periodic storms and drought conditions.  Thus, in deriving
TMDLs, wasteload allocations in the absence of TMDLs and wasteload allocations
for the purpose of determining the need for WQBELs,  it is necessary to
establish the stream conditions under which applicable criteria  and values'
will be implemented.  The volume of water flowing through the tributary in a
given time period at the design flow condition is the volume available  to
dilute .all pollutants present or introduced into the water body  and thus is a
necessary factor in developing a TMDL,  wasteload allocation in the absence of
a TMDL, and a preliminary wasteload allocation for the purpose of determining
the need for WQBELs using a steady-state model.  The proposed Guidance
specified different design flows for chronic aquatic life, acute aquatic life,
wildlife, and human'health criteria because of differences in how the criteria
were developed.  A detailed discussion of these flows and the basis for
choosing these flows can be found in the preamble to the proposed guidance  (58
FR 20933).

      ii.   Comments:  Several commenters suggested that the restriction on
stream low flow quantity for dischargers of non-BCCs is not scientifically
defensible and recommended that EPA not specify design flows.  Another
commenter suggested that specifying design flows simply adds a further  level
of conservatism in TMDL development.  They believe that this conservatism,
coupled with the margin of safety (MOS) may result in overly stringent  WLAs
and LAs.

      iii.  Final Guidance:  The final Guidance provides that the specified
stream design flows be used as a default assumption in developing TMDLs,
wasteload allocation in the absence of a TMDL and preliminary wasteload
allocations for the purpose of determining reasonable potential, but allows
the use of alternative stream design flow under certain conditions discussed
below.  The final Guidance adds new language clarifying that stream design
flows are appropriate for TMDLs, wasteload allocations in the absence of a
TMDL and wasteload allocations for the purposes of determining the need for
WQBELs established using steady-state models but are not likely  to be
applicable for those calculated using dynamic modeling.

      EPA retains language from Option A that the loading capacity is
initially calculated at the furthest downstream location in the  watershed
drainage basin.  The maximum allowable loading consistent with the attainment
of the appropriate criteria or value is determined by multiplying the
criterion or value by the flow at the farthest downstream location in the
tributary basin at the appropriate design flow condition.  States could
calculate the loading capacity at interim points in the basin.   However,
States and Tribes must include the total load capacity for the entire basin
when establishing the TMDL.  Even though the flow at the farthest downstream
point on an effluent-dominated stream may be largely effluent, the loading
capacity for the water in the stream is still the product of the criterion and
the total flow in the stream.

      The final Guidance specifies the 7-day, 10-year low flow  (7Q10) or the
4-day, 3-year biologically-based design flow  (4B3) for chronic aquatic  life
criteria or values; the 1-day, 10-year low flow  (1Q10) for acute aquatic life
criteria or values; the 90-day, 10-year low flow  (90Q10) for wildlife criteria
or values; and the harmonic mean flow for human health criteria  or values.
The final Guidance also stipulates that the lowest load is then  selected as
the loading capacity.

      Although EPA received numerous comments suggesting that flows other  than
those specified in the proposal be adopted, none of the commenters supplied
any scientific data supporting their proposed alternative flows.  Many
commenters supported the proposed flows.  In the interest of promoting  greater
consistency among States and Tribes in the Great Lakes System, EPA is
retaining, with the exception of  the design flow specified for wildlife (see
discussion below), the proposed design flows in the final Guidance.  These
design flows are default values that must be used  in developing  TMDLs,  WLAs

-------
                               Section Vm.C: TMDLs                           279
calculated in the absence of TMDLs, and preliminary WLAs for purposes of
determining reasonable potential under procedure 5 of appendix F.  EPA
recognizes that in some instances, these flows may be overly conservative, or,
in other situations, may not be protective enough. Thus, the final Guidance
allows States and Tribes to use alternative stream design flows when data
exist to demonstrate that such an alternative is appropriate for stream-
specific and pollutant-specific conditions, such as using seasonal flows to
obtain seasonal WLAs.  Allowing alternative stream design flows is especially
necessary when a dynamic model is used to calculate the TMDL.  Dynamic models
use the entire flow record, not just one design flow, in making TMDL
calculations.  States and Tribes may also adopt more stringent design flows
than those specified here in accordance with section 510 of the CWA.

      The criteria and values derived pursuant to the final Guidance are not
designed to be never-exceeded values.  Rather, EPA has determined based on
scientific analyses that they may be exceeded at varying frequencies and
durations without irreparable injury to human health, wildlife, or aquatic
life.  Current EPA guidance recommends stream design flows for chronic and
acute aquatic life and human health  (see p. 79 of the 1991 TSD).  Until today,
EPA has not implemented wildlife criteria, nor has it recommended a design
flow for wildlife criteria.

iv.   Wildlife

       (A)   Proposal:  For TMDLs, WLAs calculated in the absence of TMDLs, and
preliminary WLAs for purposes of determining reasonable potential under
procedure 5 of appendix, based upon wildlife criteria or values, the
hydrological-based, 30-day, 5-year low flow (30Q5) flow was specified in the
proposed guidance.  EPA also specifically asked for comments on using the 90-
day, 10-year (90Q10) low flow, and the harmonic mean flow for wildlife
criteria or values in the preamble to the proposal.

      Both the 30Q5 low flow and the 90Q10 low flow include a factor
representing the rate-limiting step between the exposure to the pollutant and
the effect on the organism  (30 days and 90 days, respectively).  For wildlife,
the rate-limiting step is chemical bioaccumulation.  The 30-day and 90-day
period were proposed as representing reasonable time periods for chemical
bioaccumulation.  The 30Q5 low flow and the 90Q10 low flow also include a
value representing the rate at which the affected organisms recover  (a 5 year
and 10 year return frequency, respectively).

       (B)   Comments:  Several commenters claimed that the proposed 30Q5 low
flow is not scientifically defensible for wildlife criteria and asserted that
the low flow should be the harmonic mean flow.  Commenters suggested that it
was inappropriate to use a short term low flow such as the 30Q5 and that the
harmonic mean stream flow is more consistent with the long-term nature of
bioaccumulation processes.  Another commenter recommended the use of the 7Q10
low flow for implementing wildlife criteria.

      One commenter pointed out that both the 30Q5 low flow and 90Q10 low flow
are consistent with life cycles of small water mammals  (otter and mink).
Several commenters support the use of a 90Q10 low flow for the implementation
of the wildlife criteria because it allows a reasonable time period for
chemical bioaccumulation (90 days) with a reasonable return frequency  (10
years).

       (C)   Final Guidance;  The final Guidance only establishes wildlife
criteria for BCCs (see section VI of this document).  Therefore, the stream
design flow specified in the final Guidance for wildlife criteria would apply
when a mixing zone for a BCC is authorized under procedure 3.C.6 of appendix
F.

      The final Guidance specifies that a 90-day, 10-year low flow be used for
the implementation of wildlife criteria in tributaries.  This is the lowest

-------
280    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

90-day average flow that would occur, on average, one year in every ten years
based on a statistical review of historic flow data.  EPA recognizes, as some
commenters suggested, that a 30-day averaging period may be viewed as
conservative for some pollutants, given the long time it may take for
bioaccumulation.  EPA agrees with commenters that a 30-day period is too short
to represent bioaccumulation and is instead specifying the use of a 90-day
averaging period when no data exist to suggest an alternative.

      EPA disagrees with the commenter's recommendation to, use the 7Q10 low
flow for wildlife.  As discussed in the preamble to the proposal, for
wildlife, unlike for aquatic life, the impacts of chemicals with a high
propensity to bioaccumulate in aquatic organisms are of greatest concern
because aquatic organisms comprise a major portion of the diet of many
wildlife species.  Because of the relatively slow rate of uptake by aquatic
organisms of bioaccumulative chemicals, residues in the food chain have a
delayed response to increases in ambient concentrations of chemicals during
short-term periods, such as during low flow events.  The Steering Committee
thus judged a longer term averaging period to be more appropriate for wildlife
than the 7-day averaging period used for aquatic life.

      EPA recommends the 90-day averaging period for implementing wildlife
criteria as a reasonable estimate that can be used to establish limits that
are protective of wildlife.  EPA suggests that the 90 day period is
appropriate because concentrations of BCCs in the water column are not
expected to fluctuate excessively; BCCs all have very high bioaccumulation
factors  (BAFs), and the toxicological data used to establish wildlife criteria
are not based on acute effects.  A 90-day averaging period also coincides with
the length of seasons.  Some studies have documented seasonal variability in
fish tissue concentrations.

      EPA agrees with the commenter's statement that the 10-year period
represents a reasonable return frequency-  EPA also agrees with commenters'
suggestions that a five year return frequency is too short.  EPA disagrees
with commenters who recommended the harmonic mean be used.  EPA believes that
the harmonic mean is too long and may not be protective of shorter lived
wildlife species. The harmonic mean may not be an appropriate proxy for
wildlife because the lifespan of wildlife is highly variable and may be very
short. The harmonic mean is used for the protection of humans with an average
exposure of 70 years  (e.g., an average lifespan), substantially longer than
any of the wildlife species.  In addition, wildlife criteria focus on
reproductive endpoints, a subset of toxicological endpoints, to protect
against population effects, while human health criteria cover a broad range of
effects on individuals.

      EPA believes specifying the 90Q10 low flow as a default and allowing the
use of site-specific data balances the need for consistency while allowing the
best scientific approach to be used.  In response to comments that food chain
effects attenuate the effects of fluctuations in ambient concentrations, the
final Guidance will allow the use of an alternative stream design flow where
data exist to demonstrate that such an alternative flow is appropriate for
stream-specific and pollutant-specific conditions to be protective of
wildlife.  EPA recognizes that in some situations in the Great Lakes System,
internal loadings of BCCs may dominate over external, or point source,
loadings.  These types of internal loadings  (e.g., sediment resuspension) tend
to be constant over long periods of time, and depending on local mass ratios,
may buffer the fluctuations from point source loadings.  However, because the
design flow is an important parameter in establishing TMDLs, wasteload
allocations in the absence of TMDLs and preliminary wasteload allocations for
the purposes of reasonable potential, it is important to specify a default
value that is protective of wildlife  in the absence of site-specific data.
EPA recognizes the 90Q10 low flow may be conservative for certain pollutants
for certain streams, and encourages dischargers to work with States and Tribes
in generating site-specific data.

-------
                               Section Vm.C: TMDLs                           281
v.    Chronic Aquatic Life

       (A)   Proposal;  The proposal specified the 7-day, 10-year low flow
 (7Q10) or the 4-day, 3-year biologically-based design flow  (4B3) for chronic
aquatic life criteria or values.

       CB)   Comments:  Several commenters supported the design flow for
aquatic life criteria and noted that these stream design flows are consistent
with EPA's 1991 TSD.  One commenter agreed that for protection from chronic
effects, the 7Q10 low flow is appropriate.  Several commenters recommended
that 30Q10 be used as an alternative.  One commenter asserted that the 7-day,
10 year flow is overly conservative because the chronic water quality
standards are based on toxicity tests of at least 24 days, and stated that the
30-day, 10- year low flow would be more appropriate.

       (C)   Final Guidance;  Like the proposal, the final Guidance authorizes
the use of either the 4B3 biologically-based design flow or the 7Q10
hydrologically-based design flow as the stream design flow for chronic aquatic
life criteria.  Unlike in the proposed guidance, however, the final Guidance
also provides additional flexibility by allowing the use of an alternative
stream design flow where data exist to demonstrate that the alternative is
appropriate for stream-specific and pollutant-specific conditions.  In the
absence of such data, EPA continues to specify the 4B3 or the 7Q10 stream
design flow to ensure protection of aquatic life from chronic effects.

      The 4B3 is that flow, determined on a case-by-case basis, that would
provide for an excursion of chronic aquatic life criteria, over a 4-day
averaging period, only once every three years, on the average.  This flow is
selected because EPA has determined that criteria developed on that basis may
be exceeded over a 4-day averaging period once every three years without
injury to the aquatic ecosystem.  (See 1991 TSD).  A 4B3 flow can be
calculated using the computer program DFLOW supported on EPA's computers at
the National Computer Center in Research Triangle Park, NC.  Further
information may be obtained from Assessment and Watershed Protection Division,
U.S. Environmental Protection Agency, 401 M St, S.W., Washington, D.C. 20460.

      EPA also allows, as an alternative, the hydrological-based 7Q10 low
flow.  The 7Q10 is the lowest 7-day average flow expected to occur on the
average one year in every ten, based on the period of record.  Empirical data
from approximately 60 streams show that the 7Q10 low flow provides a degree of
protection approximately equivalent to the 4B3 flow. The U.S. Geological
Survey routinely publishes statistics that commonly include estimates of the C
for most riverain systems.

vi.   Acute Aquatic Life

       (A)   Proposal:  In the preamble to the proposal, EPA solicited comments
on whether the final rule should specify a design flow for the purposes of
implementing acute aquatic life criteria.  The preamble discussed the
recommended use of the 1Q10 low flow for acute aquatic life in existing EPA
guidance (See the 1991 TSD, available in the docket).

       (B)   Comments:  One commenter suggested that use of the 1Q10 low flow
for acute aquatic life criteria is too conservative and that the final rule
should specify use of the 7Q10 low flow.

       (C)   Final Guidance:  In the final Guidance, EPA specifies the 1Q10 low
flow for purposes of implementing acute aquatic life criteria.  This design
flow would be used in determining whether the FAV cap is sufficient to protect
against acute aquatic life effects.   The 1Q10 low flow is consistent with the
recommended design flow specified in existing EPA guidance  (e.g., TSD).  EPA
agrees that this design flow may be overly conservative in some instances but
this flow should be used unless data exist to demonstrate that an alternative
stream design flow is appropriate for stream-specific and pollutant-specific

-------
282    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

conditions.  This is also consistent with the TSD, which recommends allowing
for si te~- specif ic or chemical-specific 'conditions.  States and Tribes may want
to use the biologically based 1B3 as an alternative flow for acute aquatic
life.  The 1B3 is also discussed briefly in EPA's TSD.  In addition,
alternative averaging periods can be developed from data on the  time course of
mortality in acute toxicity tests.

vii.  Human Health

      (A)   Proposal:  In the proposal, EPA specified the use of the long-term
harmonic mean flow to implement human health criteria.

      (B)   Comments:  Several commenters supported the design flow for human
health criteria and pointed out that it is consistent with existing EPA
guidance.  However,.several commenters suggested that there was no scientific
justification beyond the limited references in existing EPA guidance for using
the harmonic mean instead of the arithmetic mean for human health criteria.
One commenter suggested that the cost of statistically generating the harmonic
mean statistic for the numerous surface water discharges in the basin could be
prohibitive.  The commenter also suggested that the harmonic mean flow
estimate may be more error-prone than other flow estimates because statistics
such as the harmonic mean flow are only useful where stream flow is highly
variable.  One commenter recommended the use of a mean annual flow as an
alternative.  Another commenter suggested that the 7Q10 low flow or 30Q10 low
flow should be required rather the harmonic mean flow.

      (C)   Final Guidance:  The final Guidance retains the use of the  long-
term harmonic mean flow to implement human health criteria as supported by
current EPA guidance.  EPA has determined that such a level will ensure that
criteria will not be exceeded under stream conditions that represent long-term
average conditions.  The harmonic mean flow is the sum of the reciprocals of
individual flow measurements divided into the total number of individual flow
measurements.

      The harmonic mean was chosen as a design flow for human health criteria
because human health criteria are designed to protect an individual over a
lifetime of exposure.  Human health criteria based on cancer potencies  and
risk levels are based on models which extrapolate animal data to a human
lifetime.  Similarly, a human non-cancer criterion is based on an Rf>  (or ADE,
as it is referred to in the final Guidance which is an acceptable daily
exposure over a lifetime.  Therefore, EPA has attempted to match the longest
stream flow averaging period (using harmonic mean) with the criterion which is
protective over a human lifetime.  EPA disagrees with the suggestion that an
arithmetic mean rather than a geometric mean be used.  For carcinogens,  it is
appropriate to determine the long-term mean exposure concentration.  Because
flow is not normally distributed, using the arithmetic mean flow for design
purposes will underestimate the mean concentration.  Using the downstream
harmonic mean flow will more closely estimate the mean concentration.

      In rare instances where a human health criterion or value  is based on a
short term toxicological effect  (i.e., the critical effect upon  which the
criterion/value is based is significantly less than lifetime and may be an
acute effect), the design flow should be adjusted accordingly.   This does not
pertain to ADEs  (RfDs)  in which a short term study has  been  used as  the  ADE
basis and an uncertainty factor has been used to account for less than
lifetime study results.  This pertains only to those situations  where the
critical effect is fhe  short term effect and no additional uncertainty  factor
has been used to account for less than lifetime exposure.  A good example of
this is EPA's R£> for nitrate.  The  critical effect,  upon which the  RfD  is
based, is toxicity to children after a short term exposure.  In  this case, a
harmonic mean would be  an  inappropriate design flow for  such a  short term
effect.  In this case',  a 7Q10 or  a 4Q3 design flow may be more  appropriate.

-------
                               Section Vm.C: TMDLs                           283
      EPA is setting the default design flow for human health as the harmonic
mean.  The Jiarmonic mean can be calculated using the computer program DFLOW
supported on EPA's computers at the National Computer Center in Research
Triangle Park, NC.  Further information may be obtained from Assessment and
Watershed Protection Division, U.S. Environmental Protection Agency, 401 M St,
S.W., Washington, B.C. 20460.

      Because EPA recognizes that there may be situations, like those
discussed above, where a different design flow is more appropriate, the final
Guidance allows the use of an alternative design flow for human health
criteria where data exists to demonstrate than an alternative stream design
flow is appropriate for stream-specific and pollutant-specific conditions.

c.    Mixing Zones for Non-BCCs

      i.    Proposal:  In the proposed guidance, Option A did not provide
specific requirements for mixing zones for either chronic or acute criteria.
Rather, under Option A, site-specific cross-checks would be conducted at each
source location to ensure that water quality standards including acute and
chronic aquatic life, wildlife, and human health, are attained at the edges of
applicable mixing zones, or if mixing zones are not allowed under State law,
throughout the basin.  Option A did not specify the size of mixing zones but
suggested that mixing zone requirements, if any, adopted by the various States
will be used for the cross-checks.

      Option B specified for both new and existing sources that WLAs based on
acute aquatic life criteria shall not exceed the Final Acute Value (FAV) in
order to ensure protection of aquatic life from acute effects.  The provision
is identical to the'provision for Open Waters of the Great Lakes System.  For
WLAs based on chronic aquatic life, wildlife and human health criteria, Option
B specified different requirements for new and existing sources.  For existing
sources, Option B provided a formula to derive the dilution fraction based on
the relationship of the effluent flow of the point source to the flow of the
receiving waters and an assumption regarding how rapidly mixing occurs.  The
dilution fraction is the fraction of the 7Q10 that is available for dilution
in the WLA calculation.  Under the formula proposed in Option B, the dilution
fraction varied from 10 to 25 percent.  The proposed guidance allowed an
opportunity to demonstrate that a larger mixing zone is acceptable subject to
a mixing zone demonstration conducted in accordance with section E of proposed
procedure 3B.  This provision in the proposal specified that in no case could
the dilution fraction exceed 75 percent.  For new sources, option B specified
that WLAs based upon chronic aquatic life, wildlife and human health criteria
or values shall equal the criteria or values unless a mixing zone
demonstration is provided, approved and implemented in accordance with
proposed procedure 3B.E.  The proposal also specified that in no case should
the demonstration result in a mixing zone greater than the dilution fraction
established for existing sources.

      ii.   Comments;  Several commenters suggested that a dilution fraction
of 25 percent is ovferly conservative based on the type and level of wildlife
and human health exposure which are likely to occur and suggested the use of a
larger fraction of the design flow for dilution.

      Several commenters suggested that option A, by not establishing a
dilution fraction and, in effect, allowing 100% of the design flow for
dilution, does not provide sufficient margin of safety and is inconsistent
with the Steering Committee's recommendation that only 10-25 percent of the
design flow be allowed for dilution.

      Several commenters suggested that Option B is inconsistent with the
Steering Committee proposal, insofar as that proposal did not provide the
increased mixing zone option to existing discharges of BCCs to tributaries.
Only the default dilution was allowed (10-25 percent of design flow).

-------
284    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      A number of commenters disagreed with the provision requiring
differential treatment for new and existing dischargers of non-BCCs.
Commenters suggested that new dischargers, like existing dischargers, should
be able to adjust the mixing zone based on a mixing zone demonstration  to a
dilution fraction higher than the 10-25% default specified for existing
dischargers in the proposal.

      iii.  Final Guidance:  The final Guidance adopts the Option B provision
that TMDLs, wasteload allocations in the absence of TMDLs and preliminary
wasteload allocations for purposes of reasonable potential shall not exceed
the PAV, unless a mixing zone demonstration is conducted and approved pursuant
to procedure 3.F of appendix F.  This is intended to ensure protection  of
aquatic life from acute effects.  The rationale described in the discussion of
Acute Mixing Zones for OWGLS applies here.

      In the final Guidance,  for TMDLs, wasteload allocations in the absence
of TMDLs and preliminary wasteload allocations for purposes of determining the
need for WQBELs based on chronic criteria to ensure protection of aquatic
life, wildlife, and human health from chronic effects, the dilution fraction
should be set at no greater than 25 percent of the appropriate stream design
flow (e.g., for aquatic life, human health or wildlife criteria). Unlike the
proposal, the dilution fraction is established at 25 percent of the stream
design flow rather than calculated using a formula.  The final Guidance does
retain the proposed provision allowing the opportunity to demonstrate that a
larger mixing zone is acceptable subject to a mixing zone demonstration
conducted in accordance with procedure 3.F of appendix F.  Unlike the
proposal, the final guidance allows the dilution fraction to go up to 100% if
a mixing zone demonstration is completed and approved pursuant to procedure
3.F in appendix F.  Procedure 3.F of appendix F requires a site-specific
analysis of local conditions around the vicinity of the discharge to ensure
that unacceptable impacts do not occur.  If the information and analysis
justifies a dilution fraction greater than 75%, as a general rule it should
not be prohibited.

      EPA is retaining 25 percent as the maximum dilution fraction unless a
mixing zone demonstration suggests that an alternative dilution fraction is
appropriate  (i.e., in the absence of site-specific data).  The 25 percent
dilution fraction is consistent with existing EPA guidance. As described in
the preamble to the-proposal, the concept of the fraction of the stream design
flow is based upon recommendations found in the Water Quality Criteria  -
Report of the National Technical Advisory Committee to the Secretary of the
Interior, April 1968 (Green Book) and upon guidance from EPA's 1983 Water
Quality Standards Handbook, both of which are available in the docket.  The
Green Book recommended that in order to prevent the initial mixing of point
source wastewater from erecting a barrier to aquatic organisms, only 25
percent of the cross-sectional area of the river should be used for mixing.
The Standards Handbook suggests that the value of 25 percent of total river
flow is a rational estimate of the amount of river flow in 25 percent of the
cross-sectional area.

      This proposal is consistent with several States' current mixing zone
policies.  For example, Michigan uses a straight 25 percent of the stream
design flow for all categories of criteria or values with an opportunity
demonstrate for a larger percentage.  Ohio uses a graduated scale for the
dilution  fraction that ranges between 10 percent and 100 percent of stream
design flow.  The use of a constant dilution factor as a default should
support a more consistent permitting approach throughout the Great Lakes
System.  Flexibility is retained, however, by allowing an alternative mixing
zone to be used when site-specific information and analysis support it  (i.e.,
through a mixing zone demonstration).

      EPA agrees with•commenters and has removed the distinction between new
and existing discharges for purposes of calculating TMDLs, wasteload
allocations in the absence of TMDLs and preliminary wasteload allocations  for

-------
                               Section Vffl.C:  TMDLs                           285
purposes of determining the need for WQBELs using chronic aquatic  life,
wildlife, and human health criteria and values.  Under the final Guidance,  for
protection of aquatic life, wildlife and human health from chronic effects,
TMDLs, WLA calculated in the absence of TMDLs, and preliminary WLAs for
purposes of determining the need for WQBELs under procedure 5 of appendix F,
shall be calculated using a dilution fraction no greater than 25 percent of
the  stream design flow unless a mixing zone demonstration is conducted.  -In no
case shall a State or Tribe grant a mixing zone which exceeds the  area of
discharge-induced mixing.  This provision applies to both .new and  existing
discharges of non-BCCs to tributaries.  EPA suggests that while differential
treatment for new and existing discharges is warranted for BCCs, because a
primary goal of this initiative is to reduce loadings of BCCs to the Great
Lakes, for non-BCCs, treatment of new and existing discharges will be the
same.

7.    Procedures for High Background Concentrations

      a.    Proposal:  Under both Options, the proposal specified  that when
ambient water quality concentrations exceed narrative or numeric criteria or
Tier II values, any discharge that has a reasonable potential to cause or
contribute to an excursion above such a criterion or value should  either be
prohibited, i.e., WLAs set equal to zero, or a multiple source TMDL should  be
established that ensures the attainment of that criterion or value.  Under
both options, the procedures used in developing multiple source TMDLs for
discharges were to be developed on a case-by-case basis, consistent with
applicable State or Tribal regulatory requirements.

      b.    Comments;  A number of commenters disagreed with the proposed
approach to set WLAs equal to zero when background exceeds criteria because it
would, in effect, force all point sources to achieve zero discharge.
Commenters suggested that in addition to the use of multiple source TMDLs,  EPA
should make more use of readily available water quality variances, site-
specific criteria, and intake credits in development of WLAs when  background
concentrations exceed criteria.  Commenters suggested that the administrative
burden of these existing mechanisms is a significant deterrent to  using them.
Commenters advocated a range of alternatives, from setting the WLA equal to
the  most stringent' criterion up to setting the WLA equal to the background
concentration of the receiving stream.  Others suggested that WLAs be set at
the  greater of either the criteria or the background concentration.

      Many commenters supported the use of multiple source TMDLs to prevent
point sources from bearing a disproportionate share of the burden  in achieving
water quality goals when nonpoint source contributions dominate.   Some
commenters were concerned that  developing multiple source TMDLs would be very
resource intensive,  and encouraged EPA to specify reasonable limits in the
interim while TMDLs are developed.

      c.    Final Guidance:  In response to numerous comments disagreeing with
the  proposal to set WLAs equal to zero when background exceeds criteria, EPA
has  removed this provision from the final Guidance.  EPA first and foremost
recommends developing TMDLs to address discharges to non-attained  waters.
However, EPA also recognizes the multitude of factors that need to be
considered in the absence of a TMDL when background water quality
concentrations exceed chronic narrative or numeric criteria, or Tier II
values.

      When uncertainty regarding loadings and load reductions are  a
consideration, a phased approach to TMDL development may be appropriate.  For
a more extensive discussion of multi-source,  multi-media TMDLs, see the
introduction to section VIII.C in this document.  Permitting decisions for
discharges to non-attained waters are addressed more fully in the  provisions
and  accompanying supplementary information document discussion for eliminating
mixing zones for BCCs (section VIII.C.4), considering intake water pollutants

-------
286    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

(section VTII.E), arid in the supplementary information document  discussion on
the basis for developing WQBELs at section VIII.E.2.h.

8.    Pollutant Degradation

      a.    Proposal:  Both Options A and B allowed TMDLs to account for
degradation of a pollutant provided two conditions were met.  The first
condition was -that the regulatory authority must have information regarding
the rate of degradation of the pollutant in the form of field studies or other
relevant information.  As discussed in the preamble to the proposal, field
studies, if used, must document that degradation of the pollutant will  occur
under the full range of critical conditions expected to be encountered, and
should quantify the degradation.  Critical conditions should include the
design conditions that are established for the implementation of criteria in
ambient waters as well as other conditions such as periods of stratification
of the water body and variability of the facility effluent flow  rate.   The
preamble to the proposal also indicated that if field study information was
not available, the regulatory authority could use other relevant information
such as literature references from similar sites.  Regardless of the type of
information used, all information would have to be reviewed by the regulatory-
authority and found to be scientifically valid.

      The second condition was that the studies take into account factors
other than pollutant degradation that may affect the concentration of the
pollutant in the water column including but not limited to resuspension of
sediments, speciation and transformation.

      b.    Comments;  Several commenters supported the procedures that
provided for consideration of the environmental fate of a pollutant in  the
development of TMDLs. One commenter suggested that fate and transport should
be considered in the development of TMDLs whenever suitable data such as
existing literature^or field data from similar sites are available.  One
commenter suggested"that EPA should direct States to gather site-specific
information in scientifically sound studies.  Another commenter  suggested that
the regulatory agencies be responsible for collecting the necessary data.

      Several commenters suggested that the final guidance  specify that losses
from the water column due to physical transfer to other media  (i.e., through
volatilization, bioaccumulation, sorption to sediments) are not  acceptable
fate processes for increasing TMDL allocations, since the pollutants may
ultimately be re-released to the water column.  Other commenters suggested
that no transport processes should be precluded from consideration in the
development of TMDLs and WLAs. One commenter suggested that pollutant
degradation should not be accounted for unless rigorous studies  concerning
sediment re-suspension, speciation and transformation are also  incorporated
into the calculations.

      One commenter suggested deleting the  section on pollutant  degradation
from the final Guidance because it was not  discussed in enough  detail by the
technical work group.

      One commenter fully  supported consideration of degradation and  transport
outside the mixing zone.  The commenter recommended that existing EPA guidance
such as Interim Guidance on  Interpretation  and  Implementation of Aquatic Life
Criteria for Metals  (May 1992) be used.

      c.    Final Guidance:  The  final Guidance  retains the provision that
TMDLs, wasteload allocations in the absence of  a TMDL and wasteload
allocations for purposes of  determining the need for WQBELs should  be based on
the general assumption  that  a pollutant does not  degrade.   Like the proposal,
however, it also allows degradation to be taken into account  on the basis  of
information from scientifically valid field 'Studies or  other  relevant
information,  including  the results of properly  calibrated water quality
modeling.

-------
                               Section Vm.C: TMDLs                           287
      Each of the Great Lakes States has already adopted a narrative criterion
specifying that waters shall be free from pollutants that settle to form
objectionable deposits.  EPA's existing NPDES requirements  (40 CFR 122.44(d))
require establishment of permit effluent limitations to meet these narrative
and other criteria.- Upon adoption by States and Tribes or promulgation by
EPA, general condition 6 in the final Guidance requires that TMDLs prevent the
accumulation of pollutants in sediments to levels causing impairment of
designated or existing uses.  Inclusion of this provision in the final
Guidance reflects EPA's concern about sediment quality in the Great Lakes
System and a recognition that it may often need to be considered.

      EPA is currently developing new methods for preventing sediment
contamination.  The first step is to develop numeric sediment criteria
guidance.  Once a State adopts sediment criteria into its State water quality
standards, regulatory authorities will need to factor such criteria into the
TMDL and NPDES permitting process, in addition to applicable narrative
criteria pertaining to the formation of objectionable deposits.

      To the extent that volatilization does not represent a permanent loss
from the Great Lakes System, current atmospheric loadings of volatile
pollutants will be accounted for in determining background concentrations.  In
fact, atmospheric transport and degradation processes will influence the
amount of volatiles available for re-entrainment in the water.  Accordingly,
volatilization losses can be considered when setting TMDLs, wasteload
allocations in the absence of TMDLs and preliminary wasteload allocations for
the purposes of determining the need for WQBELs.  It would be extremely
difficult to establish a significant loss of ambient pollutants as a result of
bioaccumulation.  The confounding factors, including the potential loss of
pollutants from the water column by bioaccumulation into plants,
invertebrates, fish tissue biodegradation via depuration are not expected to
be quantifiable enough to meet the second condition of biodegradation.

9.    Mixing Zone Studies

      a.    Proposal;  Option B allowed any interested party to prepare a
mixing zone demonstration and allowed the permitting authority to modify the
dilution fraction described above in accordance with such studies.  Proposed
procedure 3B.E described several required elements of a mixing zone study, all
designed to address the area of mixing that can be allowed consistent with
attainment of water quality standards.

      b.    Comments;  Commenters raised questions about specific components
of the mixing zone demonstration requirements.   Several commenters questioned
the requirement for documentation of the substrate and geomorphology of the
mixing zone.  Other commenters suggested that the analysis of attraction of
organisms to the mixing zone is difficult to assess or predict.  Another
comtnenter questioned the requirement to determine whether the habitat supports
endemic species or naturally occurring species, and asserted that it is
essentially a useless exercise.  The commenter suggested that by definition,
the surface water into which the discharge occurs will support whatever
aquatic organisms inhabit the area, and whatever species is protected by the
criteria may be present and may pass through the mixing zone.

      c.    Final Guidance:  The final Guidance adopts the mixing zone
demonstration language as the alternative to the mixing zones specified for
both OWGLs and tributaries.  The language has been modified to require
consideration of potential impacts to threatened and endangered species
consistent with the*Endangered Species Act and otherwise to enhance clarity.
The mixing zone demonstration provision provides flexibility to allow a
greater dilution fraction than otherwise provided in sections D and E, as well
as an exceedance to the FAV cap, to better reflect site-specific
considerations.

-------
288    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      EPA believes that characterizing the substrate  and  geomorphology of any
potential mixing zone site is necessary to characterize potential  impacts on
water quality.  Similarly, the effects of any potential mixing zone on endemic
or naturally occurring species must also be  considered.   EPA recognizes that
dischargers may be required to collect additional  data  and perform additional
analyses in order to qualify for a mixing zone exception;  however,  EPA
believes it is reasonable and appropriate to require  this information if
dischargers wish to use values greater than  the default values specified.

10.   Pollution Trading Opportunities

      As described in the proposal, the TMDL process  provides an opportunity
for pollution trading in the water quality program as long as CWA  goals and
requirements are met.  Effluent limits and nonpoint source controls,  for
example, must be designed, maintained and enforced so that water quality
standards and other statutory and regulatory requirements are met.   For
purposes of the final Guidance, trading refers to  approaches which introduce
market incentives into water quality control decisions  by acknowledging the
ability of a point source to achieve water quality-based  loading reductions
through creative, enforceable market mechanisms.

      The Guidance encourages States to look for pollution trading
opportunities as TMDLs are established.  However,  trading opportunities may be
limited by the general condition s and specific requirements (e.g.,  mixing
zones for BCCs) that apply to all TMDL development.

-------
                              Section VELD:  Additivity                           289
D.    Additivitv

1.    Background

      EPA has traditionally developed numerical water quality criteria on a
single pollutant basis.  While some potential environmental hazards involve
significant exposure to only a single compound, most instances of
contamination involve mixtures of two or more pollutants.  The individual
pollutants in such mixtures can act or interact in various ways which may
affect the magnitude and nature of potential risks or effects on human health,
aquatic life and wildlife.  The potential actions or interactions between
multiple pollutants in a mixture can be divided into the following general
categories.

      1.  The effects to an organism from exposure to a mixture of two or more
pollutants may be greater than simply adding the predicted effect from
exposure to each pollutant ("synergistic effect").

      2.  The effects to an organism from exposure to a mixture of two or more
pollutants may be less than simply adding the predicted effect from exposure
to each pollutant ("antagonistic effect").

      3.  Pollutants in mixtures may exert an "additive effect."  Two types of
additive effects are possible.

      a.  "Response addition" refers to situations where pollutants in a
mixture each independently cause adverse effects to an organism without
significant interaction.  "Response addition" is generally considered to be a
valid assumption in the absence of contrary data when the response rate from a
pollutant is low (e-.g., number of organisms showing an adverse effect from a
pollutant is small).   In these circumstances, the expected toxicity of the
mixture can be estimated by adding the predicted response (risk) from each
individual pollutant.

      b.  "Dose addition" refers to situations where multiple pollutants may
effect an organism through a similar mechanism of action.  The Toxicity
Equivalence Factor approach in procedure 4.A of appendix F to part 132 is an
example of dose addition. In these circumstances, the expected toxicity of the
mixture can be estimated by converting the individual doses of the pollutants
to an "equivalent" dose of a reference chemical  (e.g., 2,3,7,8-TCDD)  and
adding these amounts.  The "equivalent" dose is based on the relative toxicity
of the individual pollutants to the reference chemical.

      EPA's current regulations and the final Guidance generally account for
the active and interactive effects of discharged pollutants on aquatic life
through direct exposure of test organisms to a point source effluent in whole
effluent toxicity (WET) tests (procedure 6 of appendix F to part 132; 40 CFR §
122.44(d)).  The use of such tests to determine effects of multiple pollutants
on aquatic life is a well established component of both Federal and State
regulatory programs,  and is discussed in subsection 5 below.

      EPA currently has no specific National guidance regarding consideration
of additive or interactive effects of pollutants on wildlife, and has not
included such provisions in the final Guidance.  This issue is discussed in
subsection 8 below.

      The remaining discussion in this section is devoted primarily to
addressing the effects on human health resulting from exposure to pollutant
mixtures.  As discussed further below, the final Guidance includes provisions
addressing the additive effects on human health, but does not include
provisions addressing any possible synergistic or antagonistic effects.

      In order to address the effects on human health resulting from exposure
to pollutant mixtures, EPA published principles and procedures for conducting

-------
290    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

human health risk assessments for multiple pollutants in 1986  ("Guidelines  for
the Health Risk Assessment of Chemical Mixtures," 51 FR 34014),("Guidelines
for Chemical Mixtures").  The Guidelines for Chemical Mixtures  recommend using
"dose addition" to estimate the combined effects from threshold toxicants
acting by similar mechanisms or affecting common organs  (generally non-
carcinogens) and "response addition"  for non-threshold toxicants acting
independently  (generally carcinogens), in the absence of contrary information
on the specific mixture.  An assumption of dose or response addition  could
result in errors in risk estimates if synergistic or antagonistic interactions
occur  (i.e., additivity assumptions could result in overestimates or
underestimates of the actual risks).  Thus, an assumption of additivity is  not
a "worst-case" assumption, but a reasonable assumption when specific
information on pollutant interaction  is not available.

2.    Additivitv Considerations in Other EPA Programs

      Several programs within EPA address the problems of exposure to multiple
chemicals.  For example, the Office of Solid Waste regulations  for Hazardous
Waste Burned in Boilers and Industrial Furnaces require permitting authorities
to add the individual cancer risk for several carcinogenic metals  (the
"response addition"' approach) when developing permit limits for those metals,
and establish a maximum aggregate cancer risk level of no greater than  10"5
((40 CFR 266.106(b)(2),  (c)(2), and  (d)(3); 56 FR 7134, 7165-66 (Feb. 21,
1991)) .

      The risks from exposures to multiple pollutants are also  considered in
the Superfund program.  The Risk Assessment Guidance for Superfund Volume 1
Human Health Evaluation Manual  (Part  A)  (RAGs) requires that at most  Superfund
sites it is necessary to assess potential health effects of more than one
chemical.  This is necessary because  considering only one chemical at a time
might significantly-underestimate the risks associated with simultaneous
exposures to several substances.  To  assess the overall potential for cancer
and non-cancer effects posed by multiple chemicals, the Superfund RAGs
recommends using EPA's Guidelines on  Chemical Mixtures referenced above.
Specifically, for carcinogens, RAGs recommends that the individual cancer risk
from each carcinogen in a mixture should generally be added to  estimate the
total cancer risk  (the  "response addition" approach).  For non-carcinogens,
RAGs recommends using the hazard index approach which is discussed below.

3.    Existing State Water Quality Standards

      Four Great Lakes  States  (Illinois, Minnesota, Wisconsin,  and
Pennsylvania) have adopted provisions in their water quality standards  to
address additive effects to human health from exposure to multiple
carcinogens.  Minnesota and Wisconsin water quality standards use a  "response
addition" approach by incorporating an assumption in criteria development that
the risk from  a  combination of  carcinogens  in a mixture  is equal to  the sum of
the risks associated with exposure to each  individual pollutant in the
mixture, unless  an alternative model  is supported by credible  scientific
evidence.  Pennsylvania water quality standards provide that the State  may
consider synergistic, antagonistic and additive toxic impacts when developing
their water quality.standards.  In addition, three of the Great Lake States
also  specifically address possible additive effects of pollutants by
establishing acceptable maximum cancer risk levels for mixtures. Minnesota  and
Wisconsin have adopted  an acceptable  cancer risk level of 1 in  100,000  (10'5)
for exposures  to either individual pollutants or to mixtures of pollutants.
Illinois water quality  standards do not specify a particular assumption for
considering additive effects, but  include  a maximum acceptable  cancer risk
level  of 1  in  one million  (10"*) for individual pollutants and  10'3 for mixtures
of substances.

-------
                              Section VELD:  Additivity                          291
4.    Proposed Guidance Overview

      The preamble for the proposed Guidance discussed several possible
approaches to address additive effects from multiple pollutants.  Proposed
regulatory language was provided for two specific options, each with separate
provisions related to aquatic life, wildlife and human health.  One approach
was developed by the Great Lakes Initiative Committees, modified to delete the
application of toxicity equivalency factors  (TEFs) for PCBs to wildlife.  The
other approach was developed by EPA.  Neither approach addressed the possible
toxicologic interactions between pollutants in a mixture  (e.g., synergism or
antagonism) because of the limited data available on these interactive
effects.  The quantitative significance of toxic interactions between
chemicals in mixtures at environmental levels of exposure is often difficult
to assess.  For example, most of the data available on toxicant interactions
are derived from acute toxicity studies using experimental animals.  The use
of acute toxicity data to assess the potential interactions in chronic
simultaneous exposures is also difficult unless the same mechanisms of
interaction are known to apply.  Additionally, the limited data available on
toxicant interactions from both chronic and acute studies indicate that the
chronic interactions can be either greater or less than the observed acute
interactions. (Technical Support Document on Risk Assessment of Chemical
Mixtures, 1990.  EPA/600/8-90/064).  Due to these data limitations, neither
specific approach presented in the preamble included procedures to estimate
synergistic or antagonistic effects from mixtures of pollutants.  Differences
between the preamble approaches for addressing additive effects are discussed
below.

5.    Aquatic Life

      Both approaches in the proposal accounted for the additive effects on
aquatic life through use of whole-effluent toxicity (WET) limitations.
Commenters on the proposal generally supported the use of WET to account for
the effects of additivity on aquatic life.

      EPA continues,to believe that the WET provisions in procedure 6 of
appendix F to part 132 are a reasonable mechanism to account for additive
effects to aquatic life and, therefore, are retaining those provisions in
appendix F to part 132.   Because the provisions for WET have been adequately
incorporated in procedure 6 of appendix F to part 132, however, EPA has
removed the references to WET testing from the additivity provisions in
procedure 4 of appendix F to part 132.

6.    Human Health

      The preamble to the proposed Guidance presented options for addressing
the additive effects to human health from both carcinogens and
noncarcinogens.   The following sections discuss the proposed options for both
carcinogens and noncarcinogens, the major comments received on these
provisions, EPA's response, and the provisions for additivity in the final
Guidance.

      a.    Carcinogens.

      i.    Proposal:   The preamble to the proposal presented two specific
approaches for implementing an additivity provision for carcinogens.  One
approach was developed by the Initiative Committees and specified a procedure
for considering the additive effects of human health carcinogens during
development of water quality based effluent limits (WQBELs).  Under this
approach, the permitting authority would identify those carcinogens that had
been detected in the effluent; had a Tier I criterion or Tier II value; and
for which a WQBEL was required.  The permitting authority would then establish
waste load allocations for those carcinogens at levels that would ensure that
the total cancer risk for those carcinogens did not exceed 10"5 in the
effluent.

-------
292    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      The second approach was developed by EPA and addressed additive effects
of chemical mixtures through interpretation of State and Tribal narrative
water quality criteria.  The narrative criteria of each of the Great Lakes
States currently require that all waters be free from toxic substances  (e.g.
substances that injure or produce adverse responses in humans, wildlife, or
aquatic life).  This approach, similar to the approach of the Initiative
Committees described above, specified a maximum total cancer risk of one 'in
one hundred thousand  (10~5)  and assumed that the incremental cancer risk of
each carcinogen in the mixture was additive in the absence of a contrary
determination by the State or Tribe.  Discharges would be controlled to ensure
attainment of a 10"5 total cancer risk in ambient waters.

      ii.   Discussion of Significant Comments on the Assumption of Additivitv
            for Carcinogens.

      Comment:  Many commenters supported the assumption of additivity  for
carcinogens arguing that it was necessary to be consistent with existing EPA
guidance that recommends an assumption of additivity unless there is specific
information to the contrary.  In addition, commenters stated that this
approach would be consistent with provisions of the Great Lakes Water Quality
Agreement  (GLWQA) calling for the consideration of interactive effects  of
toxic substances, and that it would further the GLWQA's goal of  "virtual
elimination" of toxic pollutants.

      Numerous other commenters disagreed with applying an assumption of
additivity for carcinogens.  Some argued that the procedure for deriving human
health criteria for'carcinogens already has multiple conservative assumptions
and that the assumption of additivity is just one more conservative assumption
that is not based on sound science.

      Many commenters who disagreed with applying an assumption of additivity
in the absence of specific data for a group of chemicals, cited the 1992
Science Advisory Board  (SAB) report,  "Evaluation of the Guidance for the Great
Lakes Water Quality Initiative," which recommended that multiple carcinogens
be considered on a case-by-case basis instead of incorporating additivity as a
default mechanism in all circumstances.  The SAB report recommended a case-by-
case consideration' of possible additive effects because carcinogens are known
to act by a wide variety of mechanisms and to target different organs.  The
SAB report also stated, however, that an assumption of additivity could be
appropriate for compounds that act at the same receptor  (such as dioxin,
furans and PCBs).

      Response:  Although EPA believes an assumption that carcinogens exert
additive effects in the absence of contrary data is generally reasonable and
recommends it's use, for the reasons  discussed below, the final Guidance does
not require States or Tribes to adopt an assumption of additivity for
carcinogens.  This decision  is consistent with the SAB comments  discussed
above, and with comments by  the Great Lake States that flexibility was
necessary in order to address any potential difficulties in implementation of
an additivity provision for  specific  chemicals in their respective water
quality programs.

      EPA carefully considered comments that an assumption of additivity is
overly conservative.  As discussed in the preamble to the proposed Guidance,
an additivity assumption could result in overestimates or underestimates of
the actual risks from multiple pollutants if synergistic or antagonistic
interactions occur.  Thus, EPA maintains that if an assumption of additivity
is used, that it would  not be overly  conservative or a  "worst-case"
assumption, but a reasonable assumption when specific information on pollutant
interaction  is not available.

      EPA also recognizes, however, that there is some  scientific debate on
the assumption of additivity for  carcinogens and agrees with  comments  in the
SAB report that multiple carcinogens  in a mixture should be considered on  a

-------
                              Section VELD:  Additivity                           293
case-by-case basis to determine whether there is adequate data to characterize
the potential interactions in the mixture because carcinogens can act through
a wide variety of mechanisms and target different organs.  The final Guidance,
consistent with these comments and with the 1986 Guidelines on Chemical
Mixtures, does not preclude States or Tribes from using any such data, when
available, to characterize the potential carcinogenic effects from the
interaction of pollutants.  In the majority of cases, however, these data will
not be available.  In these situations, EPA believes it would be appropriate
for States and Tribes to assume that the total upper bound carcinogenic risk
from a mixture is equal to the sum of the upper-bound incremental risk levels
for the individual components of the mixture.  EPA also believes this
assumption is valid as long as the carcinogens are acting independently.  EPA
has determined that this assumption of independence of action for carcinogens
is a valid assumption at the concentration levels of concern in the Great
Lakes System.  At these low concentrations, the competition for receptor sites
will be reduced and the likelihood of significant interactions (e.g.,
synergistic or antagonistic effects) between pollutants will be minimal.  In
other words, if the probability of developing cancer from one pollutant is
independent of the probability of developing cancer from another pollutant,
then the probability of developing cancer from both substances may be obtained
from summing the individual probabilities.  Therefore, in the absence of data
on the interactions among the carcinogens, EPA believes that it would be
reasonable for States and Tribes to estimate the total upper-bound incremental
cancer risks to human health by adding the separate upper-bound incremental
cancer risks from each pollutant in the mixture.  This assumption is
consistent with the 1986 Guidelines for Chemical Mixtures.

      This assumption of additivity for carcinogens at low concentrations has
been adopted in regulations and reports developed by other federal agencies.
For example, the Food and Drug Administration procedures governing
carcinogenic impurities in color or food additives assumes in the absence of
specific contrary information on the interactions among the carcinogenic
impurities that the risks incurred from the presence of multiple carcinogenic
impurities in a color or food additive are additive, and sum the estimated
upper bound risks of these pollutants  (53 FR 33118, August 30, 1988).

      This assumption of additivity is also supported by information in the
National Research Council Report, "Complex Mixtures: Methods for In Vivo
Toxicity Testing" National Research Council, 1988.  The Committee Report
analyzed epidemiologic studies and current models to predict toxicity of
mixtures containing multiple carcinogens, and concluded that effects of
exposures to pollutants with low response rates usually appear to be additive
(Executive Summary, at p.3) '. The report also summarizes data demonstrating
additive effects from multiple carcinogens at low chemical concentrations.
This data was based on several additivity models, including the two models
most commonly used by EPA for low-dose extrapolation: (1) multistage and (2)
the Moolgavkar models (Chapter 5, and appendix E, at p.  200).  The final
Guidance recommends use of the linearized multistage model to determine cancer
potencies.  The cancer potencies are used in the derivation of human health
criteria  (see section V.C.2 of this document for a complete discussion of the
use of the linearized multistage model).

      iii.  Discussion of Significant Comments on Application of Additivity
            Provision for Carcinogens.

      Comment:  Several Great Lake States commented that both of the
additivity approaches for human health presented in the preamble would be very
difficult to implement,  particularly if applied to the ambient waters, when
developing discharge permits and assessing ambient water quality.  They stated
that it would be difficult to apply in the ambient environment because of the
variability of concentrations of chemicals in the ambient environment due to
changing environmental conditions, and the varied and sometimes uncontrolled
sources of contaminants.  Some recommended the continued application of
numerical criteria on a single pollutant basis for discharge permit

-------
294    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


development and ambient water quality assessment.  Other States  commented that
it would be more efficient to require the use of a more stringent risk  level
for individual carcinogens (e.g., 10"*) to address additivity than to require
the use of a specific implementation provision.  Three Great Lake States
currently use a risk level of 10"6 for individual carcinogens (Illinois, New
York and Pennsylvania).  Some Great Lake States supported an effluent oriented
approach to account for additive effects of carcinogens.  Other  commenters
urged that the additivity provisions need to be structured to specifically
account for carcinogens in both permitted discharges and ambient water
quality.

      Regarding the appropriate risk level, several commenters stated that
when more than one carcinogen was in an effluent or a surface water, human
health criteria for those pollutants should be developed based on an
assumption of dose addition with a total maximum risk of 10"*. Others believed
that to be consistent with goals associated with human health protection, that
the total cancer risk from the-mixture should be 10"5 (e.g.,  the maximum risk
provided for risks from individual pollutants).  Other commenters, while
continuing to disagree with provisions for an assumption of additivity, argued
that if an aggregate risk level ceiling is established, then that ceiling
should be set at a risk level no greater than 10"4.

      Response:  Based on careful consideration of the comments, EPA has
determined that it is necessary and appropriate to provide adequate
flexibility to States and Tribes to adopt and implement provisions addressing
the additive effects of multiple carcinogens tailored to their individual
water programs.  Accordingly, the final Guidance does not specify a detailed
methodology for implementing additivity similar to the methodologies for
developing criteria to protect human health, aquatic life, and wildlife.  EPA
considered specifying in detail how States and Tribes would need to implement
a general additivity provision, but decided for the reasons cited below that
it was appropriate to provide sufficient flexibility at this time to ensure
the provisions are fully implementable.  EPA has concluded that  this approach
will result in State and Tribal water quality standards that will best  ensure
that human health is protected from potential adverse additive effects  from
chemical mixtures.

      As discussed above, the Critical Programs Act requires that EPA specify
numerical limits on pollutants in ambient Great Lakes waters to  protect human
health and provide guidance on minimum water quality standards and
implementation procedures.  EPA has interpreted this language as requiring a
minimum level of protection for human health throughout the entire Great Lakes
System.  In light of the statutory requirements that the Great Lakes System be
protective of human health, and the potential for adverse effects from
exposure to multiple carcinogens in mixtures, EPA believes that  measures must
be taken to ensure that human health  is protected from the additive effects of
carcinogens.  However, EPA recognizes, as many commenters noted, that there
are a number of difficult issues involved in attempting to implement measures
to ensure that human health is protected from the potential adverse additive
effects of carcinogens; and that States and Tribes are in the best position to
ensure, within their existing State programs, that human health  is protected
from the additive effects of carcinogens.  EPA was concerned that provisions
that are difficult to incorporate into existing water programs would
discourage or significantly impede State or Tribal development of procedures
to address the potential effects from multiple chemicals.  On the other hand,
EPA was concerned that  if too much flexibility was allowed that  the provisions
would become meaningless and not fulfill the statutory goal of improving
consistency within the Great Lakes System.  The additivity provisions in
procedure 4 of appendix F to part 132 described below have attempted to
balance these two competing demands to ensure that minimum additivity
provisions will be developed by the Great Lakes States and Tribes that  are
both implementable and  will provide appropriate protection of human health.

-------
                              Section Vffl.D:  Additivity                          295
      iv.   Final Guidance.  Procedure 4 of appendix F specifies that the
Great Lakes States and Tribes shall adopt provisions to protect human health
from the potential adverse additive effects from the carcinogenic components
of chemical mixtures in effluents.  EPA limited the procedure 4 of appendix F
to part 132 to effluents because of potential uncertainties or technical
difficulties in attempting to quantify how chemical mixtures act in the
environment.  The techniques for modeling the fate of multiple pollutants  in
the ambient water are not as well developed as for individual pollutants.
Because the science is still developing and because of concerns raised by  the
Great Lakes States responsible for implementing the final Guidance, EPA has
decided to limit the requirement for additivity to effluents.  This is
consistent with the approach advocated by the Committees of the Initiative in
the proposed Guidance.  In addition, since this is the first time EPA has
required Great Lakes States and Tribes to adopt an additivity provision for
specific chemicals into their water quality standards, EPA believes it is
reasonable to initially limit the provisions to effluents to reduce the
potential implementation difficulties raised by commenters.  EPA believes  as
States, Tribes, and EPA gain more experience in considering multiple
pollutants in establishing permit limits, that the provisions could be extend
to the ambient waters.

      EPA believes States and Tribes have several options that will provide
protection to human health from the potential adverse additive effects from
carcinogens in effluents in the Great Lakes System.  One option States and.
Tribe could consider would be to require that the total cancer risk in
mixtures cannot exceed an incremental cancer risk of one in 10,000  (10"4)  to
protect human health,  EPA recommends  an upper bound lifetime incremental
cancer risk to an individual of more than 10"4 for several reasons.  First, EPA
believes that the establishment of this minimum level will improve consistency
in permit limits within the Great Lakes System.  Improvement in the
consistency of water quality standards and permit limits in the Great Lakes
System was a primary goal of the Great Lakes Critical Program Act  (CPA)
amendments to section 118 of the Clean Water Act  (CWA).

      Second, as noted in section V (Human Health) of this document, EPA
believes that ensuring protection to human health in the risk range of 10"4 to
10"6 is acceptable and consistent with the CWA's objectives.  Adoption of this
provision would result in a maximum incremental cancer risk of 1 x 10"4 for
mixtures of carcinogens to protect all populations   Specification of 1 x  10"4
as the minimum acceptable level of protection for human health from exposure
to multiple carcinogens is intended to ensure that all populations are
sufficiently protected once the final Guidance provisions are fully
implemented in the ambient water and in individual permits, not simply those
individuals consuming 15 grams/day of fish and consuming 2 liters of water.

      EPA has long maintained that 1 x 10"4 is within an acceptable range of
risks.  For example, the Superfund program uses 10"* as its point of departure
when developing its preliminary remediation goals for a site, but selects
remedies that reduce the threat from carcinogenic contaminants at a site such
that the excess risk from any medium to an individual exposed over a lifetime
generally falls within a range from 10"4 to 10"6.   The U.S.  Court of Appeals for
the District of Columbia Circuit held that risk levels in a range between  10"4
and 10"* that are used as part of the National Contingency Plan (Superfund) are
adequately protective of human health (State of Ohio v. EPA, 997 F.2d 1520
(D.C Cir. 1993)) .  In addition to the Superfund program, the Office of
Drinking Water uses, a risk range of 10"4 to 10"* in setting the maximum
contaminant level goals (MCLGs)  for contaminants in drinking water  (56 FR
3531, Jan. 30, 1991).

      As an alternative to the above approach.  States and Tribes could
provide considerations for the additive effects of multiple carcinogens by
lowering the human health criteria for individual carcinogens to levels
corresponding to an upper-bound incremental cancer risk of one in one million

-------
296    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

( 1 x 10"6)  using the methodologies adopted pursuant to  132.4.  There are
currently three Great Lakes States that use  l x 10"6 in  determining criteria
for individual carcinogens  (New York, Illinois, and Pennsylvania).  This
approach is simple to implement, and  based on an analysis  completed for the
regulatory impact analysis  (RIA), should generally provide protection  to  human
health, including the health of any highly exposed populations, equal  to  or
greater than the level of protection  provided by the 1  x 10"4 cancer risk
level.

      Evidence exists that some individuals  in the Great Lakes area  (less than
1% of the population) may be consuming up to 150 grams  per day of fish, or ten
times more than the 15 grams per day  of fish that would be used in deriving
individual pollutant criteria at a l  x 10~* risk level.  Assuming that  all of
the 150 grams of fish consumed contain the maximum residues of pollutants
permissible after implementing individual pollutant criteria established  at  a
10"* risk level for a. 15 gram per day  consumer ("10"* criteria" for purposes of
this discussion), the risk posed to the 150-gram per day consumer as the
result of exposure to one chemical would be  1 x 10"5.  If there were two
carcinogens in the water at such levels, the total risk to the 150-gram per
day consumer would be 2 x 10"5, assuming the  additivity  of  carcinogenic risk.
There would have to be a carcinogenic risk in ambient waters greater than that
which exists when ten carcinogens are present at 10"* criteria levels for  the
additive carcinogenic risk to the 150-gram per day consumer to be greater than
10"4.  As discussed, EPA has determined that  a carcinogenic risk of 10^ is
within a range of risk levels that is protective of human  health.

      EPA believes based on information in the RIA that it will be unlikely
that mixtures of carcinogens in the Great Lakes System  will, after full
implementation of individual pollutant criteria in the  ambient water and
individual permits, contain a carcinogenic risk equivalent to that which
exists when ten  carcinogens are present at 1 x lO'* criteria levels.  It is
even more unlikely that those relatively few individuals who consume up to 150
grams per day of fish will catch fish exposed to the few waters containing
such high risks.  Accordingly, it will be extremely unlikely that after
implementing individual pollutant criteria State or Tribal selection of this
approach will lead to risks to highly exposed subpopulations that exceed  a
cumulative cancer risk from exposure  to water-borne pollutants of 1 x  10"*.
Additionally, in the event that it is demonstrated that this approach  is
insufficient to protect human health, including the health of any highly
exposed subpopulations, then procedure 1 of  appendix F  to  part 132 provides
that States or Tribes must modify the human  health criteria to provide
additional protection appropriate for these  subpopulations.       Finally,
States and Tribes can adopt any other scientifically defensible approach  that
will protect humans^ from the potential adverse additive effects from
carcinogens.   EPA will need to review any such provisions, on a case-by-case
basis when evaluating whether State and Tribes submissions are consistent with
procedure 4 of appendix F to part 132.

      .b.    Non-carcinogens.

      i.    Proposal:  As discussed in the preamble for the proposed Guidance
and in the 1986  Guidelines for Chemical Mixtures, the use  of the additivity
assumption for non-carcinogens is most appropriate when the pollutants in a
mixture elicit the same type of effect by the same mechanism of action.
However, because information on the mechanism of action is normally limited,
the 1986 Guidelines for Chemical Mixtures recommended that when two or more
compounds produce adverse effects on  the same target organ that the effects
should be considered additive.

      To estimate the potential non-carcinogenic risk from a mixture,  the 1986
Guidelines for Chemical Mixtures  recommended the use of the Hazard  Index   (HI)
approach.  The HI provides  a  rough measure of likely toxicity.   It  does not
define dose-response relationships, and its  numerical value  should  not be

-------
                              Section Vffl.D:  Additivity                          297
construed to be a direct estimate of risk.  The equation used to estimate the
HI sums the ratios of the actual exposures to the chemicals to the RfD  (see 58
PR 20941) .   The proposed Guidance requested comments on the use of the hazard
index approach.

      An alternative approach to the HI discussed in the proposed Guidance and
in the Technical Support Document on Risk Assessment for Chemical Mixtures is
the use of toxicity equivalency factors (TEFs) approach.  This approach
involves estimating the potency of less well-studied components in a mixture
relative to the potency of better studied components, using data from
comparable types of. in vitro and short-term in vivo assays.  This approach has
been used to estimate the toxicity of mixtures of chlorinated dioxins and
dibenzofurans by using extensive data on the in vitro activity of these
compounds.   To date, 17 TEFs for the Chlorinated Dibenzo-p-dioxins (CDDs) and
Chlorinated Dibenzofurans (CDFs) have been developed.  The data supporting
these TEFs are summarized in a monograph of EPA's Risk Assessment Forum
entitled "Interim Procedures for Estimating Risks Associated with Exposures to
Mixtures of Chlorinated Dibenzo-p-Dioxins and -Dibenzofurans and 1989
Update"(USEPA, 1989), which is available in the public docket for this
rulemaking.  These interim procedures were reviewed by EPA's Science Advisory
Board in 1986 and adopted as interim Agency policy in 1987.  These procedures
were also adopted for international use by the North Atlantic Treaty
Organization.

      The two additivity approaches for non-carcinogens presented in the
preamble to the proposal differed in the application of these procedures.  One
approach would require, if adopted, that non-cancer effects be considered
additive for those pollutants for which available scientific information
supports a reasonable assumption that pollutants produce the same adverse
effects through the same mode of action, and for which TEFs could be
calculated.  Thus, this approach if subsequently promulgated would establish a
general requirement, for States and Tribes to develop additivity protocols for
classes of pollutants when sufficiently supported by scientific information.
The second approach, if promulgated, would require the application of this
additivity assumption only for those pollutants for which TEFs are set forth
as part of the final Guidance.  This would initially include CDDs and CDFs,
but more pollutants could be addressed through any future revisions to the
final Guidance or State and Tribal provisions.

      ii.   Discussion of Significant Comments.

      Comment:  Several commenters supported the EPA's proposed approach to
limit the assumption of additivity for non-carcinogenic effects to those
situations when data are available to demonstrate that the chemicals produce
the same adverse effects through the same mechanism of action.  Other
commenters stated that the Hazard Index approach has little scientific support
because pollutants with concentrations below the RfD are considered to have
zero risk and summing the ratios from the risk for these pollutants below the
RfD should also be zero.  Other commenters objected to the use of the HI
approach because it fails to incorporate the shape of the dose-response curves
for the chemicals.  Other commenters supported the use of the HI approach to
address additivity for non-carcinogens.  Specific comments on the use of TEFs
are addressed in section VIII.D.4 below.

      Several commenters stated that while neither of the proposed approaches
was entirely adequate, the first approach (58 FR 20943) was preferable because
it would require regulators to quickly develop new additivity procedures as
new scientific information emerges, without waiting for formal revision of the
Great Lakes Guidance or State or Federal laws.

      Response: Consistent with the discussion above for carcinogens, the
final Guidance does not specify an approach for considering the potential
adverse additive effects from noncarcinogenic components of mixtures.
However, EPA agrees, with commenters that the non-carcinogenic effects of

-------
298    Water Quality Guidance for the Great Lakes System -- Supplementary Information Document


individual pollutants should be considered additive only for pollutants for
which available scientific information supports a reasonable assumption that
the pollutants produce the same adverse effects through the same mechanisms of
action.  This approach is consistent with the 1986 Guidelines on Mixtures and
the proposed Guidance.

      EPA carefully considered but does not agree with the comments that the
HI approach has little scientific support.  The RfD used in the HI equation is
an estimate (with uncertainty spanning an order of magnitude) of a daily
exposure to the human population  (including sensitive subgroups) that is
likely to be without appreciable risk of deleterious effects during a
lifetime.  Because the RfD estimate may span an order of magnitude, the
potential risk below the RfD may not be zero as implied by the commenter.  EPA
does acknowledge, however, that there are several shortcomings with the HI
approach as discussed in the Technical Support Document on Risk Assessment for
Chemical Mixtures including the problems identified by commenters on defining
a dose-response relationship.  Because of these shortcomings, EPA does not
believe it is appropriate to require States or Tribes to use the HI approach
when assessing the effects from multiple non-carcinogenic pollutants.

      EPA agrees with commenters who supported the proposed provision that
States and Tribes should  develop TEFs when sufficiently supported by
scientific information.  EPA encourages States to adopt this approach to
ensure that the most recent scientific information will be used for revising
the additivity provisions when it becomes available.  The States and Tribes
will not have to wait for EPA to develop TEFs for different classes of
pollutants or to revise the final Guidance.  EPA anticipates that the
Clearinghouse discussed in section II of this document will provide a
mechanism for sharing information among States and Tribes and in developing
TEFs for classes of compounds other than the CDDs and CDFs.

      iii.  Final Guidance:  Procedur'e 4 of appendix F to part 132 specifies
that Great Lakes States and Tribes shall adopt provisions to protect human
health from the potential adverse additive effects from the noncarcinogenic
components of chemical mixtures in effluents.  As with carcinogens, the final
Guidance does not specify how States and Tribes must account for the potential
adverse additive effects from noncarcinogens.  However, as discussed above,
EPA believes a reasonable approach for States and Tribes would be to include
in their water quality  standards that the noncarcinogenic effects of
individual pollutants be considered additive for pollutants for which
available scientific information supports a reasonable assumption that the
pollutants produce the  same adverse effects through the same mechanisms of
action.   EPA continues to support the HI approach as a valid and
scientifically credible approach, but EPA considers the use of TEFs as more
appropriate in most circumstances because it allows the use of all data to
assess the potential risks from non-carcinogens.

7.    Toxicitv Equivalency Factors/Bioaccumulation Equivalency Factors

      a.    Proposal:   Both approaches in the proposal presented the TEFs for
the 17 CDDs/CDFs included  in the Risk Assessment Forum report cited above.
Both approaches, if,adopted, would require that the concentration of each of
the 17 CDDs/CDFs in an  effluent be converted to a 2,3,7,8-TCDD equivalent
concentration by multiplying the concentrations of the CDD or CDF by the
appropriate TEF.  All resultant concentrations would then be added to produce
an equivalent 2,3,7,8-TCDD concentration.  Both approaches would require the
TEFs to be used  to address both cancer and non-carcinogenic effects.  The
preamble for the proposed  Guidance also requested comment on whether TEFs for
"dioxin-like" PCBs should  be included in the use of any additivity provisions
developed to address wildlife effects.

      The TEFs for the  17  CDDs/CDFs address  the toxicity of various chemicals
as compared to 2,3,7,8-TCDD, but do not address differences  in bioaccumulation
potential between the chemicals.  The first  approach presented by EPA in the

-------
                              Section VEtl.D: Additivity                          299
preamble for the proposed Guidance recognized these differences in the
bioaccumulation potentials of CDD/CDF by including specific bioaccumulation
equivalency factors  (BEFs) for the 17 CDD/CDF congeners.  These 17 BEFs for
CDD/CDF congeners were updated and the new values were provided in an August
30, 1994 notice of data availability for public comment  (59 FR 44687) .  The
Committee's approach discussed in the proposal did not include use of BEFs.
That approach assumed that BAFs for all CDDs/CDFs were identical to that '
calculated for 2,3,7,8-TCDD.

      b.    Discussion of Significant Comments.

      Comment:  Several commenters supported the use of the 17 TEFs for
CDDs/CDFs for human health and wildlife.  Others supported the use of TEFs for
human health but not for wildlife, stating that it is unclear to what extent
TEFs developed for mammalian systems are applicable to avian or other wildlife
species.  Other commenters advocated also including TEFs for "dioxin-like"
PCBs for wildlife due to their adverse impact upon wildlife in the Great Lakes
ecosystem, while others stated that there is currently an insufficient
scientific basis for the use of TEFs for "dioxin-like" PCBs.  Many commenters
questioned the use of TEFs in general and argued that the scientific
understanding is not sufficiently developed.  Other commenters argued that it
is not appropriate to assume the effects from the different congeners are
additive because it has not been demonstrated that the different congeners
affect the same receptor organ and cause toxicity through the same mechanism.
Finally, several commenters stated that the TEFs may be appropriate for non-
cancer effects, but not for cancer effects because the TEFs were developed
from toxicity data, not cancer data.

      Several commenters supported the use of proposed BEFs.  Other commenters
generally opposed any additivity requirements, including the proposed TEFs,
but stated if TEFs were adopted, they would support the use of BEFs to account
for the congener-specific BAFs of CDD/CDF congeners.  Other commenters stated
that BEFs should not be used because the data base for development of the BEFs
is not sufficiently developed and the BEFs have not been peer-reviewed or
widely accepted.

      c.    Response for Application of TEFs/BEFs to Human Health:  EPA
carefully considered but does not agree with comments that there is
insufficient data to support the TEFs for dioxins and furans for humans.  The
TEFs for the 17 CDDs/CDFs were developed over several years in collaboration
with experts from throughout the world.  Adoption of the TEF approach for the
17 CDDs/CDFs for human health risk assessment has also been recommended by the
Risk Assessment Forum as discussed above.  EPA is aware that the data
available from long-term in vivo studies are limited for the majority of CDDs
and CDFs.  However, a much larger body of data is available on short-term in
vivo studies and a variety of in vitro studies.  These experiments cover a
wide variety of end points,- e.g., developmental toxicity, cell transformation,
and enzyme induction (aryl hydrocarbon hydroxylase [AHH]).  While the doses
necessary to elicit the toxic response differ in each case, the data
demonstrate that the relative potency of the different compounds compared to
2,3,7,8-TCDD is generally consistent from one end point to-another.

      This information,  developed by researchers in several laboratories
around the world, reveals a strong structure-activity relationship between the
chemical structure of a particular CDD or CDF congener and its ability to
elicit a biological or toxic response in various in vivo and in vitro test
systems.  (Bandiera et al., 1984; Olson et al., 1989; U.S. EPA 1989; NATO/CCMS
1988a,b).  Research has also demonstrated a mechanistic basis for these
observations.  That is,  a necessary (but not sufficient)  condition for
expression of much of the toxicity of a given CDD or CDF congener appears to
be a function of the relative ability of these compounds to bind to a specific
cellular receptor located in the cytoplasm of the cell that mediates most, if
not all, of the toxic end points for these compounds.  This receptor complex
then migrates to the nucleus of the cell, where it initiates reactions leading

-------
300    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


to expression of toxicity  (Poland and Knutson, 1982; Safe, 1986; Nebert et al.
1991; Birnbaum, 1994).  Further discussion of the demonstrated use of TEFs for
these congeners is provided in the preamble for the proposed Guidance  (58 FR
20942).

      EPA also carefully considered but does not agree with comments that it
is inappropriate to assume that the effects from the different identified
congeners are-additive.  As with most chemicals, there is limited data on the
effects of interactions of these congeners.  However, there is ample evidence
indicating that there  is a common mechanism of action for the expression of
toxicity for the 17 dioxin congeners for which TEFs have been developed.  As
discussed above and in the 1986 Chemical Mixture Guidelines, in  situations
where there is a lack of data on the effects of interactions but it can be
shown that the chemicals act through the same mechanism of action, then the
use of dose addition is a reasonable and scientifically supported mechanism to
address potential adverse impacts from multiple pollutants.  EPA also
maintains that it is appropriate to use TEFs for estimating the  dioxin
equivalent concentrations for both cancer and non-cancer effects.  The
mechanistic basis for the development of either cancer or non-cancer effects
from 2,3,7,8-TCDD and  its associated congeners appears to be the same  (Ah
receptor mediated).•  As discussed above, there are a common series of
biological steps necessary for most, if not all, of the observed effects of
dioxin and related compounds.

      Finally, EPA has determined that future State or Tribal program
submissions or EPA promulgations implementing the final Guidance should also
include the use of BEFs for the 17 CDD/CDF congeners in the final Guidance
when a TEF for that congener is used.  This decision is supported by
scientific studies demonstrating that CDDs/CDFs, other than 2,3,7,8-TCDD, have
different and generally smaller bioaccumulation factors.  EPA believes it is
appropriate to use factors accounting for the different BAFs in  converting
concentrations of CDDs and CDFs to equivalent concentrations of  2,3,7,8-TCDD.
Great Lakes States and Tribes are not precluded, however, from utilizing the
more conservative BAF  for 2,3,7,8-TCDD for these related congeners as a
simplifying approach,  as described in the preamble to the proposed Guidance
 (58 FR 20942).  EPA acknowledges that the data base for developing BEFs is
limited to data from Lake Ontario, but believes the data is sufficient to
develop BEFs for the Great Lakes System. The BEF approach has been peer-
reviewed as part of the process of review on the "Interim Report on Data and
Methods for Assessment of 2,3,7,8-Tetrachlorodibenzo-p-dioxin Risk to Aquatic
Life and Associated Wildlife"  (March 1993, EPA/600/R-93/055).  The technical
rationale for each of  the BEFs included in the final Guidance is provided in
the "Technical Support Document for the Procedure to Determine Bioaccumulation
Factors," which is available in the public docket for this rulemaking.

      d.    Response to Application of TEFs to Wildlife:  EPA agrees that the
data supporting TEFs for the 17 dioxins/furans and for the PCBs  is not
sufficiently developed at this time to require States and Tribes to adopt
provisions mandating their use for development of wildlife criteria.  This
judgement is supported by a report on the March 19-20, 1992, Dioxin Ecotox
Subcommittee of the Ecological Processes and Effects Committee of the Science
Advisory Board which met to review EPA's research proposals to support the
development of an ambient aquatic life water quality criterion for 2,3,7,8-
TCDD  (USEPA, August 1992 at p.11, EPA-SAB-EPEC-92-024).  At that meeting, the
Subcommittee addressed the general issue of research needed to support the use
of TEFs for aquatic life and wildlife.  In  their final report, the Committee
stated that the TEF approach appears promising for aquatic life  and wildlife
but more studies are needed to show phylogenetic variability.  The Committee
concluded that at the  present time there are  insufficient data available to
judge the reliability  and  the accuracy of the TEF approach for protection of
aquatic life and wildlife.

       Additionally,  it  is difficult  to evaluate whether the TEFs developed
for protection of human health are also appropriate without modification for

-------
                              Section VELD:  Additivity                          301
the protection of wildlife.  Although the mechanism of action of CDDs and CDFs
for wildlife may be similar to the mechanism for humans, the endpoints of
concern for wildlife may be different than those for human health.  Because of
this uncertainty and the concerns raised by the SAB, EPA believes it is not
appropriate at this time to require the use of TEFs for wildlife until further
research can be conducted on these issues.

      Research is currently underway to develop and evaluate TEFs for "dioxin-
like" PCBs, (Safe, 1990; Walker and Peterson, 1991; Devito et al . , 1993;
Ahlborg et al . 1994) .  A set of established TEFs for these PCBs similar to the
17 TEFs for CDDs and CDFs, however, has not been adopted by EPA.  (See,
"Estimating Exposure to Dioxin-Like Compounds"  (External Review Draft, June
1994, EPA/600/6 -88/005Ca) .  EPA is continuing research in this area and
expects to develop a set of TEFs for "dioxin-like" PCBs in the near future
that could be used for risk assessment purposes.  Until the science is more
fully developed in this area, EPA has determined that it is appropriate to
include TEFs in the final Guidance for only the 17 established CDDs/CDFs for
protection of human health.  This does not preclude States or Tribes, however,
from developing TEFs for protection of wildlife for CDDs/CDFs or for the
dioxin-like PCBs based on any available supporting scientific data.

      e.    Final Guidance :  For the reasons stated above, EPA has decided to
limit the use of TEFs to the protection of human health and to only the 17
CDDs/CDFs included in the proposed Guidance.  In addition, the final Guidance
allows the use of the BEFs for the 17 CDDs/CDFs with TEFs.  The TEFs in Table
1 and BEFs in Table 2 must be used when calculating a 2,3,7,8-TCDD toxicity
equivalence concentration in effluent to be used when implementing both human
health noncancer and cancer criteria.  The chemical concentration of each CDDs
and CDFs in the effluent shall be converted to a 2,3,7,8-TCDD toxicity
equivalence concentration in effluent by  (a) multiplying the chemical
concentration of each CDDs and CDFs in the effluent by the appropriate TEF in
Table 1 below, (b) multiplying each product from step (a) by the BEF for each
CDDs and CDFs in Table 2 below, and (c) adding all final products from step
(b) .  The equation for calculating the 2,3,7,8-TCDD toxicity equivalence
concentration in effluent is:

                         ( TEC) tcdd=L ( C) x ( TEF) x (BEF) x


where :
       td     =     2,3,7,8-TCDD toxicity equivalence concentration in effluent

  (C),       =     concentration of total chemical x in effluent


 (TEF),      =     TCDD toxicity equivalency factor for x

 (BEF)^      =     TCDD bioaccumulation equivalency factor for x

       An equation specifying how to estimate the 2,3,7,8-TCDD toxicity
equivalence concentration in effluent was added to procedure 4 of appendix F
to part 132 to assist permitting authorities.  The 2,3,7,8-TCDD toxicity
equivalence concentration in effluent must be used when developing waste load
allocations under procedure 3 of appendix F to part 132, preliminary waste
load allocations for purposes of determining reasonable potential under
procedure 5 of appendix F to part 132, and for purposes of establishing
effluent quality limits under procedure 5 of appendix F to part 132.

      8.     Wildlife

      a.     Proposal :   Both approaches in the proposed Guidance presented
additivity provisions for wildlife which were very similar to those proposed
for non- cancer human health effects.  No additivity provisions were presented

-------
302    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

for cancer effects because cancer  is  not  a recognized endpoint of concern for
wildlife.

      b.    Comments:  As discussed above, several  commenters supported the
application of the additivity provisions  to  wildlife,  while others opposed the
application.

      c.    Final Guidance:  As discussed above,  the final Guidance does not
include provisions addressing any  possible additive effects of pollutant
mixtures on wildlife.  However, as discussed in section II.G (Implementation
of the Endangered Species Act) of  this document,  EPA,  in cooperation with the
Fish and Wildlife Service  (FWS) will  host a  workshop on the subject of TEFs
for those PCDDs, PCDFs, and  PCBs that have been identified as exhibiting
toxicity similar to 2,3,7,8-TCDD to wildlife.   The  workshop will examine
existing toxicity data, as it relates to  TEFs  and research and data needs, and
the use of TEFs when establishing  total maximum daily loads (TMDLs) or WQBELs.
The findings of these workshops will  be used to evaluate the feasibility of
utilizing TEFs in the development  of  wildlife  criteria.  Any methodologies
developed by EPA as a result of these efforts  will  be submitted to the EPA SAB
for review and distributed for public comments.

-------
                              Section Vffl.D: Additivity                          303
                                    Table 1
Toxic Equivalency Factor Values for CDDs and CDPs

Congener                              TEF
2,3,7,8-TCDD                        1.0
1,2,3,7,8-PeCDD                     0.5
1,2,3,4,7,8-HxCDD                   0.1
1,2,3,6,7,8-HxCDD                   0.1
1,2,3,7,8,9-HxCDD                   0.1
1,2,3,4,6,7,8-HpCDD                 0.01
OCDD                                0.001
2,3,7,8-TCDF                        0.1
1,2,3,7,8-PeCDF    ,                 0.05
2,3,4,7,8-PeCDF    '                 0.5
1,2,3,4,7,8-HxCDF                   0.1
1,2,3,6,7,8-HxCDF                   0.1
2,3,4,6,7,8-HxCDF                   0.1
1,2,3,7,8,9-HxCDF                   0.1
1,2,3,4,6,7,8-HpCDF                 0.01
1,2,3,4,7,8,9-HpCDF                 0.01
OCDF                                0.001

                                    Table 2
             Bioaccumulation Equivalency Factors for CDDs and CDFs

Congener                             BEF

2,3,7,8-TCDD                        1.0
1,2,3,7,8-PeCDD                     0.9
1,2,3,4,7,8-HxCDD                   0.3
1,2,3,6,7,8-HxCDD                   0.1
1,2,3,7,8,9-HxCDD                   0.1
1,2,3,4,6,7,8-HpCDD                 0.05
OCDD                                0.01
2,3,7,8-TCDF       '                 0.8
1,2,3,7,8-PeCDF                     0.2
2,3,4,7,8-PeCDF                     1.6
1,2,3,4,7,8-HxCDF                   0.08
1,2,3,6,7,8-HxCDF                   0.2
2,3,4,6,7,8-HxCDF                   0.7
1,2,3,7,8,9-HxCDF                   0.6
1,2,3,4,6,7,8-HpCDF                 0.01
1,2,3,4,7,8,9-HpCDF                 0.4
OCDF                                0.02

-------
304    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

9.     References

      Ahlborg et al.  1994.  Toxic equivalency factors for dioxin-Like  PCBs.
Chemosphere.  28 (6):1049-1067.

      Bandiera et al., 1984.  Polychloronated dibenzofurans  (PCDFs): effects
of structure on binding to the 2,3,7,8-TCDD cytosolic receptor protein, AHH
induction and toxicity.  Toxicology 32:131-144.

      Birnbaum, L.  1994.  Evidence for the role of Ah receptor  in  responses
to dioxin.  In: Spitzer, H.L.; Slaga, T.J.; Greenlee, W.F.; Mclain, M.  eds.
Receptor-mediated biological processes: implications for evaluating
carcinogenesis.  Progess in Clinical and Biological Research, vol.  387. New
York, NY: Wiley-Liss, pp. 139-154.

      Devito et al.  1993.  Comparative ability of various PCBs, PCDFs, and
TCDD to induce cytc-chrome P450 1A1 and 1A2 activity following 4 weeks of
treatment.  Fundam. Appl. Toxicol. 20:125-130.

      NATO/CCMS. 1988a.  North Atlantic Treaty Organization, Committee  on  the
Challenges of Modern Society.  International toxicity equivalency factor  (I-
TEF) method of risk assessment for complex mixtures for dioxins and related
compounds.  Report No. 176.

