NOTICE
This document has not been formally released by EPA and should not now be construed to represent Agency
policy. It is being circulated for comment on its technical accuracy and policy implications.
EPA-450/5-85-001
Inorganic Arsenic NESHAPS:
Response to Public Comments on Health,
Risk Assessment, and Risk Management
Strategies and Air Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
April 1985
-------
r iis report has been reviewed by the Strategies and Air Standards Division of the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names or commercial products
is not intended to constitute endorsement or recommendation for use. Copies of this report are available
through the Library Services Office (MD-35), U.S. Environmental Protection Agency , Research Triangle
Park, N.C. 27711, or from National Technical Information Services, 5285 Port Royal Road Springfield
Virginia 22161.
-------
TABLE OF CONTENTS
Chapter Page
Chapter 1 - INTRODUCTION AND LISTING OF DOCKET REFERENCES 1-1
Chapter 2 - HEALTH 2-l
Chapter 3 - LISTING OF ARSENIC 3_1
Chapter 4 - EXPOSURE AND RISK DETERMINATION 4-1
4.1 COMMENT SUMMARIES 4^
4.2 RESPONSE TO COMMENTS ON THE EXPOSURE AND RISK
ESTIMATION PROCEDURE 4_u
4.3 ADDITIONAL COMMENTS AND RESPONSES ON RISK
DETERMINATION AND MANAGEMENT ISSUES 4-19
Chapter 5 - PIECEMEAL APPROACH 5_!
Chapter 6 - EPA'S STATUTORY OBLIGATION UNDER SECTION 112 6-1
6.1 ACCEPTABLE RISK/AMPLE MARGIN OF SAFETY 6-1
6.2 BAT APPROACH 6_22
6.3 ECONOMICS AS A DECISION-MAKING CRITERION UNDER
SECTION 112 6_41
6.4 RECOMMENDED ACTION IN FACE OF UNCERTAINTY 6-44
6.b JOBS VS. HEALTH 6_51
6-6 OTHER 6_56
Chapter 7 - QUALITY OF LIFE 7-1
Chapter 8 - VICTIM COMPENSATION 8_!
-------
-------
1.0 Introduction
The Administrator has decided to regulate certain low-arsenic copper
smelters and glass manufacturing plants and not to regulate primary and
secondary lead smelters, primary zinc smelters, zinc oxide plants, cotton
gins and arsenic chemical plants. The EPA is publishing this document in
support of those decisions by providing detailed consideration and response
to comments received during the proposal/public comment process. In
particular, this document provides detailed responses to comments which are
related to the following general topics:
1. The listing of inorganic arsenic as a hazardous air pollutant
under section 112;
2. The health effects associated with arsenic exposure;
3. The risk management approach used as a basis for the proposal and
4. The risk assessment methodology.
Also, the A-gency has produced companion documents that contain other
background information and detailed responses to comments for the specific
source categories. The reader is referred to the following list for
complementary information:
1. Low-Arsenic Copper Smelters
2. Glass Manufacturing Plants
3. Primary and Secondary Lead Smelters, Cotton
Gins, Primary Zinc Smelters, Zinc Oxide
Plants, Arsenic Chemical Plant
EPA-450/3-83-010b
EPA-450/3-83-011b
EPA-450/5-85-002
In addition to the above documents, the reader is also referred to the
Agency's health effects document, "Health Assessment for Inorganic Arsenic,"
EPA-450/3-83-021F, from which many of the Agency's responses for comments in
Chapters 2 and 3 were drawn.
1-1
-------
LIST OF COMMENTERS ON PROPOSED .
COPPER SMELTER ARSENIC STANDARDS
Docket Item
Number
IY-D-1; IV -0-95; IY-D-677,
IV-F-10
IV -D -2; IV -0-3 7; IV-D-90
IV-D-3
IV -0-4
IV-6-5; IV-D-93; IV-D-530,
IY-D-673
IV-D-6
IV -0-7
IV -0-8
IV-D-9; IY-F-9
IV -0-10
IV-0-11; IY-D-127, IY-D-677
IV -0-12
IY-D-13
IV -0-14
IY-D-15
IV -0-16
IV-D-17
Commenter and Affiliation3
Susan and Robert Adams
Ms.
Mr.
Ms.
Ms.
Mr.
Ms.
Mr.
Mr.
Teresa Doyle
Hugh Kimball
Susan Anderson
Sheri Reder
Eugene Fuji mo to
Marilyn Muller
Craig D. Hi! born
John T. Konecki
Chris Connery and Mary Scott
Dr.
Mr.
Ms.
Mr.
Ms.
Mr.
Mr.
Robert E. Sul livan
Thomas M. Skarshaug et al.
Virginia Nichols
Philip H. Abel son
Nathall ie Fitzgerald
James J. Mason
T.C. White
IY-0-18; IV-0-19; IY-0-S9;
IV-0-64; IV-0-222; IV-0-445;
IV-D-602; IV-0-603; IV-D-620,
IV-0-621; IV-0-649; IV-0-691;
IV-0-702; IY-0-703; IY-0-714;
IY-0-716; IV-0-787, IV-0-792;
IV-0-793; IV-F-2b
ASARCO, Inc.
Mr. L. W. Lindquist
ASARCO, Inc.
1-2
-------
Docket Item
Number
Commenter and Affiliation*
IV-D-20
IY-D-21
IY-D-22
IY-D-23
IY-D-24; IY-D-136
IV-0-25; OAQPS-79-8/IY-D-3
IY-D-26
IV-0-27
IV-D-28
IY-D-29
IV-D-30; IV-D-283; IY-D-383
IV-0-31
IV-0-32; IV-0-677
IV-0-33
IV-0-34
IY-0-35; IV-D-593; IV-F-9
IV-D-36
IV-0-38; IV-F-10
IV-0-39
Mr. Duncan Berry
Mr. Terry Sullivan
Mr. Hans Zeisel
The University of Chicago
Law School
Mr. Hollis Day
Day's, Inc.
The Warnaco Group
Mr. Harvey S. Poll
Puget Sound Air Pollution Control
Agency
George and Adriana Hess
Mr. Steve Burcombe
Mr. Arnold Cogan
Cogan & Associates
Mr. Frank M. Parker, III
Southwest Occupational Health
Services, Inc.
Mr. John J., Sheehan
United Steelworkers of America
Mr. Edward S. Watts
Mrs. Delores Keating
Ms. Sharon Rue
Ms. Joy Nelsen
"A Concerned Citizen"
Mr. Ralph K. Garrison
Ms. Barbara Jensen
B. J. Kanagy
Ms. Elise Muller Lindgren
1-3
-------
Docket Item
Number
Commenter and Affiliation3
IV-0-40
IV-D-41 -
IV-D-42
IY-D-43; IV-0-114; IV-D-438
IV-0-44
IV-0-45
IV-0-4 6
IY-D-47
IY-D-48
IV-D-49; IV-D-375
IV-0-50
IY-D-51
IV-0-52
IY-D-53
IV -0-54
IY-D-55; IY-0-329; IY-0-687
IV-0-56
IV-D-57
IV-D-58; IV-0-253; IV-0-621;
IV-0-683
IV-0-60
IV-D-61
IV-0-62
IV-0-63; IY-0-435; IY-0-721;
IV-F-11
Ms. Patricia Ives
Ms. Rebecca L. Graves
Jam's and Gregory McElroy
Fred and Sue Campbell
David and Ann Beckwith Boberg
Ms. Susan Koneckl
Yernon and Christine Trevellyan
Erica and Michael Meade
Mr. Richard L. Swenscn
Ms. Elaine Taylor
Mr. Paul J. Braune
Ms. Hymen Diamond
Ms. Nancy Sosnove
Ms. Terry Patton
Ms. Patricia Bauer
Mr. E. Zahn
Mr. David Burcombe
Mr. Michael Higgins
Mr. Glenn L. Boggs.
Mr. Toby Holmes
Ms. Laurie E. Martin
Mr. and Mrs. Donald R. Jopp
Ms. Irene Blackford
1-4
-------
Docket Item
Number
Commenter and Affiliation3
IV-0-65
IV-0-66
IV-0-67
IV-0-68
IV-0-69
IV-D-70
IV-0-71
IV-0-72
IV-0-73; IY-D-105; IV-0-302:
IV-0-5 75
IY-0-74
IV-D-75
IV-0-76; IY-0-117; IV-D-443;
IV-D-757; IV-F-11
IV-0-77
IV-0-78
IV-D-79
IV-D-80; IY-D-677
IY-D-81; IY-D-121
IV-0-82
IY-0-83
IV-0-84; IY-D-677
IV-0-85
Ms. Ellen Ostern
Mr. David Parent
Mr. and Mrs. Leo A. Yuckert
Ms. Mildred Schiffor
Mr. R.W. Neuser
Mr. Douglas P. Coleman
Coland, Inc.
Mr. and Mrs. Al Booze
Ms. Olivia Watt
Mr. Robert Krimmel
Nancy Morgan and Michael Barnes
Mr. Noel Daley
Mr. Frank W. Jackson
Vashon-Maury Island Community
Council
Ms. Tammi L. Contris
Ms. Frances Wotton
Mr. and Mrs. Fuller
Ms. Caroline Hunter Davis
Mr. Robert Lipp
Mary and Stephen Daniel
Mr. Stanley C. Smith
Norene, Vince, and Patricia Gallo
Mr. Timothy Walsh
Greenpeace Northwest
1-5
-------
Docket Item
Number
IV-D-86
IV-D-87
IV-D-88; IY-D-109; IV-D-676;
IV-F-11
IV -0-89
IV-D-91
IV -0-92
IV-D-94
IV-D-96
IV-D-97
IV-D-98
IY-D-99
IV -0-100
IV -0-101
IV-D-102
IV-D-103; IY-D-111
IY-0-104; IV-D-677
IV-D-106; IV-0-677
IV -0-107
IV-D-108; IV-0-589
IV -0-110
IV-D-112
IV -0-113
IV-D-115; IV-0-429
IY-D-116; IV-0-433
Commenter and Affiliation3
Ms.
Mr.
Ms.
Mr.
Ms.
Mr.
Ms.
Ms.
Mr.
Ms.
Ms.
Ms.
.Ms.
"Ms.
Ms.
Ms.
Mr.
Mary Lane
William Breitenbach
Diane Harris
Michael Maskule
Harriet Strasberg
J. Brady
Cheryl Owings
Deborah J. Mills
G. R. Finden
Laura H. Vaughn
Gertrude Quinn
Mona Brady
Rose Owens
Carol Howell
Dana Larson
Dorothy J. Sivertson
Scott Sruly
Terry Graves
Mr.
Ms.
Ms.
Percy W. Lewis
Pat Burke Tischler
Sandra Ellis
Katharine and Theodore Kowalski
Ms.
Rev.
Torn* Beckman
Merry Kogut
1-6
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-118, IV-0-126; IV-F-11
IV-0-119; IV-D-446; IY-D-648;
IV-D-710, 710a, 710b; IV-D-745;
IV-D-749; IV-0-759; IV-F-2&;
OAQPS-79-8/IY-D-33, 33a, 33b
IV-D-120; IV-D-621
IV-0-122; IV-D-723
IY-D-123
IV-D-124; IY-D-670
IV-D-125
IY-D-127
IV-0-128
IY-D-129
IV-0-130
IV-D-131
IV-D-132
IY-0-133; IY-D-485; IV-D-621
IV-0-134
IY-D-135
IY-D-137
IY-0-138
IY-0-139
Dr. Ruth Weiner
Sierra Club, Cascade Chapter
Mr. David D. Doniger
Natural Resources Defense
Council, Inc.
Dr. Gilbert S. Omenn
University of Washington
School of Public Health and
Community Medicine
Ms. Rose Orr
Ms. Gail L. Warden
Group Health Cooperative of
Puget Sound
Mr. Ted Dzielak
Greenpeace Northwest
Mr. Phillip A.M. Hawley
Robert and Petra Sullivan
Mr. C.R. Myrick
Ms. Dana Griffin
Mrs. S.C. Sandize
Ms. Kathleen Hobaugh
Mrs. 6.R. Byrski
Mr. Russell I. Lewis
Ms. Jenny Binder
Mrs. John E. Erickson
Mr. Gene Alberts
Pacific Sun Ltd.