      NATO/CCMS. 1988b.  North Atlantic Treaty Organization, Committee  on  the
Challenges of Modern Society.  Scientific basis for the development of
international toxicity equivalency factor  (I-TEF) method of risk assessment
for complex mixtures for dioxins and related compounds.  Report No. 178.

      National Research Council.  1988.  Complex Mixtures: Methods  for  In  Vivo
Toxicity Testing.  Committee on Methods for the In Vivo Toxicity Testing of
Complex Mixtures. Board on Environmental Studies and Toxicology.  Commission
on Life Sciences.

      Nebert et al.• 1991.  Human Ah locus polymorphism and cancer:
inducibility of CYPIA1 and other genes by combustion products and dioxin.
Pharmacogenetics.  1:68-78.

      Olson et al. 1989.  Reexamination of data used for establishing toxicity
equivalency factors  (TEFs) for chlorinated dibenzo-p-dioxins and dibenzofurans
 (CDDs and CDFs).  Chemosphere.  18(1-6):371-381.

      Poland, A and J.C. Knutson.  1982.  2,3,7,8-Tetrachlorodibenzo-p-dioxin
and related aromatic hydrocarbons: examination of the mechanism  for toxicity.
Annu. Rev. Pharmacol.  Toxicol.  22:517-554.

      Safe, S.H.   1986.  Comparative toxicology and mechanisms of action of
polychlorinated dibenzo-p-dioxins and dibenzofurans.  Annu. Rev. Pharamacol.
Toxicol.  26:371-398.

      Safe S.  1990.  Polychlorinated biphenyls  (PCBs), dibenzo-p-dioxins
 (PCDDs), dibenzofurans  (PCDFs) and related compounds: Environmental and
mechanistic considerations which support the development of toxic equivalency
factors.  CRC Crit. Rev. Toxicol.  21:51-88.

      USEPA. 1986. Guidelines for the Health Risk Assessment of  Chemical
Mixtures. 51 FR  34014.

      USEPA.  1989.  Interim procedures for estimating risks associated with
exposures to mixtures of chlorinated dibenzo-p-dioxins and  -dibenzofurans  and
1989 update.  EPA/625/3-89/016. Risk Assessment Forum.

-------
                              Section VTH.D:  Additivity                          305
      USEPA. 1989. Risk Assessment Guidance for Superfund Volume 1 Human
Health Evaluation Manual  (Part A). EPA/540/1-89/002  (8-6 to 8-15}. PB90-
155581.

      USEPA. 1990. Technical Support Document on Risk Assessment of Chemical
Mixtures. EPA/600/8-90/064.

      USEPA.  1992.  SAB Report: Review of Rational for Development of Ambient
Aquatic Life Water Quality Criteria for TCDD  (Dioxin). Dioxin Ecotox
Subcommittee of the Ecological Processes and Effects Committee. EPA-SAB-EPEC-
92-024.

      USEPA. 1993. Interim Report on Data and Methods for Assessment of
2,3,7,8-Tetrachlorodibenzo-p-dioxin Risk to Aquatic Life and Associated
Wildlife. EPA/600/R.-93/055.

      USEPA.  1994.  Estimating Exposure to Dioxin-Like Compounds. Volum 1:
Executive Summary. External Review Draft.  EPA/600/6-88/005Ca.

      Walker M.K. and R.E. Peterson.  1991.  Potencies of polychlorinated
dibenzo-p-dioxin, dibenzofuran, and bipheyl congeners, relative to 2,3,7,8-
tetrachlorodibenzo-p-dioxin, for producing early life stage mortality in
rainbow trout (Oncorhynchus mykiss). Aq Toxicol. 21:219-238.

-------
                           Section Vm.E: Reasonable Potential                       307
E.    Reasonable Potential for Exceeding Numeric Water Quality Standards

      The purpose of this section is to define the procedures for determining
whether an NPDES permit for discharges to the Great Lakes System must include
a water quality-based effluent limitation for a particular pollutant parameter
(not including whole effluent toxicity).  Considerations related to whole
effluent toxicity and the basis for such considerations are addressed
separately in section G of this document.  This final Guidance requires
permitting authorities to follow specific procedures where facility-specific
effluent monitoring data are available.  Where these data are not available,
including when all available effluent data for a pollutant or pollutant
parameter are below the applicable analytical detection level, this Guidance
does not establish any new or specific requirements, and permitting
authorities will continue to follow existing Federal, State or Tribal
regulations and guidance.  Existing guidance for determination of reasonable
potential in the absence of facility-specific effluent monitoring data are
discussed in section VIII.E.I of this document, below.

1.    Existing National Rules and Guidance

      EPA's existing regulations require NPDES permits to contain effluent
limitations necessary to meet applicable technology-based requirements of
Federal and State law.  These technology-based limitations are derived
directly from application of National effluent limitation guidelines or on the
basis of the permitting authority's best professional judgment (40 CFR 125.3).
States are currently required to adopt regulations consistent with these
provisions as part 'of their approved NPDES State permitting program  (40 CFR
123.25(a) (36)).  EPA is not, in this final Guidance, addressing the
requirements governing the establishment of technology-based limitations.

      In addition to these technology-based requirements, EPA's existing
regulations require NPDES permits to include water quality-based effluent
limitations (WQBELs) to control all pollutants or pollutant parameters which
the permitting authority determines are or may be discharged at a level which
will cause, have the reasonable potential to cause, or contribute to an
excursion above any water quality standard, including numeric and narrative
criteria for water .quality  (40 CFR 122.44(d) (1)) .  When determining whether a
discharge  causes, has the reasonable potential to cause, or contributes to an
excursion above any State or Tribal water quality standard, the permitting
authority must use all relevant available data, including facility-specific
effluent monitoring data where available.  Additionally, the permitting
authority must use procedures which account for existing controls on point and
nonpoint sources of pollution; variability of the pollutant or pollutant
parameter in the effluent; and, where appropriate, the dilution of the
effluent in the receiving water (40 CFR 122.44(d)(1)(ii)).  If the permitting
authority determines that a discharge has the reasonable potential to cause or
contribute to an excursion of an applicable numeric or narrative water quality
criterion,  it must 'include a WQBEL for the individual pollutant in the permit
(40 CFR 122.44(d)(1)(iii)).  In the absence of a numeric water quality
criterion for an individual pollutant under these circumstances,  the
permitting authority must derive appropriate WQBELs from the State or Tribal
narrative water quality criterion by: using a calculated numeric criterion for
the pollutant that attains the applicable narrative criterion and protects
designated uses; establishing effluent limitations on a case-by-case basis
using EPA's water quality criteria developed under section 304(a) of the Clean
Water Act,  supplemented with other information where necessary; or
establishing effluent limitations on an indicator pollutant (40 CFR
122.44(d) (1) (vi)). ,

      EPA has provided guidance on how to apply these requirements in the
"Technical Support Document for Water Quality-based Toxics Control (TSD)"
(EPA/505/2-90-001,  March 1991), which is available in the administrative
record for this rulemaking.  In the TSD,  EPA recommends that facility-specific
effluent monitoring data be used,  where available, to project receiving water

-------
308    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


concentrations, which are then compared to water quality criteria.  This
comparison in the TSD guidance is comprised first of calculating  the pollutant
concentration in trie receiving water after considering dilution  (if allowed by
the water quality standards regulation), the contributions of other point and
nonpoint sources, and the potential for effluent variability to justify
assuming higher effluent concentrations than have actually been measured; and
second, comparing this calculation to the applicable water quality criterion.
The TSD guidance allows the permitting authority the flexibility  to determine
the appropriate approach for assessing reasonable potential.  For example, an
authority may opt to use a stochastic dilution model that incorporates  both
ambient dilution and effluent variability rather than use a steady state
dilution model with a statistically defined maximum effluent concentration.
Also, a permitting -authority may develop a WQBEL in the absence of facility-
specific effluent monitoring data.  Whatever approach is selected by the
authority, it must satisfy all requirements of 40 CFR 122.44(d)(1)(ii)
summarized above.

      One of four outcomes will be reached when using the TSD protocol:

      -- Excursion Above the Water Quality Standard.  If the permitting
authority determines that pollutants or pollutant parameters in a. facility's
discharge are or may be discharged at a level which causes or contributes to
an excursion above a narrative or numeric water quality criterion, it must
establish a WQBEL in the permit for those pollutants  (40 CFR 122.44(d)(1)(i).

      -- Reasonable Potential for Excursion Above the Water Quality Standard.
If the permitting authority determines that pollutants or pollutant parameters
in a facility's discharge are or may be discharged at a level which has the
reasonable potential to cause or contribute to an excursion above a narrative
or numeric water quality criterion, it must establish a WQBEL in  the permit
for that pollutant   (40 CFR 122.44(d)(1)(i)).  EPA believes that  reasonable
potential is shown where an effluent, in conjunction with other sources of a
pollutant, is projected to cause an excursion above the water quality
criterion.  This projection is based upon an analysis of available data that
accounts for limited sample size and effluent variability.   EPA's guidance in
the TSD does not, however, constrain the determination of reasonable potential
to a projection of an excursion above a water quality criterion based solely
on effluent variability.  The guidance recognizes that reasonable potential
determinations include consideration of the factors in 40 CFR
§122.44(d)(l) (ii) and any other appropriate factors based on the  professional
judgement of the permitting authority.  These other factors may include the
existing data on toxic pollutants; type of receiving water and designated uses
(e.g., high-use fishery); relative proximity of the measured effluent
concentrations to the water quality criteria; existing controls on point and
nonpoint sources; compliance history of the facility; and type of treatment
facility.

      -- No Reasonable Potential for Excursions Above the Water Quality
Standards.  If the permitting authority determines that the pollutants  or
pollutant parameters in a facility's discharge are not discharged at a  level
that has the reasonable potential to cause or contribute to an excursion above
a narrative or numeric water quality criterion, then a WQBEL for  those
pollutants is not necessary.  In these situations, EPA's guidance recommends
that effluent monitoring for the pollutants or pollutant parameters be
repeated at a frequency of at least once every five years  (see TSD at p. 64).
This usually occurs as part of the permit application.

       -- Inadequate Information.   If a permitting authority has  inadequate
information to determine whether a discharge contains pollutants  or pollutant
parameters which are or may be discharged at a level which has  the reasonable
potential to cause or contribute to an excursion of a narrative  or numeric
water quality criterion, EPA's existing guidance recommends  that  the permit
contain appropriate monitoring requirements and a reopener clause (see  TSD at
p. 64).  This clause would allow reopening of the permit and establishment of

-------
                           Section Vffl.E: Reasonable Potential                        309
a WQBEL based upon 'any monitoring results  or  other new  factors  which indicate
that the effluent causes, has  the reasonable  potential  to  cause,  or
contributes  to an excursion above water  quality  standards.

2.    General Requirements of  Procedure  5

      Procedure 5 of this Guidance provides that the permitting authority
would be required to include a WQBEL in  an NPDES permit whenever a pollutant
is or may be discharged into the Great Lakes  System at  a level  which will
cause, have  the reasonable potential to  cause, or contribute  to an excursion
above any Tier I criterion or  Tier II value.  Procedure 5  of  appendix F  to
part 132 sets forth a two-step process for determining  whether  the discharge
of a pollutant will cause, have the reasonable potential to cause, or
contribute to an excursion above any Tier  I criterion or Tier II value.

      First, under procedure 5.A of appendix  F,  permitting authorities must
develop preliminary effluent limitations (PEL) that are calculated such  that
the discharger, if meeting the PEL, would  not cause or  contribute to any Tier
I criterion  or Tier II value being exceeded in-stream after discharge, where
there is sufficient data to develop such criteria or values.  If such data do
not exist, permitting authorities must apply  the provisions in  procedure 5.C
of appendix  F to determine whether such  data  must be generated.   Second,
procedure 5.B of appendix F sets forth procedures to be followed to  determine
the projected effluent quality (PEQ) of  the effluent that  will  be discharged
and whether  a WQBEL must be established  based on a comparison between the PEQ
and the preliminary effluent limitation.   If  such effluent data do not exist
or if all such effluent data for a pollutant  or  pollutant  parameter  are  below
the analytical detection level for that  pollutant or pollutant  parameter,
permitting authorities will continue to  apply existing  Federal,  State or
Tribal regulations and guidance for making reasonable potential
determinations.  Finally, section F of procedure 5 provides that, regardless
of the manner in which the reasonable potential  determination is made, all
effluent limitation's would have to also  comply with all other applicable
State, Tribal and Federal requirements.  In particular, section F specifies
that all water quality-based effluent limitations must  be  consistent with the
wasteload allocations calculated as part of a TMDL, or  in  the absence of a
TMDL, consistent with wasteload allocations generated consistent with the
provisions specified in the final Guidance and described in more detail  below.

      This Guidance provides permitting  authorities with specific criteria for
making reasonable potential determinations based on facility-specific effluent
monitoring data consistent with the provisions of 40 CFR 122.44(d)(1)(i),
(ii), (iii), and (vi).  The procedures of  this section  do  not contain
requirements pertaining to whole effluent  toxicity.  Instead, those
requirements are set forth in  procedure  6  of  appendix F.   Furthermore, the
procedures of this section do  not affect the  permitting authorities'  existing
obligation to implement the regulations  at 40 CFR 122.44(d)(1)(vii)  which
pertain to expression of WQBELs.

      As a. preliminary matter, before explaining the specific aspects of
procedure 5, discussion of two general comments  and EPA's  responses  is
necessary.

      The first paragraph of procedure 5,  appendix F of the proposed guidance
contains the following sentence as the last sentence of the paragraph, "In all
cases, the permitting authority shall use  any relevant  information that
indicates a reasonable potential to exceed any Tier I criterion or Tier  II
value."  Some commenters suggested that  the meaning of  the word "relevant" in
this statement is unclear and  that this  statement could be read as implying
that relevant data indicating  a reasonable potential to exceed  a criteria or
value is to be used,  but that  data indicating an absence of reasonable
potential should not be used.

-------
310    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      In response, in the cited statement, EPA contemplated that a broad  scope
of data and information could be relevant in determining reasonable potential
for a given discharge, including: effluent pollutant concentration data,
receiving water background data, ambient fish tissue data, Tier I criteria,
Tier II values and in addition, a variety of factors and information when
determining reasonable potential where facility-specific effluent monitoring
data are unavailable.  The variety of factors and information can include:
dilution, type of industry, type of POTW, existing data on toxic pollutants,
history of compliance problems and toxic impact, and type of receiving  water
and designated use'(See TSD at 50-55.) .  Any of the above could be "relevant"
within the meaning of the final Guidance and, of course the permitting
authority must exercise its judgement in determining the relevance of any type
of data or information to the discharge and receiving water in question.   EPA
regulations in some cases define the parameters for the exercise of this
discretion.  For example, all representative effluent pollutant concentration
data--not just data points that alone would trigger the need for a WQBEL--(see
40 CFR 122.41 (j) (1)) for a particular discharger are clearly relevant to  that
discharger and must, under final procedure 5 be used in the determination of
reasonable potential.  In addition, to clarify the implicit premise in
proposed and final .procedure 5 that only valid data should be used when
determining reasonable potential, EPA inserted the word "valid" into the  cited
statement immediately prior to the word relevant.  EPA's intent is for  the
permitting authority to exercise its judgement in determining the validity of
data and information and their relevance to the discharge and receiving water
in question.

      Paragraph F.2 of proposed procedure 5 stated that "When determining
whether water quality-based effluent limitations are necessary, information
from chemical-specific, whole effluent toxicity and biological assessments
shall be considered independently."  Several commenters suggested that  this
provision should not be adopted, stating that the policy and philosophy of
independent application is flawed, not scientifically valid, and should be
replaced by a "weight of evidence" approach to determining the need for
WQBELs.

      In response, EPA believes that the provision at F.2 of proposed
procedure 5 and F.3 of the final procedure 5 requiring independent
consideration of chemical-specific, whole effluent toxicity and biological
assessments is grounded in the requirements of 40 C.F.R. § 122.44(d)(1),  which
requires, among other things, that the permitting authority  (1) establish
chemical-specific permit limits where a discharge has the reasonable potential
to cause or contribute to a violation of a numeric water quality criterion,
and  (2) establish a whole effluent toxicity limit where a discharge has the
reasonable potential to cause or contribute to an exceedance of a numeric WET
criterion.  Under these provisions, if "reasonable potential" is found  with
regard either of these aspects of standards, then a corresponding permit  limit
is required.  There is no indication  in the language of this provision  that
one type of information  (e.g., biological assessment or WET testing) can  be
used to  "negate" a reasonable potential finding based on another type of
information  (e.g., chemical  specific  analysis).  One principle behind the
policy on independent application as  it pertains to determining  the need  for
WQBELs is that WET -testing does not always measure all potential toxicity in
complex mixtures and, likewise, chemical analysis does not always measure
toxicity from single components of the effluent.  Hence, it is necessary  to do
both kinds of analysis on effluents and consider the results independently.
The regulations  do permit, however, the use of chemical-specific limits in
lieu of WET limits in cases  where there a discharge has the reasonable
potential to  cause or contribute  to a violation of a narrative water quality
criterion, provided  it is demonstrated that chemical-specific limits are
sufficient  to attain and maintain applicable narrative and numeric water
quality  standards.

-------
                           Section Vm.E: Reasonable Potential                       311
a.    Developing Preliminary Wasteload Allocations

      Procedure 5.A.I of this Guidance describes how the permitting authority
must establish preliminary wasteload allocations for purposes of calculating
preliminary effluent limitations.  Preliminary wasteload allocations are
solely for the purpose of determining preliminary effluent limitations that
are then compared to projected effluent quality to determine reasonable
potential.  The preliminary wasteload allocation is an essential component of
the reasonable potential analysis and serves no other purpose.

      i.    Proposal:  Procedure 5.A of appendix F of the proposed Guidance
provided that the permitting authority would develop preliminary wasteload
allocations based upon and consistent with the wasteload allocation procedures
defined in procedure 3 of appendix F of the proposed Guidance, and then
develop preliminary effluent limitations based on the preliminary wasteload
allocations.  The proposed guidance did not elaborate on or cross reference
the specific wasteload allocation procedures in procedure 3 of appendix F to
be used in calculating preliminary wasteload allocations.

      ii.   Comments:  EPA received numerous comments suggesting that the
procedures to be used for calculating wasteload allocations (in the absence of
a TMDL approved by EPA under 130.7) and preliminary wasteload allocations
needed to be clarified.   The proposal provided that preliminary wasteload
allocations would be calculated in accordance with procedure 3, the TMDL
procedure.  Some commenters read this to mean that a TMDL would have to be
prepared prior to determining reasonable potential, and hence, prior to
issuance of any new or revised permit.  That was not EPA's intent.  Instead,
EPA's intent in the proposal was that the same minimum procedures that would
be required under procedure 3 for calculating source-specific TMDLs would be
used for calculating preliminary wasteload allocations for purposes of
determining reasonable potential, and wasteload allocations for purposes of
calculating permit limits.  The proposal did not acknowledge that reasonable
potential determinations often are made, and WQBELs required,  in the absence
of a TMDL approved by EPA under 40 CFR 130.7.  Instead, the proposal suggested
that TMDLs would be required for any water receiving effluent from a
discharger found to exhibit reasonable potential.  This is not the case in the
final Guidance.  As noted in the final procedure 3 and the accompanying
discussion at section VIII.C of this document, the final Guidance specifies
the conditions under which a TMDL is required for a waterbody.  The final
Guidance does not, 'like the proposal would have, require a TMDL for every
water receiving effluent from a discharger found to exhibit reasonable
potential.  EPA recognizes that, given this change, reasonable potential
decisions often will be made, and WQBELs required, in the absence of a TMDL
approved by EPA under 40 CFR 130.7.  This is currently the case in the
national NPDES program.

      iii.  Final Guidance:  Section 5.A of appendix F of the final Guidance
clarifies the specific provisions within procedure 3 that would be the basis
for determining preliminary wasteload allocations in the absence of a TMDL
approved by EPA under 40 CFR 130.7.  Section 5.F of appendix F of the final
Guidance also specifies that  in the absence of a TMDL approved under part
130.7,  the permitting authority must use procedures consistent with the
referenced provisions of procedure 3 as the basis for determining wasteload
allocations for the purpose of deriving WQBELs when the permitting authority
determines under this procedure that a WQBEL must be included in a NPDES
permit.   These specific provisions within procedure 3 that the final Guidance
requires permitting authorities to use as the basis for determining
preliminary wasteload allocations, and wasteload allocations  (in the absence
of a TMDL approved by EPA under 40 CFR 103.7) are: 3.B.9, Background
Concentrations of Pollutants; 3.C, Mixing Zones for Bioaccumulative Chemicals
of Concern, sections C.I.a, C.3, C.4.a, C.5, C.6; 3.D, Deriving TMDLs for
Discharges to Lakes when the receiving water is an open water of the Great
Lakes,  an inland lake or other water of the Great Lakes System with no
appreciable flow relative to its volume; 3.E, Deriving TMDLs,  WLAs and

-------
312    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

Preliminary WLAs, and LAs for Discharges to Great Lakes System Tributaries
when the receiving water is a tributary or connecting channel of the Great
Lakes that exhibits appreciable flow  relative to its volume; and, 3.F, Mixing
Zone Demonstration Requirements.

      EPA selected -the portions of procedure 3 noted above and cross-
referenced in 5.A and 5.F of appendix F of the final Guidance  (cross-
referenced provisions) because they closely parallel the core factors EPA has
consistently required be accounted for, pursuant to 40 CFR 122.44(d)(1)(i) and
(ii), when determining reasonable potential and wasteload allocations for
deriving WQBELs in the absence of a TMDL.  Specifically, when determining when
a WQBEL is necessary and in setting the limit, these regulations require that
existing controls on point and nonpoint sources and dilution where
appropriate, be accounted for.  EPA has interpreted these regulatory
requirements through preamble language  (54 FR 23868, June 2, 1989) and through
various pieces of Agency guidance  (see TSD in particular).  EPA has
consistently required that, to meet the requirements of 122.44(d)(1)(i) and
(ii), permitting authorities must account for ambient background pollutant
concentrations in receiving waters and the available dilution in a receiving
water that results in mixing of effluent and receiving water.  The cross-
referenced provisions of procedure 3, which address how to evaluate background
concentrations, and available dilution and mixing are therefore now
specifically cross-referenced in the  reasonable potential section.

      It is important to note that the cross-referenced provisions, serve
essentially three purposes in the final Guidance.  First, outside of the
context of developing a TMDL under §  130.7, or in the absence of such TMDL,
wasteload allocations for purposes of calculating WQBELs must be determined
consistent with the cross-referenced  procedures.  Second, for purposes of
conducting a reasonable potential determination, preliminary wasteload
allocations must be determined  consistent with the cross-referenced
provisions.  And of course, third, when a TMDL under 130.7 is being developed,
all provisions,  including the cross-referenced provisions of procedure 3,
apply.

b.    Developing Preliminary Effluent Limitations

      Procedure  5.A.2 of appendix F of the final Guidance specifies the
procedure for developing preliminary  effluent limitations based  on the
preliminary wasteload allocations.

      i.    Proposal:   Procedure 5.A.2 of appendix F of the proposed guidance
set out the minimum requirements for  developing preliminary effluent
limitations and  specified that  the preliminary effluent limits are required  to
be developed consistent with these minimum requirements and in accordance with
existing State or Tribal procedures for converting wasteload allocations  into
water quality-based effluent limitations.  Procedure 5.A.2 of appendix  F  of
the proposal provided that the  preliminary effluent limitations  would be
expressed as either a single day value, a weekly average, or a monthly
average, and would be used in determining if a facility causes,  has  the
reasonable potential  to cause or  contributes to excursions above water  quality
criteria by being compared to actual  effluent information in procedure  5.B of
appendix F  of the proposed Guidance.  Because the preliminary effluent
limitations must be compared with  actual effluent information, the proposed
Guidance provided  that  the preliminary  effluent limitations would be  expressed
in  the  same form that effluent  data are typically available to permitting
authorities.  Effluent  information is typically available to permitting
authorities either  in the permit  application or in the Discharge Monitoring
Reports  (DMR).   Both  the application  forms  and DMRs require effluent
concentrations  to be  reported as  weekly and monthly averages for publicly
owned treatment  works  (POTWs) and as  single  day values  and monthly  averages
for  non-POTWs.   The use of these  single day values, weekly  averages,  and
monthly averages allows for  direct comparison of preliminary effluent
limitations to  effluent data without  requiring additional manipulations or

-------
                           Section VHI.E: Reasonable Potential                       313
conversion of the effluent data.  In the preamble of the proposed Guidance,
EPA discussed its belief that this reduces the burden  to the permitting
authorities and facilities in reviewing and using effluent concentration data
in determining if a WQBEL is necessary.

      Each preliminary wasteload allocation has a corresponding preliminary
effluent limitation that is based on the criterion  (or value) and dilution
basis used to develop the wasteload allocation.  The preliminary effluent
limitation based on wildlife criteria was proposed  in  section 5.A.2 as a
monthly average because the wasteload allocation is calculated using a 30-day
(monthly) average flow under proposed procedure 3 of appendix F.  The
preliminary effluent limitation based on human health  criteria was proposed to
be expressed as a monthly average because, although the wasteload allocation
is calculated using a harmonic mean  (annual) river  flow, the monthly averaging
period is the closest expression of the preliminary effluent limitations to an
annual average.  The preliminary effluent limitation based on acute aquatic
life criteria was proposed to be expressed as a daily  value to reflect that
the criteria themselves are expressed as one-hour averages and the wasteload
allocation is calculated using a one-day  (daily) average river flow.  The
preliminary effluent limitation based on chronic aquatic life criteria was
proposed to be expressed as a weekly average value  to  reflect that the
criteria themselves are expressed as four-day averages and the wasteload
allocation is calculated using a seven-day  (weekly) average river flow.  In
addition, the preliminary effluent limitation based on chronic aquatic life
criteria could, as ,an option under the proposal, be expressed as a monthly
average value to reflect that weekly average effluent  data may not be
available for non-POTW facilities.

      The proposal also explained that dilution, nonpoint sources of
pollution, and the potential for effluent variability  (all factors required by
40 CFR 122.44(d)(1)(ii) to be considered when determining whether a point
source must have a WQBEL(s) in its permit) would be adequately addressed and
accounted for under proposed procedures 5.A.I and 2, and 5.B.

      ii.   Comments:  EPA received few comments on the PEL procedures
proposed at 5.A.2 .of appendix F.  A few commenters suggested that the PEL
procedures specify that the PEL for chronic aquatic life protection be
specified as a monthly average only and that permitting authorities not be
given the option to express this PEL as either a weekly average or monthly
average.  Commenters noted that such an approach would be more consistent with
the PEQ procedures that specify that distribution of monthly averages be used
when determining PEQ with respect to protection of aquatic life from chronic
effects.  In response to this last comment, and as explained in the preamble
to the proposal and reiterated above, effluent data for a discharger will not
always be readily available to the permitting authority in both the weekly and
monthly average format, and for purposes of screening  for the need for WQBELs,
a comparison of either a weekly or monthly average to  a PEQ derived from a
distribution of monthly averages is acceptable.  Because both formats, weekly
and monthly average,  reflect averages of daily measurements, EPA does not
expect there to be substantial differences between the weekly and monthly
average values.  As noted above, dischargers often will have data expressed in
either one format or the other.  In the interest of minimizing the data
collection burden on States,  Tribes and dischargers, EPA believes it is
unnecessary for either new data to be collected to fit a single format or
existing average data to be transposed into the other  format when existing
data is expressed in only one of the two formats.

      One commenter expressed concern that the proposed PEL procedure would
require WQBELs to be derived such that effluent limits based on human health
and wildlife protection would be expressed as monthly  limitations; limits
based on protection of aquatic life from chronic effects would be expressed as
monthly or weekly limitations; and limits based on protection of. aquatic life
from acute effects would be expressed as daily limitations.  The commenter

-------
314    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


pointed out that such an approach to deriving permit limits would depart from
the approach recommended by EPA in the TSD and followed by some  states.

      In response to this comment, EPA notes that the PEL procedures in the
proposal and in the final Guidance do not address how permitting authorities
ultimately set WQBELs--that is left to existing State and Tribal procedures.
The PEL procedure is simply a component of the larger reasonable potential
screening procedure for determining when WQBELs are needed.  Similarly, the
reasonable potential guidance in Chapter 3 of the TSD does not address  how
WQBELs should be calculated--Chapter 5 of the TSD provides guidance on  that
matter.

      It is worth noting for purposes of this discussion that the proposed and
final PEL procedure' refines and expands on the TSD reasonable potential
guidance.  The TSD guidance for determining reasonable potential, in essence,
is a comparison of a projected daily maximum effluent quality, after mixing
with the receiving water, to the applicable water quality criteria.  Under the
TSD guidance, if the projected effluent value, after mixing, exceeds any of
the applicable criteria, reasonable potential exists.  The proposed and final
PEL procedure takes the TSD guidance a step further.  The PEL procedures, in
essence, are a comparison of a PEQ to a PEL.  The PEL essentially is the
applicable water quality criterion adjusted to reflect mixing and, most
importantly here because it is not a feature of the TSD guidance, the duration
of the criterion and the averaging period of the effluent data.  The PEL
procedures specify that effluent data  (PEQ) used for comparison  to the  PEL be
specified in a format that approximates the requisite averaging  period  of the
criterion being protected, and likewise, that the PEL be specified in the same
format.  So for example, where the chronic aquatic life criterion is the
applicable criterion  (exposure duration = 4 days), the PEL and the PEQ  are to
be specified in the format that the data are currently in that best
approximates a 4-day averaging period: weekly or monthly average.  Similarly,
where the acute aquatic life criterion is the applicable criterion  (exposure
duration = 1 hour), the PEL and PEQ are to be specified in the format that
most closely approximates the duration period of the criterion:  daily maximum.
The TSD guidance, While a technically sound screening approach,  is not  this
refined.  It does not provide guidance on specifying the averaging periods of
effluent data after mixing to approximate the exposure duration  of the
applicable water quality criteria when determining reasonable potential.
Instead, the TSD recommends a simple conservative screening procedure that
compares an observed or estimated maximum daily effluent value with the most
stringent applicable water quality criterion.  The PEL procedures in the
proposed and final Guidance simply take the TSD approach a step  further by
incorporating a way to more closely match the averaging period of the effluent
data with the exposure duration of the water quality criteria.

      In response to the comment that the PEL procedures depart  from TSD
guidance on deriving WQBELS, EPA notes that the reasonable potential screening
procedure for deriving PELs differs from the TSD guidance on deriving WQBELs
in the same way that the TSD reasonable potential guidance differs from the
TSD guidance on deriving WQBELs.  This difference is intentional both in the
TSD and in this final Guidance.  The difference exists because the procedure
for determining when a WQBEL is needed and on what basis, and the procedure
for deriving the WQBEL are fundamentally different types of procedures. When
determining when a WQBEL is needed and on what basis, the PEL procedure
specifies that PELs be calculated in different formats  (i.e., daily maximum,
weekly average, and monthly average) and that these values be compared  to PEQ
values expressed in the same formats.  The resulting analysis, while a
screening procedure, does attempt to match effluent averaging periods to the
exposure duration of each criterion that applies in the receiving water.  Once
the analysis is  complete, one will know whether a WQBEL is needed, and  whether
the need for the WQBEL  is based on a projected excursion of acute or chronic
aquatic life criteria, human health criteria or wildlife criteria.  Once the
need for the WQBEL has been established, the next step is to calculate  the
wasteload allocations and generate the actual WQBEL.  While the  proposed and

-------
                           Section Vm.E: Reasonable Potential                       315
final guidance contains minimum requirements for calculating wasteload
allocations, it is 'virtually silent on translating wasteload allocations into
WQBELs.  However, the TSD does contain guidance on this step.  The TSD
guidance recommends, in short, that the permitting authority calculate the
wasteload allocations that will be protective of the various applicable water
quality criteria  (acute and chronic aquatic life, human health and wildlife),
calculate the long term average effluent concentrations that will meet, those
wasteload allocations, select the most stringent long term average effluent
concentration, then translate that long term average into maximum and average
WQBELs that, if met by the discharger, will result in protection of all of the
applicable criteria.

      In summary, the reasonable potential procedures in the final Guidance
are a refined version of the reasonable potential guidance in the TSD, but
nevertheless still screening procedures.  Under these screening procedures,
effluent data in each format is compared to each applicable criterion to
assess the potential for each type of criterion to be exceeded.  In contrast,
the WQBEL guidance in the TSD is more precise.  Under the TSD WQBEL derivation
guidance, once the need for a WQBEL is established, the WQBEL is calculated
such that it results in an average effluent concentration that is low enough
to be protective of each of the applicable criteria and is expressed in both
the maximum and average formats.  EPA repeats that this fundamental difference
between the two procedures, one a screening procedure and the other a more
precise approach to calculating WQBELs, is intentional and appropriate.   Were
EPA to make the PEL procedure entirely consistent with TSD guidance on
deriving WQBELs, and thus a more precise WQBEL derivation procedure, the
screening function of the PEL procedure would be entirely lost.

      iii.  Final Guidance:  Procedure 5.A.2 of appendix F of the proposal
remains unchanged in this final Guidance.  The procedure appearing at 5.A.2 of
appendix F of the proposal has been renumbered in the final Guidance and is
located at 5.A.3.  The procedure in section 5.A.3 of appendix F of the final
Guidance is limited to calculation of preliminary effluent limitations for
purposes of comparison to projected effluent quality to determine the need for
a WQBEL.  Except for the provision in 5.E concerning intake pollutants,
procedures for converting wasteload allocations into WQBELs and for expressing
effluent limitations in NPDES permits shall continue to be governed by
existing State, Tribal and Federal requirements and guidance  (see 40 CFR
122.45(d) and  (e)).

c.    Determining Reasonable Potential to Exceed the Preliminary Effluent
      Limitations Using Pollutant Concentration Data

      Procedure 5.B of appendix F of this final Guidance specifies procedures
for determining the PEQ for discharges to waters of the Great Lakes System
based on facility-specific effluent pollutant concentration data.  Available
effluent monitoring data includes information from discharge monitoring
reports  (DMRs), data from NPDES permit application forms 2A and 2C, and other
data requested of or submitted by the facility or available to the permitting
authority.  Sections 5.B.I and 2 in this final Guidance specify that a
statistical procedure must be used to estimate PEQ.

      Sections 5.B.2.a-c specify the characteristics that the permitting
authority's PEQ procedure must have, which are the same essential
characteristics of'the proposed PEQ procedures.  Sections 5.B.2.a-c provide
that the PEQ is to be specified as: no less than the 95th percentile of the
distribution of the projected population of daily data; no less than the 95th
percentile of the distribution of the projected population of monthly
averages; and no less than the 95th percentile of the distribution of the
projected population of weekly averages.  Sections 5.B.2.a-c do not specify
the exact statistical procedure required, leaving flexibility to the States to
adopt a PEQ procedure that conforms with the characteristics described in
5.B.2.a-c.

-------
316    Water Quality Guidance for the Great Lakes System -- Supplementary Information Document

      Section 5.B.I specifies an alternative method for specifying PEQ (i.e.
the "TSD" procedure).   The TSD procedure is the procedure EPA would promulgate
in a permitting authority's program if the permitting authority  fails to adopt
a procedure consistent with the provisions described 5.B.2.a-c.

      i.    Proposal:   In the proposed rule, the estimated maximum
concentration would have been calculated, in most applications,  as the upper
bound (99th) percentile of the distribution of the projected population of
effluent concentrations.   The 99th percentile was proposed as a  reasonable
measure of the maximum effluent concentration.  In addition, the proposed rule
provided that PEQ would be specified as the maximum observed daily effluent
value where the observed maximum is greater than the statistically estimated
upper bound (99th percentile) value.

      The proposed Guidance also offered a second method, set forth in the
proposal at procedure S.B.l.d of appendix F to part 132, for calculating PEQ.
This second method provided that the PEQ could be calculated as  the upper 95
percent confidence level of the 95th percentile based on a log-normal
distribution of the effluent concentration data.  This statistical procedure
is consistent with the procedure described in section 3.3 of the TSD and with
the general characteristics of the PEQ procedure required at 5.B.I of the
final Guidance.  Each proposed PEQ provision specified that a WQBEL would be
required if the PEQ exceeded any of the preliminary effluent limitations
developed in accordance with section 5.A.

      The proposed guidance also contained a procedure 5.C to address
situations where at least one but less than ten data points exist.  The
proposed statistical procedure at 5.C is essentially the same as the proposed
procedure at S.B.l.d described above, except that, instead of specifying that
the coefficient of variation be calculated, the procedure specified that the
coefficient of variation is assumed to be 0.6.

      Finally, the proposal contained a procedure 5.B.2, described below,
specifying a more conservative approach to comparing projected effluent
quality  (PEQ)  with preliminary effluent limits  (PEL) for use in  low dilution
situations that would have required a comparison of PEQ to only  one-half of
the PEL, instead of the whole PEL.

      ii.   Comments:   EPA received many comments on the reasonable potential
procedures using effluent data.  Numerous commenters expressed support for the
reasonable potential procedures.  Many commenters requested various
simplifications and clarifications of the reasonable potential procedures,
pointing out that there are numerous valid statistical procedures and
assumptions, different and less complicated than those proposed  by EPA, that
could be used to estimate PEQ.  Many commenters suggested that States should
have more latitude in determining reasonable potential and that  best
professional judgement should play a more prominent role in reasonable
potential assessments.  Other commenters suggested that the entire procedure
for estimating PEQ should be guidance rather than required procedures.  Other
commenters expressed strong support for detailed required statistical
procedures for determining reasonable potential, suggesting that such
procedures are desirable to achieve consistency among the Great  Lakes States
in the way that reasonable potential determinations are made.  Several
commenters specifically recommended additional flexibility for determining PEQ
using small data sets, while others suggested that small data sets  (i.e.,
fewer than ten data points) should not be used to make PEQ determinations.  In
contrast, other commenters expressed support for the TSD statistical PEQ
procedure, pointing out that it is an improvement over the current approaches
of some  states with regard to small data sets.

      Several commenters suggested that single effluent data points are
insufficient to trigger the automatic inclusion of WQBEL in a permit, and that
a larger data set should be required.  These commenters noted that a single
data point for an effluent could be non-representative or an outlier.

-------
                           Section Vm.E: Reasonable Potential                       317
Commenters raised the concern that reasonable potential determinations based
on a single data point, using the statistical procedure proposed and
essentially retained in the final Guidance, will result in conservative
projections of effluent quality and unnecessary permit limits.

      In response to these comments, EPA notes that the TSD procedure in 5.B.I
of appendix F'of the final Guidance and the general characteristics at 5.B.2
of appendix F are designed to estimate the 95th percentile effluent value
based upon whatever representative effluent data is available.  When a small
data set is being used to make this projection, or in the extreme case, where
only one data point is being used, the projected effluent quality using the
EPA procedure, will be 6.2 times the observed value  (assuming a coefficient of
variation of 0.6).  EPA as well as many of the commenters on the proposed
guidance recognize this statistical procedure to be valid, even for very small
data sets.  However, EPA also recognizes that the more data there are for a
discharge, the more accurate the effluent pollutant distribution will be and,
generally, the closer the maximum value will be to the projected 95th
percentile value'.  Where dischargers are concerned that the result of the
statistical analysis using a single data point will be too conservative, the
discharger can certainly remedy the situation.  Effluent data in the vast
majority of cases becomes available to the permitting authority via reporting
by the discharger.  In other words, the discharger almost always has the same
effluent data that the permitting authority has.  Where the discharger has
only a single data point, the discharger may always collect and report more
effluent samples to the permitting authority prior to permit issuance or
reissuance.  EPA encourages this practice.  However, where the discharger
reports only a single data point, EPA's position is that such data must not be
ignored.  The final guidance provides flexibility to States to adopt a
reasonable potential statistical procedure that among other attributes,
accounts for and captures long term effluent variability and accounts for
limitations associated with sparse data sets.  Where a State fails to adopt
such a procedure, the final Guidance specifies the statistical procedure EPA
would promulgate for a State should it become necessary (procedure 5.B.I. of
appendix F).  It is essentially the same procedure that was proposed for data
sets of ten or less data points.  The final guidance leaves room for State
procedures to differ from EPA's as long as the basic characteristics outlined
in section 5.B.2 of appendix F are adhered to.  The procedure at 5.B.I is
offered as one alternative, and would only be required, where a State failed
to adopt a PEQ procedure consistent with the characteristics outlined in 5.B.2

      With regard to the comment suggesting that only representative data
should trigger the need for a permit limit, EPA notes that an implicit and
obvious premise in the proposed and final PEQ procedure is that the effluent
pollutant concentration data used to project maximum effluent quality are
valid data that are representative of the effluent.  Permittees should ensure
they are reporting valid, representative data  (see 40 CFR 122.41(j)(l)).
Where the permittee believes certain effluent measurements to not be
representative of the effluent, the permittee should bring this to the
permitting authority's attention.  EPA's position is that valid,
representative effluent data must not be ignored.  To clarify this point in
the final procedure 5,  EPA has inserted the word "representative" into the
first sentence of paragraph B of final procedure 5.  It now reads, "If
representative facility-specific effluent monitoring data samples are
available for a pollutant discharged from a point source to the waters of the
Great Lakes System, the permitting authority shall apply the following
procedures:..."         Another commenter suggested that EPA Guidance should
specify that data obtained prior to and affected by significant treatment,
pretreatment,  or pollution prevention modifications should not be used for
making reasonable potential determinations.  The commenter makes the point
that effluent data used as the basis for characterizing projected effluent
quality should be representative of the discharge and that data obtained prior
to installation of 'treatment, pretreatment or pollution prevention
modifications should not be used.  In response to this comment,  EPA agrees
that effluent data used as the basis for effluent characterization should be

-------
318    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


representative of the discharge under current conditions with current
treatment and management practices at the plant.  The permitting authority
should use judgement in determining whether available effluent data is
representative of the current operating conditions at the facility.  Where
such data is found to be no longer representative of the current discharge,
the permitting authority may choose to not use such data based on a
determination that .the data pre-dates current operating conditions and
treatment at the facility.

      Several commenters suggested that use of the 99th percentile to specify
PEQ would be too extreme and would not be consistent with guidance in the TSD
that recommends use of the 95th percentile.  Commenters also pointed out that
the proposal was unclear about how to calculate the 99th percentile value,
some comments simply reflecting confusion about whether EPA intended in the
proposal for the 99th percentile value to be the 99th percentile of the
observed data points or the 99th percentile of the projected population of
data.

      Several commenters suggested, as an alternative to comparing a
statistically derived "worst case" PEQ to a PEL derived using steady state
worst case assumptions, the final Guidance should recognize dynamic modeling
techniques which provide probability estimates of receiving water
concentrations rather than a single, worst-case condition which rarely  occurs.
Further, the commenters contended that use of these statistical techniques
allows the predicted frequency and duration of exceedances to be directly
compared to the duration and frequency associated with the water quality
criterion, as described in and EPA's TSD.

      Several commenters requested clarification on how to manage effluent
data points that are "non-detect" or  "non-quantified."  Commenters pointed out
that there will be cases where all effluent data points are below detection,-
where some data points are below detection and some are above detection, but
below quantitation; where some data points are above quantitation and some are
below quantitation, and finally, cases where effluent data points fall  into
all three categories.  Commenters specifically noted that the final Guidance
should clarify what values, if any, should be placed on values below detection
or quantitation, and on whether such values should be counted as data points
when determining reasonable potential.

      One commenter questioned the correctness of the statistical procedure
for estimating PEQ at S.B.l.d and 5.C.I  (TSD procedure), stating that the
procedure is fundamentally mathematically incorrect.  In particular, the
commenter states that the two equations presented in the preamble to the
proposal at 58 FR 20949, the first for determining Pn and the second for
determining the ratio of Pn to the 95th percentile of the projected population
of effluent values, contradicted each other.  The commenter also suggests that
the second equation is incorrect in that it is based on a. 99% confidence level
rather than as stated in the proposal, the 95% confidence level.

      Finally, one commenter suggested that the final Guidance should contain
a statistical PEQ procedure that could be applied in all circumstances. EPA
interpreted this comment to mean that a single procedure would simplify the
provisions for determining PEQ and that a single procedure should be available
that would apply to both small and large effluent data sets.

      iii.  Final Guidance:  The final Guidance specifies at 5.B.I, the
essence of the procedure for estimating PEQ proposed at S.B.l.d.  This  is
commonly known as the TSD procedure.  The final guidance also specifies, at
5.B.2, more general characteristics of a PEQ procedure.  The TSD procedure
under 5.B.I is specified as one option for  calculating PEQ.  EPA would
promulgate this option in a permitting authority's program should it become
necessary under 40 CFR 132.5 of this  Guidance.  EPA has concluded that  the
procedure proposed at S.B.l.d, the TSD procedure, and specified  in  this
Guidance at S.B.I, is consistent with the essential characteristics of  a PEQ

-------
                           Section Vin.E: Reasonable Potential                       319
procedure specified in sections 5.B.2.a-c of this Guidance.  In specifying
alternative procedures, one  specific and one more general, the final Guidance
provides the States the flexibility to either select and apply a method for
determining PEQ that is consistent with the more general provisions 5.B.2.a-c
of appendix F, or the more specific TSD procedure at 5.B.I.  EPA believes that
either a State method consistent with 5.B.2.a-c, or the specific method
specified at 5.B.I would satisfy the requirements of 40 CFR 122.44(d)(1)(ii)
because both require valid statistical procedures to characterize effluent
variability in defining a reasonable maximum effluent concentration to
characterize the PEQ.  Alternative methods may result in different conclusions
depending upon the number of data points characterizing an effluent, but
neither of the methods will always provide a more stringent basis for
determining reasonable potential.

      An important change from the proposal is that in the final  Guidance,
proposed procedures S.B.l.d and 5.C (The TSD statistical procedures) have been
merged and are presented in the final guidance under 5.B.I  The effect of this
change is to establish the same PEQ statistical procedures for both large and
small effluent data sets.  Final procedure 5.B.I, unlike proposed procedures
S.B.l.d and 5.C, does not establish different (PEQ) procedures for when there
are 10 or fewer data points and when there are greater than 10 data points.
Instead, the final Guidance provides a statistical procedure that is
appropriate for use with both small and large data sets (i.e., fewer than ten
data points, and greater than 10 data points).   EPA notes that the procedure
at 5.B.l is an optional statistical procedure for determining PEQ that EPA
believes conforms with the statistical procedures described in section 3.3 of
the TSD, proposed at 5.B and C, and the characteristics of a PEQ procedure
described at 5.B.2.a-c of the final Guidance.  Under this final Guidance,
States have the flexibility to adopt a PEQ procedure other than the procedure
at 5.B.I, as long as the procedure conforms to the characteristics of a PEQ
procedure described at 5.B.2.a-c of the final Guidance.  The alternative
procedure described at 5.B.2.a-c of the final Guidance is really less a
specific procedure than it is a set of required characteristics of a State PEQ
procedure.  Under 5.B.2.a-c, PEQ is to be specified as the 95th percentile of
the distribution of the daily, weekly, or monthly values of the facility-
specific effluent monitoring data projected using a scientifically defensible
statistical method-that accounts for and captures the long-term variability of
the effluent quality, accounts for limitations associated with sparse data
sets, and, unless otherwise shown by the effluent data set, assumes a
lognormal distribution of the facility-specific effluent data.

      Once the PEQ is determined by the permitting authority, the final
Guidance, like the proposal, requires the PEQ to be compared to the
preliminary effluent limit  (PEL) by the permitting authority to determine the
need for a WQBEL.

      The changes to the PEQ provisions from the proposal are intended to
simplify the proposed PEQ procedures,  to maintain the flexibility of the
proposal for permitting authorities to select PEQ procedures that fit the
essential characteristics of the PEQ procedures proposed by EPA, and to
clarify that where EPA determines it is necessary to promulgate a PEQ
procedure in a State's program under 40 CFR 132.5, EPA would promulgate the
procedure at 5.B.l of the final Guidance.

      Under the method described in section 5.B.l of this final Guidance,  the
PEQ is to be specified as the 95 percent confidence level of the 95th
percentile based on a log-normal distribution of the effluent concentration;
or the maximum observed effluent concentration,  whichever is greater.  In
calculating the PEQ, the permitting authority would identify the number of
effluent samples and the coefficient of variation of the effluent data, obtain
the appropriate multiplying factor from Table 1 of procedure 6 of appendix F,
and multiply the maximum effluent concentration by that factor.  The
coefficient of variation of the effluent data would be calculated as the ratio
of the standard deviation of the effluent data divided by the arithmetic

-------
320    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


average of the effluent data, except  that where there are fewer  than ten
effluent concentration data points, the coefficient of variation shall  be
specified as 0.6.  If the PEQ exceeds any of the preliminary effluent
limitations developed in accordance with section A.3 of procedure 5, the
permitting authority would establish  a WQBEL in an NPDES permit  for such
pollutant.

      The PEQ procedures set forth in 5.B.I account for effluent variability,
an important component in determining whether a discharge 'will cause, have the
reasonable potential to cause or contribute to an excursion above a water
quality criterion.  Effluent quality varies over time.  An effluent
measurement taken on any given day may or may not be representative of  the
reasonable upper bound effluent concentration in that particular discharge.
The PEQ procedure in the final Guidance (like the proposed PEQ procedures)
ensures that PEQ will be estimated as the reasonable upper bound effluent
concentration by requiring the PEQ to be specified as the upper  bound  (no less
than the 95th percentile) of the distribution of the projected population of
effluent concentrations.  The proposal would have provided that  the upper
bound estimate be specified as the 99th percentile.   As noted above, several
commenters expressed concern that the 99th percentile was too extreme,  and
that it could result in estimates of  PEQ that were unrealistic.  While  EPA
does not entirely agree with the basis of these comments, EPA also recognizes,
as described in chapter 3 of the TSD, that the 95th percentile upper bound
estimate of effluent data is an acceptable upper bound for purposes of  making
reasonable worst case estimates of effluent quality.  The 99th percentile
would, for practical purposes, be the highest one could specify  the worst case
estimate.  Instead of requiring this  estimate to be specified as the 99th
percentile, the final Guidance establishes a "floor" at the 95th percentile.
States of course have the flexibility to set PEQ at higher levels  (e.g., the
99th percentile.  Requiring the PEQ to be specified as no less than the 95th
percentile is also consistent with EPA's longstanding guidance in the TSD.

      Another change from the proposal is that the final PEQ procedure  under
5.B.2. differs from the proposal with respect to specifying the  PEQ as  the
greater of the 95th percentile value  or the maximum observed value.  The final
procedure does not., like the proposed procedure, require PEQ to  be specified
as the greater of the maximum observed effluent value and the projected upper
bound value  (99th percentile in the proposal) of the distribution of the
projected population of daily value effluent values.  Instead, as noted above,
the final procedure is simplified.  It requires the PEQ to be specified as no
less than the 95th percentile of the  distribution of the projected population
of effluent concentrations.  It is important to note that while  the
specification of PEQ as the greater of the maximum observed value or the
projected upper bound value would not be required if a State adopts a PEQ
procedure under the more flexible provisions at 5.B.2., such a provision is
recommended later in this document, and is a feature of the more specific TSD
procedure at 5.B.I.

      Some commenters expressed confusion over what the proposal meant  when it
stated that PEQ would be expressed as the greater of the 99th percentile of
the distribution of daily values of effluent concentration data  or the
observed maximum effluent value.  Some commenters thought this meant that PEQ
would, in some cases, be the 99th percentile of the distribution of  "observed"
effluent values.  EPA's intent in the proposal on this provision was for the
permitting authority to compare the maximum observed effluent value with the
99th percentile of the distribution of the  "projected" population of daily
values, and specify PEQ as the greater of the two.  In short, the term  "99th
percentile" in the proposal referred  to a projected, not an actual value.  To
clarify this point, the final PEQ procedures include the word "projected"
immediately prior to the word "population"  in the sections 5.B.I and 2.

      In addition, upon further analysis, EPA has concluded that in the vast
majority of cases, the  statistically  projected upper bound value will be
greater than the maximum observed value.  This relationship is more pronounced

-------
                           Section Vm.E: Reasonable Potential                       321
with few data points than it is with numerous data points.  In other words,
the more data points that one has, in general, the closer the observed maximum
value will be to the greater projected upper bound value.  EPA notes, however,
that in rare instances, the maximum observed effluent value will be greater
than the projected upper bound value.  Under this circumstance, EPA recommends
that the permitting authority specify the PEQ as the maximum observed value.
EPA makes this recommendation based on the principle that permitting
authorities should rely on all valid and representative effluent data, and
should not ignore such data where it is available.  Where an observed effluent
value exceeds a statistically projected upper bound value, EPA believes the
observed value should take precedence.  In short, the comparison of the
observed to the projected values and the specification of PEQ as the greater
of the two is recommended by EPA, but not a required feature of State PEQ
procedures.

      As noted above, the final Guidance retains the TSD approach to
calculating PEQ that was proposed at S.B.l.d.  It appears at 5.B.I of appendix
F of the final guidance.  The guidance specifies the procedure at 5.B.I as one
option a permitting authority may select in adopting a PEQ procedure, and as
the specific procedure EPA would promulgate in a permitting authority's
program should it become necessary to promulgate such a procedure in a
permitting authority's' program under 40 CFR 132.5.  EPA has concluded that
the procedure at 5.-B.1 of the final Guidance is consistent with the provisions
set out at 5.B.2.a-c of the Guidance.  Procedure 5.B.I of appendix F of this
Guidance is based on the principles expressed in the TSD guidance document.
All effluent assessment approaches for individual pollutants have some degree
of uncertainty associated with them.  The more limited the amount of test data
available, the larger the uncertainty and the lower the precision of the
methodology for characterizing the maximum effluent concentration.  Because of
this uncertainty, EPA developed the guidance in the TSD to provide a
statistical approach to better characterize the effects of effluent
variability and reduce uncertainty in the process of deciding whether to
require a WQBEL for a particular pollutant.  The TSD guidance combines
knowledge of efflue'nt variability as estimated by a coefficient of variation
with the uncertainty due to a limited number of data to project an estimated
maximum concentration for individual pollutants in a facility's effluent.  The
estimated maximum concentration is calculated as an upper bound of the
expected lognormal distribution of effluent concentrations at a high
confidence level.  The information is then used by the permitting authority to
determine the need for a WQBEL.

       Under procedure S.B.'l of appendix F, the PEQ is calculated by
multiplying the maximum observed effluent concentration value by a factor
which represents the uncertainty in the degree of variability in the effluent.
The specific value of this factor depends upon the number of effluent
concentration values and the variability of the effluent.  The final Guidance
provides these factors in Table 1 of procedure 6 of appendix F.

      The calculation of the factors in Table 1 of procedure 6 of appendix F
of the Guidance has two parts.  The first is characterization of the highest
measured effluent concentration based on the desired confidence level.  The
relationship that describes this is:

      pn 2 (1 -  confidence level)""

      where "pn"  is  the  lower  bound  ("worst case")  percentile  represented by
the highest concentration in the data and "n" is the number of samples.  The
second part of this calculation is a relationship between the percentile
described above and the selected upper bound of the lognormal effluent
distribution.

      As noted above, one commenter questioned the correctness of the
statistical procedure for estimating PEQ at S.B.l.d and 5.C.I of the proposal
(TSD procedure),  stating that the procedure is fundamentally mathematically

-------
322    Water Quality Guidance for the Great Lakes System ~ Supplementary Information Document

incorrect, stating that the two equations presented in the proposal at  58 FR
20949 contradict each other, and suggesting that the second equation is
incorrect in that it is based on a 99% confidence level rather than as  stated
in the proposal, the 95% confidence level.  EPA partially agrees and partially
disagrees with these comments.

      The commenter's analysis concluded that the two equations presented in
the preamble to proposed procedure 5, the first for determining Pn and the
second for determining the ratio of Pn to the 95th percentile  of the projected
population of effluent values, contradicted each other.  EPA disagrees.  The
first equation, as'noted by the commenter, is correctly presented as:

            Pn a (1  -  confidence  level)"".

This equation is used to characterize the percentile of the maximum observed
data point.  As noted by the commenter, in the second printing of the TSD and
in the preamble to proposed procedure 5, EPA replaced the = sign with a 2 sign
in this equation.  The commenter takes issue with the second equation,  stating
that it contradicts the first equation and that the implicit assumption in the
second equation is that the maximum value of any set of n samples is exactly
equal to and no greater than the p(subscript n)th percentile.  EPA agrees that
the intent behind the second equation is to set the maximum value in any set
of n samples equal to Pn.   EPA does not agree that doing so contradicts the
first equation.  EPA always meant the reasonable potential statistical
analysis in the TSD and the proposal to be "worst case."  The first equation
establishes with a 95% level of confidence the lowest percentile that the
maximum observed value reasonably could represent.   The second equation sets
the percentile to this lowest value thereby establishing the "worst case"
assumption.  EPA does not believe the equations contradict each other.  On the
contrary, the equations work in tandem to achieve a statistically sound
projection of the 95th percentile effluent quality.   The sign in the second
equation below therefore remains as it was in the proposal.

      The commenter correctly pointed out that the example equation presented
in the proposal at 58 FR 20948 (second equation) is based upon the 99%
confidence level as shown by the reference to Pn as  the 40th percentile of the
population and that it should instead be based on the 95% confidence level.
EPA agrees and has revised the second equation below.  This equation which
corresponds to the statistical portion of final procedure 5 is based on the
95% confidence level.  The equation now references the 55th percentile  instead
of the 40th, reflecting that the maximum of five samples represents equal to
or greater than the t percentile at 95% confidence.   The equation also  now
contains the normal distribution value for the t percentile (0.126) in  the
denominator.  The commenter will note that the solution to this equation is
2.3, the same value presented in the flawed equation in the proposal.   This
fact at once demonstrates that the multiplier table at 6-1 of the proposal was
correct and that, while EPA correctly presented the solution to the
calculation used to determine the ratio of the 95th to the t percentiles at
95% confidence  (the multiplier), EPA did not correctly present the equation
itself.  The multiplier table at 6-1 therefore remains as it was in the
proposal.

      As explained-in the proposal, EPA's industrial treatment effluent data
base, which was used by EPA to develop and promulgate effluent limitations
guidelines and standards, suggests that the  lognormal distribution
characterizes effluent concentrations well.  For example, if five samples were
collected  (of which the highest value represents at least the 55th percentile
at the upper 95 percent confidence level), the coefficient of variation is
0.6, and the desired upper bound of the effluent distribution is the 95th
percentile, then the two percentiles can be  related using the coefficient of
variation  (CV) as shown below:

-------
                           Section Vm.E: Reasonable Potential                       323
      c,5          exp (1.6450 - 0.5a2)
                  	-	  = 2.3
      c55          exp(0.126ff - 0.5a2)

      Where cr2 = ln(CV2 + l), and 1.645 and 0.126 are the normal distribution
values for the 95th and 55th percentiles, respectively.

      The coefficient of variation  (CV) of the effluent data is calculated as
the standard deviation of the effluent data divided by the arithmetic average
of the effluent data.  The CV is, thus, a dimensionless measure of the
relative variability of a distribution.  Estimates of the CV can be used when
the available data set for calculating the CV is small.  Typical values of the
CV for effluent data range from 0.2 to 1.2.  EPA recommends in the TSD that
when fewer than ten data points are available, a conservative estimate
(assumes relatively high effluent variability) of the CV is 0.6.  Section
5.B.l of the final Guidance specifies that the CV be calculated as described
above, except that where there are fewer than ten effluent concentration data
points, the coefficient of variation would be specified as 0.6.  Section 5.B.2
of the final Guidance, the more general characteristics of a PEQ procedure, do
not require the CV to be estimated as 0.6 when there are fewer than ten data
points.  Section 5.B.2 does, however require the PEQ procedure to account for
and capture the long-term variability of the effluent quality.  EPA notes that
estimating the CV as 0.6 when there are fewer than ten effluent data points is
a longstanding EPA recommendation contained in the TSD  (chapter 3 and appendix
E) and reiterated here.

      The assumption that the facility-specific effluent data are lognormally
distributed is based on practical experience as discussed in Chapter 3 and
appendix E of EPA's TSD.  For environmental data, EPA has found the lognormal
distribution is usually appropriate.  Although the lognormal distribution does
not provide an exact fit to environmental data sets in all cases, it usually
provides an appropriate and functional fit.

      Although the 95th percentile represents a measure of the upper bound of
an effluent distribution, the TSD states that other percentiles are acceptable
provided they have been demonstrated to provide a similar estimate of effluent
variability.  Procedure 5.B.I of appendix F of the final Guidance sets this
percentile at the upper 95th percent confidence level and the upper bound of
the 95th percentile.  EPA recognizes that there is always uncertainty in
making decisions based on limited or sparse data sets, and that, under this
approach, there is -a possibility of requiring a WQBEL where one may not be
necessary, as well as not requiring a WQBEL where one is needed.  The specific
TSD approach at 5.B.I of the final Guidance minimizes the possibility of not
requiring a limitation where one is actually needed by selecting the upper-
bound 95 percent confidence level of the 95th percentile or the maximum
observed effluent value, whichever is greater.  EPA has concluded that use of
the upper 95 percent confidence level of the 95th percentile effluent
concentration is a reasonable mechanism to assure that the permitting
authority will calculate an effluent concentration that appropriately
characterizes the PEQ.

      Several comme'nters suggested that the final guidance provide permitting
authorities the flexibility to use dynamic modelling techniques to determine
when a discharge causes, has the reasonable potential to cause or contributes
to and excursion of the applicable water quality criteria or values.  EPA has
provided substantial guidance, in the TSD and elsewhere, on dynamic modelling
techniques for generating WLAs.  The essence of wasteload allocation
calculations is to understand the distribution of effluent concentrations of a
pollutant that can be discharged from a facility over time and not cause the
applicable water quality criteria to be exceeded frequently enough to, long
enough to, or by amount which would, result in impairment in the receiving
water.  Dynamic modelling provides a sophisticated probability-based
methodology for quantifying wasteload allocations.  The same essential concept
is at work in understanding reasonable potential.  The reasonable potential

-------
324    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

test, however, is intended to answer whether the distribution of effluent
concentrations of a pollutant from a current discharge demonstrates that the
discharge will, on occasion, be in amount that could cause the applicable
water quality criteria to be exceeded.  The calculation of PEQ and its
comparison to PEL is designed to answer this question.  The PWLA procedures
specified at cross-referenced at 5.A.2 of the final Guidance, and specified in
procedure 3 of the final Guidance already provide the flexibility to the
permitting authority to use dynamic modelling to calculate the PWLA.  Because
the PEL is based on the PWLA, where the PWLA is determined using dynamic
modelling, the PEL also will based on dynamic modelling.  EPA notes that the
flexibility also exists under the PEQ procedures at 5.B.2 of the final
guidance for states to adopt a reasonable potential procedure that uses
dynamic modelling techniques to determine whether under worst case effluent
concentrations and receiving water flows, there is a reasonable potential for
the applicable water quality criteria to be exceeded.  EPA notes that the
dynamic modelling approach used would need to be consistent with the basic
characteristics for specifying PEQ at 5.B.2 of the final Guidance.

      As noted above, several commenters requested that the final Guidance
provide clarification on how "non-detect" and "non-quantified" effluent data
points should be managed for purposes of determining reasonable potential.
EPA is currently in the process of developing a national strategy on how to
manage such values for purposes of determining reasonable potential,
specifying permit limits, and measuring compliance.  Because the strategy is
under development at this time, EPA is not providing definitive guidance on
how "non-detect" and "non-quantified" effluent data points should be managed
for purposes of determining reasonable potential.  EPA notes that permitting
authorities will need to exercise discretion and careful judgement in managing
such effluent data "points.  One option that permitting authorities could
exercise is to assign a value of zero to all effluent values below the
quantitation level  (the Minimum Level  [ML] is described elsewhere in this
document and the final Guidance as an appropriate quantitation level) and
count the values as data points.  Another option that permitting authorities
could exercise is to assign values, such as one-half the detection level to
values below the detection level and one-half the difference between the
detection level and the quantitation level to values that fall below the
quantitation level, but above the detection level.  Another option is to
estimate the values below the quantitation level using a statistical model of
effluent concentrations.  EPA recognizes that there are still other
scientifically defensible approaches to managing "non-detect" and "non-
quantified" effluent data points for purposes of determining reasonable
potential.

d.    Determining Reasonable Potential Using Pollutant Concentration Data
      Where the Effluent Flow Rate is Equal to or Greater than the 7O10.

      Procedure 5.B.2 of appendix F of the proposed Guidance would have
established requirements for situations where the effluent flow  rate is equal
to or greater than the  critical low flow of the stream  (7Q10).   In such
effluent dominated-discharge situations, the requirements for determining the
need for a WQBEL in the proposal were identical to those in procedure 5.B.I of
appendix F with one exception:  the upper bound value of daily samples and the
upper bound value of weekly and monthly averages were to be compared to 50
percent of the preliminary effluent limitations based on wasteload allocations
 (instead of to the full 100 percent of the preliminary effluent  limitations).
The final Guidance does not contain this  "low dilution" provision.

      i.    Proposal:  The proposal contained a requirement at 5.B.2 to
compare the PEQ to 50 percent of the PEL  instead of the entire PEL   (low
dilution provision).  The proposal pointed out that this low dilution
requirement would riot increase the stringency of a WQBEL; instead it would
better ensure that a. WQBEL would be included in a NPDES permit in effluent
dominated situations in the Great Lakes System.  The proposal made the point
that because the procedures in section  5.B of appendix F were based on

-------
                           Section Vffl.E: Reasonable Potential                       325
statistical estimates, there was a small potential for a facility to discharge
pollutants at higher concentrations that would exceed a water quality
criterion.  The proposal suggested that this potential is offset in higher
dilution waters because the simultaneous occurrence of the high effluent
concentration and low stream flow is rare.  In contrast, the proposal
maintained that there is little substantial ambient stream flow in streams
with low dilution capacity.  The proposal stated that the 50 percent factor in
procedure 5.B.2 of appendix F provided a reasonable level of assurance that a
WQBEL would be imposed where appropriate.

      The low dilution provision proposed at 5.B.2 also precluded use of the
TSD PEQ procedure (proposed at S.B.l.d and promulgated in this final Guidance
at 5.B.I.a) in low dilution waters (where effluent flow is equal to or greater
than 7Q10 flow).  The proposal explained the basis for precluding the TSD
procedure in low dilution waters:  that the statistical approach described in
procedure B.B.l.d of appendix F of the proposal, the TSD approach, does not
provide a separate explicit mechanism to account for the need for additional
assurances in low dilution streams.  Rather, the approach addresses this
factor implicitly in the selection of the confidence level used.  In instances
of low dilution streams, a permitting authority could use a higher confidence
level to increase the likelihood that a WQBEL will be required and thereby
provide the appropriate level of additional assurance.  EPA proposed
precluding the use of the TSD approach in low dilution waters because the TSD
does not include specific guidance on how high to adjust the confidence level
in these situations.  EPA invited comment on whether the TSD approach should
also be available as an option to the permitting authority.

      ii.   Comments:  EPA received numerous comments on the use of only one-
half of the PEL for comparison to PEQ in low dilution waters  (low dilution
provision).  None of the comments EPA received on the low dilution provision
expressed support for it.  The comments generally opposed the low dilution
provision proposed at 5.B.2, suggesting that it is excessively conservative,
without technical merit, and unwarranted.  Commenters pointed out that the PEQ
procedure is already conservative and that the extra layer of conservatism of
the low dilution procedure would be unnecessary.  Commenters also pointed out
that the allowable dilution, which is required to be factored into the PEL
calculation, automatically results in discharges to low dilution streams being
judged differently than discharges to high dilution streams.  EPA interpreted
this comment to mean that the amount of available dilution in the receiving
water is already accounted for when calculating the PEL.  Commenters further
pointed out that even if the low dilution provision were necessary, and they
argued that it is not, reducing the PEL by a factor of two would be arbitrary.

      iii.  Final Guidance:  EPA was persuaded by the numerous comments
opposing the low dilution provision.   Commenters pointed out that EPA had not
shown this extra conservative assumption to be warranted, especially because
the available dilution,  or lack thereof, is expressly accounted for when
calculating the preliminary wasteload allocation.  Commenters pointed out, EPA
believes correctly,  that the actual amount of dilution available in any
particular receiving water is specified in the dilution calculations required
by procedure 5.A.1-2 of the final Guidance.  Where little dilution is
available due to the small size of the receiving water, this fact is accounted
for in the dilution calculation.  Furthermore, the PEQ statistical procedure
is, by design,  conservative.  As noted above,  the PEQ will in the vast
majority of cases be greater than the observed maximum effluent concentration.
Because the PEL calculation includes a dilution factor and because PEL is
compared to an intentionally conservative PEQ, EPA is persuaded that the low
dilution provision's not warranted as a general practice.  EPA believes it is
more appropriate to leave to the permitting authority the discretion to adopt
a more conservative approach to address, on a case-by-case basis, those rare
instances additional caution is warranted, for example when PEQ is less than
the actual effluent concentration and dilution is very low.

-------
326    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Therefore, the final Guidance does not require States to adopt the
proposed low dilution provision that would require the permitting authority to
compare the PEQ to 50% of the PEL when the effluent flow is equal to or
greater than the stream seven-day, 10-year flow.  EPA notes, however, that
States have the flexibility to exercise caution on a site-specific basis
where, for example, due to lack of available dilution, there is a possibility
that a discharge could result in environmental harm.  Where such circumstances
exist, State procedures can include the flexibility to make a finding of
reasonable potential, where such a finding would not be made using the
procedures specified in the final Guidance.

e.    Determining Reasonable Potential in the Absence of Facility Specific
      Effluent Monitoring Data.

      i.    Proposal:  The proposal would not have established any new  or
specific provisions addressing how to determine the need for WQBELs in  the
absence of effluent data for a specific facility.  Instead, the proposal would
have relied on the existing regulations and procedures for determining  the
need for WQBELs contained at 40 CFR 122.44(d)(1) and guidance in Chapter 3 of
the TSD.

      The proposal solicited comments on whether existing guidance is
sufficient for determining the need for WQBELs in the absence of facility
specific effluent monitoring data, and on whether minimum data requirements
should be specified in the final Guidance.  EPA also solicited comments on any
alternative procedures for making the determination of the need for WQBELs in
the absence of facility-specific effluent data.

      ii.   Comments:  EPA received a few comments on guidance for determining
the need for WQBELs in the absence of facility specific effluent monitoring
data.  In general, commenters agreed  that existing EPA procedures and guidance
for determining the need for WQBELs in the absence of facility specific
effluent monitoring data are sufficient and provide an appropriate level of
flexibility to permitting authorities to use professional judgement.

      EPA received "one  comment suggesting an alternative procedure for  making
the determination of the need for WQBELs for human health protection in the
absence of facility-specific effluent data.  The commenter suggested that
depending on the extent of data available on a chemical in question,
reasonable potential could be determined based on an appropriate surrogate
chemical.  The commenter suggested that such a determination may be suitable
for regulatory purposes, provided there is  sufficient certainty that the
surrogate chemical will estimate an acceptable environmental concentration for
the chemical lacking sufficient data.  The  commenter suggested that this
surrogate chemical approach would not be accurate and defensible enough to use
in establishing  a  specific numeric value for a regulatory purpose, i.e.,
permit limit, cleanup limit, etc., but that  the process would be suitable for
determining the  necessity  of generating further data on the chemical in order
to calculate a permit limit, or, to determine if a sufficient margin of safety
exists in establishing  a permit limit based on another toxicity  characteristic
of a  chemical  (e.g., aquatic toxicity), which is sufficiently protective of
human health and wildlife.  The commenter noted that the determination  of the
need  for additional  data collection,  and reasonable potential, could be made
using an appropriate surrogate chemical provided there is  sufficient data, the
investigator is  qualified, and there  is sufficient certainty that  the
surrogate chemical estimate  is an acceptable environmental  concentration for
the pollutant in question.

      EPA agrees that such determinations may be made vising data and
information other  than  specific effluent data for the pollutant  of concern
provided there  is  sufficient certainty that surrogate chemical data estimate
an acceptable estimated concentration of the pollutant in  question.  The final
guidance maintains the  provision  at  B.C.3  that provides permitting authorities
with  the flexibility to determine reasonable potential and to  incorporate

-------
                           Section YULE: Reasonable Potential                       327
WQBELs into permits, in the absence of.chemical-specific data for the
pollutant of concern.  In addition, section 5.C.I of appendix F of the final
guidance specifies that permitting authority use all available, relevant
information, including Quantitative Structure Activity Relationship
information and other relevant toxicity information, to estimate ambient
screening values that will protect humans from health effects other than,
cancer, and aquatic life from acute and chronic effects, for pollutants that a
permittee reports as known or believed to be present in its effluent, and for
which pollutants, data sufficient to calculate Tier II values for non-cancer
human health, acute aquatic life and chronic aquatic life do not exist.  The
surrogate pollutant approach suggested by the commenter is an approach that
could be used to determine the ambient screening values under this provision.

      Several commenters raised concerns about using very small effluent data
sets as the basis for specifying PEQ.  Some commenters suggested that EPA
should add minimum data requirements to the final Guidance.  Section E.2.c.ii
above discusses these comments in more detail.

      iii.  Final Guidance:   Consistent with the proposal, this Guidance does
not establish any new or specific provisions addressing how to determine the
need for WQBELs in the absence of effluent data for a specific facility.  In
these instances, the permitting authority must continue to apply existing
regulations and procedures consistent with 40 CFR 122.44(d)(1) to determine on
a case-by-case basis whether WQBELs are necessary.  EPA's existing guidance
recommends that the regulatory authority use a variety of factors and
information when determining whether or not a discharge will cause, has the
reasonable potential to cause, or contributes to an excursion of a water
quality standard if facility-specific effluent monitoring data are
unavailable.  (See TSD at pp. 50-55.)  At a minimum, existing regulations
require the permitting authority to consider the four factors identified in 40
CFR 122.44(d)(1)(ii) in making a reasonable potential determination regardless
of the availability of facility-specific effluent monitoring data.

      If the permitting authority, after evaluating all available information
on the facility, is not able to determine whether the discharge will cause,
has the reasonable potential to cause, or contributes to an excursion above a
water quality standard, existing EPA guidance recommends that the authority
should require whole effluent toxicity or chemical-specific effluent
monitoring to acquire additional data.  The permitting authority should
require the monitoring prior to permit issuance,  if sufficient time exists, or
as a condition of the issued or reissued permit.   If monitoring is required
after permit issuance,  the permitting authority should also include a specific
reopener clause to allow for subsequent modification of the permit to include
a WQBEL if the monitoring establishes that the discharge causes, has the
reasonable potential to cause, or contributes to an excursion above a water
quality criterion.  (See TSD at p. 55)

f.    Determining Reasonable Potential for Pollutants When Tier II Values are
      Not Available.

      i.    Proposal:  Procedure 5.D of appendix F of the proposed Guidance
specified procedures for determining whether permitting authorities must
generate, or require permittees to generate, data sufficient to calculate Tier
II values when pollutants on Table 6 are known or suspected of being
discharged into the Great Lakes System,  but neither Tier I criteria nor Tier
II values have been derived due to a lack of toxicological data.  EPA
recognized in the proposal that it is preferable to have Tier I criteria
available to compute WQBELs in all circumstances.  However, the development of
Tier I criteria is often costly and time-consuming.  In the absence of a Tier
I criterion, the permitting authority must have some mechanism with which to
interpret and ensure that the narrative prohibition against the discharge of
toxic substances in toxic amounts is reflected in permits  (40 CFR
§122.44(d)(1)(vi)).

-------
328    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      The proposed Guidance included  the use of a Tier II methodology  to
derive values in the absence of Tier  I criteria.  Consistent with this
decision, procedure 5.D of appendix F of the proposal contained  a methodology
for determining whether Tier II data must be generated by the permitting
authority or discharging facility to  determine the need for a WQBEL for a
pollutant in the NPDES permit.

      Procedure 5.D.I of appendix F of the proposed Guidance stated that the
permitting authority would use all available, relevant information including
Quantitative Structure Activity Relationship (QSAR) information  and other
relevant toxicity information to develop "ambient screening values" for each
of the following water quality criteria categories: aquatic life  (acute and
chronic); wildlife; and non-cancer human health for pollutants included in
Table 6 of the proposed Guidance.  These ambient screening values were
proposed to be specified at a level which would not be expected  to cause an
excursion of the narrative water quality standard.  The proposal also
referenced examples of development of ambient screening values in "Technical
Support Document:  Establishment of Ambient Screening Values under the  Great
Lakes Water Quality Initiative," February 1993, which is available in  the
administrative record for this rulemaking.

      The proposed Guidance would have required that the permitting authority
apply the appropriate procedure described in procedure 5.A of appendix F of
the proposal to calculate a preliminary wasteload allocation and preliminary
effluent limitation using the ambient screening value.  If, based on this
information, the permitting authority concluded the discharge would cause, had
the reasonable potential to cause, or contributed to an excursion above an
ambient screening value, the regulatory authority, under the proposal,  would
be required to  either generate or require the permittee to generate the data
necessary to derive Tier II values for the protection of aquatic life,
wildlife, and human health for the pollutant in Table 6.  Once sufficient data
were generated to calculate a Tier II value, the proposal required the
permitting authority to follow the procedures set forth in procedures  5.A
through 5.C of appendix F to determine whether a WQBEL must be incorporated
into an NPDES permit based on the Tier II value.

      EPA proposed procedures 5.D.I and 5.D.2 of appendix F to implement the
existing NPDES regulations at 40 CFR  122.44(d)(1)(vi).  These regulations
direct permitting authorities, in the absence of an applicable numeric water
quality criterion, to establish effluent limitations for pollutants that
cause, have the reasonable potential  to cause,  or contribute to  an excursion
of a narrative water quality criterion using one or more of the  following
options: calculate a site-specific numeric criterion; use EPA's  water  quality
criteria (developed in accordance with section 304(a) of the Clean Water Act)
supplemented where necessary by other relevant information; or establish
effluent limitations on an indicator  pollutant.

      The proposed Guidance in procedure 5.D.I of appendix F to  part 132 did
not require the permitting authority  to estimate ambient screening values or
to generate or require  the generation of data sufficient to develop a  Tier II
value for human health based on carcinogenic effects.

      The proposed Guidance in procedure 5.D.2 of appendix F also did  not
require the development of a Tier II  value for the protection of aquatic life
if the permittee de'monstrated through a biological assessment that there are
no acute or chronic effects on aquatic life  in the receiving water and that
the whole effluent does not exhibit acute or chronic  toxicity based on the
requirements in procedure  6 of appendix F.  The proposed procedure 5.D.2 of
appendix F did not allow this aquatic life Tier II exception for
bioconcentratable chemicals of concern  (BCCs) as defined in 132.2 of the
proposed Guidance, because whole  effluent toxicity tests are not designed to
measure important impacts  from these  pollutants resulting from elevated tissue
concentrations over time.

-------
                           Section Vm.E: Reasonable Potential                       329
      Procedure 5.D.3 of appendix F of.the proposed Guidance stated that,
where there was insufficient information to develop a Tier II value, nothing
in procedure 5.D of appendix F would preclude or deny the right of a State or
Tribe to determine in the absence of the data necessary to derive a Tier I
criterion or a Tier II value, that the discharge of a pollutant will cause,
have the reasonable potential to cause or contribute to an excursion above the
State's narrative criterion for water quality or incorporate a WQBEL for that
pollutant in a NPDES permit.  This provision was consistent with section 510
of the Clean Water Act which expressly retains the State's authority to adopt
and enforce standards, limitations or requirements more stringent than those
in effect under the. Clean Water Act.  Proposed procedure 5.D.4 of appendix F
clarified that if the permitting authority developed a WQBEL pursuant to
procedure 5.D.3 of appendix F under other more stringent authority, it would
not be obligated to generate or require the permittee to generate the data
necessary to derive a Tier II value for that pollutant.

      ii.   Comments:  EPA received numerous comments objecting to several
aspects of the Tier II data collection and value generation requirements
proposed at 5.D of appendix F.  In addition to raising concerns, some
commenters requested clarification on the extent to which professional
judgement can be the basis for determining when there is sufficient data to
generate a Tier II -value.

      Commenters suggested that requiring the discharger to generate Tier II
values for each type of criteria end point regardless of whether the
available, relevant information indicates that all of these target populations
(i.e., aquatic life, humans, or wildlife) would actually be at risk would be
inappropriate.  Commenters explained that Tier II values should be developed
by the permitting authority only for those target populations which are
actually at risk.  For example, where screening values clearly show the
pollutant to be primarily an aquatic life concern, data to calculate a Tier II
human health value ^should not be necessary.

      Several commenters explained that the data collection necessary to
calculate Tier II values and the calculation of such values would be expensive
and time consuming and as a result, difficult for States to accomplish,
especially given that the ambient screening values are loosely defined.  Other
commenters explained that because Tier II data collection and value generation
would be expensive and time consuming, States would tend to pass the
responsibility on to the dischargers.  Commenters questioned whether the
expense of Tier II data collection and value generation should be passed on to
dischargers.   Commenters also pointed out that the Tier II data collection and
value generation requirements would tend to result in duplicative efforts,
i.e., multiple states and dischargers would collect Tier II data and generate
values unbeknownst to the other states and dischargers.  One side effect of
this would be that conflicting Tier II values would be generated.

      In addition, commenters suggested that the Tier II values should not be
used as the basis for permit limits, claiming that the Science Advisory Board
did not support use of Tier II values as the basis for permit limits, but
rather,  supported their use only as screening values to determine when
additional toxicity data for a particular chemical needed to be collected.  In
addition, some commenters suggested that Tier II-based limits would, by
design,  be overly conservative.  Several commenters objected to the proposal
that would require dischargers to spend money for additional effluent controls
to meet Tier II values that the commenters suggest are overly conservative and
in some cases incorrect.

      In response to commenters concerns about the scientific validity of
ambient screening values, EPA acknowledges that the establishment of ambient
screening values involves a considerable amount of judgment by the permitting
authority in the face of scientific uncertainty.  EPA believes, however, that
the mere absence of a Tier II value for a particular pollutant does not take
away from the permitting authority the obligation to ensure that a pollutant

-------
330    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


discharge that has the reasonable potential to cause or contribute to an
excursion above water quality  standards is subject to appropriate
restrictions.  See 40 CFR 122.44(d)(1).  It therefore behooves the permitting
authority to use the best available information to make this judgment.  EPA
also emphasizes that ambient screening values are just that - "screening"
values that do not in and of themselves result in the establishment of any
enforceable conditions on the permittee.  Rather, the projected  exceedance of
the screening value is a trigger for the development of more information  (Tier
II data collection .and values) that would enable the permitting  authority to
determine the need for a WQBEL and derive the WQBEL if necessary.  As for the
concern about the burden associated with generating Tier II values, as
discussed in the section II.C.2 of this document, Adoption and Application of
Tier II Methodologies, EPA believes that there should not be a significant
number of pollutants for which additional Tier II data will need to be
collected, in light of the fact that most of the pollutants of initial focus
already have, in EPA's view, enough information available to generate Tier I
criteria or Tier II values.

      EPA also received a few comments on the exception to Tier  II data
collection and value generation requirements proposed at 5.D.2 of appendix F
(Tier II exception).  Commenters generally supported the Tier II exception,
noting that it would provide needed flexibility to permitting authorities to
utilize bioassessment and whole effluent assessments to characterize impacts
to aquatic life in receiving waters.  In voicing support for the Tier II
exception, two commenters explained that the whole basis of this regulatory
program should be the reasonable protection of humans and valued resources.
They went on to explain that even though EPA sometimes refers to independent
applicability of the various measures of protection, the real measure of the
protection is whether there is any instream impact.  When considering aquatic
life criteria, impact is demonstrated not to exist where whole effluent
toxicity testing arid biological assessment shows no acute or chronic toxicity.
At that point, the goal has been reached.  Any additional protection based on
statistical procedures and/or aberrations is, according to the commenters, a
waste of resources.

      EPA does not share the commenters' belief that a measurement of whole
effluent toxicity instream is the only valid measure of adverse  water quality
impacts.  Water quality-based permitting requirements are intended, as the
commenters suggest, to prohibit discharges that cause environmental impacts
such as acute and chronic aquatic life toxicity.  Such controls, however, are
also intended to prevent environmental impacts "before" they occur.  If one is
detecting adverse impacts in the receiving water, it often is an indication
that the preventative aspect of the program has failed.  EPA notes that
procedure 5 of appendix F of the final Guidance, like the proposal, is
intended to result in WQBELs to ameliorate discharges that are already causing
measurable environmental impacts, as well as discharges that demonstrate the
reasonable potential to cause such impacts.

      EPA notes that the exception to Tier II value generation for aquatic
life protection has been retained in the final Guidance at 5.C.2.  EPA notes
that, as in the proposal, this exception is for aquatic life Tier II values
only and that in addition to the WET and biological assessment conditions
described in the comment,  the third condition that must be met in order to
exercise the exception is  that there must be insufficient data to calculate a
Tier I criterion or a Tier II value for aquatic life.  Where such data are
available, the exception to Tier  II value generation does not apply.

      One commenter suggested that the Tier II exception should  not be limited
to non-BCCs and  should be  expanded to include BCCs.  In response and as noted
above, and in the preamble to the proposal, the proposed procedure 5.D.2 of
appendix F did not allow this aquatic life Tier II exception for
bioconcentratable chemicals of concern  (BCCs) as defined in 132.2 of the
proposed Guidance, because whole  effluent toxicity tests are not designed  to

-------
                           Section VIQ.E: Reasonable Potential                       331
measure important impacts from these pollutants resulting from elevated tissue
concentrations over time.

      iii.  Final Guidance:  The proposed requirements remain essentially
intact in today's guidance and would require permitting authorities to develop
ambient screening values, develop PELs based on the ambient screening values,
and compare PEQ with PEL for purposes of determining when permitting
authorities are required to collect Tier II data, generate tier II values, and
develop permit limitations to protect the Tier II values from being exceeded.
There is, however, one important change from the proposal.  Unlike the
proposal, the final Guidance does not require use of the Tier II methodology
for wildlife.  Like the proposal, S.C.l.d of the final Guidance provides that
the permitting authority, in the absence of a Tier I criterion, would generate
or require the permittee to generate data to derive a Tier II value.  Like the
proposal, the final Guidance provides that the permitting authorities would
use the Tier II methodology to generate Tier II values for protection of non-
cancer human health, acute aquatic life and chronic aquatic life, but unlike
the proposal, use of the Tier II methodology for wildlife protection by the
permitting authorities is optional.  Likewise, S.D.l.e of the final Guidance,
unlike the proposal, does not require Tier II permit limitations for wildlife.

      EPA would like to clarify that States and Tribes may use the scientific
defensibility exclusion to avoid the data generation requirements in procedure
S.C.l.d of appendix F for pollutants which are not amenable to toxicological
testing as commonly practiced.  In these situations, a State, Tribe, or
discharger would need to demonstrate experimentally that the chemical is not
amenable to testing and provide documentation for this finding.  For a more
thorough discussion on the use of the Tier II methodology, Tier II data
quality assurance, the steps EPA is taking to minimize the burden on States
and dischargers in the collection of Tier II data and calculation of Tier II
values, and the development of Tier II values for wildlife, see section II of
this document on Tier II regulatory requirements and Tier II methodologies.

      In addition, an important clarification of the exception from Tier II
data collection, value generation and permit limit requirements for the
protection of aquatic life ("Tier II aquatic life") is necessary.  The final
Guidance, at 5.C.2, maintains the exception to Tier II aquatic life contained
in the proposal and discussed above.  A permitting authority would not be
required to generate Tier II data or values, or establish Tier II permit
limitations in a permit for an existing discharge, for other than
bioaccumulative chemicals of concern if:  there is insufficient data to
calculate a Tier I criterion or Tier II value for aquatic life for such
pollutant; the permittee has demonstrated through a biological assessment that
there are no acute or chronic effects on aquatic life in the receiving water;
and the permittee has demonstrated in accordance with procedure 6 of this
appendix that the whole effluent does not exhibit acute or chronic toxicity.
      The preamble to the proposed guidance discussed this three-part test in
the context of EPA's existing regulations governing the use of indicator
pollutants in establishing WQBELs.  Upon further consideration, EPA believes
that this discussion in the preamble to the proposal was misplaced and
therefore may have resulted in confusion regarding the precise context for the
exception under section C.2 of procedure 5.  In implementing this provision,
EPA believes that the permitting authority would simply determine, in
accordance with procedure 6,  whether the effluent exhibited whole effluent
toxicity.  If toxicity was not found (and the other two conditions in section
C.2 were met), then the permittee would be exempt from tier II data generation
requirements.  The discussion in the preamble to the proposed rule of the
imposition of a WET limit under section C.2. was therefore besides the point,
since such a limit would not be appropriate if the discharge were found not to
exhibit whole effluent toxicity.

      In order to provide clarification on this point, the condition at
5.D.2.C of the proposed Guidance that would have to be met for a discharge to
be excepted from Tier II aquatic life requirements is maintained in the final

-------
332    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


Guidance at 5.C.2.C with one change.  The condition at 5.C.2.C has been
clarified to say that, to fulfill the condition at 5.C.2.C of the final
Guidance, the permitting authority must determine in accordance with procedure
6 of the final Guidance not only that the discharge does not exhibit WET, but
also that limits for whole effluent toxicity are therefore not required.

      Because the preamble discussion in the proposed Guidance discussed the
condition at 5.D.2.C  (final 5.C.2.C) in the context of EPA's indicator
parameter regulation, it was unclear regarding whether satisfying condition
2.c, alone, would exempt the permitting authority from the requirement to
generate Tier II data and calculate a Tier II value.  EPA intended that the
proposed conditions for exemption from Tier II aquatic life at 5.D.2.a-c would
"all" have to be met in order for the permitting authority to exempt a
discharger from tier II data collection, value generation and limits.  EPA
received no significant  comments suggesting that a WET limit alone, or
indicator parameter limit alone, should except a permitting authority from the
Tier II requirements.  Therefore, the final Guidance, at 5.C.2.a-c, maintains
the conditions for exception from the Tier II requirements.  Neither a finding
of no WET in the discharge, nor use of WET as an indicator parameter limit,
alone, is sufficient basis to exempt the permitting authority from Tier II
data collection and value generation requirements.  The permitting authority
can only be exempted from Tier II data collection and value generation
requirements by satisfying each of the conditions in 2.a-c.  The permitting
authority may, as described in the preamble to the proposal, include an
indicator parameter limit in the permit in lieu of the tier II limit provided
the requirements in the existing regulations at 40 CFR 122.44(d)(1)(vi)(C) are
met.  EPA notes that the Tier II procedure for aquatic life in the final
guidance is the required method for interpreting state narrative criteria that
protect aquatic life.  Therefore, once a Tier II value is determined for a
pollutant of concern, an indicator parameter limit used in lieu of a Tier II-
based limit for the pollutant of concern must, in order to be used, be shown
to attain and maintain the Tier II value for the pollutant of concern.  Such a
showing must be made in order to fully meet the requirements of
122.44(d)(1)(vi)(C).

      The proposal ^tated that EPA believes that WET can serve as an indicator
parameter under appropriate circumstances.  Having received no public comments
to the contrary, EPA stands by its statement.  EPA notes that WET is by no
means the only parameter that can serve as an indicator parameter under
122.44(d)(1)(vi)(C).  EPA also notes that whenever an indicator parameter
limit is used for a WQBEL, the conditions at 122.44(d) (1) (vi) (C) must be met.

      Finally, the proposed guidance specified at 5.D.3 and 4 is retained
essentially intact  in the final Guidance.  These proposed provisions stated
that, "3.  [n]othing in  sections D.I or D.2 shall preclude or deny the right of
a permitting authority  to: a.  [d]etermine, in the absence of the data
necessary to derive a Tier II value, that the discharge of the pollutant will
cause, have the reasonable potential to cause or contribute to an excursion
above a narrative criterion for water quality; and b.  [t]o incorporate a WQBEL
for the pollutant into  an NPDES permit.  4.  If the permitting authority
develops a water quality-based effluent limitation consistent with  section D.3
of this procedure,  it shall not be obligated to generate or require the
permittee to generate the data necessary to derive a Tier II value  or values
for that pollutant."  The provision at 5.C.4,  (proposed 5.D.4) has  been
changed to clarify  an issue raised by commenters.  Commenters pointed out that
the proposed provisions at 5.D.3 and 4 seemed to provide maximum flexibility
to permitting authorities to derive and require WQBELs using procedures in
lieu of the Tier II provisions for generating data and values.   The preamble
to the proposal explained that the proposed provision at 5.D.4 could only be
exercised by a permitting authority where the result would be a  WQBEL more
stringent than one  based on a Tier II value.  The change to the  language at
C.4 of procedure 5  of appendix F of the final Guidance clarifies that an
alternative WQBEL under C.4, one not based  on a Tier II value, must be shown
to be more stringent  than a Tier II-based limit would have been  in  order for

-------
                           Section VHI.E: Reasonable Potential                       333
such a limit to negate the need for a Tier II-based limit.  One effect  of this
clarification is that it eliminates the possibility that this provision would
result in less stringent requirements than would have been obtained under the
approach in the final Guidance.  Under C.4, a permitting authority could not,
for example, choose to forgo generating a Tier II value without showing that
the WQBEL derived using alternative procedures under C.3 would be at  least as
stringent as the tier II-based limit would have been.  This would occur where
the alternative limit under C.3 is zero.  In this example it is clear that the
alternative limit is at least as stringent as the tier II-based limit would
have been.  This clarification is consistent with the principle that  States
can, under 510 of the Act, be more stringent than EPA.

g.    Determining Reasonable Potential Using Fish Tissue Data

      i.    Proposal:  Procedure 5.F.3 of appendix P of the proposal  would
require that permitting authorities establish a WQBEL if the discharger has a
pollutant in its effluent at detectable levels and fish tissue from the water
body also contains the pollutant at levels that exceed the tissue basis of the
water quality criteria.  This provision would apply to instances where
proposed procedures 5.B and 5.C of appendix F did not project the reasonable
potential of a discharger to cause or contribute to an excursion above  a Tier
I criterion or Tier II value but tissue data from ambient fish sampling
demonstrates an excursion.  These instances occur when ambient water
concentration monitoring either does not include the pollutant of concern or
else the pollutant .is present in ambient waters at a level below the  ability
of analytical chemical methods to detect or quantify.  Nevertheless,  the
presence of the pollutant in fish tissue at levels that exceed the tissue
basis of the Tier I criterion or Tier II value demonstrates that the  criterion
or value is not met.  Under NPDES regulations at 40 CFR 122.44(d)(1)(i), a
WQBEL is required for that pollutant or pollutant parameter whenever  there is
information that demonstrates that the discharge of a pollutant causes  or
contribute to such an excursion.  The provisions of proposed procedure  5.F.3
of appendix F would implement the requirements of 40 CFR 122.44(d)(1)(i) with
respect to ambient fish tissue data.

      ii.   Comments:  EPA received several comments on the provision proposed
at 5.F.3 of appendix F specifying the use of fish tissue data for purposes of
determining reasonable potential.  Several commenters opposed retaining the
provision that would require fish tissue as the basis for reasonable  potential
determinations, citing various difficulties with the approach.  Other
commenters supported the use of tissue data, but suggested that the approach
needed to be better explained, particularly in the areas of determining the
relevance of the tissue data to the discharge and whether the discharge is
actually contributing to the tissue contamination.  One commenter strongly
supported the use of fish tissue data as a required basis for determining
reasonable potentia.1.

      The commenter that strongly supported the use of fish tissue data as a
required basis for determining reasonable potential suggested the provision
should be expanded to require the development of numeric permit limits  to
implement narrative criteria for any discharge of a pollutant that has
resulted in or contributed to the issuance of a fish consumption advisory
downstream of the discharge (including in the open waters of the Great  Lakes).
The commenter explained that the issuance of fish consumption advisories by a
State,  Tribal or federal agency represents, in effect, a determination  that
the subject waters have not achieved the "fishable waters" goal of the  Clean
Water Act,  i.e. that narrative criteria prohibiting toxic pollutants  in toxic
amounts are being violated.  The commenter concluded that  any discharge of
such a pollutant is, by definition,    contributing to a criteria violation and
requires a WQBEL.

      Commenters that' opposed the use of use of fish tissue data as a required
basis for determining reasonable potential explained that using fish  tissue
data in permitting is technically difficult due to the variability of the data

-------
334    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


and the difficulty of pinpointing contributing sources.  These commenters also
suggested that the use of the word "detectable" in the fish tissue provision
incorrectly implies that there is no threshold of effect, i.e.,  one molecule
is enough to cause 'unacceptable bioaccumulation.

      The commenters that supported the use of tissue data, but  suggested that
the approach needed to be better explained, pointed out that permitting
authorities should have the flexibility to determine the relevance of tissue
data to any particular discharger in question.

      One commenter suggested that EPA needed to clarify what is meant by
"detectable level" in a facility's discharge, pointing out that  some
analytical methods are more innovative and sensitive than others.  The
commenter suggested that EPA should clarify its intent as to whether
innovative methods should be employed in a reasonable potential  context.

      Another commenter requested clarification on whether fish  tissue data
under proposed procedure 5.F.I should be obtained through caged  or ambient
fish tissue studies.

      Other commenters suggested that the use of resident fish tissue data is
without sufficient scientific basis for use in the reasonable potential
procedures.  These commenters expressed concern that fish tissue data is
unreliable.  The commenters also suggested that such data does not lead to
those sources which   contribute to the problem.  The commenters suggested
that this procedure, if it is used, should cause permit limits to be generated
for those facilities contributing a level of the pollutant high  enough to be
of concern.  The commenters expressed concern that the proposed procedure
would result in the generation of many unnecessary permit limits throughout
the region resulting in monitoring costs that would squander resources that
would better be applied to other uses.

      The preamble to the proposal discussed the need for permitting
authorities to use care in determining what tissue data are representative of
ambient conditions..  Care in judgement by the permitting authority should be
taken in assuring .that the fish tissue data is representative of the ambient
conditions in the discharger's receiving water, and relevant to  the
discharger's effluent.

      iii.  Final Guidance:  This provision on fish data remains essentially
unchanged from the proposal and is located in the "Other Applicable
Conditions" section at 5.F.4 of appendix F of the final Guidance.  However,
the words "to that water" have been added to 5.F.4 to clarify that a
discharger must discharge detectable levels of the pollutant to  the waterbody,
the condition of which is reflected  by the fish tissue data, to be required
to have a WQBEL for such pollutant under this provision.  This change simply
clarifies the intent of the proposal.

      This provision does not exclusively link the need for WQBELs for  a
discharge to the existence of a fish advisory on the receiving water, but it
does require a WQBEL for pollutants found in the discharge that  are also found
in tissue of fish from the receiving water at levels that exceed the tissue
basis of the criteria or values.  EPA expects there to be a correlation
between fish advisories and tissue levels that exceed the tissue bases  for the
criteria or values.  However, it  is possible that a fish advisory could be in
place, where tissue data from the receiving water shows the mean tissue value
for a particular pollutant to be  below the tissue bases for the  applicable
criterion or value for that chemical.  Therefore, the provision  at F.4  does
not automatically require WQBELs  for discharges containing the chemical for
which a fish advisory has been  issued.

      EPA recognizes that  the existence of a pollutant at a detectable  level
in an effluent that is also found in tissue from the receiving water at levels
that exceed the  tissue bases of the applicable criteria or values is not

-------
                           Section Vm.E: Reasonable Potential                       335
necessarily an indication that the discharge of the effluent is solely the
cause of the fish qontamination, or even a substantial contributor of such
contamination.  However, such a finding is an indication that the discharge of
the effluent is a possible contributor of such contamination and therefore
exhibits the reasonable potential to cause or contribute to the excursion
above applicable water quality standards.  EPA notes that the reasonable
potential procedures as a whole are intended as a conservative screening
procedure to determine where a discharge should contain a WQBEL to either
prevent possible future contamination or curtail the contribution to existing
contamination.  EPA also notes that the reasonable potential procedure, while
specifying whether a discharge must have WQBELs, does not specify the actual
value of the permit limitation.  Hence, where reasonable potential is found
under this tissue provision, the permit limit will not necessarily be required
to be set at the detection level, but could, as noted in the discussion in
section h below, be set at the applicable criteria level, or, through a TMDL
at a different level if appropriate.

      EPA recognizes that when evaluating reasonable potential under this
tissue provision, the permitting authority will need to exercise discretion
and careful judgement in determining whether fish tissue data are
representative of ambient conditions and in determining the relevance of fish
tissue data to any particular discharger.  EPA guidance on these
considerations is provided in "Assessing Human Health Risks from Chemically
Contaminated Fish and Shellfish:  A Guidance Manual" (USEPA September 1989,
EPA-503/8-89-002), which is available in the administrative record for this
rulemaking.

      EPA recognizes that there can be wide differences in the detection level
of different chemical analytical methods and analysts,  and that different
methods and analyses for detecting and measuring the same chemical will have
different detection levels.  EPA notes that the reasonable potential
procedures in the final Guidance do not require the generation of effluent
data, only that existing effluent data be used when determining reasonable
potential.  In using existing effluent data, EPA recommends that the detection
level of the analytical method employed be recorded in the record of the
reasonable potential decision by the permitting authority.  In addition, when
requiring collection of effluent data for purposes of determining reasonable
potential, EPA recommends that the permitting authority specify the use of
analytical methods that are sensitive enough to detect in the range of the
tissue basis of the criteria or value, or where commonly used analytical
methods are not sensitive enough to detect at such level, specify the most
sensitive commonly used method.

      As discussed in section VIII.C of this document,  EPA recognizes that
both caged fish and resident fish studies can yield valid results indicating
the level of contamination in fish in a waterbody.  As noted above, EPA
recognizes that permitting authorities will need to exercise careful judgement
in determining the scientific validity of any fish tissue study, whether the
study results are representative of the condition of the receiving water, and
the relevance of the tissue data to any particular discharge.

      Procedure 5.F.4 of appendix F compares the geometric mean of tissue
samples collected from ambient fish to the tissue basis of the Tier I
criterion and Tier II values for human health and wildlife protection.  The
tissue basis is equal to the bioaccumulation factor that was used to calculate
the Tier I criterion or Tier II value multiplied by the Tier I criterion or
Tier II value.  The tissue basis for the same pollutant may differ for human
health and wildlife criteria and values; if any tissue basis is exceeded,
reasonable potential exists with respect to facilities discharging detectable
levels of the pollutant.  The mean of the ambient data is used in the
comparison to be consistent with the assumptions of the criteria, that is,
wildlife and human consumers of fish eat an assemblage of fish.  A mean best
reflects this assemblage.  The geometric mean is used as the most
representative way to reflect the average of environmental samples.

-------
336    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      Procedure 5.F. 4 of appendix F also recognizes that there may be
differences in tissue concentrations between fish samples collected from a
specific water body.  The reasons for this include differences in lipid
content between fish, the ages of fish, and the actual exposure  of individual
fish.  The use of a geometric mean in the comparison serves to overcome some
of the inherent variability because the mean reduces the effect  of any  one
sample.  However, there may be still be some variability associated wi.th using
fish tissue data.  Therefore, procedure 5.F.4 of appendix F directs the
permitting authority to consider the variability of a pollutant's
bioconcentration and bioaccumulation in fish.  The assessment of the
variability may be .accomplished by applying specific factors to  adjust  for
differences in lipid content or age, or by applying an overall factor based on
review of the variability in literature or field data.  Whatever method is
used by the permitting authority must be described in the administrative
record supporting the permit decision.

      Procedure 5.F.4 of appendix F applies to all facilities that discharge
detectable levels of a pollutant into a water body where the pollutant  is
found in the fish tissue in the water body at levels exceeding the tissue
basis of a Tier I criterion or Tier II value.  This provision recognizes that
all facilities that discharge detectable levels of the pollutant into the
water body are contributing the pollutant and therefore meet the requirements
Of 40 CFR 122.44(d)(1)(i).

h.    Basis for Effluent Limitations

      As noted in the discussion of general requirements of procedure 5 above,
sections C and D of appendix F of the final Guidance provide that, regardless
of the manner in which the reasonable potential determination is made,  all
effluent limitations must comply with all other applicable State, Tribal and
Federal requirements.  This statement was in the proposal and remains intact
in the final Guidance.  However, as noted above, EPA has clarified, in
response to public comments, the connection between procedure 3  (TMDL
procedure) and procedure 5  (Reasonable Potential procedure).  The final
Guidance omits from procedure 3 the proposed provision that would have
required a TMDL under 40 CFR 130.7 for each pollutant  (and waterbody) for
which there is reasonable potential that a discharge causes, has the
reasonable potential to cause, or contributes to an excursion above the
applicable water quality standards.  Therefore, while a finding  of reasonable
potential continues to require a WQBEL, it no longer requires a  TMDL as the
basis of that WQBEL.   In light of this change, EPA concluded it  was necessary
to include in procedure 5 the basis for developing wasteload allocations from
which to derive WQBELs  in the absence of a TMDL under 40 CFR 130.7.
Therefore, the final Guidance contains a new provision at section F.2 in
procedure 5 of appendix F.  This provision specifies that once the permitting
authority has determined in accordance with procedure 5 that a WQBEL must be
included in an NPDES permit, the permitting authority shall:  1) Rely upon the
wasteload allocation established  for  the point source as part of any TMDL
prepared under procedure 3 of this appendix F and approved by EPA pursuant to
40 CFR 130.7, or,  in the absence  of such TMDL, calculate wasteload allocations
for  the protection  of  acute and chronic aquatic life, wildlife and human
health using, at a minimum, the procedures set forth in section  A.I of
procedure 5 for  developing preliminary wasteload allocations; and 2) develop
effluent limitations consistent with  these wasteload allocations in accordance
with existing State  or Tribal procedures for converting wasteload allocations
into water  quality-based effluent limitations.  Similar conforming changes
have also been made  to procedure  4.C  to address TMDLs, wasteload allocations,
and  preliminary  wasteload allocations.  In making this clarification, EPA  is
remaining consistent with  its intent  expressed in the proposal that WQBELs be
consistent  with  calculated wasteload  allocations.

      By including a separate provision in  the final guidance addressing
procedures  to be followed  in deriving WQBELs in the absence  of a TMDL,  EPA has
not  made a  substantive change from  the approach contained  in the proposal.  As

-------
                           Section Vffl.E: Reasonable Potential                       337
discussed above, the structure of the proposed guidance would have called for
the development of a TMDL for the purpose of deriving wasteload allocations
where the permitting authority determined reasonable potential existed.  The
final Guidance has simply "moved" those procedures into a new subsection,
5.F.2.a., of appendix F of the Guidance.  This "move" is necessitated by the
fact that, under the final Guidance, the actual development of a TMDL is not a
prerequisite to the establishment of a wasteload allocation and permit limits.

      Finally, it is important to note that, as discussed in section VIII.C of
this document, the final Guidance does not, like the proposal, require
wasteload allocations to be set equal to zero in cases where background
concentrations of the pollutant in the receiving water exceed criteria or
values  (non-attained waters), and a multiple source TMDL has not been
completed.  As noted in section VIII.C of this document, EPA did not include
this provision  (high background provision) in the final Guidance because
setting a wasteload allocation at zero as a default, in the absence of a TMDL,
may not be appropriate in many situations.  EPA recognizes that many factors
need to be considered when background water quality concentrations exceed
criteria or values.  Furthermore, many commenters objected to a mandate of
setting wasteload allocations equal to zero in non-attained waters unless a
multiple source TMDL has been completed.  Commenters pointed out that such a
mandate would, in effect, force all point sources to achieve zero discharge of
pollutants to non-attained waters.

      Once EPA concluded that it was inappropriate to include the high
background provision in the final Guidance, EPA then had to determine if there
is an appropriate alternative to the high background provision.  Commenters
suggested a range of alternatives for setting wasteload allocations for
discharges to non-attained waters in the absence of a multiple source TMDL.
The suggested alternatives ranged from setting the wasteload allocation to the
most stringent applicable criterion up to setting the wasteload allocation
equal to the background concentration of the receiving stream.  Others
suggested that the wasteload allocation be set equal to the greater of the
most stringent applicable criterion or the background concentration.  EPA
examined these suggested alternatives to determine which of them were
permissible readings of the national program requirements under the CWA.

      Upon review of the alternatives suggested by commenters, EPA notes that
in the absence of a TMDL under 40 CFR 130.7, there are several reasonable
interpretations of national program requirements under the CWA.  One
reasonable interpretation of national program requirements is that in non-
attained waters and in the absence of a TMDL under 130.7, the wasteload
allocation for a pollutant for which the waterbody is in non-attainment, may
be set equal to the most stringent criterion or value applicable to the
waterbody (criteria end-of-pipe).  The concept of a mixing zone to provide for
dilution obviously is not relevant where the stream already exceeds the water
quality criterion.  EPA believes that this approach is consistent with
existing regulatory provisions relating to water quality-based permitting, as
well as the goals and objectives of the Clean Water Act to restore and
maintain the biological integrity of U.S. waters.

      EPA's existing NPDES regulations require that, where a wasteload
allocation has not "been prepared by a state and approved by EPA under 40 CFR
130.7,  water quality-based effluent limits must insure that the "level of
water quality to be achieved by limits on point sources established under this
paragraph is derived from, and complies with all applicable water quality
standards."  40 CFR 122.44(d)(1)(vii)(A).  Consistent with this provision,
water quality based effluent limits set at the water quality criteria end-of-
pipe are "derived from" the applicable state water quality standards.
Moreover, the water quality that would "be achieved by point sources" will be
no greater than the applicable numeric water quality criteria, since all point
sources will be limited to discharging at no' greater than the criteria end-of-
pipe.   EPA recognizes that,  due to contributions from nonpoint sources and
other media (e.g., air deposition of mercury or PCBs), the level of a

-------
338    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

pollutant in the receiving water from all sources combined may exceed numeric
water quality criteria.  EPA believes that limiting discharges from point
sources to criteria end-of-pipe is nonetheless appropriate in these
circumstances, as discussed below.

      Numeric criteria are concentration-based standards designed to protect
the aquatic ecosystem and humans from the adverse effects of pollutant
discharges that would occur at levels above the criteria.  Where the
background level of the pollutant in the receiving water is greater than the
criteria, the stream is in non-attainment and the aquatic environment or human
health is adversely impacted.  A point source discharging at criteria end-of-
pipe in such situations, however, will contain a lower concentration of the
pollutant than the receiving water, and therefore will not increase the
pollutant concentration in the waterway.  Such a discharger may, in fact,
cause the ultimate pollutant concentration in the receiving water to decrease.
Where the environmental effects of a pollutant on the aquatic ecosystem or on
human health are associated with the concentration of the pollutant in the
waterway, limiting discharges from point sources to criteria end-of-pipe in
these circumstances should therefore result in no further degradation of the
waterbody, and may in fact improve the water quality of the waterbody  (special
environmental considerations are present with regard to bioaccumulative
[persistent] compounds, which are addressed separately under the final rule
and discussed further below).  The Agency therefore believes that establishing
limits on point sources under these circumstances at criteria end-of-pipe is
consistent with the underlying environmental objectives of the CWA.

      The Agency recognizes that establishing limits at the criteria end-of-
pipe will not alone result in the attainment of water quality standards in the
receiving water for pollutants that are present mainly due to contributions
from nonpoint sources and other media.  In the absence of a TMDL addressing
comprehensively such sources and corresponding controls on such sources,
however, the water'quality-based permitting process for point sources cannot
achieve compliance with standards in such a waterbody.  Even if the Agency
were, for example, to prohibit discharges from point sources entirely under
these circumstances, standards would not be attained in the waterbody.
Indeed, where effects on aquatic life or human health are due to the
concentration of the pollutant in the water column, allowing discharge at
criteria end-of-pipe may actually improve water quality as compared with
prohibiting any discharge at all since the former approach may ultimately
reduce the pollutant concentration in the receiving water.

      For the reasons explained above, EPA believes that, as an  interim
approach until a TMDL can be developed, establishing WQBELs to meet criteria
end-of-pipe is a permissible permitting approach to address adverse
environmental and health effects that are due to the concentration of
pollutants in the water column in non-attained waters.   Allowing such a
discharge means that additional mass of a pollutant may be added to the
waterbody and consideration of adverse effects due to increases  in mass is
well suited to the TMDL development process.  In the interim before a TMDL has
been established, EPA believes that any environmental concerns associated with
such additions of mass  can appropriately be addressed by the permitting
authority through interpretation of the "toxics" narrative criterion contained
in state water quality  standards.  For example, where an addition of mass is,
in and of itself, of environmental concern because of the loadings of  such
pollutants in sediments, the permitting authority could interpret the
narrative criterion to  require more stringent limitations than criteria end-
of-pipe  in order to provide  a requisite level of protection.  Therefore, the
permitting authority retains the ability to address circumstances where
additions of mass alone may  be of environmental concern.

      While the Agency  recognizes that the  criteria end-of-pipe  approach may
not result in attainment of  water quality standards in the near  term on  some
waterbodies, the Agency views this as a reasonable interim approach to water
quality-based permitting until a TMDL is developed for such waterbodies.  EPA

-------
                           Section VHI.E: Reasonable Potential                       339
believes that the TMDL process is the appropriate means of effectively
addressing ubiquitous pollutants in the Great Lakes basin where background
levels exceed standards.  Once a TMDL is established, point sources will have
to have limits consistent with their wasteload allocation established under
the TMDL (which could be lower or higher than criteria end-of-pipe).  EPA
recognizes, however, that TMDLs have not been established for many waterbodies
where background exceeds criteria and that, given the technical difficulties
and financial resources it takes to develop some  TMDLs, the States will not
be able to establish TMDLs everywhere they are needed in the immediate future.
Under these circumstances, the Agency believes that setting wasteload
allocations equal to criteria provides the best way of restricting additional
discharges of pollutants from point sources in the period until a TMDL can be
developed.

      EPA also examined the approach suggested by commenters to set wasteload
allocations equal to background concentrations in non-attained waters in the
absence of a TMDL  (background end-of-pipe).  EPA believes that setting limits
at background for discharges to non-attained waters is not an approach that
would be consistent with national program requirements under the CWA.  EPA
notes again that existing NPDES regulations require that, where a wasteload
allocation has not 'been prepared by a State and approved by EPA under 40 CFR
130.7, water quality-based effluent limits must ensure that the "level of
water quality to be achieved by limits on point sources established under this
paragraph is derived from, and complies with all applicable water quality
standards."  40 CFR 122.44(d)(1)(vii)(A).  In circumstances where a waterbody
is in non-attainment for a particular pollutant, EPA believes that  (with the
exception of certain discharges of intake pollutants allowed under procedure
5.D and E)  it would not be consistent with this provision to establish a WQBEL
allowing discharges of the pollutant at levels exceeding the most stringent
applicable water quality criterion.  On its face, EPA believes that a WQBEL
allowing discharges into a waterbody already exceeding such criteria would not
ensure that the water quality achieved by point sources was either  "derived
from" or "complies with" applicable water quality standards.  EPA also
believes that such a permitting approach would be fundamentally at odds with
the water quality-based permitting requirement contained in section
301(b)(1)(C) of the CWA, since such an approach would allow point sources to
contribute to the excursion above water quality standards in the waterbody.

3.    Consideration of Pollutants in Intake Water

a.    Introduction

      Appendix F, procedure 5.A-C, provides a means for permitting authorities
to determine if a discharge causes, has the reasonable potential to cause, or
contribute to an excursion above a State or Tribal numeric or narrative water
quality criterion.  These procedures require the permitting authority to
establish a water quality-based effluent limitation  (WQBEL) upon a
determination that a pollutant is or may be discharged at sufficient levels to
cause, have the reasonable potential to cause, or contribute to an excursion
above any Tier I criterion or Tier II value.

      The baseline .procedures for conducting "reasonable potential"
determinations in procedure 5.A-C do not provide special consideration for
pollutants contained in a facility's intake water.  Procedures 5.D and 5.E of
appendix F of the final Guidance provide separate mechanisms for considering
the presence of intake water pollutants in a facility's discharge when
determining the need for WQBELs and in establishing such limits.

      In some situations, the sole or primary origin of a pollutant in a
discharge may be the intake water for a facility. For example, the origin of
many pollutants in once through cooling water is the water body where the
facility obtains the water rather than an industrial process or other activity
of the facility itself.  Where the intake water contains pollutants at levels
that exceed water quality criteria, facilities which use and discharge that

-------
340    Water Quality Guidance for the Great Lakes System - Supplementary Information Document


intake water may face the need for WQBELS even where the facility otherwise
does not contribute that pollutant to the wastestream, unless special
consideration of the source of the pollutant is taken into account.  As has
been emphasized by States throughout development of the GLWQI, and many others
during the public comment period, a discharge of pollutants whose source is
the intake water is a special circumstance that warrants careful examination
in establishing regulatory controls.

      Intake pollutants are a particular concern in the Great Lakes region.
Concerns mentioned by numerous commenters include: elevated ambient levels of
"ubiquitous" pollutants like PCBs and mercury, as well as other common
pollutants of concern like copper, zinc, nickel and cadmium, whose continued
presence at levels of concern in waters that serve as water supplies is
believed to be due primarily to nonpoint sources of pollution such as
atmospheric deposition and contaminated sediments; relative lack of regulatory
control over nonpoint sources and ability to achieve reductions from nonpoint
sources sufficient to relieve point sources of additional reduction
responsibilities; increased stringency of water quality criteria, which will
increase the number of waters needing protection from pollutant sources; and
new analytical methods which may find previously undetected pollutants  that
exceed WQS.

k-    Existing Mechanisms

      EPA's NPDES permitting regulation at 40 CFR 122.45(g) currently provides
a mechanism for adjusting technology-based effluent limitations to account for
pollutants in a discharger's intake water in certain situations.  The
regulation provides that technology-based limitations shall be adjusted where
the applicable effluent limitations guidelines direct that limitations  be
applied on a net basis, or where the  discharger demonstrates that the presence
of intake water pollutants prevents compliance with the applicable technology-
based limitations despite proper installation and operation of the treatment
system(s).  The regulation also identifies four specific conditions
restricting the use of net credits:

       (l)  Net credits for generic or indicator pollutants are not allowed
unless the permittee demonstrates that the constituents of the generic  measure
in the effluent and influent are substantially similar or unless appropriate
additional limits are placed on process water pollutants.

       (2)  Credit may be granted only to the extent necessary to meet the
applicable technology-based  limitation, up to a maximum value equal to  the
influent value.

       (3)  Credit is generally limited to discharges to the same body of water
from which the intake water  is drawn  although the permitting authority  may
waive  this requirement if  no environmental degradation will result.

       (4)  Credit is precluded for  return of materials generated from the
treatment  of  intake, water  (e.g.,  raw  water  clarifier sludge.)

       The  provision granting credit only to  the extent necessary to achieve  a
technology-based limitation  assures that a  discharger uses the appropriate
technology-based level of  treatment in  removing pollutants that  originate from
the discharger's facility.   This  provision  in essence assures the proper
operation  of  treatment technology.

       When promulgating  the  net  credit  adjustment for technology-based  limits,
EPA declined  to  develop  a  similar mechanism to adjust water quality-based
effluent  limitations to  reflect  credit  for  intake water  pollutants.   EPA
explained  that  " [t]-he  Clean  Water Act's requirement  to protect  and enhance
water  quality is not 'conditioned on factors  such  as  intake water quality and
it would be inappropriate  for  EPA to  impose  such  a  condition.   Eligibility  for
a net  credit  under  these [technology-based]  regulations  does  not imply  any

-------
                           Section Vffl.E: Reasonable Potential                       341
right to violate water quality standards."  (49 FR 37998, 38027  (September 26,
1984)) .  EPA recognized the complexity of water quality-based permitting,
however, and indicated that permit writers may take the presence of intake
water pollutants into account, as appropriate, in individual permitting
decisions.  In all cases, EPA noted that permit limits "must be adequate to
meet the water quality objectives of the Clean Water Act when considered.along
with control requirements for other discharges to the stream."  (49 FR 38027;
September 26, 1984).  The existing mechanisms for simultaneously considering
control requirements for all dischargers to a single body'of water are total
maximum daily loads  (TMDLs) and NPDES permits written to implement TMDLs.
The preamble to the proposed Guidance at 58 FR 20954-6  (April 16, 1993)
described in detail the four mechanisms available under existing National
regulations and guidance which allow the permitting authority to determine
appropriate WQBELs when the receiving water exceeds a water quality criterion.
In addition to TMDLs, these mechanisms include:  temporary variances to WQS,
removal of non-existing uses, and site-specific modifications to water quality
criteria.  While each of these mechanisms directly or indirectly allows
consideration of intake pollutants in deriving permit limitations for
individual dischargers by adjusting the standards or wasteload allocations to
achieve standards, none is a permit-based mechanism  (i.e.,  based directly on
CWA requirements to include WQBELs in NPDES permits necessary to attain WQS).

c.    Summary of Proposal

      i.    Proposed Guidance

      In the proposed Guidance, EPA included procedure 5.E of appendix F to
provide a new procedure for considering the presence of intake water
pollutants in water quality-based permitting decisions in addition to the
available mechanisms described above.  The proposed Guidance would allow the
permitting authority to determine that there is no reasonable potential for
the discharge of a particular intake water pollutant to cause or contribute to
an excursion above a narrative or numeric water quality criterion, without
application of procedure 5.A-D of appendix F of the proposed Guidance, based
on the permittee's demonstration of specified conditions.  If these conditions
are demonstrated, .the permitting authority would not be required to include a
WQBEL for the pollutant in the facility's permit.  If these conditions are not
satisfied, the permitting authority would follow the baseline reasonable
potential procedures in proposed procedure 5.A-D of appendix F to determine
whether a WQBEL is necessary for these pollutants.

      Proposed procedure 5.E of appendix F provided a separate mechanism for
determining whether WQBELs are necessary for facilities that return unaltered
intake water pollutants to the source of the intake water.   The underlying
premise for this proposal, based on technical considerations of the nature of
pollutants and the effects of pollutants on surface water quality, is that
determinations whether a discharge of intake water pollutants should be
limited by a WQBEL and,  if so, the scope of such limitations, must be
determined after consideration of site-specific factors.  These factors
include consideration of the applicable water quality criteria, the quality of
the receiving water relative to the criteria,  additional pollutant loadings
from other point and nonpoint sources, and evaluation of the facility's
effluent.  As discussed further below, the effect of the discharge of intake
water pollutants may also vary substantially depending on the location of the
outfall in relation to the intake point, the time interval between intake and
discharge ,  alterations of the pollutant by the wastewater treatment process,
synergistic or additive interactions between the intake water and other
wastewater pollutants,  or the chemical nature of the pollutant in the
environment.  In proposing this new alternative, EPA acknowledged that States,
Tribes,  and dischargers had serious concerns about the feasibility of relying
on the existing mechanisms for adjusting standards and wasteload allocations
described above.

-------
342    Water Quality Guidance for the Great Lakes System - Supplementary Information Document


      EPA proposed procedure 5.E of appendix F as a reasonable mechanism for
evaluating the site-specific water quality effects from the discharge of
intake water pollutants.  This procedure allowed permitting authorities to
conclude that the return of unaltered intake water pollutants to the same body
of water under identified circumstances does not cause, have the reasonable
potential to cause, or contribute to an excursion above water quality
standards andA therefore, WQBELs for that pollutant were not needed.  The
permittee would be eligible for the reasonable potential procedure in proposed
procedure 5.E of appendix F upon demonstration of five conditions.

      -- First, the permittee would need to demonstrate that it withdraws 100
percent of the intake water containing the pollutant from the same body of
water into which the discharge is made.

      -- Second, the permittee would need to demonstrate that it does not
contribute any additional mass of the specified intake water pollutant to its
wastewater.   In other words, the pollutant present in the discharge must be
due solely to its presence in intake water from the receiving water body.

      -- Third, the permittee would need to demonstrate that it does not alter
the identified intake water pollutant chemically or physically in a manner
that would cause adverse water quality impacts to occur from the discharge
that would not occur if the pollutant were left in-stream.  Alterations  could
occur as long as they do not cause adverse water quality impacts.

      -- Fourth, trie permittee would need to demonstrate that the pollutant is
not concentrated at the edge of any available mixing zone after discharge from
the facility.

      -- Fifth, the permittee would need to demonstrate that the timing and
location of the effluent discharge do not cause adverse water quality impacts
to occur that would not occur if the pollutant were left in-stream.

      If the permittee demonstrated the five conditions to the satisfaction of
the permitting authority, the proposed procedure further identified three
conditions that the permitting authority  would have to address:   (l) the
permitting authority must summarize the basis for the determination that there
is no reasonable potential for the discharge of an identified intake water
pollutant to cause or contribute to an excursion above a narrative or numeric
water quality criterion within a State or Tribal WQS in the NPDES permit fact
sheet or statement of basis, including an evaluation of the permittee's
demonstration of the five specified conditions described above;  (2) the permit
must require all monitoring of the influent, effluent and ambient water
necessary to determine  that the conditions of procedure 5.E of appendix F are
maintained during  the permit term; and  (3) the permit must contain a reopener
clause authorizing the permitting authority to modify or revoke and reissue
the permit if new  information indicates that changes in any of the conditions
of procedure 5.E of appendix F have occurred.

      Finally, the proposed procedure addressed the relationship between the
option and any available wasteload allocation  (WLA) or TMDL prepared and
approved pursuant  to 40 CFR 130.7.  The proposed provisions of procedure 5.E
of appendix F stated that it would not alter the permitting authority's
existing obligation to  develop effluent limits consistent with the assumptions
and requirements of any WLA  (which is a part of a TMDL) that is developed and
approved in accordance  with 40 CFR 130.7.  (40 CFR 122.44(d)(1)(vii)).   The
preamble further explained that the proposed intake pollutant reasonable
potential procedure also would not alter a State's obligation to identify
water quality-limited segments and establish priorities for conducting  TMDLs
for those waters under  40 CFR 130.7.  The required evaluation of existing and
available water  quality data to make  these determinations under 40 CFR
130.7(b)(5) would  include consideration of the information submitted or
generated to  support permit decisions under procedure  5.E of appendix F.

-------
                           Section Vffl.E: Reasonable Potential                       343
      Procedure 5.E of appendix F was proposed as an alternative to the
reasonable potential procedures under proposed procedure 5.A-D of appendix F.
Under the proposal, ineligibility of a facility for the simple pass-through
determination of procedure 5.E of appendix F did not affect that facility's
ability to request 'the application of existing mechanisms for consideration of
intake water pollutants in setting WQBELs  (e.g., TMDL, a variance from water
quality standards, and modifications to designated uses and criteria),

ii.   Other Options

      In addition to the proposed alternative reasonable potential
determination for intake pollutants, the preamble to the proposed Guidance
discussed four other options EPA would consider in developing the final
Guidance.  Three of the four options, described briefly below, assumed that
WQBELS would be needed and addressed how intake pollutants could be taken into
account in establishing WQBELs.

      (A)   Option 1:  Option 1 reflected the current National approach.
EPA's existing regulations do not provide a specific mechanism to allow
special credit or consideration for pollutants present in a facility's intake
water when setting WQBELs comparable to the intake credit provision for
technology-based limitations at 40 CFR 122.45(g).  If the permitting  authority
determines that a facility's discharge causes,  has the reasonable potential to
cause, or contribute to an excursion above water quality standards for any
pollutant in the effluent, the NPDES permit must include an appropriate WQBEL
for that pollutant.-  Under Option 1, the permitting authority relied  on the
mechanisms available under EPA's existing regulations and guidance to derive
any WQBEL necessary to control discharges of pollutants to receiving  waters
that exceed water quality standards, including discharges containing  those
pollutants found in a discharger's intake water.  As noted previously, these
mechanisms include TMDLs, temporary variances from water quality standards,
and changes in the designated use of the water body, or site-specific criteria
modifications.

      (B)   Option 2:  Option 2 allowed the permitting authority to directly
modify WQBELs to reflect a credit for intake water pollutants if the
pollutants are discharged to the same body of water as the intake water.  A
specified level of credit was allowed under this approach even when the
facility contributes an additional amount of the intake water pollutant from
its process waste stream.  Basically, the discharger would be allowed to
discharge an amount of the pollutant equivalent to the mass of the pollutant
in its intake water.  Removal of the pollutant from the intake water  could be
offset by increases in the amount added through the facility's process water
or other activities.  However, credit was precluded under Option 2 if the
facility failed to demonstrate the remaining conditions specified in  section
5.E.I.a, c, d, and e of the proposed intake pollutant reasonable potential
procedure.  This option is commonly called the "no net addition" option (and
was referred to as Option 2a in the proposal).

      The preamble to the proposed Guidance also discussed a variation to the
basic "no net addition" option.  Option 2b allowed a facility to discharge an
effluent containing, at a maximum, the same mass of the pollutant in  the
intake water after deduction of the amount removed by the facility's  raw
water, or intake water treatment system .   Thus, if a facility removed any of
the pollutant originating in the intake water prior to use at the facility,
the facility would not be able to offset this reduction.  If a facility
removed any of the intake water pollutant during the wastewater treatment
process, however,  it would be able to provide a commensurate increase in the
amount of the pollutant contributed in the process wastewater.

      As with the proposal, neither Option 2a nor 2b altered the authority of
the  permitting authority to develop WQBELs to account for the presence of
intake water pollutants pursuant to a TMDL, temporary variance, or other
allowable modifications to WQS pursuant to State and EPA regulations  and the

-------
344    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


proposed Guidance.  The proposed options 2a and 2b did not authorize
exceedance of a TMDL or alter a State's obligation to identify water quality-
limited segments and establish priorities for conducting TMDLs for those
waters under 40 CFR 130.7.  EPA specifically asked for comment on whether this
option should be limited to one permit term to encourage timely  completion and
implementation of TMDLs.

       (C)    Option 3:   Option 3 allowed the permitting authority to directly
modify WQBELs to reflect a credit for intake water pollutants regardless where
the intake water source is located.  In other respects, Option 3 was similar
to Option 2.  The preamble discussed three variations to this Option.

      Option 3a allowed a facility to discharge an effluent containing, at a
maximum, the same mass of the pollutant that the facility receives from any
water source including sources other than the receiving water.   If a facility
removed any of the pollutant from the intake water, the facility could  offset
this reduction by increasing the amount of the pollutant contributed by the
process wastewater.

      Option 3b  allowed a facility to discharge an effluent containing, at a
maximum, the same mass of the pollutant contained in intake water from  any
source after deduction of the amount removed by the facility's intake water
treatment system   (analogous to Option 2a).  If a facility removed any  of the
pollutant originating in any intake water through wastewater treatment, the
facility could similarly offset this reduction by increasing the amount of the
pollutant contributed by the process wastewater.

      Option 3c allowed a facility to discharge an effluent containing, at a
maximum, the same concentration of the pollutant that is present in the
receiving water.  If a facility removed any of the pollutant from the intake
water, the facility could offset this reduction by increasing the amount of
the pollutant contributed by the process wastewater.

      Like the proposal and Options 2a and 2b, Options 3a, 3b, and 3c did not
alter the authority of the permitting authority to develop WQBELs to account
for the presence of intake water pollutants pursuant to a TMDL,  temporary
variance,  or other allowable modifications to WQS pursuant to State,Tribal or
EPA regulations and the proposed Guidance.  If a TMDL were developed, effluent
limitations derived using Option 3 would need to be adjusted to  be consistent
with the TMDL.

       (D)    Option 4:  The fourth Option discussed in the preamble to the
proposed Guidance was the initial procedure developed by the Great Lakes
Technical Work Group and was originally included in the Great Lakes
Implementation Guidance as part 11.B  (which became, in part, proposed
procedure 3 of appendix F to part 132 controlling development of TMDLs).
Option 4 represented a combination of Options 2a and 3c.  In addition,  it
applied only when the background level of pollutant in the receiving water
exceeded an applicable water quality criterion.  Like Options 2  and 3  (and
variations), Option 4 required adjustment to limits to ensure consistency with
any applicable TMDL.  The procedure provided mechanisms for accounting  for
pollutants in a facility's intake water under two circumstances.

      First, when at least 90 percent of the intake water source is from
ground water  (except ground water withdrawn from a location of contaminated
ground water) or a public drinking water supply, Option 4 allowed a facility
to discharge an effluent containing a concentration of a pollutant ranging
from, at the low end, the water quality criterion, to, at the high end, the
concentration of the pollutant in the receiving water.  The permitting
authority would use its professional judgment and consider  reasonable,
practical, and otherwise required methods to minimize addition of toxics in
deciding where to establish the effluent limitation within the specified
range.

-------
                           Section VID.E: Reasonable Potential                       345
      Second, when a minimum of ten percent of wastewater is from the same
waterbody into which the effluent is discharged, Option 4 allowed a facility
to discharge an effluent at a concentration equal to the receiving water or
containing a mass of a pollutant equal to the mass the facility receives from
the waterbody.  This Option would apply even if 90 percent of the wastewater
was from the process waste stream or from waters other than the receiving
stream.  POTWs which discharge to the same surface water from which the public
water supply is withdrawn would be covered under this provision as well.

d.    Structure of Remaining Discussion

      Section 4 below addresses the key features of the Final Guidance and,
where applicable, how they differ from the proposed Guidance, including a
discussion of and response to major comments.  Section 5 addresses EPA's legal
authority to regulate intake pollutants.  Section 6 explains the relationship
between the intake pollutant provisions in today's final Guidance and existing
mechanisms to adjust WQBELs.  Section 7 addresses other, more detailed aspects
of the final Guidance in a similar fashion.  In addition, a separate response
to comment document, which has been included in the docket for the final
Guidance, contains responses to all comments that were received on the intake
pollutant procedure in the proposed Guidance, some of which may not be
specifically addressed in this document.

4.    Summary of Intake Pollutant Considerations in Final Guidance and Overall
      Rationale

      EPA has carefully considered the numerous comments received on the
intake credit issue in establishing new mechanisms for considering intake
pollutants in water quality-based permitting.  Like the proposal, the final
Guidance includes a procedure for considering intake pollutants from the same
body of water into which the effluent is discharged when evaluating whether a
discharge causes, has the reasonable potential to cause, or contributes to an
excursion above applicable water quality standards.  Unlike the proposal, the
final Guidance also  establishes procedures for considering intake pollutants
in developing WQBELs.  The consideration of intake pollutants arises in three
factual scenarios, and the final Guidance addresses each of the following
circumstances:   (1) when the source of the intake pollutants is the same body
of water as the receiving water for the discharge; (2) when the source of the
intake pollutants is a different body of water than the receiving water for
the discharge; and  (3) when the discharger has multiple sources of intake
water that contain the identified pollutant of concern.  These provisions
appear under procedure 5.E of appendix F.  In addition, the final Guidance
defines "same body of water."

a.    Intake Pollutant Reasonable Potential Procedure

      As described above, the proposed Guidance included a separate procedure
5.E of appendix F for considering intake water pollutants when determining
whether a discharge causes, has the reasonable potential to cause, or
contributes to an excursion above a water quality standard and therefore
needed a WQBEL in the permit ("intake pollutant reasonable potential"
procedure).    The proposal generated numerous comments as to the details of
the procedure and most significantly, to EPA's decision not to  propose an
additional procedure to adjust WQBELs directly for intake pollutants.
However, most commenters supported the need for a reasonable potential
procedure that would account for intake pollutants.  Support was not
universal.   Some commenters objected, asserting that TMDLs were the only
mechanism authorized by the CWA for allocating loads when waters do not meet
standards and that in the absence of a TMDL, discharge of the pollutant(s)
exceeding standards either should be prohibited or limited to the most
stringent applicable criteria or value.

      The final Guidance includes a procedure 5.D of appendix F which is
essentially the same as the proposal and allows the permitting authority to

-------
346    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

determine that a WQBEL is not needed for a particular pollutant when a
discharger returns unaltered intake pollutants to the same body of water, and
the discharge does not cause adverse water quality effects that would not have
otherwise occurred if the pollutants were left in-stream.  This procedure
represents a comprehensive approach for conducting a site-specific analysis of
the potential for a discharge to cause or contribute to an excursion above a
water quality standard, which can lead to a decision not to require a WQBEL.

b.    No Net Addition Interim Approach for Setting WOBELs

      In response to numerous comments and as explained in more detail  in
section 5.b below, the final Guidance also provides for a direct adjustment to
WQBELs when the discharge contains intake pollutants from the same body of
water as the discharge and the other requirements for the intake pollutant
reasonable potential demonstration are met.  The only difference between the
no net addition procedure and the intake pollutant reasonable potential
procedure is that, under the no net addition procedure, a discharger may add
mass of the identified pollutant of concern to its wastestream as long  as its
discharge contains no more mass of the pollutant than was contained in  the
intake water.  In a sense, the permittee must demonstrate what may be viewed
as the functional equivalent of the intake pollutant reasonable potential
determination, or simple pass through situation because this procedure
establishes WQBELs at a level which ensures the discharge has no greater
impact on the receiving water than if the discharger had not removed and
returned the intake pollutants to the same body of water.

      A major concern expressed by EPA in the proposal about options that
would provide for direct adjustment of permit limits for intake pollutants
when the discharger contributes additional amounts of the pollutant of  concern
to its wastestream, and the receiving water exceeds the criteria for that
pollutant, was that the availability of a permit-based mechanism would
discourage development of TMDLs that meet the requirements of, and are
approved or established by EPA in accordance with, 40 CFR 130.7.  As discussed
in section VIII.E above, TMDLs are a. mechanism for determining the
assimilative capacity of a waterbody and fairly allocating that capacity among
sources of a pollutant.  CWA section 303(d); 40 CFR 130.7.  Using a permit-
based adjustment for intake pollutants imposes an allocation scheme without
the assurance that other sources contributing to an exceedance of a water
quality standard will be controlled so that the standard is attained, as would
be the case with a TMDL.

      While EPA does not regard TMDLs as the only available mechanism under
the CWA for adjusting point source controls, TMDLs are clearly the preferred
mechanism for determining the appropriate load allocation scheme to ensure
that a waterbody which does not meet  standards is brought into attainment of
WQS, particularly when multiple sources contribute the pollutant that exceeds
water quality criteria.  Indeed, section 303(d) of the CWA requires States to
develop TMDLs for waters that are not expected to meet water quality standards
despite implementation of technology-based requirements or other existing and
planned controls.  To address the concern that providing a direct adjustment
to permit limits for intake pollutants would discourage development of  TMDLs,
EPA solicited comments on whether allowing a "no net addition" or other
approach for developing WQBELs should be limited to one permit term  (i.e.,
five years). In addition, EPA requested comment on the reasonableness of using
one permit  term as the maximum duration of relief in this case.

      Commenters who specifically addressed this issue supported intake
credits based on the premise that dischargers should not be held responsible
for the intake pollutants  in their discharges.  They similarly opposed
limiting intake pollutant  relief  to one permit term.  On the other hand,
commenters  who opposed  intake credits asserted that TMDLs should be required
from the outset.  Commenters supporting intake credits asserted  that intake
credits should be available as long as the  site conditions leading to the need
for intake  credits -existed and that dischargers should not be penalized for a

-------
                           Section VEQ.E: Reasonable Potential                       347
State's failure to develop a TMDL.  In commenting more generally on whether a
TMDL could be an adequate substitute for a permit-based approach for
considering intake pollutants, commenters expressed skepticism  whether TMDLs
and, by implication, reductions of the intake pollutant from other sources
contributing to background levels specified in the TMDL, could be completed in
time to result in limits at or above the background level for a downstream
discharger, particularly where pollutant reductions are needed from nonpoint
or natural sources.

      Consistent with EPA's determination that TMDLs are the preferred
mechanism for developing control strategies to bring impaired waters into
attainment of WQS, the final Guidance limits the time period for which a
permitting authority can use direct adjustment of permit limits for intake
pollutants.  Under procedure 5.E.2.a of the final Guidance, no net addition
limitations may be included in permits for discharges of intake pollutants for
a period of up to twelve years after publication of the final Guidance.  After
that period of time, WQBELs would be based either on:   (1) the "baseline"
procedures for developing WLAs in procedure 5.F.2 of appendix F; or, when
available, (2) WLAs in TMDLs developed and approved in accordance with 40 CFR
130.7; or  (3) WLAs .contained in an assessment and remediation plan for the
receiving water submitted by the State or Tribe and approved by EPA as meeting
the requirements of procedure 3.B-F, as provided in procedure 3.A of appendix
F and section D.l.c of intake pollutant procedures.  EPA remains concerned
that the unlimited availability of intake credits through the permitting
program could discourage the development of TMDLs where they may be most
needed, i.e., for non-attained waters where multiple sources contribute to the
exceedance of water quality standards.  EPA recognizes that point source
dischargers are understandably concerned when their discharge control
requirements depend in part on contributions from other sources and,
consequently, on the ability and willingness of regulatory agencies and other
sources to reduce cither loadings that impact water quality.  However, an
unalterable fact of water quality-based controls is that they will vary
depending on the particular condition of the receiving water.  The TMDL
provides a specific mechanism for determining how best to achieve water
quality objectives considering such factors as the ability to achieve
reductions from other sources.

      EPA has included this time limitation on the availability of no net
addition limitations in order to help ensure that States and Tribes are
evaluating comprehensively the root sources of non-attainment in the receiving
water.  As noted previously, EPA believes that, in light of the unique
considerations posed by discharges of intake pollutants from the same body
water, no net addition limitations are appropriate as an interim measure to
help ensure that such point source discharges are not aggravating a
waterbody's non-attainment problem.  As discussed below, however, such non-
attainment is not an acceptable long-term status in light of the requirements
and goals of the CWA.  Ultimately, it will be critical for EPA, the States and
Tribes to address comprehensively the sources of non-attainment in these
waterbodies.   When these problems are properly addressed, of course, the need
for a specialized permitting mechanism to address intake pollutants will no
longer be necessary.

      As emphasized by EPA in the proposal and elsewhere in the SID, the best
means for States and Tribes to address comprehensively the root causes of non-
attainment is the TMDL development process.  EPA recognizes, however, that
States and Tribes may seek to address comprehensively point and nonpoint
source contamination in a waterbody through means other than TMDLs.As
explained in section VIII.C above, which discusses procedure 3.A of appendix
F, EPA recognizes that alternative assessment and remediation plans that meet
the requirements of procedure 3 of appendix F can serve the same purpose as
TMDLs in providing a holistic assessment of all sources contributing toward a
particular water quality impairment problems, identify remediation activities
which are reasonably anticipated to result in load reductions that will
achieve attainment of WQS within a reasonable period of time, and establishing

-------
348    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


wasteload allocations for point sources that are consistent with these water
quality objectives.  Therefore, EPA believes that it is appropriate for the
permitting authority to have the flexibility to continue to establish no net
addition limitations provided that they are consistent with assessment and
remediation plans submitted and approved in accordance with procedure 3.A of
appendix F.  As described in section  I.D.4 of this document, the States and
Regions in the Great Lakes basin have undertaken significant assessment and
remediation planning efforts through  the development of RAPs and LaMPs.  Where
such assessments and remediation plans are approved by EPA in accordance with
procedure 3 .A of appendix F, EPA believes that the State or Tribe has
demonstrated that non-attainment will not be the indefinite status quo of the
waterbody.  Under these circumstances, EPA believes, as a matter of policy,
that WLAs consistent with those alternative plans, including those that result
in no net addition limits, can be an  appropriate permitting approach.

      For discharges of pollutants for which intake pollutants are not at
issue, wasteload allocations would continue to be derived in accordance with
procedure 5.F.2 or an EPA approved TMDL or assessment and remediation plan
under procedure 3.A of appendix F, where applicable.  While commenters
addressed generally whether a time limitation on the availability of intake
credits was appropriate, none specifically addressed whether one permit term
would be a reasonable period of time  or suggested specific alternative
durations.  Upon further consideration of the effort that might be needed to
develop TMDLs or comparable assessment and remediation plans in the Great
Lakes region, particularly considering that more waters may be in non-
attainment based on new water quality criteria adopted pursuant to the final
Guidance, EPA has concluded that a one permit term limitation would be unduly
restrictive.  Therefore, the final Guidance adopts a time limit, which is
based on ten years from when States must adopt the Guidance, i.e., 12 years
from the publication date of the final Guidance  (approximately two permit
terms).  At the end of this time period, intake pollutant relief in the form
of permit limits based on no net addition will no longer be available, unless
the limits are consistent with wasteload allocations in an EPA-approved or
prepared TMDL under 40 CFR 130.7 or an alternative assessment and remediation
plan as provided in procedure 3.A.  It is important to note that wasteload
allocations in a. TMDL, and resulting  permit limits, may be more or less
stringent than previously allowed under a no net addition approach for
establishing permit limits.  Because  TMDLs or comparable assessment and
remediation plans are designed to ensure attainment of water quality standards
in the waterbody, such attainment would eliminate the need for special
consideration of intake pollutants.

      EPA recognizes that developing  TMDLs may be a difficult task in the
Great Lakes region because water quality problems are widespread, result from
numerous sources, many of which have  not been extensively regulated, and may
be due to past, rather than current,  activities.  However, the States and
Tribes in the Great Lakes region have already undertaken many of the tasks
necessary to develop TMDLs through efforts to develop LAMPs and RAPs, as
described in section I.D.4 of this document.  The final Guidance specifically
recognizes these efforts by allowing  them to serve in lieu of TMDLs when they
meet the fundamental elements of a TMDL as established in procedure 3.  The
final Guidance alsd recognizes, in procedure 3.A of appendix F, that
comprehensive plans for assessing and remediating non-attainment waters should
be tailored in the level of detail and magnitude for the watershed and
pollutant being assessed.

      In response to commenters' concerns about the difficulties in developing
TMDLs in the near term, it is important to realize that under the phased
approach, TMDLs can be  developed in the absence of complete information and do
not necessarily require full implementation of all necessary controls before
adjustments to wasteload allocations  for point source contributors can be made
available.  See procedure  3.B.I of appendix F, which acknowledges that it may
not be possible to attain WQS immediately in all cases and that  specific
controls on individual  sources may need to be implemented in stages. The

-------
                           Section Vffl.E: Reasonable Potential                       349
phased approach to TMDLs is described in detail in section VIII.C above.
Further, as discussed in more detail in section VIII.E above on the baseline
reasonable potential procedures, unlike the proposed Guidance, the final
Guidance does not require States to develop a TMDL every time the permitting
authority determines that an individual discharger causes, has the reasonable
potential to cause, or contribute to an excursion above a water quality
standard.  Thus, States retain flexibility in determining priorities for TMDL
development, consistent with section 303(d) of the CWA and 40 CFR 130.7.  The
final Guidance provides additional flexibility by allowing States or Tribes to
develop appropriate approaches for comprehensively addressing non-attainment
in accordance with mechanisms comparable to TMDL  (e.g., LaMPs) and use these
approaches as an alternative basis for developing WLAs when intake credits are
not available  (e.g.,  when the 12-year period allowed for "no net addition"
limits ends).  Because of this flexibility in the timing of TMDLs,  the
efforts currently underway to identify the sources contributing to impairment
of waters, and the provision recognizing mechanisms comparable to TMDLs, EPA
believes that 12 years is a reasonable period of time for States and Tribes to
develop TMDLs or comparable approaches.

      The final Guidance limits how long permit-based intake pollutant credits
are available, but consideration of intake pollutants in setting limits does
not necessarily end on March 23, 2007.  Within the context of the TMDL or
comparable assessment and remediation plan, States and Tribes have
considerable discretion to determine how the necessary loading reductions will
be achieved.  Thus, States and Tribes can continue to provide for
consideration of intake water pollutants in establishing wasteload allocations
as part of a TMDL or comparable mechanism, provided that the requirements for
a TMDL in 40 CFR 130.7 are met  (or in the case of a comparable mechanism, the
requirements established in procedure 3.A of appendix F are met), most
fundamentally the requirement to show that water quality standards will be
attained through a combination of load and wasteload allocations, together
with a margin of safety.  In addition, dischargers may seek adjustment of
permit limits through other available mechanisms such as a temporary variance
from water quality standards, as addressed in procedure 2 of appendix F and
discussed more fully in section VIII.B of this document.

c.    Consideration of Intake Pollutants from a Different Body of Water

      The proposed Guidance limited intake pollutant relief to instances where
the source of the intake pollutant was the same body of water as the receiving
water for the discharge.  EPA explained that the "same body of water"
restriction was appropriate, to ensure consistency with the structure and
function of State or Tribal water quality standards.  Without such a
restriction, dischargers could transfer pollutants from one waterbody to
another without determining whether the discharge causes, has the reasonable
potential to cause, or contribute to an excursion above an applicable water
quality criteria based on consideration of site-specific factors, including
the condition of the receiving water and contributions by other point or non-
point sources.

      The preamble to the proposed Guidance also discussed options that would
allow intake credits when pollutants were transferred from one waterbody to
another.  Option 3 allowed consideration of intake water pollutants without
regard to waterbody source and based limits on one of two approaches.  The
first approach  (Options 3a and 3b)  based limits on the amount of the pollutant
in all sources of intake water,  with the further option of not extending
credit for pollutants removed from the intake water before use at the
facility.  This approach could result in improved water quality, but could
also allow further degradation of the receiving water, depending on the
quality of the intake water relative to the receiving water.  The second
approach  (Option 3c)  based limits on the concentration of the pollutant in
the receiving water.  At best, this option would maintain the status quo.
Option 4 used a combination of approaches.  Under Option 4,  where the
discharge was to a different body of water, limits would be set at the water

-------
350    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

quality criteria of the receiving water and could be adjusted up to the
concentration of the pollutant in the receiving water if the permittee
demonstrated that the concentration of the substance at the point of  intake
exceeds the applicable standard or criterion for that substance and that
reasonable, practical or otherwise required methods are implemented to
minimize the addition of the toxic substance to the wastewater.

      Several commenters stated that intake credit relief should be limited to
the same body of water and if interbody transfers were allowed, the discharge
should have to meet the standards of the receiving stream.  EPA also  received
numerous comments supporting intake credits when the source of the intake
pollutant is a different body of water.  Some commenters favored relief in
this situation based on the general notion that dischargers should not be
responsible for pollutants originating outside their facilities regardless of
the impact on the receiving water.  Others, however, endorsed the more  limited
rationale that interbody transfers of pollutants should be allowed if the
discharge would substantially maintain or improve the existing receiving water
quality in terms of the concentration of the pollutant in the waterbody.
Several commenters supported basing limits on receiving water quality or in
instances of net environmental improvement because the "technical work group
[composed of the Great Lakes States] could not scientifically conclude whether
adding water from another source with lower pollutant concentration helped or
hurt the Great Lakes."  One commenter suggested dropping the distinction
between same and different body of water and advocated instead a requirement
that the source and receiving waters have similar concentrations of the
pollutant and apply other restrictions to ensure protection of the receiving
water from other possible adverse effects of the discharge  (e.g., timing and
location, chemical and physical alternation, etc.)  Others claimed that the
distinction was irrelevant because of the application of uniform water  quality
criteria throughout the basin.  Finally, several commenters stated that the
distinction between same and different bodies of water in the proposal would
unfairly and unnecessarily preclude relief in the many instances where  a
discharger had multiple sources of intake water.

      The final Guidance retains a distinction between situations where the
source of the intake water pollutant  is the same or different body of water as
the waterbody receiving the discharge.  (See related discussion on the
definition of "same body of water" in section VIII.E.7.iv below.)  As
explained below in the discussion of EPA's legal authority, EPA does  not agree
with commenters who argued that the CWA  does not authorize regulation  of
pollutants in a discharge that do not originate with the discharger.  EPA
maintains, moreover, that requiring consideration of the whole discharge in
evaluating whether water quality based limits are needed is  essential, as a
technical matter.  The distinction between same and different bodies  of water
ensures that each time a particular mass of a pollutant is introduced into a
waterbody for the first time  (i.e., would not otherwise be in the waterbody
but for the discharge), its impact is evaluated to determine whether  the
discharge of the pollutant would cause, have the reasonable potential to
cause, or contribute to an exceedance of the water quality standards
applicable to the receiving water, and it is controlled through  limits  that
implement water quality standards  where appropriate.  Similarly, where WQBELs
are found to be necessary, EPA believes that direct adjustment of limits to
account for pollutants  in the intake  water should be restricted  to those
pollutants that would be in the receiving water with the same effect  even if
the discharger had not withdrawn and  subsequently discharged those pollutants.

      EPA recognizes that in  some instances discharges from other bodies of
water that exceed applicable water quality  criteria for the receiving water
but have a lower concentration of the pollutant than the receiving water
could, theoretically,  improve the overall water quality from the standpoint of
water column concentrations in the receiving water.  These  instances  could
occur if a facility discharges  a lower concentration of a pollutant  than is
present  in  the receiving water.  Although the resulting ambient  concentration
could be lower,  the mass of a pollutant would increase by the transfer  of

-------
                           Section Vffl.E: Reasonable Potential                       351
pollutants to a different body of water.  Whether this discharge would  result
in overall water quality improvements would depend on several factors
including, the magnitude of the actual decrease in the pollutant concentration
in the water column, the lowered concentration in the water column as compared
to the water quality criterion, consideration of the factors identified under
the proposed procedure 5.E of appendix F  (e.g., alteration of the pollutant,
concentration at the edge of any applicable mixing zone, and time and location
of the discharge), the impacts of the additional mass on pollutant levels in
sediment and fish tissue, and the transfer of the additional mass through
volatilization and sedimentation into nonpoint sources of atmospheric
deposition and sediment resuspension.  In particular, the additional mass of a
persistent pollutant may offset the environmental benefits of lowering  water
column concentrations because the additional mass, if cycled through sediments
by deposition and resuspension or through the food chain, could negatively
impact the waterbody so as to ultimately prolong the non-attainment status of
the waterbody.  In other words, whether additions of mass, considered alone,
will further degrade  a receiving water, needs to be considered on a site-
specific basis taking into account a comprehensive evaluation of the state of
the waterbody, its pollutant sources, and the fate and effect of pollutants
within the waterbody, which is the type of evaluation that occurs during
development of a TMDL or comparable assessment and remediation plant as
provided in procedure 3A.

      In the absence of a TMDL or comparable mechanism appropriately
considering these factors, EPA is not comfortable with concluding
categorically, as advocated by some commenters, that a discharge is by
definition environmentally acceptable if it will not cause an increase  the
concentration of a pollutant in the receiving water.  Without such an analysis
as part of the TMDL or comparable process, therefore, EPA believes that there
is not a sound scientific basis to conclude that such a discharge would never
cause, or have the reasonable to cause, or contribute to an excursion above
water quality standards.  EPA believes, moreover, that such an approach would
be inconsistent with the language and structure of the CWA, under which States
or Tribes establish water quality criteria to ensure that designated uses are
protected.  If, as some commenters asserted, the discharge of a pollutant does
not pose "reasonable potential" provided it will not increase the
concentration of the pollutant in the receiving water, then water quality
criteria would be entirely irrelevant to the permitting process.  EPA believes
such a result cannot be reconciled with the CWA. See CWA 301(b)(1)(C) and 40
CFR 122.44(d).

      In the absence of a TMDL or comparable assessment and remediation plan
approved under procedure 3.A of appendix F, the permitting authority needs to
determine whether the discharge causes, has the reasonable potential to cause,
or contribute to an excursion above an applicable WQS, and therefore should
have a WQBEL.  Evaluating such discharges using the procedures in 5.A-C
adequately considers the impact of the discharge on the receiving water.
Where "reasonable potential" exists, the permitting authority must develop
limits that implement water quality standards.

      EPA agrees that maintaining or providing a net improvement to receiving
water is preferable to further degradation of the receiving water, but  does
not agree that "net improvement" is the appropriate standard for deriving
WQBELs once "reasonable potential" has been established.  Similarly, EPA does
not agree that simply maintaining the status quo, as would be the case  when
limits are based on the background water quality of the receiving water, is
appropriate when the source of the pollutants is a different body of water,
even if all other requirements applicable to the intake pollutant reasonable
potential or "no net addition" approaches for the discharge of the intake
water pollutants from the same body of water are met.  Instead,  EPA
regulations at 40 CFR 122.44(d)(1)(vii) require that water quality-based
permit limits be derived from and comply with all applicable water quality
standards.  Also see. CWA 301(b)(1)(C).  The fundamental basis for providing
special allowance for intake pollutants in the same body of water situation,

-------
352    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

i.e., that the pollutant would have reached the vicinity of the discharge
point despite its removal and subsequent discharge, does not exist when the
discharge is from a different body of water.  Therefore, procedure 5.E.4 of
appendix F of the final Guidance provides that, when the permitting authority
finds that reasonable potential exists  (using the baseline reasonable
potential procedures in 5.A-C), discharges that contain intake pollutants from
other bodies bf water must meet water quality criteria end-of-pipe when the
receiving water exceeds the criteria for that pollutant.  In essence, the
final Guidance denies the special consideration for intake water pollutants
from a different body of water that has been adopted for intake pollutants
from the same body of water.  EPA's rationale for determining that criteria
end-of-pipe, as an interim basis for setting effluent limits in the absence of
a TMDL or comparable assessment as provided in procedure 3.A of appendix F,
meets CWA and NPDES regulation requirements for WQBELs is explained in section
VIII.E.2.h of this document.  States and Tribes of course, may require more
stringent limits.

      Using water quality criteria end-of-pipe as the basis for permit limits
when the intake pollutant originates in a different body of water than the
discharge, is consistent in part, with Option 4.  That option, however, would
provide for less stringent limits--up to the concentration of the pollutant in
the receiving water--if the intake water exceeded criteria and the permittee
could demonstrate that  it has implemented reasonable, practical or otherwise
required methods to minimize the addition of the toxic substance to the
wastewater.  In effect, Option 4 would substitute the feasibility of pollution
control for consideration of water quality  standards as the basis for deriving
WQBELs.  Comments in support of Option 4 suggest that it could be read to
require simply that the discharger has implemented technology-based controls.
EPA does not  believe,  however, that this approach adequately implements CWA
requirements to require limits "more stringent" than technology-based limits
when necessary to implement State water quality standards.  See CWA section
301(b)(1)(C).

d.    Combined Approach for Multiple Intake Sources

      The proposed Guidance provided for consideration of intake water
pollutants only when all such pollutants in the discharge were from the same
body of water as the receiving water.  As noted above, some commenters
objected, asserting that the proposal would unnecessarily deny relief where a
facility has multiple sources of  intake water which contains the pollutant of
concern.  Some commenters  specifically endorsed a  "combined wastestream"
approach that would use flow-weighted averages to develop an end-of-pipe limit
when a facility had multiple sources of intake water.  EPA agrees that  it
would be reasonable to  provide for a combination approach using flow-weighted
averages.  Accordingly, the  final Guidance  specifically provides the
permitting  authority discretion to develop  limits using a  "combined
wastestream" approach.  Under procedure 5.E.5 of appendix F, a permitting
authority could  develop limits by applying  the no net addition approach
established in procedure 5.E.3 of appendix  F, for that portion of the
wastestream containing  intake pollutants from the  same body of water, and
applying the criteria end-of-pipe approach  established  in procedure  5.E.4  of
appendix F  to that portion of the wastestream containing intake pollutants
from a different body of water, and use flow-weighted averages to develop  end-
of-pipe limits.  The permitting authority has the  discretion  to not  use this
approach  if it determined  that development  of  such  limits, or appropriate
compliance  monitoring,  is  infeasible.  In addition,  State or Tribes may  impose
more stringent requirements  pursuant to CWA section 510.

       The discussion  in section  7.a.v  below, addressing the requirement that
100  percent of the  intake  pollutant be  from the  same body  of  water  as  the
discharge as a condition for relief under the  intake pollutant  reasonable
potential procedure and for no net addition limits,  also discusses multiple
sources of  pollutants  in a facility's  discharge.

-------
                           Section Vm.E: Reasonable Potential                       353
5.    Legal Authority

      EPA received more comments on its authority to regulate intake
pollutants than any other aspect of the intake pollutant procedures in the
proposed Guidance.  While some agreed.that EPA had authority to regulate any
pollutants in a discharge (and further argued that EPA did not have authority
to provide any special allowances for intake pollutants in water quality-based
permitting), many commenters who addressed this issue maintained that
discharge of intake pollutants was not an "addition" of a pollutant as defined
in the CWA and therefore fell outside of EPA's jurisdiction under the NPDES
program.  Others argued that even if EPA had the statutory authority to
regulate intake pollutants,  the proposal violated EPA's existing regulations,
which require WQBELs only when the discharge causes, has the reasonable
potential to cause, or contributes to an excursion above applicable water
quality standards.

      Evaluating options responsive to commenters' concerns led EPA to examine
the appropriate basis for determining the responsibility of a point source
discharger when its receiving water exceeds criteria for a pollutant due
largely to other sources, its discharge contains that pollutant, and the
pollutant originates, at least in part, in the discharger's water supply.
Section 101(a) of t;he CWA establishes the overall objective of the CWA to
restoreiand maintain the chemical, physical,and biological integrity of the
Nation's waters.  The CWA also establishes a comprehensive scheme for
designating uses of surface waters and establishing criteria for pollutants
that ensure attainment of those uses  (section 303(a)-(c)).  Section 303(d)
further requires States to identify waters that will not meet water quality
standards after implementation of technology-based controls on point source
discharges, establish a priority ranking of those waters, and in accordance
with the ranking, develop TMDLs which establish the pollutant level that
cannot be exceeded in order to protect the applicable water quality standard
for that pollutant.

      Despite increased recognition of the impact of nonpoint sources on water
quality, the CWA scheme for regulating sources of pollutants that can impair
water quality focuses on point source controls.  Sections 301(a) and 402 of
the CWA require limits in NPDES permits for point sources more stringent than
technology-based limits when necessary to meet water quality standards.  The
CWA is silent on how compliance with limits "necessary to meet   water quality
standards"  (CWA section 301(b)(1)(C)) should be determined for a particular
discharger, particularly when controls on other sources contributing
pollutants to the same waterbody may be necessary to attain standards in the
waterbody.  While TMDLs provide a mechanism for allocating loadings among all
sources in determining how to attain water quality standards, the CWA does not
specify how those other sources are to be considered in developing permit
limits when a TMDL does not exist.

a.    Discharge of Intake Pollutants is an Addition of Pollutants Under the
      CWA

      Some commenters objected to EPA's position articulated in the preamble
to the proposed Guidance that release of intake pollutants into waters of the
United States, where those pollutants had previously been removed from such
waters, constitutes an "addition" of pollutants subject to regulation under
the CWA.  Commenters asserted that relevant case law (in particular NWF v.
Gorsuch. 693 F.2d 580 (D.C.Cir.  1982), NWF v. Consumers Power Co.. 862 F.2d
580 (6th Cir. 1988), Appalachian Power Co. v. Train. 545 F.2d 1351  (4th Cir.
1976))  establishes that such activities do not come within the scope of CWA
jurisdiction.  Some commenters also argued that the release of intake
pollutants does not constitute an addition of the pollutants provided that the
concentration of the pollutants in the facility's effluent is no greater than
the concentration in the waterbody.  In the view of these commenters, EPA must
look to the environmental effects of a particular discharge to determine
whether it is an addition of pollutants.  Citing cases relied upon by EPA in

-------
354    Water Quality Guidance for the Great Lakes System ~ Supplementary Information Document


the proposal such as U.S. v. M.C.C. of.Florida. Inc.. 772 F.2d  1501,  1506
(llth Cir. 1985) and Avovelles Sportsmen's League v. Marsh. 715  F.2d  897,  923-
924 (5th Cir. 1983), these commenters contended that a discharge of pollutants
must necessarily be accompanied by a fundamental disturbance or  substantial
alteration to the receiving water in order to be subject to regulation  under
the CWA.

      EPA's basic position, detailed in the preamble to the proposed  Guidance
at 58 FR 20956-7  (April 16, 1993), that discharge of any pollutant, including
those in  a facility's intake water, is an addition of pollutants within the
meaning of the Clean Water Act, remains unchanged.  In EPA's view, the
commenters' assertion that the statute narrowly circumscribes the EPA's
discretion to interpret "addition" to include the discharge of intake
pollutants is without  support in the CWA.  Several of the cases principally
relied upon by these commenters -- Gorsuch and Consumers Power  --in  fact
support EPA's broad discretion in construing this statutory term.  These cases
found that the CWA did not speak directly to the question whether dams  "add"
pollutants within the meaning of the CWA and therefore deferred  to EPA's
position that dams are not subject to regulation because they do not
themselves "physically introduce a pollutant into water from the outside
world."  See Gorsuch, 693 F.2d at 175.  Similarly, there is no evidence in the
language or legislative history of the CWA that Congress ever considered the
precise question whether the release of intake water pollutants  into  waters of
the United States is subject to regulation under the CWA.  As explained below,
in deciding that the discharge of intake pollutants is subject to the CWA, EPA
has adopted an interpretation of "addition" that the Agency believes  is
consistent with the language in the statute, and its underlying  intent  and
purposes.