George and Norma Newcomb
Ms. Sue Hanson
1-7
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-140; OAQPS-79-8/IV-D-6
IY-D-141; OAQPS-79-8/IY-D-7
IV-D-142
IV-D-143
IV-D-144; IV-D-719
IV-0-145
IV-D-146
IY-D-147
IV-D-148; IV-0-667
IV-0-149; IV-D-621;
OAQPS-79-8/IV-0-2
IV-D-150
IV-D-151; OAQPS-79-8/
IV-0-10
IV-0-152
IV-0-153
IV-0-154
IV-0-155
IV-D-156
IY:D-157
IV-D-158
Mr. J.W. George
Tennessee Chemical Company
Mr. David C. Roberts ,
Mr. Del Langbauer
Ms. Diane Kay Davis
Mr. Noel McLane
Mr. Paul F.'Munn
City of Toledo
Dept. of Public Utilities
Mr. Jeffrey P. Davis
Ms. Johanna H. Mason
Mr. Joe Geier
Mr. Douglas Frost, Ph.D.
Dr. Douglas A. Smith
Walter and Dorothy Pelech
Ms. Leah Quesenberry
Mr. Bill Stewart
Mr. R.J. Kirrage
National Blower & Sheet Metal Company
Mr. Peter K. Schoening
Chemical Proof Corporation
Mr. C. W. Bledsoe
Canal Industrial Supply Company
Mr. Richard B. Barrueto
Carl F. Miller & Company
Frank and Deborah Jackson
1-8
-------
Docket Item
Number
IV -0-159
IY-D-160; IV-D-316; IV-D-453;
IV -0-5 77; IV-D-658i IV -0-695
IV-D-161
IY-D-162
IV-D-163
IV-D-164, IV-0-666
IV-D-165
IY-0-166
IV-D-167
IY-D-168
IY-0-169
IY-0-170
IY-D-171
IV -0-172
IV-D-173
IY-0-174
IV-D-175
IY-D-176
IY-0-177
IV-0-178
IY-0-179; IV-0-621
IV -0-180
IV -0-181
Commenter and Affiliation3 .
Ms.
Mr.
Mr.
Mrs.
Mr.
Ms'.
Ms.
Ms.
Mr.
Ms.
Ms.
Mr.
Mr.
Mr.
Mr.
Avel
Mr.
Mr.
Mr.
Mr.
Mr.
Mr.
Mr.
Paula Bond
Robert Bloom
David Hakala
, Richard Tallman
Thomas Jay Allen
Mary G.L. Shackelford
Mildred E. Blandford
Elsie Wood
Donald E. White
Claudia Hurd
B.J. Hartman
Ralph Brock
Charles E. Hochmuth
John F. Mattes
Richard L. Barney
ino and Amelita Soareuas
and Mrs. George Kahl
Harold T. Rock
David Walkup
Raymond Garner
Owen T. Gallagher
Joe E. Bartosch
Richard Balles
1-9
-------
Docket Item J
Number
Commenter and Affiliation3
IV-D-182
IY-0-183
IY-D-184
IV-D-185
IV-D-186; IV-D-3S2
IV-D-187
IY-D-188
IY-D-189
IY-0-190
IV-D-191
IY-D-192
IY-0-193
IY-0-194
IV-D-195
IV-0-196
IY-D-197
IV-D-19 8
IY-D-199
IV-0-200
IY-0-201
IV-0-202
IY-D-203
IV-0-204
IY-0-205
Mr. Al Cook
Mr. Eric Zeikel
Mr. Ben R. Petrie
Ms. Mary LaPlant
Mr. Lee R. Carl
Mr. Stanton Neut
Mr. Stephen J. Romanovich
Mr. and Mrs. Dennis F. Keating
Mr. Glenn E. Enzler
Mr. and Mrs. Richard Rader
Mr. Maurice C. Killenbeck
Mrs. L.G. Tallman
Mr. Warren Mattson
Mr. Marion Beach
Mr. Robert L. Sprague
Mr. R. Andress
Clarence and Lorene Borell
Mr. D.L. Bean
Mr. Robert D. Hughes
Mr. and Mrs. Roy Mybeck
Mr. John Fuller
Mr. Norman 0. Bond
Mrs. C.W. Koski
Mr. Gerald E. Johnson
1-10
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-206
IV-D-207
IY-D-208
IV-D-209
IY-D-210
IY-D-211 :
IV-D-212
IV-D-213
IY-D-214
IV-D-215
IV-0-216
IV-D-217
IV-D-218
IV-D-219
IV-0-220
IV-0-221
IV-D-223
IV-D-224; OAQPS-79-8/IV-D-11
IY-0-225
IV-0-226
IV-0-227; IV-D-621
IV-0-228
IV-0-229
Mr. Kenneth R. Leffler
Mr. Emil H. Novis
-Mr. Robert A. Bowman
Ms. Shirley Welch-
Mr. and Mrs. Jay Hensley
Mai, Van Nguyen
Mr. Harold E. Jorgenson
Mr. John Bentson Vale
Mr. Ron Streich
Streich Bros. Engineering
Mr. Arthur J. Dunaway
Minnie and Al Greco
Doug and Kristy Funkley
Mr. Joseph Udovich
Mr. Robert Zimmerman
Mr. William Lobeda
Mr. Bill D. Roumel
Mr. Arnold Kese
Ms. Karen S. Kamp
Mr. Ben H. Roseberry
Mr. Daniel S. Dean
Mr. John C. Larsen
Mr. and Mrs. Pete McDonell
Mr. Harry D. Maxwell
1-11
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-230
IY-D-231
IY-D-232
IY-D-233
IY-D-234
IY-D-235
IY-D-236
IY-D-237
IV-D-238
IV -0-239
IY-D-240
IV-D-241; OAQPS-79-8/IY-0-12
IY-D-242
IY-D-243
IY-D-244
IV-0-245
IV-0-246
IV-D-247
IY-0-248
IV-D-249
IY-0-250
IY-D-251
Mrs. Matt Gunovich
Mr. Adam S. Kreisman
Mr. B.K. Arnberg, Jr.
Mr. Alfred N. Johnson
Mr. Henry Cox
Mr. Homer T. Brown
Ross and Mildred Rice
Mr. Joseph M. S tad tier
Mr. Robert F. Sylvanus
Mr. Wallace H. Larson
Mr. Art Alsos
Carl T. Madsen, Inc.
Ms. Alice Spears
Ms. Adah Green
Mr. Charles E. Allen
Mr. Robert Z. Primm
Candid Photo Service, Inc.
Ms. Kathleen M. Brainerd
Mr. Raymond R. Webster
Mr. F. Mil lard White
Thomas and Rosemary Arnold
Willis and Edith Powers
Mr. & Mrs. Arthur Keug
Ms. Eleanor Schaffer
1-12
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-252
IV-0-254
IV-D-2S5; IY-D-337
IY-D-256
IY-D-257
IV-0-258
IV-D-259
IV-D-260
IV-0-261
IV-0-262
IV-0-263
IV-D-264
IV-0-265
IV-D-266
IY-D-267
IV-D-268; IV-D-518
IY-0-269
IV-0-270
IY-0-271
IY-D-272
IV-0-273
IV-0-274
IV-0-275
Mr. Frank C. Hansen
Unico Service & Engineering
Mrs. Lorette Prettyman
Mr. Richard Tallman
P.J. Dougherty
Mr. Edward R. Kiehlmeier
Ms. A Ha F. 'Hyde
Mr. Ernest. Cooper
Mr. Charles Mattheson
Mrs. Ellen Manweiler
R.D. Gallagher
Richman and Forestbyne McNeil
Mrs. Joe Sunich
Mr. Ed Michalski
Ms. Luvina Johnson
Mr. Michael Mclntyre
Mr. John Henderson
Mr. Michael Evans
Mr. Doss Bridges
A. P. Konick
Ms. Mae Brown
Mr. Lowell Jorgenson
Mr. Paul DiMaio
Frank and Del ores Keating
1-13
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-276
IY-D-277
IY-D-278
IV-D-279
IV-D-280
IY-0-281
IV-D-282
IV-D-284
IY-D-285
IV-0-286
IV-D-287
IY-0-288
IV-0-289
IY-D-290
IY-D-291
IV-D-292; IY-D-582, IY-0-668
IY-0-293
IV-0-294
IV-0-295
IV-D-296
E.M. Krisman
Mr. Jack Stutler
Erwin and Patricia Myers
K.S. Hammond
Mrs. F.M. Larson
Mr. Frank Diane
Florence Irvin and John Jurovich
Mr. William Dearborn
Mr. Leon Cunningham
Mr. Richard Lowery
Electric Motor Service Co.
Mr. Fred Young
E. A. Wilcox Co.
Mr. Kenneth Sprong
' Harbison-Walker Refractories
Mr. C.M. Bevis
Bevis & Assoc., Inc.
Mr. Laurence Evoy
Pierce County Medical
Mr. George Leonhard
Mount Rainier Council
Boy Scouts of America
Mr. Mike Cooney
Mr. Joseph Prinse
Mr. Lee Fedderly
Ms. Marge Kunschak
Mr. John Vipond
Girard Wood Products
1-14
-------
Docket Item
Number
Commenter and Affiliation3
IV-0-297
IY-D-298
IY-D-299
IV-D-300
IY-D-301; OAQPS-79-8/IY-D-13
IV-D-303
IV-D-304
IV-D-305
IV-0-306
IV-D-307
IV-D-308
IV-D-309
IV-D-310
IV-0-311
IV-0-312
IY-D-313
IV-0-314
IY-D-315
IY-0-317
IV-0-318; IV-D-621
IY-D-319; IV-0-621
IY-0-32Q
Mr. Kenneth Griswold
Mr. Robert Laughlin
Mr. Walter Ivey
Mr. William Taylor
Flohr Metal Fabricators
Ms. Sally Davidson
Mr. R.E. Wendlandt
Reliable Steel Fabricators
Ms. Roxie Skidmore
Mr. William Leonard
Mrs. Robert Schanzenbach
Mr. Hugh Williamson
Pierce County Medical
Mr. H. Eugene Quinn
Mr. B.W. Truswell
Wenatehee Silica Products, Inc.
Mr. Don Zemek
Mr. Justice Ashwell
Harold and Anne Ransom
Dr. and Mrs. Robert Knapp
Dr. Richard G. Schoen
Herbert and Charlotte Weston
Mr. John Susanj
Mr. Coy Brown
Mr. Bill Weston
Mr. and Mrs. John Reed
1-15
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-321
IY-D-322
IV-D-323
IY-D-324
IV-D-325
IY-D-326
IV-D-327
IV-D-328
IV-0-330
IV-D-331
IV-D-332
IV-0-333
IV-D-334
IY-D-335
IY-D-336
IV-0-338
IY-D-339
IY-D-340
IV-D-341
IY-0-342
IV-0-343
IV-0-344
IY-0-345
IY-0-346; OAQPS-79-8/IY-0-14
1-16
Ms. Ruth Brown
Mr. George Austin
Austin Mac, Inc.
Mrs. Ivy Blackburn
Mrs. Robert Kling
Malcolm and Laurel Ross
Mr. Floyd Martin
Mrs. Elizabeth Pedersen
Ms. Laure Nichols
Mr. John Dyer
Mr. Kenneth Taylor
Mr. and Mrs. Fredrick Young
Mrs. Robert Guddes
Charles and Thelma Modie
Ms. Mary L. Mull in
Mr.1John Daly
Mr. Arlander Bell
Mr. Walter Kunschak
Mr. Donald Angle
Pete and June Zaferin
Mr. Allan Weydahl
Nalco Chemical Co.
Ms. Greta Dotson
Mr. Charles Shaw
Mr. Frank Puz
Ms. Shermaine Celine
-------
Docket Item
Number
Commenter and Affiliation3
IY-D-347
IV-0-348
IV-D-349
IV-D-350
IV-D-351
IV-0-353
IY-0-354
IV-D-355
IV-D-356
IV-0-357
IV-0-358
IV-0-359
IV-0-360
IV-D-361
IV-D-362
IV-0-363
IV-0-364
IV-0-365
IV-0-366
IV-D-367
IV-D-368
IY-0-369
IV-0-370
';:*'
Mr. Warren Harvey
Mr. Frank Storizic
W. Phelps
Mrs. Chris Mortensen
Mr. Robert Ellener
Ms. Ella Phillips
Mrs. Marjorie McMenamin
Patrick and Nora Duggan
Ms. Mary McCormack
Mr. and Mrs. Ervin Lee
Mr. J.M. Will
Tarn Engineering Corp.