      As an initial matter, EPA  rejects the notion advanced by  some
commenters that EPA's statutory authority to regulate a discharge depends  on
the discharge exceeding a threshold of detrimental environmental effects on
the receiving water  (however such a threshold is defined).  The  statute
provides simply that "any addition of any pollutant to navigable waters from a
point source" is a discharge subject to the CWA  (emphasis added).  CWA
502(12).  EPA does not see how this broad statutory language could be read as
mandating that EPA exclude from regulation discharges of pollutants that do
not meet a specified threshold of environmental impact on the receiving
waters.  EPA has never, since the enactment of the CWA, adopted  such  a
constricted view of its statutory authority, and  declines to do so now.
Certain cases cited by commenters  (such as Avovelles and MCC of  Florida)
merely noted the detrimental environmental effects of the activities  in those
cases in the course of determining that they were regulated under the CWA.
However, those decisions in no way indicate that such environmental effects
are a jurisdictional prerequisite to subjecting an addition of pollutants  to
the CWA's requirements.

      Construing the CWA as mandating such an approach would, moreover, be at
odds with the basic structure of the CWA, which contemplates that the
environmental effects of a. discharge will be evaluated in the context of the
permit issuance process.  Whether the discharge of a particular  pollutant  will
adversely impact the waterbody is a question appropriately addressed  in the
context of determining compliance with applicable water quality standards.
See CWA 301(b)(1)(C).  Requiring such an environmental analysis  up front,  as a
prerequisite to  the assertion of any regulatory authority whatsoever  over  an
activity, would  turn this  statutory scheme on its head.

      In EPA's view, whether a pollutant is discharged to waters of the United
States turns on  a common sense notion of what the term "addition" means:   the
simple, physical act of introducing a pollutant into a water of  the United
States from the  outside world.  Some  commenters  argued that the regulation of
intake water pollutants does not meet this test  (which was adopted by the
courts in the Gorsuch and  Consumers Power cases) because the pollutants had
already been contained  in  the waterbody prior to its removal and use  by the

-------
                           Section VHI.E: Reasonable Potential                       355
discharger as intake water.  These commenters contended that regulating intake
pollutants therefore reflected a sudden break with EPA's historic policies and
practices, and was contrary to these judicial precedents.

      EPA's position here is entirely consistent with the legal interpretation
advanced by  EPA, and upheld by the courts, in these cases.  EPA believes that
the pivotal fact for determining whether an addition has taken place for
purposes of section 402 of the CWA is simply whether a pollutant is physically
moved from outside of the waterbody into the waterbody by the discharger via a
point source.  In SPA's view, the appropriate analytical scope for answering
this question need go no further than the end of a facility's discharge pipe.
If, immediately prior to the discharge activity, the pollutant was not
contained in waters of the United States, then the release of the pollutant
into the waterbody is, quite logically, an "addition" of that pollutant to the
waterbody.  EPA sees nothing in the language of section 502(12} that compels
the Agency to look more broadly at the entire life history of the pollutant
(i.e., whether the pollutant was in the waterbody at some previous time) in
order to answer the straightforward question whether a facility has "added" a
pollutant to the waterbody.  This interpretation is entirely consistent with
EPA's position in the Gorsuch and Consumers Power cases, where the Agency
advanced the position that dams did not "add" pollutants because the
pollutants never left the waterbody in the course of being diverted by the
facilities.  (It should be noted that, in the case of discharges of dredged
material regulated under section 404 of the CWA, the dredged material is by
definition, contained in waters of the United States, and redeposition of such
materials can be subject to the permitting requirement even though they are
not introduced into the waterbody "from the outside world."  See AvoYelles.
715 F.2d at 924, n.43.)

      Moreover,  adopting an exceedingly broad analytical scope to determine
what constitutes an "addition" (to try to trace the ultimate source of the
pollutant) would run counter to Congress' goal of effective water pollution
control.  Such an approach would transform a simple factual inquiry regarding
whether a facility has added a pollutant to the waterbody into a more complex
analysis of the ultimate source of the pollutant, and its relationship to the
receiving water. .Ascertaining whether a pollutant in a facility's effluent
was originally contained in the receiving water can in some cases be a
complicated task subject to uncertainties due to variability of pollutant
levels in the intake and effluent, and limitations in available analytical
techniques for detecting and quantifying such pollutants at low levels.  EPA
believes that it wquld make no sense  (and certainly would be inconsistent with
the comprehensive nature of the CWA's permitting requirement) to exempt
entirely a pollutant discharge from regulation based on such an analysis.
Moreover, EPA sees nothing in the text, structure or legislative history of
the CWA to support the view that Congress expressly decided to require EPA to
undertake this complex technical analysis as a prerequisite to regulating a
discharge of pollutants by a point source.  EPA believes, rather, that it is
far more consistent with the structure and purposes of the CWA to apply
regulatory scrutiny whenever pollutants are added by a point source to waters
of the United States, and to address the particular environmental issues
associated with discharges of intake pollutants through special permitting
procedures such as-those contained in the final Guidance.

      Focusing the "addition" analysis on the narrow factual question whether
the pollutant was outside the waterbody immediately prior to its discharge is
an approach that has been expressly supported by some of the case law relied
upon by commenters challenging EPA's authority to regulate intake pollutants.
In Consumers Power, the court found that the entraining of fish by a pumped
storage facility did not constitute an "addition" because the facility "never
removes the fish from waters of the United States."  862 F.2d at 585. The
court contrasted such a situation with seafood processors, which "add"
pollutants because .they remove the fish from waters of the United States prior
to releasing the re'mains of processed fish.  The court in Consumers Power also
contrasted releases by dams, where the water containing entrained fish "never

-------
356    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


loses its status as water of the United States," with diversion  of cooling
water by industrial steam/electric operations, where removal of  pollutants
from waters of the United States and  their subsequent discharge  was subject  to
the CWA.  Id. at 589.

      One commenter pointed out that  the court's discussion in Consumers Power
of steam electric operations focused  on the fact that use of cooling water by
these facilities results in the absorption of "heat and other minerals
produced by the plant or electric generator before being added to waters of
the United States."  Id.  This commenter argued that it was the  alteration of
the intake water that,  in the court's analysis, made its release constitute  an
"addition" of pollutants subject to regulation, and that therefore this
reasoning did not apply to other discharges of intake pollutants by other
types of facilities.  EPA has two responses to this comment.  First, the
cooling water process associated with steam/electric facilities  is, in EPA's
view, typical of industrial use of cooling water generally, as is the effect
of this process on the  water that is  used  (i.e., "cooling" water is
necessarily heated and  can also absorb minerals in the process).  Thus, EPA
believes that the court's citation of this example provides guidance that use
of intake water as cooling water by any industrial facilities generally, and
the subsequent discharge of those cooling water, is an "addition" subject to
CWA regulation.  Second, as discussed elsewhere, EPA does not believe that it
is the  "alteration" of  intake water quality that renders its removal and
subsequent discharge an "addition" of pollutants.  Rather, the simple fact
that the pollutants were withdrawn by the facility so that they  were no longer
in waters of the U.S. means that the  subsequent release of those pollutants
into the waterbody is an addition of  pollutants from the "outside world."
This analysis is consistent with the  discussion in Consumers Power, which
found that use of cooling water was an addition because the water "loses its
status as waters of the United States."  In EPA's view, this is  the critical
fact for determining whether an "addition" has occurred, and this position is
consistent with the discussion in Consumers Power.

      Consistent with this line of reasoning, EPA  maintains its position
that, once pollutants contained in intake water are removed from waters of the
United States, the pollutants are "outside" such waters, and the subsequent
release of those pollutants back into jurisdictional waters via  a point source
is an "addition" subject to regulation under the CWA.  A contrary reading of
the statute is not compelled by the language of the CWA and would, moreover,
complicate and hamper the administration of the comprehensive permitting
scheme  created by Congress under section 402 of the CWA.

      In light of the growing body of case law finding that EPA  has broad
discretion to interpret what constitutes an addition subject to  the CWA, EPA
does not agree with commenters who relied upon the decision in Appalachian
Power to argue that EPA is wholly without statutory authority to regulate
releases of intake pollutants into waters of the United States.  The court in
this case found that the withdrawal,  use, and subsequent discharge of
pollutants originating  in the receiving water does not constitute an addition
of pollutants.  The court based this  conclusion on its finding that it  is
beyond  the scope o£ EPA'e authority to require a facility to remove pollutants
"other  than those added by the plant  process."  Appalachian Power. 549  F.2d  at
1377.   For the following reasons, EPA does not agree that this  court's
decision and reasoning  preclude EPA  from continuing its longstanding practice
of asserting regulatory jurisdiction  over the discharge of intake pollutants.

      First, as noted in the preamble to the proposed Guidance,  several courts
have upheld EPA's regulatory approach for considering intake pollutants in
establishing technology-based effluent limitations; this approach can in some
cases lead to subjecting the discharge of intake pollutants to CWA regulation,
but allows for net .limits to be established under certain, defined
circumstances.   (See American Petroleum Institute v. EPA, 540  F.2d 1023, 1034-
35  (10th Cir. 1976); Hooker Chemicals and Plastics v. Train, 537 F.2d 620, 633
 (2d Cir. 1976)) .  Second, it is important to note that the Appalachian  Power

-------
                           Section Vm.E: Reasonable Potential                       357
decision predated decisions of the U.S. Supreme Court  delineating the
appropriate standard of review of agency actions where Congressional  intent
regarding the meaning of a statutory term is not clear.  See, e.g.. EPA v.-
National Crushed Stone Association. 449 U.S. 64  (1980); Chevron U.S.A. v.
NRDC. 467 U.S. 837  (1984).  Because the Fourth Circuit's decision in
Appalachian Power contains no discussion of the deference due to EPA's
statutory interpretation, and the court's conclusory analysis does not.
expressly make a finding that Congress had spoken to the precise question at
issue, EPA questions whether the court's analysis is consistent with  current
legal principles governing review of agency statutory interpretations.  As
discussed above, court decisions subsequent to Appalachian Power have upheld
EPA's interpretation of addition based upon application of the proper,
deferential standard of review.  See, e.g.. Gorsuch. Consumers Power.

      Finally, more recent cases, including a case decided by the same court
which had issued the Appalachian Power opinion, have not adopted the  approach
or reasoning of Appalachian Power.  For example, as discussed above,  the
decision in Consumers Power focused on whether the intake pollutants  in
question had been removed from waters of the United States in order to
determine whether the reintroduction of those pollutants into those waters
constituted an addition subject to the CWA.  That court's analysis was
therefore a sharp departure from the reasoning which some commenters  read into
the Appalachian Power case (i.e., that pollutants contained in the waterbody
can never be subject to CWA regulation, even when they are withdrawn  from
those waters and subsequently discharged into waters of the United States via
a point source).

      In a recent decision in U.S. v. Law. 979 F.2d 977 (4th Cir. 1992), the
Fourth Circuit itself adopted a line of reasoning similar to EPA's approach
here.  This case involved an appeal of a criminal conviction of a mine
operator for violating the CWA by discharging pollutants into waters  of the
United States from a water treatment system that had collected contaminated
run-off from the mining operation.  The defendant challenged his conviction on
the grounds that the CWA imposes liability only upon the generators of
pollutants, and not upon persons over whose property pre-existing pollutants
pass before flowing into navigable waters.  Arguing that the headwaters of the
receiving stream were already polluted prior to entering his treatment system,
the defendant cited Gorsuch.  Consumers Power, and Appalachian Power to argue
that he had no duty to remove pre-existing pollutants.  The Fourth Circuit
held that the discharge of pollutants from the facility's water treatment
system was subject to the CWA's permitting requirement because the water
treatment system was not part of waters of the United States.  Because the
"origin of pollutants in the treatment and collection is therefore
irrelevant," the court found that release of pollutants from the treatment
system into waters of the United States required a permit.  979 F.2d  at 979.

      While the Law case addressed pollutants contained in surface run-off  (as
opposed to intake pollutants originating from the receiving water), the Fourth
Circuit found in this case that the critical fact for determining whether an
addition of pollutants has occurred is whether the pollutants were already in
the waters of the United States at the time  of discharge through the point
source and therefore not being added to waters of the United States.  Since,
in this case,  the water treatment system releasing pollutants was not part of
waters of the United States,  the ultimate origin of those pollutants  was
irrelevant, and the discharge of those pollutants constituted an "addition"
for purposes of the CWA.  This is precisely the line of reasoning that EPA
advocated in the "dams" cases, and continues to adopt here.  Under this mode
of analysis, EPA asserts that reintroduction of intake pollutants that have
been withdrawn from waters of the United States clearly constitutes an
addition of pollutants.  Thus, notwithstanding how some commenters may
construe the reach of the Appalachian Power decision, EPA maintains that its
position that discharges of intake pollutants by point sources are subject to
the CWA is consistent with the current state of the law in the Fourth Circuit.

-------
358    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      Numerous commenters also argued that the proposed Guidance was
inconsistent with the court's decision in American Iron and Steel Institute et
al. v. EPA. 526 F.2d 1027  (3rd Cir. 1975), which held that EPA was required to
adjust technology-based effluent limitations if it could be shown by a
discharger that it was unable to meet the limitations due to significant
amounts of pollutants in intake water.  The court stated that "[s]uch an
adjustment would seem required by due process, since without it a plant could
be subjected to heavy penalties for circumstances beyond its control."  526
F.2d at 1056.  EPA believes that this decision does not undercut the final
Guidance's intake pollutant provisions.  First, under appropriate
circumstances, the final Guidance authorizes a special reasonable potential
determination procedure and the adoption of "no net addition" limitations
that, in effect, would not hold the permittee responsible for pollutants that
had been in the waterbody prior to use of the intake water by the facility.
Second, the focus of the American Iron and Steel decision was on technology-
based limitations which, by definition, can only require permittees to take
all technologically and economically feasible steps to reduce pollutants in
its discharge.  See CWA § 304(b).  If, as the court noted, a discharger is
simply unable to meet such limitations due to the presence of intake
pollutants, then an adjustment of those limitations is appropriate, and EPA's
permitting regulations recognize this fact.  See 40 CFR §122.45(g).  Under
section 301(b)(1)(C) of the Act, however, limitations necessary to meet state
water quality standards must be included in NPDES permits, without regard to
economic or technological feasibility.  See U.S. Steel Corp. v. Train, 556
F.2d 822, 838 (7th Cir. 1977) ("It is clear from §§ 301 and 510 of the Act,
and the legislative history, that States are free to force technology ....
If the States wish to achieve better water quality [than can be achieved
through technology-based controls], they may, even at the cost of economic and
social dislocation caused by plant closings.").  Thus, in determining the
appropriate limitations on discharges of intake pollutants necessary to meet
the requirements of section 301(b)(1)(C) of the Act,  the feasibility
considerations relied upon principally by the court in American Iron and Steel
are not legally relevant.

      Finally, as to the statement in the court's opinion that adjustment of
limitations would seem required by due process, the precise legal analysis
supporting this conclusion is unclear.   Notably, the court in that case did
not analyze the process by which intake credits would have been determined for
permittees subject -to the regulations at issue in that case, and whether that
process would meet constitutional requirements.  See Mathews v. Eldridqe. 424
U.S. 319, 333 (1976)  ("The fundamental requirement of due process is the
opportunity to be heard at a meaningful time and in a meaningful manner.").
EPA believes that procedural protections satisfying the requirements of the
due process clause are afforded to NPDES permit applicants, since the
applicant has the right to receive notice of draft permits developed by the
permitting authority and to comment on the contents of the draft permit.  See
40 CFR §124.10.  Thus, any permit conditions relating to the discharge of
intake pollutants would only be adopted by the permitting authority after the
permittee has had a timely opportunity to provide comments on the permitting
authority's decision.  In EPA's view, these procedures provide an opportunity
"to be heard in a meaningful time and in a meaningful manner."  Mathews. 424
U.S. at 333.

      Underlying the court's conclusion in American and Iron and Steel was the
concern that a permittee could be held responsible for circumstances beyond
its  "control."  As discussed above, to the extent the notion of  "control"
reflects concerns about technological feasibility, such considerations are not
legally germane to determining water quality-based effluent limitations.  EPA
disputes, however, the notion advanced by some commenters that facilities do
not have any  "control" over discharges of intake pollutants, and that the
discharge of such pollutants is a wholly  "passive" act which should never
subject dischargers to regulatory requirements or legal liability for their
discharge.  While it is true that a point source does not necessarily have
control over the background level of pollutants in surface waters used as a

-------
                           Section VTII.E: Reasonable Potential                       359
source of intake water by the facility, this does not mean that the source
automatically does not have any control of the pollutant levels that it
discharges.  The process of withdrawing intake water from surface waters for
use by a facility is unquestionably an affirmative, volitional act by the
facility.  It reflects an affirmative decision by the point source to use
intake water from waters of the United States instead of utilizing other water
sources  (e.g./ public water supply).  While such decisions may reflect
reasonable economic calculations  (i.e., it may be more expensive to obtain the
necessary intake water from the public water utility), it can hardly be argued
that such decision-making is wholly passive in nature.  Moreover, once the
facility has decided where to obtain its intake water, the facility's use of
the water as part of its processes makes it impossible, in EPA's view, to
argue that the water is not subject to the "control" of the facility.  It is
not uncommon for facilities to pretreat their water to make it suitable for
use by the facility, and EPA's technology-based intake credit regulations
recognize, and require, that intake pollutants be reduced through treatment to
the extent feasible.  See 40 C.F.R. § 122.45(g)  (allowing credit "only to the
extent necessary'to meet" technology-based limitations).  EPA acknowledges
commenters' concerns that treatment of intake pollutants may in certain cases
be difficult or costly.  This does not mean, however, that discharges of such
pollutants are wholly outside the discharger's "control."  EPA also recognizes
that removal of intake pollutants may not be necessary to meet water quality
standards in certain cases, and accordingly has included appropriate
provisions in the final Guidance that would allow such discharges when
appropriate environmental safeguards are met.

b.    EPA's Authority and Rationale for Establishing Interim Permitting
      Procedures Allowing "No Net Addition" Limitations for Intake Water
      Pollutants

      In contrast to commenters who argued that EPA has no jurisdiction to
regulate intake pollutants, some commenters argued the intake provision in the
proposed Guidance, as well as the other options discussed in the preamble to
the proposal, were outside EPA's legal authority under the CWA.  In the view
of these commenters, section 301(b)(1)(C) of the CWA, which requires that
dischargers achieve any "limitation .  . . necessary to meet water quality
standards," does not authorize EPA to allow intake pollutants to be discharged
at a level exceeding water quality standards.  These commenters maintain that
section 301(b)(1)(C) flatly precludes the permitting authority, under any
circumstances, from finding that a discharge of pollutants into non-attainment
waters at levels above applicable water quality criteria "meets" water quality
standards under the CWA.

      EPA disagrees with these commenters that EPA is without statutory
authority to provide for the establishment of special permitting procedures
for the discharge of intake pollutants being discharged into the "same body of
water" as defined the final Guidance.  EPA believes that it has the legal
authority to adopt the "no net addition" permitting procedures contained in
the final Guidance.  The Agency believes, moreover, that it is appropriate, as
a matter of policy, that these procedures be followed for an interim period of
time (i.e.,  12 years from publication of the Guidance) to provide an
opportunity for States and Tribes to develop TMDLs or comparable assessment
and remediation plans as provided in procedure 3.A of appendix F which ensure
attainment of water quality standards in the waterbody.

      EPA agrees with the general principle that section 301(b)(1)(C) narrowly
limits permissible discharges into waters of the United States to those which
meet water quality standards.  Consistent with this statutory requirement,
EPA's regulations require the imposition of WQBELs that are derived from, and
comply with,  all applicable water quality standards.  See 40 CFR
122.44(d)(1)(vii)(A).

      While section 301(b)(1)(C)  clearly requires that all dischargers meet
water quality standards, the CWA does not expressly address the specific

-------
360    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


technical question of how EPA should determine whether a particular discharge
is, in fact, meeting such standards.  Ascertaining a discharger's  compliance
with water quality standards is a technical determination that is  necessarily
made on a case-by-case basis, based on consideration of the pollutants  in the
discharge, as well as the discharge's relationship to, and interaction  with,
the conditions in the receiving water.  In the case of intake pollutants
discharged into the same body of water from which they originated, there is a
highly unique relationship between those discharged pollutants and the
receiving waterbody.  In some cases, discharge of such a. pollutant merely
transfers the pollutant downstream to a location that the pollutant would have
reached in any event, and the environmental impact of such a discharge  may be
no different than that which would have occurred had the pollutant remained
in-stream.

      EPA believes that, to ascertain accurately whether, and under what
conditions, such a-discharge meets water quality standards, a site-specific
evaluation of the relationship between the discharge and the waterbody  is
necessary.  Where it can be demonstrated that particular intake pollutants in
a discharge meet all conditions necessary to ensure that the discharge  would
have no different impact on the receiving water than the impacts that would
occur in the absence of withdrawal and discharge of the pollutant, EPA
believes that, as a matter of policy, it has the authority to authorize the
permit writer to  account for the presence of that pollutant in determining a
discharger's compliance with water quality standards, and in deriving
limitations that are necessary to meet such standards.  Further, for the
reasons explained below, EPA believes that it is appropriate for this special
allowance of intake pollutants to be effective only for an interim period, to
provide continued incentives for the permitting authority to develop
comprehensive solutions necessary to reduce background concentrations of
pollutants in non-attainment waters.  However, EPA believes that   accounting
for such pollutants in making a determination of compliance with water  quality
standards must only result from a rigorous, case-specific inquiry  carefully
evaluating the relationship of the intake pollutant discharge with the
waterbody, and a finding that the presence of the discharge does not alter the
impacts that would have occurred in the absence of the discharge.

       A case-specific inquiry is  required in the final Guidance, which
provides that, in order to be eligible for "no net addition" limitations, the
permitting authority must find that the intake pollutant is discharged  into
the same body of water from which it originated, and that the facility  does
not increase the mass or concentration of the pollutant, alter the pollutant
chemically or physically, or cause any increased adverse effects due to the
timing and location of the discharge.  In light of the clear mandate in
section 301(b)(1)(C) that all discharges meet water quality standards,  EPA
believes that it would be necessary for a discharger to meet all of these
conditions in order for the presence of intake pollutants to be accounted for
in determining whether a discharge meets water quality standards and in
deriving WQBELs.  Only when these conditions are met is EPA satisfied that
discharge of a particular pollutant would not result in any impacts that would
not have occurred if the pollutants had remained in-stream, and that
discounting of the pollutant in the standards compliance determination  may be
therefore appropriate, as a matter of policy, on a case-by-case basis.

      Some commenters argued that intake pollutant relief should also be
available where a facility has intake pollutants from a different  body  of
water  (or from other sources allegedly not the  "responsibility" of the
discharger), provided that the concentration of the pollutant in the discharge
is no greater than^the concentration  in the receiving water.  In the view of
these commenters, such a discharge should be allowed on the grounds that it
does not worsen the condition of  the waterbody.  The relevant test for
determining consistency of a water quality-based permitting approach with the
requirements of the CWA is not, however, whether a discharge worsens existing
conditions, but whether the discharge  "meet[s] water quality standards."  CWA
section 301(b)(1)(C).  EPA does not believe that there would be  any basis,

-------
                           Section Vm.E: Reasonable Potential                       361
either legally or technically, to account for the presence of pollutants in a
discharger's effluent where those pollutants only reach the receiving water
because of the presence of the discharge.  Since the introduction of such
pollutants into waters of the United States is precisely the type of activity
that the  CWA is intended to regulate, EPA does not believe that accounting
for the presence of such pollutants in determining a discharger's compliance
with water quality standards and in deriving effluent limitations would be
permissible under section 301(b)(1)(C) of the CWA.

      Since EPA has determined that the presence of intake pollutants can, as
a matter of policy and under certain circumstances, be accounted for when
determining what limitations are necessary to meet water quality standards,
the question arises as to how to derive the appropriate WQBEL that implements
section 301(b)(!)(C).  One option considered by EPA would have allowed a
straight "credit" for intake pollutants.  Under this option, an initial WQBEL
would be derived following the normal permitting procedures, and that
limitation would then be adjusted upwards to allow the facility to discharge
any additional amount of an intake pollutant meeting all the conditions
specified in the final Guidance.  Thus, for example, if the permitting
authority were to determine that the appropriate WQBEL for a discharge into a
non-attainment water would be criteria end-of-pipe  (in the absence of intake
pollutants), the discharger would receive a limitation corresponding to that
limitation plus the amount of the intake pollutant  (assuming all other
conditions in the final Guidance were met).  EPA rejected this option because
facilities would be allowed to discharge pollutants into non-attainment waters
at levels exceeding the background level of the pollutant in the receiving
water.  Such a discharge would therefore actually be making the quality of a
non-attainment water worse, a result that cannot be reconciled with the
requirement in section 301(b)(1)(C) that dischargers meet water quality
standards, or with the CWA's goal of restoring and maintaining the integrity
of the nation's waters.

      Therefore, as discussed in previous sections, EPA decided that the
appropriate method of taking account of intake pollutants in developing WQBELs
would be to establish limitations based on the concept of "no net addition."
Under this approach, the permitting authority would establish limitations
allowing the facility to discharge the amount of the intake pollutant, but not
any additional amount.  EPA recognizes that, under this approach, a facility
without pollutants in its intake water may be permitted to add an additional
amount of the pollutant from its process  (e.g., a concentration based on
criteria end-of-pipe)  whereas a facility with intake pollutants would be
precluded from adding to the total net amount of the pollutant discharged.
However, rectifying this difference by allowing a facility with intake
pollutants to add an additional amount to the total loadings in the effluent
would not, as noted above, be consistent with section 301(b)(1)(C), or the
goals of the CWA.

      While EPA believes it has discretion to account for the presence of
intake pollutants in deriving WQBELs under certain circumstances, EPA believes
that it is vital that this discretion be exercised in a manner that  advances
the underlying objectives of the CWA:  to restore the integrity of the
nation's waters.  CWA section 101(a).  Therefore, as discussed in previous
sections, the final Guidance only allows the "no net addition" approach for a
period of approximately two permit terms.  After that time, WQBELs in a non-
attainment water would be based on the WLAs in an EPA-approved TMDL or
comparable assessment and remediation plan approved under procedure 3.A of
appendix F.  In the continued absence of a TMDL or comparable plan,  WQBELs
would be developed without special consideration for intake water pollutants
in accordance with the provisions in procedure 5.F.2 of appendix F.    (See
section VIII.E for a discussion of deriving WQBELS for discharges to non-
attainment waters in the absence of a TMDL.)

      While EPA believes that "no net addition" limitations are appropriate as
an interim measure for dealing with intake pollutants in the permitting

-------
362    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


process, EPA recognizes that this approach contributes to the mere maintenance
of status quo for waterbodies that are currently out of compliance with water
quality standards.  This status quo is not an acceptable long-term result
under the CWA.  As has been discussed extensively during the development of
the Guidance, the root causes of non-attainment with water quality standards
in the Great Lakes basin are complex, and for some pollutants of concern, non-
attainment  may be due predominantly to nonpoint sources of pollution.  The
best, indeed in some cases the only, means of bringing such waterbodies into
compliance with standards is to address comprehensively all the  relevant
sources of contamination.  The CWA mechanism for comprehensively evaluating
these sources is the TMDL development process, which establishes wasteload
allocations for point sources and load allocations for nonpoint  sources in
order to calculate the pollutant reductions from point sources necessary to
achieve water quality standards in the waterbody.  Section 303(d) of the CWA
requires States to identify waters that are not attaining water  quality
standards and in accordance with the State's priority rankings,  develop TMDLs
for those waters.  As discussed above, States may also undertake other  efforts
that identify sources causing or contributing to a water quality impairment
and develop load reduction activities for those sources as necessary for
attainment of water quality standards.  When these efforts satisfy the
provisions of procedure 3.A of appendix F, they may serve in lieu of a  TMDL as
the basis for developing WLAs for discharges of intake pollutants from  the
same body of water-after the time period for "no net addition" limits ends.

      By limiting the time period during which the "no net addition"
permitting approach will be available, EPA is establishing incentives for
States and Tribes to develop TMDLs or comparable assessment and  remediation
plans that will comprehensively assess all of the causes of non-attainment in
the affected waterbody and establish a plan for achieving attainment, a step
which EPA believes is critical if the objectives of the CWA are  to be met.  As
to the appropriateness of precluding intake pollutant relief after this
interim period, it is important to emphasize that EPA does not view the CWA as
mandating that special allowance be made for intake pollutants.  EPA is,
rather, allowing such consideration by the permitting authority  for an  interim
period of time as a matter of policy, for the reasons explained  above.
Dischargers do not, as some commenters asserted, have a "right"  to discharge
intake water pollutants since EPA believes that the discharge of intake
pollutants by a point source constitutes an  "addition" of pollutants subject
to regulation under the CWA.  Among the many policy options available to
address intake pollutants, EPA believes that allowing the no net addition
approach where the discharge is not altering the impact of the pollutant on
the waterbody, but limiting the time period during which this procedure is
available so that the root causes of non-attainment can be addressed through
TMDLs or comparable assessment and remediation plans, best complies with the
mandates of the CWA and will most effectively achieve its goals.

6.    Continued Availability of Existing Mechanisms

      In the preamble to the proposed Guidance, EPA stated that  existing
mechanisms  could adequately address intake pollutants, particularly in  non-
attainment  waters, and provided numerous examples.  See 58 FR  20953-56  (April
16,  1993).  At the same time, EPA proposed  the new reasonable  potential
procedure addressing intake water pollutants in recognition that States and
dischargers had  concerns about administrative burdens associated with existing
mechanisms,  including increased paperwork, permit issuance delays, and
increased costs  for both dischargers  and the States.  Numerous commenters
echoed  these  concerns,  with  several asserting that the environmental result
would be the  same regardless of which mechanism were chosen to adjust the
limits.  Some  commenters also objected to the use of existing  mechanisms on
the  grounds that  they may  not be  available  under existing State  laws, and  even
if they were,  would be  discretionary  with permit writers, and  none of them
would necessarily offer relief considered by these commenters  to be adequate.

-------
                           Section Vm.E: Reasonable Potential                       363
      The final Guidance addresses many of these practical concerns with
existing mechanisms by adopting an alternative reasonable potential procedure
that focuses on intake pollutants and by establishing methods to consider
intake pollutants directly in developing WQBELs.  The intake credit provisions
in the final Guidance are designed to provide for consideration of intake
water pollutants through the NPDES permit program in certain circumstances,
but do not replace the need for other mechanisms to address the underlying
problem, i.e., waters that do not meet water quality standards.  Indeed, one
of EPA's major concerns with adopting a permit-based mechanism to adjust
WQBELs for intake pollutants was that it could discourage the development of
TMDLs.  When non-attainment is due largely to significant loadings from
nonpoint sources or. historical contamination, as is the case for some
pollutants in many of the waters in the Great Lakes basin, a TMDL is the
mechanism best suited for characterizing the problems and possible solutions,
and establishing a comprehensive plan to bring the water into attainment.
Other mechanisms may accomplish this same objective as TMDLs and thus serve in
lieu of TMDLs as described above.  To provide incentives for comprehensively
evaluating sources contributing to excessive background levels of a pollutant
and identifying steps necessary to reducing those levels to acceptable levels,
EPA is limiting the availability of no net addition limits to 12 years after
the publication date of the final Guidance.

       A TMDL is the preferred mechanism for identifying all sources of a
pollutant to a waterbody and devising a load allocation scheme to achieve
attainment of water quality standards throughout the waterbody.  In all cases,
TMDLs should be considered initially when the background concentrations in the
waterbody exceed applicable criteria before turning to intake credits.
Indeed, section 303(d) of the CWA requires States to identify waters that do
not meet water quality standards after implementation of technology-based
controls on point sources discharges, establish a priority ranking of those
waters, and prepare TMDLs.  A phased approach to TMDL development, described
more fully in section VIII.E of this document, provides flexibility in cases
where data are limited.

      All options discussed in the preamble to the proposed Guidance preserved
the paramount role of TMDLs in allocating load reductions among various
sources of pollutants, as reflected in the requirement to implement limits
consistent with a TMDL.  The final Guidance retains this feature.  The
provision addressing the relationship between the intake pollutant procedures
and TMDLs has been moved to the front of the procedure and reworded slightly
to highlight the fact that when a TMDL exists, it governs wasteload
allocations for individual point sources and intake pollutant procedures do
not apply.  See procedure B.D.l.c of appendix F of the final Guidance.  This
section also now recognizes that assessment and remediation plans approved in
accordance with procedure 3.A of appendix F may serve in lieu of TMDLs as the
basis for establishing WLAs.  Where such a plan has been approved, it would be
used instead of the intake pollutant procedure to determine WQBELS.  The TMDL
or comparable plan itself may take into account the presence of intake water
pollutants in determining the appropriate load allocations for nonpoint
sources, wasteload allocations for point sources, a margin of safety required
by CWA section 303(d), and any reserved allocation for future growth
determined desirable by the State or Tribe.  Nothing in the final Guidance
alters a State's or Tribe's existing responsibility to identify and provide
priority ranking for waters that do not meet water quality standards in
accordance with section 303(d) of the CWA and 40 CFR  130.

      Some commenters argued that the final Guidance should require States or
Tribes, in developing a TMDL, or permit writers in implementing a TMDL, to
grant intake credits despite the existence of a TMDL that allocates loads
differently than would the intake credit procedures.  These comments are
consistent with other numerous comments that dischargers should not be
responsible for pollutants in their discharge that originate in the intake
water.  One  suggestion was that WQBELs in waters exceeding WQS not be more
stringent than the larger of the criterion or the background concentration.

-------
364    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


Another suggestion-was that, given the.number of conservative assumptions in
addressing scientific uncertainty in  developing criteria or values and  TMDLs,
the WQBEL calculation becomes unworkable without considering that intake
waters frequently are "contaminated"  with trace concentrations;  and stated
further, since trace levels are being proposed to be unacceptable i.e., the
new GLI criteria), reality demands a  sound procedural solution for this phase
of national water quality protection.

      EPA does not agree that States  and Tribes should be'constrained by the
intake credit provisions when establishing TMDLs or comparable assessment and
remediation plan designed to achieve  water quality standards in  the waterbody.
Nothing in the CWA 'constrains the authority of EPA or the States to require
whatever load reductions are necessary to attain water quality standards.  As
a matter of policy in implementing the NPDES program, EPA has established
intake credit procedures to restrict  the load reduction responsibilities for
individual dischargers in the absence of a TMDL or comparable plan.
Historically, State flexibility in allocating load reduction responsibilities
has been a key feature of the TMDL process.  However, imposing mandatory
intake credit provisions on a State TMDL or comparable process would eliminate
the flexibility to apportion available loadings among sources in a manner
which best achieves CWA objectives, accounts for sources that cannot feasibly
be reduced, and meets State policy goals such as cost-effectiveness and
reserving loading capacity to accommodate future growth.  This same
flexibility would allow a State to consider the presence of intake water
pollutants in establishing wasteload  allocations for point source dischargers,
as long as controls on other sources  are sufficient to attain water quality
standards and provide a margin of safety.  Consideration of intake water
pollutants within the context of a TMDL or comparable plan could result in
wasteload allocations, and resultant  permit limits, that are more or less
stringent than limits developed under the intake pollutant procedures in the
final Guidance.

       EPA acknowledges that meeting  the CWA requirement for TMDLs to attain
standards with a margin of safety may be difficult in some cases given
stringent new standards and significant loadings from nonpoint sources.
States face difficult choices in determining priorities for TMDL development
and for devising a workable mix of load allocations, wasteload allocations,
and margin of safety necessary to attain water quality standards when
establishing a TMDL.  It would be inappropriate  (and inconsistent with  CWA
section 510 and EPA regulations at 40 CFR 122.44(d)(1)(vii)(B))  for EPA to
limit States' or Tribes' flexibility  in undertaking this task or to undermine
decisions made through  the TMDL process by adopting a permitting policy that
ignores WLAs established by a TMDL.

       Waters may exceed standards for numerous reasons and all  available
mechanisms should be considered to determine which provides the  most suitable
approach to addressing  a particular situation.  A common feature of all intake
credit options discussed in the preamble to the proposed Guidance was the
continuing availability of existing mechanisms for adjusting standards  or load
allocations that would  allow adjustment to WQBELs and provide relief to
dischargers using polluted intake water.  In addition to TMDLs,  these
mechanisms include site-specific modifications to criteria, temporary
variances to water quality standards, and changing the designated use of the
waterbody.  Also,  in certain circumstances, a permittee may qualify for a
compliance schedule in  its permit, which does not adjust the limits but rather
provides a period  of time to come into compliance with new effluent limits,
under procedure  9  of appendix F.  Under the final Guidance, these mechanisms
continue to be available and are discussed  in detail elsewhere in this
document.  States  and Tribes, of  course, may choose to be more stringent under
section 510 of the CWA, and restrict  the availability of these mechanisms.
Depending on the  circumstances, application of these mechanisms  at the  outset
could preclude the need for WQBELs.   For example, a site-specific modification
to criteria under  procedure 1 of  appendix F may  result in the waterbody
attaining standards, at  the existing levels  and projected discharges from a

-------
                           Section Vm.E: Reasonable Potential                       365
facility may not be at levels determined to cause, have the reasonable
potential to cause, or contribute to an excursion above the newly modified
standard.   In some cases, a combination of mechanisms may be necessary.  For
example, a facility with intake pollutants from a different body of water that
discharges to a waterbody which exceeds standards for a pollutant in the
discharge does not qualify for "no net addition" limits and if the discharge
causes, has the reasonable potential to cause, or contribute to an excursion
above State or Tribal water quality standards, WQBELs for that facility will
be based on the most stringent applicable criteria for the pollutant of
concern  (see procedure 5.E.4 of appendix F) or, where the facility has
multiple sources of intake water pollutants, on a flow-weighted combination of
"no net addition" and criteria end-of-pipe approaches  (see procedure 5.E.5 of
appendix F).  WQBELs derived from these procedures in the final guidance may
be more stringent than those in the facility's existing permit.  In this
situation, the facility may qualify for a compliance schedule  (see procedure 9
of appendix F).  Alternatively, the facility may be eligible for a temporary
variance under procedure 2 of appendix F).

7.    Final Intake Credit Provision and Response to Maior Comments

a.    General Issues

      This section discusses issues common to the reasonable potential
determination and qualifying for limits based on no net addition.

      i.    Pollutant-by-Pollutant. Outfall-by-Outfall Analysis

      In the preamble to the proposed Guidance, EPA explained that the intake
pollutant procedures would apply on a pollutant-by-pollutant, and outfall-by-
outfall basis.  The reasonable potential analysis need only be done for
pollutants in the intake that, absent special consideration for their
origination in the same body of water, could potentially cause or contribute
to an excursion above the water quality standard applicable to the receiving
water.  The presence of an intake pollutant of concern in the discharge would
not require automatic submission of data and an analysis of, or limits for,
all other pollutants in the discharge.  Similarly, if a facility used
receiving water for cooling purposes that exceeded criteria for copper, for
example, and also added copper to a separate process stream, the added copper
to the process stream would not be relevant  to the no additional mass
requirement or for the reasonable potential procedure for the cooling water
stream if the process and cooling water streams were discharged from separate
outfalls.  Each discharge should be evaluated separately.

      Despite the discussion in the preamble to the proposal, some commenters
appeared confused on this point.  One commenter requested clarification to
avoid problems during implementation. EPA agrees that clarification in the
Guidance is appropriate because misunderstanding could be costly.  Although
treatment of one pollutant in a discharge might incidentally remove other
pollutants, as several commenters pointed out, treating different pollutants
could also require installation of new or different technology at significant
expense.  Accordingly,  EPA has added language in procedure S.D.l.a which
states that the determination made under the intake pollutant provisions shall
be made on a pollutant-by-pollutant and outfall-by-outfall basis.

      ii.   Pollutants Covered

      The final Guidance,  as  did the proposed Guidance, provides that Great
Lakes States and Tribes may,  at their discretion, apply any of the
implementation procedures to the pollutants and pollutant parameters listed in
Table 5 of part 132 ("Pollutants Subject to Federal, State and Tribal
Requirements," referred to as the "excluded" pollutants in the proposal).  EPA
solicited comments _on whether the pollutants eligible for intake pollutant
relief should be more or less inclusive.  Specifically, EPA stated that
application of proposed intake pollutant reasonable potential to the

-------
366    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


pollutants in Table 5, including generic pollutant parameters  (e.g.,
biochemical oxygen demand  and total  suspended solids ), is technically
feasible as long as the proposed requirements of procedure 5.E of appendix F
are demonstrated.  EPA asked for comment on whether any intake pollutant
procedures should be applied to all pollutants, including identification of
pollutant characteristics that may prevent demonstration of any of the
proposed requirements of procedure 5.E.  In addition, EPA requested comments
on whether any procedures in the final Guidance that provided for direct
adjustment of permit limits for intake pollutants should be restricted only to
those pollutants that, due to nonpoint source contributions such as
atmospheric deposition, are present throughout the Great Lakes basin at about
the same concentration, and that concentration already exceeds the water
quality criterion.  With respect to Option 3, which would allow relief for
pollutants from other bodies of water, EPA asked whether such a provision
should be limited to non-BCCs.

      EPA received few comments on this issue.   Some commenters stated that
no distinctions should be made based  on the type of pollutant because
regulating intake pollutants was not  legal or equitable for any pollutant  .
Similarly,  some commenters asserted  that any option chosen should be
applicable to all pollutants subject  to regulation by the final Guidance, not
just non-BCCs, because EPA had no sound legal, policy or other reason to limit
intake credits in that manner.  With  regard to limiting relief to ubiquitous
pollutants,  some commenters stated that such an approach would not be
scientifically justified because background concentrations will vary
considerably throughout the basin, as will sediment contributions to
background concentration, which are limited to a few specific areas.  Several
commenters submitted data showing varying levels of background pollutants, in
many cases at levels above the proposed criteria.  Another commenter added
that limiting relief to ubiquitous pollutants would be meaningless anyway,
because any additional treatment necessitated  for the remaining pollutants
would likely remove the ubiquitous pollutants as well.

      Commenters who generally opposed intake credits favored efforts to limit
the availability of intake pollutant  relief based on the type of pollutant.
One commenter stated that although intake pollutant relief for BCCs generally
would be contrary to the purposes of  the GLWQA and the CPA  to advance
"virtual elimination" of those pollutants; nevertheless,  relief would be
appropriate where, because of atmospheric deposition, the pollutant is present
throughout the Great Lakes at similar levels.  One commenter advocated a
prohibition on the transfer of pollutants to another medium for  intake
credits, e.g., an intake credit for atmospheric deposition must not be
allowed, unless provisions are written that guarantee that industry will not
benefit by burning toxics in order to get an intake credit.

      The final Guidance is the same  as proposed with respect to the
pollutants covered by the intake  credit procedures.  That is, States or Tribes
can adopt the intake pollutant procedures in procedure 5 for any pollutant for
which the State or Tribe determines that such procedure is appropriate.
Although certain pollutants are excluded from the Guidance under Table 5, EPA
believes that States and Tribes can choose  to apply the intake pollutant
procedures to any of the Table 5 pollutants, where the State or Tribes
believes that application of those procedures is appropriate.  Moreover,
States and Tribes can elect not to adopt intake pollutant procedures for any
pollutants, since the final Guidance  does not mandate that these procedures be
adopted by States or Tribes.  States  and Tribes can, of course,  choose to
adopt intake pollutant  procedures that are more stringent than the procedures
in the final Guidance,  consistent with section 510 of the CWA.

      While EPA believes that it  has  discretion to develop different
approaches based ori the nature of the pollutants, it declines  to do so in  this
instance.  The reasons  supporting EPA's decisions on intake pollutant relief
do not readily support  distinctions among different types of  pollutants,
e.g., BCCs vs. non-BCCs.  Ubiquitous  pollutants, particularly  PCBs and

-------
                           Section Vffl.E: Reasonable Potential                       367
mercury, are of concern when considering intake pollutant relief, but
available data suggest that it would be difficult to support a finding that
any pollutant occurs at roughly the same level throughout the Great Lakes.
EPA believes that the intake pollutant procedures in the final Guidance are
defensible on legal and technical grounds regardless of the pollutant at
issue.  When the source of the intake pollutants and the discharge are within
the same body of water, the final Guidance provides relief when the permittee
can demonstrate that the pollutant of concern in the discharge essentially has
no effect on the receiving water quality greater than would occur if the
pollutant were left in-stream.  When the source of the intake water pollutant
is a different body of water, EPA does not believe that intake pollutant
relief cannot be reconciled with the requirement to establish limits that
implement water quality standards, even if the pollutant of concern can be
characterized as "ubiquitous."  Relief might be appropriate through other
mechanisms, however, such as variances from water quality standards, as
provided in procedure 2 of appendix F.

    Few comments specifically addressed whether intake pollutant relief would
be appropriate or inappropriate for pollutants on Table 5.  The only objection
to making intake pollutant procedures available to Table 5 pollutants are that
they are not needed for those types of pollutants.  Because States have
discretion in this matter, no changes have been made, which would limit a
State or Tribe's ability to apply the intake pollutant procedures in the final
Guidance to pollutants on Table 5.  However, as a practical matter, EPA does
not expect much demand for intake pollutant relief for many of these
pollutants because they biodegrade quickly in water or are relatively easy to
treat and thus do not present the same problems as BCCs, for example.

      iii.  Required Demonstration

      The proposed Guidance  required the permittee to demonstrate that it met
each of the five requirements for intake pollutant relief.  The permitting
authority was required to document how this demonstration was made in the fact
sheet or statement of basis for the permit and to establish monitoring to
ensure that the requirements continued to be met throughout the term of the
permit.

      Most comments" on the necessary demonstration focused on the specific
elements that needed to be demonstrated and are addressed in the sections
addressing those requirements.  In addition, several commenters objected
generally that the intake pollutant procedures were cumbersome and onerous,
put an unfair burden on permittees and, because of the "absolute nature" of
the required demonstration, effectively denied any relief.  These commenters
requested that permitting authorities be given more discretion to exercise
their best professional judgment in making the reasonable potential
determination.  The most extreme application of this approach, advocated by
some, was to simply state that intake pollutants may be considered in WQBEL
development and leave it up to States to determine how.  Others, in contrast,
urged relative specificity because consistency among the Great Lakes States is
a major goal of the CPA and the Guidance.

      In the final intake pollutant procedures, EPA has balanced the goal of
consistency with flexibility for permitting authorities to exercise judgment
in making the ultimate determination of whether WQBELs are needed.  EPA
recognizes that determining whether a discharge causes or has the reasonable
potential to cause or contribute to an excursion above a WQS is a complex
determination based on site-specific factors and calls for the exercise of
judgment, particularly in ascertaining the impact of a particular discharge in
light of other sources and stressors on the receiving water .  At the same
time, leaving all aspects of this determination to the permitting authority
could lead to very different results in comparable situations, a result the
Guidance was intended to minimize.  EPA also' recognizes that while many
aspects of this evaluation are site-specific, there are bounds which can be
prescribed for conducting such an evaluation.  EPA has balanced these

-------
368    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


competing considerations by providing a. general framework for assessing the
impact of intake pollutants on a case-by-case basis that considers the
possible impacts resulting from the discharge while providing discretion to
the permitting authority to determine the details of how the required
demonstration can be met.  This decision is reflected in the discussion below
of each of the conditions necessary to establish eligibility for special
consideration "for intake water pollutants.

      EPA does not agree with numerous comments that a discharger should not
be required to demonstrate that its discharge meets the conditions in
procedure 5.D of appendix F.  Section 301(b)(1)(C) of the CWA requires  that
NPDES permits include any limitation necessary to meet state water quality
standards.  Here, a facility seeks to discharge an effluent  containing a
pollutant at levels above the applicable water quality criteria into a
receiving water that may not be in compliance with WQS.  Despite these
circumstances, EPA has found that such a discharge does not have the
reasonable potential to cause or contribute to a violation of an applicable
WQS provided certain conditions are met.  In light of the potential for harm
to the aquatic environment if those conditions are not met, and in view of the
strict statutory prohibition against discharges that do not meet water  quality
standards, EPA believes that it is entirely appropriate to require
dischargers to demonstrate that they qualify for special consideration  of
intake pollutants under procedures 5.D or E.

       These demonstrations may not be as onerous as some commenters believe,
since information already on hand may be used.  The permitting authority can
look to available data, such as permit application data, in deciding whether
the demonstration Has been made adequately, rather than requiring the
permittee to supply additional data.   In all cases, however, the discharger
bears responsibility for establishing eligibility for special consideration of
intake water pollutants in procedure  5.D and E.  The permitting authority's
responsibility is to document the basis for the determination in the fact
sheet or statement of basis for the permit.

      Additionally, EPA believes the  intake pollutant procedures generally are
less burdensome than those associated with existing mechanisms for addressing
intake pollutants, e.g., variances, downgrading uses.  The level of effort
needed to make the .required demonstration for special consideration of  intake
pollutants will vary depending on the characteristics of the effluent and the
receiving water.  Where the intake water contains numerous pollutants at high
concentrations in relation to the water quality criteria, the demonstration
may require analysis of numerous pollutants.  The final Guidance allows the
permitting authority to require the use of the baseline reasonable potential
procedure  (procedures 5.A-C) before intake pollutant relief is considered.
Using procedures 5.A-C  as a relatively simple screening process has the
advantage of limiting the number  of pollutants needing fuller analysis
 (particularly since most applicants are required  already to provide a
comprehensive characterization of their effluents in the permit  application
and effluent characterization is  the  major  information needed for the baseline
reasonable potential determination.)  Similarly,  the procedures  for adjusting
permit limits for consideration of intake water pollutants apply only to
pollutants that exceed  water quality  standards  in the background of the
receiving water.  This  limits the number of pollutants for which the
demonstration is required.

      iv.   Definition  of  Same Body of Water

       Special allowance for intake water pollutants under the final Guidance
depends on whether .the  source of  the  intake pollutant of concern is from  the
same body of water or a different body of water than the receiving water.   If
the intake pollutant is from the  same body  of  water, the discharger may be
eligible for  two types  of  intake  water pollutant  consideration:   (1)  a  finding
that its discharge does not  cause, have the  reasonable potential to  cause,  or
contribute  to an excursion above  water quality standards when  the  intake  water

-------
                           Section Vm.E: Reasonable Potential                       369
is the sole source of the pollutant and;  (2) if the discharger additionally
contributes that pollutant to the wastestream during its operations, it may be
eligible for limits based on "no net addition."  When the pollutant originates
from a different body of water, WQBELS will be based on the most stringent
applicable water quality criterion  if the discharge causes or has the
reasonable potential to cause or contribute to an excursion above water
quality standards.

      Consistent with the approach taken with the intake credit provision for
technology-based limits, the proposed Guidance did not define "same body of
water."  However, the preamble discussed two  options for defining more
specifically "the same body of water":   (1) define as water segments
designated in State or Tribal water quality standards; and  (2) define "same
body of water" on a case-by-case basis.  The preamble further solicited
comments on what factors should be taken into account when using the case-by-
case approach, including whether the intake and outfall points are within the
same water segment 'identified in State or Tribal water quality standards;
whether the discharge is upstream or downstream from the intake point in
flowing waters or in close proximity to the intake point in open lake waters;
whether the intake pollutants would reach the outfall point within a
reasonable time period in the absence of the removal and discharge back to the
receiving water; or whether the water chemistry  (e.g. hardness and pH) are
similar.

      One option discussed in the preamble to the proposed Guidance provided a
different approach for distinguishing between relief available based on the
source of the pollutant.  Option 4 provided for "no net increase" limits when
the source of more than 10 percent of the wastewater for any discharger is
from the same receiving water and for POTWs which discharge to the same
surface water from which the public water supply is withdrawn.  "Same
receiving water" was not defined.  Under this option, different limits might
apply when the source of at least 90 percent of the wastewater is from ground
water or a public drinking water supply system (except where groundwater is
contaminated ).

      The following discussion addresses three aspects of defining "same body
of water:"  (a) general issue of whether and how "same body of water" should
be defined in the Guidance; (b) whether intermediate use of the water should
preclude intake pollutant relief, i.e., whether dischargers who receive their
water from the water supply system rather than directly from the receiving
water should be eligible for intake pollutant consideration; and  (c) whether,
and to what extent, ground water should be considered as "the same body of
water."

       (A).  General.  Commenters took divergent positions on whether, and how,
same body of water should be defined.  One position was that a definition was
needed in order to provide consistency throughout the Great Lakes System and
to make the distinction meaningful.  Further, in these commenters' view, "same
body of water" should be defined to preclude any requests where the discharge
is not to the same designated segment from which the intake water is drawn, to
require dischargers to demonstrate environmental equivalence of the source
water and the receiving water,  and to preclude any allowance for intake water
pollutants in complex hydrological situations, such as discharges to the open
Lakes or to rivers, such as the lower Fox River and St. Louis River, that
experience seiche events, essentially wind-induced tides, or where there is no
consistent unidirectional flow.  Other commenters who advocated defining "same
body of water" urged broader definitions including:  the entire Great Lakes
Basin  (to be consistent with the basin-wide approach taken generally in the
Guidance and because relatively uniform water quality standards would apply
throughout the basin)  or the same watershed as defined by the State.  Several
commenters advocated finding same body of water whenever the background
concentrations of the source water and receiving water were similar.  This
approach would prevent discharges of relatively polluted intake waters to
cleaner receiving waters and therefore prevent further degradation.

-------
370    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      Another position taken by several commenters was that permit writers
should have discretion in making this determination, with some commenters
advocating that permit writers should have the same discretion to determine
that the same body ,of water restriction need not apply as they do when
technology-based limits are at issue.  In contrast, some commenters argued
against the case-by-case approach because it would be difficult  to administer
given the site-specific nature of the determination and could lead to
inconsistency among the States, which is contrary to the intent  of the  CPA.
In these commenters' view, no definition could be adequate, because a
discharge may have adverse water quality effects due to differences in  aquatic
communities, differences in the physical and chemical characteristics of the
affected waters apart from the pollutant in question, or simply  because of
increased loading in the receiving water, whether the discharge  was from the
same or different body of water; therefore, a better approach, in the
commenter's view, would be to drop the distinction and address these
possibilities through the requirement that the permitting authority consider
whether any adverse water quality impacts will occur due to the  "location" of
the discharge.

      As discussed above, EPA's position is that intake pollutant relief
generally should be available only when the discharge containing the
identified intake pollutant of concern effectively has no impact on the
receiving water that would not otherwise occur if the pollutant  were left in-
stream.  This requires an evaluation of the receiving water and  the
discharge's impact on that water which is necessarily site-specific.  Any
definition of "same body of water" should reinforce and facilitate this
approach.  For this reason, EPA disagrees with commenters who argued that the
distinction between same and different body of water should be dropped  or that
"same body of water" should be defined broadly to include the entire Great
Lakes basin.  Application of uniform standards across the basin  is not
relevant in this context, because receiving water quality is not necessarily
uniform across the basin.  Thus, EPA disagrees that defining same body  of
water on a watershed basis is appropriate.  Also, how States define watersheds
may differ and may be based on administrative considerations, such as units of
manageable size, rather than solely on hydrological considerations.
Similarly, definitions that focus narrowly, such as  only on comparisons of
background concentrations of the intake and receiving waters, could be  over-
or under-inclusive in   identifying situations EPA considers appropriate for
relief, i.e., where the discharge of the intake pollutant has no different
impact that would occur if the pollutant had remained in-stream.

      In the preamble to the proposed Guidance, EPA discussed using water body
segments as a way to identify the same body of water.  Commenters did not
support this option as  the sole way to define same body of water.  However,
some commenters  advocated requiring the intake source and receiving water to
be within the same water segment as a minimum requirement.  EPA  has not
adopted the stream'segment approach for two reasons.  First, as  stated  in the
preamble to the proposal, the way in which States designate segments may vary,
which could lead to inconsistency in how the "same body of water"
determination is made.  Second, as with other narrow approaches, relying
solely on stream segment designations could be over- or under-inclusive for
purposes of defining when the discharge of the intake pollutant  has no
different impact that would occur if  the pollutant  remained  in-stream. The
stream segment designation can be a useful factor to consider, however, in
making the  same  body of water determination.  In some instances, stream
segment designations correlate to the boundary determinations made by States
in establishing  the use of a particular body of water  (and thus, the water
quality criteria that apply to that waterbody).  In this situation, the
permitting  authority could find that  a  "same body of water" determination
would be inappropriate  because,  for  example, the use designation of the
waterbody and corresponding criteria  applicable to the intake water source  is
less stringent than the use designation applicable to the receiving water.   In
this case,  not only could the concentrations of the intake and receiving water
differ  (potentially allowing discharges of water that contain higher levels  of

-------
                           Section VHI.E: Reasonable Potential                       371
the pollutant of concern than the receiving water), but a finding of the  "same
body of water"  would undermine the standards setting scheme established  by
States or Tribes arid required by the CWA.

      The "same body of water" definition in procedure 5.D.2 of the final
Guidance balances the need for consistency with the flexibility to consider
factors on a site-specific basis by defining the basic scope of the same  body
of water determination and listing factors the permitting authority must
consider on a case-by-case basis in making that determination.   The same body
of water definition in the final Guidance is structured as follows.  First,
procedure 5.D.2.b states the underlying definition of same body of water:  an
intake pollutant is from the "same body of water" if the pollutant would  have
reached the vicinity of the outfall point in the receiving water within a
reasonable period of time had it not been removed by the permittee.  In order
to determine whether this underlying definition is met, the Guidance provides
that the permitting authority must consider the three specific factors listed
in procedure 5.D.2.b.i-iii (discussed below), and may consider other site-
specific factors relevant to the fate and transport of the pollutant.  This
structure to the same body of water definition is designed to provide clear
guidance to the permitting authority as to the underlying finding that must be
met in any finding of "same body of water," but preserves flexibility in  the
factors that the permitting authority can consider in making this
determination.

      Under procedure 5.D.2.b.i-iii, the permitting authority may find that an
intake pollutant is from the same body of water if all three of the following
conditions are met:   (i) the background concentration of the pollutant in the
receiving water (excluding any amount of the pollutant in the facility's
discharge) is similar to that in the intake water; (ii) there is a direct
hydrological connection between the intake and discharge points; and  (iii)
water quality characteristics (e.g., temperature, pH) are similar in the
intake and receiving waters.   When these three factors exist, the permitting
authority may make a same body of water determination.

      EPA selected these factors from those listed in the preamble and
suggested by commenters as those conditions which reflect that  an intake
pollutant would have reached the outfall area  within approximately the same
time frame and,  to a lesser extent, with approximately the same effect, as
would have occurred if the pollutant had been left in-stream.  EPA also
decided to include these factors in order to simplify the process for the
permitting authority in making a same body of water finding, since these
factors should be relatively easy to determine based upon easily obtainable
monitoring information.

      The definition in procedure 5.D.2.C also allows the permitting authority
to consider other site-specific factors related to the transport and fate of
the pollutant that may be relevant in determining whether the intake
pollutants would have reached the receiving water in any event within a
reasonable period.   "Reasonable period" is site-specific and may vary
depending on the type of pollutant; therefore, determining a "reasonable
period" is left to the permitting authority's best professional judgment.
Procedure 5.D.2.C of the final Guidance is related to procedure 5.D.2.b in
several ways.  First, even if all three conditions listed in procedure 5.D.2.b
are met, the permitting authority retains the discretion to conclude, based on
other site-specific factors,  that the underlying definition of "same body of
water" has not been satisfied.  While EPA believes that such circumstances are
relatively unlikely, the Agency did not want to preclude the permitting
authority from considering all relevant site-specific conditions in making the
underlying "same body of water" determination required by procedure 5.D.2.b
For example,  the discharge point may be upstream of the intake point.  In this
case,  although the three factors in procedure 5.D.3.a have been satisfied, the
permitting authority could find that the intake pollutant is from a different
body of water because water generally does not flow upstream and the basic
showing that the pollutant would have reached the outfall area in any event

-------
372    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


would not be met.  Similarly, unusual hydrological characteristics  such as
seiche events  mentioned by one commenter may be appropriate to  consider  in
making the required determination.  As discussed above, stream segment
designations may be a useful factor to consider in certain situations.

      Second, under procedure 5.D.2.C, the permitting authority  could based,
on site-specific factors relating to the fate and transport of a pollutant,
find that an intake pollutant is from the "same body of water" even if  one or
more of the criteria listed in procedure 5.D.2.b is not met.  For example, in
a case where the intake point is upstream from the discharge point, but there
is a. tributary that joins the waterbody between the two points,  the permitting
authority could conclude that the pollutant would have reached the  outfall
point within a reasonable period of time, even though the concentration of the
pollutant at the outfall point differs from the intake point due to the
confluence of the waterbodies.  In this circumstance, the absence of one  or
more of the conditions  listed in paragraph 5.D.2.b. would not necessarily
preclude the permitting authority from deciding that the underlying definition
of same body of water is met.  In order to apply the intake pollutant
procedures in this circumstance, however, the permitting authority  must still
find that other applicable requirements in procedure 5.D.3 been  met, e.g.,
that the timing and location of the discharge or chemical or physical changes
to the intake water pollutants that occur at the facility would  not cause
adverse water quality impacts to occur that would not occur if the  pollutant
had been left in-stream.

      Some commenters advocated creating an exception to the same body  of
water requirement similar to the one found in the intake credit  provision for
technology-based limits.  That provision allows the permitting authority  to
waive the same body of  water requirement if  he or she determines that  no
environmental degradation would result  (see 40 CFR 122.45(g) (4)) .   EPA
disagrees that this exemption is appropriate when intake pollutant  relief from
WQBELs is at issue.  When it promulgated 40 CFR 122.45(g), EPA explained  that
it could justify providing this exemption  because  protection of the
receiving water could occur, in part, through imposition of WQBELs  where
necessary to meet applicable water quality standards  (49 PR 38026,  September
26, 1984) .  In other words, WQBELs are a "safety net" which protect receiving
water quality when intake credits are considered for technology-based limits.
There is no comparable  safety net when intake pollutant relief is considered
for WQBELs.  As explained previously, EPA believes that special  allowance of
intake pollutants can only be squared with the water quality-based
requirements of the CWA where it  is demonstrated that the discharge of  the
pollutant does not alter the environmental impact of leaving the pollutant in-
stream.  If this condition is not met, EPA believes that a discharge into a
non-attainment water can be permitted only if it does not exceed water  quality
criteria.  These are the applicable requirements under the water quality
program, not the more general standard of "environmental degradation" provided
under 40 CFR 122.45(g)(4). Therefore, an exemption from the  "same body  of
water" requirement comparable to  the  one available for technology-based limits
is not appropriate.

       (B).   Intermediate Use/Public Water Supply.  The proposal  was silent on
whether  intermediate use of  intake water would preclude a discharger from
seeking  intake credit relief.  Numerous commenters endorsed Option  4 because
it specifically  provided for  situations where a facility used a  public  water
supply rather than-drew its water supply directly from the receiving water.
Many  commenters  stated  that  intake pollutant relief  should be available in
this  situation as well  because a  discharger  should only be responsible  for
pollutants  it originally  introduces into the wastestream and not be required
to clean up  its  source  water.  Others claimed that it would be  inequitable  to
treat direct and indirect  users of the  same  water differently, which would be
the case if  intermediate use  disqualified a discharger from  eligibility for
intake pollutant relief.   A  trade association representing  industries  which
use significant  amounts of cooling water  suggested that  smaller  industries
could be disproportionately  impacted  because they were more  likely to  use

-------
                           Section Vm.E: Reasonable Potential                       373
public water supplies than large industries which could draw from the rivers
or lakes directly. - Another commenter stated: "Virtually all of the water
provided by a public water supplier is ultimately discharged to ground or-
surface water.  It is inequitable to hold a limited number of dischargers
responsible for removing pollutants from such water while leaving other
dischargers of the same water unregulated.  Such pollution should be
controlled at the source, the water utility, or by a general limit adjustment
granted to all dischargers of such water."   Several commenters noted that,
unless intermediate use was taken into account, all POTWs would be ineligible
for intake pollutant relief because a POTW's "intake" water comes from users
of the POTW's system.

      EPA agrees th'at indirect users of the receiving water should be eligible
for special consideration of intake water pollutants to the same extent as
direct users.  As long as all requirements of the intake pollutant provisions
are met, EPA can find no reason for excluding indirect users of the receiving
water from intake pollutant relief.  As noted above, however, EPA rejects the
theory that dischargers generally should not be responsible for pollutants in
their dischargers that they do not themselves add to the wastestream. The
final Guidance addresses the intermediate use issue in two ways.  First,
5.D.2.e. states that intake pollutants  include those withdrawn from the
waters of the United States by any facility supplying the discharger with
intake water  (e.g.,, public water supply) .  Thus, users of  such waters
supplies are eligible for intake pollutant relief to the same extent as direct
users of surface waters.  In applying the criteria in section 5.D.2.b to
determine whether a user of a public water supply is discharging to the same
body of water, the permitting authority would compare the surface water where
it is withdrawn from the waterbody by the water supplier with the user's
outfall location.

      The final Guidance addresses intermediate use more specifically in
5.E.2.C, which details how no net addition limits would be developed for water
supply users.  That section states that intake water shall be determined at
the point where the raw water supply leaves the waters of the United States,
except that it shall be the point at which the water enters the water
supplier's distribution system where the  water treatment system removes any
of the identified pollutants from the raw water supply.  This provision
further specifies that, for purposes of establishing limits, the applicable
concentration of the pollutant should be determined at the intake point, and
that the mass  of the intake pollutant will be determined by multiplying the
applicable concentration times the facility's influent flow from the public
water supply.

      By defining the intake point as the point where the water leaves its
source in waters of the United States, this provision does not extend intake
pollutant relief to pollutants added at the water treatment plant or in the
distribution system.  To allow relief for pollutants added after the intake
water leaves the same body of water would be inconsistent with the rationale
for allowing special consideration of intake water pollutants, i.e., that
special consideration is appropriate only for those pollutants whose discharge
has no different impact than the impact that would have occurred had the
pollutants been left in place,  which is in part reflected by the fact that the
pollutant would have reached the discharge point in any event.  Obviously,
pollutants which are introduced by a public water supply or during
distribution to its users are not initially present in waters of the United
States.  At the same time, EPA does not think it is appropriate to extend
relief automatically to  water supply users for pollutants removed by the
water supply system before reaching the user.

      Several commenters requested relief for pollutants added as part of
water treatment.   Some commenters noted that water which meets drinking water
standards may not be of sufficient quality to meet water quality standards.
For the reasons discussed in previous sections,  EPA disagrees that complete
relief is required for any pollutant in a facility's water supply or that

-------
374    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


generally it should hold water suppliers, rather than users, responsible for
pollutants in the water supply.   However, if the water supplier adds
pollutants for purposes of water treatment into a water of the United States
(e.g., a dammed impoundment or other waterbody meeting the CWA definition of a
water of the United States), and the permitting authority finds that the
impoundment and the discharger's receiving water are the "same body of water,"
the mass and concentration of the pollutant for purposes of the intake
pollutant procedures could include the pollutants added by the water supplier
to the impoundment  (which should have been subject to an NPDES permit).  This
is no different than a situation where the background concentrations of a
pollutant in a facility's intake water includes pollutants  due to other
discharges to waters of the United States subject to the NPDES permit program.

      In most cases, however, treatment of the water supply occurs after the
water is removed from waters of the United States.  As noted earlier,
pollutants added during treatment or in the distribution system are not from
the receiving water initially and therefore would not reach the receiving
water except for the discharge by the water supply user.  While not providing
a special intake pollutant allowance in this instance may impose an additional
burden for the users of the water supply who do discharge to waters of the
United States, EPA does not believe, for the reasons stated above, that
special consideration of intake pollutants is appropriate in this
circumstance.  Likewise, EPA does not agree that the fact that water quality
standards may be more stringent than drinking water standards is a reason to
provide special consideration of intake water pollutants.  Providing intake
pollutant relief based on the difference between drinking water and surface
water quality standards would not be appropriate in the context of ensuring
that the requirements of the CWA are met.  EPA encourages users to work with
their water suppliers, as they would other suppliers of raw materials, to
minimize pollutants of concern in the water supply.  For example, one POTW,
which experienced difficulty in meeting zinc limits in its NPDES permit,
successfully worked with the water supply managers to substitute sodium
metaphosphate for zinc orthophosphate for purposes of addressing problems
created by an aggressive water supply.

       (C).  Ground water.  As with intermediate use, the proposed Guidance did
not specifically address whether pollutants in ground water would be eligible
for intake pollutant relief.  Option 4, however, specifically included ground
water by providing  limited relief  (limits based on water quality criteria with
the possibility for limits up to background in specified circumstances) when
the source of at least  90 percent of the wastewater is from ground water or a
public drinking water supply system  (except where ground water is
contaminated).

      EPA received numerous comments advocating intake pollutant relief for
ground water, echoing the argument that dischargers should not be responsible
for pollutants already  in their intake water.  Several commenters objected
that the proposal would preclude any  relief if the discharger used any amount
of ground water for its water supply.  Some commenters who supported providing
relief for pollutants in ground water also  stated that it would  be appropriate
to distinguish between  ground water naturally contaminated and that
contaminated by human activity.  In addition, commenters argued  that relief
for pollutants was  appropriate because ground water may be hydrologically
connected to the surface water into which the pollutants are discharged and
therefore are, in effect,  the same body of  water.

      EPA does not  agree with  commenters who advocated intake credits for all
pollutants  in ground water  even with  suggested limitations, such as not
allowing  credit  for pollutants added  by human activity, or  including ground
water if  there is any hydrological  connection to surface water.  However, EPA
agrees that relief  for  intake pollutants  in ground water is appropriate for
situations  that  fall within the general  rationale for relief, i.e., when the
pollutants  would have reached the  vicinity  of the outfall in approximately the
same period of time, in the  natural course  of things regardless  of removal and

-------
                           Section VULE: Reasonable Potential                       375
use by the discharger.  Thus, the final Guidance specifies in procedure
5.D.2.d that the permitting authority can find that ground water is the same
body of water as the receiving water using the test specified in 5.D.2.  EPA
also agrees that ground water contaminated by human activity should not be
considered for relief because it would simply transfer the problem without
abating it.  By human activities, EPA means activities conducted by
industrial, commercial, and municipal entities resulting from operations,
disposal actions, or treatment processes.  Thus, ground water at CERCLA
remediation sites would not be eligible for intake pollutant relief.  EPA
would also consider contaminated ground water that has been intentionally
diverted to form a direct hydrological flow to surface water to be covered by
this restriction.  This restriction may avoid the potential for a discharger
to obtain relief for pollutants added as a result of the discharger's activity
 (e.g., when ground water has been contaminated by mining activities and the
ground water serves as the mine's water supply).

      v.    100 Percent Requirement

      Related to the definition of "same body of water" is the requirement in
the proposed Guidance that 100 percent of the intake pollutant is from the
same body of water 'as the discharge in order to be eligible for a finding that
WQBELs are not needed.  The purpose of this requirement, as explained in the
preamble to the proposed Guidance, was to avoid inter-waterbody transfers of
pollutants without making a reasonable potential determination and developing
appropriate WQBELs.  The preamble also clarified that multiple sources of
water did not preclude a discharger from relief automatically.  However, if
the discharge included water from a different body of water which contained
the pollutant of concern, intake pollutant relief was not available, and
reasonable potential would be determined by the proposed baseline procedures
in 5.A-D of appendix F.  Two other options discussed in the preamble were less
limiting.  Option 3. applied intake pollutant relief regardless of the source
of water. In Option 4, the provisions applicable to same body of water
scenario applied as long as at least 10 percent of the wastewater was from the
same body of water.

      According to many commenters, the practical implication of this
requirement would be to deny relief in many situations because dischargers
commonly use multiple sources of water, especially smaller facilities.  Of
particular concern to some commenters was the situation where the discharge
was primarily non-contact cooling water from the same body of water but also
included process wastewater or pollutants added by the infrastructure.
Alternative approaches suggested for this situation included changing 100
percent to "substantially all" and categorically excluding non-contact cooling
water from the reasonable potential determination.  Several commenters also
asserted that EPA had no scientific basis for taking a different approach than
that contained in Option 4, i.e., using a no net addition approach in
situations where the discharges had other sources of water, such as ground
water and public water supplies.  Some commenters supported Option 4 in this
regard as a reasonable interim measure and asserted that States and EPA always
have the authority to impose tighter limits if they believe they are needed to
protect the designated use of a water which provides a sensitive habitat.
Some commenters went further, objecting to limiting Option 4 to situations
where at least 10 percent of the wastewater was from the same body of water.
One commenter stated that the proper test should be whether or not the
discharge,  taken as a whole, has a negligible impact on the receiving water.

      In the final Guidance, EPA has retained in procedure 5.D.3.b.i the
requirement that 100 percent of the intake pollutant come from the same body
of water to qualify for a determination that a WQBEL is not needed.   EPA's
rationale for distinguishing between the same and different bodies of water in
the intake pollutant context has been discussed previously.  The 100 percent
requirement in 5.D.3.b.i implements this decision.  However, unlike the
proposal, the final- Guidance provides additional flexibility where a facility
has multiple sources of the pollutant of concern in two respects.  First, as

-------
376    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


explained in more detail elsewhere, procedure 5.E.5 allows the permitting
authority to apply a flow-weighted average approach for deriving limits when
the pollutant of concern in the discharge originates in both the same and
different bodies of water.  Second, if the facility that has intake water
containing the pollutant of concern   from the same body of water and the
facility also introduces additional amounts  of the pollutant to the
wastestream, e.g., from process waste, 5.E.4. of the final Guidance authorizes
the use of a no net addition approach.  These changes will address many of the
situations identified by commenters as warranting relief.

      EPA does not agree that non-contact cooling water should be
categorically excluded from any reasonable potential determination because it
would deny relief  if the facility co-mingled its non-contact cooling water
with other wastestreams containing the pollutant of concern, as asserted by
some commenters.  EPA is not aware of a factual basis for determining that
non-contact cooling water, co-mingled with other wastestreams, as a class of
discharges, will not have the reasonable potential to cause or contribute to
an excursion above water quality  standards.  In some cases, such discharges may
have a detrimental .impact on the  receiving water, for example, because
pollutants are added to those in  the  intake  water via the cooling water
wastestream  (e.g., through corrosion  or water treatment) or concentration of
the pollutants occurs due to evaporation.  Likewise, the mass and
concentration of the pollutant of concern that may be contributed by the co-
mingled, non-cooling water wastestream, and  consequently the impact of  the
discharge on the receiving stream, will vary from facility to facility.
Evaluating such circumstances and determining whether a discharge poses
"reasonable potential" must necessarily be made on a case-by-case basis.
Nonetheless, EPA believes that the underlying concern of commenters--that some
allowance for intake pollutant be allowed where non-contact cooling water is
commingled with other wastestreams--is addressed through the provision  in the
final rule allowing "no net addition" limits.

      vi.   No Increase in Concentration Requirement

      To qualify for special consideration of intake pollutants, the proposed
Guidance required the discharger  to demonstrate that the identified pollutant
of concern is not concentrated at the edge of any available mixing zone after
discharge from the facility.  The preamble explained that facilities which
further  concentrate pollutants at the edge of a mixing  zone may  contribute to
the excursion abov^ water quality standards.  If no mixing zone  is allowed by
a State's water quality standards, the appropriate comparison is instead to
the point of discharge.  EPA further  explained that proposed procedure  5.E of
appendix F allowed consideration  of increased concentrations at  the edge of
any allowable mixing zone to accommodate water conservation,  as consistent
with the proposed provisions in procedure 3  of appendix F that allowed  the
continued use of mixing zones for BCCs when  water conservation measures
resulted in an overall reduction  in mass loading even though higher
concentrations occurred-  EPA invited comment on all aspects of  this provision
including the interpretation of  "no concentration"; whether a particular
statistical methodology for measuring "no concentration" should  be included  in
the final Guidance; and whether  any provision in the final Guidance for intake
water pollutants  should allow consideration  of a maximum increased
concentration resulting from evaporation of  cooling water.

      EPA  received numerous  comments  opposing the  "no  increased
concentration" requirement based on competing policy considerations but none
on the  grounds that the requirement was  inappropriate  on the basis of  water
quality considerations.   Some commenters disputed whether the water quality
impacts  were sufficiently significant to justify the condition.  As with the
no additional mass requirement  discussed below, numerous  commenters advocated
a general  exemption from  the no increase in concentration requirement  for de
minimis  additions  of mass.   Similarly,  some  commenters  advocated an exemption
for  slight  increases due   solely to evaporation  (but did not  define  "slight.")
Others  requested a blanket exemption  for cooling water,  both  once-through and

-------
                           Section Vffl.E: Reasonable Potential                       377
recycled.  Several commenters asserted.that a "no increased concentration"
requirement would effectively preclude recycled cooling water from relief
because concentration of pollutants is inherent in the design of cooling
towers; furthermore, relief would be denied for BCCs where it was most needed,
even if mass were not increased  (a result that would be further exacerbated by
the proposed phase-out of mixing zones for BCCs).   Commenters suggested that
mass limits only be used to avoid concerns about concentration.  Another
suggestion was to limit this requirement to situations where adverse water
quality impacts would occur.

      EPA has retained the "no increased concentration" condition in the final
guidance with  minimal changes from the proposal.  It appears at 5.D.3.b.iv.
EPA is retaining the "no increased concentration" requirement because it is
integral to its rationale for allowing special consideration of intake
pollutants, i.e., that the removal and subsequent discharge of intake
pollutants would not create adverse water quality impacts that would not have
occurred if the pollutant were left instream.  Discharges which are more
concentrated than the receiving water will  increase the degree to which the
waterbody exceeds the applicable criterion for that pollutant, thereby
exacerbating the any non-attainment problems.  For this reason, EPA rejects
suggestions that imposing mass limits only is appropriate. For waters already
attaining water quality standards, increases in concentration in the discharge
that do not result in an increased concentration of the pollutant at the edge
of the mixing zone would be allowed where mixing zones are available under
State provisions.  This scenario could occur only for purposes of the intake
pollutant reasonable potential procedure.  Applying the "no increased
concentration" requirement at the edge of the mixing zone allows for dilution
of the pollutant in the receiving water and therefore for some increased
concentration in the discharge itself.    In the final Guidance, EPA has also
clarified that, in the attained water scenario, an increase in concentration
at the edge of an available mixing zone is allowed if the concentration at the
edge of the available mixing zone does not cause or contribute to an
exceedance of an applicable WQS.

      Allowing increased concentrations in these limited situations does allow
discharges that potentially cause adverse effects in the receiving water
quality that would not occur if the pollutant were left in-stream.  However,
prohibiting increased concentrations at the edge of an available mixing zone
that result in an exceedance of the applicable WQS when applying the intake
pollutant reasonable potential procedure is consistent with EPA's and States'
use of mixing zones in the water quality standards program and with limiting
"reasonable potential" finding to situations where the discharge causes, has
the reasonable potential to cause, or contribute to an exceedance of an
applicable WQS.  EPA believes that the final Guidance adequately addresses the
circumstances under which an increase in pollutant concentration within the
mixing zone would itself be of environmental concern by precluding the
application of the intake pollutant reasonable potential procedure if the
timing and location of the discharge would cause adverse effects that would
not occur in the absence of the removal and discharge of the pollutant.  See
procedure 5.D.3.b.v of appendix F.  EPA also believes this approach adequately
defines when an increased concentration is significant enough to preclude
special consideration of intake pollutants (i.e., when it causes or
contributes to an exceedance of the criteria at the edge of an available
mixing zone)  and therefore has not adopted some commenters' suggestion that
the "no increased concentration" requirement be modified to apply only when
the increased concentration creates adverse effects that would not occur if
the pollutant were left in-stream.

      Applying the "no increased concentration" requirement may differ
depending on whether eligibility for the intake pollutant reasonable potential
test or eligibility for "no net addition" limits is at issue.  Because
availability of "no net addition" limits in the final Guidance is limited to
situations where the level of the pollutant in the background water exceeds
the criteria for that pollutant, mixing zones will be unavailable when

-------
378    Water Quality Guidance for the Great Lakes System ~ Supplementary Information Document

evaluating whether increased concentration has occurred.  Application of a
mixing zone assumes that water is available to dilute the discharge  so  that
compliance with water quality standards is determined at the edge  of the
mixing zone rather than within the mixing zone.  As noted by some  commenters,
dilution to meet a standard in-stream for a particular pollutant is  not
possible when the pollutant already exceeds the criteria in the background
waters'.  In short,''mixing zones do not exist for non-attainment waters, i.e.,
where background waters exceed criteria even, if a State has a policy which
allows mixing zones.  Unavailability of mixing zones when' background
pollutants levels exceed the criteria also means that the exemption  from the
"no increased concentration" requirement due to competing water conservation
considerations, which is based on the mixing zone provisions in procedure 3 of
appendix F, would necessarily be unavailable when "no net addition"  limits are
at issue.

      EPA recognizes that States or Tribes may develop TMDLs which project
attainment of WQS f,or a water body currently in non-attainment.  In  this
instance, a mixing zone can be allowed as long as it is consistent with the
provisions and assumptions of the TMDL or comparable remediation and
assessment plan approved under procedure 3.A of appendix F.  Since "no  net
addition" limits assume the absence of a TMDL or comparable plan,  the
possibility of mixing zones in non-attainment waters, where no TMDL  or
comparable plan approved under procedure 3.A of appendix F exists, is not
relevant.

      Also, explained above, increased concentration beyond the edge of an
available mixing zone may be allowed when the intake pollutant  reasonable
potential procedure is being applied and the receiving water meets in-stream
standards, if the increase does not cause or contribute to an exceedance of an
applicable WQS.  When the receiving water already exceeds an applicable WQS,
however, as is the case when "no net addition" limits are at issue,  in  an
ideal sense, any increase in concentration exacerbates em existing problem and
therefore is unacceptable because it contributes to an exceedance  of the
standard.

      The "no increased concentration" requirement is necessary to implement
the modest interim goal of holding discharges to non-attainment waters  at a
level that maintains the status quo and does not  aggravate a situation that
already is inconsistent with the CWA's goals.