Mr. S. Evan Davies
S. Evan Davies & Associates
Ms. Betty J. Roberts
Mr. and Mrs. Garland Cox
Ms. Janet Jacobson
Ms. Frances Coats
Ms. Ellen Herigstad
Mr. Fred Wise
D.M. Manning
Mr. and Mrs. W. Rieck
Ms. Olga Williams
Mr. Bill Merrill
Mr. and Mrs. Ray Lunger
1-17
-------
Docket Item
Number
IV-D-371
IV-D-372
IY-D-373
IV-D-374
IY-D-376
IV-D-377
IV-0-378
IV-D-379
IV-D-380
IY-D-381
IV-D-382; IY-D-621
IY-D-384
IV-0-385
IY-D-386
IV-D-387
IV-D-388
IV-D-389
IY-D-39QC
IV-D-391
IV-0-392
IY-D-393
IV-D-394
Commenter and Affiliation3
Mr. William Hanar
Mr. Albert DiLoreto, Sr.
Anne and Grant Whitley
Mr. Louis Burkey
Mr. Gerald Copp
Public Utility District #1
of Chelan County
Ms. Eva Malovich
S. Behrman
Mr. Raymond .Wall
Mr. Jim Wllhelral, Jr.
The Stationers, Inc.
Mr. George Jewell
Mr. Floyd Williams
Mr. Michael Fabb
Mr. Ken Reaj
Mr. William Cammarano, Jr.
Cammarano Bros, Inc.
Mr. M.J. Burgess
Mr. D.S. Skeie
Industrial Mineral Products, Inc.
Mr. P. McOougal
Mr. Victor Selvig
G.O. Shipley
Robert and Jan Van de Mark
G.S. and Bernice Tallman
Ms. Mildred Wall
1-18
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-395
IY-D-396
IV-D-397
IY-0-398
IV-0-399
IV-D-400
IV-0-401
IV-D-402
IV-0-403
IV-0-404
IV-0-405; IV-D-621
IV-D-406
IV-0-407
IV-0-408
IV-0-409
IV-D-410
IY-0-411
IV-D-412
IV-0-413; IY-0-677
IY-0-414; IV-D-677
IV-0-415
IY-D-416
IV-0-417; IV-F-9
Mr. James Jacobsen
M.C. Teats
Mrs. June Gil son
H.C. Bauman
Mr. Ronald Roman
Virginia and John Weaver
Mr. Manfred Bell
Mr. Edwin Briggs
Mr. David Griffiths
Cornell, Weinstein & Griffiths
Ms. Kathryn Keller
Mr. Theodore Kennard
B.A. McKenzie & Co.
A.J. and Emily Charap
Mrs. Edna Carlson
S. Mladervich
Mr. Glenn Roberts
Mr. Frank D. Pupo
Sam's Tire Service
Ms. Carol Van Ginhoven
Mr. Lloyd Skinner
Ms. Helen Gabel
Mr. Phillip Notermann
Mr. Charles Wie
Mr. Charles W. Olsen, Jr.
Mr. James Garrison
1-19
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-418
IV-D-419
IY-D-420
IY-D-421
IV-D-422; IV-D-584
IY-D-423
IY-D-424
IV-D-425
IY-D-426
IY-0-427
IY-0-428
IY-D-430
IY-D-431
IY-D-432
IY-D-434
IY-D-436
IV-0-437
IV-D-439; IV-D-662; IY-D-676
IV-0-440
IY-0-441; IV-D-664; IY-D-676
IV-0-442
IV-0-443
Mr. F. Andrew Bartels
Mr. Philip Yolker
Ms. Patricia Howard
Ms. Marianne Edsen
Demelza Costa, e_t al.
Robert and Elnora Turver
Mrs. Cheryl Curtis
Mr. and Mrs. Harold Feley
Walt and Kathy Hansen
Rev. John Keliner
Old St. Peter's Church
Mr. Robert Burns
Oleta Kerns
Mr. Jon Fayst
Mr. John Ellingson
M. J. Bunnell
Mr. G. Patrick Healy
Ms. Joan Peterson
Margie and Jeff Goulden
Devitt and Debby Barnett
Dr. John Van Ginhoven
Mrs. Ray Hund
Jeanne Snell and Frank Jackson
Yashon-Maury Island Community
Council
1-20
-------
Docket Item
Number
Commenter and Affiliation3
IV-Q-444
IV-0-447; IY-D-786
IY-D-448
IV-D-449; IY-0-620;
IV-0-621; IV-F-2b
IV-D-450
IV-0-451
IY-D-452
IV-D-454
IY-0-455
IV-D-456
IV-0-457
IY-D-458
IV-0-459
IV-0-460
IV-D-461
IV-D-462
IV-0-463
IY-0-464
IV-0-465
IY-0-466
IY-0-467
Mr. David A. Frew
Mr. Stephen Cant
State of Washington Dept. of
Labor & Industries
Ms. Anita Fries
Ohio State Clearinghouse
Mr. Donald Robbins
ASARCO, Inc.
Mr. Ron Johnson
s
Mr. Marion Brannon
Ms. Cora Tolstrup
Mr. Wayne Yanderflute
Mr. F. Steven Doman
Mr. Mark Peterson
Mr. Robert Daniel
Pat Frostad
Motors & Controls Corp.
Mr. Robert Lawson
Mr. William Scott
Mr. Bailey Nieder
Tacoma Steel Supply
Mr. Hugh Wild
Ms. Elaine Thomas-Sherman
Mr. Sidney Peyton
Mr. Paul Foslien
Mr. Sam Smyth
Mr. Bill Cope
1-21
-------
Docket Item
Number
Commenter and Affil1ationa
IY-D-468
IV-D-469
IY-D-470
IY-D-471
IY-D-472
IV-D-473
IV-D-474
IY-D-475
IY-D-476
IY-D-477
IV-D-478
IV-0-479
IV-D-480
IV-0-481
IV-0-482
IY-D-483
IY-D-484
IY-D-486
IV-0-487
IV-0-488
IV-D-489
IV-0-490
IY-D-491
Mr. Albert Behar
Pierce County Medical
Ms. Sheila McCanta
Mr. Edgar E. King
Ms. Mary Chouinard
Rose and Floyd Murphy
Mr. .Russell Johnson
Ms. Helen Carnahan
Ms. Lucille Olsen
Beatrice and George Peterson
Mr. and Mrs. Carroll Thompson
Ms. Norma Rozmen
Ms. Marian Ganz
Mr. John Gaul
Ms. Molly LeMay
Mr. Joseph Petranovich
Mr. Rohn Burgess
Mr. Jack McGuirk
Mr. John Watson
Mrs. Georgann Gallagher
Ms. Alvinia Hagen
Mr. C. Mark Smith
Tacoma-Pierce County Economic
Development Board
Mrs. 'Virginia Loomis
Delmer Pitts
1-22
-------
Docket Item
Number
IV -0-492
IV -D -493
IV -D -494
IV-D-495
IV -0-496
IV -0-49 7
IV -D -498
IV -0-499
IY-D-500
IV-D-501
IY-D-502
IV-0-503
IV-D-504
IY-D-505
IV-D-506
IV-D-507
IV -0-508
IV-0-509
IY-0-510
IY-0-S11
IV-0-S12
Commenter and Affiliation3
Mr. Robert Heaton
Dr. Michael J. Jarvis
Mr. Kenneth J. Haagen
Mr. E.P. Stiles
Pierce County Medical Bureau, Inc
Beverly and Lawrence Sawtelle
Mr. and Mrs. K.W. Mueller
Ms. Frances Johnson
Inte.rAcc Co.
P. Fischer
Ms. Betty M. Susan
Mr. and Mrs. Duane Puyear
Ms. Marie Bean
Mr. Thomas G. Stoefae
Mr. Malcolm N. Thompson
United Steelworkers of America
Local 25
Ms. Doris Adams
Smelterman's Federal Credit Union
Mr. John Fink
Mr. Wayne Harkness
Herb and Shirley Godfrey
Mr. and Mrs. A.R. Glenn
Mr. Donald S. Leinum
Mr. Paul A. Schulz
Gary and Nancy Ackman
1-23
-------
Docket Item
Number
IY-D-513
IY-D-514
IY-D-515
IY-0-516
IY-D-517
IV-D-519
IV-D-520; IY-F-9
IY-0-521
IV-D-522
IY-D-523 ' -
IY-D-524; IV-0-554; IY-D-660
IY-D-525
IV-D-526
IY-D-527
IY-D-528
IY-D-529
IY-D-531
IV-D-532
IV -0-533
IY-D-534
IV-0-535
IY-D-536
Commenter and Affiliation3
Mr. Bailey Nieder
Columbia Energy Co., Inc.
Mr. E.T. McGrath
Ms. Beverly M. Migliore
Brown University
Department of Geological Sciences
Mr. Fred H. Smith
Cochrane Northwest, Inc.
Ms. Margaret J. Rowan
Mr. Robert R. Treanton
Pick Foundry Co.
Ms. Rayna Holtz
James and Jerry Brandfas
Mr. Jerry Michael Carlson
Mr. Wayne S. Moen
Mr. Richard L. Franklin
Mrs. E. Gerie Fortier
Ms. Cheryl Kirkwold
Mr. James D. Gray
Mr. and Mrs. Al Wegleitner
Ms. Carol A. Krona
John and Doris Achman
Mr. Robert D. Hall
Mr. and Mrs. W.H. Buzzell
Ms. Ruth M. Johnson
Mr. Howard 0. Huggard
Mr. Kenneth Merrsching and Family
1-24
-------
Docket Item
Number
Commenter and Affiliation3
IY-D-537
IV-D-538
IV-0-539
IV-D-540
IV-D-541
IV-D-542
IV-0-543
IV-D-544
IV-D-545; IV-0-621
IY-D-546
IY-D-547
IV-0-548
IV-0-549, OAQPS-79-8/IV-0-15
IV-D-550
IV-0-551
IV-0-552
IV-0-553
IV-D-555
IV-0-556
IY-D-557
IV-0-558
IV-0-559
Mr. Robert D. Budd
Mr. Gregory B. Curwen
Gierke, Curwen, Metzler & Bobrick
Mr. and Mrs. Richard Perkins
Mr. R.M. Kennard et aj.
Mr. T. Russell Mager
Ronald and JoAnn Roberts
Mr.'and Mrs. Austi-n £. Atwood
Ms. Ruby M. Martin
Mr. Clyde H. Hupp
Pierce County Central Labor Council
AFL-CIO
Mr. Mike D. Perkins
Don H. .Perkins, Inc.
Mrs. Leonard Berglund
Mr. Marion W. Samuel son
i
Mr. Kenny Scott
Mr. W.E. Lightfoot
Coffman Engineers, Inc.
Mr. Robert Reinhart
Mr. Robert F. Griffith
Mr. W.A. Palmer
Mr. and Mrs. Clifford Lakin
Ms. Stephanie Colony
Mr. Don H. Hinkley
Mrs. Allan Lindstrom
Mr. Bob L. Marshall
1-25
-------
Docket Item
Number
Commenter and Affiliation3
IV-0-560
IV-D-561
IV-D-562
IY-D-563
IV-D-564 . ;
IY-D-565
IY-D-566
IV-D-S67
IV-D-568
IV-D-569
IY-D-570
IY-D-571
IY-D-572
IY-D-573; OAQPS-79-8/IY-D-17
IY-0-574
IV-0-576; IY-D-699
IV-D-578
IY-D-579; IV-F-9
IY-D-580
IV-D-581
IV-0-583
Mr. Kim de Rubertis
Mr. A.B. Berg.
Industrial Mineral Products, Inc.
Mr. David A. Pitts
Mr. Paul E. Miller
Mr. Duane A. Lindoff
Mr. Richard Fundly
Mr. Robert M. Helsell
Wright Schuchart, Inc.
Mr. R. Eccles .
Mr. Stephen F. Politeo
Lilyblad Petroleum, Inc.
Mr. Stan Sable
Ms. Mary Susanj
Ms. Katherine Spiratos
Brown University
Ms. Gretchen C. Gerish
Ms. Mary E. Cosaboom
Ms. Ellen McComb Smith
Mr. Alf G. Anderson
Adm. James S. Russell
Ms. Laurie Lehman
Ms. Jennifer Paine
Dr. Colleen R. Carey
St. Luke's Medical Bldg.