      As with the other conditions, the final Guidance does not establish how
"no increased concentration" should be determined and leaves this  to the
discretion of the permitting authority.  Language has been added to  the final
Guidance to clarify that the permitting authority has this discretion.  As a
general rule, increases in concentration can be determined easily  by comparing
measurements of intake  levels of the pollutants with those in the  effluent to
determine whether there is any statistically significant difference. This
approach can incorporate principles of averaging  (provided, of course that the
averaging period is appropriate for the circumstances) and also can  have the
end result of accounting for relatively insignificant mass increases or
"slight" increases due  to evaporation where such differences are not
statistically significant.  In addition, increases due to evaporation that
cannot be measured or are not statistically significant, as discussed below,
adequately account for  "insignificant" increases due to evaporation.

      EPA is aware that measurable increases in concentration due  solely to
evaporation may require some dischargers to remove intake pollutants or take
other appropriate measures to ensure  that intake water pollutants  are not
concentrated at the point of discharge.  However, EPA believes that  such
measures are necessary  to ensure that discharges at levels exceeding water
quality criteria into non-attainment waters are allowed only in  those
instances where such a  discharge has no greater adverse effect on  the
receiving water than that which would occur if the pollutant were  left  in-
stream.  Furthermore, these increases would need to be larger than those

-------
                           Section Vffl.E: Reasonable Potential                       379
allowed by the permitting authority as.they use their approach to assess
increased concentrations.

      EPA has not adopted the numerous suggestions for categorical exemptions
for certain types of discharges, particularly recycled cooling water blowdown,
that would be more -concentrated than the intake water.  As discussed above,
EPA believes t;hat an increase in pollutant concentration caused by a discharge
of intake pollutants generally should disqualify a discharger from receiving a
special allowance for intake pollutants.  EPA has no basis to conclude that
certain categories of discharges will not cause increased concentrations of
pollutants in a waterbody.  The impact of  discharges that have increased
concentrations of pollutants as compared to the intake water must therefore be
evaluated on a case-by-case basis.

      vii.  Chemical/Physical Alteration Requirement

      The proposed 'Guidance also required that the permittee demonstrate that
it does not alter the identified intake water pollutant chemically or
physically in a manner that would cause adverse water quality impacts to occur
from the discharge that would not occur  if the pollutant were left in-stream.
Only alterations that cause adverse water quality impacts would be prohibited
under this requirement.  Simple removal of a pollutant, which could be
considered a chemical or physical alteration, would not be considered to cause
adverse effects.  The purpose of this requirement was to exclude from relief
situations where changes in the pollutant occur within the facility that could
render the pollutant more harmful to the receiving water than before its
removal by the facility.  For example, a change in pH or temperature that
affects the structure or valence of some intake pollutants contained in non-
contact cooling water could increase the toxic effects associated with the
pollutants in the discharge.  EPA invited comment on all aspects of this
provision including:  the interpretation of "chemical and physical alteration;
" the interpretation of "adverse water quality impacts;" the specific
environmental and pollutant parameters needing evaluation for making this
determination; the use of statistical methods to make this determination;
whether minimum parameters for consideration, such as hardness, pH, total vs.
dissolved fractions, should be specified in the final Guidance; and whether
the final Guidance should specify the maximum extent to which these parameters
can change without -causing an adverse impact on water quality.

      EPA received numerous comments on this provision. The comments generally
did not dispute the validity or need for the condition, but rather focussed on
the details of how it would be applied.  Some commenters stated that the
burden should be on the permitting authority to demonstrate that a physical or
chemical alteration that adversely affects water quality has occurred, rather
than on the permittee to demonstrate that it has not occurred.  One commenter
requested that "chemical and physical alteration" be defined with sufficient
flexibility to permit adequate consideration of site-specific factors, such as
the comparative qualities of intake and receiving water and the nature of the
discharger's activities so as to avoid disqualification due to altered
pollutants that result in less toxic impacts.  Several commenters advocated
recharacterizing this condition as "no net increase in pollutant
bioavailability" as protecting receiving water while providing necessary
flexibility to deal with different situations reasonably (e.g., where a
facility softens water before use, which may increase the toxicity of
pollutants,  but rehardens the water before discharging).  Similarly, several
commenters suggested requiring whole effluent toxicity tests or water effects
ratio tests to meet this demonstration.  With respect to chemical-specific
testing,  some commenters suggested measuring the bioavailable form of the
pollutant to determine if changes have occurred (as well as basing criteria on
the bioavailable form), but other commenters noted that it would be difficult
to measure the bioavailable form of pollutants at low levels.  Other
commenters expressed concern that changes such as increases in temperature
that result when water is used for cooling purposes, or evaporation that might
increase concentration would eliminate them from relief.  A municipality

-------
380    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


expressed concerns about how this requirement might be applied to POTWs:
"With a sewage treatment plant as our 'end of process,' we would never  be able
to argue that the physical, biological, and chemical processes we employ did
not change the chemical form of the pollutant...."

      The final Guidance in procedure 5.D.3.b.iii retains the requirement for
the permittee to demonstrate that the identified intake pollutant of concern
will not be altered chemically or physically in a manner that would cause
adverse water quality effects that would not  occur  if the pollutant were
left in-stream.  This provision is necessary to identify instances where use
of the intake water at a facility alters the pollutant sufficiently to  cause
additional adverse impacts on the receiving water.  Although several
coimventers seemed to read this requirement to prohibit any alteration of the
pollutant (and even of the chemistry or other characteristics of the
discharge),  EPA's intent, as reflected in the final Guidance , is to address
only physical or chemical alterations to the pollutant that would cause
increased adverse water quality effects as compared to that pollutant in the
intake water before removal and use by the facility, i.e., "adverse water
quality impacts...that would not occur if the pollutants were left in-stream"
(emphasis added).  In this sense, the requirement already focuses only  on
"net" changes as advocated by some commenters.  Temporary changes within the
facility that do not affect the pollutant as discharged are not prohibited.
For example, temporary increases in the toxicity of metals due to water
softening would not fail this condition if the toxicity were reduced-to the
same level as in the intake water by hardening of the water before discharge.
For this reason, the discharger need not demonstrate that no changes have
occurred to the pollutant at the facility as feared by a municipal commenter,
but only whether "net" changes have occurred that create an adverse effect in
the receiving water that would not have occurred if the pollutant had been
left in-stream. The focus of this requirement is on changes to the form of the
pollutant.  It does not duplicate the requirement prohibiting increases in
concentrations or mass, which apply independently of the "no chemical or
physical alteration" requirement.

       EPA has not included additional specificity in the Guidance on what
data are necessary to demonstrate this requirement.  As with the other
conditions, what data are appropriate will vary with the pollutant of concern,
the facility processes, other pollutants entering the wastestream, and
characteristics of the receiving water.  No one test will capture all possible
alterations that could create additional adverse affects for the receiving
water.  As noted by one commenter, toxicity or water effects ratio measures
could indirectly measure changes  in toxicity but could not provide information
on specific pollutants, such as BCC effects on human health.  Flexibility is
also important because EPA expects that scientific advances in understanding
the effects of, and ability to measure pollutant alterations, will change over
time.  EPA is aware that lack of  specificity may create  uncertainty from the
permittee's perspective.  However, as addressed in the discussion on
demonstrating that the  "no additional mass" requirement  under the intake
pollutant reasonable potential procedure is met, uncertainty should not create
unknown liability as some commenters  feared.  For this reason, EPA encourages
permitting authorities  to develop a strategy or additional guidance on  what  is
needed to make  this demonstration.  As discussed in section VIII E.7.a.iii
above, EPA believes that it  is appropriate  to require the permittee to
establish eligibility for special consideration on intake pollutants.

      viii. Timing/Location Requirement

      The final  condition  in the  proposed Guidance that the permittee had to
demonstrate to  be eligible for intake pollutant relief is that the timing and
location of the  effluent discharge does not cause adverse water  quality
impacts to occur from the  simple  pass-through of an intake water pollutant
that would not   occur   if the pollutant were left in-stream.  EPA explained
that it did not  expect  the timing or  location of the discharge to cause
adverse water quality impacts in  most instances, but that the possibility

-------
                           Section Vffl.E: Reasonable Potential                       381
should be considered in each instance to ensure protection of the receiving
water.  An example where timing could adversely affect the receiving water
would be if intake water were withdrawn at high flow conditions and discharged
at low flow conditions or withdrawn at high tide and discharged at low tide.
EPA invited comments on all aspects of this condition including whether the
final Guidance should specify a maximum distance between the intake and
outfall or a maximum time interval between intake and discharge to be eligible
for the proposed intake pollutant relief.

      Although EPA received several comments on this condition requesting
additional guidance on its application, no commenter advocated more
specificity in the final Guidance, particularly with regard to identifying the
maximum location between the intake and outfall points.  One commenter
advocated against the timing and location condition as regulation or guidance
because its purpose was purely for protection of the stream from ill effects
from the withdrawal and discharge of the water not associated with industrial
processes.  For similar reasons, another commenter noted adverse affects due
to timing and location of the discharge is likely to be a very site-specific
phenomena; therefore, rather than adopt a firm rule, require consideration
solely  of whether there is an adverse change.

      The final Guidance at B.D.S.b.v. is the same as the proposed.  EPA
agrees with commenters that determining whether timing or location cause
adverse impacts that otherwise would not occur requires consideration of site-
specific factors that cannot be easily reduced to a general rule.  EPA does
not agree that the site-specific nature of making this determination is
sufficient reason not to require the analysis.  Similarly, EPA does not
consider the fact that the "timing/location" requirement focuses on an aspect
of the discharge not directly related to an industrial process to be
sufficient reason  to  disregard potential adverse effects caused by the
timing or location of the discharge, especially for deciding whether the
discharge causes or has the reasonable potential to cause or contribute to an
excursion above an applicable WQS.

      ix.   Relationship Between this Section and Other Reasonable Potential
            Sections

      The proposed Guidance included the reasonable potential procedure for
intake pollutants as an alternative to the  reasonable potential procedure in
proposed sections 5.A-D of appendix F  (the baseline reasonable potential
procedures).  In discussing options for considering intake pollutants that
addressed limit derivation (options 2, 3 and 4), the preamble did not
separately address the question of reasonable potential.  In other words,
these options assumed that reasonable potential would exist and WQBELS were
necessary.

      In most cases where the final Guidance now allows for consideration of
intake pollutants in limit derivation, reasonable potential will be found to
exist regardless whether the baseline reasonable potential procedure or the
intake pollutant reasonable potential procedure is used.  However, in some
cases where intake pollutants are at issue, the baseline reasonable potential
test may find that reasonable potential does "not" exist, even in non-
attainment waters where the pollutant of potential concern is present in the
discharge.  An example is where the discharge contains extremely low
concentrations of the pollutant of concern, the discharge has a high flow rate
when compared to the receiving water flow rate, and the impact of the
discharge on the receiving water is sufficient to bring it back into
attainment with the applicable WQS.  This result is most likely in the
different body of water scenario,  but could occur in the same body of water
scenario if the facility has a significant source of clean water  (i.e.,
without the pollutant of concern)  that contributes to the discharge or its
treatment system removes most,  but not all o'f the pollutant.  The intake
pollutant provision is intended to be an exception to the usual reasonable
potential procedures in appropriate circumstances; therefore, it would not

-------
382    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


make sense to require WQBELs in those situations where reasonable potential
would not be found under the baseline reasonable potential procedures
notwithstanding the result that would be reached under the intake pollutant
procedures.

      Under the final Guidance, States and Tribes have the discretion to make
both reasonable potential procedures available or, alternatively, to determine
the order in which they apply the baseline and intake pollutant  reasonable
potential procedures.  States may also opt not to require a separate
reasonable potential determination before imposing WQBELs in those situations
covered by the intake pollutant limit derivation procedures (i.e., background
of the receiving water exceeds criteria and the discharge contains the
pollutant for which criteria is exceeded).  Where the discharge  does not
qualify for special consideration of intake water pollutants under procedure
5.D.3.b, e.g., the intake water pollutants are from a different  body of water,
procedure 5.D.3.d provides that "reasonable potential" will be determined
using the procedures in 5.A-C of appendix F.

b.    Issues Specific to the Intake Pollutant Reasonable Potential Procedure

      i.    No Mass Added Requirement

      The proposed Guidance limited the intake pollutant reasonable potential
procedure to those .situations where the presence of the pollutant of concern
in the discharge was due solely to its presence in the intake water from the
same body of water as the discharge  (sometimes referred to as the "simple pass
through scenario).  Proposed procedure S.E.l.b stated that the facility must
not contribute any additional mass of the identified intake water pollutant to
its wastewater.  The preamble further explained that this determination should
be made based on monitoring data and information on the kinds of pollutants
generated by the particular type of facility.

      The question whether contributing additional mass of the pollutants
should preclude intake pollutant relief absolutely generated many comments.
As discussed above,' EPA carefully considered these comments and  the final
Guidance provides intake pollutant relief in situations that are
environmentally the functional equivalent of the simple pass through scenario
outlined in the proposal.  That is, where all requirements for a finding of
"no reasonable potential" are met, but the facility additionally contributes
the pollutant of concern to the wastestream, the permitting authority may
establish limits based on a  "no net addition" principle.  Thus,  a facility can
contribute additional mass of a pollutant to its wastestream in  any amount,
but permit limits will be set to ensure that the discharge contains no more
mass of that pollutant than had been  in the intake water.

      Where a facility  is simply  "passing through" a pollutant taken from the
waterbody and other conditions under the final Guidance are met, discharge of
intake pollutants may not cause or contribute to violations of water quality
standards.  Where, however, a facility is adding some of the pollutant of
concern to its wastestream, and the effluent as a whole contains levels of
pollutants that exceed  applicable criteria, then EPA believes that it is
reasonable to conclude  that the facility is or may be contributing to the
exceedance of water quality standards in the waterbody.  EPA believes that the
latter  conclusion is reasonable as a  logical matter:  i.e., a contribution by
a facility of some amount of a pollutant to a wastestream exceeding applicable
criteria into a waterbody that is in non-attainment can reasonably be viewed
as contributing to the  continued violation  of the standards for  the waterbody.

      EPA also believes that this conclusion is appropriate as a technical and
policy  matter.  EPA's regulations provide that WQBELs must be established for
all pollutants that  "are or may be discharge at a level which will cause, have
the reasonable potential to  cause, or contribute to" a violation of water
quality standards.  40  C.F.R.  § 122.44(d)(1)(i)  (emphasis added).  The
emphasized  language in  this  regulation underscores the cautionary nature of

-------
                           Section Vm.E: Reasonable Potential                       383
the reasonable potential inquiry, which necessarily asks the questions whether
there may be a possibility that a discharge will cause or contribute to a
violation of standards.  Such a cautionary approach is consistent with the
protective nature of the CWA, and EPA does not believe that it would be
consistent with the statutory scheme to establish an unduly high threshold for
the reasonable potential determination, and require certainty that a
particular discharge will cause or contribute to a violation of water quality
standards prior to impose a water quality-based effluent limitation.  Where
the permitting authority determines that, in fact, a particular discharger is
not contributing any additional mass of a pollutant to a wastestream
containing intake pollutants exceeding water quality standards but is simply
"passing through" those pollutants, EPA believes there is a reasonable degree
of certainty that the discharge of the facility will not contribute to further
exacerbating the ndn-attainment in the waterbody.  Where, however, it is
demonstrated that the facility is adding some amount of the pollutant to its
wastestream, EPA is not comfortable concluding that the facility will not ever
contribute to a violation of water quality standards.  Even if particular
monitoring results indicates that the total amount of a pollutant in the
discharge is no greater than the amount in the intake water (because, for
example, the facility has both added and removed some of the pollutant of
concern), the fact that the facility is itself contributing some amount of
that pollutant to the effluent creates a reasonable possibility that the
effluent may exceed the amount of the pollutant in intake water under some
circumstances.  In .the latter case, EPA believes that the prudent and
reasonable approach is to impose a WQBEL to ensure that this eventuality does
not occur.

      Most comments on the "no additional mass" requirement focussed on the
fundamental issue whether an addition of mass should preclude all intake
pollutant relief.  These comments are addressed above.  However, the question
whether the facility contributes additional mass remains important for
purposes of determining whether or not a WQBEL is necessary.  In the preamble
to the proposed Guidance, EPA solicited comments on several aspects of
determining how this requirement should be demonstrated, including:  how
"contribution of nd additional amount" should be interpreted; what data would
be needed; the  use of statistical information; and whether the final Guidance
should specify minimum data required.     Commenters suggested a variety of
approaches for determining whether additional mass has been added.  Some
endorsed the idea of using statistical procedures to compare influent and
effluent data.  One commenter submitted a detailed statistical methodology,
supported by several other commenters, for making this determination and
another commenter specified that long-term averages should be considered in
this analysis.  One commenter suggested that the requirement be changed to no
"quantifiable" addition.  In a related vein, a commenter stated that
procedures for handling non-detects must be established, especially
considering that detection limits may differ between intake and effluent data
sets.  Commenters were divided on whether the Guidance should establish
minimum data requirements.  Some stated that more specificity was needed to
provide adequate notice of what was expected of dischargers, as well as their
potential liability for failure to act.  Others commented that "adequate" data
was the type of determination best left to the permitting authority's
discretion.

      Another group of comments involve suggestions for different
interpretations of "addition" to exclude certain types of additions from a
reasonable potential determination, i.e., to exclude certain discharges from
the need for WQBELs.  Many of the suggested categories overlap.  The broadest
categories suggested  were "de minimis" additions (defined in numerous ways)
and cooling waters in general, although most commenters would only accord
special consideration for non-contact cooling water.  Some commenters put
limitations or conditions on excluding cooling waters, such as allowing only
"tiny amounts" of metal-based algicides, or allowing exemption from WQBELs
only when all State technology-based requirements or industry design standards
to minimize corrosion have been met.  Other suggested categories for an

-------
384    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


exemption from WQBELs included:  pollutants added as result of corrosion or
erosion, with several suggestions that EPA use the  approach adopted in the
amended  effluent guidelines for Organic Chemicals, Plastics and Synthetic
Fibers  (OCPSF)  (57 PR 41836, 9/11/92) for constituents due to corrosion or
materials of construction that are not reasonably avoidable; pollutants not
added "intentionally" or "voluntarily," including "trace" amounts of
pollutants in raw materials; increases in mass due solely to. evaporation;
additions of mass that do not increase concentration; and automatic
determination of "no reasonable potential" for all pollutants but those added
by the facility as part of normal operations.

      The final Guidance, in procedure 5.D.3.b.ii, retains the requirement
that the discharger not add additional mass of the identified pollutant of
concern in order to qualify for a determination that its discharge does not
cause, have the reasonable potential to cause, or contribute to an excursion
above an applicable WQS and therefore does not need a WQBEL.  When the
receiving water exceeds standards for a pollutant, WQBELS are reasonable to
ensure that there are no additional loadings of the pollutant to the receiving
water beyond that already in the intake water.   Requiring that WQBELS be
established in cases of additions of mass is consistent with the CWA goal of
restoring non-attainment waters.  For this reason, EPA disagrees with the
comment that additions of mass should be allowed as long as concentrations do
not increase.  EPA also rejects the suggestion that additions of mass due to
evaporation be exciuded because evaporation does not increase mass  (although
it may be easier to detect mass when concentration is increased).

      As reflected in the comments, "addition" can be determined in a number
of ways depending on the situation.  In some cases, knowledge about a
facility's process and infrastructure is enough to determine that an addition
will occur.  For example, dioxin is a known by-product of certain processes
used in pulp and paper mills and it would be reasonable for a permitting
authority to find that dioxin is added even if it is not detectable in the
effluent.  For these situations, a reasonable approach would be to rely on
application data describing the processes used at the facility.  Similarly, as
noted by many commenters, certain metals are known to enter a wastestream
through corrosion or erosion when particular materials are used in a
facility's infrastructure.  Applying best professional judgment to make an
addition determination in these situations has the advantages of simplicity
and avoiding expensive monitoring.  Similarly, knowledge that certain
industrial processes do not use a particular pollutant and that the pollutant
would not be added in other ways also may be a sufficient basis for concluding
that a pollutant is not added.  In many cases, however, it may not be known
whether a pollutant is added.  In these cases, a statistical approach using
commonly acceptable procedures may be the best approach.  To provide
flexibility to deal with a variety of situations, the final Guidance does not
establish the minimum data needed to determine whether the "no addition"
requirement has been met.  The permitting authority must determine the most
appropriate approach, using its best professional judgment.  EPA expects that
this flexibility will allow permitting authorities to minimize the burden on
dischargers in many instances.  EPA strongly encourages permitting authorities
to develop a policy addressing how "addition" determinations will be made.
This policy can serve as advance notice to dischargers of what is expected for
them to demonstrate that the "no addition" requirement is met.  Note, however,
that EPA does not agree that leaving the question of demonstrating addition to
the discretion of the permitting authority deprives permittees of adequate
notice  of what is expected of them with regard to their discharge activities
and associated liability.  Such notice would be provided through any requests
for information made by the permitting authority authorized by existing
regulations, e.g., 40 CFR 122.21(e) and 122.41  (h), and/or by the terms of the
permit.

      EPA declines to adopt the many  suggestions for categorical exclusions of
certain types of discharges from the  reasonable potential determination and
would not approve these if included in a State or Tribal submission under part

-------
                           Section Vm.E: Reasonable Potential                       385
132.  All these suggestions acknowledge that the discharge contains additional
contributions of the pollutant beyond that already in the intake water, but
argue that these contributions should nonetheless be excluded from
consideration for water quality-based controls for various reasons.  Because
these suggestions, in some cases, argue that additions by the permittee should
not be regulated, such suggestions are not, in fact, related to the intake
pollutant issue.  Many of the suggestions seem based on the principle that
some discharges, while having an impact on the receiving water, are so
insignificant that they should be ignored.  EPA disagrees' that it is
appropriate to judge certain discharges as "insignificant" on a categorical
basis.  Instead, "significance" needs to be evaluated on a site-specific,
discharge-specific'basis, considering such factors as the status of receiving
water and other loadings of the pollutant, which will obviously vary.
Existing procedures available to the States, such as the baseline reasonable
potential procedures, already take into account the significance of the level
of the pollutant in the discharge in relation to receiving water quality.  EPA
and the States have acquired significant experience in determining "reasonable
potential" based upon the policies and procedures reflected in the TSD and
similar guidance, and this general approach is reflected in the baseline
reasonable potential procedures in procedure 5.A-C in appendix F of the final
Guidance.  EPA does not believe that it would be appropriate to graft onto
these technically sound procedures the additional evaluations of whether a
particular discharge would cause "significant" environmental effects.  As
demonstrated by the widely varied views of commenters about when and what
types of discharges would cause such effects, a "significance" test that is
acceptable, objective and detailed would be difficult to craft.  In EPA's
view, the baseline reasonable potential procedures provide the appropriate
mechanism for determining whether the discharge is environmentally significant
in the sense that it causes, has the reasonable potential to cause, or
contribute to an excursion above an applicable WQS.  As noted previously,
further consideration of the discharge's significance and related questions of
appropriate load reduction responsibilities are best evaluated through the
TMDL or comparable'process, as provided in procedure 3 .A of appendix F.  With
regard to pollutants added to the discharge because of corrosion and erosion,
EPA is not aware of a factual basis, nor was one provided by commenters, for
concluding that pollutants added through processes such as corrosion and
erosion or in cooling water have less of an impact than the same pollutants
added in other ways, or that their amounts are necessarily small in all cases.
This further supports EPA's decision not to categorically exclude such
pollutants from water quality-based controls.

      Suggestions that exclusions or de minimis determinations should be based
on the discharger meeting existing State technology-based standards or
industry standards to minimize leaching of metals from corrosion or erosion
are inconsistent with the fundamental premise of the CWA that more stringent
limits are required if technology-based limits are not sufficient to attain
water quality standards.  Similarly, suggestions that EPA adopt the approach
for regulating "incidental" metals used in the OCPSF technology-based
guidelines are not appropriate when WQBELs are at issue.

      Other suggestions for categorical exclusions, including those arguing
for exclusion of pollutants found in raw materials or added unintentionally,
not deliberately, or outside the normal operations of the facility, are
variations on the argument that dischargers should not be responsible for
certain pollutants in their discharge because they would be difficult or
costly to control.  As reflected by the allowance in the final Guidance for
limits based on "no net addition" and  based on a "combined wastestream"
approach when pollutants are added from other bodies of water, EPA agrees that
the addition of such pollutants should not preclude intake pollutant relief in
all situations.  EPA  believes, however, that the other circumstances cited by
commenters should not exempt discharges containing these pollutants from the
need for WQBELs if they pose "reasonable potential" under procedure 5.A-C.  As
discussed elsewhere, EPA does not believe that the intent of the discharger or
other factors cited by these commenters relieve in any way the responsibility

-------
386    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


of the discharge to comply with section 301(b)(1)(C) of the CWA.  In all cases
cited by cornmenters, facilities would be adding pollutants to a  waterbody that
would not be there but for the action of the discharger.

      EPA agrees that questions regarding measurements when the  levels of
pollutants in the wastestream are below detection  are important and may be
significant both when determining whether reasonable potential exists and when
determining compliance with permit limits.  These issues have become
increasingly common as  criteria below analytical detection levels have been
adopted and more permits include WQBELs, some of which are below detection
levels.  However, there is nothing unique to the intake pollutant situation
that would require'a different approach to these issues.  Procedures to manage
data when values are above and below detection levels or levels  of
quantification are addressed elsewhere in the final Guidance and this
document.  For example, procedure 3 of appendix F, discussed in  section VIII.C
of this document, addresses these issues when determining whether the ambient
background water exceeds criteria.  Section VIII.E addresses these issues when
determining whether a discharge causes, has the reasonable potential to cause,
or contribute to an exceedance of WQS.  Procedure 8 of appendix  F, discussed
in section VIII.H of this document, addresses WQBELs below the levels of
quantification.

      ii.   Other Requirements

      The proposed Guidance included three additional requirements for  intake
pollutant relief imposed on the permitting authority.  First, the permitting
authority must summarize in the NPDES permit fact sheet or statement of basis
the reason for determining that there is no reasonable potential for the
discharge of an identified intake water pollutant to cause or contribute to an
excursion above a applicable narrative or numeric water quality  criterion,
including an evaluation of the permittee's demonstration of the  five specified
conditions in the proposed Guidance.  Second, proposed procedure 5.E of
appendix F required that the permit require all monitoring of the influent,
effluent and ambient water necessary to determine that the required conditions
are maintained during the permit term.  Third, the proposal required that the
permit contain a reopener clause authorizing the permitting authority to
modify or revoke and reissue the permit if new information indicates changes
affecting any of the required conditions, e.g., obtaining a different source
of intake water or relocating the discharge to a different body  of water.

       (A).  Documentation.  EPA received few comments on the requirement to
document the basis for  the -reasonable potential determination in the NPDES
permit fact sheet or statement of basis.   Commenters stated that the proposed
intake pollutant reasonable potential procedure would deny the public an
opportunity to review and comment on decisions that would relieve dischargers
of pollution control obligations because it does not require an  opportunity
for public comment on any of the proposed required demonstration, nor does it
require public participation in the decision to grant an exemption; rather,
according to this commenter, the proposal merely required an after-the-fact
notice to the public of the agency's determination and its basis.   Commenters
further objected that EPA does not explain how this public notice will  be
provided when the discharge contains only intake pollutants.

      The final Guidance at procedure 5.D.3.c.i retains the requirement to
document  the intake pollutant reasonable potential finding in the statement of
basis or  fact sheet, as proposed.  This requirement is consistent with
existing  requirement at 40 CFR 124.7  and 124.8 to document all calculations
and the  rationale for the conditions  in a draft permit.  The requirement to
document  the intake pollutant reasonable potential determination ensures that
the fact  sheet or statement of basis  explains why certain limitations may not
be in the permit that would otherwise be expected absent special consideration
for intake pollutants  (i.e., WQBELs for pollutants that would be found  to
cause, have the  reasonable potential  to cause, or contribute to  an  excursion
above applicable water  quality standards under the baseline procedures).

-------
                           Section Vm.E: Reasonable Potential                       387
Procedure 5.D.3.c.i only applies to the reasonable potential determination.
Existing procedures governing development of NPDES permits and supporting
documentation sufficiently cover instances where intake pollutants are
considered in developing permit limits in accordance with today's final
Guidance.

      Contrary to commenters' assertion, the proposed, and final, provision
addressing the fact sheet and statement of basis ensures that the public
receives advance notice of a proposed finding that WQBELS will not be required
for particular pollutants.   Existing NPDES regulations require the permitting
authority to prepare a fact sheet  (40 CFR 124.8(a)) or statement of basis  (40
CFR 124.7) for each draft permit and make them available to any requester.
Further, the permitting authority must public notice the preparation of the
fact sheet  (40 CFR 124.10(a) (ii) )  and, in addition, send the public notice to
any person requesting to be on a State's mailing list  (40 CFR 124.10(a)(ix)).
The public notice must include the name of the person to contact to obtain
further information, including copies of the draft permit, the statement of
basis or fact sheet, and the application (40 CFR 124.10(d)(iv)).  Existing
regulations at 40 CFR 122.7(b) and  (c) also require that certain information
such as the permit application be publicly available.  All interested persons
have an opportunity to comment on the draft permit  (40 CFR 124.11) and to
submit oral or written statements at any public hearing (40 CFR 124.12(c)).
These procedures apply even if the permit does not include permit limits for
all pollutants in the discharge.

       (B).  Monitoring.  The preamble to the proposed Guidance explained that
monitoring is necessary to ensure that the conditions supporting a finding of
"no reasonable potential"  continue to be met throughout the term of the
permit.  Appropriate monitoring is necessary, for example, to identify changes
in the mass or concentration of an intake water pollutant or to identify a new
source of the pollutant of concern from a facility's processes.  The proposed
Guidance left the selection of appropriate monitoring parameters and
frequencies to the -discretion of the permitting authority to allow
consideration of the individual circumstances at each facility or within the
receiving water.  EPA also invited comment on whether the final Guidance
should specify minimum monitoring requirements for all facilities, whether
permitting authorities should be required to consider specified factors in
making this determination, or whether other permit conditions would be
adequate in lieu of the proposed monitoring provisions.

      EPA received comments on two aspects of the monitoring requirement:
whether more specificity was needed in the Guidance and what appropriate
monitoring methodologies should be.  Several commenters endorsed the approach
in the proposed Guidance to leave the details of monitoring requirements, to
the discretion of the permitting authority so that individual circumstances
could be considered.  One commenter requested that the permitting authority be
given discretion to remove the influent and receiving stream monitoring
requirements after the permittee demonstrates it is not responsible for the
pollutant.  Other commenters requested that the Guidance specify that
monitoring be based on a statistical comparison of the influent and effluent,
as opposed to straight comparisons of influent and effluent based on single
samples, to account for normal variability in analytical methods, particularly
when values are close to the detection limit.

      EPA has  finalized the monitoring requirement as proposed.  EPA agrees
that factors such as variability and detection limits are important
considerations in establishing appropriate monitoring requirements. However,
establishing a set approach in the Guidance would unduly hamper the ability of
permitting authorities to determine appropriate monitoring for a variety of
situations.  EPA also declines to adopt the suggestion to specifically allow
the permitting authority to remove monitoring requirements when it determines
that the permittee is "not responsible" for the pollutant.  As discussed
elsewhere, it is EPA's view that a discharger is "responsible" for whatever
pollutants are in its discharge.  While special allowances may be appropriate

-------
388    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


under certain circumstances for intake pollutants, EPA does not  endorse the
view that the reasonable potential finding and related monitoring provisions
should be adjusted generally based on the facility's  "responsibility" for the
pollutant.  Moreover, the final Guidance gives considerable discretion  to the
permitting authority to determine the appropriate requirements regarding
monitoring parameters and frequency.  The purpose of monitoring  is to ensure
that the conditions for intake pollutant relief will  continue to be met
throughout the term of the permit.  Traditionally, such factors  as historical
data on the consistency of effluent quality and whether conditions affecting
the discharge are static, or subject to change, are considered in developing
appropriate monitoring frequencies.  Permitting authorities continue to have
discretion to consider such factors under the final Guidance.

      The monitoring provisions at 5.D.3.c.ii in the  final Guidance apply only
when the permitting authority has determined that the permittee's discharge
does not cause, have the reasonable potential to cause, or contribute to an
exceedance of applicable water quality standards, i.e., when WQBELs are not
included in the permit.  Existing NPDES regulations govern development  of
monitoring and reporting requirements when permit limits developed under
procedure 5.E.3-5 of the final Guidance are included  in the permit  (see, for
example, 40 CFR 122.41, 122.44, and 122.48).  Also, see procedure 5.E.3.3 of
appendix F with respect to compliance monitoring for  "no net addition"  limits
and related discussion in section VIII.E.7.c.iii of this document.

       (C).  Reopener.  EPA received no comments on the proposed  requirement to
include a reopener provision in the permit to allow the permitting authority
to modify or revoke and reissue the permit if changed conditions at the
facilities require different permit limits or conditions.  For example, a
facility may obtain a different source of intake water or may relocate  its
discharge into a different receiving water.  In these instances, limits may be
needed in accordance with procedures 5.E.3-5.  Similarly, monitoring may
demonstrate that an intake water pollutant is altered by some change in the
process wastestreara subsequent to permit issuance.  Here, also,  a WQBEL may be
appropriate given the changed circumstance.  The final Guidance  includes the
reopener provision as proposed to ensure that permitting authority can  take
steps necessary to evaluate new information and adjust permit requirements for
changed conditions without waiting for the permit term to expire.  This
reopener provision is consistent with the authority to modify permits because
of new information under 40 CFR 122.62(a)(2).

      As with the documentation and monitoring provisions, the reopener
provision at 5.D.3.c.iii is specific to the situation where the  permitting
authority has determined that a WQBEL is not needed for an intake pollutant in
a facility's discharge.  Existing NPDES regulations require reporting of
changed conditions  (40 CFR 122.42(1) (1)) and allow inclusion of  a reopener
provision in the permit, which authorize the permitting authority to modify or
revoke and reissue the permit if changed conditions warrant new  or different
permit requirements  (40 CFR 122.62).  These existing  regulations would  govern
permits that include permit limits based on the procedures in 5.E.3-5.

c.     Issues Specific to Consideration of Intake Pollutants in the Derivation
       of WOBELs

       i.    Availability Only for Non-Attainment Waters

       Of the options discussed in the preamble addressing development of
WQBELs, option 2 was silent on whether it was restricted to situations  where
the background concentration of the receiving water exceeded criteria,  while
option 4 was specifically  restricted  to these situations.  Some  commenters
endorsed  limiting intake credits  to situations where  the intake  water exceed
criteria, with some  stating this was  appropriate when the intake source was
the same body of water  as  the receiving water while others did not make this
distinction.  Other  commenters who generally favored  option 4 stated that  it
should not be limited to non-attainment waters and requested the same relief

-------
                           Section VHI.E: Reasonable Potential                       389
for waters close to exceeding the criteria.  Commenters further explained that
the basis for this 'request was that conservative assumptions used in
determining reasonable potential meant that WQBELs could be required in some
cases even if the receiving water was in attainment, and that relief in this
situation was needed to avoid compliance problems dealing with variability and
to avoid requiring dischargers to remove pollutants they did not add.  Another
commenter stated that it may be legitimate to impose a permit limit or.
condition restricting an increase in the discharge beyond that contained in
the intake water, except where the pollutant concentration in the intake water
is below the applicable water quality criterion, in which case, the discharger
should be allowed to increase the pollutant load up to the criteria and a
WQBEL will not always be justified.  Finally, several commenters argued in
effect that WQBELs should not be required at all in non-attainment waters
because "the technology-based limits would become essentially useless because,
in addition to dealing with its own pollution, the facility would be required
to have technology to combat many forms, and varying degrees, of pollution
without workable industry standards to guide them."

      The intake pollutant reasonable potential procedure can be used
regardless of the status of the receiving water since it looks solely to
whether a pollutant discharge that originates in the same body of water has
any adverse effect that would not have occurred if the pollutant were left in-
stream.  Whether silch an effect occurs is not related to the status of the
receiving water.  However, in the final Guidance, EPA is limiting no net
addition limits to the situation where the background concentration of a
pollutant in the receiving water (i.e., the quality of the water upstream of
the discharger) exceeds the applicable criterion for that pollutant.  This
situation is also referred to as non-attainment waters.  To ensure
consistency, the procedure for determining whether background exceeds criteria
is the same as for determining background concentration in developing a TMDL,
as established in procedure 3 of appendix F of the final Guidance.

      EPA's decision to limit intake credit relief to non-attainment waters is
closely linked to its rationale for making intake pollutant relief available,
that is, special consideration for intake pollutants is reasonable when other
sources are the primary cause of the impaired water body to which the point
source is discharging and the discharge itself effectively has no further
adverse impact on the receiving water than that which already existed.  Not
providing special consideration for intake pollutants when the background
water quality exceeds criteria in this situation would require the discharger
to remove intake or background pollutants.  Although EPA believes that
requiring removal of background pollutants could be justified as furthering
the restoration goals of the CWA, it believes that generally a more efficient,
effective or equitable approach for attaining the desired water quality would
be the fuller consideration of all sources and control strategies contemplated
by the TMDL or a  comparable assessment and remediation plan as provided in
procedure 3.A of appendix F.  In contrast, special consideration of intake
pollutants is not needed to avoid requiring removal of background pollutants
when the background quality of the receiving water is in attainment with water
quality criteria.  Indeed, in attainment waters, applying the procedures for
considering intake pollutants in setting WQBELs would produce more stringent
limits than would result from applying "baseline" procedures for setting
limits that do not account for intake pollutants.  This can be seen by
comparing the wasteload allocation that would result from application of EPA's
Technical Support Document for Water Quality-Based Toxics Control  (TSD) to the
wasteload allocation that would result from application of the no net addition
approach in the final Guidance.  (The TSD was used for this analysis because
it formed the basis for the GLI "baseline" reasonable potential and WLA
procedures.)  The wasteload allocation is the discharge concentration that is
necessary to meet the WQS applicable to the receiving water.  EPA found that
the TSD steady-state wasteload allocation procedures would always generate a
less stringent wasteload allocation than background.  Specifically, using the
TSD resulted in limits above background (and above criteria) when waters are
in attainment with WQS whereas the intake pollutant procedures would result in

-------
390    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

limits at background.  Limits below criteria (but above background) are
possible using the TSD procedures, but only in situations where the State or
Tribe has reserved assimilative capacity for future growth or for similar
reasons  (e.g., used very conservative assumptions to account for significant
uncertainty).  EPA thinks it is appropriate to support a State's decision to
reserve assimilative capacity for attained waters.  Additionally, reservation
of assimilative capacity usually occurs in the context of TMDL development;
the assumptions used and WLAs developed as part of an approved TMDL (or
comparable plan approved under procedure 3.A of appendix F) would govern over
the intake credit provision in any event.

      As a practical matter, using the TSD procedures in attainment waters to
develop WQBELS rather them the intake water pollutant provisions is consistent
with the comment that limits above background  up to criteria should be
allowed in attainment waters  as long as compliance with such limit eliminates
the discharge from causing, having the reasonable potential to cause or
contribute to an excursion above the applicable State or Tribal WQS. When the
source of the intake water pollutant is the same body of water as the
receiving water, no net addition does not require removal of background
pollutants.  In response to a commenter's concern about compliance problems
associated with variability, EPA notes that variability is always a potential
problem in determining ambient pollutant concentration values.  However, the
procedures for determining background levels in procedure 3 of appendix F  (the
TMDL procedures) adequately account for consideration of variability for
purposes of determining whether the ambient levels exceed criteria.
Variability also may be considered in limit derivation, as discussed in more
detail below.  In addition, concerns about variability can be overcome in most
cases through collection of additional data, which a discharger is free to
collect and submit for consideration.

      EPA does not agree with comments that WQBELs should not be required in
non-attainment waters because they would effectively make technology-based
limits useless and dischargers would not have workable industry standards to
guide them given the varying amounts of pollutants they might be expected to
control.  The need for WQBELs in non-attainment waters is implicit in the
requirement of section 301(b)(1)(C) of the CWA to require more stringent
limits  (i.e., water quality-based limits) when technology-based limits are not
sufficient to attain the applicable WQS.  Because the need for WQBELS is
inherently site-specific, and varies over time as ambient water quality
changes, it is not feasible to develop national guidelines on control
strategies for particular industries.

      ii.   Partial Credit at Discretion of Permitting Authority

      In the proposal, EPA solicited comments on whether options that would
provide a direct adjustment to permit limits to account for intake  pollutants
should preclude credit for pollutants removed from the intake water prior  to
use at the facility.  EPA explained that partial credit in this situation
would be consistent with the existing approach for technology-based limits,
which only allows partial credit.  In a related vein, EPA requested comments
on whether options 2 or 3 should be limited to situations where the discharge
of intake pollutants improves receiving water quality because the  discharge
contains a lower mass or concentration of the pollutant than that  found in the
intake water.  The rationale for limiting intake pollutant relief  in these
situations would be to  advance the restoration goals of the CWA.

      EPA received no comments on the option of limiting permit  limit
adjustments for  intake pollutants to instances where the discharge produced  a
net improvement  to the  receiving water.  Comments on partial credit came
primarily from industry and were divided.   Some commenters advocated partial
credit as a way  to address EPA's concerns that a no net addition approach  for
developing WQBELS would not necessarily  advance attainment of WQS  in non-
attainment waters, i.e., the  restoration goal.  According  to these commenters,
partial  credit could allow  a  facility to enjoy a  "windfall" that could occur

-------
                           Section Vffl.E: Reasonable Potential                       391
as a result of the mechanics of wastewater treatment  (i.e., when dirtier water
makes it easier or less costly to remove a higher percentage of a pollutant),
and still produce a net discharge of pollutants that is less than the amount
in its intake water.   Commenters further asserted that partial credit could
also be used to "level the playing field" if full credit would give a facility
an undue advantage over other facilities.  Several other commenters who
supported partial credit urged the Agency not to focus exclusively on
pollutants removed through pretreatment of intake water because if the
discharger were facing new, more stringent limits to meet no net addition, it
may be more cost effective or technically simpler to achieve the required
removals through pre-treatment rather than post-treatment.  For this reason,
these commenters advocated giving the permit writer discretion to consider the
overall system in determining whether full credit was appropriate.

      Commenters who opposed the partial credit concept also focussed on a
discharger's potential lack of flexibility in devising treatment strategies if
intake pollutant relief were unavailable for pollutants removed through pre-
treatment of intake water.  One commenter objected that this would "result in
the Agency becoming involved in minor elements of plant engineering, beyond
that appropriate for a regulatory agency."  Similarly, another commenter
stated that allowing any removal of intake pollutants by the facility's
wastewater treatment system to be replaced by an increase in the process
pollutant loading to the wastewater treatment system is not unique to
dischargers with pollutants in their intake water; all permittees can increase
the loading to their wastewater treatment systems so long as those systems can
effectively remove that additional loading.  Other commenters asserted that
EPA can only require removal of intake pollutants when removal is incidental
through required treatment of pollutants added by a facility.  Several also
argued that adjusting intake credits for pollutants removed through pre-
treatment of intake water effectively penalizes dischargers which pre-treat.
These commenters asserted that this would be particularly inappropriate given
that some dischargers which pre-treat intake water will discharge less than
the mass of the pollutant removed from the intake water, thereby creating a
potential for cleaner effluent as compared to the influent.

      EPA has determined that allowing only partial credit for intake
pollutants where a facility would remove those pollutants in any event is
consistent with the CWA goal of restoring non-attainment waters.  The decision
to authorize limits based on a no net addition approach in the absence of a
TMDL or comparable assessment and remediation plan for a period of up to 12
years from the publication date of the Guidance reflects in part a recognition
that requiring dischargers to remove intake pollutants under certain
conditions will not necessarily result in attainment of water quality
standards in non-attainment waters.   However, a policy of not compelling a
discharger to remove intake pollutants is not relevant when the discharger
would remove intake pollutants in any event.  Similarly, EPA believes that not
extending full credit when a discharger incidentally removes intake pollutants
in the course of normal operation and maintenance of its treatment facility is
appropriate.  Other policy considerations come to the forefront when
considering full credit.  Here, EPA believes that extending full credit, that
is, allowing the discharger to add more process pollutants to offset the
amount that would be removed from its wastewater, unnecessarily impedes the
goal of restoring impaired waters.  Therefore, procedure 5.E.2.b of the final
Guidance provides that the permitting authority may establish limits lower
than background levels of the pollutant in the intake water where removal of
the pollutant occurs as a result of the normal operation and maintenance of
the facility's treatment system.  At the same time, EPA agrees that it should
avoid, where possible, interfering with plant-level determinations of whether
reductions to meet new discharge limits can best be met through pre-treatment
or post-treatment.  Permit writers should have discretion to take facility-
specific factors into account when determining whether full credit is
appropriate.  In addition, EPA is aware that determining whether, and to what
extent, a facility .would remove an intake pollutant in the course of normal
operations can be technically difficult or infeasible for certain facilities

-------
392    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

with complex processes.  Accordingly, the final Guidance in 5.E.2.b provides
that, in determining whether to grant "partial" credit, the permitting
authority has the discretion to consider factors such as removal of the
pollutant achieved by proper operation and maintenance of the treatment system
and the feasibility of establishing  limits that reflect a "partial" credit.

     'iii.  Develdt)ing the Limit

      (A) .  Technol'ocrv-based vs. Water Quality-based Limits.  Nothing in the
final Guidance changes the fundamental principle of permitting under the NPDES
program that the effluent limit for any pollutant in the discharge should be
the more stringent of the technology-based limit or the WQBEL.  The permit
writer should develop both types of the limits for the pollutant of concern to
determine which is the more stringent.  In developing the limits for
comparison, the permit writer should consider all applicable implementation
procedures for developing each type of limit independently before determining
which is more stringent.  For example, if the permittee has requested and
qualifies for intake credits for a technology-based limit under 40 CFR
122.45(g), the permitting authority would determine the appropriate adjustment
to the technology-based limit considering the partial credit provision in
122.45(g)(3).  If the permitting authority further determines that the
pollutant is being discharged at a level that causes, has the potential to
cause,  or contribute to an excursion above an applicable water quality
standard and the background of the receiving water exceeds the applicable
criteria, the permitting authority would determine eligibility for special
consideration of intake pollutants and develop a WQBEL using the procedures in
5.E.3 of the final Guidance, including the  "partial credit" provision in
5.E.2.b, and 5.E.2.C if appropriate.  After this analysis is completed, the
permitting authority would then determine which limit is the more stringent
and include it in the permit.

      (B).  Expression of a Numeric Limit.  The preamble to the proposed
Guidance discussed different options for developing WQBELs considering the
presence of intake water pollutants, but did not discuss the specifics of
implementing those options.  The option chosen in the final Guidance, limits
based on "no net -addition," essentially requires that the permittee discharge
no more of the pollutant than that in the intake, or background, water supply.
One commenter suggested that the geometric mean of background water quality
can provide the basis for establishing daily maximum and monthly average
permit limits.  However, other commenters expressed concern that limits based
on background water quality could create compliance problems unrelated to the
discharger's actions because of variability in background water quality,
especially when there is a limited data base.

      The final Guidance leaves the details on actual development of permit
limits to the discretion of the permitting  authority.  Numerical limits may be
derived from data on ambient water quality  or influent samples, with
appropriate consideration for variability.  Permitting authorities have
considerable experience in judging the adequacy of data and in using available
data to develop defensible limits.  Seasonal limits can be developed to
account for anticipated seasonal  changes in background water quality.
Similarly, permit limits can be modified if warranted by changed conditions.
In some instances where variability and data available or quality is of
particular concern, the permitting authority can develop limits to implement
"no net addition" that establish  the numerical limit as "zero," with required
compliance monitoring based  on the differences between influent and effluent
samples.   (EPA recognizes that developing appropriate compliance monitoring
provisions in this situation may  not be a simple task because of such  issues
as detection levels and appropriately paired samples.  However, these  details
are more appropriately  left  to the permit writer's best professional judgment
based on the site-specific factors at the permitted  facility).  The final
Guidance at procedure 5.E.3.a requires the  permit to contain provisions
specifying how compliance with any limitations will be assessed.  This
language was added to the final Guidance in response to concerns that

-------
                           Section Vm.E: Reasonable Potential                        393
permittees would face uncertainty about  their obligations with respect to
potentially fluctuating levels of pollutants  in the intake water.   Permitting
authorities may also require additional  data  collection to better characterize
intake water quality, although the decision that more data is needed is not a
reason in this case for not including  a  limit in the permit that implements no
net addition.  In sum, considerable  flexibility exists in developing no net
addition limits that takes into account  the types of concerns raised by
commenters.

-------
394     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
 	Section Vm.F: Whole Effluent Toxicity	395

 F.     Whole  Effluent Toxicitv

 1.     Background

       As  discussed in section VIII.D of this document on Additivity,  a. focus
 on  individual pollutants  may not  provide complete protection of water quality
 because of the  potential  interactions between pollutants.   See procedure 4 of
 appendix  F to part 132.   Moreover, as a practical matter,  sufficient  data does
 not exist at this  time to develop chemical-specific criteria for each chemical
 that may  be  discharged in the waters of the Great Lakes System,  or to
 categorically conclude that  such  chemicals  are not harmful to aquatic life.

       Procedure 6  of appendix F to part 132 accounts for chemical interactions
 and for data gaps  regarding  individual chemicals  in an effluent by
 establishing procedures for  limiting the toxic effect on aquatic life from an
 effluent  as  a whole (known as "whole effluent toxicity" or "WET").  The whole
 effluent  approach  to toxics  control  for the protection of aquatic life
 involves  the use of acute and chronic toxicity tests to measure the toxicity
 of  wastewaters. An acute WET test is a comparative study in which organisms
 that are  subjected to different treatments,  such  as different amounts of
 effluent, are observed for a short period -- usually not constituting a
 substantial  portion of their life span.   Typically acute WET tests are run for
 a period  of  less than 96  hours and are evaluated  using measures of mortality.
 A chronic WET test is a comparative  study in which organisms that are
 subjected to different treatments are observed for a long period or a
 substantial  portion of their life span.   Typically chronic WET tests  are long-
 term tests in which sublethal effects,  such as impaired fertilization,  growth,
 or  reproduction, are measured, in addition  to lethality or immobilization.
 Aquatic organisms  used in the tests  may include,  but need not be limited to,
 invertebrates,  fish,  and  plants.

       Terms  commonly used to express the toxicity of an effluent include the
 lethal concentration (LC)  and the no observed effect concentration (NOEC).
 The LC50  is  the concentration of  an  effluent at which 50 percent of test
 organisms die in an acute WET test (e.g., if 50 percent of the test organisms
 die in 20 percent  effluent,  the LC50 = 20).    The NOEC is defined with respect
 to  a toxicity test with aquatic organisms,  as the highest concentration of
 toxicant  to  which  organisms  are exposed that causes no observable adverse
 effects.  The procedures  appropriate for calculating the NOEC and LC50 are
 described in the Technical Support Document for Water Quality Based Toxics
 Control,  March  1991 (March 1991,  TSD).   Typically,  the permitting authority
 can use a series of tests-run at  different  dilutions of effluent to determine
 the NOEC, or derive the Inhibition Concentration  at 25 percent effect,  IC25,
 from a statistical analysis  of the raw test data.   The IC25 is defined as the
 toxicant  concentration that  would cause a 25 percent reduction in a non-
 quantal biological measurement for the test population.   The data presented in
 appendix  A of the  March 1991 TSD  demonstrate that the IC25 is comparable to
 the NOEC  derived using the hypothesis testing of  a dilution series.   Other
 commonly  used terms are acute toxic  units (TO.) and chronic toxic units  (TU0) ,
 which  are defined  as follows:

       TO, =  100/LC50

       TUC =  100/NOEC or 100/IC25

       For example,  an acute  WET test with an LC50 at 20 percent effluent
 translates to 5 TU.'s.  A chronic WET test with an NOEC or IC25 at 10 percent
 effluent  translates to 10 TUc's.

     Procedure  6 of appendix F to part 132  provides specific provisions for
 controlling  the WET of discharges to the Great Lakes System.   The procedure
•contains  four sections: criteria  for WET, appropriate test methods to measure
 WET, permit  conditions, and  reasonable potential  procedures for determining
 whether or not  limits for WET are necessary.  The final Guidance on WET only

-------
396    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

supplements for the Great Lakes States.and Tribes, and does not  replace, the
regulations at 40 CFR 122 .44 (d) .(1) .   In addition, the March 1991 TSD contains
guidance relevant to topics not addressed in procedure 6 of appendix F  to part
132.

      The WET provisions apply to all facilities regardless of the cause of
toxicity, including toxicity caused by any "excluded pollutants" listed in
Table 5 of part 132 (see 58 FR 20969) and also including pollutants other than
the Table 6 of part 132 pollutants of initial focus.  A contrary result would
seriously limit the effectiveness of  the WET procedure since the purpose of
WET control is to limit the toxic effect of the combination of all pollutants
in a wastestream, without the need for identifying the individual pollutants
contributing to the toxicity when assessing compliance.

      WET is in many ways a unique pollutant parameter that requires different
implementation techniques than are used for other pollutants.  Procedure 6 of
appendix F to part 132 includes a complete reasonable potential  section that
is tailored to WET determinations.  The Total Maximum Daily Load (TMDL)
section, procedure 3 of appendix F to part 132, in general is not applicable
to WET control because it does not account for the peculiarities of WET.  For
example, techniques have not yet been developed for adding the whole effluent
toxicities associated with individual dischargers.  Individual toxicity
measurements may be based on different species, and may be independent  of one
another because they involve widely different classes of chemicals.
Accordingly, EPA is not specifying in this rule a TMDL methodology applicable
to WET.  Nevertheless, procedure 6 of appendix F to part 132 incorporates by
reference certain important components of the TMDL procedure, including
applicable design flows for aquatic life protection and provisions for  chronic
mixing zones.  In addition, the WET procedure addresses site-specific
conditions associated with WET and, therefore the site-specific  provisions in
procedure 1 of appendix F to part 132 are not applicable to WET.

2 .    Criteria for WET

      a.    Proposal: The proposed Guidance prohibited any discharge from:  (1)
exceeding 1.0 acute toxic unit (TO.)  at the  point of  discharge,   (2)  causing or
contributing to a receiving water exceeding 1.0 chronic toxic unit  (TUC)
(subject to certain exceptions) and  (3) causing or contributing  to an
excursion above any numeric WET or narrative criterion, such as  the free-from
toxics narrative, within State or Tribal water quality standards.  The
proposed Guidance did not require Great Lakes States or Tribes to adopt
numeric criteria for WET.  The proposed Guidance, rather, specified minimum
criteria objectives, 1.0 TU, (end-of-pipe)  and 1.0 TUC (edge of chronic  mixing
zone) that applied when either narrative criteria or numeric WET criteria were
involved.  Like the Federal regulations at 40 CFR 122.44(d)(1)(iv) or  (v) ,
which reflect the ability of States and Tribes to control WET with either
numeric or narrative criteria, respectively, the proposed Guidance allowed the
Great Lakes States or Tribes to choose the preferred form of criteria to
implement in their State or Tribal procedures.

      b.    Comments:  Several commenters expressed support for  allowing a
State or Tribe to adopt either numeric WET criteria or an interpretation of  a
narrative criterion into their water  quality standards.  These commenters
expressed a desire to maintain the flexibility currently provided  at 40 CFR
122(d)(1)(v) when implementing narrative as opposed to numeric WET criteria.
These regulations provide that a permit need not  contain a  limitation for WET,
notwithstanding a determination that  a reasonable potential exists for  the
discharge to cause or contribute  to an exceedance of a narrative criterion,  if
the permitting authority  can demonstrate that controls on individual
pollutants will ensure attainment  of  the narrative criterion.

      Numerous commenters supported the establishment of specific  minimum
criteria objectives that  States or Tribes would  need to achieve  through
implementation of either numeric  or narrative criteria.  A  few commenters

-------
                   	Section Vm.F: Whole Effluent Toxicity	397

supported the required adoption of numeric WET criteria to ensure consistency
among States and Tribes.

      c.    Final Guidance:  Based on existing rules and guidance and upon
consideration of the comments received, EPA has decided to allow the States
and Tribes to adopt into their water quality standards either numeric WET
criteria or narrative criteria providing for the protection of aquatic life
from toxicity. However, those States or Tribes choosing to rely on narrative
criteria must also adopt into their water quality standards specific numeric
interpretations of those criteria.  The effect of either approach would be the
same -- control of acute toxicity to 0.3 TUa and  chronic  toxicity to 1.0  TUC,
either end-of-pipe or at the edge of appropriate mixing zones.  This approach
will ensure consistency in application of WET requirements among the Great
Lakes States and Tribes, and will allow the States and Tribes to maintain the
discretion they currently have under existing rules to control toxicity
through implementation of either a narrative or a numeric criterion.  An
explanation of EPA's choice of the 0.3 TUa  and 1.0  TU0values to protect
aquatic life at the edge of mixing zones is provided in the following two
sections of this document.

3.    Acute Toxicitv Control

      a.    Proposal:  The proposal provided that no discharge be allowed to
exceed 1.0 TO, at  the  end-of-pipe  to protect aquatic  life in the  receiving
stream and ensure minimal exposure to acutely toxic conditions.  This
provision was based on the current EPA guidance that 0.3TUa is necessary  to
protect aquatic life, and assumed a maximum 3:1 dilution by the receiving
water.

      b.    Comments:  Many commenters disagreed with the proposed lack of
flexibility regarding the use of acute mixing zones, also referred to as zones
of initial dilution, when developing water quality-based effluent limitations
(WQBELs) for acute WET.  They stated that there was no justification for
imposing more stringent limitations on acute toxicity in the Great Lakes
System than recommended in the March 1991,  TSD because the impacts from acute
toxicity would not be unique to the Great Lakes System.  Also, several
commenters stated that the proposed 1.0 TO, value would be an  overly
protective value when applied end-of-pipe and would not be necessary to meet
the 0.3 TO. value  in the  receiving water.   Several  commenters  supported the
use of acute mixing zones to account for site-specific conditions and believed
the 0.3 TO. criterion  applied  to the edge of the  acute  mixing  zone provides
sufficient protection of the receiving water.  Other commenters supported the
proposed limitation on acute mixing zones for WET based,  in part, on their
view that the Great Lakes Water Quality Agreement  (GLWQA) recommends the
elimination of acute toxicological effects in all portions of the receiving
water.

      Based upon a review of the comments and the existing National rules and
policies, EPA has determined that acute WET mixing zones are appropriate for
use in the Great Lakes System.  The current national policy,  as articulated in
the March 1991, TSD, envisions the use of mixing zones in developing WQBELs
for WET, provided the acute mixing zone effectively minimizes aquatic life
exposure to acutely toxic concentrations of pollutants.  EPA considers acute
mixing zones to be protective of water quality when the physical conditions
within the acute mixing zone bar organisms from being present for sufficient
time to elicit a toxic response.   It is important to keep in mind that an
acute mixing zone is not appropriate in all instances.  Lack of flow in the
receiving water for dilution or proximity of the discharge to areas of
ecologic importance may preclude placement of an acute mixing zone in some
cases.  However, within these constraints,  dilution may be considered in
developing acute WET limits in NPDES permits.

     EPA will review State and Tribal mixing zone policies as part of its
periodic review of State and Tribal water quality standards.   Moreover, EPA

-------
398    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

will review individual State and Tribal mixing zone determinations when it
reviews State and Tribal NPDES permits.  EPA has the authority to object to
State and Tribal permits it considers inadequate to meet the requirements of
the Clean Water Act  (CWA)..

      It should also be noted that by establishing an FAV cap for individual
pollutants, WET in'each discharge should be substantially reduced, and  fewer
WET limits are likely to be required.  As a result, less variability in
regulation of WET discharges should ensue, notwithstanding the flexibility
allowed under the final Guidance with respect to establishment of acute mixing
zones.

      EPA has determined that the provision to adopt a numeric or narrative
criterion of 0.3 TUa with the allowance  of  acute  mixing zones where
appropriate, and the ability to select test species appropriate  for the local
biological community, will provide sufficient site-specific flexibility as
well as promote the consistent application of acute WET permit limits.  The
0.3 TO,  criterion is based upon  the  evaluation  of over  1200  toxicity tests
with over 100 chemicals and species from several families.  EPA  has determined
that at least 90 percent of the species subjected to an acute WET test  would
have survival rates of 99 percent if exposed to 0.3 TUa.   EPA believes  that
the level of protection associated with 0.3 TO. is protective of  aquatic life
and is strongly supported by the large body of data referenced in the March
1991, TSD.

      c.    Final Guidance:  EPA has established under procedure 6.A.I  of
appendix F to part 132 a provision that a numeric acute WET criterion of 0.3
TU.,  or  a  numeric interpretation of  a narrative at  least as  stringent as the
acute WET numeric criterion, be adopted by States and Tribes.  This criterion
would apply to the receiving water and would allow for the use of acute mixing
zones consistent with EPA approved State mixing zone provisions.  As described
in the TSD, an acute mixing zone is designed to minimize aquatic life exposure
to acutely toxic conditions.  This is accomplished by:  (1) locating the acute
mixing zone to avoid contact with immobile organisms,  (2) use of high rate
diffusers or other means to create a well mixed and sufficiently turbulent
flow regime to discourage fish and other organisms from entering the acute
mixing zone and  (3) requiring that drifting organisms such as daphnia will be
exposed for only a .short period of time, usually less than one hour.

      In cases where the available receiving stream dilution provides less
them a 3:1 dilution ratio, the WQBEL for acute WET can be no greater than  1.0
TO. -- the lowest acute  toxic value  that can be directly measured.   In  these
situations, the acute WQBEL itself will not be sufficient to ensure attainment
of 0.3 TU. in  the receiving water.   However control  of  chronic toxicity in
low-flow situations will generally prevent acutely toxic WET impacts.

4.    Chronic Toxicitv Control

      a.    Proposal:  The proposed Guidance,  required that a value of  1.0 TUC
be maintained at all points of the receiving water except  (i) within a  mixing
zone for aquatic life as defined in  section VIII.C of this document, or (ii)
in any portion of the receiving water for which a permitting authority  has
demonstrated that due to the site-specific physical and hydrological
conditions, it is unnecessary to apply any  chronic WET requirements  to  protect
aquatic life.  The  1.0 TU0 value is,  by definition,  the point at which no
effect is observed  in the  test  species when they are exposed to  undiluted
effluent.  The no effect determination can be performed using either
hypothesis testing or the  IC25  statistical procedure.

       The proposed provision for site-specific modifications was similar  to
that included in the proposed procedure 1 of • appendix  F to part  132, which
provided that States and Tribes could develop  site-specific  modifications  to
chronic aquatic  life criteria/values  for individual pollutants to reflect

-------
                         Section Vm.F: Whole Effluent Toxicity                      399

local physical or hydrologic conditions  (see section VIII.A of this document
on Site Specific Modifications to Criteria).

      b.    Comments:  Many commenters were concerned that the chronic WET
testing procedures were not sufficiently validated to be used in developing
NPDES permits; however no information was presented to support this assertion.
Other commenters supported the proposed flexibility for States and Tribes to
adopt either numeric or narrative chronic WET criteria.  Many commenters
stated that if the chronic limit requirements are retained, the States and
Tribes should be given flexibility as to how the NOEC is determined such as
using the IC25 or other methods supported in the March 1991, TSD.  Also some
commenters contended that the 1.0 TU0  value  could be  over-protective  because
of the apparent lack of correlation between the chronic lab tests and the
impacts observed in the receiving water.  Many commenters supported the site-
specific provisions for implementing the chronic WET limits.

      EPA believes that the chronic WET testing procedures are suitable for
supporting the inclusion of chronic WET limits in NPDES permits.  EPA has
documented in the TSD several studies which demonstrate that chronic toxicity
tests are comparable to chemical analyses in their reliability and
reproducibility.  In addition, EPA has provided in the final Guidance
flexibility in how the TDC value  can be  determined, and  supports  any  method
consistent with the provisions discussed in the TSD including the use of the
IC25.        EPA agrees with the commenters that site-specific factors are
appropriate to consider in matters regarding chronic WET.  EPA has decided,
however not to finalize the language addressing site-specific modifications in
the proposal.  The 'proposal would have allowed States and Tribes to completely
eliminate chronic WET requirements as a result of site-specific physical and
hydrologic conditions.  Upon further consideration of this matter, EPA has
determined that the general provision for documenting scientific defensibility
at 132.4(g), for application of any procedure in the final Guidance,  is
sufficient to address those rare instances where a permitting authority can
document that there are no aquatic species exposed to acute or chronic
toxicity.

      The final Guidance requires that States and Tribes use WET methods
published in 40 CFR part 136 for all purposes in implementing chronic WET
criteria.   These test methods allow the use of test species representative of
those capable of surviving under the physical and hydrologic conditions
present in the receiving waters.   Moreover, the part 136 test methods allow
site-specific aquatic chemistry to be taken into account in certain situations
by using site water to dilute effluent being subjected to WET tests.  EPA
believes that through use of the flexibility provided in the part 136 chronic
WET test methods, States and Tribes can appropriately take site-specific
situations into account in implementing WET controls under most conditions.

      c.    Final Guidance:  This provision is generally consistent with the
proposal,  with the-exception of the elimination of the site-specific
modification language, as described above.  The final Guidance allows the
States and Tribes to adopt either a numeric chronic WET criterion of 1.0 TOC
with provisions for allowing a chronic mixing zone for aquatic life,  or a
numeric interpretation of a narrative criteria that establishes 1 TUC as  the
necessary value to protect aquatic life from WET.  The 1.0 TDC value  is
defined as 100 percent effluent/NOEC or 100 percent effluent/IC25. The NOEC
and the IC25, are described in the March 1991 TSD.  By definition, the 1.0 TO,.
criterion as it would apply to the receiving water indicates that there is no
statistically significant or measurable chronic toxicity.  Clearly, this
criterion represents the threshold of measurable chronic toxicity and field
studies documented in the March 1991,  TSD indicate that this criterion is
protective of natural aquatic populations.

-------
400    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

5.    Numeric and Narrative Criteria

      a.    Proposal:  The proposed Guidance prohibited any discharge from
causing or contributing to an excursion above any State or Tribe adopted
numeric or narrative criteria for WET.  EPA added procedure 6.A.3 of appendix
F to part 132 to make it clear that the proposed Guidance on WET merely
supplemented f.or dischargers into the Great Lakes System, rather than
replaced, the requirements of 40 CFR 122.44(d)(1).  EPA had determined that
procedure 6.A.3 of appendix F to part 132 was necessary because there could be
instances where a Federally-approved State or Tribal water quality standard
had additional or more stringent requirements pertaining to toxicity than
those contained in the proposal.  Procedure 6.A.3 of appendix F to part 132
made it clear that these additional requirements would still need to be met.

      b.    Final Guidance:  There were no significant comments on this
provision.  However, procedures 6.A.I and 6.A.2 of appendix F to part 132,
explicitly require States and Tribes to adopt (1) numeric criteria or  (2) a
numeric interpretation of a narrative criterion as protective of aquatic life
as the numeric criteria.  These modifications remove the need for the proposed
provision, since States and Tribes are now explicitly required to adopt
numeric or narrative criteria as the basis for WET controls.  Moreover,
procedure 6.C of appendix F to part 132 ensures that the limitations will be
included in NPDES permits to attain and maintain these criteria.

6.    WET Test Methods

      a.    Proposal:  The proposed Guidance required that all WET tests be
performed in accordance with test procedures approved under 40 CFR 136.  The
proposed provision was consistent with current NPDES regulations at 40 CFR
136.1, requiring that analytical methods promulgated at 40 CFR 136 are used in
the NPDES permit program.  In the case of WET, there were no promulgated part
136 analytical methods at the time this rule was proposed.  When there is no
analytical method promulgated, permitting authorities have the discretion to
specify the method for use as allowed at 40 CFR 136.3.  EPA's recommended test
methods were referenced in the proposed Guidance.

      b.    Comments:  Some commenters requested that approved WET test
methods be adopted into 40 CFR 136 or be clearly documented and made
available.  In addition, some of  these commenters requested that States and
Tribes be allowed to exercise  flexibility  in selecting test species and other
parameters associated with the test procedures as currently allowed in
approved EPA test methods, -while  other commenters requested that the bounds of
flexibility in method application be defined.

      EPA agrees with the commenters who suggested that WET test methods be
adopted in 40 CFR 136 and intends to do shortly.  The methods EPA expects to
adopt under part 136 are those that were referenced in the proposed Guidance.
The methods provide some flexibility in conducting the tests.  For example,
the permitting authority can use  grab or composite samples in the WET  tests.
Best professional judgement is needed to determine what type of sample is
appropriate for a given discharge.  Also, the tests can be run by using
continuous flow-through of. the sample and dilution water or by periodic
replacement of the sample and dilution water.   Costs and the variability of
the effluent are key factors the  permitting authority may consider when
prescribing test procedures.   EPA also agrees  that there must be some bounds
on the State and Tribal flexibility to conduct  WET tests and has determined
that the test procedures adopted  under part 136  sufficiently define the  bounds
of flexibility.

     As discussed in the proposal, EPA expects  that WET monitoring will  be
performed consistent with  the March 1991, TSD.   At least three  species
representing three families  should be used  to assess WET toxicity, or
documentation must be provided  that one  or  two  species  is  sufficient to
adequately  characterize the  toxicity  of  an  effluent.  Compliance with  the WET

-------
                          Section Vm.F: Whole Effluent Toxicity                      401

NPDES permit limits is determined by using the test results of the most
sensitive species tested.

      c.    Final Guidance:  The final Guidance requires that part 136 methods
be used when assessing compliance with WET permit limits.  EPA expects to
publish part 136 WET test methods shortly.  In the unlikely event that such
methods are hot published prior to State or Tribal implementation of the final
Guidance, EPA recommends that States and Tribes use the following WET test
methods: "Methods for Measuring the Acute Toxicity of Effluents and Receiving
Waters to Freshwater and Marine Organisms," EPA/600/4-90/027; and "Short-Term
Methods for Estimating the Chronic Toxicity of Effluents and Receiving Waters
to Freshwater Organisms," EPA/600/4-89/OOla (except Method #1001 and #1003,
marine methods, for both acute and chronic test methods).  The WET test
methods include provisions requiring that they be appropriate for the species
selected and the approved end points for assessing the LC50 for acute, and the
NOEC or IC25 for chronic, toxicity.

7.    Permit Conditions

      The proposed Guidance specified permit provisions for three situations:
1) when sufficient data demonstrated that the reasonable potential to exceed
the provisions of proposed procedure 6.A of appendix F to part 132 existed; 2)
when sufficient data are not available to determine whether the discharge had
the reasonable potential to exceed the provisions of proposed procedure 6.A of
appendix F to part'132; and 3) when adequate data demonstrated that reasonable
potential to exceed the provisions of proposed procedure 6.A of appendix F to
part 132 did not exist.

      a.    Data Indicates Reasonable Potential.

      i.    Proposal:  The proposal required that permitting authorities
include effluent limitations for WET when sufficient effluent-specific data
demonstrate that a reasonable potential exists for an exceedance of the 1.0
TU. and  1.0 TUC provisions.  The proposal also included three other provisions:
1) chronic WQBELs were to be calculated based upon the TMDL design flow and
mixing zone requirements of procedure 3.B of appendix F to part 132; 2) a
schedule of compliance consistent with proposed procedure 9 of appendix F to
part 132 could be included in the NPDES permit; and 3) when regulating using a
narrative criterion for water quality, a specific WQBEL for WET may not be
necessary if it could be shown  (and documented in a fact sheet or statement of
basis for a NPDES permit) that chemical-specific WQBELs would ensure
compliance with the requirements of procedure 6.A of appendix F to part 132.

      ii.   Final Guidance:  There were few comments opposing these
provisions.  EPA has retained the essence of these provisions, but has
modified the language somewhat to account for the changes to procedure 6.A of
appendix F to part 132.  In light of the requirements in the final Guidance
that States and Tribes adopt criteria for WET protection, the final Guidance
adheres to the reasonable potential language in 40 CFR 122.44(d) more closely
than in the proposal.

     The effluent limitations for acute WET will be calculated taking into
account the allowable dilution for an acute mixing zone to ensure that 0.3 TDa
is met at the edge of the acute mixing zone.  The effluent limitations for
chronic WET will be derived using the TMDL chronic mixing zone provisions for
aquatic life protection as discussed in procedure 3 of appendix F to part 132.
EPA intends that the mixing zone provisions in procedure 3 of appendix F to
part 132 that are applicable to non-BCCs apply to the derivation of chronic
WET mixing zones.  Therefore, even if there are BCCs in a discharge, there is
no prohibition of chronic mixing zones for WET.

      A provision-has been added to the final Guidance which states that the
permitting authority may specify in the NPDES permit the conditions under
which a permittee would be required to perform a toxicity reduction evaluation

-------
402    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

(TRE).  ..This provision is consistent with existing EPA guidance  and rules
regarding the use of TREs in the control of WET.  By including this provision
in the final Guidance, EPA is encouraging the States and Tribes  to address
elevated levels in WET before they trigger a significant non-compliance
determination.

     Because some'existing dischargers may not be able to meet WET limits at
the time that their permits are reissued or modified to include  new WET
limits, EPA believes that allowing some permittees time in which to achieve
compliance may be appropriate where allowed in State or Tribal water quality
standards.  The final Guidance requires that compliance schedules, however, be
developed in accordance with procedure 9 of appendix F to part 132.  The
provision at procedure e.C.l.d of appendix F to part 132 mirrors the existing
regulation at 40 CFR 122.44(d)(1)(v).

        EPA wishes to emphasize that there are situations when a permitting
authority may determine without facility-specific effluent data  that the
reasonable potential to cause an excursion above a WET criterion or the
narrative criterion exists.  For example, effluent data from similar
industrial operations can be used to evaluate a facility for which no
effluent-specific data exist.  This information, within the judgement of the
permitting authority, can be used as a basis for evaluating whether the
reasonable potential to exceed the WET criteria exists.                      (

      In addition, the factors at 40 CFR 122.44(d)(1)(ii) which  apply to WET
need to be evaluated in reasonable potential determinations, including: the
variability of the pollutant parameter in the effluent, the sensitivity of the
species to expected pollutant toxicity testing and, where appropriate, the
dilution of the effluent in the receiving water.

       Finally, the provision at procedure S.C.l.e of appendix F to part 132
mirrors the existing regulations at 40 CFR 122.44(d)(1)(v).  EPA is including
this provision to eliminate any confusion about the applicability of 40 CFR
122.44(d)(1)(v) to facilities covered by the final Guidance.

      b.    Insufficient Data to Determine Reasonable Potential.

      i.    Proposal:  As previously discussed in this document, 40 CFR
122.44(d)(1)(i) requires a permitting authority to impose effluent limitations
whenever it finds that a facility has the reasonable potential to cause or
contribute to an excursion above a State's or Tribe's numeric or narrative
water quality criterion.  Procedure 6.C.2 of the proposed Guidance recognized
the potential for a permitting authority to have insufficient information
prior to permit issuance to determine reliably whether a facility causes, has
the reasonable potential to cause, or contributes to  such an excursion.  In
this instance, the proposed Guidance  required permitting authorities to
collect sufficient information by requiring effluent monitoring  in permits.

      Recognizing that  the approach of collecting effluent monitoring data as
a permit condition could delay effluent controls necessary to achieve State  or
Tribal  numeric and narrative water quality criteria, the proposed Guidance
required that such effluent monitoring be combined with a permit requirement
that the permittee initiate a toxicity reduction evaluation  (TRE) if the
monitoring demonstrated reasonable potential as determined by the permitting
authority.

       ii.   Comments:   Several commenters expressed  confusion regarding  the
provisions of this section because  the discussion  in the proposed preamble  and
current national policy allows State  and Tribal discretion regarding the
imposition of WET monitoring  and TRE  requirements  while  the proposed Guidance
used "shall require"  language.   Many  commenters supported  the use of State  and
Tribal discretion in  making  these decisions.'  Specifically, the commenters
were generally supportive  of  the use  of permit  conditions  to  collect the
necessary data provided States and  Tribes can  be  flexible  in  establishing  the

-------
                         Section Vm.F: Whole Effluent Toxicity                      403


appropriate monitoring frequency for a.given facility.  Several commenters
felt that the TRE provisions as proposed did not provide adequate flexibility
for the permitting authority to determine when a TRE was needed to ensure, that
water quality standards will be achieved.

      EPA did not intend to limit State or Tribal discretion regarding WET
monitoring requirements when there is insufficient information available to
make a reasonable potential determination.  In this respect the preamble to
the proposal actually reflected EPA's intent, while the proposed regulatory
language did not.

      In deciding which of the facilities lacking sufficient data for
reasonable potential determinations should be required to collect effluent WET
monitoring data, permitting authorities should consider a number of factors
including the type of facility, the potential sources of toxic contaminants,
the presence of individual toxic pollutants in the effluent, and known impacts
on the receiving water.  These decisions are best left to a case-by-case
analysis and, therefore, EPA is not requiring WET testing for all such
facilities in the final Guidance.  In deciding the type of monitoring that
comprises a sufficient data set, EPA expects permitting authorities to require
WET tests using multiple aquatic species to be consistent with the provision
of 40 CFR 122.44(d) (1) (ii), requiring consideration of aquatic species
sensitivity.  The amount of information to be collected is left to the
permitting authority to determine.  However, EPA guidance in the TSD
recommends that toxicity testing to characterize an effluent should require
testing of three species, quarterly for one year.  The means to account for
the uncertainties posed by infrequent monitoring are addressed in procedure
6.D of appendix F to part 132 in this document.  EPA will review State and
Tribal determinations in this area when it reviews individual NPDES permits.
EPA is authorized to object to NPDES permits it finds inconsistent with the
requirements of the CWA, and is also authorized to take over permit issuance
authority in such circumstances.

     Similarly, EPA has modified the proposed requirement that a TRE be
performed if effluent monitoring indicated that a reasonable potential exists
to exceed State or Tribal water quality standards.  EPA believes it is more
appropriate to include a reopener provision to first add WET permit limits if
the additional monitoring data indicate that a reasonable potential exists to
exceed the State or Tribal water quality standards.  This provision recognizes
that it will be to the permittee's advantage to conduct a TRE as soon as
possible to identify the cause of the toxicity and avoid the penalties
associated with the violation of the permit limitations.

      iii.  Final Guidance:  Procedure B.C.2 of appendix F to part 132 has
been modified to clarify that the permitting authority has discretion to
decide whether to impose WET monitoring requirements in permits of facilities
for which insufficient information exists to make a reasonable potential
determination at the time of permit issuance.  In addition, for the reasons
described above, procedure 6.C.2.b. of appendix F to part 132 was modified to
provide that States and Tribes should consider establishing a permit reopener
clause to establish WET limits when WET monitoring data collected under a
permit indicates that there is reasonable potential to cause or contribute to
an exceedance of applicable criteria.  The final procedure does not include a
provision for requiring TREs, but EPA expects that States and Tribes will
require them in NPDES permits under appropriate circumstances.

      c.    Data Indicates No Reasonable Potential.  Commenters generally
supported these provisions.  EPA has decided to finalize this part of
procedure 6 of appendix F to part 132 with only those modifications necessary
to conform to other sections of this procedure that have been modified.
Procedure 6.C.3 of appendix F to part 132 restates the current authority for a
permitting authority to establish monitoring requirements for WET in an NPDES
permit for dischargers for which it does not find a reasonable potential to
exceed numeric or narrative water quality criteria.  Where the permitting

-------
404    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


authority concludes that a continued monitoring requirement is warranted based
upon the particular circumstances of a discharge, the permitting authority may
require continued testing for a reasonable period of time and then evaluate
the monitoring results at the conclusion of this period.  For example,  a
permitting authority may decide to impose WET monitoring, prior to the  next
permit reissuance, on a discharger whose current effluent WET data indicate no
reasonable potential to determine if the facility will require WET limits in
the next NPDES permit.  Under sections 308 and 402 of the CWA, a permitting
authority can require NPDES permittees to provide WET testing data that will
assist in the development of future effluent limitations.

8.     Reasonable Potential Determinations

      a.    Proposal;  The proposed Guidance at procedure 6.D of appendix F to
part 132, provided that the factors described in 40 CFR 122.44(d)(1)(ii) be
evaluated when making a determination whether reasonable potential to exceed
the provisions of procedure 6.A of appendix F to part 132 existed.  These
factors need to be considered in all evaluations pursuant to 40 CFR
122.44(d)(1).  In addition, the proposed Guidance included procedures and
decision criteria to use in cases where facility-specific WET data existed to
determine whether the discharge had the reasonable potential to cause or
contribute to causing an exceedance of the proposed 1.0 TUa  or 1.0 TUC values.

      i.    Characterization of the Discharge.  The proposal specified  that
all acute toxicity values collected during the same day were to be averaged
for each species.  Also, the proposal specified that chronic toxicity test
results from samples collected during the same month were to be averaged.
When either chronic or acute toxicity values were unavailable, the States and
Tribes were required to estimate the missing WET test data by using an
effluent-specific acute-chronic ratio.  If there was no effluent-specific
acute-chronic ratio, then the missing WET test data was to be predicted using
a default acute-chfonic ratio of 10.  For example, using an acute-chronic
ratio of 10, a discharge with 5 TUa would  be equivalent  to 50  TU0.

      ii.   Specific Acute WET Procedure.  The proposed Guidance  specified
that a discharge has the reasonable potential to cause, or contribute to
causing an exceedance of the 1.0 TO. value when sufficient effluent-specific
information demonstrates that:

             50%/ %effect in 100% effluent < B

Where the effect is immobilization or mortality of the test organism in the
WET test and B is the multiplying factor taken from Table F6-1 of  this
procedure.  The B factor is derived from the number of samples and their
coefficient of variation, to estimate the 95th percentile toxic  unit value
when multiplied by the maximum sampled toxic unit value,  to determine if an
effluent has the reasonable potential to exceed the 1.0 TU,  value.

      iii.  Specific Chronic WET Procedure.  The proposed Guidance included a
provision to determine if a discharge has the reasonable  potential to cause,
or contribute to causing an exceedance of the 1.0 TCC value  using the chronic
aquatic life mixing zone provisions in the TMDL procedures  in proposed
procedure 3 of appendix F to part 132.  The  following equation was proposed
for assessing reasonable potential for exceeding the 1.0  TUC value:

      [chronic toxicity (TU0)  of the effluent] > 1/(B X RWC)

where B is the multiplying factor and RWC  is the receiving  water contribution,
or available dilution, as allowed under the  provisions of the section VIII.C
of the proposed preamble.

      b.    Comments:  The majority of comments  focussed on the  frequency  of
WET testing and  the total amount of data  necessary  to make  a  reasonable
potential determination.  Several  commenters were concerned that the proposed

-------
                          Section Vm.F: Whole Effluent Toxicity                      405

requirement to average acute WET test values taken on the same day, and
chronic WET test values taken in the same month, effectively required that
more than one acute WET test be conducted per day and that more than one
chronic WET test be conducted per month for each species tested to satisfy the
reasonable potential requirements.  They were concerned that such a
requirement would be unnecessary and too expensive to implement.  Some
commenters raised concerns regarding the validity of acute-chronic ratios, but
said that they could concur with a default acute-chronic ratio of 10.  Some
commenters recommended that only data representative of the discharge be used
in reasonable potential determinations.