Toshio and Suzanne Akamatsu
St. Joseph Hospital
1-26
-------
Docket Item
Number
Commenter and Affiliation*
IV-D-585
IV-D-586
IV-0-587
IV-D-588
IV-D-590
IV-D-591
IV-0-592
IV-0-594
IV-D-595
IV-D-596
IV-0-597
IV-0-598
IY-D-599
IV-0-600
IV-D-601
IV-0-604; IV-0-609
IV-D-605
IY-D-606; IY-D-689
IY-0-607
IY-D-608; OAQPS-79-8/IV-D-18
IV-D-610
Mr. Frank B. Terrill
Ms. Lidona Shelley
Mr. Brent Hartinger
Ms. Constance Northey
Mr. Michael J. Curley
Ms. Susan M. Hodge
Ms. Miriam Bishop
Mr. John Candy
Mr. Daniel M. Nelson
Princeton University
Department of Religion
Mr. Dwight Hoi combe
Mr. Bruce Hoeft
Mr. Lloyd D.' Morrell
Mr. Elliott McLean
Ms. Betsy Allen
Mr. Robert A. Erickson
Mr. Gerald S. Pade
Friends of the Earth,
Northwest Office
Mr. and Mrs. A. Derby
Chris Combs
Mr. Floyd Oles
Mr. Michael Gregory
Sierra Club, Grand Canyon Chapter
Paul and Sally Borgen
1-27
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-611
IV-D-612
IV-D-613
IV-D-614
IV-D-615
IV-0-616
IV-D-617; OAQPS-79-8/IV-D-19
IV-D-618
IV-0-619
IV-D-620
IV-D-620; IV-F-2b
IV-0-621
IV-D-623
IV-D-624
Mr. Ake Nygren
Boliden Metal! AB
Sweden
Mr. Lloyd Dodd
L-M-D Electro-Silver Resource
Ms. Virginia Mitchell
Mr. James Tracht
Pennwalt Corporation
Mr. Marvin Williams
Washington State Labor Council
AFL-CIO
Mr. Arne Bjornberg
Mr. Rolf Svedberg"
Boliden Metal! Ab
Sweden
Mr. David F. Zoll
Chemical Manufacturers Assoc.
Mr. Christopher DeMuth
Office of Management ft Budget
Mr. James H. Boyd
Newmont Mining Corporation
Mr. R. J. Moore, F. C. Schafrick,
and J. C. Martin
Shear & Gardner (for ASARCO)
Dr. Ian T.T. Higgins (for ASARCO)
Mr. M. 0. Varner, C. K. Guptill,
C. R. Counts, and D. E. Holt
ASARCO, Inc.
ASARCO, Inc.
*See footnote at end of this section
Mr. William Mitchell
Mr. William Woolf
1-28
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-625; OAQPS-79-8/IV-0-20
IY-D-626; OAQPS-79-8/IY-D-21
IY-D-627
IY-D-628
IV-D-629
IV-D-630
IV-0-631
IV-D-632
IV-D-633
IV-D-634
IY-D-634; IV-F-2b
IV-D-635
IV-D-636
IV-D-637
IV-0-638
IV-0-639
IV-0-640
IY-D-640; IV-F-2&
IV-0-640; IV-F-6b
Mr. J.F. McKenzie
Pacific Gas & Electric
Mr. Richard Kamp
Smelter Crisis Education Project
Mr. Thomas C. White
Mr. E.E. Ives
Steams-Roger Engineering Corp.
Mr. Brian Baird
Mr. John Thomas
Mr. Harmon Rulifson
Mr. Robert Matthews
Mr. Dennis Crumbley
Mr. A.Y.J. Prather and K.E. Blase
Prather, Seeger, Doolittle & Farmer,
Dr. S.H. Lamm (for Kennecott)
Mr. R.A. Malone, Dr. L.S. Salmon,
Dr. H.A. Lewis (for Kennecott)
Mr. and Mrs. LeRoy Annis
Ms. Evelyn Hildebrand
Ms. Lucy Fitch
Ms. Julie Reimer
Mr. Larry Jones
Mr. Floyd Hoffman, R.E. Johnson,
and W.N. Miller
Phelps Dodge Corporation
Dr. S.H. Lamm, Mr. T.L. Cogut
(for Phelps Dodge)
Mr. F.-P. Mendola
Phelps Dodge Corporation
1-29
-------
Docket Item
Number
Commenter and Affiliation3
IY-D-640; IV-D-704; OAQPS-79-8/
IY-D-22; OAQPS-79-8/IV-D-32
IY-D-641; OAQPS-79-8/IY-D-23
IY-D-642; IV-D-750
IY-D-643
IY-D-644
IY-D-645; IY-0-763; IV-D-770
IV-D-546; IY-D-708 and 708a;
IY-0-712; IV-0-767
IY-D-647
IY-D-650
IV-D-651; IY-D-653
IY-D-652
IY-D-654
IV-D-655
IY-D-6S6
IV-D-657
IY-D-659
IY-0-661
IY-D-663
A. Coy and S. Christiansen
Evans, Kitchel & Jenckes (for
Phelps Dodge)
Mr. Steven Kuhrtz
New Jersey Dept. of Environmental
Protection
Ms. Yvonne Thomas
Ms. Jeanette Wakeman
Ms. Catherine German
Dr. Thomas Douglas
Allied Medical Examiners
Mr. Michael Wright
United Steelworkers of America
Mr. Victor Gawley
Mr. William Evan
Wharton School of Finance
University of Pennsylvania
Mr. James Nolan
Puget Sound Air Pollution Control
Agency
Washington State Department of
Social & Health Services
Mr. Doug Sutherland
Tacoma-Pierce County Board of Health
Mrs. P.A. Aarrestad
Mr. Joseph Shopin
Mr. Warner Matson
Mr. Dwight Kipp
Mr. O.ouglas Branson
David and Marti Lambert
1-30
-------
Docket Item
Number
Commenter and Affiliation3
IV-0-664
IV-D-665
IV-D-669
IV-D-671
IV-D-672
IV-D-674
IV-D-675
IY-D-676; IY-D-677; IV-D-777
IV-0-678
IV-D-679
IV-D-680; IV-D-681
IY-0-682; IV-0-773
IY-0-684; IY-D-754;
IY-0-780
IV-0-685
IV-D-686
IV-D-688
IY-D-690
IY-D-692; IV-0-787; IV-D-792;
IY-D-793
IY-0-693; IY-D-764; IV-0-791
Dr. John Van Ginhoven
Mrs. Harold Hartinger
Mr. Bradley Nakagawa, et al.
Mr. Warren Wotten
Ms. Annabelle Reed
James and Debra Mains
Jonlee Joseph
Sen. Slade Gorton
U.S. Senate
Ms. Susan Macrae
Sierra Club
Mr. Bernard Clouse
Mr. Leonard Roberts
Office of Budget and Management
Ohio State Clearinghouse
Mr. Floyd Frost, Ph.D.
Washington Department of Social and
Health Services
Ms. Darcy L. Wright
Mr. Jon Muxoll
Mrs. T.L. Radke
Ms. Mary Clark Lee
Mr. Jack Callinsky
Mr. Gerald McGrath
Mr. Arthur Dammkoehler
Puget Sound Air Pollution Control
Agency
1-31
-------
Docket Item
Number
Commenter and Affiliation3
IY-D-694
IY-D-696
IV-D-697
IV-0-698; IV-0-731; IV-D-766;
OAQPS-79-8/IV-0-26; OAQPS-79-8
/IV-D-31; OAQPS-79-8/IV-D-34
IV-D-700
IY-D-701
IY-D-704a; OAQPS-79-8/IY-D-28
IY-D-705
IY-D-706
IV-D-707
IY-0-709
IY-D-711
IV-0-713
IV-D-715
IV-D-717; IY-0-722
IY-0-718
IV-0-720
IY-D-724
1-32
Donald and Shirley Ferris
Ms. Gail Nordstrom
Mr. Everett Lasher
Mr. Robert Abrams
Ms. Mary Lyndon
New York State Department of Law
Sven and Arvi Halstensen
Star Electric
Mr. Jon Hi nek
Greenpeace, U.S.A.
Dr. Steven Lamm
Consultants in Epidemiology Si
Occupational Health, Inc.
Iskra Johnson
Mr. John Roberts
Engineering Plus, Inc.
Ms. Margaret Wolf
Mr. Larry Weakly
Mr. Kurt Blase
Prather, Seeger, Doolittle & Farmer
Mr. Francis Hull
Mr. Phil Nelson
Washington State
Department of Ecology
Mr. James Harris
Ms. Eileen Goldgeier
Brown University
Ms. Lizabeth Brenneman
Mr. William Rodgers, Jr.
University of Washington
School of Law
-------
Docket Item
Number
Commenter and Affiliation3
IY-D-725
IY-D-726
IV-0-727
IY-D-728
IV-D-729
IV-D-730
IV-D-732
IY-0-733
IY-D-734
IY-D-735
IV-0-736
IV-0-737
IY-D-738; IV-D-751; IV-F-9
IV-D-739
IY-D-740
IY-0-741
IY-0-742
IV-0-743
Mr. Hugh Mitchell
Mr. Peter Andrews
Mr. John Calnan
Mr. Paul Karkainen
Mr. Timothy Larson
University of Washington
Department of Civil Engineering
Ms. Debbie Huntting
Mr. Peter Murray
Vashon Business Assoc.
Mr. Dan Schueler
Joseph and Karen Sartle
Mr. Frank Hagel
Mr. Robert Evans
Purified Air Systems
Washington Fair Share
Ms. Jeanne Snell
Yashon-Maury Island Community
Council
Mr. Douglas Easterling
University of Wisconsin
Department of Psychology
Mr. Bruce Mann
University of Puget Sound
Department of Economics
Dr. Jesse Tapp
Seattle-King County Department of
Public Health
Mrs. Anna Marie Champlain .
Mr. Brian Kameus
1-33
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-744
IV-D-746
IV-D-747; OAQPS-79-8/IV-D-24
IV-D-748
IY-D-752
IY-D-753
IY-0-755; IY-0-758;
OAQPS-79-8/IY-0-25
IY-0-756
IY-0-760; IV-D-774
IV-D-761
IV-Q-762
IY-0-765
IY-D-768
IV-D-769
IV-D-771
IV-D-772; OAQPS-79-8/IV-D-16
IY-D-775
Ms. Lin Noah
Kelly Wheat
Dr. Thomas Godar
American Lung Association
Ms. Karen Langbauer
Mr. Daniel Carlson
Ms. Kathleen R. Harkins and
.Mr. Vernon W. Harkins
Dr. W. Dale Overfield
Neurology and Neurosurgery Associates
of Taccma, Inc., P.S.
Ms. Penny Perka
Mr. Nils Lucander
Ms. Mary-Win O'Brien
United Steelworkers of America
Mr. Richard Dale Smith
Port of Tacoma
Mr. G.D. Schurtz
Kennecott
Ms. Marjorie L. Williams and
Ms. Fern Stephan
Mr. Lance Neitzel
Mr. Jeffrey Morris and
Ms. Cheryl Platt
Dr. Philip J. Landrigan
Centers for Disease Control NIOSH
Robert A. Taft Laboratories
Mr. Norman 0. Dicks
Member of Congress
1-34
-------
Docket Item
Number
Commenter and Affiliation3
IV-D-776
IY-D-778
IV-D-779
IV-0-782
IV-D-783
IV-D-784
IV-D-785
IV-D-788
IV-D-789
IV-0-790
IV-D-795; OAQPS-79-8/IY-D-9
IV-0-810
IV-0-811
IY-D-812
IV-0-813
IV-0-814
Mr. Rod Chandler
Member of Congress
Mr. John McCain
Member of Congress
Ms. Katherine M. Hayes
Mr. Ross Schlueter
Mr. Gary A. Preston
Mr. Dave Bateman
Mr. Richard W. Rice
Phelps Dodge Corporation
Mr. R.A. Malone
Kennecott
Mr. M.O. Varner
ASARCO, Inc.
Mr. Richard W. Rice
Phelps Dodge Corporation
Ms. Eve R. Simon
Ms., Denise Fort
State of Mew Mexico
Environmental Improvement
Division
Mr. F.C. Schafrick
Shea & Gardner (for ASARCO)
Mr. K.E. Blase
Prather, Seeger, Doolittle & Farmer
(for Kennecott)
Mr. S.J. Christiansen
Evans, Kitchel & Jenckes
(for Phelps Dodge)
Mr. Gordon Venable
State, of New Mexico
Environmental Improvement Division
1-35
-------
Docket Item
Number
Commenter and Affiliation9
IV-F-1**
IV-F-3, -4, -5**
IV-F-6^
OAUPS-79-8/IV-D-1
OAQPS-79-8/IV-D-4
OAQPS-79-8/IV-D-5
OAQPS-79-8/IV-D-8
OAQPS-79-8/1V-D-27
OAQPS-79-8/IV-D-29
OAQPS-79-8/IV-D-30
OAQPS-79-8/IV-D-35
UAQPS-/9-8/IV-D-36
Public Hearing transcript
Thomas Jefferson Auditorium
Department of Agriculture
Washington, D.C.