      EPA agrees that the language in the proposal was not as clear as it
should have been with respect to the data requirements for assessing
reasonable potential.  EPA did not intend to require that the regulated
community take multiple samples in any given time period for the same species;
rather the proposal simply addressed situations where multiple samples were
collected and analyzed in the specified time period.  Sampling from several
different time periods will produce a more representative analysis of a
facility's performance than several samples from the same time period.  Also,
the proposal specified that data from one species should not be averaged with
data from other test species.

       The number of WET tests to be used in the reasonable potential
determinations depends on the number of test results that the permitting
authority considers to be acceptable to characterize a facility's discharge.
Where the permitting authority includes conditions in the permit for WET
testing for purposes of assisting in future reasonable potential
determinations, the permitting authority can balance cost considerations
associated with an increase in set of samples against the effects of having
fewer samples.  One effect of decreasing the set of samples is that the
multiplying factor .B used in the equations for assessing reasonable potential
will increase.  The effect of increasing the multiplying factor B is that the
projected maximum WET value for effluent will be larger than the one
calculated using more data points.

       EPA considers the default acute-chronic ratio of 10 suitable for
reasonable potential determinations, based on an evaluation of acute-chronic
ratios in the March 1991 TSD.  The use of either an effluent-specific or the
default acute-chronic ratio of 10 provides a cost-effective means of
augmenting a reasonable potential data set.  Where data exists for both acute
WET and chronic WET, the permitting authority should allow the augmentation of
the reasonable potential data set by: (1) allowing the use of an effluent-
specific acute-chronic ratio, (2) or allowing the use of the default acute-
chronic ratio of 10, (3) or allowing the facility to collect additional data
before a reasonable potential determination is made.

      With regard to the comment suggesting that only representative data
should trigger the need for a permit limit, EPA notes that an implicit and
obvious premise in the proposed and final WET procedure is that the WET data
used to project maximum effluent quality are valid data that are
representative of the effluent.  Permittees should ensure they are reporting
valid, representative data (see 40 CFR 122.41 (j) (1).  Where the permittee
believes certain effluent measurements to not be representative of the
effluent, the permittee should bring this to the permitting authority's
attention.  EPA's position is that valid, representative effluent data must
not be ignored.  To clarify this point in the final procedure 6 of appendix F
to part 132, EPA has inserted the word "representative" into the first
sentence of paragraph D of final procedure 6 of appendix F to part 132.  It
now reads, "Where representative facility-specific effluent WET data are
available, apply the following requirements..."

      Another commenter suggested that EPA Guidance should specify that data
obtained prior to and affected by significant treatment, pretreatment, or
pollution prevention modifications should not be used for making reasonable

-------
406    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

potential determinations.  The commenter makes the point that effluent data
used as the basis for characterizing projected effluent quality  should be
representative of the discharge and that data obtained prior to  installation
of treatment, pretreatment or pollution prevention modifications should not be
used.  In response to this comment, EPA agrees that effluent data used as the
basis for effluent characterization should be representative of  the discharge
under'current'conditions with current treatment and management practices at
the plant.  The permitting authority should use judgement in determining
whether available effluent data is representative of the 'current operating
conditions at the facility.  Where such data is found to be no longer
representative of the current discharge, the permitting authority may choose
to not use such data based on a determination that the data pre-dates current
operating conditions and treatment at the facility.

      c.    Final Guidance:  The provisions in this section of procedure 6 of
appendix F to part 132 have been modified to account for the change in the
acute WET criteria in procedure 6.A of appendix F to part 132, to clarify the
use of average WET test results, to modify the provision regarding the use of
acute-chronic ratios, and to better define the terms in the reasonable
potential equations.

       Procedures 6.D.I.a and 6-D.l.b of appendix F to part 132 have been
modified to allow the permitting authority either to average WET test results
for acute toxicity taken on the same day or use the maximum result for that
day.  Likewise, chronic WET test results taken in the same month may be
averaged or the maximum value obtained can be used in a reasonable potential
determination.  The option of using the maximum value was added  to allow for
the possibility that a State or Tribe may choose to use a more stringent data
interpretation in its reasonable potential determinations.  In addition, a
provision has been .added to ensure that the maximum daily test result for
acute WET and the maximum monthly test result for chronic WET is used in the
reasonable potential determinations.  This provision is consistent with the
intent of the proposed Guidance and existing National policy as  discussed in
the March 1991, TSD.

       Procedure 6.D.1.C of appendix F to part 132, specifying when acute-
chronic ratios shall be used in reasonable potential determinations has been
modified from the proposal.  The provision states that a default acute-chronic
ratio of 10 shall be used to estimate the values for the missing endpoint when
data exists for either acute WET or chronic WET but not for both endpoints.
This provision ensures that reasonable potential determinations  will be
performed for both acute WET and chronic WET when sufficient facility-specific
WET effluent data is available for one of the endpoints.  The requirement for
use of an effluent-specific acute-chronic ratio was deleted from the final
Guidance provision because its use is not essential, but only an option that
can be used in augmenting a reasonable potential data set as discussed in the
comments section above.

       The reasonable potential equations have been reformatted  so that the
WET criterion is on one side of the equation to facilitate interpretation of
the reasonable potential determination.  These equations are consistent with
the reasonable potential equations used in Chapter 7 of the March 1991 TSD.
The objective of the use of these reasonable potential equations is to assess
the likelihood that a facility's  effluent could discharge a pollutant at a
level that would violate a given  water quality standard.  This assessment  is
performed by estimating the 95th percentile  concentration level, TU's for WET,
from effluent sample data and determining if an allowable effluent flow into
the receiving water at  low flow conditions would violate the water quality
standard.

      The reasonable potential equations account for the effects of the
available dilution,. Qad, in the receiving water by taking into account:

-------
                   	Section Vm.F: Whole Effluent Toxicity	407

       (1) Low flow conditions used for.TMDLs, wasteload allocations  (WLAs) and
preliminary WLAs are cross referenced or  specified in procedure 6 of appendix
F to part 132;

       (2) Mixing zones as allowed pursuant to EPA-approved State and Tribal
provisions, including provisions for chronic mixing zones consistent with
provisions in the procedure 3 of appendix F to part 132;

       (3) An adjustment to the low flow of the receiving water if the
receiving stream is used by the facility  for all or part of its process water,
so that the available dilution calculated from the low flow and mixing zone
procedures does not include that portion  of the flow used by the facility.
      The revised equations used in procedures 6.D.2 and 6.D.3 of appendix F
to part 132 are presented below.

      i.    Reasonable Potential Equation for Acute WET

      The WET of an effluent is or may be discharged at a level that will
cause, have the reasonable potential to cause, or contribute to an excursion
above any numeric acute WET criterion or  numeric interpretation of a narrative
criterion within a State or Tribal water  quality standards, when effluent-
specific information demonstrates that:

       (TU,  effluent)(B)(effluent  flow/(Qad+effluent  flow))  > AC

where TO. effluent  is the  maximum measured acute  toxicity of  100 percent
effluent determined pursuant to section D.I.a. of procedure 6 of appendix F to
part 132, B is the multiplying factor taken from Table F6-1 of this procedure
to convert the highest measured effluent  toxicity value to the estimated 95th
percentile toxicity value for the discharge, effluent flow is the same
effluent flow used to calculate the preliminary WLAs for individual pollutants
to meet the acute criteria and values for those pollutants, AC is the numeric
acute WET criterion or numeric interpretation of a narrative criterion
established pursuant to section A.I of procedure 6 of appendix to part 132 and
expressed in TU, units,  and Qad  is  the amount  of  the receiving water available
for dilution calculated using: (i)  the specified design flow, 1Q10, for
tributaries and connecting channels, or where appropriate, procedure S.E.l.d
of appendix F to part 132 for use of dynamic modeling, and using EPA-approved
State and Tribal procedures for establishing acute mixing zones in tributaries
and connecting channels, or (ii) the EPA-approved State and Tribal procedures
for establishing acute mixing zones in open waters of the Great Lakes System.
Where there are less than ten individual  WET tests, the multiplying factor
taken from Table F6-1 of this procedure shall be based on a coefficient of
variation  (CV) of 0.6.   Where there are ten or more individual WET tests, the
multiplying factor taken from Table F6-1  shall be based on a CV calculated as
the standard deviation of the acute toxicity values found in the WET tests
divided by the arithmetic mean of those toxicity values.

      ii.   Reasonable Potential Equation for Chronic WET

      The WET of art effluent is or may be discharged at a level that will
cause, have the reasonable potential to cause, or contribute to an excursion
above any numeric chronic WET criterion or numeric interpretation of a
narrative criterion within a State or Tribal water quality standards, when
effluent-specific information demonstrates that:

      (TUC  effluent)(B)(effluent  flow/(Qad+effluent  flow))  >  CC

where TUC effluent  is the  maximum measured chronic  toxicity value  of 100
percent effluent determined in accordance with subsection D.l.b. of procedure
6 of appendix F to .part 132,  B is the multiplying factor taken from Table F6-1
of this procedure,  effluent flow is the same effluent flow used to calculate
the preliminary WLAs for individual pollutants to meet the chronic criteria
and values for those pollutants, CC is the numeric chronic WET criterion or

-------
408    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

numeric.interpretation  of a narrative  criterion established pursuant to
section A.2 of procedure 6 of appendix F and expressed in TUC units,  and  Qad
is the amount of the receiving water available for dilution calculated using:
(i) the design flow(s)  for tributaries and connecting channels specified in
procedure 3.E.I.a of appendix F, and where appropriate procedure S.E.l.d. of
appendix F, and in accordance with the provisions of procedure 3.E.5 of
appendix F for chronic  mixing zones, or (ii)  procedures 3.D.I and 3.D.4 of
appendix F to part 132  for discharges  to the open waters of the Great Lakes
System.  Where there are less than ten individual WET tests, the multiplying
factor taken from Table F6-1  of this procedure shall be based on a CV of 0.6.
Where there are ten or  more individual WET tests, the multiplying factor taken
from Table F6-1 of this procedure  shall be based on a CV calculated as the
standard deviation of the WET tests divided by the arithmetic mean of the WET
tests.

-------
	.	Section Vm.F: Whole Effluent Toxicity	409


                    Tables to  Procedures. 5 and  6  of Appendix F
                                      Table F6-1
     Reasonable Potential Multiplying Factors: 95% Confidence  Level and 95%
                                  Probability  Basis

Nunber of	Coefficient of Variation	
Samples 0.1  0.2  0.3 0.4  0.5  0.6  0.7  0.8  0.9  1.0  1.1 1.2  1.3  1.4  1.5  1.6  1.7  1.8 1.9  2.0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
30
40
50
60
70
80
90
100
1.4
1.3
1.2
1.2
1.2
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.1
1.0
1.0
T.O
1.0
1.0
1.0
1.0
1.0
1.9
1.6
1.5
1.4
1.4
1.3
1.3
1.3
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.1
1.1
1.1
1.1
1.1
1.1
1.0
1.0
1.0
1.0
1.0
1.0
1.0
2.6
2.0
1.8
1.7
1.6
1.5
1.4
1.4
1.4
1.3
1.3
1.3
1.3
1.3
1.2
1.2
1.2
1.2
1.2
1.2
1.1
1.1
1.0
1.0
1.0
1.0
0.9
0.9
3.6
2.5
2.1
1.9
1.8
1.7
1/6
1.6
1.5
1.5
1.4
1-.*
1.4
1.4
1.3
1.3
1.3
1.3
1.3
1.2
1.1
1.1
1.0
1.0
1.0
0.9
0.9
0.9
4.7
3.1
2.5
2.2
2.1
1.9
1.8
1.7
1.7
1.6
1.6
1.5
1.5
1.4
1.4
1.4
1.4
1.3
1.3
1.3
1.2
1.1
1.0
1.0
1.0
0.9
0.9
0.9
6.2
3.8
3.0
2.6
2.3
2.1
2.0
1.9
1.8
1.7
1.7
1.6
1.6
1.5
1.5
1.5
1.4
1.4
1.4
1.4
1.2
1.1
1.0
1.0
0.9
0.9
0.9
0.9
8.0
4.6
3.5
2.9
2.6
2.4
2.2
2.1
2.0
1.9
1.8
1.7
1.7
1.6
1.6
1.6
1.5
1.5
1.5
1.4
1.2
1.1
1.0
1.0
0.9
0.9
0.9
0.9
10.1
5.4
4.0
3.3
2.9
2.6
2.4
2.3
2.1
2.0
1.9
1.9
1.8
1.7
1.7
1.6
1.6
1.6
1.5
1.5
1.3
1.1
1.1
1.0
0.9
0.9
0.9
0.8
12.6
6.4
4.6
3.7
3.2
2.9
2.6
2.4
2.3
2.2
2.1
2.0
1.9
1.8
1.8
1.7
1.7
1.6
1.6
1.5
1.3
1.2
1.1
1.0
0.9
0.9
0.9
0.8
15.5
7.4
5.2
4.2
3.6
3.1
2.8
2.6
2.4
2.3
2.2
2.1
2.0
1.9
1.8
1.8
1.7
1.7
1.6
1.6
1.3
1.2
1.1
1.0
0.9
0.9
0.8
0.8
18.7
8.5
5.8
4.6
3.9
3.4
3.1
2.8
2.6
2.4
2.3
2.2
2.1
2.0
1.9
1.9
1.8
1.7
1.7
1.6
1.3
1.2
1.1
1.0
0.9
0.9
0.8
0,8
22.3
9.7
6.5
5.0
4.2
3.7
3.3
3.0
2.8
2.6
2.4
2.3
2.2
2.1
2.0
1.9
1.9
1.8
1.8
1.7
1.4
1.2
1.1
1.0
0.9
0.9
0.8
0.8
26.4
10.9
7.2
5.5
4.5
3.9
3.5
3.2
2.9
2.7
2.5
2.4
2.3
2.2
2.1
2.0
1.9
1.9
1.8
1.7
1.4
1.2
1.1
1.0
0.9
0.9
0.8
0.8
30.8
12.2
7.9
6.0
4.9
4.2
3.7
3.3
3.1
2.8
2.7
2.5
2.4
2.3
2.2
2.1
2.0
1.9
1.9
1.8
1.4
1.2
1.1
1.0
0.9
0.9
0.8
0.8
35.6
13.6
8.6
6.4
5.2
4.5
3.9
3.5
3.2
3.0
2.8
2.6
2.5
2.3
2.2
2.1
2.0
2.0
1.9
1.8
1.4
1.2
1.1
1.0
0.9
0.8
0.8
0.8
40.7
15.0
9.3
6.9
5.6
4.7
4.1
3.7
3.4
3.1
2.9
2.7
2.5
2.4
2.3
2.2
2.1
2.0
2.0
1.9
1.5
1.2
1.1
1.0
0.9
0.8
0.8
0.8
46.2
16.4
10.0
7.4
5.9
5.0
4.3
3.9
3.5
3.2
3.0
2.8
2.6
2.5
2.4
2.3
2.2
2.1
2.0
1.9
1.5
1.2
1.1
1.0
0.9
0.8
0.8
0.8
52.1
17.9
10.8
7.8
6.2
5.2
4.5
4.0
3.6
3.3
3.1
2.9
2.7
2.6
2.4
2.3
2.2
2.1
2.0
2.0
1.5
1.2
1.1
1.0
0.9
0.8
0.8
0.7
58.4
19.5
11.5
8.3
6.6
5.5
4.7
4.2
3.8
3.4
3.2
3.0
2.8
2.6
2.5
2.4
2.3
2.2
2.1
2.0
1.5
1.3
1.1
1.0
0.9
0.8
0.8
0.7
64.9
21.1
12.3
8.8
6.9
5.7
4.9
4.3
3.9
3.6
3.3
3.0
2.9
2.7
2.5
2.4
2.3
2.2
2.1
2.0
1.5
1.3
1.1
1.0
0.9
0.8
0.8
0.7

-------
410     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
                            Section Vin.G: Loading Limits                         411
G.    Loading Limits

1.    Proposal

      The proposed Guidance provided that water quality-based effluent limits
(WQBEL) be expressed in terms of both concentration and mass loading rates,
except for those pollutants that cannot appropriately be.expressed in terms of
mass.  These provisions clarify the application of existing Federal
regulations at 40 CFR 122.45 (f) to the Great Lakes System in order to most
effectively implement the objectives of the Clean Water Act.  The proposed
Guidance accounted for the following three factors in establishing procedure
7.

      First, the proposed Guidance provided one exception from the requirement
to express WQBELs in both concentration values and mass loading rates.
Consistent with the existing Federal regulations at 40 CFR 122.45 (f) (1) (i),
the proposed Guidance would not have required Great Lakes States and Tribes
adopting this procedure to express WQBELs as mass loading rates for pollutants
that cannot be appropriately expressed in terms of mass, such as pH, color,
temperature, or radiation.

      Second, the proposed Guidance did not apply to technology-based limits,
and therefore did not propose to affect the application of this provision to
limits established using 40 CFR 125.3.

      Third, the proposed Guidance emphasized the need to express WQBELs as
both concentration values and mass loading rates in order to implement the
antidegradation policy for the Great Lakes System.  The use of mass limits in
the Great Lakes antidegradation analysis is discussed in appendix E of the
final Guidance.

      The proposed'Guidance was consistent with the EPA March 1991, Technical
Support Document for Establishing Water Quality-Based Effluent Limits, TSD.

2.    Comments

      Many of the comments on the Loading Limit procedure addressed the need
for loading limits and concerns regarding the applicability of loading limits
to POTW discharges under wet weather conditions.  A summary of the key points
raised by commenters for these two topics is presented below.

a.    Need for Loading Limits

      Many commenters expressed concern that the inclusion of mass loading
limits is not necessary if concentration limits are also required.  They did
not see any environmental benefit to requiring both types of limits.  Other
commenters expressed support for the use of loading limits, especially as a
means to limit bioaccumulative contaminants and to aid in the implementation
of the antidegradation procedures.

      EPA has determined that mass-based limits are necessary to prevent the
use of dilution as a means of treatment.  Where water quality is impaired,
mass-based limits dre used to implement TMDL mass load reduction targets and
other procedures to establish WQBELs to control total loadings to the
receiving water body.  EPA determined from past experience that mass loading
limits are a valuable regulatory tool in many discharge scenarios and
therefore continues to support the requirement for including mass loading
limits in all NPDES permits.  For example, such mass-based limits enable POTWs
to more effectively control toxics and bioaccumulative contaminants
originating from indirect dischargers by limiting the mass of contaminants
that these facilities are allowed to discharge into the sewer system.  Such
mass-based limits prevent the facilities from using the domestic waste at the
POTW to dilute their toxic discharges.

-------
412    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

b.    Wet Weather Discharges

      Many commenters expressed concern that requiring mass loading limits for
POTWs with CSOs, especially if those loading limits are based on daily maximum
design flows and weekly average flows,  would not adequately account for the
intermittent wet weather flow conditions.  Commenters were concerned that such
loading limits would place POTWs with CSOs in non-compliance status several
times per year.  Also several commenters recognized the need to account for
wet weather conditions in compliance determinations while maintaining the use
of loading limits to protect the receiving water body.  Few commenters
supported the use of annual mass limits to account for WET weather flow
impacts on POTWs.

      EPA does not envision that dischargers will be unnecessarily restricted
in allowing elevated flows and loads under wet weather conditions using
procedure 7.  Existing Federal, State and Tribal  regulations and policies
allow flexibility in addressing wet weather conditions and intermittent
increases in flows due to wet weather events.  It is not the intent, or
design, of the final Guidance to limit existing regulatory flexibility
regarding wet weather flows.  Rather, it is in EPA's interest to promote
consistent policy objectives among programs impacting NPDES permit
requirements.  The EPA CSO Strategy encourages POTWs to treat as much wet
weather flow as possible, recognizing that treatment efficiencies will likely
decrease due to higher flows during wet weather events, EPA supports the
continued use of State and Tribal discretion in determining special NPDES
permit conditions and other appropriate mechanisms to address wet weather
flows.  Therefore, EPA sees no need to establish any additional provisions,
such as annual mass-based WQBELs, regarding wet weather discharges in the
final Guidance.

3.    Final Guidance

      After fully considering the comments and evaluating the use of mass-
based WQBELs to  date, EPA has retained, virtually unchanged, the provision for
including mass-based WQBELs as outlined in procedure 7 of appendix F, of the
proposal.  In the final Guidance procedure  7 of appendix F, EPA establishes
requirements to  calculate mass-based WQBELs  to restrict the loadings of
pollutants to the Great Lakes System.  As discussed above, procedure 7
specifies that when a WQBEL is developed using procedures for Total Maximum
Daily Loads, 3,  and Reasonable Potential, 5, of appendix F  , or other State
and Tribal procedures, the limitation must  be expressed in terms of both
concentration and mass loading rate, except  for the specific exclusions
identified in  §132.

      Procedure  7.A specifies that the concentration and mass WQBELs must be
consistent in  terms of daily, weekly, and monthly averages, or  in other
appropriate averaging periods.   For example, where a concentration-based WQBEL
is expressed in  terms of maximum daily and  average monthly  limitations, the
corresponding mass loading rate  limitations must likewise be expressed as
maximum daily  and average monthly limitations.  Existing Federal regulations
at 40 CFR 122.45(d), require  that limitations for continuous discharges be
expressed, unless impracticable,  as average weekly and average monthly
limitations for  POTWs and maximum daily  and average monthly limitations for
all other continuous discharges.  The final Guidance does not  change  these
existing requirements, but  instead ensures  consistency between  mass-based and
concentration-based WQBELs  in  individual NPDES permits.

       Procedure  7.B of appendix  F directs  the permit writer to  use  effluent
flow  rates  when developing the mass  loading rate WQBELs that are  consistent
with  those  used in procedures  3  and  5 of appendix  F,  or other  State  and Tribal
procedures, to develop the concentration-based WQBELs.   The existing Federal
regulations at 40 CFft  122.44(d)(1)(vii)(B)  and  123.25(a)(15)  require that
water quality-based effluent limits  in  State, Tribal  and Federal  NPDES  permits
be  ".  .  . consistent with  the  assumptions  and requirements  of  any available

-------
                            Section Vm.G: Loading Limits                         413
wasteload allocatidn for the discharge.  .  ."  By specifying this requirement
in the final Guidance, EPA believes that it clarifies any confusion  that might
arise regarding the proper effluent flow rate to use in developing mass  •
loading rate permit limitations.  In addition, this approach ensures greater
consistency in WQBELs among the Great Lakes States and Tribes.

      EPA recognizes that POTWs with CSOs are subject to intermittent
increases in effluent flow above the dry weather flows used to develop the
WQBELs.  EPA defers to the States and Tribes to use existing permit  procedures
to address these intermittent flow increases in evaluating compliance with  the
concentration-based and mass-based limits.

-------
414     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
                   Section Vm.H: WQBELs Below the Level of Quantification               415

H.    WOBELs Below the Level of Quantification

      Several pollutants of initial focus in Table 6 and many of the BCCs -are
known to cause unacceptable toxic effects at water column concentrations lower
than what the most sensitive analytical techniques can currently quantify.
Accordingly, for these pollutants, water quality criteria necessary to protect
the designated uses have been established at levels below quantification.
Therefore, WQBELs  calculated for these pollutants can also be below current
minimum quantification levels.  The final Guidance contains procedure 8 of
appendix F, which  establishes procedures for expressing WQBELs in these
circumstances, assessing compliance with such limits, and requiring pollutant
minimization programs.  Procedure 8 is intended, in part, to increase
consistency among  the Great Lakes States and Tribes in addressing these types
of permit limits.  As required by the Critical Programs Act, the provisions of
procedure 8 reflect existing national policy and guidance.

1.    Expressing a WOBEL Below the Minimum Quantification Level

      a.    Proposal;  The Great Lakes Guidance proposal provided that all
WQBELs included in an NPDES permit be expressed exactly as calculated and that
a compliance evaluation level  (CEL) be specified in the permit, which
signified the level of the pollutant in the effluent that was not to be
exceeded.  The proposal further provided that the CEL was the level at which
compliance with the effluent limit would be assessed.  The preamble to the
proposal also recognized the difficulty of making a definitive statement as to
whether or not the concentration of a pollutant is above or below the WQBEL,
and therefore, characterized the pollutant minimization plan as one means of
increasing the likelihood that the concentration of the pollutant is as close
to the WQBEL as possible.

      b.    Comments;  Several commenters opposed EPA's proposal to require
WQBELs to be expressed in NPDES permits exactly as calculated and the
implication that the permittee comply with such limits even if they are below
quantification levels.  Some commenters asserted that the provision in the
proposed guidance would require substances to be removed below the levels that
exist in nature.• Others contended that EPA lacks the legal authority under
the Clean Water Act (CWA) to regulate substances that are discharged below
levels that can be accurately quantified and asserted that issuance of permit
limits in such circumstances violates due process.  They asserted that an
effluent limitation below the level of quantification is constitutionally
defective insofar as it subjects the permittee to criminal sanctions while
failing, in the commenters' view, to specify clear and ascertainable standards
of conduct.  These commenters contended that permittees subject to such limits
will have no idea whether they are complying with their limits and thus will
be unable to take action to avoid noncompliance.  They further asserted that
they have no way to determine exactly what conduct is prohibited and hence
concluded that such limitations are void for vagueness.

      Other commenters supported using the WQBELs as calculated as the permit
limit irrespective of the minimum quantification level and emphasized the need
to hold facilities accountable for meeting those limits.  Some commenters felt
that those limits are especially important in controlling the discharge of
BCCs.

      c.    Final Guidance:  EPA has retained the requirement that the WQBELs
be included as calculated as the permit limit in NPDES permits, even if the
WQBEL is below the minimum quantification level.  However, EPA wants to
clarify here that it is the Agency's policy that any effluent sample analyzed
in accordance with the analytical method specified in the permit and other
applicable procedures that is found to be below the quantification level shall
be deemed in compliance with the WQBEL.

      EPA's national guidance in the 1991 Technical Support Document for Water
Quality-based Toxics Control (TSD) recommends expressing WQBELs in permits

-------
416    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

exactly, as calculated, even if they ar.e below the minimum quantification
level, in order to comply with the CWA and implementing regulations at 40 CFR
122.44(d)(1)(vii), which require that NPDES permits ensure water quality
standards be met.  Specific examples, such as the development of dioxin WQBELs
for paper mills, were discussed in the preamble to the proposed Guidance.

      EPA believes that such WQBELs are required by the CWA.  Under the CWA,
States and Tribes are required to adopt water quality standards to protect the
public health and welfare, enhance the quality of water and serve the purposes
of the Act.  See CWA section 303(c)(2)(A).  Such standards shall consider,
among other things, the waters' use and value for the propagation of fish and
wildlife.  As part-of their standards, States and Tribes are required to adopt
water quality criteria for particular pollutants that reflect concentration
levels deemed protective of the waters'  designated uses.  In addition, the CWA
further requires that, once these standards are adopted by a State or Tribe,
point source dischargers must achieve effluent limitations as necessary to
implement such standards, including any numeric water quality criteria
therein.  See CWA section 301(b)(1)(C).   These principles are reiterated in
section 118 of the CWA, under which authority this Guidance is issued.  In
that section. Congress directed EPA in its Guidance to specify numerical
limits to protect human health, aquatic life, and wildlife and to provide
guidance to the Great Lakes States on  water quality standards and other
aspects of their water quality programs.  None of these sections provides an
exception for pollutants that are known to be present in the effluent or
wastewater but not in reliably quantifiable amounts.

      EPA disagrees with the comments that EPA is regulating pollutants that
are not present in the effluent.  It is important to keep in mind that a WQBEL
is derived only when the permitting authority has determined there is
reasonable potential for a discharge to cause or contribute to an exceedance
of a water quality standard for that pollutant.  The procedures for
determining reasonable potential are presented in procedure 5 of appendix F of
this final Guidance.  Even though a pollutant may not be present at
quantifiable concentrations in the effluent, the permitting authority may have
information, such as fish tissue data, samples with detectable concentrations
or process information, that documents the presence of the pollutant in
sufficient quantity to warrant a finding that the pollutant has the reasonable
potential to cause or contribute to an exceedance of the applicable water
quality standard.  If a reasonable potential determination is made, a WQBEL
must be calculated in order to assure the water quality standard will be
attained and maintained.  The CWA provides no exception to this principal when
the pollutant of concern is present at concentrations below quantifiable
levels.

      EPA acknowledges that, in some cases, the water quality criterion
calculated to protect a particular designated use may require establishment of
a WQBEL that authorizes discharges at levels lower than those that exist in
nature.  In these cases, the State or Tribe could adopt a site-specific
modification of the criteria consistent procedure 1 of appendix F; or
alternatively, the State or Tribe could grant a variance to the  standard under
procedures 2 of appendix F, which allows for consideration of naturally
occurring pollutant concentrations which prevent the attainment of standards.

      For the reasqns set forth above, EPA has concluded that it has the legal
authority and obligation under the CWA to require the establishment of WQBELs
below the level of quantification.  EPA disagrees with those commenters that
argued that imposing permit limits in such circumstances violates due process.
Contrary to the commenters' assertions, the WQBEL expressed in the permit
establishes a clear and certain numeric standard of conduct.  In addition, a
discharger will be deemed to be in compliance with the permit, as discussed
below, if samples analyzed in accordance with the analytical method specified
in the permit, and other applicable procedures, are found to be below the
quantification level.  Furthermore, it is EPA's current policy that analytical
methods other than those specified in the permit cannot be used  to establish a

-------
                   Section VHI.H: WQBELs Below the Level of Quantification               417


violation of the WQBEL.  Instead, it is EPA's position that special  conditions
in the permit such as fish tissue sampling, WET  tests, limits and/or
monitoring requirements on internal waste streams, and monitoring  for
surrogate parameter can be used to reopen the permit to establish  more
stringent effluent limits.

      EPA acknowledges commenters' concerns that it may be difficult to
demonstrate that it is achieving these limits.   However, EPA disagrees that
this prevents permittees from assessing their compliance status or from
selecting control strategies to avoid non-compliance.  As discussed  above, it
is EPA's current policy that a permittee shall determine whether it  will be
deemed in compliance with the permit by monitoring its effluent using
authorized methods, the same methods that would  be used by an enforcement
authority in an enforcement action.

2.    Compliance Issues

      a.    Proposal:  Under the proposed Guidance, the permitting authority
would specify in the permit an analytical method, a monitoring frequency, and
a corresponding compliance evaluation level (CEL) for each pollutant with a
WQBEL below the minimum quantification level.  A CEL was defined as  the
concentration at which compliance with the WQBEL is assessed.  The preamble
discussion provided guidance that the permitting authority should  specify the
most sensitive analytical method and define the  CEL as the Minimum Level  (ML)
specified in or approved under 40 CFR 136, if available.  The guidance also
proposed that, when MLs are not available, the permitting authority  must still
specify a CEL in the permit.  Section 132.2 of the proposal defines  the ML as
the lowest level at which the analytical system  gives recognizable spectra and
acceptable calibration points.  This is the lowest level at which  the
concentration of the pollutant can be reliably measured.  The proposed
guidance also allowed adjustments for matrix interference in establishing the
ML.  In addition, the preamble to the proposal discussed several approaches
for selecting an analytical method and CEL for chemicals not addressed by 40
CFR 136, but the proposed guidance itself was silent on this matter.

      b.    Comments:  EPA received a variety of comments pertaining to this
portion of procedure 8.  Many commenters expressed confusion regarding the
relationship between the WQBEL, CEL and ML.  This was due in part  to the
several options presented in the preamble for establishing the CEL,  including
the use of the WQBEL as the CEL.

      Many commenters advocated the use of the practical minimum
quantification level (PQL) as the CEL.  The PQL  is typically defined as a
concentration 5 to *10 times the method detection level (MDL).  The MDL is
defined according to a procedure specified in 40 CFR 136 appendix  B.
Measurements at or above the MDL value are highly unlikely to be associated
with a true concentration of zero.  Their contention was that the  PQL could be
derived in a straightforward manner and would be less onerous to use in
comparison to the ML,  especially for facilities  that experience matrix
interference in analytical testing procedures.  Many commenters agreed with
the provision to allow for matrix interference when setting minimum
quantification levels in a permit.

      In addition, .for the majority of chemicals that do not have  MLs in 40
CFR 136, several commenters advocated using the  PQL in lieu of determining the
ML for these chemicals.  Other commenters supported the use of the MDL as a
CEL in cases where the WQBEL is much lower than  the ML, since any  detection of
the chemical could, in their opinion, indicate an exceedance of the  WQBEL.

      Commenters also expressed concern that the level at which compliance
would be measured would change as analytical methods improve, thus exposing
them to enforcement actions and to successively more stringent permit limits
for which immediate compliance would be required.  Some commenters contended
that this potential for change in analytical methods would prevent them from

-------
418    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

determining whether they were in compliance with their limits or from
selecting controls necessary to achieve compliance.  Others contended that, by
specifying procedures for determining compliance when WQBELs are below the
level of quantification, the Guidance will eliminate the discretion
traditionally vested in the States for determining such compliance.  Other
commenters argued that laboratory detection capabilities vary greatly
throughout the Great Lakes region, which would lead to widely disparate
treatment requirements and enforcement activities.  Some commenters expressed
concern that the long lag time between sampling and analysis could mean that
the permittee could unknowingly be out of compliance for a lengthy period.  A
few commenters stated that laboratory analyses should be done under the
Superfund Contract Laboratory Program protocol to ensure uniformity.  These
commenters also expressed concern that there will be a greater likelihood of
false readings from using equipment at the frontiers of detection capability.

      A number of commenters expressed a related concern that, if a permit
does not specifically provide that compliance with the  minimum quantification
level constitutes compliance with the permit, permittees will be vulnerable to
law suits for violations of effluent limitations in situations where a more
sensitive analytical technique is developed that shows that pollutants are
being discharged above the WQBEL but below the permit-specified minimum
quantification level.  These commenters also suggest that permittees may be
subject to frivolous law suits based upon wholly unreliable data.

      c.    Final Guidance:  In developing the final Great Lakes Guidance, EPA
acknowledges the need to develop regulatory procedures that provide the public
with a clear and unambiguous means of evaluating compliance with WQBELs below
the minimum quantification level.  EPA modified the Guidance based upon the
comments received and has revised the compliance requirements associated with
this procedure.  Th.e principal change eliminates the use of the CEL term.
This clarifies that the WQBEL as actually calculated is the enforceable permit
limit even if it is below the minimum quantification level.  Under procedure
8.B.I, once adopted by a State or Tribe or promulgated by EPA, the regulatory
authority must specify in the NPDES permit for each WQBEL that is calculated
to be less than the quantification level, the applicable method to be used for
monitoring the presence and amount of the pollutant in an effluent.  The
permit shall also specify the quantification level pursuant to procedure
8.B.2.

      Like the proposal, the final Guidance provides that the analytical
method and minimum 'quantification level is to be specified in the permit.  In
EPA's view, requiring States and Tribes to specify approved analytical methods
or if none are available, the most sensitive analytical method practicable,
and a corresponding analytically determined minimum quantification level for
reporting monitoring information, will ensure a sufficient level of
consistency between States and Tribes in evaluating compliance with WQBELs
below the minimum quantification level.

     A new provision has been added as procedure 8.B.2 in the final Guidance,
which specifies that, for analytical methods under 40 CFR 136, for those
approved under the Alternate Test Procedures at 40 CFR 136.3(d), or other
methods specified i'n the permit that do not have an ML specified under 40 CFR
136 or 136.3(d), the permitting authority is required to specify a minimum
quantification level in the permit that is as close to the WQBEL as
practicable.  The preamble to the proposed guidance discussed several options
for addressing the above condition, but the proposed Guidance was  silent as  to
how to establish the appropriate minimum quantification level in NPDES permits
in such cases.  Procedure 8.B.2 is included in the final Guidance  to increase
consistency in application of procedure 8 among States and Tribes  for cases
where the ML has not been specified in 40 CFR 136 or pursuant to 136.3(d).
This provision addresses the concerns of several commenters that felt that too
much flexibility was being granted to the permitting authorities in selecting
the minimum quantification level  in these cases.

-------
                   Section Vm.H: WQBELs Below the Level of Quantification               419

      The discharger has the opportunity to document a higher minimum
quantification level if repeated attempts to overcome matrix interferences are
unsuccessful.    EPA has provided Guidance Associated with Compliance
Monitoring  (EPA 821-B-93-001; the "Monitoring Guidance"), for dischargers
attempting to overcome matrix interference problems.  Recent evaluations of
effluent data from well-designed, well-operated waste treatment facilities
indicate that' in all instances in which a good-faith attempt was made to
overcome matrix interferences using the strategies set forth in the Monitoring
Guidance, or similar strategies, matrix interferences have not been an
impediment to achieving the minimum quantification level in EPA's analytical
methods.

      Procedure 8.B.2 allows flexibility in developing a minimum
quantification level in these instances and should not place an undue burden
on the States, Tribes, or permittees.  EPA's MLs do not get promulgated in 40
CFR 136 until after public notice and comment.  Consequently, these MLs are
the most appropriate minimum quantification levels to use, if available, for
NPDES monitoring data requirements because they represent the most stringent,
scientifically reliable minimum quantification level available.  EPA has used
the ML in Clean Water Act programs since the mid 1980's. EPA rejected the use
of the MDL and other non-quantifiable concentration levels because these
concentrations, by definition, do not represent concentrations that are both
reproducible and quantifiable indicators of the actual concentration of a
given sample, and hence are not reliable measures for permit compliance
purposes.

      However in the absence of an ML specified in, or approved under, 40 CFR
136, the permitting authority is required to specify a minimum level of
quantification at the lowest practicable quantifiable level above the level of
detection.  As with any permit condition, such level is subject to public
notice and comment and may be contested on appeal, thereby insuring its
validity.  EPA has no national guidance or policy on how to develop the lowest
quantifiable level that can be used as a tool to assess compliance with
WQBELs.  Currently States and Tribes develop their own lowest practicable
quantifiable level above the level of detection and they range from the use of
practicable quantitation levels  (PQLs), minimum quantitation levels  (MQLs),
quantitation levels (QLs), reliable quantitation levels  (RQLs), plus many
more.

      EPA's Engineering and Analysis Division  (RAD) has developed a procedure
for calculating the Minimum Level (ML) of quantification for those pollutants
that do not have MLs specified in or approved under in 40 CFR 136.  The MLs
are the lowest levels used in establishing the calibration required by the
analytical methods in the NPDES program.  The procedure, based on the "ML
concept," calculates the minimum quantification level by multiplying the MDL
by a factor of 3.18.  This factor produces an ML approximately equal to values
obtained by methods for developing the Limit of Quantitation endorsed by the
American Chemical Society.  The MDL has been established for more than 130
analytical methods and several hundred chemicals.  Another approach for
developing a quantification level in the absence of an ML specified in, or
approved under 40 CFR 136, would be to determine an MDL as specified in 40 CFR
136, appendix B,  and multiply the result by 3.18.

      These approaches are designed to produce values which approximate
specified or approved MLs, thereby providing consistency with the monitoring
requirements imposed for those chemicals with specified or approved MLs.  The
above examples for calculating a required minimum level of quantification, are
some of many approaches currently being used and are intended as guidance, not
requirements.  The permitting authorities have the flexibility under this
provision to develop other credible procedures for calculating a minimum
quantification level for use in a NPDES permit in the absence of an EPA
promulgated ML.  The 'only requirement is that the permitting authorities must
demonstrate that any minimum quantification level specified is as close to the
WQBEL as practicable.

-------
420    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Some commenters urged that the PQL be used to establish the
quantification level.  The PQL, which has been used by EPA's drinking water
and solid waste programs, is defined as the level at which reliable
measurements can be made .under routine operating conditions  (50  FR 46908; 52
FR 25699).  The PQL is typically established by multiplying the MDL by a
factor ranging from 5 to 10.  One criticism of the PQL procedure is the,.
ambiguous nature of the multiplier and the resulting levels being perceived as
too high for regulatory or compliance purposes for wastewaters.  EPA does not
endorse the use of the PQL for the NPDES program.  EPA acknowledges that the
PQL has been used by EPA's drinking water and solid waste programs; however,
during the past few years, EPA's drinking water program has been investigating
alternatives to the PQL.  Since the EPA is actively reevaluating its use of
the traditional PQL values, EPA does not endorse them for evaluating
compliance with WQBELs below the minimum quantification level.

      Commenters also expressed concern that the level at which  compliance
would be measured would change as analytical methods improve, thus exposing
them to enforcement actions and to successively more stringent permit limits
for which immediate compliance would be required.  It is EPA's intent,
however, to promote the development of more sensitive analytical techniques,
to adopt these techniques into 40 CFR 136, and to require the States and
Tribes to specify the most sensitive tests, specified in or approved under 40
CFR 136, in the NPDES permit.  In addition, the Guidance has been modified to
require that a permit subject to procedure 8 shall contain a reopener clause
authorizing modification or revocation and reissuance of the permit if new
information generated as a result of special conditions included in the permit
indicates the presence of the pollutant in the discharge at levels above the
WQBEL.  If a new analytical procedure is authorized by 40 CFR 136, either
directly or through the alternate testing procedure at 40 CFR 136.3(d), States
and Tribes have the discretion of reopening the permit to include a new
analytical method or to wait until the permit is reissued to include a new,
more sensitive analytical method.  However, because there are several factors
to consider, such as the degree of improvement in the minimum quantification
level, time remaining on the existing permit, current compliance status, and
administrative costs of modifying a permit, EPA has determined that it would
be inappropriate .to dictate the timing and implementation of more sensitive
analytical techniques.  Therefore, neither this Guidance nor EPAs' existing
regulations require permits to be reopened under such  circumstances.

      EPA also acknowledges commenters' concerns that, because analytical
methods can change, a discharger deemed to be in compliance with its WQBEL
today (as indicated by current analytical methods) has no assurance that it
will be complying with its limit in the future.  However, EPA disagrees that
this raises a due process question.  It is EPA's current policy  that new
analytical methods will not be a basis for determining compliance with the
WQBEL unless they are specifically included in the permit.  Permits can be
reopened to include more stringent limits or conditions, for example when data
generated from the special conditions of the permit support it.  Permittees
will have the opportunity to comment on the new requirements before they
become effective in their permits.  While EPA expects the sensitivity of
several analytical methods to improve, the Agency would expect that permitting
authorities, when requiring the use of new methods in NPDES permits, would
consider establishing compliance schedules, if authorized by law, to allow  the
facility time to achieve full compliance.  While the approach described above
is consistent with existing EPA guidance, EPA is currently reevaluating its
National policy regarding the implementation of WQBELs that are  below the
level of quantification; any changes to the National policy would apply to  the
Great Lakes States and Tribes.

      EPA also notes that the commenters' concerns about a lack  of  certainty
regarding future compliance status or changing permit terms are  not unique  to
this procedure.  For example, all permittees, even those not affected by this
procedure, may be subject to increasingly stringent WQBELs in successive
permits as circumstances change  (e.g., lower water quality or improved

-------
                   Section Vm.H: WQBELs Below the Level of Quantification               421

analytical methods), because CWA section  301(b)(1)(C) requires  in each permit
the establishment  of effluent limits as necessary to achieve water quality
standards.  In such cases, a permittee must comply immediately  with more
stringent limits unless a State's or Tribe's water quality standards or
implementing regulations authorize a compliance  schedule.  Opportunities  for
notice and comment -as well as administrative and judicial review however,
avoid due process  concerns.

      EPA notes that procedure 8 is consistent with EPA's guidance in the 1991
TSD.  In section 5.7.3 of the TSD, EPA recommends that special  conditions be
included in certain permits, such as those subject to procedure 8 here, to
help ensure that the limits are being met and that excursions above water
quality standards  are not occurring.  EPA has incorporated several of these
conditions into the pollutant minimization program.  Information derived  from
these conditions,  including fish tissue data and analyses and in-plant
monitoring data, can be used to help support reopening the permit to establish
more stringent limits and controls, if necessary to ensure that the WQBEL is
attained.  Procedure 8 of appendix F simply codifies existing EPA guidance.

      EPA also acknowledges corranenters' concerns regarding the  lack of
uniformity among laboratories performing  analyses under the analytical methods
and related issues.  The ML procedure used under 40 CFR 136 relies on the use
of a well-defined  MDL.  Before a particular laboratory can be employed by a
permittee to analyze its effluent for compliance with the WQBEL, an individual
laboratory would need to demonstrate that its procedures can reliably quantify
the MDL, which has been established for more than 130 analytical methods  and
several hundred chemicals.  In this way,  EPA hopes to minimize  any disparity
among laboratory performance and hence to minimize any differences across the
Great Lakes basin  in treatment requirements and  enforcement activities.   An
ML, defined as the lowest calibration point, can then be determined at an
approved laboratory for use in evaluating compliance data.  Since all
laboratories seeking to use the same analytical  procedures would need to
demonstrate that they can reliably achieve the published MDL and use approved
calibration techniques, the MDL establishes a uniform standard  that must  be
met.

      Moreover, the enforcement authority has the burden of proving non-
compliance, which  can be difficult if all of the samples analyzed in
accordance with the specified analytical methods are below quantification and
hence do not support allegations that a violation has occurred.  In addition,
even though the WQBEL is below the level of quantification, a permittee
seeking greater certainty regarding its compliance status may be able, in some
instances, to monitor its wastestreams within the plant, prior  to dilution, at
points where the presence of the pollutant may be detected directly.  Thus, a
permittee possesses the tools necessary to avoid noncompliance, and can
develop and implement appropriate pollution control or prevention strategies
as necessary to keep the presence of the pollutant below the ML.  See below
for a discussion of in-plant monitoring in connection with the  pollution
minimization program.  Furthermore, in no case can civil or criminal penalties
be imposed against a permittee without a hearing to ensure that the permittee
is accorded due process.  For these reasons, EPA concludes that requiring a
permitting authority to impose in the permit the WQBEL as calculated does not
violate due process.

3.    Compliance with the CEL

      a.    Proposal:  The proposed Guidance included a provision that a  CEL
be established for each daily, weekly, and monthly NPDES permit limit. The
proposed Guidance did not specify how to define  the CEL for the above permit
limits. The proposed procedure 8.C of appendix F, deferred to existing State
and Tribal procedures for determining how to .average compliance data that
includes non-quantifiable data.

-------
422    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      b.    Comments;  Many commenters. expressed confusion concerning the
relationship between the WQBEL, ML and CEL, as previously discussed in
Compliance Issues section of this document.  Many commenters supported the
provision to defer to existing permitting authority procedures for averaging
compliance data.  A few commenters supported a consistent approach for
assessing non-quantifiable data by using surrogate values.  These values
ranged from zero to the ML value.

      c.    Final Guidance:  EPA modified the provisions in procedure 8.B.4 to
clearly state that permitting authorities can use their existing compliance
data averaging procedures, in evaluating compliance with WQBELs below the
minimum quantification level.  Thus the permitting authority may specify that
effluent samples that are below the minimum quantification level should be
deemed equal to zero, equal to one-half the minimum quantification level, etc.
However, regardless of the averaging procedures used, the resulting value must
be compared to the WQBEL in assessing compliance.

      EPA addressed the confusion regarding the relationship between the
WQBEL, ML, and CEL by eliminating the CEL term and establishing the WQBEL
exactly as calculated as the appropriate compliance level.  The ML and other
established minimum quantification levels are specified in the NPDES permit to
clearly establish the compliance monitoring requirements.  EPA agrees with the
commenters that supported deferring to permitting authorities' existing
compliance data averaging procedures.  EPA believes that existing State and
Tribal procedures, in conjunction with the other permit requirements for
addressing WQBELs below the minimum quantification level, will result in
adequate consistency among States and Tribes.

4.    Pollution Minimization Program

      a.    Proposal:  In order to increase the likelihood that the
concentration of the pollutant in the effluent is as close to meeting the
WQBEL as possible, EPA included procedure 8.D in the proposed Guidance, which
provided that a pollutant minimization program (PMP) be specified and
implemented as a permit condition for each pollutant with a WQBEL below the
minimum quantification level.  This proposal reflected EPA's recognition that
effluent monitoring data alone is not sufficient to ensure that the WQBELs
below a minimum quantification level are being attained.  Under the proposed
PMP provision, a permittee was to develop a PMP to reduce all potential
sources of the pollutant in all internal, or indirect, wastewater streams
contributing to the permittee's wastewater collection system with the goal of
maintaining the effluent at or below the WQBEL.

      Under the proposal, the PMP included, but was not limited  to, the
following components:  annual review and semi-annual monitoring  of sources of
the pollutant; quarterly monitoring of the pollutant in the influent to the
treatment system; submittal of a control strategy for reducing loadings of the
pollutants of concern to the treatment system; implementation of appropriate
control measures which are consistent with the control strategy, as the
sources of the pollutants are discovered; and submittal of an annual status
report of activities.

      EPA expected the PMP to reflect that there are practical constraints on
treatment capabilities.  EPA did not view the PMP as a zero discharge
requirement.  Instead, EPA viewed it as a means to ensure that WQBELs were
achieved.  The effects of a PMP may be to reduce all pollutant(s) of concern
in the internal streams to non-detectable levels, but this is not equivalent
to a zero discharge* requirement.  A PMP-type requirement is consistent with
the guidance found in section 5.7.3 of the March 1991, TSD and in the May 21,
1990 "Strategy for the Regulation of Discharges of PHDDs and PHDFs from Pulp
and Paper Mills to Waters of the United States."  The proposal preamble
indicated that a permittee may be allowed to consider cost-effectiveness in
developing a pollutant minimization program.  In considering alternative

-------
                   Section Vm.H: WQBELs Below the Level of Quantification               423

elements of a PMP, the permittee was also allowed to consider the cost-
effectiveness of each element.

      b.    Comments:  EPA received many comments in opposition to or  in
support of the PMP provision in procedure 8.  Several commenters opposed the
PMP requirement on legal grounds.  One commenter questioned the need to
eliminate a pollutant merely because the WQBEL for that pollutant is below
detection.  Some characterized the PMP as an attempt to regulate pollutants
that are not being discharged by the permittee and asserted that EPA lacks the
legal authority under the CWA to regulate pollutants under such circumstances.
Others asserted that the PMP represents an unlawful attempt by EPA to  dictate
pollutant reduction strategies to the permittee.  They contend that the PMP
essentially specifies discharge levels for internal waste streams that exactly
equal the permit limits imposed on the effluent.  They assert that as  long as
the aggregated effluent discharged to the receiving water meets the WQBEL,
there is no justification for imposing a PMP on internal wastestreams.  These
commenters asserted that EPA's proposal to require PMPs conflicted with EPA's
decision in the OCPSF rule to reject in-plant limitations in favor of
traditional end-of-pipe limitations.

      Some commenters also asserted that the PMP contravenes  EPA's guidance,
particularly its May 1990 guidance pertaining to the pulp and paper industry,
by ignoring the role of wastewater treatment systems in reducing pollutant
discharge levels, especially when treatment is more efficient and cost-
effective. Some commenters criticize as arbitrary the assumption that
wastewater treatment facilities are only capable of treating to a level
between the WQBEL and the ML.  Other commenters objected that permittees would
be unable to determine what to include in their control strategies to  ensure
compliance because the guidance fails to prescribe criteria for assessing
minimum technology requirements.  Several commenters stated that the guidance
does not address how a municipality would implement a pollutant minimization
program, especially with respect to households.  Other commenters stated in
that context, that minimization programs would be equivalent to product bans
and zero discharge from indirect sources.  Other commenters stated that PMPs
should be required only when statistically quantifiable concentrations of the
pollutant are present in the discharge, based on a significant number  of
analyses demonstrating concentrations above the PQL.

      c.    Final Guidance;  Procedure 8.D of the final Guidance includes the
PMP components, largely as proposed, with certain additions.  In response to
comments and to prevent undue burden on permitees, the final Guidance
authorizes States and Tribes to consider cost-effectiveness when establishing
the requirements of a PMP.  The final Guidance also specifies that cost-
effective control measures consistent with the PMP's control strategy  shall be
implemented when sources of the pollutant of concern are discovered.   The
final Guidance also clarifies that fish tissue and other bio-uptake studies
can be used to monitor potential sources of the pollutant of concern.
Finally, it provides that any information generated as a result of this
provision can be used to support a request for subsequent permit
modifications, including revisions to or removal of the requirements of
procedure 8.D of appendix F (once adopted by a State or Tribe), consistent
with 40 CFR 122.44 and 122.62.

      In EPA's view, the PMP provisions will not result in any undue burden on
permittees since the States and Tribes will be allowed to consider a
permittee's costs and benefits analysis of any requirement that they impose
through the PMP.  Cost-effectiveness may also be considered.  EPA recognizes
that each industry type, size of facility, and type of industrial user for a
POTW will have unique considerations as to what is considered practicable.  In
addition, EPA recognizes that the household contribution of pollutants with
WQBELs below the quantification level would .best be controlled by public
education,  product bans, and establishing hazardous waste depositories.  EPA's
experience is that permittees, using these guidelines, can develop effective
PMPs that include clear measures for compliance with the PMP requirements.  It

-------
424    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

is not EPA's intent to require an open-ended PMP that cannot be  objectively
evaluated; rather EPA is promoting State and facility flexibility in
addressing facility-specific and pollutant-specific issues.

      The final Guidance also includes a provision in the TMDL section  to use
the PMP in evaluating an application for an exception to the mixing zone ban
for certain BCC discharges pursuant to procedure 3.C.6 of appendix F, once the
ban on mixing zones for BCCs is implemented.  The purpose of this language is
to clarify that the need for a mixing zone should be evaluated,  in part, based
on the extent to which a PMP can achieve a WQBEL calculated without a mixing
zone.

      In addition, -all PMPs are subject to revision as analytical methods
improve, new technologies become available, economic conditions  change, and
other factors are modified that affect the PMP.  EPA envisions that the
implementation of PMPs will be iterative where the annual results of the PMP
are used to modify subsequent updates of the PMP as appropriate, including
requiring more or less frequent monitoring or the removal of some or all of
the PMP requirements.

      EPA disagrees with the comments asserting that EPA, through the
pollutant minimization program, is unlawfully attempting to regulate
pollutants that are not present in a discharger's effluent.  The pollutant
minimization provision of procedure 8 clearly applies only when  a WQBEL for
that discharger is calculated below the level of detection.  Because no WQBEL
can be imposed with respect to a pollutant unless the permitting authority
determines that it has the reasonable potential to cause or contribute  to an
exceedance of water quality standards  (see procedure 5 of appendix F),
application of the PMP provision necessarily assumes that the pollutant is
present in the effluent, albeit in non-quantifiable amounts.  If the
permitting authority can demonstrate reasonable potential without the use of
data indicating detectable levels of the pollutant in the effluent as some
commenters wanted as a prerequisite for implementing a PMP, such as by  the use
of bio-uptake studies or documentation that the pollutant is present in the
wastestream at a concentration above the WQBEL, a PMP would be required.  If a
discharger can demonstrate that the pollutant does not have the  reasonable
potential to cause or contribute to an exceedance of water quality standards,
especially if the facility meets the provisions to obtain an intake credit to
address discharges at or below the intake water concentration as discussed in
procedure 5  (simple pass through), then no WQBEL and, therefore, no PMP would
be necessary.  Such demonstrations could include treatability studies to
document that the treatment process can remove the pollutant of  concern to the
WQBEL concentrations.

      Contrary to gome commenters' assertions, the goal of the pollutant
minimization program is not to eliminate the pollutant from a discharger's
effluent merely because it is present in  nondetectable amounts.   (EPA  notes,
however, that section 101 of the Clean Water Act establishes as  a goal  of the
Act the elimination of the discharge of pollutants into the Nation's waters.)
Rather, the purpose of the program is to ensure that a discharger's WQBELs are
achieved at the end of the pipe, a purpose that is compatible, as noted by
some commenters in apparent opposition to the pollutant minimization
provision, within EPA's overall authority to ensure that the ultimate levels
discharged after treatment are acceptable.  The applicable water quality
criteria dictate to the permitting authority what level of pollution is
acceptable in a discharger's effluent because of its effect on the surface
water.

      In order to ensure that acceptable level is reached, EPA is authorized
to require dischargers to monitor for  the presence of the pollutant  in-plant
if monitoring after treatment is not  a practical or feasible way to  evaluate
whether acceptable levels of the pollutant  in fact are present in the
effluent.  See 40 CFR 122.45(h).  Quarterly monitoring for the pollutant  in
the influent to the wastewater treatment system pursuant to the  pollutant

-------
                   Section Vm.H: WQBELs Below the Level of Quantification               425


minimization program, ideally prior to. commingling with and dilution by
wastestireams not bearing that pollutant, provides information for both the
permittee and the permitting authority regarding the potential detectable
quantity of the pollutant in the wastestream.  Among other things, this
information, in conjunction with bio-uptake studies of the effluent, allows
the permittee to evaluate the effectiveness of its treatment system  (which is
infeasible to evaluate when the pollutant is undetectable. in the wastewater
reaching the treatment system) and allows the permitting authority to estimate
the quantity of the pollutant that may be present in the 'effluent discharged
to the receiving water in order to determine whether it exceeds levels
necessary to protect the receiving waters designated and existing uses.

      Moreover, procedure 8 sets forth as a goal, not a requirement, of the
PMP that permittees reduce all potential sources of the pollutant as necessary
to achieve the WQBEL.  If the discharger can demonstrate, by means of an
analytical method specified in or approved under 40 CFR 136, that the effluent
is at or below the WQBEL, then procedure 8 does not apply and no PMP is
necessary.  However, if the discharger cannot feasibly or practically
demonstrate compliance at the end of the pipe, then in-plant compliance
monitoring is necessary and authorized by EPA's regulations and the CWA.

      The goal that a permittee's wastestreams meet the WQBEL prior to
treatment at the wastewater collection system is simply intended as a goal to
mitigate the effects of dilution, not as an explicit compliance requirement.
As noted above, however, if in-plant monitoring indicates, in conjunction with
flow and treatment data, that the pollutant is being discharged at levels
above the WQBEL but below the quantification level, then that data may be
evidence of noncompliance with the WQBEL.  The internal monitoring provision
also is intended to increase the likelihood of demonstrating that the
concentration of the pollutant in the effluent is at or below the WQBEL.

      Monitoring influent prior to treatment, as provided in the PMP, is
consistent with the longstanding principle that dilution as an alternative to
treatment is impermissible.  EPA expects that, in order to monitor for the
presence of a pollutant instream, a permittee will select a monitoring point
prior to dilution, of that wastestream by other wastestreams.  This will not
only help the permittee to identify and control -the amount of that pollutant
in the effluent, but it will also help the permittee ascertain the
effectiveness of the treatment technology and to promote the development of
even more sophisticated technologies that may remove greater quantities of the
pollutant.  Several courts have upheld EPA's authority to set effluent
limitations for a plant's wastewater stream before it is diluted internally by
other wastestreams, such as cooling waters.  See Texas Municipal Power Agency
v. EPA. 836 F.2d 1482 (5th Cir. 1988); Hercules. Inc. v. EPA. 598 F.2d 91
(D.C. Cir. 1978).

      In procedure 8, EPA does not go so far as to set in-plant effluent
limitations, but rather simply provides for internal monitoring and adoption
of control strategies with a goal of maintaining all sources of the pollutant
to the wastewater collection system below the WQBEL.  The WQBEL itself
continues to apply only at the end of the pipe, after treatment.

      Some commenters asserted that procedure 8's PMP provision conflicted
with EPA's decision in a 1987 effluent guideline regulation to reject in-plant
limitations in favor of traditional end-of-pipe limitations.  In that
regulation, which established effluent limitations and standards for the
organic chemicals,  plastics and synthetic fibers (OCPSF) category of point
sources,  EPA based it limits and standards on an in-plant treatment technology
but imposed monitoring requirements at the end of the pipe.  52 FR 42522  (Nov.
5, 1987).  EPA acknowledged in that regulation, as it does here today, that
the Clean Water Act provides no explicit authority for specifying technology.
See 52 FR 42560.  EPA further acknowledged, as it does here today, that
dischargers are allowed to select the means by which they would comply with
effluent limitations.  Id.  Finally, EPA also acknowledged that in-plant

-------
426    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

limitations would be inconsistent with, the Agency's general approach up to
that time, which was to determine compliance with effluent limitations at the
end of the pipe.  Id.

      Nevertheless, even at that time, EPA had already promulgated its
regulation authorizing, in certain circumstances, the imposition of effluent
limitations on internal waste streams.  See 40 CFR 122.45(h).  Moreover, prior
to that time, EPA had invoked that regulation to impose effluent limitations
in-plant.  While it is true, as EPA noted in the preamble to the OCPSF rule,
that EPA's general approach was to focus on end-of-pipe compliance, nowhere
did EPA say that approach was the only approach authorized by the Clean Water
Act.  Therefore, while EPA's decision to impose end-of-pipe limitations made
sense in the specific context of the OCPSF effluent guideline, it does not
prevent EPA from making a different policy decision in different
circumstances, such as these, where the WQBEL for the pollutant in question is
calculated below the level of quantification.  EPA also notes that the OCPSF
effluent guidelines were established at a level equal to or above the ML;
therefore, these effluent guidelines did not raise a quantification level
question.

      Some commenters objected that the PMP represents an unlawful attempt by
EPA to dictate pollutant reduction strategies to the permittee.  EPA disagrees
with this comment.  The PMP nowhere specifies what control measures, if any, a
discharger will need to implement in order to ensure that the effluent
discharged to the receiving water actually achieves the WQBEL.  Indeed, the
PMP provision affords the permittee considerable flexibility in meeting its
WQBEL; appropriate control strategies to reduce the particular pollutant could
include not only new treatment approaches but also source reduction and
substitution.  In short, a permittee implementing a PMP can devise any control
strategies it determines will minimize the presence of the pollutant in its
wastestreams.  EPA also recognizes that there are practical constraints on
treatment capabilities.  The PMP makes no attempt to dictate the treatment or
source reduction strategies that a permittee could or should implement.
Rather, EPA recognizes that the permittee is in a far better position to
devise innovative approaches to meeting its WQBEL.  The PMP is intended to
emphasize the opportunities afforded by source reduction up front, rather than
by traditional reliance on end-of-pipe treatment.  In this way, the pollutant
minimization program in procedure 8 is consistent with the national goals
articulated in the Pollution Prevention Act of 1990, 42 USC 13101, et seq.,
which criticizes the traditional regulatory emphasis on treatment and disposal
at the expense of source reduction.  See 42 USC 13101(a)(3).  Congress
therefore declared as part of a national policy on pollution prevention that
pollution should be prevented or reduced at the source whenever feasible and
that disposal or other release into the environment should be employed  only as
a last resort.  See 42 USC 13101(b).  By promoting the identification and
control of sources of pollution, the pollutant minimization program is
consistent with that national policy.

      As set forth in the preamble to the proposed guidance, EPA expects the
PMP specified in the final Guidance to reflect that there are practical
constraints on treatment capabilities.  EPA does not view the PMP as a  zero
discharge requirement.  Instead, EPA views it as a means to ensure that WQBELs
were achieved.  The effects of a PMP may be to reduce all pollutant(s)  of
concern in the internal streams to non-detectable levels, but this is not
equivalent to a zero discharge requirement.  A PMP-type requirement is
consistent with the May 21, 1990  "Strategy for the Regulation of Discharges of
PHDDs and PHDFs from Pulp and Paper Mills to Waters of the United States."  In
addition, in the final Guidance a permittee is allowed to consider cost-
effectiveness in developing a pollutant minimization program, including
alternative elements of a PMP.  For this reason, EPA is not prescribing
specific technology requirements for  PMPs, rather EPA will defer to the
permitting authority to determine the appropriate level of evaluation  and
controls to impose with the PMP.

-------
                  Section Vm.H: WQBELs Below the Level of Quantification               427

      EPA has determined that it is justified in requiring a permittee to
document that all practicable steps are being taken to comply with a WQBEL
that cannot be directly measured.  EPA and  State experience with such
requirements, especially in the paper industry for 2,3,7,8 TCDD and large
POTWs for PCBs, is that these PMPs lead to  reductions in the pollutant of
concern and compliance with the PMP requirements can be evaluated objectively.
Another way to looik at the PMP requirement  is as a public statement of intent
that the facility is undertaking a rigorous program to reduce the pollutant(s)
of concern to levels which will not adversely impact the 'environment.

5.    BCC Requirements

      a.    Proposal:  In the proposal, fish tissue monitoring or other bio-
uptake studies were required to assess PMP performance and trigger additional
PMP actions if elevated levels of BCCs were detected.

      b.    Comments:  Many commenters questioned the broad application of
caged fish samples, the use of resident fish studies, and other bio-uptake
studies in assessing a permittee's compliance with their PMP and, by
inference, the WQBEL.  Specific concerns  included that caged fish may not
receive a natural diet due to being physically restricted to cages, the
resident fish may be subjected to contaminated food sources not attributed to
the facility, and costs associated with these and other bio-uptake studies
would be burdensome to small facilities.

      c.    Final Guidance;  Based upon the comments received from the States
and other commenters, EPA believes that fish tissue and other bio-uptake
monitoring programs can be properly designed to provide quality data to use as
an assessment tool for reasonable potential determinations and as an
indication of ambient pollutant concentrations.  However, the cost of
performing such tes,ts may not be justified  for all facilities, especially if
they can identify all the sources of the  contaminant and demonstrate that they
are already doing everything practicable  to eliminate them.  Since the
original intent of this procedure was to  use the fish monitoring and other
bio-uptake studies to evaluate PMP performance, EPA does not believe it is
necessary to require such monitoring methods universally for all cases.
Therefore, proposed procedure 8.F has not been included in the final Guidance.

      Nevertheless, the final Guidance encourages the States and Tribes to
include fish tissue sampling or any other test procedures they deem necessary
to properly evaluate the performance of a PMP.  Fish tissue monitoring can be
an effective means -for some facilities to document that the PMP actions
accomplish the desired load reductions to achieve WQBELs and, therefore, can
provide a factual basis to allow a facility to reduce the PMP requirements or
even remove such permit conditions.  Where  facilities are expected to
discharge levels of BCCs, above the WQBEL but below the minimum quantification
level, fish monitoring data can be valuable in assessing PMP performance and
determining the need for additional pollution reduction efforts.

6.    Other Conditions

      a.    Proposal:  The proposed Guidance included a provision to allow
States and Tribes to require a permittee, on a case-by-case basis, to develop
or use analytical techniques more sensitive than the ones specified in the
permit, internal waste stream monitoring  and other methods capable of
adequately determining the compliance status of the effluent.

      b.    Comments:  EPA considers these  optional measures suitable for
inclusion in a PMP and, therefore, addressed them under the PMP comment
discussion and the analytical methods comment discussion.

      c.    Final Guidance:  Since these  measures are intended to be optional
and they are suitable for PMP purposes or special permit conditions, EPA

-------
428     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

decided.that a  separate  provision  for including these  measures  is not
necessary.

-------
                          Section Vffl.I: Compliance Schedules                       429

I.    Compliance Schedules

      For purposes of the final Guidance, a compliance schedule refers to-an
enforceable sequence of interim requirements  in a permit leading to ultimate
compliance with water quality-based effluent  limitations  (WQBELs) in
accordance with the Clean Water Act  (CWA).  This procedure allows, but does
not require, -States or Tribes to include  such compliance schedules in .permits
under appropriate circumstances.  A State or  Tribe authorized to administer
the National Pollutant Discharge System  (NPDES) may exercise discretion when
deciding if a compliance schedule is justified because of the technical or
financial infeasibility of immediate compliance.

1.    Proposal

      Procedure 9 of the proposed Guidance allowed compliance schedules only
for "existing dischargers," not "new" or  "increasing" dischargers.  For
existing dischargers whose permits were reissued or modified to contain more
stringent limitations based upon certain water quality requirements, the
permit could allow up to three years or the length of the permit, whichever
was less, to comply with such limitations.  The provision applied to effluent
limitations based on a Tier I criterion, Tier II value, whole effluent
toxicity (WET) criterion, or narrative criterion, provided the criterion or
value was adopted (or, in the case of a narrative criterion, newly
interpreted) after July 1, 1977.

      Under the proposal, where such a schedule of compliance exceeded one
year, interim requirements were to be specified and interim progress reports
submitted at least annually.

      For existing dischargers with more  stringent effluent limitations based
on Tier II values, .the proposal allowed a reasonable period of up to two years
for the permittee to conduct studies necessary to develop a Tier I criterion
or modify the Tier II value.  After completion of such studies, and upon an
appropriate showing, the permit could be modified or the compliance schedule
could be extended to the end of the permit.   The proposal stated that such
permit modifications would not be affected by the anti-backsliding provisions
of section 402(o) of the CWA.  For further discussion on anti-backsliding
provisions, see section II.C of this document.

      These compliance schedule provisions were included in the proposal
because of the potential for existing dischargers to have more stringent
effluent limitations, under the final Guidance, for which immediate compliance
would be impossible or impracticable.  Schedules of compliance to accommodate
such situations may be included in permits only if the State water quality
standards or implementing regulations authorize them.

2.    Final Guidance

      As explained below, the final Guidance  follows the general approach of
the proposal; however, several modifications  were made to address issues
raised by commenters.  These modifications include expanding the definition of
existing discharger, extending the possible length of compliance schedules
from a maximum of three years to a maximum of five years, allowing schedules
of compliance to extend beyond the term of the permit in limited situations,
and clarifying some of the language.

a.    Eligibility

      The proposal did not allow schedules of compliance for dischargers who
commenced discharging or increased flow, concentration or loading after the
effective date of the final rule.  States and the regulated community objected
to excluding from eligibility those permittees with an increased discharge or
those who commenced discharging after the Guidance becomes effective.  They
specifically argued that the availability and duration of compliance schedules

-------
430    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

should be within the permitting authority's discretion, the new  requirements
may be difficult for new, existing or increasing dischargers to  meet, and
exclusions of some dischargers may result in competitive disadvantages.  For
example, a new permittee may have already invested a lot of resources
(including time) in the planning and construction of a facility  only to be
immediately out of compliance with the final Guidance's requirements.

      As a. result of considering the comments, EPA made some changes to
increase eligibility for compliance schedules.  Specifically, EPA revised the
Guidance to eliminate the definition of "increasing discharger"  and redefined
the term "existing discharger" to include any discharger which is not a "new
Great Lakes discharger."  The definition of a "new Great Lakes discharger"  (in
§ 132.2 of the final Guidance) includes "any building, structure, facility, or
installation from which there is, or may be, a 'discharge of  pollutants', the
construction of which commenced after March 23, 1997."  The final Guidance's
revised definitions were modeled after the existing 40 CFR 122.2 definitions
for parallel terms, but with the cut-off date modified to reflect the date by
which States or Tribes must adopt provisions consistent with the final
Guidance.  Only "new Great Lakes dischargers" are required to comply
immediately upon commencement of discharge with effluent limitations derived
from a Tier I criterion, Tier II value, whole effluent toxicity  criterion, or
narrative criterion.  Therefore, existing dischargers, including those
previously defined as increasing dischargers, are eligible for schedules of
compliance to meet more stringent limitations derived from specified criteria
and values.

      EPA has included increasing dischargers within the category of existing
dischargers since they are factually closer to existing dischargers than to
new dischargers.  Increasing dischargers may be existing facilities which have
a change (an increase) in their discharge.  Such facilities may  include those
with seasonal variations.  Increasing dischargers will already have treatment
systems in place for their current discharge.  Thus, they have less
opportunity than a new discharger does to design and build a new treatment
system which will meet new water quality-based requirements for  their changed
discharge.   Allowing existing facilities with a changed discharge  (increasing
discharger) a compliance schedule will avoid placing them at a competitive
disadvantage vis-a-vis other existing dischargers, who are eligible for
compliance schedules.

      The final Guidance retains the prohibition against compliance schedules
for new Great Lakes dischargers because, as defined in § 132.2,  these
permittees are the facilities whose construction commences more  than two years
after the final Guidance is published in the Federal Register.   Therefore,
these permittees will have had ample notice of the Guidance's new requirements
and should have included the requirements in the planning of the new facility.
Continuing this prohibition is also consistent with the national regulations.

      The final Guidance does not prohibit the use of a short-term  "shake-down
period" for new Great Lakes dischargers as is provided for new sources  or new
dischargers in 40 CFR 122.29(d)(4).  These regulations require that the owner
or operator of a  CO new source;  (2) a new discharger  (as defined in 40 CFR
122.2) which commenced discharge after August 13, 1979; or  (3) a recommencing
discharger shall install and implement all pollution control equipment  to meet
the conditions of the permit before discharging.  The facility must also meet
all permit conditions in the shortest feasible time  (not to exceed  90 days).
This shake-down period is not a compliance schedule.  This approach may be
used to address violations which may occur during a new facility's  start-up,
especially where permit limits are water quality-based and biological
treatment is involved.

      Another approach is to use prosecutorial discretion as an  unofficial
shake-down period. • That is, the permitting'authority may elect  not to  take
enforcement action against a new  source which has installed the  necessary
treatment prior to discharging and is making a good faith effort to come  into

-------
                           Section Vffl.I: Compliance Schedules                       431

compliance as  soon as possible.  Alternatively, the permitting authority may
issue a compliance order  (under section 309(a) or equivalent State authority)
requiring compliance by a  specified date, where circumstances warrant.

b.    Duration of Compliance Schedules

      Paragraphs 9.B and C of the final Guidance set forth the basic
requirements for compliance schedules for existing dischargers with new or
more restrictive limitations.  As a result of the changed definitions
discussed above, these paragraphs now also apply to those dischargers formerly
considered increasing dischargers.  There are two principle changes from the
proposal:  the maximum length of compliance schedules is changed from three to
five years in  limited circumstances, and the proposed requirement that
schedules of compliance not extend beyond the term of the permit is
eliminated.

      EPA received comments from States, the regulated community and
environmental  groups on the issue of duration of compliance schedules.  States
and the regulated community commented that the proposed three-year maximum
duration for compliance schedules was too short.

      With regard to the time to meet post 1977 Tier I criterion/Tier II
values, whole  effluent toxicity criteria, or narrative criteria, the final
Guidance provides for compliance schedules for up to a maximum of five years.
EPA continues  to believe that compliance schedules of three years or less will
be sufficient  to allow facilities to make the changes necessary to meet new or
revised discharge requirements in most cases.  Such compliance periods are
consistent with analogous  provisions of the CWA including sections 301(b)(2)
and 304(1).  For example,  section 301(b)(2)  (C) -  (F) of the Act provided that
various technology-based effluent limitations shall be complied with as
expeditiously  as possible  but no later than three years after effluent
limitation guidelines are  promulgated.  Similarly, section 304 (1) provides
that sources shall comply  with individual control strategies  (water-quality
based requirements) within three years.

      However, the Agency  also recognizes the concerns raised by commenters
regarding the  amount of time and resources in some cases that may be needed
for implementing certain new treatment technologies.  Commenters have asserted
that installing state-of-the-art treatment to meet the final Guidance's new
requirements would include the following efforts: identifying the appropriate
technology,  evaluating budgetary constraints, installing the technology, and
effectively implementing the new treatment system.  The regulated community
commented that these efforts are resource and time intensive and, therefore,
would not be feasible to accomplish within the proposed three year time
period.

      EPA acknowledges that in limited situations it may be difficult to
accomplish the objectives  listed above in three years to identify, design, and
implement complex state-of-the-art treatment technology.  The Agency also
recognizes that evaluation, design, and implementation of facility-wide
comprehensive pollution prevention control strategies involving product
substitution, process line changes, new piping, new raw materials and revised
waste handling, recycling, and disposal procedures may require more than three
years at large facilities.  In addition, EPA is aware that the technical and
administrative process of  modifying and implementing revised requirements for
numerous industrial users  at POTWs, as well as planning, budgeting, and
undertaking significant new construction to change treatment processes at a
municipal treatment works, may require more than three years.

      Therefore, the final Guidance has been revised to provide that
compliance schedules may provide for up to five years to meet new or more
stringent effluent limitations in those limited circumstances where the
permittee can  demonstrate  to the permit authority that such an extended
schedule is warranted.  The Agency emphasizes its belief that in most

-------
432    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

situatipns less than three years will be required.  EPA believes that permit
authorities should consider shorter compliance schedules wherever possible or,
alternatively, not allow compliance schedules where unnecessary.

      Comments from an environmental group recommended immediate compliance or
that only one additional year be allowed for facilities to comply with the new
requirements.  Environmental groups were concerned that the three-year maximum
would become a default compliance schedule duration for facilities.  As
discussed above, the final Guidance has been revised to provide for a maximum
five-year compliance schedule.  EPA emphasizes that the five-year is a
"maximum" and that permitting authorities, based on their discretion, should
consider shorter periods of time, or alternatively, not allow compliance
schedules.  If a permit grants a schedule of compliance where the need for a
schedule has not been substantiated, the permit would be objectionable.  EPA
has review authority to ensure that the additional time period is "reasonable"
and not based upon a standard practice of granting additional time without
appropriate technical justification.  The burden is on the permit applicant to
justify a compliance schedule.

      The regulated community also commented that toxicological studies are
difficult to conduct and that facilities may not be able to complete the
research within the three-year maximum.  The regulated community also argued
that once studies are completed  (which justify modifying a permit to
incorporate the results) permittees may need additional time to modify the
operation of their facility to reflect the studies' findings.

      If permits are modified to reflect the results of studies conducted by a
permittee and/or tq allow for a changed operation, the public would have an
opportunity to comment upon the appropriateness of any extension of the
compliance schedule as well as the modified WQBEL.

      Some commenters were concerned that conducting toxicity tests to justify
new or revised criteria was not a "routine exercise" and warranted a longer
compliance schedule.  However, both the proposal and the final Guidance
address this concern for permits based on Tier II values by providing up to
two years for completion of studies and calculation of a revised limit where
appropriate, and then up to five years additional time to come into compliance
with the applicable limit.  With respect to commenters' concerns over the
length of time needed to perform toxicity tests, EPA's experience indicates
that the necessary toxicity studies can likely be accomplished in one year.

      Some States felt that limiting compliance schedules to the term of, or
expiration date of, a five-year permit was not realistic and did not provide
adequate time for permittees to implement new technologies or complete
additional research studies.  These States claimed that this scenario would be
especially true for those permittees in the third or fourth year of their
permit who would not be allowed the maximum schedule of three years for
compliance due to their permit's expiration date.  The regulated community
commented that limiting the duration of compliance schedules to the term of
the permit could create unfair competitive disadvantages and that three years
was an unrealistic amount of time for permittees to change their operations or
install new technologies to comply with the new requirements and to complete
the additional toxicological studies for developing a Tier I criterion or
modifying a Tier II value.

      EPA recognizes that permittees in the third or fourth year of a permit
may receive modified or revised permits reflecting the final Guidance's new
requirements.  If the permitting authority has no discretion to extend the
compliance schedule beyond the permit's expiration date, but immediate
compliance is infeasible, these permittees may be out of compliance and
potentially subject to enforcement actions.  Such violations may be avoided
with the discovery and implementation of new technologies or revised limits
due to completed additional studies by industry.  EPA has addressed this

-------
                          Section Vin.I: Compliance Schedules                       433

comment.by further clarifying the maximum five-year compliance schedule
approach in the next paragraph.

      While the final Guidance has been changed to all compliance schedules to
extend to a maximum duration of five years  (see discussion above), EPA
recognizes that where a permit is modified near the end of the permit term,
the permittee.may still need a full five years to comply.  The Agency finds no
persuasive reason for distinguishing between these permittees and permittees
who are in the earlier part of a permit cycle.  Therefore, the final Guidance
provides that the compliance schedule can go beyond the term of the permit.
When this occurs, an interim permit limit effective upon the permit expiration
date shall be included in the permit, in effect giving the permittee up to the
same five years.  The fact sheet and administrative record shall address the
final limit and its compliance date.

      States and the regulated community also raised issues concerning EPA's
interpretation of its anti-backsliding provisions.  States and the regulated
community commented that EPA's policies were unclear as to whether compliance
schedules would be affected by anti-backsliding provisions.  They also
questioned whether revised permit limits based on new information resulting
from the completion of additional studies by a permittee would present an
anti-backsliding issue.  The anti-backsliding requirements of section 402(o)
of the CWA do not apply to revisions to effluent limitations made before the
scheduled date of compliance for those limitations.  Anti-backsliding
requirements are discussed in section II.C.3 of this document.

-------
434     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
                           Section DC: Executive Order 12866                       435
              IX.  EXECUTIVE ORDER 12866


A.    Introduction 'and Rationale for Estimating Costs  and Benefits for the
      Great Lakes Water Quality Guidance

      Under Executive Order 12866,  [58 FR 51735 (October 4, 1993)] the Agency
must determine whether the regulatory action is "significant" and therefore
subject to Office of Management and Budget (OMB)  review and the requirements
of the Executive Order.  The Executive Order defines a "significant regulatory
action" as one that is likely to result in a rule that may:

      (1)   Have an annual effect on the economy of $100 million or more or
adversely affect in a material way the economy,  a sector of the economy,
productivity, competition, jobs,  the environment,  public health or safety, or
State, local, or tribal governments or communities;

      (2)   Create a serious inconsistency or otherwise interfere with an
action taken or planned by another agency;

      (3)   Materially alter the budgetary impact of entitlement, grants, user
fees, or loan programs or the rights and obligations of recipients thereof; or

      (4)   Raise novel legal or policy issues arising out of legal mandates,
the President's priorities,  or the principles set forth in the Executive
Order.

      The final Guidance establishes minimum water quality criteria,
antidegradation provisions,  and implementation procedures for the Great Lakes
System.   Implementation of these provisions is dependent upon future
promulgation of provisions consistent with the final Guidance by State or
Tribal agencies or, if necessary,  EPA.  Until actions  are taken to promulgate
and implement these provisions (or equally protective  provisions consistent
with the final Guidance),  there will be no economic effect of this rule on any
entities.