November 8, 1983
Mr. Blake Early
Sierra Club
Public Hearing transcripts
Bicentennial Pavillion
Tacoma, Washington
November 2-4, 1983
Mr. Rolf Svedberg
Boliden Metal! AB
Sweden
Mr. Thomas J. Koralewski
Libbey-Owens-Ford Company
Mrs. Robert 0. Hartwig
Mr. H. E. Dean
Plains Cotton Growers, Inc.
Mr. Earl W. Sears
National Cotton Council of America
Mr. J. T. Barr
Air Products and Chemicals, Inc.
Dr. Samuel Milham, Jr.
Washington State Department of
Social and Health Services
Dr. Ian Higgins
University of Michigan
School of Public Health
Comments Cross Referenced To
Other Dockets
Hunton & Williams for UARG
1-36
-------
Docket Item
Number
Commenter and Affiliation3
OAQPS-79-8/1V-D-37
Wapora, Inc., "Carcinogens from
Municipal Incinerators"
alf no affiliation is indicated, commenter is a private citizen.
bThese docket items contain the written testimonies submitted by
commenters at the public hearings, which are essentially identical to
their oral presentations.
cNot a comment on this standard.
*The IV-D-621 code indicates comments submitted by ASARCO. Numbers 1-16
following 621 indicate sections of ASARCO comments. Naumbers following
1-16 and immediately preceeded by a decimal point indicate subsections,
e.g., IV-D-621-15.1 represents comments found in subsection 1 within
section 15 of ASARCO's comments.
**In the main text, a one, two, or three digit number following a decimal
point indicates the position of the commenter within the order of the
speakers at the hearing on that particular day, e.g., IV-F-1.13 represents
the thirteenth speaker at the public hearing on November 8, 1983 in
Washington, D.C.
1-37
-------
-------
2. HEALTH
2.1 HEALTH EFFECTS ASSOCIATED WITH EXPOSURE
2.1.1 Carcinogenicity of Arsenic Emissions
Comment:
A number of comments submitted in response to the listing of arsenic
agreed with EPA's conclusion that arsenic should be regarded as a human
carcinogen (IV-D-11, IV-D-66, IV-D-158, IV-D-150, IV-D-144, IV-D-590, IV-D-164,
IV-0-292, IV-D-420, IV-D-427, IV-D-441, IV-D-152, IV-D-592, IV-D-301, IV-D-388,'
IV-D-588, IV-D-8, IV-D-411, IV-D-314, IV-D-604, IV-F-3.30*, IV-F-3.31, IV-F-3.37,
IV-F-3.38, IV-F-3.43, IV-F-3.45, IV-F-3.60, IV-F-3.103, IV-F-3.72, IV-F-4.4,
IV-F-4.6, IV-F-4.9, IV-F-4.25, IV-F-4.28, IV-F-4.31, IV-F-4.50, IV-F-4.66,
IV-F-5.15, IV-D-709, IV-D-717, IV-D-718, IV-D-756, IV-D-768, IV-D-722,
IV-D-726, IV-0-16 IV-D-772, IV-D-705, IV-D-710, IV-D-742, IV-D-746, IV-D-427,
IV-D-515, IV-D-530, IV-D-541, IV-D-622, IV-D-630, IV-D-644, IV-D-673,
IV-D-676). One writer (IV-D-66) indicated that he knows arsenic is a carcinogen
and that according to ASARCO's Michael Varner, high levels of arsenic caused
cancer in pre-WWII workers. A second writer (IV-D-411) spoke of her experiences
as a biologist and cancer victim. She is fully convinced that arsenic, in
any amount, causes cancer. A third writer (IV-D-164) referred to a Tacoma
area Veterinarian's statement that there is "an unusually high incidence of
cancer among hogs".
Comment:
Testimony was also offered on the carcinogenicity of inorganic arsenic.
One commenter (IV-F-3.52) expressed concern for the effects of arsenic on children
growing up with as high as ten times the normal amount of a known carcinogen
in their bodies. The Washington State League of Women Voters (IV-F-4.11) in
their testimony, referred to arsenic as a hazardous air pollutant and known
carcinogen.
* See footnotes in Chapter 1 for description of the decimal system used to
identify public comment.
2-1
-------
Comment:
Some commenters discussed the types of cancer caused by arsenic,
specifically, lung and skin cancer (IV-D-115, IV-D-32, IV-D-106, IV-D-412,
IV-F-3.4, IV-F-3.41, IV-F-3.42, IV-F-4.9, IV-F-4.66, IV-F-4.71, IV-D-429,
IV-D-137, IV-0-111, IV-D-621-16.12, IV-D-611, IV-D-622, IV-D-670, IV-D-676).
Some individuals felt that arsenic caused lung cancer (IV-D-4, IV-F-3.6,
IV-F-3.7, IV-F-3.5, IV-F-4.4, IV-F-4.68 IV-D-141, IV-D-146, IV-F-4.71).
According to one correspondent (IV-D-4), lung cancer is two times as common
near arsenic emitting smelters. Another correspondent (IV-D-141) indicated
that EPA cites the cancer risks of arsenic, but cannot determine how much
it takes to cause lung cancer. Another correspondent (IV-D-146) stated the
odds of getting lung cancer may be slightly above average for srnelter
workers.
Comment:
One individual (IV-F-3.73) testified that arsenic in the air would increase
the risk of lung cancer. A second commenter (IV-F-4.62) stated his belief that
arsenic caused an increased level of lung cancer in smelter workers. A
third individual, speaking as a member of a smelter union, expressed the
opinion that exposure to inorganic arsenic posed health risks. "We know
what arsenic has done to too many of our union brothers and sisters in the
Tacoma Smelter and other copper smelters. It was the deaths of our members
which provided the conclusive evidence that arsenic causes lung cancer"
(IV-F-4.4).
Response:
The EPA agrees with the commenters. The present status of inorganic
arsenic as a human and experimental animal carcinogen has been closely
investigated by agencies such as the National Institutes of Occupational
Safety and Health, scientific organizations such as the National Academy
of Science and the International Agency for Research on Cancer (IARC), and
in a number of individual assessments.
The EPA has estimated the relative carcinogenic potencies of a number
of chemical substances and has ranked arsenic within the first quartile of
52 suspect carcinogens among such other suspect human carcinogens as DDT,
2-2
-------
PCBs, Aldrin, and B[a]P. In addition, the IARC characterizes arsenic as
carcinogenic to humans.1
Epidemiological studies of copper smelter workers in the U.S., Sweden,
and Japan have strongly suggested an increased risk of respiratory cancer
resulting from exposure to airborne inorganic arsenic. The EPA's Health
Assessment Document has reviewed 12 such studies, of which 11 have shown a
positive association between exposure to arsenic and lung cancer.2 The range
of the statistical mortality rates from lung cancer in smelter workers above
the expected lung cancer mortality rates in the non-exposed population
indicates a 3-fold to 12-fold increase in the risk of lung cancer as a result
of airborne arsenic exposure.
Proportionate mortality studies of arsenical pesticide workers have
also shown an increased risk of lung cancer mortality in a range of 3 to 16
times that expected. A study of German vintners using arsenical pesticides
found a significant increase in lung cancer mortality above the expected
rate.
With respect to nonoccupationally exposed groups, arsenic contaminated
drinking water studies and studies of patients using arsenical medicinals
have demonstrated a skin cancer prevalence rate in these exposure groups.
A study of Taiwan residents consuming high levels of arsenic in drinking
water showed a 10-fold increase in the risk of skin cancer. Precancerous
hyperpigmentation and hyperkeratosis were evident in many other arsenic
contaminated drinking water studies. Keratonic lesions, hyperpigmentation,
and epitheliomas were found to be prevalent in individuals using arsenical
medicinals.
Comment:
Several commenters expressed concern about the correlation between
lung cancer and the ASARCO Tacoma smelter emisisons. Five individuals
noted that Tacoma ranked below the national average and/or as fifth among
the state's 10 largest cities for lung cancer (IV-D-264, IV-D-255, IV-D-256,
IV-D-330, IV-U-402, IV-F-4.38).
2-3
-------
Response:
While EPA agrees that studies of the lung cancer rates in the vicinity
of Tacoma generally do not show elevated lung cancer rates, the Agency does
not agree that these findings are sufficient to discount the potential cancer
risks from exposure to ambient arsenic. The power of any epidemiological
study in the detection of risk is limited. Particularly for more common forms
of cancer such as lung cancer, a large increase in observed cancers would be
necessary to distinguish the effect of a specific carcinogen from a
relaitvely high background incidence. The Tacoma data may indicate that
the problem is not "epidemic" in nature, and in this regard are not
inconsistent with EPA's risk estimates.
Comment:
A few comments (IV-D-593, IV-F-3.72, IV-D-571, IV-D-622, IV-D-630, IV-D-676)
referred only to skin cancer. One writer (IV-D-593) commented that EPA's own
scientists had acknowledged a relationship between arsenic exposure and
skin cancer but that the Agency ignored this effect because it is curable.
An individual (IV-F-3.72) thought there's a possibility of skin cancer transmitted
to human being through hand-mouth dust contact. However, two commenters
felt that arsenic does not cause skin cancer (IV-D-621-14.4, IV-D-345).
Two correspondents discussed skin cancer in relation to the smelter emissions.
One correspondent (IV-D-391) wrote that although his family had lived near
the ASARCO Tacoma smelter for a total of over 800 person-years, not one had
ever had skin cancer. Another correspondent (IV-D-597) stated that two family
members had developed skin cancer, fortunately treatable, at various times
in their lives (IV-D-597).
Response:
The EPA reviewed case-control studies of populations or individuals ex-
posed to arsenic-contaminated drinking water, and arsenical medicinals.4 Most
of the studies demonstrated a positive association between exposure, either by
inyestion or dermal absorption, and the manifestation of skin cancer. A
study of 40,000 persons in Taiwan exposed to arsenic in the drinking water
2-4
-------
found a significant prevalence of skin cancer over that of 7,500 other
Taiwanese who drank water free of arsenic. A study of a village in Chile
dependent on arsenic contaminated drinking water showed precancerous skin
formations in many of the inhabitants. Similar etiological studies in the
U. S. where relatively high arsenic contamination existed in the drinking
water have not demonstrated a prevalence of skin cancer. Case reports of
patients that were treated with arsenical medicinals showed histopathic
manifestations including hyperpigmentation, keratotic lesions and epitheliomas,
While the bulk of the evidence does correlate skin cancer with exposure to
arsenic contaminated drinking water and arsenical medicinals, the results
of the case reports were not quantitative to the extent that risk from
exposure could be modeled. Furthermore, in reviewing the primary route of
individual exposure to inorganic arsenic, EPA determined that the primary
exposure pathway affecting the most numbers of people is the airborn disper-
sion of inorganic arsenic from the emissions of certain industrial processes.
Therefore, in estimating risk only epidemiological data that involved this
exposure pathway were assessed.
Comment:
One commenter (IV-F-5.21) referred to the difference in individual suscep-
tibility to cancer from arsenic exposure. According to this commenter:
"A lot of people think that because they can get by without having cancer
from arsenic that everybody is the same and in the same boat but I think
some people can get by without this cancer...or the smelter fumes and not
get cancer but other ones will".
Response:
Individual susceptibility to carcinogenesis is a consideration in the
overall assessment of risk. The U. S. population represents a very diverse,
genetically heterogeneous group that is exposed to a variety of toxic agents.
The National Academy of Sciences has stated that:
"Genetic variability to carcinogenesis is well-documented, and it is
2-5
-------
dlso known that individuals who are deficient in immunoloyical competence
(for yenetic or environmental reasons) are particularly susceptible to some
forms of cancer. It seems, therefore, that even if we were to postulate an
average threshold for a particular cancer induced by a particular agent, we
would in practice need a series of thresholds for different individuals. It
would be extremely difficult, in practice, to establish a single threshold."^
The EPA agrees with MAS' observations on this subject and, EPA, for this
reason, tends to give less weight to view that a threshold for airborne
arsenic exposure exists below which no possibility of cancer can arise.
Considering the potential for variation in susceptibility, it is unlikely
that practical thresholds could be determined with any degree of certainty.
In addition, the inhalation exposure data base involves healthy male workers
and therefore does not provide adequate information for EPA to assess risk
to subpopulations with potentially higher susceptability.