      EPA anticipates that States or Tribal agencies will promulgate
provisions consistent with the final Guidance.   Therefore, pursuant to the
terms of Executive Order 12866,  it has been determined that this rule is a
"significant regulatory action"  because it may impose  an annual cost or
benefit to the economy of $100 million.   As such,  this action was submitted to
OMB for review.   Changes made in response to OMB suggestions or
recommendations will be documented in the public record.

      Executive Order 12866  requires EPA to submit an  assessment and
description of potential costs and benefits anticipated from the regulatory
action and reasonable alternatives considered for "significant" regulations
meeting the definition described above.   Under the Clean Water Act (CWA),
costs and benefits are not directly relevant in establishing water quality
criteria.  However, if a range of scientifically defensible criteria that are
protective of the designated use in question is identified, costs may be
considered in selecting a particular criterion within  that range.  In
addition, under EPA's regulations,  certain costs can be considered in the
context of use attainability analyses, variances,  and  antidegradation.
Moreover, as a matter of good government,  EPA considers costs and benefits in
making its decisions.  Accordingly,  consideration of costs and benefits has
been an integral part of the deliberations involving the States, environmental
interests,  the regulated community,  and EPA in the development of the Great
Lakes Water Quality Guidance.

-------
436    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      In addition to the requirements set forth in Executive Order 12866, the
States of Ohio, Michigan, and Wisconsin had specifically requested that EPA
examine the costs and benefits of the Guidance.  Some members of the Public
Participation Group also expressed a concern during the Steering Committee
meetings that the costs to point sources could be sizeable.  Other public
comments expressed .concern about the need for specifically identifying the
benefits.  During the Steering Committee deliberations in November and
December 1991, EPA expressly committed to estimate various costs and benefits
that could accrue from the proposed Guidance and to include that analysis in
the public comment process associated with the proposed regulation.

      As noted above, Executive Order 12866 requires EPA to prepare an
assessment of the potential costs and benefits for proposed and final major
rules.  The studies described below  (collectively referred to as the
Regulatory Impact Analysis) have been submitted to OMB to fulfill this
requirement.  These cost and benefit studies are summarized below.  The
documents underlying this summary are available in the administrative record
for this rulemaking.

B.    Summary of Proposal

      The following discussion briefly describes how EPA prepared the original
estimate of both costs and benefits associated with the proposed Guidance.
More detailed information regarding the procedures used by EPA to estimate
compliance costs for the proposed Guidance is included in the EPA  Assessment
of Compliance Costs Resulting from Implementation of the Proposed Great Lakes
Water Quality Guidance  (April 16, 1993).  In general, aggregate costs were
estimated for all direct  (i.e., point source) and indirect dischargers in the
Great Lakes System.  Benefits and costs were also assessed for direct
industrial and municipal point source dischargers at three case study sites in
the Great Lakes System, the results of which were not appropriate to use in
evaluating the aggregate benefits and costs of the proposed Guidance.  Section
IX.C.I of this document discusses modifications to the approach for the
proposed Guidance that were made to estimate costs for the final Guidance.

l.    Costs

      EPA focused its initial assessment on  categories of industries and
municipalities that would be likely  to be affected by the proposed Guidance.
EPA conducted a detailed review of a random  stratified sample of direct
dischargers to serve as the basis for reasonable extrapolation to the universe
of Great Lakes System dischargers.  These sample facilities were considered
representative of all types and sizes of facilities in the Great Lakes Basin.

a.    Method

      EPA selected 50 sample facilities to represent the estimated 588 major
dischargers and 9 facilities to represent the 3,207 minor dischargers in the
Great Lakes Basin. '  For major dischargers, sample facilities were selected
from each of the major  categories of facilities, which  included nine primary
industrial groups and a category for municipalities, also known as Publicly
Owned Treatment Works or  POTWs.  The nine industrial categories were Mining,
Food and Food  Products,  Pulp and Paper, Inorganic Chemical Manufacturing,
Organic Chemical Manufacturing/Petroleum Refining, Metals Manufacturing,
Electroplating/Metal Fabrication, Steam Electric Power  Plants, and
Miscellaneous  facilities  (e.g., remedial clean-up discharges, tire
manufacturers).  Sample major facilities were also selected to ensure
representation across facility size  (as measured by discharge flow volume),
through stratification  by flow within  each category.

      Because  minimal compliance costs were  anticipated by minor  dischargers,
EPA analyzed  a limited  number of randomly selected minors  to verify  that
assumption.   Furthermore,  because EPA  had limited discharge flow  data  for

-------
                           Section DC: Executive Order 12866                        437
minor dischargers, it was not possible.to adopt a flow-stratified analytical
plan similar to that used for majors.

      For each sample facility under review, the most current NPDES permit
data and background information were collected to calculate the limits that
would be anticipated from current regulatory requirements  (if not incorporated
into the current permit) and to develop additional  (hypothetical) permit
requirements based on the proposed Guidance.  EPA gathered information from
State and Regional files that included permit applications, permit fact sheets
or rationales, inspection reports, discharge monitoring reports, pretreatment
reports, short-term waste characterization studies, receiving stream low-flow
scenarios and total maximum daily loads/waste load allocation reports.  In
addition, EPA used any other readily available information including industry-
wide studies of various industrial categories used in developing effluent
guidelines.

      For each sample facility, new hypothetical permit limits and permit
conditions were developed based on the implementation procedures in the
proposed Guidance.  The proposed criteria would require some permitted
facilities to meet new limits and adopt other permit conditions such as whole
effluent toxicity testing and additional monitoring.  New limits were
calculated for those 32 pollutants for which numeric Tier I criteria were
proposed.  For a given facility, only those pollutants that were detected in
the discharge, or expected to be present in the discharge but reported as not
detected because of use of less sensitive EPA approved analytical methods,
were evaluated.  The need for whole effluent toxicity limits and monitoring
was also evaluated .in accordance with the proposal.  For each facility, limits
were calculated for the outfalls that contained or might contain observed or
anticipated loadings for the pollutants of concern.

      If the existing effluent limits for some of the permitted facilities
selected did not reflect current State water quality standards and
implementation policies, EPA calculated hypothetical, alternative permit
limits to reflect the newly-revised State standards and requirements, which
are based on the adoption of toxic water quality standards under section
303(c)(2)(B) of the CWA (referred to here as baseline requirements).  This
approach more accurately reflected differences between existing effluent
limits based on newly revised State requirements and procedures required in
the proposed Guidance.

      In determining specific requirements imposed by the proposed Guidance,
site-specific wasteload allocations  (WLAs) were calculated for discharges to
both the open waters of the Great Lakes and their tributaries using equations
set forth in the proposed implementation procedures.  Because of the general
lack of readily available background concentration data for receiving waters,
two different WLAs were calculated for each sample facility.  The first WLA
assumed zero background in the absence of background data  {WLA #1).  The
second WLA assumed a value for background concentrations where no background
data existed  (WLA $2).  The assumed background values were approximately 50
percent of the proposed Guidance water quality criteria.

      The resulting WLAs were then used to establish water quality-based
effluent limits (WQBELs) for the sample facilities; the daily maximum WQBEL
for a pollutant was set equal to the Final Acute Value, which represents the
WLA to achieve the acute aquatic life criterion; monthly average WQBELs were
set equal to the most stringent WLA calculated to protect chronic aquatic
life,  wildlife, or human health criteria.  When negative WLAs were calculated
for a pollutant (because of high background concentrations of pollutants
reported for a receiving water), two different sets of WQBELs were calculated
for each facility, which resulted in different compliance cost scenarios.  In
cases where negative WLAs were calculated using WLA #1, the WQBEL was set
equal to the background concentration  (WQBEli #1); when negative WLAs were
calculated using WLA #2, then the WQBEL was set equal to the most stringent
water quality criteria  (WQBEL #2).

-------
438    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      If either WQBEL #1 or WQBEL #2 was more stringent than the existing
effluent limits--either in current permits or calculated against current
regulatory requirements--then EPA developed costs based on options that would
likely be available to the facility to comply with the more stringent effluent
limits.

      To estimate costs to the particular facilities reviewed and to develop
potential compliance options, an engineering analysis for each facility in the
sample was conducted.  This included a review of existing treatment systems at
the facility and an assessment of the need to add new or supplement existing
treatment capabilities.  Having defined the control options, EPA estimated the
compliance costs to facilities implementing each option.  Compliance costs
generally included treatment costs, monitoring, and operations and maintenance
costs, and a variety of one-time costs of limited durations (e.g., waste
minimization audits of production processes).  Residual management costs were
also estimated for industrial and municipal facilities that were projected to
install end-of-pipe treatment and generate additional sludge (e.g., sludge
produced from chemical precipitation).

      If the analysis showed that additional treatment was the most likely
control method to be used to comply with either WQBEL #1 or #2, then EPA
generally assumed that this treatment would be added as an end-of-pipe unit
process (i.e., the treatment unit process would be added at a point just prior
to discharge to the receiving water).  While additional treatment at end-of-
pipe may be neither technically nor economically efficient in a variety of
circumstances, EPA did not have the necessary facility- or process-specific
information such as contributing wastewater flows, in-plant treatment
capabilities or opportunities, process waste characteristics,  or recycling
capabilities that would allow an assessment of other potentially less
expensive alternatives.  However, in the majority of instances, additional
end-of-pipe treatment was not projected for a facility.  This was the case
where existing treatment facilities could accomplish the required treatment,
current permit requirements or construction plans were already in place to
provide the additional treatment, the incremental amounts of pollutants to be
removed were insignificant, or waste minimization/pollution prevention control
techniques were believed to be adequate to comply with the Guidance.  In each
of these instances, appreciable treatment capital costs directly attributable
to the proposed Guidance were not anticipated.

      In the case of municipalities, or POTWs, consideration was given to the
number and types of industrial users discharging to the collection system, as
well as the size of the POTW.  If additional pretreatment controls or
modifications seemed unlikely to achieve the pollutant reductions, then
additional treatment at the POTW was considered the next most likely option.

      Monitoring costs for permitted facilities were also estimated.  In those
cases where additional parameters and limitations were deemed necessary
because of the Guidance, the monitoring regimes  (i.e., sampling  frequency)
were established consistent with the existing monitoring requirements for
other parameters.  Monitoring costs were then estimated based upon average
costs per analytical method for the more common techniques.

      Because the discharge of bioaccumulative chemicals of concern  (BCCs) are
of special concern under the proposed Guidance, EPA included monitoring-only
costs for Tier I BCCs for all affected facilities, regardless of whether Tier
I BCCs were detected or expected to be present in a discharge.

      A number of other costs were also considered depending on  the specific
circumstances surrounding a particular type  of facility.  These  were generally
one time costs related to pollutant minimization studies, bioconcentration
studies, whole effluent toxicity testing, pretreatment program revisions,
waste minimization audits, and implementing  pollution prevention techniques.
Generally, these costs were  included with the  capital costs for  purposes of
calculating annualized costs of compliance.

-------
                           Section IX: Executive Order 12866                        439
      Four different cost estimates were developed to account for differences
between limits based on WLA #1 .(zero background absent actual data) and WLA #2
 (assumed 50 percent of the most stringent Guidance criteria as background
absent actual data), as well as the potential range of costs associated with
implementation of waste and pollutant minimization studies and controls.
These scenarios are described below:

      Scenario 1:  Limits based on WLA #1 and the low end of the estimated
range of waste minimization costs for all facilities.

      Scenario 2:  Limits based on WLA #2, the middle of the estimated range
of waste minimization costs for industrial facilities, and high end of the
estimated range costs for the POTWs aggressively implementing the pretreatment
program to promote source control.

      Scenario 3:  Limits based on WLA #2, the middle of the estimated range
of waste minimization costs for industrial facilities, and end-of-pipe
treatment installation by POTWs.

      Scenario 4:  Limits based on WLA #2, high end of an estimated range of
waste minimization costs, and end-of-pipe treatment installation by POTWs.

      The major difference between Scenario 2 and Scenario 3 was the emphasis
on pollution prevention versus end-of-pipe treatment.  Assumptions underlying
Scenario 2 emphasized pollution prevention through source control.  Scenario 3
focused on end-of-pipe treatment, especially at POTWs.

      To develop a single cost estimate for each facility for each scenario
described above, the three cost categories mentioned above (treatment,
monitoring, and one-time costs) were combined into a single annualized cost,
which reflects the annual economic costs associated with recurring activities
 (e.g., compliance monitoring, and operation and maintenance), repaying capital
expenses, and special studies.  Annualized costs were calculated by assuming
that all capital costs and special study costs would be paid by borrowing
money at an interest rate of seven percent and paying it back over a 10-year
period.  Annual costs of monitoring, operation, and maintenance were added
directly.

      Given a single estimate of the annualized cost for each facility, the
procedure for extrapolating costs from the sample to the entire population was
pre-determined by the stratified random sampling procedure used to select the
subset of facilities examined in detail.  Using the single annualized cost
figure for each plant, an estimate of the cost for each category was
calculated by averaging the values for applicable (sample) plants and then
multiplying by the 'total (population) number of plants in that stratum.  The
cost estimate for the category was calculated by summing the strata in the
category.  The cost estimate for the entire universe of facilities was the sum
across categories.  This procedure was followed to estimate costs for each
scenario.

      EPA identified an estimated 3,500 indirect industrial dischargers that
discharge to POTWs in the Great Lakes System and developed preliminary
estimates of compliance costs for them.  These preliminary cost estimates were
based on the assumption that indirect dischargers affected by the proposed
Guidance would incur costs comparable to those incurred by direct industrial
dischargers.  In addition,  EPA assumed that costs to industrial users subject
to categorical pretreatment standards would be higher than the costs to non-
categorical significant industrial users.  The following four scenarios for
indirect dischargers are consistent with the four cost scenarios developed for
direct dischargers.

      Scenario 1:  Assumes that 10 percent of all indirect dischargers in the
Great Lakes Basin would install additional controls.

-------
440    Water Quality Guidance for the Great Lakes System — Supplementary Information Document


      Scenario 2:  .Assumes that 30 percent of all indirect dischargers in the
Great Lakes Basin would install additional controls.

      Scenario 3:  Assumes that 20 percent of all indirect dischargers would
install additional controls.

      Scenario 4;  Assumes that 20 percent of all indirect Dischargers would
install additional controls  (same as Scenario 3).

      The estimated percent of indirect dischargers affected by the proposed
Guidance was based on an assessment of conditions involving industrial users
and their toxic discharges to a moderately large POTW in the Great Lakes
Basin.

b.    Results

      The total annual!zed compliance costs of implementing the proposed
Guidance to direct and indirect dischargers were estimated to be between $80
million under Scenario 1 and $505 million under Scenario 4.  EPA considered
Scenario 2 to be the most likely scenario of the four scenarios evaluated
because it was assumed that dischargers would pursue more cost effective
approaches to comply with the Guidance (as opposed to always installing
treatment), and estimated the annualized compliance cost to be about $192
million.  Under Cost Scenario 2, direct dischargers accounted for 59 percent
of the total estimated compliance cost of the proposed Guidance; indirect
dischargers accounted for 41 percent of the total estimated cost.  Direct
major industrial dischargers accounted for 32 percent of the total estimated
cost under Scenario 2; direct major POTWs accounted for 21 percent of the
total cost.  Major POTWs accounted for the largest proportion of total
annualized costs borne by any of the categories of dischargers; the mining
category was estimated to incur the lowest proportion of costs for the
universe of industrial categories.

      Annualized capital costs accounted for about 7 percent of the total
annual cost for majors, but none of the costs for minors, which did not
require investment in treatment technology.  Waste/pollutant minimization
studies and implementation of appropriate controls/techniques were a very
significant portion of the total expected cost of the proposed Guidance.  The
total annualized costs of such pollutant controls made up about 54 percent of
the costs for direct dischargers under Scenario 2.  The annual cost of
monitoring, operating and maintaining equipment, etc., made up about 36
percent of the estimated annual costs for direct dischargers.  Special
monitoring studies accounted for about 4 percent.

      EPA identified several limitations in the scope of the cost study for
the proposal and performed several sensitivity analyses in an effort to
evaluate the impacts of these study limitations on the total estimated
compliance cost.  The analyses performed included:

      - - evaluating the cost impact assuming that Tier I BCCs are found
bioaccumulating as a result of Guidance monitoring requirements when Guidance-
based WQBELs were below analytical detection levels;

      - - estimating the impact of the proposed antidegradation requirements;

      -- analyzing the potential effect should BCCs be detected  in the future;

      -- evaluating the impact of eliminating mixing zones for BCCs;

      -- evaluating the prevalence of Tier  II BCCs and potential BCCs; and

      - - evaluating the potential costs associated with the proposed options
to control pollutants  in intake waters.

-------
                           Section DC: Executive Order 12866                        441
Based on these sensitivity analyses, EPA estimated that an additional $40
million annually in costs may also be attributable to the proposed Guidance.
This incremental cost represented about a 21 percent increase above the cost
study estimate.  When added to the most likely cost study estimate  (i.e., Cost
Scenario 2), the total compliance costs were estimated to reach as high as
$232 million annually.

2.    Cost-Effectiveness

      Based on the cost estimates described above, EPA evaluated the cost-
effectiveness of the proposal.  Cost-effectiveness is defined as the
incremental annualized costs of a pollution control option per incremental
pollutant removal.  Cost-effectiveness is measured using pounds of pollutant
removed weighted by an estimate of the relative toxicity of the pollutant
(referred to as pounds-equivalent).  As also described in section IX.C.l.g of
this document, EPA derives toxic weights through standardizing pollutant
criteria by using the original EPA criterion for copper.

a.

      EPA estimated pollutant loadings reductions to indicate the decrease in
pollutants  discharged due to more stringent proposed Guidance-based limits.
Baseline loadings were determined in pounds per day by multiplying the
existing permit limit or effluent concentration by the facility's average flow
rate and a  unit conversion factor.  If either the permit limit or effluent
concentration for a pollutant was reported as less than a detectable level,
the reported detection level was divided in half.  Using extrapolation
procedures  similar to those used for compliance costs, EPA averaged sample
facility baseline loadings across all facilities in each stratum, multiplied
by the total number of facilities in each stratum, and summed.

      EPA calculated Scenario 1  (reflecting WQBEL #1) and Scenario 2
(reflecting WQBEL #2) loading reductions by finding the difference between the
existing permit limits (or highest reported concentration in the absence of an
existing permit limit) and the proposed Guidance-based limit for each
pollutant.  The resulting difference was converted to pounds per day by
multiplying the difference by the facility's flow rate and a unit conversion
factor.  Several assumptions were made to calculate the loading reduction for
a pollutant:

      i.  In the absence of an existing permit limit, if the difference
between the Guidance-based WQBEL and the highest reported concentration was
negative, zero reduction was assumed.  This situation occurred because there
were instances where the reported concentration was below the Guidance-based
WQBEL.

      ii.  When the highest reported concentration was reported as below a
detection level, one-half of the reported detection level was used as the
baseline concentration.  When the Guidance-based WQBEL was below analytical
detection levels, one-half of the detection level was used for purposes of
calculating a difference.

      iii.  It was assumed that facilities are discharging at the level of the
existing permit limitation.

      Loading reductions were calculated for each direct discharging facility
in the sample and then extrapolated to all direct discharging facilities in
the Great Lakes System.

      The pollutant loading reductions from direct dischargers were weighted
using EPA toxic weights (EPA/OST 1988 Cost Effectiveness Criteria and
Weights), which result in reductions expressed in toxic pounds-equivalent per
day.  Annual pollutant reduction values were then calculated by multiplying
the pounds-equivalent per day by 365 days under each scenario.  Total

-------
442    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

annualized costs for direct dischargers in each scenario were then divided by
total annual pounds-equivalent for each scenario to estimate the cost per
pounds-equivalent per year for each scenario.

b.    Results

      EPA estimated that the proposed Guidance would reduce the baseline
toxic-weighted pollutant loadings (126 million pounds-equivalent per year) by
about 30 percent.  The cost-effectiveness of all four cost scenarios ranged
from $1.30 per pounds-equivalent to $10.40 per pounds-equivalent over the
baseline; the cost-effectiveness of the most likely scenario was $2.60 per
pounds-equivalent (the estimated costs for indirect dischargers are not
included in these calculations).  When compared with the cost-effectiveness of
various effluent guidelines, these values were at the low end of the range of
$1 to $500 per pounds-equivalent.  However, as discussed below, the cost
estimates for the proposed Guidance were reevaluated using the revised
approach for estimating costs and load reductions for the final Guidance.  For
consistency, costs and load reductions based on the revised approach will be
used when comparing the proposal to the final Guidance.

C.    Revisions to -Projected Costs

      This section discusses EPA's approach to revising its original cost
estimates to comply with the final Guidance.  The revisions to the approach
are based primarily upon the changes to the proposal discussed throughout this
document, the preamble and rule for the final Guidance, and the public
comments received on the Regulatory Impact Analysis methodology used for the
proposal.  In addition, in making final decisions related to the Guidance, EPA
performed numerous analyses of different regulatory options for the following
issues identified at the time of the proposal: intake credits, use of Tier II
values, impact of analytical detection levels, elimination of mixing zones for
BCCs, and additivity.

1.    Summary of Revised Approach to Estimate Costs and Pollutant Load
      Reductions

      In general, EPA employed the basic methodology used to estimate
compliance costs and pollutant load reductions attributable to the proposal
 (as described in IX.B.I of this document).  However, EPA revised the approach
based on comments it received to more accurately project the costs to the
regulated community and to better account for the pollutant load reductions.
Several of the significant -revisions are described briefly below.  All of the
revisions made by EPA are documented in the  "Revised Assessment of Compliance
Costs Resulting From Implementation of the  Final Great Lakes Water Quality
Guidance," which is available for review in the public docket for the
rulemaking.

      EPA has also revised its assessment of compliance costs to reflect
modifications made to the final Guidance to provide increased flexibility for
State and Tribal implementation.  For example:

      -- Site-Specific Modifications: Great Lakes States and Tribes may  adopt
either more or less stringent modifications to human health, wildlife, and
aquatic life criteria based  on site-specific circumstances.  All criteria,
however, must be sufficient  not to cause  jeopardy to threatened or endangered
species listed or proposed to be listed under the Federal Endangered Species
Act.

      -- Intake Credits:  Great Lakes States and Tribes may consider the
presence of intake water pollutants in establishing water quality-based
effluent limits.

      -- Mixing Zones:  Great  Lakes States  and Tribes may authorize mixing
zones for existing discharges  of BCCs after a 12-year phase-out period,  if  the

-------
                           Section IX: Executive Order 12866                        443
permitting authority determines, among, other things, that the discharger has
reduced its discharge of the BCC for which a mixing zone is sought to  the
maximum extent possible.  Water conservation efforts that result  in overall
reductions of BCCs are also allowed even if they result in higher effluent
concentrations.

      For purposes'of estimating compliance costs, EPA assumed that permitting
authorities would use the more stringent provisions specified in  the final
Guidance even when the Guidance provides for less stringent alternatives.  For
example, this document and the preamble to the final Guidance suggests that
the permitting authority require bio-uptake studies for BCCs when WQBELs are
below analytical detection levels; however, the regulation to the final
Guidance contains rio such requirement.  For purposes of estimating compliance
costs, EPA conservatively assumed that bio-uptake studies would be required
for BCCs.  Table IX-1 describes several other conservative assumptions that
were made in estimating compliance costs for the final Guidance.

-------
444      Water Quality Guidance for the Great Lakes System — Supplementary Information Document
                                                        TABLE IX-1
           COST STUDY ASSUMPTIONS THAT ARE MORE STRINGENT THAN REQUIRED BY FINAL GUIDANCE
                  Final Guidance Requirement
         Description of Cost Estimate Assumptions
                                               WATER QUALITY CRITERIA
  Appendix B - Bioaccumulation Factors

  Human heath and wildlife criteria to be derived using measured
  or predicted BAFs
BAFs used to develop criteria for the costing analysis were
calculated using conservative assumptions. BAFs were therefore
more stringent than those that would be calculated by States and
should result in more stringent criteria than would be developed
by States. Projected costs, therefore, will likely overestimate
actual costs.
  Appendix C - Human Health

  Requires use of drinking water factors where discharge is to open
  waters, connecting channels, or designated drinking water
  sources.

  Uses IS grams/day fish consumption rate
Assumed all receiving waters were drinking water sources.  This
resulted in more stringent criteria. Therefore, projected costs
will likely overestimate actual costs.

Assumed a fish consumption rate of 45 grams/day. States will
most likely use IS grams/day; thus, the cost projections will
likely overestimate actual costs.
                                             IMPLEMENTATION PROCEDURES
  Procedure 1: Site Specific Modifications

  Allows States to use more or less stringent aquatic life criteria
  and BAFs
Cost analyses are based on Guidance recommendations. It is
most likely that, States will use the provision to relax criteria
and BAFs, thus, projected costs will likely overestimate actual
costs.
  Procedure 3: TMDL/WLA

  All mixing zones for BCCs eliminated within 12 years of rule
  publication.  Extensions may be granted for technical and
  economic considerations.
Assumed immediate elimination of all mixing zones for BCCs.
Costs were developed assuming BCC criteria were applied at
end-of-pipe.  Since States will likely allow mixing zones for
many existing sources up to 12 years. This would allow
permittees to defer some costs over the 12 year period.
Projected costs, therefore, will likely overestimate actual costs.
  Procedure 4: Additivity

  Requires States to adopt an additivity provision. States must
  address additivity by assuming an additive risk of IxlO4, an
  individual risk of 1x10*, or other defensible approach.
WQBELs were calculated assuming that all carcinogens were
additive. Assumed an additive risk of IxlO'5. This assumption is
likely to be at the stringent end of the options selected by States.
The projected costs, therefore, should slightly overestimate
actual costs.
  Procedure 6: Whole Effluent Toxkhy (WET)

  WET limits must be determined where reasonable potential is
  determined.  States must develop and apply WET limits, but may
  defer limit application until sufficient data have been generated.
Costed WET testing requirements for all facilities where WET
data were unavailable, and where toxic pollutants were present in
discharge. Rigorous WET testing requirements will likely
exceed State requirements; thus, projected costs will likely
overestimate actual costs.
  Procedure 8: WQBELs Below Detection

  Permits will specify the most sensitive analytical technology and
  will establish the 'Minimum Level of Quantification" (MLOQ).
Cost analyses used the method detection level (MDL) as the
target concentration. Since the MDL will be equal to, or more
stringent than the MLOQ, the projected costs will likely
overestimate the actual costs.
  Procedure 9: Compliance Schedules

  Final Guidance allows States to provide a three year compliance
  period upon permit reissuance. This could allow costs to be
  deferred for up to eight years (i.e., 5-yr permit cycle + 3-yr
  schedule).
Cost analyses did not consider compliance schedules. Costs
were assumed to be incurred immediately.  Since States will
likely provide compliance periods for many existing sources,
projected costs will likely overestimate actual costs.

-------
                           Section DC: Executive Order 12866                       445
a.    Revised Criteria and Implementation Procedures

      The costs to .comply with the final Guidance were estimated based on  the
revised implementation procedures and criteria.  Following are brief
descriptions of the specific procedures that were changed from the original
cost analysis.

i.    Dissolved Metals Water Quality Criteria

      As described in sections III, V, and VI of this document, EPA revised
many of the criteria originally proposed under the Guidance.  These revised
criteria were used as the basis for calculating WQBELs for the sample
facilities and then estimating compliance costs.

      In order to apply metals criteria in the dissolved form, EPA used
conversion factors, based on toxicity testing results, to revise criteria  from
the total form to the dissolved form.  The conversion factors were based on
the ratio of dissolved to total metals present in the laboratory toxicity
tests performed to establish the criteria for toxic metals.  Conversion
factors ranged from 0.333 for trivalent chromium, to 1.0 for trivalent
arsenic.  Where conversion factors were not available, EPA assumed the most
protective conversion factor of 1.0.  The criteria for the dissolved form  of
the metals were then adjusted back to the total form using the theoretical
partitioning relationship between the total/dissolved phases described in  the
EPA policy memo from Martha G. Prothro, dated October 1, 1993, to EPA's Water
Management and Environmental Services Division Directors in Regions I-X.

      The memo reiterates EPA's position that permit limits for metals must be
established for total metals and describes three methodologies for translating
dissolved metals criteria to the total form.  Two of the three methods rely on
site-specific studies performed to determine actual in-stream partitioning
relationships.  However, since actual metals partitioning data were not
available for any of the 59 study facilities, the third alternative, based on
the theoretical partitioning relationship,  was utilized to calculate
Guidance-based WQBELs.  The theoretical partitioning relationship is based on
a partitioning coefficient, determined empirically for each metal, and the
concentration of total suspended solids (TSS) in the receiving water.  Using
this relationship, EPA determined partitioning factors from dissolved to total
metals in the range of 2.0, for nickel, to 20.55, for trivalent chromium.
Where empirically determined partitioning coefficients were not available, a
partitioning factor of 2.0 was assumed.  Therefore, partitioning factors
assumed to be 2.0 would produce the most stringent possible permit limit.

ii.   Intake Credits

      In estimating the compliance costs for the sample facilities, EPA
applied the intake credit provisions to applicable facilities.  Consistent
with the proposal, intake credits were provided in one of two general ways.

      First,  EPA evaluated whether there would be a reasonable potential for
the discharge to cause or contribute to an excursion above a narrative or
numeric water quality criterion.  For purposes of estimating compliance cost
estimates,  EPA determined no reasonable potential and did not establish WQBELs
for outfalls that met the following criteria:

      -- The facility withdrew 100 percent of the intake water containing  the
pollutant from the same body of water into which the discharge was made.

      - - The facility did not contribute any additional mass of the identified
intake water pollutant to its wastewater.

      - - The facility did not alter the identified intake water pollutant
chemically or physically in a manner that would cause adverse water quality
impacts that would not occur if the pollutants were left in-stream.

-------
446    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      -- The facility did not increase.the identified intake water pollutant
concentration compared to the pollutant concentration in the intake water.

      -- The timing and location of the discharge would not cause adverse
water quality impacts to occur that would not occur if the identified intake
pollutants were left in-stream.

      It should be noted that when EPA did not possess intake pollutant data
for a sample facility, then EPA assumed that there would be a reasonable
potential to exceed WQBELs, and appropriate compliance costs were estimated
for the sample facility.  This assumption tends to overstate the costs, since
some of the facilities for which no data was available could qualify for an
intake credit.  There were also instances when some intake pollutant data was
available, but not all the data needed to determine whether all five of the
criteria described above would be met.  In general, for purposes of estimating
compliance costs, EPA assumed there would be no reasonable potential to exceed
water quality standards if at least the first two criteria described above
(i.e., withdrawing 100 percent from the same body of water, and no additional
pollutant was added to the discharge) were met.  This assumption could
potentially underestimate the cost impact of the intake credit provisions
contained in the final Guidance.

      Second, EPA granted intake credits for sample facilities when the level
of the pollutant upstream of the discharge exceeded the most stringent
applicable water quality criterion for that pollutant.  When this situation
occurred, relief was provided by making the WQBEL for the pollutant(s) equal
to the most stringent Guidance criterion, instead of prohibiting the discharge
or making another more stringent assumption.  EPA did this for both discharges
to different and same bodies of water.  The final Guidance allows "no net
increase" (i.e., discharge at background concentrations) for up to 10 years
following promulgation of procedures by the State or Tribal agency, or until a
total maximum daily load (TMDL) is established, for discharges to the same
body of water.  EPA conservatively assumed in its cost estimates that TMDLs
justifying loads greater than criteria would not be developed and dischargers
to the same body of water would need to comply with the most stringent
criteria at end-of-pipe.

iii.  Additivitv/TEFs

      EPA's estimate of costs for the sample facilities accounted for
additivity of human carcinogenic effects of pollutants contained in a
discharge.  To estimate costs for the final Guidance, EPA assumed that  the
total carcinogenic risk of the mixture of two or more carcinogens in a
discharge would not exceed a lifetime incremental cancer risk equal to  one in
100,000  (10~5) .  The final Guidance allows States to use a less stringent
incremental cancer risk equal to one in 10,000  (10"4) for additivity, but EPA
decided to use a 10*: risk level for the mixture for estimating costs because
some States may choose to use a 10"5 risk level.  In addition, the final
Guidance allows a State and Tribe to account for additivity by establishing
individual human carcinogen doses at levels corresponding to an  incremental
cancer risk of one in 1,000,000  (10"*), or applying a scientifically defensible
method to account for the additive effects of carcinogens.

      The first step in estimating the cost attributable to additivity  was to
determine the number of potential carcinogens discharged by a study facility,
the concentration of those pollutants in the discharge, and the  background
concentrations in the ambient water for those pollutants.  The second step was
to determine the human cancer value  (HCV) associated with a lifetime
incremental cancer risk equal to one in 100,000 for those individual
pollutants identified in the discharge.  The third  step was to divide each of
the HCVs for those carcinogens identified in the discharge by the total number
of carcinogens in the discharge to determine the allowable concentration  for
each carcinogen that could be discharged.  This concentration was then
compared to the actual concentration in the discharge for that pollutant  to

-------
                           Section DC: Executive Order 12866                       447
 determine whether the facility needed to reduce that pollutant.  This approach
 results in an equivalent  (proportional) reduction for each HCV at  a  sample
 facility.  This approach was  selected to provide a consistent method for
 addressing additivity to all  sample facilities.

      EPA also considered use of toxicity equivalent factors  (TEFs)  when
 establishing wasteload allocations for both human health non-cancer  and cancer
 criteria for compounds similar to 2,3,7,8-tetrachlorodibenzo-p-dioxin
 (2,3,7,8-TCDD) in accordance  with procedure 4 of appendix F of the final rule.
 It should be noted, however,  that for those sample facilities for  which
 2,3,7,8-TCDD WQBELs were established, no concentration data existed  for the
 chlorinated dibenzo-p-dioxins (CDDs) and chlorinated dibenzofurans (CDFs) in
 an effluent.  Thus'equivalent concentrations for 2,3,7,8-TCDD-type chemicals
 were not developed.

 iv.   Acute Mixing Zones for  WET

      As discussed in section VTII.F of the document, the proposed
 implementation procedures originally required that facilities comply with an
 acute WET criterion of 1.0 acute toxic unit (TU.)  at  the  end-of-pipe  (i.e.,  no
 mixing zone allowed).  The final Guidance requires that no discharges exceed
 0.3 TU. at  the  edge  of  an approved acute mixing  zone.  As a result, for
 purposes of estimating costs, EPA allowed mixing zones to comply with acute
 WET criteria.  EPA used the WLA equations provided in procedure 3  of appendix
 F of the proposed Guidance and the 1-year, 10-day  (1Q10) critical  receiving
 water flow as recommended in  procedure 3 of the final Guidance to  calculate
 acute WET limits.

 b.    Criteria for Tier I and Tier II Pollutants

      The proposed Guidance,  while generally applying to all pollutants, was
 structured to provide an initial focus on 138 pollutants.  The 138 pollutants
 were identified as those being known or suspected of being of primary concern
 in the Great Lakes'Basin.  The proposed Guidance included numeric  criteria to
protect aquatic life, human health, and/or wildlife for 32 of the  138
pollutants.  The cost study for the proposed Guidance was based on these 32
pollutants.  Because of concern that the 32 pollutants did not represent all
 the possible pollutants that  may contribute to potential costs, EPA  evaluated
whether additional pollutants should be included in the cost analysis.  The
 evaluation used three criteria to determining whether additional pollutants
 should be included in the final analysis:  loadings, frequency of  occurrence,
 and toxicity.

      To determine .which pollutants exhibited significant loadings to the
Great Lakes Basin, the loadings for all 138 pollutants of initial  focus listed
 in the proposed Guidance at the 59 study facilities were calculated.  The
 loadings were based on facility permit limits or measured effluent
 concentrations.  The loadings were then multiplied by EPA toxic weights to
normalize the toxicity of each pollutant to that of copper.  Using the
 statistical extrapolation factors developed for the costing analysis, the
total toxic weighted loadings for the 138 pollutants were extrapolated to the
universe of major dischargers in the Great Lakes Basin.  Based on  the results
of this evaluation, EPA determined that pollutants that exhibited  "de minimis"
 loadings would be omitted from the final costing analysis.  The "de minimis"
value selected was '10 pounds-copper toxicity equivalents per day.  This value
corresponds to a total pollutant load from all major point source  dischargers
to the Great Lakes Basin of 10 pounds of copper per day.

      In addition to the loadings analysis, EPA wanted to ensure that other
pollutants that were frequently limited or required to be monitored  at
facilities,  but that might have been undetected or that exhibited  low toxicity
and thus were not captured in the loadings analysis, were also included in the
final costing analyses.  Since the loadings analyses should have captured the
most significant pollutants of concern, this evaluation was considered a

-------
448    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

"safety,net."  This analysis  captured any pollutant  that  was limited,
detected, or required to be monitored at three  or more  of the 59  sample
facilities.

      As a final "safety net" EPA ensured that  any pollutant limited,
detected, or required to be monitored at any facility that exhibited a hjLgh
toxici-ty (high toxic weight)  was included in the final  costing.   This
evaluation was performed by multiplying the  "frequency  of occurrence"  for a
given pollutant by its toxic  weight.  The resulting  value1 was designated the
"Occurrence Toxic Equivalent"  (OTE).  The OTE analysis  captured those
pollutants that might have not been detected, and so escaped the  loadings
evaluation, but that had monitoring requirements at  one or more facilities.  A
target value of 0.1 OTE was selected  to ensure  that  any pollutant with a toxic
weight of 50 or greater, and  even a single monitoring requirement at one
sample facility, would be included in the final costing analysis.

      The additional pollutant evaluation identified 76 pollutants that were
limited, detected in the effluent, or required  to be monitored at one or more
of the 59 sample facilities.  From this list of 76 pollutants,  37 were
determined to be of consequence to the loadings and  costing analyses using the
rationale described above.  This increased the  total number of pollutants
evaluated for loadings and costs in the final analysis  to 69.  The list of
pollutants included in the final analysis and those  found, but not included in
the costing analyses, are provided in the following  table.

-------
                               Section IX: Executive Order 12866
                                                                  449
                PARAMETERS IDENTIFIED FOR FINAL  COSTING ANALYSIS
PARAMETERS IN
PROPOSED ANALYSIS

2,3,7,8-TCDD
2,4-Dimethylphenol
2,4-Dinitrophenol
Arsenic(ni)
Benzene
Cadmium
Chlordane
Chlorobenzene
Chromium(ni)
Chromium(VI)
Copper
Cyanide, free
Cyanide, total
DDT
Dieldrin
Endrin
Heptachlor
Hexachlorobenzene
Hexachloroethane
Lindane
Mercury
Methylene Chloride
Nickel
Parathion
PCBs
Pentachlorophenol
Phenol
Toluene
Total Selenium
Toxaphene
Trichloroethylene
Zinc
PARAMETERS ADDED
FOR FINAL ANALYSIS

1,1-Dichloroethane
1,1-Dichloroethylene
1,1,1 -Trichloroethane
1,2-Dichloroethane
1,2-Dichloropropane
1,2-trans-Dichloroethylene
1,2,4,5-Tetrachlorobenzene
2,4,6-Trichlorophenol
3,3-Dichlorobenzidine
4,4-DDD
4,4-DDE
Acrylonitrile
Aldrin
alpha-Endosulfan
alpha-Hexachlorocyclohexane
Aluminum
Antimony
Benzidine
Benzo[a]pyrene
Beryllium
beta-Endosulfan
beta-Hexachlorocyclohexane
Carbon tetrachloride
Chloroform
Chlorpyrifos
Chrysene
Endosulfan
Fluoranthene
Fluoride
Hexachlorocyclohexane
Iron
Lead
Pentachlorobenzene
Phenanthrene
Silver
Tetrachloroethylene
Thallium
PARAMETERS FOUND
BUT NOT ANALYZED

2-Nitrophenol
4-Nitrophenol
1,2-Dichlorobenzene
1,3 -Dichlorobenzene
1,4-Dichlorobenzene
3,4-Benzofluoranthene
11,12-Benzofluoranthene
1,1,2-Trichloroethane
1,2,4-Trichlorobenzene
1,1,2,2-Tetrachloroethane
1,2,3,4-Tetrachlorobenzene
2-Chlorophenol
2,4-Dichlorophenol
2,4-Dichlorophenoxyacetic acid
Acenaphthene
Acenaphthalene
Acrolein
Anthracene
Bis(2-chloroethyl)ether
Bis(2-chloroisopropyl)ether
Bromoform
Butylbenzylphthalate
Chlorodibromomethane
Chloroethane
Dichlorobromomethane
Diethyl phthalate
Di-n-butylphthalate
Dimethylphthalate
Ethylbenzene
Fluorine
Hexachlorobutadiene
Indeno[ 1,2,3 cd]pyrene
Isophorone
Methyl bromide
Methylchloride
Naphthalene
Octachlorostyrene
Pyrene
Vinyl Chloride

-------
450    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      An example of a pollutant found at study facilities, but excluded from
the final analyses, is vinyl chloride.  Vinyl chloride was limited in the
permit for one facility and was detected in the effluent at a second facility.
Based on the permit limit- and the monitoring data, a total load for vinyl
chloride of 0.04 pounds per day was determined.  Using the vinyl chloride
toxic weight of 0.0013 and extrapolating the load to the 588 major discharges
in the Great Lakes Basin, a total toxic weighted load of 0..0018 pounds-
equivalent per day was estimated, which is below the "de minimis" criteria of
10 pounds-equivalent per day.  The occurrence frequency for vinyl chloride  (2
of 50 facilities or 4 percent), did not exceed the trigger of 5 percent.  The
OTE was then calculated by multiplying the occurrence frequency of 4 percent
by the toxic weight (0.0013).  This resulted in an OTE of 0.00005, which is
less than the 0.1 OTE trigger.  As a result of this analysis, vinyl chloride
was not considered in the final costing analyses.

      Having established the list of pollutants for the final costing
analyses, EPA developed criteria for these pollutants utilizing the Tier I and
Tier II procedures outlined in Appendices A, B, C, and D of the final Guidance
and readily available data.

c.    Data for the Sample Facilities

      The original analysis was performed based on data collected from EPA
Region 5, State permitting authorities, EPA development documents, and special
studies.  Discharge data were based on 1990 Permit Compliance System  (PCS)
data, and facility-specific permit file information were generally from 1992.
For this final rule, EPA ensured that the current information and data
(including permits, fact sheets, permit applications, and other relevant
discharge information) were used as the basis for comparison to Guidance
requirements.  In addition, State permitting authorities were requested to
review each sample facility evaluated in the original cost estimate, and to
provide comments and additional information as necessary to ensure accurate
reflection of current permit requirements and discharge conditions.

      For the cost estimate for the final Guidance, EPA specifically used 1993
PCS discharge data, as well as permit file  information and data provided by
the State permitting authorities  (generally representing permits issued as
late as through mid-1994).  As a result of use of more recent data, EPA noted
a shift in the baseline of permit requirements for the sample facilities; the
baseline was lowered based on more stringent NPDES permit requirements being
applied by permitting authorities.  As will be discussed further in section
IX.C.2 of this document, the overall effect of lowering the baseline  is that
estimated compliance costs and pollutant load reductions will not be  as
substantial as was originally projected for the proposed Guidance.

      One of the limitations of the original compliance cost study was a
general lack of  site-specific receiving water data  (i.e., background  data)  for
the sample facilities.  To fully evaluate EPA's provisions in the final
Guidance for intake credits  (i.e., determining whether discharges are to  same
or different bodies of water and for identifying non-attainment waters),  as
well as to ensure tjhat all available data were used for the cost analysis,  EPA
gathered additional background concentration data for each of the sample
facilities.  EPA also reviewed and considered data submitted as a part of the
public comments, as well as  examined the water quality files contained  in the
STORET data base.  In addition, EPA worked  closely with the State permitting
authorities to collect all applicable data.

      Consistent with procedure 3 of appendix F  to the final Guidance,  EPA
also collected fish tissue data  (either caged or resident fish  tissue data)  to
represent ambient water  column background concentrations.  When fish  tissue
data were available.for  the  pollutants being evaluated at a  sample  facility,
EPA used a  simplified approach for converting the tissue  data to  ambient  water
column concentrations.   This method entailed dividing fish tissue data  (in
mg/kg wet weight) by the pollutant-specific bioaccumulation  factor  (BAF)  used

-------
                           Section IX: Executive Order 12866                       451
 to derive Tier  I  criteria  (in  I/kg) and multiplying  the result by  1,000  to
 give the result as  concentration of pollutant  (/xg/1) .  When data for more than
 one species was available, the geometric mean  for all  species was  calculated
 and used.

 d.    Compliance  Cost Decision Matrix

      In deriving the cost estimate for the proposed Guidance, EPA assumed
 that when treatment costs became excessive in  light  of the amount  of pollutant
 to be removed,  or if information regarding the existing treatment  system was
 lacking, that waste minimization/pollution prevention  techniques would be the
 preferred control approach selected by the regulated community.  In addition,
 for the proposed  Guidance, EPA also did not consider the alternatives
 available to facilities through regulatory relief mechanisms such  as
 variances, mixing zone studies, phased-TMDLs,  site-specific criteria, etc.

      The Guidance, consistent with the CWA and NPDES  program, does not  direct
 facilities on how t;o comply with permit requirements.  Therefore,  each
 regulated facility  can consider a variety of options to comply with permit
 requirements.   In estimating compliance costs, EPA selected control options
 for the sample  facilities.  However, in an effort to ensure consistency  in
 estimating the  general types of controls that  would  be necessary for a sample
 facility to comply with the final Guidance (assuming that the Guidance
 resulted in more  stringent requirements), as well as to integrate  into the
 cost analysis the other alternatives available through the final Guidance, EPA
 developed a costing decision matrix that was used for  each sample  facility.
 The underlying  assumption of the decision matrix is  that a facility will
 examine least-cost alternatives prior to incurring the expense and potential
 liabilities associated with constructing end-of-pipe treatment.

      Under the decision matrix, costs for minor treatment plant operation and
 facility changes  were considered first.  Where it was  not technically feasible
 to simply adjust  existing operations, waste minimization/pollution prevention
 controls were considered; however, these controls were costed only where they
 were considered feasible based on EPA's understanding  of the process(es) at a
 facility.  In general, detailed treatment and  manufacturing process
 information is  not available in NPDES permit files;  therefore EPA's assessment
 of feasibility  was primarily based upon best professional judgement using
 general knowledge of industrial and municipal  operations.

      If waste minimization was deemed not feasible  to reduce pollutant  levels
 to those needed to comply with the final Guidance criteria, a combination of
 waste minimization/pollution prevention and simple treatment was considered.
 If these relatively low-cost controls could not achieve the Guidance-based
 WQBELs,  then finally end-of-pipe treatment was considered.

      However,  before assuming that treatment  would  be installed by the
 facility, EPA first considered the relationship between the cost of adding the
 treatment and other types of remedies or controls.   If EPA concluded that
 other remedies or controls would be more feasible than installing  end-of-pipe
 treatment, EPA assumed that a  facility would alternatively pursue  some type of
 regulatory relief from the WQBEL.  If the estimated  annualized cost for
 removal of a pollutant exceeded $200 per toxic pounds-equivalent then EPA
 assumed that dischargers would explore the use of other remedies or controls.
This cost trigger was based on the upper end of the  range of the costs to
 comply with promulgated effluent guideline limitations and standards for
 direct discharger industrial categories.  When EPA assumed that facilities
would pursue alternative relief, no treatment  cost was estimated for a
 facility; however, a nominal cost for some efforts to  reduce the pollutant
until the relief  is granted was included.  In  addition, EPA did not take
 credit for any load reduction  for any pollutant for  which alternative relief
was assumed.   Finally, based upon discussions  with EPA Regional and State
permitting agencies and outside experts, EPA estimated the typical cost  to
 facilities pursuing relief from Guidance-based WQBELs.  These costs will be in

-------
452    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

the form of additional monitoring,  performing special studies, etc., to
support facilities' requests for relief from the Guidance-based WQBEL.  The
costs estimated by the Regions and States for the relief mechanisms ranged
from a high of $1,000,000 per pollutant for phased-TMDLs to a low of $20,000
for criteria modifications.  For purposes of estimating compliance costs, EPA
used a mid-range cost value of $200,000 per pollutant for each time a relief
mechanism was assumed necessary.

      In developing and using the cost decision matrix, EPA acknowledges that
granting relief from WQBELs is dependent upon the specific circumstances at a
facility, as well as the judgement and implementing procedures of State and
Tribal permitting authorities.  EPA also acknowledges that opportunities for
waste minimization are dependent upon the specific circumstances at a
facility.  The use of a $200 per toxic pounds-equivalent trigger for a
"facility" assumes that the regulatory flexibility in the Guidance would be
available and granted to all facilities that exceed the cost trigger.  Section
IX.D.I of this document provides further discussion regarding the use of
regulatory relief for purposes of estimating compliance costs for the final
Guidance.

      Acknowledging that the use of regulatory relief may be limited depending
upon the particular circumstances for a "facility," EPA also estimated costs
under a higher cost scenario that assumes regulatory relief would be granted
only when the cost for the particular "category of dischargers" exceeds a cost
trigger.  Particularly, if the estimated annualized cost for a "category of
dischargers" exceeded $500 per toxic pounds-equivalent then EPA assumed that
dischargers within the "category" would be granted regulatory relief.  This
cost trigger was based on the highest costs to comply with promulgated
effluent guideline limitations and standards for direct discharger industrial
categories, which ranged from just over $1 to over $500 per toxic pounds-
equivalent per industrial category.

e.    Pollution Prevention/Waste Minimization Costs

      As discussed briefly in the section above, EPA used waste
minimization/pollution prevention techniques as controls for a number of
sample facilities.  The costs associated with the implementation of these
techniques were originally based upon a limited evaluation of information
available through the EPA Pollution Prevention Clearinghouse.  In the absence
of information for a particular category of dischargers, EPA used best
professional judgement to estimate the cost to implement pollution prevention.

      Since the time of proposal, EPA has attempted to collect additional
information to verify or replace its original estimates of pollution
prevention/waste minimization costs.  EPA particularly solicited input from
the EPA  Pollution Prevention Office and the American Institute of Pollution
Prevention.  Both of these organizations acknowledged the difficulty  in
developing generic costs because of the site-specific nature of manufacturing
processes and pollutants being removed.  In fact, the implementation  of waste
minimization/pollution prevention techniques may actually result in a cost
saving for a facility.  Because of the general lack of information related to
the cost of pollution prevention techniques, EPA retained its original
estimates for waste minimization/pollution  prevention.

f.    Indirect Dischargers

      As discussed in section IX.B.I.a of this document, EPA's basic  approach
to estimating indirect discharger costs was based on an analysis of one major,
highly industrialized, sample POTW.  Based  on this evaluation, it was assumed
that the number of -indirect  dischargers that could be affected ranged from 10
to 30 percent.  To further verify this range, EPA analyzed information for an
additional eight POTWs based on data collected from the Michigan DNR  and
Wisconsin DNR, as well as  reevaluated the original sample POTW based  on
changes  to the final Guidance  (as reflected in estimated WQBELs for the POTW).

-------
                           Section DC: Executive Order 12866                       453
      Since not all of the eight POTWs,were  selected as a  sample  facility and
analyzed under the study, EPA assumed  for the purpose of this  analysis,  that
the pollutants limited by each  POTW's  existing NPDES permit would be  the same
as those that would require regulation under the Guidance  (i.e.,  the  Guidance
would not result  in additional  pollutants being regulated, but would  result in
more stringent permit limits).  Based  on the results provided  in  the  EPA Draft
Analytical Survey 'of Nine POTWs from the Great Lakes Basin (December  15,'
1994), this assumption was considered  reasonable in light  of the  limited
detection of pollutants, particularly  BCCs.  It should also be noted  that
information for three of the additional POTWs was not sufficient  for  EPA to
determine the number of industries potentially affected.

      For each POTW, the potential indirect  dischargers of each regulated
pollutant were identified from  among the POTW's list of indirect  dischargers,
as well as the number of industrial  users found to be violating the POTW's
permit limits for any of the pollutants of concern over a  1-year  period.
Based on these data, the range  of potentially affected indirect users is
estimated to be 8 to 44 percent of the total number of the indirect
dischargers to a  POTW.  The results  show that EPA's assumed range of  indirect
dischargers affected (10 to 30  percent) had  a reasonable basis.

      For purposes of developing costs for the final Guidance, EPA assumed
that 10 to 30 percent of all indirect  dischargers in the Great Lakes  Basin
would be impacted by source control  efforts  by POTWs as a  result  of more
restrictive Guidance-based WQBELs.   EPA also updated the average  compliance
cost per direct discharger facility, based on the revisions made  to the  sample
facilities as a result of the final  Guidance, and used this updated compliance
cost to estimate  costs for indirect  dischargers.

g.    Revised Toxic Weights

      EPA used toxic weights to derive cost-effectiveness  estimates for  the
proposed Guidance, as well as to compare the relative loadings of the 138
pollutants of concern analyzed  for the cost  study (as discussed in section
IX.B.2.a).  Toxic weights are used by  EPA as normalizing factors  that relate
the toxicity of any pollutant to the toxicity of copper.   The  factor  considers
the aquatic toxicity and the human health effects of a pollutant  and  is
calculated using  the following  formula:

Toxic Weight =  5.6/[fresh water chronic criteria (/xg/l)]  +
   5.6/[human health criteria (/zg/1)]

The value of 5.6  /ig/1 was the original national chronic water  quality
criterion for copper; thus, the toxic  weight for copper was equal to  one.  A
pollutant with a  toxic weight of 10, therefore, would be considered 10 times
as toxic as copper.

      The national chronic water quality criterion for copper  has since  been
revised to 12 /ig/1; however, the 5.6 value is retained by  EPA  for consistency.
This results in copper currently having a toxic weight of  5.6/12,  or  0.47.

      EPA used toxic weights from 1988 in calculating baseline pollutant loads
and load reductions for the proposed Guidance.  These loads and load
reductions are expressed in toxic pounds-equivalent.  In analyzing the impact
of the final Guidance,  toxic weights were developed or recalculated for  all 69
pollutants included in the cost study, using the most recent criteria and
toxicity information available  to EPA.  These updates resulted in both raising
and lowering of the 1988 toxic  weights, depending on the toxicity data
available for a specific pollutant.

2.    Results

      Based upon  the adjustments made  to the cost study approach  as described
in section IX.C.I above, EPA revised its estimate for the  compliance  cost of

-------
454    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

the Guidance, as well as its estimate of pollutant reductions attributable to
the final Guidance.  This section presents these results and discusses them in
relation to EPA's estimates for the proposed Guidance.

a.    Estimated Compliance Costs

      As shown in Table IX-2, the total annualized costs of .implementing the
final Guidance to direct and indirect dischargers is estimated to range from
$61 million dollars to $376 million dollars.  As shown in Table IX-2, under
the low-end estimate, direct dischargers account for 67 percent of the total
estimated compliance cost and indirect dischargers are estimated to incur 33
percent of the total cost.  Under the high-end estimate, direct dischargers
account for 98 percent of the total estimated cost, and indirect dischargers
account for 2 percent.  This shift in proportion of costs between direct and
indirect dischargers between the high and the low estimates is due to the
increased use of end-of-pipe treatment for direct dischargers under the high-
end estimates.  In addition, it was assumed that a smaller proportion of
indirect dischargers  (10 percent) would be impacted under the high-end
estimate, since municipalities are adding end-of-pipe treatment which should
reduce the need for source controls (i.e., reduce the need for increased
pretreatment program efforts).

i.    EPA's Low-End Estimate

      Under the low-end estimate for the direct dischargers, municipal majors
are expected to incur 39 percent of total costs and industrial majors account
for 26 percent of total costs.  Minor direct dischargers are estimated to
incur just under 3 percent of the total costs.  The two major industrial
categories with the largest total annualized cost are the pulp and paper  (14
percent of total) and miscellaneous (8 percent) categories.  The food and food
products and metal finishing categories are estimated to incur less than 1
percent of the total annualized cost.

      Although the municipal major category accounts for almost 40 percent of
the total estimated cost, the average annual cost is just over $75,000 per
facility.  Average annualized costs for industrial majors vary widely across
categories, with the highest average cost estimated for the miscellaneous
 ($168,000 per plant) and pulp and paper  ($151,000 per plant) categories.  For
minor facilities, average costs are negligible at an estimated $500 per
facility.

      Costs to direct dischargers for developing and  implementing pollutant
minimization plans  (required when WQBELs are below detection levels) account
for most of the costs  (39 percent of  total  annual costs).  Annualized capital
and operation and maintenance costs make up  just over 14 percent of total
annual costs; waste minimization  costs account for almost 8 percent.

      Common to both the reevaluated proposal and the final Guidance, the
lowering of the permit baseline  (discussed  previously in section IX.C.l.c of
this document) also accounts for  an overall  decrease  in compliance costs and
load reduction.  The  lowering of  the permit baseline  was expected due to  State
implementation of the requirements of section 303(c)  of the CWA, which
required all States to promulgate water quality criteria for certain toxic
pollutants.  To ensure  that the  requirements of section 303(c) are met, EPA
promulgated the National Toxics  Rule  (57 FR 6084; 12/22/92) to provide water
quality criteria for pollutants  for which States did  not promulgate  criteria.
More important, each  of the Great Lake States has been actively  involved  in
the Great Lakes Water Quality Initiative since 1989,  acting as co-partners  and
major participants  in developing the  Guidance.

      By pollutant,  controls for mercury account for  over  20 percent of annual
costs  (attributable primarily to pollutant  minimization plan-related costs).
Other pollutants that account for significant  costs include methylene
chloride, aluminum, benzene, and copper.

-------
                                    Section IX: Executive Order 12866
455
                                            TABLE LX-2
                                SUMMARY OF ANNUALIZED COSTS
COST CATEGORIES
Major Direct Dischargers— Industrial
Major Direct Dischargers— Municipal
Minor Direct Dischargers
Indirect Dischargers
TOTAL
NUMBER OF
FACILITIES
272
316
3,207
3,528
7,323
LOW-END
ESTIMATED COSTS
(First quarter 1994 $,
Millions)
15.7
23.8
1.6
19.9
61.0
HIGH-END
ESTIMATED COSTS
(First quarter 1994 $,
Millions)
108.2
259.8
1.6
6.6
376.2
Source:  Revised Assessment of Compliance Costs Resulting from Implementation of the Final Great Lakes Water Quality Guidance.

-------
456    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

ii.   EPA's High-End Estimate

      Under the high-end estimate for the direct dischargers, municipal majors
are expected to incur just under 70 percent of total costs and industrial
majors account for 29 percent of total costs.  Minor direct dischargers are
estimated to incur less than 1 percent of the total costs.  The  two major
industrial categories with the largest total annualized cost are the pulp and
paper (23 percent of total) and miscellaneous (3 percent) categories.  Even
under the high-end, the food and food products and metal finishing categories
are estimated to incur less than 1 percent of the total annualized cost.

      The municipal major category accounts for almost 70 percent of the total
estimated cost, the average annual cost is just over $822,000 per facility.
Average annualized costs for industrial majors vary widely across categories,
with the highest average cost estimated for pulp and paper ($1,583,000 per
plant) and miscellaneous ($433,700 per plant) categories.  For minor
facilities, average costs are negligible at an estimated $500 per facility.

      For the high-end scenario, costs to direct dischargers shifted away from
developing and implementing pollutant minimization plans and waste
minimization to capital, operating and maintenance costs  (over 52 percent of
total annual costs) associated with construction and application of end-of-
pipe treatment.  Annualized costs for developing and implementing pollutant
minimization plans'make up just over 6 percent of total annual costs; waste
minimization costs account for less than 1 percent.

      By pollutant, controls for lead account for over 60 percent of annual
costs (attributable primarily to end-of-pipe treatment related costs).  Other
pollutants that account for significant costs include heptachlor,
pentachlorophenol, lindane, and mercury.

iii.  Comparison of Estimated Costs for the Final Guidance to Costs of
      Proposed Guidance

      Table IX-3 provides a comparison of EPA's cost estimate for the final
Guidance to costs of the original proposal.  The original proposal estimates
were revised to reflect changes made in the approach to estimating costs for
the final Guidance.  As shown in Table IX-3, reevaluation of the original
proposal resulted in an increase in costs of about $240 million  under the low-
end scenario and $265 million for the high-end when compared to  the final
Guidance.

      As shown, EPA's annual cost estimate for the final Guidance is
significantly lower than the revised estimates for the proposed  Guidance.  EPA
attributes some of 'this reduction to the final Guidance  intake credit
provisions, which provide relief to several significant  dischargers that
discharge to non-attained waters, and to the use of dissolved metals  criteria,
which also tends to lower the costs for the final Guidance.

      Consequently, most of the Guidance and current State water quality
standards have a wide number of similarities.  In fact,  some States have
already elected to promulgate more stringent requirements for a  variety of
Guidance-related provisions in anticipation of the Guidance.  For example, New
York, Minnesota, Illinois, and Wisconsin all use a higher fish consumption
rate than required ,by the final Guidance for deriving human health  criteria.
Many of the Great Lakes States also currently have provisions in their water
quality programs to account for additivity  of risk from  carcinogens.  As  a
result of these and many other efforts by  States, the stringency of NPDES
permit requirements continues to  increase,  which decreases the incremental
difference between the  current State permit  limits and Guidance-based WQBELs.

-------
                                      Section DC: Executive Order 12866
457
                                                 TABLE IX-3

              COMPARISON OF PROPOSED AND FINAL COMPLIANCE COST ESTIMATES
COST CATEGORY
Major Direct
Dischargers— Industrial
Major Direct
Dischargers— Municipal
Minor Direct
Dischargers
Indirect Dischargers
TOTAL COST
ANNUAL COMPLIANCE COST ESTIMATE ($ Millions)
ORIGINAL PROPOSAL
(LOW ESTIMATE)1
(1st Quarter 1994)
20.7
259.9
1.6
19.9
302.1
ORIGINAL PROPOSAL
(HIGH ESTIMATE)2
(lot Quarter 1994)
376.5
259.9
1.6
6.6
644.6
FINAL GUIDANCE
(LOW ESTIMATE)'
(1st Quarter 1994)
15.7
23.8
1.6
19.9
61.0
FINAL GUIDANCE
(HIGH ESTIMATE)4
(1st Quarter 1994)
108.2
259.8
1.6
6.6
376.2
1 The proposed Guidance cost estimate revised to reflect the low-end cost assumptions made for the Regulatory Impact Analysis for the final
Guidance.
2 The proposed Guidance cost estimate revised to reflect the high-end cost assumptions made for the Regulatory Impact Analysis for the
final Guidance.
9 Estimated costs to comply with the final Guidance using low-end assumptions.
4 Estimated costs to comply with the final Guidance, but assuming limited regulatory relief is available.

-------
458    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

b.    Estimated Pollutant Reductions  ..

      Table IX-4 presents the toxicity-weighted pollutant baseline and
reductions estimated for .the final Guidance.  As shown in Table IX-4, the
baseline pollutant loadings are projected to be approximately 36 million toxic
pounds-equivalent per year  (Ibs-eq/year).  This baseline pollutant loading
represents a 71 pefecent reduction in the baseline projected by EPA for its
original analysis of the proposed Guidance  (126 million Ibs-eq/year, April 16,
1993) .

      This downward shift in the baseline pollutant loadings is particularly
significant in light of the fact that over 35 more pollutants were added for
the analysis of the final Guidance.  As discussed above, EPA attributes the
shift to the fact that the existing permit baseline also moved downward  (i.e.,
existing permit limits for the sample facilities were found to be more
stringent).  This downward shift in the permit baseline is due, in part, to
increased efforts by States to protect water quality.  EPA also attributes the
difference to the Use of dissolved criteria for metals for the final Guidance,
which tended to eliminate metals from the cost and loading analyses.  As
discussed in section IX.C.l.d of this document, EPA did not take load
reduction credit for pollutants for which regulatory relief was assumed
necessary to comply with Guidance-based WQBELs.

      Upon implementation of the final Guidance, EPA estimates that pollutant
loadings under the low-end estimate would be reduced by 5.8 million Ibs-
eq/year, which represents a 16 percent reduction of the baseline pollutant
loadings.  Under the high-end cost estimate, pollutant loading reductions
would increase by \.8 million Ibs-eq/year to a total of 7.6 million Ibs-
eq/year, which represents a 22 percent reduction of the baseline pollutant
loadings.

      The percent reductions estimated for the final Guidance are also lower
than projected for the proposed Guidance reevaluated using the revised
approach for estimating costs and load reductions.  Pollutant loadings under
the proposed Guidance would be 8.4 million Ibs-eq/year  (24 percent reduction)
and 10.1 million Ibs-eq/year  (29 percent reduction) for the low- and high-end
scenarios, respectively.  The drop in estimated pollutant loadings can also be
credited to the changes made by EPA to the criteria for the final Guidance
 (e.g., adjusting bioaccumulation factors, use of dissolved criteria for
metals) and the toxic weights.  The combined result of these changes was
essentially less stringent  criteria, which would tend to reduce the difference
between existing permit limits and the Guidance-based WQBELs.

      Under the low-end estimate  for the final Guidance, the largest pollutant
load reductions occur for dieldrin and  lead, which account for over  50 percent
of the toxic weighted load  reduction.  Chlordane, heptachlor, and
pentachlorobenzene were also reduced by  significant amounts from the baseline.
Under the high-end estimate, the  largest pollutant load reductions occur for
heptachlor, dieldrin, and lead which account for about 70 percent of the  toxic
weighted  load reduction.

-------
            Section IX: Executive Order 12866
459
                    TABLE LX-4




POUNDS-EQUIVALENT POLLUTANT LOADING REDUCTIONS
POLLUTANT
Acrylonitrile
Aldrin
Aluminum
Antimony
Arsenic(ni)
Benzene
Benzidine
Benzo[a]pyrene
Beryllium
Cadmium
Carbon tetrachloride
Chlordane
Chlorobenzene
Chloroform
Chlorpyrifos
Chromium(ni)
Chromium(VI)
Chrysene
Copper
Cyanide, free
Cyanide, tout
4,4-DDD
4,4-DDE
DDT
3,3-Dichlorobenzidine >
1 , 1 -Dichloroethane
1 ,2-Dichloroethane
1 , 1 -Dichloroethy lene
1 ,2-trans-Dichloroethy lene
1 ,2-Dichloropropane
POLLUTANT LOADING (lbs-eq/year)
BASELINE
	 *
—
2,379,890
—
21,975
164
—
—
—
344,827
648
975,523
—
129
—
—
—
—
2,402
95,940
—
45
21
88,152
110,466
—
29
62
—
5
REDUCTION
(LOW ESTIMATE)
—
—
25,419
—
21,556
1
—
—
—
0
527
664,604
—
7
—
—
—
—
744
10,623
—
23
10
212
68,619
—
19
0
—
5
REDUCTION
(HIGH ESTIMATE)
-
-
25,419
-
21,556
1
-
-
-
0
562
664,604
-
7
-
-
-
-
744
10,623
.
23
10
212
90,465
-
19
0
.
5

-------
460     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

POLLUTANT
Dieldrin
2,4-Dimethylphenol
2,4-Dinitrophenol
alpha-Endosulfan
beta-Endosulfan
Endosulfan
Endrin •
Fluoranthene
Fluoride
HepUchlor
Hexachlorobenzene
alpha-Hexachlorocyclohexane
beta-Hexachlorocyclohexane
Hexachlorocyclohexane
Hexachloroetfaane
Iron
Lead
Linda ne
Mercury
Methylene Chloride
Nickel
Parathion
PCBs
Pentachlorobenzene
PenUcblorophenol
Phenanthrene
Phenol
Selenium, toul
Silver
1 ,2,4,5-Tetrachlorobenzene
2,3,7,8-TCDD
Tetrachloroethylene
POLLUTANT LOADING Obs-eq/year)
BASELINE
3,190,719
—
—
—
—
—
189,557
—
73,584
2,324,390
542,816
82,945
23,117
34,675
—
17,732
1,794,813
5,366
519,286
5
88
—
454,908
443,840
6,742
—
—
—
426,685
388,895
3,989,245
—
REDUCTION
(LOW ESTIMATE)
2,092,368
—
—
—
—
—
183,778
—
0
434,659
195,908
81,721
22,423
33,172
—
0
1,080,141
0
66,304
2
76
—
0
441,528
0
—
—
—
0
376,802
0
—
REDUCTION
(HIGH ESTIMATE)
2,092,368
-
-
-
-
-
183,778
-
0
2,201,441
195,908
81,788
22,423
33,172
-
0
1,080,141
5,289
67,878
2
76
-
0
441,528
5,891
-
-
-
0
386,536
0
-

-------
                                         Section DC: Executive Order 12866
461
POLLUTANT
Thallium
Toluene
Toxaphene
1,1,1-Trichloroethane
Trichloroethylene
2,4,6-Trichlorophenol
Zinc
TOTALS
POLLUTANT LOADING Obs-eq/year)
BASELINE
—
12
16,833,496
—
8
—
1,735
35,364,934
REDUCTION
(LOW ESTIMATE)
—
0
36,956
—
6
—
76
5,838,289
REDUCTION
(HIGH ESTIMATE)
-
2
36,956
-
6
-
76
7,649,510
*—indicates that there was no reasonable potential for the pollutant to exceed Guidance-based WQBELs for any of the sample facilities.

-------
462    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      Approximately 80 percent of the pollutant load reduction for the final
Guidance, regardless of the scenario, is attributable to reducing
bioaccumulative pollutants of concern (BCCs) as a result of pollution
minimization plans and end-of-pipe treatment.  However, it should be noted
that for several BCCs (e.g., PCBs, 2,3,7,8-TCDD, mercury, toxaphene), little
or no reduction from the baseline is estimated.  This phenomenon occurs
because of the method used by EPA to derive load reductions.  As described in
section IX.B.2.a, when an existing permit limit or a Guidance-based WQBEL is
below the analytical detection level, one-half of the method detection level
is used for each.  The result of this approach is that no pollutant reduction
would be estimated, regardless of whether the Guidance-based WQBEL was further
below detection levels than the existing permit limit.  Therefore, in effect,
for several of the -toxic pollutants of concern, the lack of estimated
reduction is due to the downward shift in the permit baseline  (i.e., more
stringent existing permit limits).  As discussed below in the benefits section
(section IX.E), this results in fewer benefits than originally anticipated for
the April 16, 1993, proposal.

c.    Estimated Cost-Effectiveness

      EPA estimates that the cost-effectiveness of the final Guidance under
the low-end estimate is just under $7.00/lbs-eq for the direct dischargers
only; with the cost for indirect dischargers, the cost-effectiveness rises to
$10.30/lbs-eq.  Under the high-end estimate, the cost-effectiveness increases
to just over $49.00/lbs-eq.

      The estimates for the final Guidance are considerably more cost-
effective than those estimated for the proposed Guidance using the revised
approach ($35.96/lbs-eq and $63.82/lbs-eq; low-end and high-end scenarios,
respectively).  For comparative purposes, cost-effectiveness values for
effluent limitations guidelines and standards range from just over $1.00/lbs-
eq/year to over $500/lbs-eq/year.

D.    Maior Issues/Comments and Responses Related to Estimated Costs

      EPA's cost estimate for the proposed Guidance generated extensive
comments related to the potential costs of numerous aspects of the Guidance.
In terms of the total estimated cost for the proposed Guidance, most of the
regulated community, driven by the uncertainty of future impacts of the
Guidance, claimed the costs were orders of magnitude higher than those
estimated by EPA.  Alternatively, there were concerns from several States that
the estimated impact of the proposed Guidance, in relation to their current
program to regulate point source dischargers, was too high.

      The remainder* of this section discusses the major issues raised by
commenters and EPA's approach to evaluate and address these issues.  In
evaluating the impact of these issues on the cost to comply with the final
Guidance, EPA used both the low- and high-end compliance cost estimates as the
basis for comparison.

1.    Use of Pollution Prevention/Waste Minimization to Achieve Guidance-Based
      Limits

a.    Comments

      Many comments from the regulated community stated that EPA's  compliance
costs relied too much upon pollution prevention techniques.  Commenters
maintained that pollution prevention techniques may not always be feasible  for
some industries and processes.  This was particularly noted for mercury and
PCBs that are apparently present  in intake waters  (deposited to surface waters
through atmospheric deposition and precipitation, or release from historically
contaminated  sediments) for which pollution prevention would be infeasible.
In the absence of process-specific information, many commenters felt that  it

-------
                           Section IX: Executive Order 12866                        463
is impossible for EPA to determine whether pollution prevention would reduce
pollutants in process wastewaters.

b.    Response

      As discussed in section IX.C.l.e, EPA attempted to collect some data
related to the cost and effectiveness of pollution prevention techniques for
the pollutants being regulated under the final Guidance.  The result of these
efforts, which generally constituted an extensive review of the EPA Pollution
Prevention Information Clearinghouse  (PPIC), was that very little is
documented regarding the effectiveness of pollution prevention to remove many
of the pollutants subject to the Guidance.  The limited data did, however,
suggest that there are facilities that have reduced toxic pollutants to below
analytical detection levels using pollution prevention techniques.

      As also discussed in section IX.C.l.d., EPA developed and used a
decision matrix for purposes of estimating the types of controls and costs
associated with these controls to avoid unjustified use of waste
minimization/pollution prevention techniques to achieve Guidance-based WQBELs.
Under the decision matrix, waste minimization/pollution prevention was
considered only after consideration of modifying existing treatment systems to
achieve Guidance-based WQBELs.  Further, waste minimization/pollution
prevention controls were only considered when a relatively insignificant
amount of pollutant needed to be removed  (i.e., less than 10 to 25 percent of
current discharge levels) and when EPA considered the production process or
source generating the pollutant to be amenable to pollution prevention
techniques.

      As an alternative to the use of waste minimization/pollution prevention,
EPA also considered the use of the flexibility provided through the Guidance
(i.e., regulatory relief) as a control alternative in estimating costs for the
final Guidance.  However, the use of regulatory relief was limited to only
those facilities (under the low-end scenario) and categories  (under the high-
end scenario) where the estimated cost was disproportionately high as compared
to the resulting estimated pollutant reduction.  The primary difference
between the low-end and high-end cost estimates is that under the high-end
scenario, EPA's decision matrix relied on end-of-pipe treatment to achieve
Guidance-based WQBELs by limiting compliance costs based upon the use of waste
minimization/pollution prevention techniques and the costs associated with
pursuing a relief mechanism (e.g., standard variance, phased-TMDL, etc.).

      In summary, in estimating costs to comply with the final Guidance, EPA
has taken steps to ensure that the use of waste minimization/pollution
prevention as a method to comply with Guidance-based WQBELs was limited to
only those facilities where its use was considered technically feasible.  This
is particularly true under the high-end scenario where end-of-pipe treatment
formed the basis of compliance cost estimates.

2.    Future Impact of Detection Levels

a.    Comments

      Many commenters disagreed with the cost study approach for the proposed
Guidance to only use current analytical detection levels as the basis for
estimating costs.  Further, many commenters stated that regular improvements
in analytical detection levels should be expected over time, resulting in a
dramatic impact on the costs to comply with the final Guidance in the future.
Commenters felt that they would need to design and operate treatment systems
and incur the costs now to achieve the WQBELs, in order to reduce the chance
that the technology will have to be replaced when analytical detection levels
improve in the future.  Several commenters also stated that EPA underestimated
the cost associated with implementing pollution prevention and waste
minimization (through the required pollutant minimization plan) to
consistently keep discharge levels below detection levels.

-------
464    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

b.    Response

      In recent years, several States in the Great Lakes System have
promulgated water quality- criteria for various toxic pollutants that are more
restrictive than the level of analytical detection.  Implementation of these
existing water quality criteria by these States do take into account the
ability to detect the pollutant in the wastestream.  Likewise, procedure 8,
appendix F, of part 132 clearly provides that the water quality-based effluent
limit must be derived from the water quality criterion; compliance with that
limit, however, will be based on the Minimum Level (ML) where available.  When
a promulgated ML is not available, compliance with that limit may be based on
the lowest level of quantification (at the State's discretion) defined in
procedure 8 of part 132.

      In estimating the compliance cost for the final Guidance, EPA
conservatively used the method detection level (MDL)  as the compliance level.
In actuality, the State permitting authority is only required to use the ML
(as defined under 40 CFR part 136), which is generally higher than the MDL, as
the basis for reporting compliance with the Guidance-based WQBEL.  Although
EPA used the pollutant MDL for costing purposes,  the Agency acknowledges that
estimating treatment costs for WQBELs below the MDL,  and most likely below the
ML, would be speculative, particularly as such estimation relates to expected
performance of treatment processes.

      However, EPA does believe that an aggressive pollutant minimization plan
can successfully result in compliance with WQBELs below detection levels.  In
fact, several of the sample facilities examined as part of the cost study have
successfully performed studies, required as part of their current NPDES
permit, to effectively reduce all detectable amounts of particular pollutants
of concern from their discharge.  For example, the State of Wisconsin required
the Fort Howard Paper Company, as part of an NPDES permit special condition,
to perform a PCB reduction study "to reduce PCBs to the maximum extent
possible with a goal of zero discharge."  Resulting effluent concentrations of
PCBs allowed the State of Wisconsin to recommend reduced permit requirements
for PCBs in the subsequent draft reissued permit.

      EPA agrees that some facilities will want to ensure compliance with
WQBELs below detection levels through the use of additional or enhanced end-
of-pipe treatment.  But as shown by the above examples, pollution minimization
plans can be effective in achieving WQBELs below detection levels.

      As discussed .in section IX.C.2 of this document, EPA estimates that
pollutant minimization plans account for a significant proportion of the total
compliance cost under the low-end scenario.  EPA evaluated the impact of these
requirements by deriving cost estimates assuming that permitting authorities
would only require increased monitoring for any pollutant for which a
Guidance-based WQBEL was below analytical detection levels.  Under this
scenario, EPA estimates that annual compliance costs for direct dischargers
will decrease by over 60 percent to $16.6 million.  EPA did not take pollutant
reduction credit when Guidance-based WQBELs were below detection levels, and
as a result the estimated pollutant load reductions decrease to  1.3 million
Ibs-eq/day, which is almost 80 percent less then the reduction estimated for
the final Guidance.'  Under the high-end, compliance costs do not drop as
dramatically as the low-end costs due to the shift towards end-of-pipe
treatment, however, the pollutant load reductions decrease by over 50 percent.
Because of the impact on pollutant load reductions, EPA concludes that
retaining the pollutant minimization plan requirements is justified for the
final Guidance.

      EPA also evaluated the potential impact improvements to analytical
detection levels would have on compliance cost estimates.  EPA particularly
estimated costs-and pollutant load reductions under two scenarios, one that
assumes MDLs improve 10-fold over time and another that assumes  MDLs improve
100-fold over time.  The results of the analysis show  conceivable increases  in

-------
                           Section IX: Executive Order 12866                        465
estimated compliance costs.  When MDLs,become 10 times more stringent, annual
costs increase by over $500 million dollars.  Pollutant load reductions also
increase when MDLs decrease 10-fold by over 12 million toxic pounds-equivalent
per year.  When MDLs become 100 times more stringent, the annual compliance
costs are estimated to increase by just over $880 million and pollutant load
reductions would increase by approximately 19 million toxic pounds-equivalent
per year above the final Guidance estimates.  These results, indicate that as
analytical detection levels improve, the costs to implement the final Guidance
increase.  However, the projected increase in compliance costs are offset by
comparable pollutant load reductions, which in the future could significantly
improve water quality in the Great Lakes Basin.

3.    Intake Credits

a.    Comments

      As described in section VIII.E.3 of this document, there were many
comments related to whether EPA can and should regulate pollutants in a
discharge that originate in the discharger's water supply.  Many commenters
particularly believed that the compliance costs would be very high if intake
credits would not b.e provided for once-through, non-contact cooling water.  In
fact, most of the regulated community studies assumed that the proposed
Guidance would not exempt once-through, non-contact cooling water, and these
wastestreams were treated as process wastestreams for purposes of estimating
compliance costs.

b.    Response

      In generating cost estimates for the proposed Guidance, EPA accounted
for the provision that intake pollutants would not present reasonable
potential at facilities that merely passed through the pollutants unchanged.
As discussed in section IX.B.I.a, EPA also developed two different cost
scenarios to account for the lack of ambient background concentration data and
for negative wasteload allocations.  As also discussed in section IX.C.I.a.ii,
EPA revised its approach to developing cost estimates to reflect the intake
pollutant provisions of the final Guidance.

      In an effort to evaluate the impact of intake pollutants on estimated
compliance costs, EPA developed compliance costs under a number of different
intake pollutant scenarios.  For discharges to different bodies of water, EPA
estimates no significant impact  (less than 0.5 percent) to either the
compliance costs or pollutant load reduction estimates under both the low- and
high-end scenarios,' regardless of whether intake credits are relaxed  (no net
increase) or made more stringent  (no intake credit allowed).  This result
occurred because discharges occurred infrequently to different bodies of water
that were not attaining water quality standards for the pollutant or
pollutants requiring intake credits.

      Alternatively, the form of intake credits does impact discharges to the
same body of water.  When intake credits were not allowed for discharges to
the same body of water, the annual compliance costs for direct dischargers
increased by $245 million, representing over a 600 percent increase from the
final Guidance low-end estimate.  In addition, pollutant load reductions
increased to 6.4 million Ibs-eq/year, which represents a 9 percent increase
from the final Guidance low-end estimate.  The cost-effectiveness of the low-
end scenario increases to almost $45/lbs-eq.  EPA projects the same trend
using high-end scenario costs where the costs increase by over 60 percent, but
pollutant reductions increase by only 7 percent.

      As a result of this analysis, EPA agrees that the absence of intake
credit provisions, particularly for discharges to the same body of water, will
have an impact on the cost to comply with the final Guidance.  EPA, however,
has included intake credits as part of the final Guidance.

-------
466    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

4.     Tier II Criteria

a.     Comments

      EPA stated in its cost estimate for the proposal that evaluating  the
costs to comply with only numeric Tier I criteria was a limitation of its cost
study.  Many commenters agreed with EPA and stated there would be a
significant cost to comply with Tier II values.  This cost would first  include
monitoring to generate the data for the Tier II value itself, and then  the
cost to comply once the values were used as the basis for a Guidance-based
WQBEL.

b.     Response

      EPA agrees that compliance costs should have been estimated for
pollutants for which Tier I criteria or Tier II values could be developed and
that were present in effluent at levels of concern.  As a result, EPA's cost
estimate for the final Guidance is based upon evaluation of compliance  for 69
pollutants  (see discussion in section IX.C.I.a above).  To determine the
potential impact of the use of Tier I criteria versus Tier II values, EPA
developed compliance costs under a variety of scenarios.