Comment:
Examples of community members (mostly employed by and/or liviny in the
vicinity of the ASARCO Tacoma smelter who had developed cancer (IV-D-79,
IV-D-133, IV-D-139, IV-D-428, IV-F-4.52, IV-F-5.18) were provided. Examples were
also given of those, in similar circumstances, who hadn't developed cancer
(IV-D-30, IV-D-133, IV-D-139, IV-D-277, IV-D-326, IV-D172, IV-D-181, IV-D-208,
IV-D-210, IV-D-265, IV-D-229, IV-D-383, IV-D-457, IV-D-345, IV-D-306, IV-D-333,
IV-D-356, IV-D-359, IV-D-282, IV-D-324, IV-F-4.5, IV-F-4.21, IV-F-4.52, IV-D31,
IV-D-58, IV-D-134, 481,8343, IV-D-504, IV-D-193, IV-D-179, IV-D-601, IV-D-485,
IV-D362, IV-D-372, IV-D-270, IV-F-4.44, IV-F-4.49, IV-F-4.58).
Comment:
Several commenters disagreed with the potential positive correlation
between arsenic and lung cancer (IV-F-4.60, IV-D-304, IV-D-316, IV-D-338,
IV-D257, IV-D-274, IV-D-303, IV-D-312, IV-D-502, IV-D-185, IV-D-196, IV-D-160,
IV-D-167, IV-D-168, IV-D-311, IV-D-242, IV-D-494, IV-D-232, IV-D-621-14.12,
IV-D-621-16.11, IV-U-343). One writer (IV-D-494) stated that arsenic emitted
by the smelter has been in the air for over sixty years and research has
shown a below average rate for lung cancer in the Tacoma area. Good scientific
work would not ignore past real experience and make life and death predictions
2-6
-------
based on short term measurements and estimates. Another writer (IV-0-232)
asserted that no link has been established between arsenic and lung cancer.
An individual (IV-F-4.60) testified that arsenic air pollution has not been shown
to produce high lung cancer rates in the study populations.
Comment:
Many of the commenters claimed that there was no connection between
the smelter emissions and lung cancer (IV-D-212, IV-D-213, IV-D-215, IV-D-354,
IV-D-367, IV-D-355, IV-D-349, IV-D-382, IV-D-340, IV-F-3.2, IV-D-621-12.6,
IV-D-621-12.il, IV-D-621-12.17, IV-D-621-12.22, IV-D-621-13, IV-D-621-14.4,
IV-D-621-14.7, IV-0-621-15.9, IV-D-621-16.2, IV-D-695, IV-D-697, IV-D-718,
IV-D-621-12.8, IV-0-621-15.9, IV-D-621-16.11, IV-D-350, IV-D-518, IV-D-523,
IV-D-550, IV-D-555, IV-D-568, IV-D-612, IV-D-607, IV-D-522, IV-0-561,
IV-D-621-5, IV-0-625). Dr. Samuel Milham (IV-F-3.2) investigated lung cancer
mortality by census tract in the Tacoma area and found no difference in lung
cancer mortality in census tracts closest to the smelter when workers were
removed (although smelter emissions were not mentioned in the description
of this study).
Comment:
Other writers felt that ASARCO emissions do not cause cancer (IV-D-335,
IV-D-487, IV-D-381, IV-D-472, IV-D-154, IV-D-166, IV-D-233, IV-D-226,
IV-D-250, IV-D-489, IV-D-251, IV-D-377, IV-D-68, IV-D-621-12.3, IV-D-621-12.15,
IV-D-62112.11, IV-D-621-12.23, IV-D-621-14.3, IV-D-621-6, IV-D-343, IV-D-534, '
IV-D621-14.9). According to one writer (IV-D-489), epidemiological studies
have not proven a direct relationship between cancer and the levels of
emissions currently found in the areas around primary copper smelters.
Comment:
An individual (IV-F-3.15) commented that attributing a certain number of
deaths per year from cancer due to smelter emissions was "sheer nonsense"
and that while body chemistry can trigger cancerous conditions in certain
individuals, determining the degree of tolerance and the amount of chemical
intake the average individual could withstand could be pure guesswork.
2-7
-------
Comment:
Some people disagreed with an association between arsenic and cancer
on other grounds (IV-D-249, IV-D-187, IV-D-407, IV-D-192, IV-D-149, IV-D-358,
IV-D-275, IV-F-3.18, IV-D-45, IV-F-3.52, IV-F-3.59, IV-F-5.8, IV-D-16
(IV-D-772), IV-D-621-15.6). A chronic disease epidemiologist (IV-D-46)
maintained that the relationship of dose and cancer risk which supports the
NESHAP may be nothing more than a relationship between age and risk of
cancer. A correspondent (IV-D-149) indicated that many experiments in
animals have shown trivalent inorganic arsenicals incapable of causing
cancer. Five individuals (IV-D-621-120, IV-D-621-15.6, IV-D-621-12.8,
IV-D-621-16.12, IV-D-621-12.15), commented that arsenic is actually a
nutrient and/or has a protective effect on the human system.
Comment:
Testimony regarding the lack of a positive relationship between arsenic
and cancer was also provided. One individual (IV-F-3.11) noted that while
certain epidemiological studies based on industrial exposures seem to implicate
arsenic as a carcinogen, most community based studies have not provided con-
firmation and that there is uniform support from the animal literature denying
the carcinogenicity of arsenic. Nancy Frost (IV-F-4.72), in support of her
father, Douglas V. Frost, Ph.D., a nutrition biochemist, submitted a newspaper
article that he wrote in which he described arsenic as an essential nutrient
and not a pollutant. Dr. Frost concluded that "no arsenical had been found
to cause cancer experimentally in animals and the presumed link between
arsenic and cancer in human's was an unproven and untestable association.
Ms. Frost also submitted a paper published by her father entitled "What
Do Losses in Selenium and Arsenic Bioavailability Signify for Health?" in
which Dr. Frost asserted that, "there are many more likely causes for the
cancers in humans for which arsenic is blamed".
Community Health Studies:
Kennecott (IV-D-634) submitted testimony by two epidemiologists stating
that review of the community epidemiology studies referenced in EPA's draft
2-8
-------
health assessment and other such studies shows no support for EPA's state-
ment that excess mortality or morbidity exists for populations living near
arsenic emitting sources. One commenter believed that in total, these
studies represent a sufficient population group to have shown such effects
if they existed. The commenters reviewed 10 lung cancer mortality studies.
They claimed there were only three with any positive findings, and they
questioned these findings. These 10 studies included the following:
1. Blot and Fraumeni, 1975. Lung cancer mortality was shown to be
significantly higher among males and females in 36 U.S. counties with
copper, lead, and zinc smelters and refineries than in the rest of U.S.
counties. The increase, corrected for demographic variables, was 17 percent
for males and 15 percent for females over the years 1950-1969.
2. Lyon, et al. 1977. Using a population based cancer registry,
addresses at diagnosis of lung cancer cases are compared to malignant
lymphoma controls to assess the possible carcinogenic effect of the Salt
Lake City copper smelter. The distribution of distances from the Smelter
of lung cancer cases and lymphoma controls was similar.
3. Rom, et al. 1982. Using the same methodology as Lyon, lung cancer
cases around the El Paso, Texas, smelter were shown to have the same distance
distribution from the smelter as breast and prostate cancer controls.
4. Greaves, et al. 1981. Greaves, using the same methods as Lyon and
Rom, studied the distances of residences at diagnosis or death of lung
cancer cases and controls (prostate, colon and breast cancers) from ten
copper smelters and one lead-zinc smelter. The distance distribution of
lung cancer was not significantly different from the distribution of the
control cancers in any of the areas studied.
5. Pershagen, et al. 1977. Mortality in the region around the Ronnskar
Smelter in northern Sweden was studied. The population residing within 15
km of the smelter was compared to the population residing 200 km away. The
lung cancer mortality in the exposed population (<15 km) was significantly
higher in men than in the comparison population, but not significantly
different in comparison to national rates. When the occupationally exposed
2-9
-------
cases are removed, the lung cancer Standard Mortality Rate (SMR) was reduced
and was no longer statistically significantly different than the comparison
population.
6. Matinoski, et al. 1976. Cancer mortality reported on death
certificates was studied in census tracts in Baltimore around a chemical
plant producing calcium and lead arsenate, arsenic acid, cupric acetoarsenite
(Paris green), and sodium arsenite. An increase in lung cancer was seen in
the census tract containing the plant in the years 1966-1974 in males only.
No increase was seen in an earlier time period (1958-1962). Residents of
the city who died elsewhere were not ascertained. In the census tract
where the plant was located, the average soil arsenic level was 63 ppm.
Removing plant workers from the high lung cancer census tract did not
eliminate the high male lung cancer mortality rate.
7. Polissar, et al. 1979. Lung cancer mortality by census tract was
examined around the Tacoma, Washington, copper smelter. Two surrogates for
arsenic exposure were used: (1) distance of the census tract to the smelter,
and (2) concentration of sulfur dioxide over background for each census
tract. There was no excess risk of lung cancer for persons living near the
smelter.
8. Hartley, et al . 1982. Lung cancer mortality in the 35 census
tracts in Tacoma, Washington, was examined for the 21 years 1950-1970,
using the death certificate address for assignment to census tract. Lung
cancer mortality was no higher in the census tracts near the smelter than in
those farther away.
9. Mil ham, et al. 1982. Class rosters of children enrolled at the
Uuston elementary school (100 yards from the Tacoma, Washington, smelter)
were examined. A cohort of 283 children who were enrolled for three or
more years during the years 1900-1919 was developed. Surviving cohort
members were contacted and death records were obtained for decedent members.
Using life table comparisons, mortality of men in this cohort was shown to
be favorable (more survivors to 1980 than expected). It also did not appear
that lung cancer was increased in the male cohort (1 lung cancer death
among 20 for whom death certificates were obtained). Forty percent of the
men in this cohort were employed at the smelter at some time.
2-10
-------
10. Newman, et al. 1976. Although this was primarily a study of lung
cancer cell type in two Montana copper mining and smelting counties, it
demonstrated an increase in lung cancer incidence in both men and women in
the towns of Butte and Anaconda, but the same increase was not seen in the
counties as a whole.
Also, commenters identified several morbidity studies and they are
summarized below.
Community Morbidity Studies
1. Mil ham and Strong, 1974. In the population around the Tacoma
smelter, children were shown to have increased levels of arsenic in hair
and urine. Urinary arsenic decreased with distance from the smelter. Mean
urinary arsenic for children living within .5 miles of the smelter was .30
ppm (normal .014). Vacuum cleaner dust and attic dust contained over 1000
ppm of arsenic.
2. Morse, et al. 1979. Children exposed to arsenic in air and drinking
water in Ajo, Arizona, near a copper mine and smelter were studied. Hair
and urinary arsenic were elevated in children and decreased with distance
from the smelter. No clinical or hematologic abnormalities attributable to
arsenic were found.
3. Baker, et al. 1977. In 19 U.S. towns with primary nonferrous
smelters, one to five year old children were studied for arsenic, lead and
cadmium absorption. Urine arsenic was elevated near 10 of 11 copper smelters.
4. Mil ham 1977. Hearing, hematologic status and school attendance of
children living in Ruston, Washington (near the Tacoma Smelter), were the
same as children living further away from the smelter. The Ruston children
have increased levels of urinary and hair arsenic.
5- Nordstrom, et al. 1978. Frequencies of congenital malformations
were studied in offspring of female employees of the Ronnskar Smelter and
in the populations living near the smelter. In the offspring of the
employees, the frequency of multiple malformations was increased. However,
there was no increase in total frequency of malformations or in type of
malformations in the population around the smelter.
2-11
-------
6. Nordstrom, et al. 1978. Frequency of spontaneous abortion and
birthweight distributions in female smelter employees and women who lived
near the smelter were examined. Women working at the smelter had an
increased frequency of spontaneous abortion and low birthweight infants.
Women living near the smelter showed no increase in spontaneous abortions,
but had a tendency to have infants slightly lighter than women who lived at
a distance from the smelter.
The commenter said the Matanoski census track study, showing an
increase in lung cancer near a plant in Baltimore, MD, producing arsenical
compounds, is probably an aberration. The study found increased lung
cancer rates only in males and only between the years 1966-1974. Increases
were not seen in the years 1958-1962 despite the fact that the smelter had
been operating since about 1900. Furthermore, the commenters claimed that
studies of communities with higher potential arsenic exposure showed no
increase in lung cancer in persons residing near higher arsenic emitting
sources. Three studies they cited to show this were the Pershagen et al.,
1977 study of the region around the Ronnskar smelter in northern Sweden,
and studies by Polissar, et al., 1979 and Hartly et al., 1982 of census
tracts around the ASARCO-Tacoma copper smelter. The studies of Tacoma
showed no increase in lung cancer. The Pershagen study showed no significant
increase once occupationally-exposed men were removed. The commenters said
that emissions of arsenic are lower from the Baltimore plant than from the
smelters, and that maximal soil arsenic levels near the Baltimore plant are
only 10 percent of those in the Tacoma area. Thus, they said that any
increases in lung cancer mortality should also have been seen at these
plants, and that the Matanoski study results are an aberration.