      If only Tier I criteria are used, the annual compliance costs for direct
dischargers would drop by $5 million, which is just under 12 percent of the
final Guidance low-end estimate.  The pollutant load  reductions would also
decrease by about 8 percent of the estimate for the final Guidance.  Under the
high-end scenario, both costs and pollutant load reductions decrease similarly
 (2 percent drop in costs and 6 percent drop in pollutant load reduction).

      If Tier I criteria and Tier II values are used  for all pollutants, the
annual compliance costs increase insignificantly at both the low- and high-
end.  This result was expected since  the scenario only added Tier II wildlife
values.  Although many of these additional Tier II wildlife values are  more
stringent than other Guidance criteria, the impact is insignificant since both
the Tier II wildlife values and the other Guidance criteria are  below
analytical detection levels.

5.     Wildlife Criteria/Mercury Criteria

a.     Comments

      A significant number -of comments were received  stating that  the wildlife
criteria, and particularly the criteria for mercury,  would result  in
significant compliance  costs for  the  regulated community.  Of particular
concern to commenters was the ubiquitous nature of mercury in the Great Lakes
System, and, as a BCC,  the potential  for high costs associated with installing
treatment to achieve the mercury  criteria after mixing  zones are eliminated.

b.    Response

      EPA agrees that the wildlife  criteria proposed  under the Guidance could
impact the  costs to comply with the Guidance.  As  discussed  in section  V.I  of
this  document, the .final wildlife criteria  for mercury  has increased  from 180
picograms per liter  (pg/1) to 1,300 pg/1.   Further,  the final Guidance  limits
the use of  the wildlife criteria  methodology to the Tier I procedure  for the
22 BCCs for which  sufficient data exist.

      The final  Guidance cost estimate includes Tier  I  wildlife  criteria for
mercury, 2,3,7,8-TCDD,  PCBs, and  DDTs.   In  addition,  hypothetical  Tier  I
 criteria were developed for  three additional BCCs  and hypothetical Tier II
values for  21 other pollutants which  were used to  assist in  estimating  costs
 for the final Guidance.  Six of the 21 pollutants  are not BCCs,  which tends to
 further overstate  the  impact of the wildlife criteria,  as the  final Guidance
 limits development'of Tier I wildlife criteria for only BCCs.   Further, these

-------
                           Section DC: Executive Order 12866                        467
values were estimated using many simplifying and conservative assumptions that
are expected to be generally more stringent than the values that would be
derived by permitting authorities using the wildlife methodology in the final
Guidance.

      However, using the additional wildlife criteria results in an
insignificant increase in annual compliance costs.  Alternatively, excluding
all wildlife criteria also results in essentially no difference in compliance
cost estimates at both the low- and high-end.  These results indicate that
factors other than the wildlife criteria tend to drive the costs of the final
Guidance.  In the absence of wildlife criteria, the Guidance human health
criteria would form the basis for Guidance-based WQBELs.  The Guidance human
health criteria for most pollutants are below analytical detection levels and,
as such, the costs for treatment and pollutant minimization plans would be
incurred by a facility.  Although the wildlife criteria in general are more
stringent than the Guidance human health criteria, they would -also result in
Guidance-based WQBELs below analytical detection levels and, therefore, the
same treatment and pollutant minimization plan requirements and costs.

6.    Elimination of Mixing Zones for BCCs

a.    Comments

      EPA received numerous comments that the study neglected the costs
related to the eventual elimination of mixing zones provision for BCCs.
Further, most commenters stated that the elimination of mixing zones for BCCs
in 12 years will impose enormous costs without commensurate benefits.

b.    Response

      As discussed in section VIII.C of this document, and as promulgated in
procedure 3 of appendix F to part 132, the final Guidance retained the
requirement for elimination of mixing zones within 12 years, but does provide
some flexibility to allow limited mixing zones for BCCs if the facility can
show that all prudent and feasible treatment technologies are being
implemented to reduce the discharge of BCCs to the maximum extent possible.

      EPA began to address this issue in the sensitivity analyses performed
for the proposal.  In general, if analytical detection limits remain the same,
it was concluded that a cost would be incurred infrequently for a BCC after
mixing zones have been taken away.  This conclusion was based on the fact that
many of the WQBELs and associated criteria for BCCs were already below
analytical detection levels.

      In estimating costs for the final Guidance, EPA conservatively assumed
that no mixing zones were allowed for BCCs.  The cost for the final Guidance
was presented previously in section IX.C.2.  To determine the impact of this
requirement on facilities (in terms of cost) and the environment (in terms of
pollutant load reductions),  EPA revaluated the sample facilities allowing the
same mixing zones for BCCs as are allowed for non-BCCs.

      EPA estimates that the addition of mixing zones for BCCs results in an
incremental annual .cost savings to direct dischargers of just over $200,000,
which is less than a 0.5 percent increase above the final Guidance low-end
estimate.   In terms of pollutant load reductions, the addition of mixing zones
results in lowering reductions by 15,700 Ibs-eq/year  (which is less than 1
percent of the total estimated pollutant reductions for the final Guidance).
Slight reductions in cost and pollutant load reductions were also found under
the high-end scenario.

      The relatively small impact associated .with the elimination of mixing
zones for BCCs is primarily because the criteria for most BCCs are relatively
stringent,  and usually well below analytical detection levels.  Even with the
dilution afforded by the mixing zones allowed under the final Guidance, the

-------
468    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

resulting WQBELs are still below analytical detection levels and, as a  result,
do not drastically impact the costs and load reductions  (i.e., the pollutant
controls would not change if both WQBELs were below analytical detection •
levels).  As a result of this analysis, EPA disagrees that the elimination of
mixing zones for BCCs will have an adverse impact on the regulated community.

7.    Antideqradation

a.    Comments

      EPA received -numerous comments that the regulatory impact  analysis for
the proposed Guidance neglected the costs related to the antidegradation
provisions.  Although EPA performed a sensitivity analysis for the
antidegradation issue at the time of proposal, many comments stated that the
analysis only estimated the cost of the demonstration process.   Many
commenters felt that although the demonstration process can assist in
compliance, it does not ensure compliance or necessarily make it cheaper.
Finally, many commenters stated that the antidegradation provisions will
inhibit economic growth in the region, because facilities would  be prohibited
from returning to full production or increasing current production capacities.

b.    Response

      EPA agrees with the commenters that the antidegradation provision of the
final Guidance, as promulgated under appendix E to part 132, may impact the
regulated community.  However, EPA disagrees that the impact will be
significant.  EPA also agrees that costs other than demonstration cost  may be
incurred by a facility in the form of lost opportunities for business.

      In an effort to estimate what the lost opportunity cost could be  related
to implementation of the antidegradation provision of the Guidance, EPA
performed a separate cost analysis.  This analysis was based on  the
antidegradation requirements for BCCs contained in the final Guidance,  that
requires an antidegradation review if there is a deliberate action on the part
of a facility that results in a significant lowering of water quality  (i.e.,
the activity results in an increase in BCC loadings).  The general premise
behind EPA's analysis was that the economic growth in the region (as indicated
by total value of shipments for six major categories of direct dischargers in
the basin) would continue at a pace equal to the average growth  over the last
8 years  (1987-1994) .  The resulting estimated incremental annual growth ($864
million) served as the baseline from which impacts were estimated.

      Under the worst case, it was assumed that all facilities with BCCs in
their discharge  (approximately 5 percent of all facilities) requested an
antidegradation review and were denied; thus, 5 percent, or $43.2 million, of
the incremental annual growth would be lost due to the Guidance. More
realistically, if it was assumed that half of the facilities requesting
antidegradation reviews for BCCs were allowed to increase discharges, only
$21.6 million of opportunity would be lost each year.  Finally,  assuming that
only 10 percent of the facilities discharging BCCs request an antidegradation
review, and only half are denied, then the opportunity lost for  growth  would
be approximately $2.2 million.

      EPA does not expect to see an increase in baseline loadings for BCCs
because for many, their use is already banned or severely restricted by the
Agency.  A study performed for EPA shows that 14 of the 28 BCCs  are banned or
severely restricted, and another four of the 28 are by-products  of banned or
severely restricted BCCs.  The remaining 10 BCCs have some limited
restrictions for use or are not restricted at all, or no data were found for
them.  EPA therefore believes that the mid- and high-estimates  of lost
opportunity are unlikely because the  increase of banned or restricted BCCs
should not occur .due to releases from the manufacture or use of  the BCC.  In
fact, EPA assumes that the levels of  these BCCs will decrease over time in
point source discharges and in the environment.  Several other  BCCs are

-------
                           Section EX: Executive Order 12866                        469
present, as contaminants or by-products.of banned or restricted BCCs  (e.g.,
heptachior epoxide is a metabolic breakdown product of heptachlor),  and for
the same reason, the levels of these BCCs should also decrease over  time.-
Therefore, EPA does not anticipate antidegradation reviews as a result of an
increase in loading levels for BCCs that result in a significant  lowering of
water quality.

      If the low-end estimate is used, then EPA would expect a modest 3
percent increase in the low-end annual compliance cost.  The potential
benefits, although not quantified by EPA, could be relatively significant for
some receiving waters because additional discharges of BCCs would be denied.

8.    Additivitv

a.    Comments

      EPA received several comments related to the potential impact  of the
additivity provision discussed in the preamble to the proposed Guidance.  Most
of these commenters felt that the additivity provision will increase the
estimated compliance costs of the Guidance.

b.    Response

      In an effort to evaluate the impact of the additivity provision on the
compliance cost of the final Guidance, EPA developed cost estimates  for two
scenarios.  Under one scenario, EPA assumed that additivity would be
controlled if the total carcinogenic risk in a discharge was less than 10~5 and
accounted for by assuming that individual criteria were based on  a 10"5 risk
level.  Under the second scenario, EPA assumed that the additive  effects from
carcinogens would be accounted for if individual criteria were based on a 10"*
risk level.

      The impact of the first scenario was relatively insignificant  (less than
0.5 percent decrease in costs and just over 1 percent decrease in pollutant
load reductions at both low- and high-end estimates).  The relatively
insignificant change in cost and load reductions is based on the  fact that
most facilities did not detect more than a few carcinogens in their  discharge.
As a result, the final Guidance estimates  (based upon a total carcinogenic
risk of 10*s but accounted for by distributing the risk across all  carcinogens
in the effluent) did not represent more stringent WQBELs for carcinogens, as
compared to only accounting for the risk through the individual criteria.

      When the individual criteria risk level is adjusted down to a  10"*, a
more dramatic increase in costs occurs.  EPA estimates that a 10"* risk level
for individual criteria would increase the low-end estimate of annual
compliance costs for direct dischargers by $10.4 million.  The associated load
reductions do not increase as dramatically, accounting for only an additional
6,000 Ib-eq/year.  The reason a large pollutant reduction did not accompany
the large increase in costs under the low-end scenario is the assumption that
a significant number of facilities would pursue some sort of regulatory
relief,  for which there is no pollutant reduction credit, to meet the more
stringent criteria based on a 10"* risk level.

      The same trend occurs at the high-end, where EPA estimates  that costs
will increase by over 30 percent, but pollutant load reductions will decrease
by less than 1 percent.  However, under the high-end scenario where  variances
are limited to categories that exceed the high-end cost trigger,  the
significant increase in costs is due to the costs associated with installing
and maintaining end-of-pipe treatment for pollutants impacted by  the more
stringent criteria.  The insignificant load reductions associated with the
large increase in costs is due to the fact that some regulatory relief was
still justified under the high-end.  Further, for some pollutants with
criteria below the analytical detection level, the shift from criteria based
on a 10"5 risk level* to criteria based on a 10"* risk level resulted in criteria

-------
470    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

further, below analytical detection levels, which had no impact on pollutant
load reductions.

      In summary, the additivity provision included in the final Guidance
could have a significant impact on the cost to comply, depending upon how
States and Tribes decide to implement the final additivity provisions.  As a
result, EPA p'rovided implementation flexibility in the additivity provisions
for the final Guidance to afford States and Tribes a broad range of
alternatives, which will allow cost-effective decision-making under a variety
of circumstances without sacrificing human health or environmental protection.

9.    Other Study Costs

a.    Comments

      The majority of comments received by EPA related to the regulatory
impact analysis for the proposed Guidance stated that EPA severely
underestimated the costs to comply.  One commenter stated that compliance
costs are "likely to be 10, 20, or even 100 times higher" than the $232
million annually predicted by EPA.  Many commenters supported the results of
an independent impact study that estimated compliance costs to be between $710
million and $2.3 billion per year.

      Many commenters provided their own estimates of the cost of the proposed
Guidance, and in most instances, the cost estimates provided were far greater
than the total cost estimated by EPA.  The cost estimates provided by
commenters for many individual facilities were most frequently on the order of
$10 to $50 million in total capital investment costs alone; two different
municipalities estimated total capital costs in excess of $100 million.

      Many industry groups also provided compliance cost estimates that were
significantly higher than those projected by EPA.  One group representing the
pulp and paper industry in the Great Lakes concluded that the proposed
Guidance would require $1.25 billion for new capital equipment and $284
million in annual costs.  Another group, representing Michigan POTWs,
estimated that annual costs for the proposed Guidance would be $110 million.
Another group representing the chemical industry in New York estimated that
the proposed Guidance would cost from $45.2 million to $76.1 million in total
capital costs, and between $9.6 million to $17.7 million in annual operation
and maintenance costs.

b.    Response

      While EPA did not evaluate in detail the process-specific  and plant-
specific information submitted my the many commenters, EPA believes that
several general observations can be made regarding these studies and how  they
differ from the EPA cost study.

      EPA noted that most studies assumed that an intake pollutant provision
would not be available in the Guidance, and thus several studies include  costs
to treat intake water.  As discussed  in section IX.C.3.C of this document, the
final Guidance includes provisions for intake pollutants for discharges to
both the same and different bodies of water.  Further, in its analysis of
intake pollutant options, EPA concluded that including intake pollutant
provisions for discharges to the same body of water would decrease costs  six-
fold from the estimated cost of compliance.

      Many commenters also assumed that the  "no addition of mass" provision of
the proposed Guidance would include,  for example, metals leaching from intake
pipes.  As a result, these commenters assumed that there would be reasonable
potential to violate water quality standards and the  resulting outfalls would
be subject to the proposed Guidance.  The final Guidance retains the
prohibition against "no addition of mass" to qualify  for a determination  that
a reasonable potential  to violate water quality standards does not exist.  The

-------
                           Section DC: Executive Order 12866                        471
final Guidance provides facilities in ..this situation relief in the form of
allowing  "no net addition" limits for discharges to non-attained waters if the
source of the intake water pollutant and the point of discharge is to the same
body of water.  The final guidance gives permitting authorities discretion to
determine how to assess compliance with a "no net addition" limit.

     ' The majority of cost estimates provided by commenters include the 'costs
for the addition of end-of-pipe treatment to achieve proposed Guidance-based
WQBELs.  This was particularly the case when WQBELs were•expected to be below
analytical detection levels.  EPA disagrees that end-of-pipe treatment is
necessary to achieve Guidance-based WQBELs in all cases.  As discussed in
section IX.C.3.b, tfhere are documented cases where waste minimization
techniques have been used to comply with existing permit limits established
below detection levels.  Although waste minimization is not always applicable,
EPA assumes that a facility would first evaluate whether process changes or
modifications are feasible, prior to incurring costs for adding treatment.
Even so, as discussed in section IX.C.2.a, EPA estimates that the annual cost
for all direct dischargers would be about $370 million with limited use of
regulatory relief mechanisms and waste minimization as control options.

      Finally, EPA agrees with the many commenters who included costs to
comply with Tier II values.  As discussed in section IX.C.l.b of this
document, EPA included cost estimates to comply with the most prevalent and
toxic pollutants for which Tier II values could be derived.

E.    Benefits

1.    Summary of Proposal

a.    Introduction

      The following discussion describes the methodologies and results of the
benefits analysis Chat accompanied proposal of the Guidance in April, 1993.
The benefits analysis was intended to provide insight into both the types and
potential magnitude of the economic benefits expected to arise as a result of
the proposed Guidance.  A qualitative assessment of these benefits was
provided.  In addition to the qualitative assessment, empirical estimates of
the potential magnitude of the benefits from controlling point sources were
developed to the extent feasible, and then compared to the estimated costs of
the proposal.  This discussion was intended to demonstrate data needed and a
methodology suitable for comparing benefits and costs.  The qualitative and
quantitative benefits assessments are summarized below.

b.    Qualitative Assessment of Benefits

      A qualitative assessment of the anticipated benefits of the proposed
Guidance focused on: (a) the sensitivity and unique attributes of the
receiving waters, (b)  the nature of the toxic pollutants addressed by the
proposed Guidance and some implications for human health and ecological risk
reductions, and  (c)  an overview of exposed and sensitive populations.
Substantial benefits were predicted from the proposed Guidance due to the
significant health and ecologic risks posed by the chemicals addressed when
combined with physical characteristics of the Great Lakes that cause them to
be particularly vulnerable to bioaccumulative toxic pollutants.  Given the
long retention time, low sedimentation, low productivity in the Great Lakes
System and the presence of self contained, vulnerable populations  (and hence
the persistence of toxic contaminants in this ecosystem), loadings reductions
realized as a result of the proposed Guidance were expected to have lasting
impacts on mortality risk and the reproductive success of many aquatic, avian,
and mammalian species of concern.  These benefits included increased
productivity and protection of biological diversity of Great Lakes species
including salmonids and other fish species, cormorants, eagles, osprey, and
otters.

-------
472    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

      The persistence and toxicity of the compounds to be directly controlled
from point sources under the proposed Guidance had important implications for
the benefits analysis.  The principal benefits are difficult to forecast
because most of the direct benefits  (1) are likely to be delayed for many
years, and  (2) are largely ecologic in nature.  For example, the Guidance
provides added protection for endangered species with provisions for site-
specific criteria modification.

      Fish and waterfowl consumption advisories are likely to be lifted as
concentrations of toxic compounds are reduced.  Such actions, and water
quality improvements leading to those actions, would result in increased
recreational fishing and hunting opportunities and increased values for
recreational fishing and hunting days.  According to the U.S. Fish and
Wildlife Service's 1985 National Survey of Fishing, Hunting, and Wildlife-
Associated Recreation  (U.S. Department of Interior, Washington, D.C., 1989),
the Great Lakes supported more than 46 million angler days in that year.  Even
a small increase in the number of angler days or the value associated with
improvements in Great Lakes angling would provide significant annual benefits.
Other recreational opportunities, including boating, swimming, and wildlife
observation would also be expected to be enhanced as water quality and
ecosystem health improve.  Health risk reduction benefits are likely to be
generated through Deduced exposure from the consumption of fish and wildlife
(particularly among subsistence populations relying on Great Lakes fish and
wildlife as a primary food source).  The value of these improvements would
accrue not only to direct users of the Great Lakes, but also to nonusers who
ascribe values to the ecologic benefits resulting from the implementation of
the proposed Guidance.

c.    Quantitative Assessment of Benefits Analysis

      Quantitative benefits estimates were prepared for three case study
sites:  (1) the lower Fox River drainage, including Green Bay, located on Lake
Michigan in northeastern Wisconsin;  (2) the Saginaw River and Saginaw Bay,
located on Lake Huron in northeastern Michigan; and  (3) the Black River,
located on Lake Erie in north-central Ohio.  The case studies focused on
empirically tractable benefit categories, and omitted several types of
potential benefits.  In addition, the numerical results were based on limited
assessments of the extent to which the Guidance might contribute to
improvements beyond baseline levels, and were not intended to illustrate
sensitivity to modest changes in either the timing or stringency of water
quality criteria and associated discharge permit limits.

      Rather, the benefits analysis was geared toward indicating:  (1) the
types of benefits to be anticipated;  (2) a  general approach for describing
and, as feasible, estimating these benefits;  (3) the general magnitude of the
monetized worth of several categories of benefits; and  (4) an indication of
how benefits compared to costs.  These concepts and results are summarized
below.

i.    Economic Concepts Applicable to the Benefits Analysis

      The term economic benefits  refers to  the dollar value associated with
all of the expected direct positive  impacts of the proposed Guidance; that  is,
all outcomes that lead  to higher  social welfare.  Conceptually, the monetary
value of benefits is embodied by  the  sum of the predicted changes  in  consumer
(and producer) surplus.  These surplus measures are standard  and widely
accepted terms of applied welfare economics,  and reflect the  degree of well-
being enjoyed by people given different levels of goods and prices  (including
those associated with environmental  quality).

      This  conceptual economic foundation raised three  relevant issues and
potential limitations for  the benefits analysis of  the proposed Guidance.
First, the  standard economic approach to estimating environmental  benefits  is
anthropocentric--all benefit values  arise from how  environmental  changes are

-------
                           Section IX: Executive Order 12866                        473
perceived and valued by humans.  Second, benefits of all future outcomes are
valued by the present-day human population.  Third, all near-term as well as
temporally distant future physical outcomes associated with reduced pollutant
loadings need to be predicted and then translated into the framework of
present-day human activities and concerns.

     • The potential benefits associated with the proposed Guidance included
two broad categories: use benefits and nonuse benefits  (also referred to as
passive use, or intrinsic benefits).  The use benefit category embodies both
direct and indirect uses of the impacted waters, and the direct use category
embraces both consumptive  (e.g., fishing) and non-consumptive  (e.g., wildlife
observation) activities.  In most applications to pollutant reduction
scenarios, the most prominent use benefit categories are those related to
recreational fishing, boating and/or swimming.

      Recreational use benefits may or may not reflect society's prime
motivation for environmental protection measures; however, recreational
activities are amenable to various non-market valuation techniques  (e.g.,
travel cost models) and, accordingly, have received considerable empirical
attention from economic researchers over the past two decades.  Thus, there is
a considerable body of knowledge relating to recreational fishing and related
activities, and the.se generally indicate that water-based recreation is a
highly valued activity in society.  Accordingly, many benefits analyses focus
on recreational values because they are well understood, there is a large body
of empirical research to draw upon, and the associated benefits tend to be
quite large.

      Improved environmental quality can also be valued by individuals apart
from any past, present or anticipated future use of the resource in question.
Such nonuse  (or intrinsic) benefits include aesthetic, bequest  (preservation
for future generations), and existence  (e.g., ecologic) values.  Nonuse values
may be of a highly significant magnitude; but the benefit value to assign to
these motivations often is a matter of considerable debate.  Whereas human
uses of a resource can be observed directly and valued with a range of
technical economic techniques, nonuse values can only be ascertained from
directly asking aurvey respondents to reveal their values.  The inability to
rely on revealed behavior to ascertain nonuse values has led to debates as to
whether they exist for applicable changes in environmental quality and, if so,
whether they are of an appreciable magnitude relative to use values.  As
described below, nonuse benefits were considered relevant and of appreciable
magnitude for the proposed Guidance.

      The category of nonuse values considered most important for the proposed
Guidance was ecological benefits associated with decreasing the level of toxic
compounds found in Great Lakes waters and sediment.  Such ecological benefits
are likely to embody reduced risks of direct mortality, and increased
reproductive success, in a range of important fish and wildlife species.  The
species include, but are not limited to, bald eagles, cormorants, and other
piscivorous avian species; mink, river otter and other mammalian species that
feed on fish and crustaceans; and a wide range of aquatic species such as lake
trout and other salmonid species.

ii.   Benefits Methodologies

      The quantitative case study benefits analyses utilized benefits transfer
techniques.  Benefits transfer is an approach to estimating benefits in which
benefits estimated for one site are "transferred" to another site.  The use of
benefits transfer is not new to EPA; the resource-intensive demands associated
with developing benefits estimates often leads EPA to use existing data
sources and studies to prepare benefits estimates.

      The empirical results presented in the case study analyses utilized
benefits estimates from relevant research on water quality improvements.
Several of these estimates included results derived from contingent valuation

-------
474    Water Quality Guidance for the Great Lakes System - Supplementary Information Document

methodology  (CVM)  (as well as the travel cost methodology and other
techniques).  CVM is an approach in which hypothetical markets are constructed
and presented to individuals in a survey format, with the responses used-to
infer prices and values for the goods and services being evaluated  (such  as
those associated with different levels of environmental quality).  Because the
CVM approach relies on survey responses to hypothetical conditions and
markets, there is controversy surrounding its validity.  However, CVM is  the
only method available for estimating nonuse benefits.

      Furthermore, a U.S. Department of Commerce, National Oceanic and
Atmospheric Administration (NOAA) Blue Ribbon Panel  (K. Arrow, R. Solow,  P.R.
Portney, E.E. Learner, R. Radner, and H. Schuman, Report of the NOAA Panel on
Contingent Valuation, NOAA, Rockville, Maryland, 1993) review of CVM to
measure nonuse values concluded that CVM can produce reliable estimates for
use in the litigation process to determine natural resource damages.  The
panel set forth a number of guidelines for CVM surveys considered to generate
more reliable results.  However, the NOAA panel evaluated CVM as a means  of
estimating nonuse values for use in litigation to determine the liability of a
specific party.  In comparison, the CVM estimates utilized in the case  study
analyses are used in an informational context, to compare benefits to costs.
Strict conformance to the NOAA guidelines is not essential when using the
methodology as a source of benefits information in a policy setting.

iii.  Results

      For each of the case studies, benefits estimates were derived in  one of
two fashions, depending on data availability:  (1) the benefits of a discrete
change in water quality beyond present day conditions were estimated  (wherever
feasible), and then a share of those benefits was apportioned to the proposed
Guidance; or (2) the current value  (at present water quality) of a benefit
category was assessed, and then the benefits of the proposed Guidance were
estimated based upon a range of plausible percentage increases in the level of
benefits  (as plausibly attributable to the proposed Guidance) beyond the
current baseline.

      Costs for the case studies were derived using the methodology developed
to estimate the total costs of the proposed Guidance, but with application to
the dischargers present in each specific watershed.  Case study costs included
costs to direct dischargers only.  The results, as summarized in Table  IX-5,
indicated that the potential benefits of the proposed Guidance under the
attribution assumptions applied appeared to be commensurate with the most
likely cost estimate  (Scenario 2).  These results were, as noted throughout,
sensitive to how water quality benefits were attributed to the proposed
Guidance as opposed to other water quality-impacting actions that are in
progress, or anticipated, absent the Guidance.  In addition, the results  in
Table IX-5 are portrayed in annualized terms, and do not account for the
discounting of potentially delayed benefits.  However, a more standard  form of
net present value calculation, embodying delays in the realization of
benefits, generated comparable results  (e.g., present value costs being within
the range of present value benefits).  These results are summarized in  Table
IX-6.

-------
                                Section IX: Executive Order 12866
475
                                         TABLE IX-5

                   CASE STUDY BENEFIT-COST RESULTS FOR GREAT LAKES
                                WATER QUALITY GUIDANCE"
                  (MILLIONS OF 1992 [FIRST QUARTER] DOLLARS PER YEAR)
BENEFIT CATEGORY
Recreational Fishing
Recreational Boating and Swimming
Subsistence Fishing
Commercial Fisheries
Waterfowl and Other Hunting
Nonconsumptive Use
Human Health Benefits
Nonuse/Ecologic Values
Total Benefits
Annualized Costs'*
CASE STUDY SITE
FOX RIVER/
GREEN BAY
0.9-6.1
+
+
0.2 - 0.3
+
1.3- 1.8
+
0.5 - 3.7
2.9- 11.9
5.1
SAGINAW
BAY
0.9-8.1
+
+
0.2 - 0.7
0.1 -0.1
0.2 - 0.7
+
0.5-7.1
1.9- 16.7
6.0
BLACK
RIVER
0.1-0.6
0.0-0.1
+

+
+
+
0.0 - 0.6
0.1 - 1.2
1.0
+ Positive benefits anticipated, but not estimated in monetary terms.
"  See body of report for explanation of assumptions employed to attribute benefits to the Guidance, and their
application to specific benefit estimates.
k  Source:  SAIC 1993, Cost Scenario 2.
Note: Numbers may not add to total due to rounding.
Source: Regulatory Impact Analysis of the Proposed Great Lakes Water Quality Guidance, April 1993.

-------
476    Water Quality Guidance for the Great Lakes System — Supplementary Information Document
                                          TABLE DC-6
                 COMPARISON OF GUIDANCE POTENTIAL BENEFITS TO COSTS"
                        (MILLIONS OF 1992 [FIRST QUARTER] DOLLARS)
FOX RIVER CASE STUDY
Direct Annualized Comparison
Discounted Benefits and Costs'1
10 Year Phase-in of Benefits
20 Year Phase-in of Benefits
SAGINAW RIVER/BAY CASE STUDY
Direct Annualized Comparison
Discounted Benefits and Costs4
10 Year Phase-in of Benefits
20 Year Phase-in of Benefits
BLACK RIVER CASE STUDY
Direct Annualized Comparison
Discounted Benefits and Costs*
10 Year Phase-in of Benefits
20 Year Phase-in of Benefits
BENEFITS
RANGE
$2.8 -$11.9
$45.2 - $185.7
$34.3 - $140.6
BENEFITS
RANGE
$1.9 -$16.7
$29.6 - $260.5
$22.4 - $197.3
BENEFITS
RANGE
$0.1 - $1.2
$1.6 -$18.7
$1.2 - $14.2
COST RANGEk
$2.7-$14.1
$24.8 -$194.1
$24.8- 194.1
COST RANGEb
$2.1 - $29.3
$31.5 - $537.8
$31.5 - $537.8
COST RANGE"
$0.3 - $7.6
$4.5 - $142.8
$4.5 - $142.8
SCENARIO 2 COST
$5.1
$44.8
$44.8
SCENARIO 2 COST
$6.0
$76.2
$76.2
SCENARIO 2 COST
$1.0
$11.7
$11.7
" See body of report for explanation of assumptions employed to attribute benefits to the Guidance, and their
application to specific benefits estimates.
k Range over 4 cost scenarios (SAIC, 1993).
c Most likely cost scenario.
d Present value over 30 years, using a 7 percent discount rate.
Source: Regulatory Impact Analysis of the Proposed Great Lakes Water Quality Guidance, April 1993.

-------
                           Section DC: Executive Order 12866                       477
      In conclusion, estimating the benefits of the proposed Guidance
presented significant challenges.  It was difficult to  fully account for  and
portray the types of long-term ecologic benefits anticipated from  the  control
of the toxic compounds addressed on the Tier 1 list.  This was  largely a
result of the persistence and potential for bioaccumulation of  these
compounds, and the  "nonuse" nature of benefits to Great Lakes species  of
concern  (which are  less easily quantified and monetized than direct
consumptive uses).  It also was difficult to empirically base attribution of
the benefits of water quality improvements between current conditions  and the
baseline appropriate to the proposed Guidance.  Given these caveats, however,
the case study results suggested that under the attribution assumptions made
explicit in the analyses and the most likely cost scenario, the projected
benefits of the proposed Guidance were commensurate with its costs.

2 .    Ma-ior Is sues/Comments and Responses

      This section  provides a summary of the major issues and public comments
related to the benefits analysis and the economic impact of the proposed
Guidance, and EPA's responses to these issues/comments.  The discussion is
organized by benefits topic areas.

a.    Attribution of Benefits to the Guidance

      The RIA described the basis by which a portion of the potential  benefits
illustrated in the  case study analyses were attributed  to the proposed
Guidance, and the numerous sources of uncertainty surrounding these estimates.
In general, baseline resource values and the value of water quality
improvements were based on available data and applied research.  However,  data
were not available  on the potential contribution of the proposed Guidance
towards such changes.

      One uncertainty in the attribution of benefits relates to knowing the
water quality baseline relevant to implementation of the Guidance.  Assumably,
this baseline is at; some point cleaner than current conditions  because of a
number of regulatory actions that have been promulgated, but not yet fully
implemented (e.g.,  efforts to comply with 303(c)(2)(B)  of the Clean Water Act,
guidance for nonpoint source controls in coastal areas).  Similarly, at the
other end of the water quality spectrum, it is not clear what water quality
benchmark the Guidance will attain with respect to removing fish consumption
advisories and moving towards a "toxic free" status.

      Another important uncertainty in the attribution  issue lies  in the
relative contribution of point sources to the problems  addressed by the
Guidance.  That is, although the reduction in point source loadings expected
to result from the -Guidance is estimated, the contribution of point sources  to
total loadings of the relevant pollutants in the basin  is not known.

      Because of the lack of information on the attribution issue, benefits
were attributed to  the proposed Guidance in the case study analyses for
illustrative purposes based on general information about the sites.  For
example,  for the Pox River case study, the proposed Guidance was expected to
reduce loadings of  several chemicals of primary concern in the  watershed,
including PCBs,  dioxins, and mercury.  For the most significant benefits
categories,  50 percent of future toxics-oriented benefits were  attributed to
the proposed Guidance.  In addition, a sensitivity analysis was performed to
show the impact of 'alternative attribution assumptions  on the benefits
estimates.

i.    Comments

      EPA received numerous comments related to attribution uncertainties.
Some commenters stated that loadings reductions attributable to the proposed
Guidance would be negligible, which will cause economic benefits to be
negligible.   It was stated that the benefits to Wisconsin from  enhanced Lake

-------
478    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

Michigan trout and salmon fisheries from a reduction in PCBs resulting  from
the proposed Guidance were overestimated.  Similar remarks were  directed at
the Saginaw River/Saginaw Bay and Black River case studies.

      One commenter believed that the proposed Guidance would be responsible
for only very modest improvements in designated uses for aquatic life and fish
consumption, and that the conclusion that benefits and costs, are of the same
order of magnitude arises from arbitrary and overly optimistic attribution
assumptions.  While the RIA assumes in one case study that the proposed
Guidance might account for "as much as 50 percent of the potential incremental
use value of a waterbody or as much as 20 percent of the current use value,"
one commenter stated that the loadings data suggest the value is at least an
order of magnitude less.

      Commenters felt that the benefits of the proposed Guidance were
uncertain since EPA could not identify reliably the level of pollution
reduction.  Further, they noted that the RIA did not adequately  consider the
loadings reductions that will be achieved by existing programs and
regulations, or those that are soon to be implemented.

ii.   Response

      EPA acknowledged the uncertainty surrounding the attribution of benefits
to the Guidance, and subsequent research efforts were directed towards  better
quantification of the potential impact of the Guidance in bringing about
future benefits from reducing toxicity.  Efforts were focused on determining
the potential contribution of point source loadings to the toxic-related
problems in the basin.  These efforts include the review of available
literature and data on loadings of toxic pollutants in the basin, research to
identify sources that might be ruled out as significant contributors to
loadings of toxic pollutants in the basin, and the development of a
generalized Great Lakes exposure and bioaccumulation model to estimate  changes
in fish tissue concentrations that might be expected to result from the
Guidance.  The results of these analyses are described in  section IX.E.3.a.

b.    Risk Assessment

      In addition to the three economic benefit case study analyses, the RIA
contained a preliminary assessment of health-related risks to Great Lakes
Basin sport anglers, and potential risk reductions resulting from the proposed
Guidance.  This assessment was based  largely on U.S. EPA's 1991  review  draft
Great Lakes Basin Risk Characterization Study.  Carcinogenic and
noncarcinogenic  (systemic) risks due  to PCB, DDT, mercury, and dieldrin
exposure were evaluated.

i.    Comments

      Comments addressed two major components of the recreational angler risk
assessment.  First, it was noted  that the scope of the project did not  allow
sampling of fish throughout the Great Lakes Basin.  Data from 1989 on
concentrations of contaminants in fish were  applied as representative
averages.  Second, commenters questioned the appropriateness of  the  cancer
slope factors used for  PCBs and dieldrin.

ii.   Response

      The scope of the  risk assessment did  not enable sampling of fish
throughout the basin to calculate average PCB and dieldrin concentrations.
EPA acknowledges that there may be several  orders of magnitude of difference
between species and across lakes, and revised the analysis for the RIA  of  the
final Guidance to use lake-specific  fish tissue contaminant  concentrations.
Although contaminant concentrations  in fish may vary within  individual  lakes,
EPA's lake-specific approach  is more  representative of  the potential exposure
faced by sport fishermen then using  a basinwide average  concentration for  all

-------
                           Section IX: Executive Order 12866                        479
five lakes.  EPA used standard cancer slope factors for PCBs and dieldrin.
Epidemiological evidence of carcinogenicity is not available for dieldrin
exposure.  For PCBs, there are data that suggest PCBs may cause cancer in
humans.  Although it would be highly informative to have data regarding the
carcinogenic effects of toxics in humans, it is impossible to perform studies
of carcinogenic potency in humans.  Therefore, it is standard practice to have
the carcinogenic potency of compounds determined using experimental animals,
and compounds causing cancer in animals are generally considered to be
carcinogenic in humans unless the mechanism of carcinogenicity is not likely
to occur in humans.  It also is standard EPA policy to use available data to
generate a dose-response curve from which the 95th percent confidence limit of
the slope is calculated as the slope factor (the slope factor represents an
upper 95th percent .confidence limit on the probability of a carcinogenic
response per unit intake of a chemical over a lifetime).  This slope factor is
used to determine cancer risks in accordance with EPA risk assessment
guidelines.

      EPA also updated the sport angler risk assessment to incorporate new
data and information, including data on sport-fishing licenses sold in the
Great Lakes Basin, minority and income-adjusted fish consumption levels, and
fish tissue concentrations for additional pollutants.  In addition to the
chemicals listed above, risks were addressed for chlordane, hexachlorobenzene,
2,3,7,8-TCDD, and toxaphene.  EPA also conducted a separate risk assessment
for Native Americaris engaged in subsistence fishing in the basin, but who
would not be included in the exposed population of sport anglers; Native
Americans are not required to purchase fishing licenses to exercise treaty
fishing rights.

c.    Valuation Approaches Used for the Proposed Benefits Analysis

      The case study benefits analyses utilized benefits transfer methodology.
Much of the estimated benefits are based on contingent valuation studies,
including a study of Lake Michigan recreational anglers by Lyke  (1992) .

i.    Comments

      The benefits analysis was criticized for not including any original
research.  The statement was also made that the basis for the methodologies
used in determining use and nonuse values varied so greatly that use and
nonuse benefits should not be summed to calculate total benefits estimates.

      One commenter noted that the use of methodologies such as the CVM forces
the conclusion that costs and benefits are commensurate.  Another commenter
suggested that CVM suffers from serious flaws making it wholly inappropriate
as a source of benefits information for the RIA.

ii.   Response

      EPA acknowledges that the benefits analysis contains no original
research.  However, the use of benefits transfer and other uses of existing
research findings are often the only viable approaches available to EPA for
estimating benefits.  It is standard, accepted practice in economics to
calculate benefits for different categories using different methods. CVM is
currently the only method accepted by the U.S. Department of the Interior
(DOI)  to estimate nonuse values.  The 1993 Blue Ribbon Panel convened by NOAA
evaluated CVM and found it to be an appropriate methodology for measuring
nonuse values, and the method has withstood Federal Court review for its use
in litigation contexts.  Hence, its use in a policy evaluation is clearly
justified.  The DOl" also accepts the travel cost method as a best-available
procedure to estimate use values.  Economic values for different categories
can be summed to calculate total value, regardless of the extent to which the
conceptual bases for the different methodologies used to calculate the values,
or the values themselves, may vary.  In fact,  the DOI regulations for natural
resource damage assessment clearly state that "nothing in this section

-------
480    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

precludes the use of a combination of methodologies so long as the authorized
official does not double count ..." (43 CFR § 11.83  (c)(2)).  Care was taken
not to double count benefits for any category.

d.    Limitations of the Case Study Approach and Specific Case Study Critiques

      The benefits'analysis of the proposed Guidance is based on a case study
approach, using benefits transfer applied to three case study areas.  The case
study approach was used because it is more amenable to meaningful benefit-cost
analyses than are studies of larger aggregate areas.  Although the results
obtained for a case study site may not apply uniformly to the entire Great
Lakes Basin, the case study approach does provide policymakers with a
pragmatic and realistic perspective of how a proposed program can generate
benefits, the types of benefits anticipated, and how these benefits compare to
costs.

      The case studies were selected from a list of candidate sites
(designated Area's of Concern in the Great Lakes Basin) on the basis of data
availability and the relevance of the water quality problems to the Guidance
(i.e., areas in which problems were more likely to be associated with on-going
point source discharges rather than historic loadings from Superfund sites and
other sources).  Geographic diversity was also considered in selecting the
sites so that the analyses might better promote a broad perspective of the
Guidance's benefits and costs.

i.    Comments

      Commenters asserted that the benefits analysis is not representative of
region-wide benefits because it is based on case studies of the three "hot
spots." Further, it was suggested that contamination at the case study sites
is related to historic practices and not necessarily current discharges from
point sources.

      One commenter stated that the value of fishery enhancements are inflated
as ". .  . most of the fish caught and consumed are in  the unlimited
consumption category, and  .  .  . stressed fish compensate  in various ways to
maintain their population."  Another commenter asserted that the yellow perch
decline in the Fox River/Green Bay area was due to issues other than toxics,
such as overharvesting, and that the Green Bay wildlife sanctuary is not
suitably located to become "a haven for eagles, otters, minks, etc.  .  ."
Commenters stressed that benefits from copper removal  are "grossly
overstated."  Finally, a commenter asserted that the fish tissue concentration
data for PCBs and dieldrin is inaccurate.

ii.   Response

      An inherent limitation of the case study approach  is the inability to
extrapolate from a limited set of river-based sites to the Great Lakes Basin
as a whole.  The choice of three  of the basin's RAP areas was motivated by
data availability, as noted above.  RAP areas are typically  well studied,
thus, a wealth of relevant data are available.  Data  limitations usually
preclude conducting case studies  of less well known sites.   However,  there is
no reason to believe that  the  selected sites are not  reflective of  other  sites
in the basin.

       While RAP areas  are hot spots  and can be expected to  have a  higher
proportion  of potential benefits  (as  well  as  costs) associated with total
cleanup, other sites are expected to  have  a greater share of benefits
attributable  to  the Guidance.  This is because contamination at hot spots
typically results from  historic problems and highly contaminated  sediments
would not be  eliminated by the Guidance alone.  As a  result,  the potential
benefits attributable to the  Guidance for  hot  spots are  expected  to be  lower
than for other sites, everything  else constant.

-------
                           Section DC: Executive Order 12866                        481
      EPA investigated two additional case study sites for possible inclusion
in the benefits analysis:  (1) the Ashtabula River in Ohio and  (2) the St.
Louis River in Minnesota.  EPA determined that adding case studies would only
offer limited insights, because sites with readily available data have
profiles similar to the existing case studies  (e.g., large historic sediment
loads).  Instead, EPA conducted an analysis of the representativeness of the
case study sites. ' These results are presented in section IX.E.3.c.v.

      The April 1993 RIA notes that the yellow perch fishery declines in Fox
River/Green Bay are due to multiple causes and not dominated by Guidance-
relevant toxics.  The fishery values cited are reported as indicative of
fishery values in the basin and are not the primary basis for the benefit
estimate (which is largely based on a salmonid fishery study of Wisconsin's
Great Lakes waters)'.

      The assumption of a 50-percent reduction in pollutant loads was not
applied uniformly at each case study site; loadings reductions were estimated
for each case study relative to point source baselines, and evaluated based on
the relative toxicity of the contaminants.  The benefits estimates focused on
the toxicity-impacted salmonid fishery, specifically, the impact of current
consumption advisories on those fisheries.  Potential benefits to "unlimited
consumption" fisheries were not included (but may be appreciable).  Fisheries
benefits were based on an increase in consumer's surplus attributable to the
reduction of toxic .pollutants in fish; these benefits would accrue regardless
of whether stressed fish compensate for pollution to maintain their
populations.

      The wildlife benefits in Green Bay were not based on the sanctuary
becoming a "haven" for eagles and other wildlife, but on the potential
increased participation in nonconsumptive recreation such as nature viewing
resulting from ecologic improvements.  The current level of activity at the
sanctuary,  which is on the order of one million visitor days per year,
provides a benchmark basis for measuring potential increases in activity at
that site.

      EPA did not- estimate significant benefits from the removal of copper.
Copper reductions were estimated as incidental to controls of Tier I
pollutants, but the toxicity weighted loadings reductions place relatively
little weight on such reductions.  In fact, copper reductions accounted for
less than 0.6 percent of the toxic-weighted pollutant reductions in the Fox
River/Green Bay case study  (although higher percentages applied in other
sites).  Benefits from reduced loadings of copper were not monetized.

e.    Economic Impacts of the Guidance

      Economic impact analysis examines the direct and indirect effects of a
stimulus throughout an economy through an analysis of complex interindustry
linkages.   Spending by industries impacted by the Guidance can result in
positive impacts in the region by stimulating construction and/or output from
pollution control technology producers located in the basin.  However,
Guidance-imposed costs may also result in the closure of some plants and a
loss of spending by these firms and their employees in the local economy.  The
economic impact on the region must be evaluated by looking at both the
positive and negative impacts of the imposed costs.

i.    Comments

      Commenters suggested that the Guidance will require immense
expenditures, and that these costs will have drastic negative economic impacts
in the Great Lakes Basin such as decreased competitiveness; inhibited economic
growth, including the discouragement of efforts to expand production to
prerecession levels; a loss of markets and jobs; increased manufacturing
costs; and limits to existing zone businesses.  One commenter suggested that
the Guidance will "land a fatal blow" to already declining industries in the

-------
482    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

region such as steel, auto, paper, and. petroleum.  Commenters also indicated
that costs were greatly underestimated in the RIA, and that major costs were
omitted.  Generally, commenters believed that the Guidance would impose
enormous compliance costs- on businesses, municipalities, and taxpayers, which
will have'a negative impact on the quality of life in the Great Lakes Basin.

ii.   Response

      Estimating the impact of the Guidance on the economy of the Great Lakes
region requires a detailed econometric model of the region's economy.  An
econometric analysis was performed independent of the RIA for the Council of
Great Lakes Governors (The Great Lakes Water Quality Initiative: Cost
Effective Measures to Enhance Environmental Quality and Regional
Competitiveness.  DRI/McGraw-Hill, San Francisco, California, July 1993).
This analysis showed a nearly imperceptible impact of the Guidance on the
region's economy for a worst case scenario, a scenario with costs far
exceeding those estimated by EPA.  Manufacturing output was estimated to fall
by 0.008 percent to 0.337 percent over a range of four scenarios evaluated,
while personal income loss was estimated at between 0.002 percent and 0.094
percent for these scenarios.  As a result, the region was considered able to
"afford" the Guidance.

3.     Revised Benefits Estimates

      This section presents the additional research and analysis conducted by
EPA in response to the comments described above, and the revised case study
benefit-cost analyses. In addition, EPA updated the preliminary basinwide
human health risk assessment presented in the April 1993 RIA. These results
are also presented below.

a.     Introduction

      Based on the comments received on the benefits analysis of the proposed
Guidance, EPA conducted several analyses related to the attribution issue
which focused on better quantifying the potential impact of the Guidance in
bringing about future toxic-oriented benefits in the basin.  Specifically, EPA
reviewed studies and data related to quantifying the contribution point
sources make to total loadings of the relevant contaminants in the basin;
conducted a screening analysis to gain information on other potentially
significant sources of the contaminants in the basin; and modeled the relative
contribution of point sources to fish tissue contaminant concentrations  (for
one site and one pollutant).

      Inferences regarding the point source contribution to total loadings
from all sources  (e.g., sediments, atmospheric deposition) of Guidance-
impacted contaminants can be made from studies and data related  to the  inputs
of contaminants in the Great Lakes Basin.  A significant amount  of research
has been conducted on the mass balancing of contaminants in the  Great Lakes;
however, appreciable data gaps remain.  In general, this research indicates
that there is insufficient data available to estimate total loadings  (and thus
calculate the point source contribution) for almost all of the  contaminants
impacted by the Guidance, and that results are likely to be highly site- and
contaminant-specific.

      For example, assessing the  contribution of point  sources  to the toxic-
related problems in the basin requires an estimate of total basin loadings for
the relevant contaminants.  Of the 138 pollutants of initial focus for  the
Guidance, EPA had sufficient information to estimate basinwide  loadings for
only four of these: mercury, lead, cadmium, and  PCBs.   Even for these
contaminants, EPA considers the limitations of the studies used to develop the
estimates to be significant.  However, if these  results are utilized to
provide preliminary  indication of the relative contribution of  point  sources
 (based on Permit Compliance System data for point source  loadings),  the

-------
                           Section IX: Executive Order 12866                        483
estimated relative contribution of point sources ranges from approximately 2%
to 40%, depending on the contaminant.

      Research by Strachan and Eisenreich  (1988) indicates that the lower
Great Lakes receive a greater percentage of total loadings from point sources
than the upper Great Lakes.  The upper lakes are estimated to receive a much
greater fraction of their total loadings of toxic contaminants from
atmospheric sources than the lower lakes due to the relative lack of local
sources and the larger surface area of the upper lakes.  In comparison, the
lower lakes receive extensive loadings from sources on the Detroit-St. Clair
and Niagara River systems.  Strachan and Eisenreich developed estimates of the
relative contribution of atmospheric deposition to total loadings for PCBs,
DDT, benzo(a)pyrene, lead, mercury, and mirex.  Their results indicate that
the role of atmospheric deposition (and, thus, the role of other sources)
varies greatly by contaminant and lake.

      For example, atmospheric deposition accounts for 90 percent of PCB
loadings to Lake Superior but only 7 percent of PCB loadings to Lake Ontario;
for benzo(a)pyrene, atmospheric deposition is attributed with 96 percent of
loadings to Lake Superior, 80 percent of loadings to Lake Huron, and 72
percent of loadings to Lake Ontario.   Moreover, Strachan and Eisenreich's
results for PCBs indicated atmospheric deposition accounts for 58 percent of
PCB loadings to Lake Michigan, however, the results of the Green Bay Mass
Balance Study (GBMBS)  showed a different site-specific result on the lake.
The GBMBS,  an estimated $12 million dollar study, showed sediment to be the
dominant source of PCB loadings to Lake Michigan's Green Bay (over 90
percent).  For the relative contribution of point sources to total loadings of
lead and PCBs, Strachan and Eisenreich (1988)  report a range of 0.7 - 1.5
percent for Lake Superior, and a range of 2.0 - 7.0 percent for Lake Huron.
They suspect that their data underestimates point source loadings.

      Based on the above findings, EPA developed assumptions regarding the
relative contribution of point sources to total loadings for each lake.  For
Lake Superior, EPA assumed point sources account for between 1-2 percent of
total loadings.   For Lakes Michigan and Huron, EPA assumed point sources
account for between 5-10 percent of total loadings.  For Lakes Erie and
Ontario, EPA assumed point sources contribute 10-15 percent of total loadings.
These assumptions were used in the benefits analysis to attribute benefits to
the final Guidance.

      EPA also developed a generalized Great Lakes exposure and
bioaccumulation model to estimate changes in fish tissue concentrations in
response to model scenarios of reduced point source loadings and applied the
model to PCBs in Green Bay.   Reductions in point source loadings in Green Bay
were shown to have a modest impact on fish tissue PCB concentrations and
exceedences.  For example, a 50% reduction in point source loadings reduces
baseline exceedences from 10.6% to 8.9%.  This result occurs because for this
contaminant and this site, existing sediment contamination is the dominant
source of PCB exposure to fish (point sources contribute only 9.4% to
loadings),  and baseline exceedences are only 10.6%.  However, loading
reduction estimates for the Fox River/Green Bay case study area indicate that
the Guidance may reduce point source PCB loadings by 89.6%. A 90% reduction
lowers exceedences of the human health threshold from 10.6% to 6.8%.

      The model-estimated change in exceedences of the human health threshold
of 2 ppm (relative percentage reductions) in response to point source loading
reductions are 2%, '16%, and 36% under scenarios of 10%, 50%, and 90%
reductions in point source loadings,  respectively.  Application to other sites
and conditions may show greater benefits from point source reductions  (i.e.,
Green Bay is considered a "worst case" scenario).  Sites where point source
loadings represent a greater percent of total loadings are expected to show a
greater change from baseline conditions.  For example, under a scenario of a
50% reduction in point source loadings, the estimated reduction in exceedences

-------
484    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

may increase from about 16%  (with point sources contributing 9.4% of total
loadings) to about 35%  (where point sources contribute 20% of total loadings).

b.    Updated Risk -Assessments for Great Lakes Anglers

      Executive Order  (EO) 12898 established a presidential policy for
incorporating; environmental justice into Federal agency missions by directing
agencies to identify and address, as appropriate, disproportionately high and
adverse human health or environmental effects of its programs, policies, and
activities on minority populations and low-income populations.  In order to
assist in identifying the need for ensuring protection of populations who
principally rely on fish and/or wildlife for subsistence, the EO directs
agencies, whenever practicable and appropriate, to collect, maintain, and
analyze information on their consumption patterns and to communicate to the
public the risks of those consumption patterns.  In accordance with EO 12898,
EPA collected data and information on the consumption of Great Lakes Basin
fish to conduct risk assessments for two populations at risk:  Great Lakes
sport anglers (including minority and low-income anglers) and Native Americans
engaged in subsistence fishing in the basin.

i.    Sport Angler Risk Assessment

      EPA used data on fishing license sales in the basin to estimate the
number of potentially exposed recreational anglers.  Data from 1991 through
1993 (the most recent sales reports available for each state) indicated
approximately 2.69 million fishing licenses were sold to residents in Great
Lakes Basin counties.  To the extent that anglers purchase their licenses
outside the basin or share their catch with unlicensed family members, these
data may result in an underestimate of the potentially exposed population.

      EPA estimated the consumption of sport-caught fish based on a review of
the literature on licensed angler consumption patterns in the Great Lakes
Basin.  These studies reflected a range of fishing locations, and water
quality conditions that included the presence of fish consumption advisories.
Nonetheless, fish consumption was shown to be significant.  In addition, a
relationship between consumption levels and socio-economic characteristics was
found: minorities exhibited higher fish consumption levels than whites in the
basin; and the combination of minority status and relatively low income
 (annual income less than $25,000) resulted  in the higher consumption levels
 (West et al., 1993) .

      Using census data to- divide anglers by minority and income status, EPA
applied the average minority- and income-adjusted consumption levels found by
West et al.  (1993) (43.1 grams/person/day  (gpd) for low-income minorities and
11.7 gpd for other minorities) to sport anglers in counties in closest
proximity to the Great Lakes  (the contiguous lakeshore counties).  For the
remaining sport-angler population in the lakeshore counties and in the
remaining counties of  the basin, fish intake was assumed to be 16.7 gpd  (also
based on West et al.,  (1993)).  There are an estimated 92,000 low-income
minorities in lakeshore counties; 65,000 other minorities in lakeshore
counties; and 2.5 million other sport anglers in the basin.

      Health risks were calculated based on exposure to  chlordane, DDT,
dieldrin, hexachlorobenzene, mercury, PCBs,  2,3,7,8-TCDD, and toxaphene  in
fish tissue.  These chemicals were chosen based on their potential to  cause
adverse human health effects  (i.e., cancer  or disease) and the availability  of
information on fisH tissue contaminant concentrations.   Chemical-specific
toxicity factors were  drawn primarily from  EPA's Integrated Risk Information
System  (IRIS) ,- other sources were utilized  as needed.  Fish tissue contaminant
levels were estimated  by  lake, based on data from several  sources.  To the
extent that not  all fish may contain all contaminants at the  concentrations
shown, exposures and risks may be overestimated; however,  data were  only
available for a  small  portion of  the contaminants covered by  the Guidance,
which could result in  the  underestimation of risks.  Standard EPA assumptions

-------
                           Section IX: Executive Order 12866                        485
were used regarding length of residence  (i.e., 70 years; by convention) and
body weight  (70 kilograms).

      Baseline cancer risks for low-income minorities ranged from 2.5xlO"3
 (Lake Superior) to 1.2xlO'2 (Lake Michigan); for other minorities, baseline
cancer risks ranged from 6.5x10"* (Lake Superior)  to S.OxlO"3 (Lake Michigan);
and for all other sport fishermen, these risks ranged from 9.7x10"*  (Lake
Superior) to 4.5xlO"3 (Lake Michigan).  Baseline cancer risks were driven by
fish tissue PCB concentrations, which were highest in Lake Michigan and lowest
in Lake Superior.  For all lakes combined at baseline fish contaminant
concentrations, low-income minorities in lakeshore counties would have 10.1
potential excess cancer cases per year, other minorities in lakeshore counties
might expect 1.9 excess cancer cases per year, and for other sport  fishermen,
100.5 excess cancer cases per year might be observed.

      EPA calculated the potential reduction in baseline risk levels using
revised estimates of the reduction in point source contaminant loadings due to
the Guidance and the relative contribution of point sources to total loadings
in the basin.  As described above, the relative contribution of point sources
was estimated by lake, based on available data and EPA's analyses related to
the attribution issue (1-2 percent for Lake Superior; 5-10 percent  for Lakes
Michigan and Huron; and 10-15 percent for Lakes Erie and Ontario).  Estimated
point source loadings reductions due to the Guidance ranged from 68 percent
for chlordane to 0 percent for PCBs and dioxin.

      Table IX-7 presents the results of the sport angler risk assessment.
The Guidance is estimated to result in a reduction of annual excess lifetime
cases of 2.2 to 4.1 for low-income minorities in lakeshore counties; 0.4 to
0.8 for other minorities in lakeshore counties; and 21.9 to 41.9 for all other
sport anglers.  These results represent a total reduction of 24.6 to 46.8
excess lifetime cancer cases )potential cancer cases assuming a 70-year
lifetime exposure period).  On an annualized basis, these risk reductions
represent potential benefits attributable to the Guidance of between $0.7
million and $6.6 million per year  (based on the estimated value of  a
statistical life of between $2.0 million and $10.0 million).  Not all excess
cancer cases will necessarily result in mortality, therefore, the monetized
benefits estimate may be overstated.

       The estimated reductions in risks due to the Guidance are small since
baseline risks are driven by PCB exposure, and the modeled basinwide loadings
reductions show no reduction of PCBs due to the Guidance.  However, these
results are based on 59 sample facilities, and may result in conservative
estimates of actual basinwide reductions.  Indeed, estimated loadings
reductions for the Fox River and Green Bay and Saginaw River/Bay case study
areas, which are based on examination of all facilities in the areas, show an
89 percent reduction in PCBs from baseline levels.  Thus, the potential risk
reduction benefits of the Guidance may be underestimated at the basin level.

      For example, recalculating the above results using an average of the
estimated loadings reductions for the three case study areas instead of the
modeled basinwide reductions results in a greater estimate of benefits.  PCB
loadings are estimated to be reduced by 89.6% in the Fox River case study
area, 89.4% in the Saginaw Bay/River case study area, and by 0.0% in the Black
River case study area, giving an average PCB reduction of 59.7%.  Using this
result, and the resulting average reductions for the additional contaminants
included in the risk assessment, sport anglers are estimated to have a
potential reduction of 3.3 to 6.0 excess cancer cases per year.  This
corresponds to potential benefits of between $6.6 million and 60.1  million per
year.

-------
486      Water Quality Guidance for the Great Lakes System — Supplementary Information Document
                                                  TABLE K-7

                        POTENTIAL REDUCTION IN SPORT ANGLER EXCESS CANCER CASES
                             DUE TO THE GREAT LAKES WATER QUALrTY GUIDANCE
.
•
Low Income Minorities'
Other Minorities*
Other Sportfishermen
POST-GUIDANCE
INDIVIDUAL EXPECTED RISK
LEVEL*
REDUCTION IN CANCER CASES"
YEARLY CASES
LIFETIME CASES
LAKE ERIE
3.7 x 10-'
9.5 x 10"
1.4 x 10's
Total
0.01
0.00
0.03-0.04
0.04-0.05
0.27-0.41
0.04-0.07
1.80-2.80
2.11-3.28
LAKE HURON
Low Income Minorities'1
Other Minorities'
Other Sportfishermen
5.5 x 10-'
1.4 x ID'5
2.1 x 10"'
Total
0.00
0.00
0.04-0.08
0.04-0.08
LAKE MICHIGAN
Low Income Minorities'
Other Minorities'
Other Sportfishermen
1.2 x 10'2
3.0 x 10"s
4.5 x 10'J
Total
0.03-0.05
0.01
0.20-0.40
0.2*0.46
0.03-0.07
0.01-0.02
2.70-5.40
2.74-5.49

1.70-3.40
0.34-0.69
15.00-30.00
17.04-34.09
LAKE ONTARIO
Low Income Minorities'
Other Minorities'
Other Sportfishermen
7.5 x 10-'
1.9 x 10-'
2.9 x 10"'
Total
0.00
0.00
0.03-0.05
0.03-0.05
0.17-0.25
0.03-0.05
2.30-3.40
2.50-3.70
LAKE SUPERIOR
Low Income Minorities'
Other Minorities'
Other Sportfishermen
2.5 x 10"'
6.5 x 10"4
9.7 x 10"
Total
0.00
0.00
0.01
0.01
0.00-0.01
0.00
0.14-0.28
0.14-0.29
ALL LAKES
Total Low Income Minorities
Total Other Minorities
Total Other Sportfishermen
Grand Total All Lakes
0.04-0.07
0.01
0.31-0.58
0.36-0.66
2.17-4.14
0.42-0.83
21.90-41.90
24.50-46.90
* Based on estimated reductions in basinwide loadings which suggest that PCBs will not be reduced by the Guidance. Results for two case
studies, however, indicate that PGBs are reduced by 89% by the Guidance.
k Based on upper estimate of reduction in fish tissue concentrations.
* Range over estimated range of reduction in fish tissue concentrations.
' Lakeshore counties only.
Note: Numbers may not add to total due to rounding.

-------
                           Section IX: Executive Order 12866                        487
ii.   Native American Risk Assessment

      Another group of anglers in the basin, Native Americans, are not
required to buy fishing licenses and, therefore, would not be accounted for by
the sport angler risk assessment.  Native American tribes hold extensive
fishing rights in the basin and many practice a traditional subsistence
fishing lifestyle.1

      EPA researched the treaty rights of the Great Lakes1 Tribes and contacted
tribal representatives to confirm which tribes are engaged in subsistence
fishing.  Of the 38 tribes in the basin, 24 are currently engaged in
subsistence fishing.  The potentially exposed population for these tribes was
estimated to be 13,648 based on the on-reservation population reported in the
1990 census.  The tribes engaged in subsistence fishing are located in
Minnesota, Michigan, and Wisconsin; tribes located in New York are not
subsistence fishing.  For at least one of the New York tribes, the St. Regis
Mohawk, subsistence fishing has been curtailed by fish consumption advisories.

      EPA estimated consumption of fish by Native Americans based on a review
of the literature on fish consumption by Native Americans and subsistence
anglers.  Based on'these studies, low, moderate, and high estimates of fish
intake  (31.5, 57.8, and 140.0 gpd, respectively) were used to estimate risks.
Native Americans were assumed to reside permanently in the area.  Standard EPA
assumptions were used regarding body weight.

      Risks were estimated by lake to more accurately match fish tissue
concentrations with the exposed population.  The chemicals used to evaluate
risks were chlordane, DDT, dieldrin, hexachlorobenzene, mercury, PCBs,
2,3,7,8-TCDD, and toxaphene.  These chemicals were chosen based on their
potential to cause adverse human health effects  (i.e., cancer or disease) and
the availability of information on fish tissue contaminant concentrations.
Fish tissue concentrations were based on data from several sources which
indicated that individual fish were, in fact, contaminated with numerous
pollutants.

      EPA calculated lifetime cancer risk due to the ingestion of contaminated
fish at baseline conditions for the exposed population, using the low,
moderate, and high estimates of fish ingestion.  Average baseline cancer risks
for all tribes using these ingestion assumptions are 5.3xlO"3, 9.5xlO"3,  and
2.3xlO*2, respectively.  For the exposed population, these cancer risks
translate into potential excess cancer cases over a lifetime of 51.5, 93.4,
and 225.4, for low, moderate, and high estimates of fish ingestion,
respectively.  This is equivalent to 0.7, 1.3, and 3.2 excess cases per year,
respectively.  The greatest cancer risk is due to PCB contamination.

      EPA estimated the potential reduction in baseline risk levels in the
same manner as for sport anglers.  Table IX-8 presents the results for Native
Americans.  The Guidance was estimated to result in a reduction of excess
lifetime cancer cases of 0.1 to 0.3 for the low fish ingestion scenario; 0.2
to 0.5 for the moderate fish ingestion scenario; and 0.5 to 1.1 for the high
fish ingestion scenario.  On an annualized basis, these risk reductions
represent potential benefits of the Guidance of between $20,000 and $100,000
per year  (based on the estimated value of a statistical life of between $2.0
million and $10.0 million).  Not all excess cancer cases will necessarily
result in mortality, however, and as such, the monetized benefits estimate may
be overstated.

-------
488      Water Quality Guidance for the Great Lakes System - Supplementary Information Document

                                                  TABLE K-8

                       POTENTIAL REDUCTION IN NATIVE AMERICAN EXCESS CANCER CASES
                             DUE TO THE GREAT LAKES WATER QUALITY GUIDANCE
FISH CONSUMPTION SCENARIO
POST-GUIDANCE
INDIVIDUAL EXPECTED
RISK LEVEL"
REDUCTION IN CANCER CASES'
YEARLY CASES
LIFETIME CASES
LAKE MICHIGAN
Low (32 gpd)
Moderate (58 gpd)
High (140 gpd)
8.6 x 10s
1.6 x 10"2
3.8 x lO'2
0.00
0.00-0.01
0.01
0.11-0.23
0.20-0.41
0.49-0.98
LAKE SUPERIOR
Low (32 gpd)
Moderate (58 gpd)
High (140 gpd)
1.9 x 10-'
3.4 x 10-*
8.1 x 10-'
0.00
0.00
0.00
0.01-0.02
0.02-0.04
0.05-0.09
TOTAL
Low (32 gpd)
Moderate (58 gpd)
High (140 gpd)
0.00
0.00-0.01
0.01
0.12-0.25
0.22-0.45
0.54-1.10
* Based on estimated reductions in basinwide loadings which suggest that PCBs   will not be reduced by the Guidance. Results for two
case studies, however,  indicate that PCBs are reduced by 89% by the Guidance.
k Based on upper estimate of reduction in fish tissue concentrations.
° Range over estimated range of reduction in fish tissue concentrations.
Note: Numbers may not add to total due to rounding.

-------
                           Section IX: Executive Order 12866                        489
      As for the sport angler risk assessment, the potential risk reductions
due to the Guidance are small since baseline risks are driven by PCB exposure,
and the modeled basinwide results show no reduction in PCBs.  However, these
results are based on 59 sample facilities.  To the extent that the sample
underestimates potential basinwide reductions  (results for both the Fox River
and Green Bay and Saginaw River/Bay case study areas showed an 89 percent
reduction in PCBs from baseline levels), potential benefits of the Guidance
may be underestimated.  If the average loading reduction for the three case
study areas is used instead of the modeled basinwide results, the Guidance
would be attributed with a reduction in excess annual cancer cases ranging
from 0.01 to 0.03 for the low fish ingestion scenario and 0.06 to 0.11 for the
high fish ingestion scenario.

c.    Revised Case Study Benefit-Cost Analyses for the Final Guidance

      EPA revised the original case study benefits estimates to reflect its
revised assumptions regarding the relative contribution of point sources to
total basin loadings  (5-10 percent for the lower Fox River/Green Bay site, 10-
15 percent for the Saginaw River/Bay site, and 10-15 percent for the Black
River site) and revised estimates of the reduction in point source loadings
due to the Guidance.  In addition, based on the data and information collected
to conduct the basin wide risk assessments, human health risk reduction
benefits were calculated for the Fox River and Saginaw Bay case study areas.
These results are presented below in 1994  (first quarter) dollars.

i.    Fox River and Green Bav Case Study

      The final Guidance is expected to reduce toxic-weighted loadings in the
Fox River case study area by an estimated 28.2 percent, including significant
reductions in aluminum, benzo(a)pyrene, dieldrin, hexachlorobenzene, mercury,
and other chemicals.  Revised benefit categories for this case study include
human health, recreational fishing, nonconsumptive recreation (e.g., wildlife
viewing), commercial fishing, and nonuse  (e.g., intrinsic benefits placed on
natural resources).  EPA estimates that the annual potential benefits of the
Guidance will range from $27,000 to $3.8 million for recreational fishing,
$22,000 to $173,000 for nonconsumptive recreational use, $19,000 to $120,000
for commercial fishing, and $32,000 to $1.9 million for nonuse values.  In
addition, potential annual human health benefits of between $250,000 and $2.5
million are estimated to result from a reduction of between 0.12 and 0.25
excess cancer cases per year.  Thus, the annual potential benefits of the
Guidance associated with these categories total $349,000 to $8.5 million.

ii.   Saginaw River/Saginaw Bav Case Study

      The final Guidance is anticipated to reduce toxic-weighted loadings by
an estimated 60.5 percent in the Saginaw River case study area, including
significant reductions in aluminum, arsenic (III), DDT, lindane, mercury,
PCBs, and other chemicals.  Revised benefit categories include human health,
recreational fishing, nonconsumptive recreation, waterfowl and other hunting,
commercial fishing, and nonuse.  EPA estimates that potential annual benefits
of the Guidance are $60,000 to $4.7 million for recreational fishing, $8,000
to $66,000 for nonconsumptive uses, $2,000 to $11,000 for hunting, $7,000 to
$72,000 for commercial fishing, and $30,000 to $2.3 million for nonuse values.
In addition, potential annual human health benefits of between $60,000 and
$580,000 are estimated to result from a reduction of between 0.03 and 0.06
excess cancer cases per year.  The total potential benefits of the Guidance
associated with these categories range from $168,000 to $7.7 million.

iii.   Black River Case Study

      The final Guidance is anticipated to reduce toxic-weighted loadings by
an estimated 36.6. percent for the Black River case study area, including
significant reductions in fluoride, lindane, lead, and mercury.  Revised
benefit categories include recreational fishing; recreational boating,

-------
490    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

waterskiing, sailboarding, and swimming; and nonuse.  Annual benefits  of  the
Guidance are estimated to range from  $251,000 to $719,000 for  recreational
fishing, $33,000 to $67,000 for nonconsumptive water-based recreation,  and
$126,000 to $667,000 million for nonuse values.  Total annual  potential
benefits of the Guidance associated with these categories range  from $0.4
million to $1.5 million.

iv.   Comparison of Benefits and Costs for the Case Studies

      Two methods were used to compare the estimated case study  benefits  to
estimated compliance costs in the April 1993 RIA: l) a direct  comparison  of
annualized benefits and costs, and 2) a comparison of discounted benefits and
costs.  EPA discounted benefits and costs to incorporate a 10- and  20-year
phase-in of annual benefits and the present value of a stream  of annual costs.
Capital costs were .annualized using a 7 percent real interest  rate, and annual
costs and benefits were discounted by 3 percent each year.  A  comparison  of
the revised case study benefits and costs using these methods  is presented in
Table IX-9.

      Benefits ranges across case study areas are roughly comparable.   Annual
benefits range from approximately $200,000 to several million  dollars,
reflecting the uncertainty in the benefits estimates.  Annualized costs are
commensurate with annual benefits; costs are approximately $2-$3 million  per
year for each of the case studies.  The net present values of  streams  of
benefits and costs over 30 years are  also generally similar.

v.    Case Study Representativeness

      The representativeness of the case study sites was assessed by comparing
the percentage of total benefits estimated to accrue in the case study areas
to the percentage of basinwide costs  they will incur.  Benefits-related
measures (such as population, recreational angling days, and nonconsumptive
recreation days) were used in place of total benefits for this analysis
because there is no estimate of benefits for the entire Great  Lakes Basin.

      Overall, there is no evidence to suggest that the three  case  studies
reflect an unrepresentative level of  benefits relative to costs. The  three
case studies combine to account for nearly 14 percent of the Guidance  total
cost, nearly 17 percent of the total  loadings reductions, and  between  4
percent and 10 percent of the benefits proxies  (basin wide population,
recreational angling, etc.).  Thus, it may be that the three case studies
represent a reasonably proportionate  share of costs and benefits.

-------
             Section DC: Executive Order 12866
491
                       TABLE DC-9
COMPARISON OF THE POTENTIAL BENEFITS TO THE POTENTIAL COSTS
         OF THE GUIDANCE FOR THE CASE STUDY AREAS
         (MILLIONS OF 1994 [FIRST QUARTER] DOLLARS)
FOX RIVER CASE STUDY
Direct Annualized Comparison'
Discounted Benefits and Costs2-
10-Year Phase in of Benefits
20- Year Phase in of Benefits
SAOINAW RIVER CASE STUDY
Direct Annualized Comparison1
Discounted Benefits and Costs2
10- Year Phase in of Benefits
20-Year Phase in of Benefits
BLACK RIVER CASE STUDY
Direct Annualized Comparison1
Discounted Benefits and Costs2
10-Year Phase in of Benefits
20-Year Phase in of Benefits
BENEFITS
RANGE
$0.3 - $8.5

$5.4 -$133.9
$4.1 -$101 .4
BENEFITS
RANGE
$0.2 - $7.7

$2.6 - $120.9
$2.0 -$91. 5
BENEFITS
RANGE
$0.4 -$1.5

$6.4 - $22.7
$4.8 - $17.2
MIDPOINT OF BENEFITS
RANGE
$4.5

$69.7
$52.7
MIDPOINT OF BENEFITS
RANGE
$4.0

$61.7
$46.8
MIDPOINT OF BENEFITS
RANGE
$0.9

$14.5
$11.0
COSTS
$3.6

$71.8
$71.8
COSTS
$2.6

$53.0
$53.0
COSTS
$2.1

$42.7
$42.7
1 Based on annualized costs assuming a 10-year capital life and reflecting a 7 percent real interest rate on capital.
2 Present values (1994) over 30 years. Annualized costs (assuming a 10-year capital life and 7 percent real interest rate on
capital) and benefits are discounted at a 3 percent real discount rate.

-------
                          Section X: Regulatory Flexibility Act                      493
         X.  REGULATORY FLEXIBILITY  ACT


      Under the Regulatory Flexibility Act (RFA),  EPA generally is required to
conduct a final regulatory flexibility analysis (FRFA)  describing the impact
of the regulatory action on small entities as part of the  final rulemaking.
However, under section  605(b) of the RFA, if EPA certifies that the rule will
not have a significant  economic impact on a substantial number of small
entities, EPA is not  required to prepare a FRFA.

      Implementation  of the final Guidance is dependent upon future
promulgation of provisions consistent with it by State or  Tribal agencies or,
if necessary, EPA.  Until actions are taken to promulgate  and implement these
provisions (or equally  protective provisions consistent with the final
Guidance), there will be no economic effect of this rule on any entities,
large or small.   For  that reason, and pursuant to section  605(b) of the RFA,
EPA is certifying that  this rule itself will not have a significant economic
impact on a substantial number of small entities.

      Although EPA is certifying that this rule will not have a significant
economic impact on a  substantial number of small entities,  and therefore is
not required to prepare a FRFA, it is nevertheless including a discussion for
public information in the RIA of the possible economic effects to small
entities that could result, from State or Tribal implementation of the final
Guidance.  As discussed above, small facilities are projected to incur costs
of only approximately $500 per facility to comply with subsequently
promulgated requirements that are consistent with the final Guidance.
Accordingly,  EPA believes there will be no significant economic impact on a
substantial number of small entities as a result of State  or Tribal
implementation of the final Guidance.

-------
494     Water Quality Guidance for the Great Lakes System — Supplementary Information Document

-------
                          Section XI: Executive Order 12875                       495
 XI.   ENHANCING THE INTERGOVERNMENTAL
    PARTNERSHIP UNDER EXECUTIVE  ORDER
                                 12875
      In compliance with Executive Order 12875 (58 FR 58093, October 28,
1993) ,  EPA has  involved State,  Tribal, and local governments in the
development of  the final Guidance.

      As described in section II above, the core elements of the Guidance were
developed by the Great Lakes States, EPA, and other Federal agencies in open
dialogue with citizens, local governments, and industries in the Great Lakes
ecosystem over  a five-year period through the Great Lakes Water Quality
Initiative (GLWQI).  The Initiative process marks the first time that EPA has
developed a major rulemaking effort in the water program through a regional
public forum.   The process is described further in the preamble to the
proposed Guidance (58 FR 20820-23).

      In addition to the participation by State and local governments in the
initial development of the Guidance and in the public comment process, several
activities have been carried out since the publication of the proposed
Guidance.   These include:

      (1)   On April 26, 1994, EPA held a public meeting  to solicit additional
information from interested parties on the proposed Guidance.  As part of
EPA's outreach  efforts to State, Tribal and local governments, a special
invitation was  sent inviting elected officials and other State, Tribal and
local representatives to participate in the public meeting.  EPA specifically
welcomed Tribal and local officials and opened the floor to them to hear and
discuss their specific concerns and views on the final Guidance.

      (2)   A series of meetings and teleconferences were held with Great Lakes
States in early 1994 to discuss their comments on several issues, including
development of  water quality criteria, state adoption requirements, whole
effluent toxicity, bioaccumulation factors, additivity,  compliance schedules,
anti-backsliding, nonpoint sources, and international concerns.

      (3)   In October, 1994, EPA met with each individual State in the Great
Lakes basin to  discuss the nature, form, and scope of the draft Guidance, and
State concerns  with implementation of the provisions under consideration.  The
following issues were discussed at each of the meetings: intake credits,
antidegradation and existing effluent quality, wildlife  criteria, excluded
pollutants (ammonia and chlorine), elimination of mixing zones, site-specific
variances,  fish consumption, appropriate degrees of flexibility for
implementation  (e.g., guidance vs. regulation), and implementation procedures.

      (4)   In 1994 and 1995 EPA met with representatives of the National
Wildlife Federation to discuss EPA's activities in developing the final
Guidance in accordance with the terms of a consent decree governing the
schedule for development of the final Guidance.

      (5)   In 1994 EPA also met with elected officials and other
representatives from several local communities in the Great Lakes basin to
discuss issues  regarding the economic impact of the proposed Guidance on local
communities and publicly owned treatment works.  Issues  discussed include cost
impacts associated with implementing water quality criteria, methodologies,
and implementation procedures;  dealing with pollution from nonpoint sources;
public outreach to control pollutants such as mercury instead of costly end-

-------
496    Water Quality Guidance for the Great Lakes System — Supplementary Information Document

of-pipe.treatment; and applicability  of provisions  in  the  Guidance to the
National water quality program.

       (6)  EPA held an additional 18  consultations  with the  regulated
community throughout 1994.  Such meetings allowed representatives  of
dischargers to share additional data, which has been placed  in the docket for
this rulemaking, and concerns about a range of issues  the  dischargers expect
to arise in implementation of the Guidance, including  cost concerns.

       (7)  In 1994 EPA met with State representatives  to conduct  initial
planning for implementation of the GLI Clearinghouse.   All Great  Lakes States
agreed to participate in  this effort, which will involve sharing  of
toxicological and other data to assist in development  of additional water
quality criteria and values.

      The results of the  above efforts have assisted in development of the
final Guidance through broad communication with a full range of interested
parties, sharing of additional information, and incorporation of  features to
improve the implementation of the Guidance.

      EPA has estimated the total annual State government  burden  to implement
the Guidance as approximately 5,822 hours, resulting in a  State government
cost of $156,029 annually.  Such burden and costs were estimated  based upon
the burden and costs associated with  developing water  quality criteria,  review
of antidegradation policy demonstrations, reviewing approvable control
strategies and bioaccumulative chemicals of concern monitoring data, and
review of variance requests.  The total annual local government burden is
estimated to be 5,595 hours with an associated cost of $1,858,243.  All of the
burden and costs to local governments are associated with  being a regulated
entity as an operator of  a publicly owned treatment works  (POTW).

-------
                          Section XH: Paperwork Reduction Act                      497
         XH.  PAPERWORK REDUCTION ACT


      The information collection requirements  in this final Guidance have been
submitted for approval to OMB under the  Paperwork Reduction Act, 44 U.S.C.
3501 et sea.  EPA Has prepared an Information  Collection Request  (ICR)
document (ICR No. 1639.01) .   Copies of the  ICR may be obtained from the
address given at the beginning of this document.

      The public reporting and record keeping  burden for this regulation is
estimated to be 128,901 hours for the 3,795 permittees, or an average 34
hours.   The total annual burden to local governments as publicly owned
treatment works operators is estimated to be 45,296 hours.  The total annual
burden to State governments is estimated to be 5,886 hours.

      Send comments regarding the burden estimate or any other aspect of this
collection of information,  including suggestions for reducing this burden to
Chief,  Information Policy Branch,  Mail Code 2136, U.S. Environmental
Protection Agency,  401 M St.,  S.W.,  Washington, DC 20460; and to the Office of
Information and Regulatory Affairs,  Office  of  Management and Budget,
Washington, DC 20503.

      In this rulemaking EPA is also amending  the table of currently approved
ICR control numbers issued by OMB for various  regulations.  This amendment
updates the table to accurately display  those  information requirements
promulgated under the Clean Water Act.   The affected regulations are codified
at 40 CFR parts 122, 123, 131,  and 132.   EPA will continue to present OMB
control numbers in a consolidated table  format to be codified in 40 CFR 9 of
EPA's regulations,  and in each 40 CFR volume containing EPA regulations.  The
table lists the section numbers with reporting and recordkeeping requirements,
and the current OMB control numbers.  This  display of the OMB control numbers
and their subsequent codification in the Code  of Federal Regulations satisfies
the requirements of the Paperwork Reduction Act  (44 U.S.C. 3501 et seq.) and
OMB's implementing regulations at 5 CFR  part 1320.

      The ICR for this rulemaking was previously subject to public notice and
comment prior to OMB approval.   As a result, EPA finds that there is "good
cause"  under section 553(b)(B)  of the Administrative Procedure Act  (5 U.S.C.
553(b)(B))  to amend this table without prior notice and comment.  Due to the
technical nature of the table,  further notice  and comment would be
unnecessary.

      For further information on this amendment to 40 CFR 9, contact Sandy
Farmer (telephone 202-260-2740).
                                                     U.S. Environmental Protection Agency
                                                     Region 5, Library (PL-12J)
                                                     77 West Jackson Boulevard, 12th Floor
                                                     Chicago, IL  60604-3590

-------