The commenters noted that a study by Blot and Fraumeni showed increased
lung cancer in males and females for the years 1950-1969 in 36 counties
with lead, copper, zinc smelters and refineries when compared to all U.S.
counties. The commenters said that weak points of the study include the
fact that smelting counties were not separated from refining counties and
no data on arsenic exposure are available. They also stated that since lung
cancer mortality varies by a factor of 2 from state to state, it is more
valid to compare smelting counties with other counties in the same state.
2-12
-------
They claimed that if lung cancer rates are compared with rates for all
other counties in the same state during the same time period, the lung
cancer excess disappears.
The commenters said the Newman et_ cil_., 1976 study showed excess lung
cancer rates in two cities near copper mining and smelting facilities, but
not in the county as a whole. One epidemiologist proposed that since
residential information was obtained from hospital and tumor registry files,
elevated rates in the towns may have been caused by migration from rural
areas to the towns after retirement.
The commenters' review of six community morbidity studies showed only
one finding with possible potential significance. The Nordstrom et_ aj_.,
1978 study found that women living near a Swedish smelter delivered infants
weighing slightly less on average than those at distances from the smelter.
No increase in spontaneous abortions or congenital malformations was observed
in this group. The commenter did not know if any significant health problems
could be associated with low birthweight , and could not tell if the findings
were an aberration or might be possible to duplicate in future studies.
The commenters concluded that no significant increases in mortality or
morbidity had been shown in areas around high and low arsenic emitting
sources. One commenter said he believed that, in total, the studies
represented a significant population group exposed for a number of years,
and that if there were discernable increases in lung cancer or other
morbidity, these studies should have shown them. He did not attempt to
calculate the total population represented by the studies, and his conclusion
that the studies should have been able to detect any lung cancer increased was
based on his judgment and the judgment of other epidemiologists he had
contacted.
Comment:
On behalf of Phelps Dodge (IV-D-704a), CEOH submitted a translation of
a 1981 report by the Swedish National Health Board expert committee. The
report reviewed the Nordstrom et al., studies relating arsenic exposure to
women living near a Swedish smelter to low birthweight of their babies.
The report cited several factors which may affect birthweight:
2-13
-------
- age and birth year of the mother
- Parity (number of children born previously) and earlier
pregnancies
- smoking habits
- health status
- social factors (social group membership, occupation, etc.)
The committee reported that these factors have not been adequately considered
and may account for the variation in birthweights between groups. The
committee further reported that population groups chosen in the studies
were not homogeneous in important respects. There was no reporting of an
aje factor or of a factor of previously experienced deliveries in this
study. The committee believed these factors to substantially influence
birth rates and felt them necessary to account for in such a study. Social
factors also were not controlled in the studies. Furthermore, there appeared
to be confusion in the concepts of pregnancy order and parity in the
reporting of the studies, making results difficult to interpret and
unreliable. The committee said that the deficiencies cited in the study
call into question the authors' statistical analysis showing differences
amony the groups.
The committee described Nordstrom's exposure data as vague. Exposure
is described in terms of residence location (areas A through D and parishes
in the Skellefteae area) or employment in a department at the smelter.
Information on environment lacks detail, and there is nothing in the papers
reviewed to indicate that exposures in areas A through D differ from each
other. A recent article does indicate increased urinary arsenic concentrations
in women living in the Skellefteae area and lower concentrations in women
living further from the smelter. For occupationally exposed employees, the
committee found department not to be a fine enough classification to
determine exposure.
Due to the problems discussed above, the committee could not conclude
from the studies that birthweights are lower for women living near the smelter.
Nordstrom1s study also reported increased frequency of chromosomal
abnormalities in smelter workers; however, the committee noted that the
2-14
-------
control population is not described with respect to selection, size, and
age distribution. They believed new analyses would be necessary before
this finding could be supported.
The studies found that children of mothers working at the smelter have
increased malformation frequencies. However, the committee presented data
showing that the frequency of diagnosing and reporting malformations varies
greatly between different hospitals and clinics. This makes comparison
between time periods (as Nordstrom's studies have done) suspect. Furthermore,
the committee deemed the population groups studied small and the numbers of
malformations small. Such small numbers the committee called unreliable.
The committee also noted deficiencies in the original analyses of
spontaneous abortion, and noted that reanalyses of areas around the smelter
have found no significant differences in frequency of spontaneous abortion.
Other commenters (IV-D-640, IV-D-621-16.10, IV-F-1.16) referred to the
same study. They characterized the study as the only study that alleges
health effects from community exposure. The commenters reported that the
study claims decreases in birthweight and increases in multiple malformation
frequency among offspring of residents near the Ronnskar smelter in Sweden.
The commenters pointed out that this birthweight study had found no difference
for births of parity 1 or parity 2 and that Nordstrom's analysis had given
no consideration to known factors affecting birthweight such as smoking
history, maternal age and increased parity, social class, and gestational
age. The commenters felt there were too many difficulties with the study
to accept its results. They noted that the Swedish National Health Board
expert committee report of 1981 concluded that study design and execution
problems prevented these findings from being accepted at face value.
Response:
The EPA has reviewed community mortality and morbidity studies of
areas in the vicinity of smelters emitting arsenic, and arsenic pesticide
manufacturing plants.6 For a number of reasons, these studies are confounding
or inconclusive in demonstrating either a positive or negative association of
lung cancer to community exposure to inorganic arsenic (see pages 7-50-52 in
2-15
-------
the Moalth Assessment Document). For the most part, these studies did not
observe the length of time people lived near smelters, nor did they account
for migration patterns. In addition, arsenic exposure levels correlated
with distance from the emitting source were not determined. Lung cancer
morbidity was derived by inspecting death certificates and comparing rates
of lung cancer morbidity in the community with national, county, or States
rates. Such procedures may undercount lung cancer SMRs within the community
because individuals with lung cancer may move away to receive treatment, or
patients diagnosed as having lung cancer may have died of other causes.
Generally, community studies do lack the statistical power to detect
the increased lung cancer risk to the exposed public and EPA does not
expect such studies to produce positive findings.
In the series of Nordstrom et al. studies, it has been repeatedly
shown that these studies are flawed for a number of reasons. The Health
Assessment Document for Inorganic Arsenic also cautions: "These
studies (Nordstrom's) were not designated specifically to study effects of
arsenic but rather to study the effects, in general, of the smelter work
pollutants on neighboring (proximate) populations, the diverse agents
involved preclude making conclusive statements about the specific effects
of arsenic." In addition, unbeknownst to the EPA at the time of the HAD
publication, the Swedish National Health Board Expert Committee published a
report in 1981 that questioned almost every finding in the Nordstrom studies.
It is therefore highly questionable whether the Nordstrom studies are
suitable for making determinations regarding the potential human reproductive
effects caused by arsenic exposure.
Thus, in view of the fact that the community studies did not produce a
clear understanding of risk associated with arsenic exposure near a smelter,
EPA resorted to the best etiological data base in characterizing inorganic
arsenic as a human carcinogen: smelter worker exposure studies. While
various animal studies have not demonstrated arsenic carcinogenicity despite
using different chemical forms, routes of exposure, and different experimental
species, various human epidemiological investigations have showed a consistent
2-16
-------
association between airborne exposure to inorganic arsenic, and respiratory
cancer in humans. The occupational studies compared the cancer risk (adjusted
for age, sex, and other variables) among groups or cohorts exposed to various
concentrations of inorganic arsenic with control groups not so exposed.
Furthermore, the methodology followed the exposed groups prospectively over
time to determine latency, and doseresponse relationships. Epidemiological
studies of smelter workers exposed to inorganic arsenic have demonstrated
an increased risk of lung cancer mortality 3 to 12 times the expected
mortality rate of nonexposed population groups.
2.1.2 Toxicity of Arsenic Emissions/Smelter Emissions
Comment:
Some individuals (many living near the ASARCO-Tacoma smelter) commented
on the toxicity of arsenic (IV-D-301, IV-D-108, IV-D-439, IV-D-116, IV-D-76,
IV-D-709, 1V-D-769, IV-D-720, IV-D-721, IV-D-734, IV-D-756, IV-D-698, IV-D-70b,
IV-D-16 (IV-D-772), IV-E-621-14.3, IV-D-730, IV-D-779, IV-D-736, IV-D-16
(1V-D-702), IV-D-524, IV-D-554, IV-D-576, IV-D-630, IV-D-674, IV-D-675,
IV-D-676, IV-D-662, IV-D-677, IV-D-694, IV-D-520, IV-D-433, IV-D-438). One
writer (IV-D-108) stated that the "garlic-smell" of the arsenic in the
smelter plume was indicative of its toxic character. A second writer
(IV-D-439) called arsenic a known poison and questioned the health danger
of this compound at low concentrations. A third writer (IV-D-116) asked
"why ASARCO (has) the right to daily poison (him) with its filthy arsenic
emissions." A fourth writer (IV-D-76) maintained that a dose of arsenic as
small as half the weight of a pin can be fatal to small children; a daily
dose as small as 3.5 milligrams is likewise fatal to infants. However, one
commenter (IV-D-621-16.2) noted that the body can detoxify/excrete arsenic.
Comment:
Testimony was also given on the toxicity of arsenic (IV-F-3.6, IV-F-3.11,
IV-F-3.38), calling arsenic "a very toxic substance"... and that "if you
put a teaspoon of it on your Wheaties, it will kill you right now...but not
from lung cancer (IV-F-3.73). Another individual (IV-F-3.11) noted that
concerns have been voiced in various U.S. communities and that the basis
2-17
-------
for these concerns includes a well founded appreciation for arsenic as a
lethal agent of historical as well as modern significance.
Comment:
Cases of physical ailments resulting from arsenic exposure were submitted
by community members of the ASARCO smelter (IV-D-571, IV-D-622, IV-D-755,
IV-D-758). According to one writer (IV-D-158), arsenic exposure in high
doses causes increased incidences of chromosomal aberrations and neurological
problems. Another writer (IV-D-593) referred to a statement made by Dr. Karle
Mottet of the University of Washington that arsenic may cause cardiovascular
problems. Another writer (IV-D-273), employed by ASARCO for years, felt
that the only lasting damage that he sustained from the arsenic itself was
a perforated septum.
Comment:
Commenters also discussed physical ailments (IV-F-4.6, IV-F-3.38,
IV-F-3.34, IV-F-4.45). One person (IV-F-3.38) asked about other "less
dramatic" health effects resulting from arsenic exposure including angina
and high blood pressure. Another commenter (IV-F-3.34) noted swelling and
certain described edema, especially of the lower limbs, face and ankles,
and a garlic odor to the breath and body sweat. This may be associated
with nausea, vomiting or diarrhea. There can be depression of the bone
marrow.
Comment:
Another commenter (IV-F-3.11) claimed there is also no reason to conclude,
based on either theoretical or any practical considerations, that ambient
arsenic concentrations from smelters cause or contribute to any other disease
processes.
One individual (IV-F-3.57) stated that ASARCO's emissions cause chromosomal
aberrations and a variety of neurological problems. Another person (IV-F-3.53)
maintained that community residents have unexplained breathing difficulties,
gastrointestinal problems and mysterious allergies which are attributable to
the smelter emissions.
2-18
-------
Comment:
Some comments were submitted which claimed that the fumes from ASARCO
were damaging to the public's health (IV-D-94, IV-D-430, IV-D-599, IV-D-597,
IV-F-3.20, IV-F-3.29) and caused burning of the lungs (IV-D-89), eyes, nose,
and throat (IV-D-115, IV-D-428). One writer (IV-D-89) indicated that smoke-
stack emissions caused illness, discomfort, and severe stomach upset.
Another writer (IV-D-597) commented, that his wife grew up a quarter mile
from the smelter smokestacks and neither she nor her brother have a sense
of smell (IV-D-597). One commenter (IV-F-4.19) stated that when sailing
near the smelter the smoke occasionally comes straight down onto the water
and causes sore throats and general discomfort.
Comment:
A few correspondents discussed the potential adverse health effects
associated with the inhalation of smelter emissions in general. Two
correspondents (IV-D-113, IV-D-15) asked about the health implications of
the inhalation of toxic smelter emissions. Another correspondent (VI-D-32)
commented that on windless days, she cannot jog without getting a chemical
taste in her mouth. Another individual (IV-F-3.10) stated that from a health
standpoint primary interest is in the particles inhaled into the lower
lung, approximately in the range of 0.5 microns to 5 microns. Another
correspondent (IV-D-360) submitted: "I jog every day and it causes no
problem."
Comment:
Four correspondents noted that arsenic-related health effects other
than lung cancer have not been addressed by EPA (IV-D-32, IV-D-85, IV-D-314,
IV-C-168).
Response:
The EPA agrees with the commenters that arsenic is toxic to humans. As
discussed in EPA's health asessment document, the acute and chronic toxicity
2-19
-------
of arsenic is dependent on the chemical form. Inorganic trivalerit arsenic
is more acutely toxic than inorganic pentavalent arsenic. The complex
organic arsenic form is regarded as nontoxic. Acute effects seen after oral
exposure include gastroenteritis, diarrhea, and cardiovascular effects.7
These effects can cause death. The precise lethal dose of inorganic arsenic
is unknown, however, the lethal dose of arsenic trioxide is estimated to
range from 70 to 180 milligrams.
Neurotoxic effects in humans have been observed following ingestion of
inorganic arsenic. These effects have varied with length and type of
exposure, as well as the pathway of exposure. Neuropathies have been
associated with chronic and acute exposures to high levels of inorganic
arsenic, and have included: peripheral nervous system effects characterized
by numbing or tingling in the hands and feet; neuralgia; peripheral neuritis,
muscular weakness, and memory loss. However, specific dose-response
relationships have not been established, especially to chronic low level
airborne arsenic exposure.
Cardiovascular effects of inorganic arsenic exposure have been observed.
A study in Taiwan indicated an occurrence of peripheral vasculopathy in a
population exposed to high levels of inorganic arsenic in the drinking
water, characterized by poor circulation resulting in gangrene of the feet,
legs or fingers. In epidemiological studies of smelter workers, peripheral
vascular disease has generally not been observed, although a few smelter
studies have found a significant increase in cardiovascular mortality.°
Studies of one copper smelter by Lee and Fraumeni (1969) and Lee-Feldstein
(1983) found a significant increase in cardiovascular mortality in the
workers (SMR=118 and SMR=129, respectively). No relationship to duration
of arsenic exposure was found. Higgins, et al. (1982), reported on the
same smelter workers, and found that cardiovascular mortality increased
with increasing ceiling arsenic exposure among smokers at 500-4999 ug/m^
(SMR=165). No effect was seen among nonsmokers. However, Lubine, et al.
(1981) did not find an excess of cardiovascular disease in their cohort
study of the same smelter workers (SMR=108). The confounding and conflicting
2-20
-------
findings of smelter exposure studies suggests that further research is
needed in this area.
Respiratory effects other than cancer have been observed in smelter
workers exposed to high airborne arsenic levels. Pulmonary insufficiency
and tracheobronchitis have been observed in smelter workers in the roaster
and furnace areas. Septa! perforations and rhinopharyngolaryngitis has
also been seen in copper smelter workers. However, the limited information
(dose-response data) did not permit the Agency to perform a risk assessment
for these particular health effects.
Comment:
Several general comments were submitted to the effect that arsenic is
an established health hazard (IV-D-421, IV-D-412, IV-D-419, IV-D-32, IV-D-401,
IV-D-115, IV-D-81, IV-D-112, IV-D-106, C-140, IV-D-92, IV-D-292, IV-D-62
IV-F-3.30, IV-F-3.43, IV-F-4.34). One writer (IV-D-81) expressed concern about
the effect of ASARCO arsenic emissions on the health of the citizens in the
community and particularly, on the health of his children. A second writer
(IV-D-412) simply maintained that arsenic causes pain and death. A resident
(IV-F-3.43) testified: "We are saying that we don't think arsenic is safe.
It's not sufficient for ASARCO to roll out a bunch of Eastern scientists
to come in here and tell us, 'Hey, folks, don't worry. We've read the
data; we've studied the issue.'"
Response:
Section 112 requires that EPA set standards that provide a "ample
margin of safety." Where a health effects threshold can be determined,
this requirement can be met by establishing the standard at a level that
insures that the exposure threshold is highly unlikely to be exceeded.
Where identifiable thresholds do not exist or are indeterminate, as with
carcinogens including inorganic arsenic, any level of control selected
short of an absolute ban on emissions, may pose a finite carcinogenic risk.
2-21
-------
The EPA believes that the final inorganic arsenic standards which permits
some level of residual risk provides an ample margin of safety to protect
public health.
Comment:
Correspondents (IV-D-76, IV-D-51, IV-D-32) discussed the potential
adverse health effects associated with the inhalation of arsenic. One
correspondent (IV-D-76) said that the risks from inhalation of (arsenic) in
the air and dust are also substantial and that the primary concern is for
infants, children, and pregnant women. Another correspondent (IV-D-51)
questioned to what extent the inhalation of arsenic is lethal. One
correspondent (IV-D-32) expressed concern about the possibility of children
absorbing arsenic by breathing playground dust.
Comment;
Many correspondents submitted comments referring to studies which have
shown high levels of arsenic in the blood, hair and/or urine of children
living near the copper smelters (IV-D-112, IV-D-76, IV-D-9, IV-D-11, IV-D-21,
IV-D-106, IV-D-107, IV-D-166, IV-D-422, IV-D-426, IV-D-417, IV-D-164,
IV-D-90, IV-D-33, IV-D-66 IV-D-404, IV-D-375). One writer (IV-D-166) claimed
claimed that (his) six year old son (had) the highest content of arsenic found
in the urine of all the children tested in the Olympia-Tacoma-Vashon Island
area. A second writer (IV-D-164) mentioned that urinary arsenic levels twice
normal were found in Island children and referred to an article in a local
paper in which children aged 0-5 months showed the highest arsenic levels.
Comment:
Testimony regarding arsenic tissue levels in children was also provided
(IV-F-3.2, IV-F-3.4, IV-F-3.15, IV-F-3.41, IV-F-3.53, IV-F-3.57, IV-F-3.60,
IV-F-3.74, IV-F-4.15, IV-F-4.49, IV-F-4.60, IV-F-4.62, IV-F-4.68, IV-F-5.8,
2-22
-------
IV-F-b.lB, IV-F-4.71). Samuel Milham Jr., M.D., M.P.H. (IV-F-3.2) discussed
a study he had conducted in which he determined that children who lived
near ASARCO had higher urinary and hair arsenic levels than those who lived
farther away. He attributed these high levels to the inhalation route of
exposure. Graphs on urinary and hair arsenic levels in children were
submitted by Dr. Milham (IV-F-3.81 IV-F-3.95).
Comment:
The authors of a published report entitled Monitoring and Reducing
Toxic Intake of Children Near the Tacoma Smelter and in South Park, Seattle
(IV-F-4.73) discussed in their testimonies the main results of this study.
the Tacoma Smelter and concluded that children with pica, or who eat dirt
and other materials they shouldn't, may have a significant arsenic intake.
For example, some children who live near the smelter have three times the
normal amount of arsenic in their urine. Also, some hair samples contained
20 times the usual amount of arsenic. Hair analysis can give a doctor an
idea of the long-term ingestion of arsenic". Mr. John Roberts (IV-F-4.10),
coauthor of this study, testified that both ingestion and breathing are
important routes of entry of arsenic, into children.
Comment;
Comments were also made that no illness resulted from an increase in
arsenic tissue levels. Dr. Milham (IV-F-3.2) indicated that hearing, chromosomal
analyses, growth and development, and blood levels were all normal and that
no anemia was found in the population studied (in contrast to other morbidity
studies of arsenic human tissue contamination). Another individual (IV-F-3.6)
testified that although arsenic levels in physiologic samples from children
are elevated close to the Tacoma smelter, no increase in illness or deaths
has been demonstrated.
Comment:
Some individuals (IV-D-604, IV-F-4.15, IV-D-593, IV-F-3.55, IV-D-719,
IV-D-726, IV-D-768, IV-D-741, IV-D-621-14.10, IV-D-579, IV-D-738, IV-D-670)
expressed concern about the cumulative effects of arsenic exposure. A
person (IV-F-3.55) stated that to accurately gauge our exposure to arsenic,
2-23
-------
UPA must include the historical and continuing accumulation of arsenic. How-
ever, two commenters indicated that arsenic does not accumulate (IV-D-621-15.6,
IV-D-621-16.12).
Response:
Individuals residing in the vicinity of sources of airborne arsenic
exposure, especially high arsenic copper smelters, may be at risk for
increased intake because of the concomitant exposure to arsenic in the air,
and arsenic deposited from the air onto soil and dust. Children may be
more susceptible than adults. A Japanese study of arsenic poisoning of
young children that had consumed arsenic-contaminated infant milk formula
showed a number of indications of central nervous system involvement.9
Follow-up studies showed significant cases of abnormal brain patterns,
masked cognitive deficiencies, severe hearing loss and behavioral problems.
Unfortunately, no specific dose-response curves were developed either
in the child poisoning studies or in the female smelter worker studies
relating arsenic exposure to the manifestation of an effect. In the latter
study no certainty was expressed that indeed airborne arsenic exposure
caused the observed spontaneous abortion rate. Although indicative of a
positive response to arsenic exposure, no extrapolation of an estimate of
risk can be done with the data base. With respect to risk to children
absorbing arsenic by inhaling playground dust, no inferences of risk can be
made from the arsenic ingestion and poisoning studies of Japanese children.
In addition, the mechanisms of inorganic arsenic deposition onto soil surfaces
from smelter emissions and consequent adsorption onto soil and dust surfaces
are not well understood. Given the extent of knowledge concerning deposition,
transport and surface clearance of inorganic arsenic as it passes from the
air media to soil and dust, EPA cannot accurately assess the cumulative
effects of arsenic exposure nor was EPA able to assess the relative risk of
these noncarcinogenic health responses. However, EPA in its decision making
process is aware of the possible risk to sensitive individuals, and does
consider this in conjunction with results of quantitative risk modeling.
2-24
-------
Comment:
A few correspondents (IV-D-60,1V-D-76, IV-D-114) discussed the potential
adverse health effects associated with the ingestion of arsenic through
sources other than vegetables grown in local gardens such as ingestion of
arsenic in seafood (IV-D-718, IV-D-739, IV-D-515, IV-D-530).
Comment:
One correspondent maintained that several populations ingest more
arsenic in their drinking water than those who live near smelters will be
exposed to, and that lifetime studies of those persons show no ill effects
(this statement was documented with the following report: EPA 600/1-81-064).
Response:
As discussed in the previous response, the final risk assessment
addresses only the inhalation of arsenic emitted by the smelter. The form
of arsenic contained in seafood and its toxicological properties are different
from the inorganic forms of arsenic regulated under the proposed air emissions
standard. While shellfish and other marine foods have the highest arsenic
level of any food category, the arsenic in marine species is stored in
complex organoarsenical forms. Based on recent reports these forms are
assimilated by man and rapidly excreted intact. They are not metabolized
like the inorganic forms being regulated. Toxicologically, the organic
forms of arsenic contained in seafood are relatively inert.
Comment:
A few people questioned the level at which arsenic presents a health
hazard (IV-D-164, IV-D-267, IV-F-4.28, IV-F-5.15, IV-D-622, 1V-D-756).
Comment:
Concern was also expressed about the health hazards of smelter emissions
(IV-D-595, IV-D-110, IV-D-105, IV-D-83, 1V-D-404, IV-D-375, IV-D-329, IV-D-592,
IV-F-3.58, IV-F-3,70, IV-F-4.10, IV-F-4.28, IV-F-4.52, IV-F-5.3). One writer
2-25
-------
(IV-D-HU), with no specific reference to arsenic, described a disease in-
volving a breakdown of the immune system which she attributed to the daily
bombardment of toxic chemicals; she referred to an article which appeared
in the L.A. Times concerning the ASARCO Tacoma smelter emissions. A second
writer (IV-D-105) indicated that smelter workers will have an obvious health
hazard. The populace within several miles will have a less obvious but
real health loss. A third writer (IV-D-83) has been concerned about the
smelter pollution and how it might affect the health of his family for
years.
Response:
Clinical pathology reports of arsenic exposure have reported on the
role of inorganic arsenic as an immunosuppressant in humans. This is
evident in the use of arsenical medicinals in the treatment of steroid-
responding disorders, and as a lytiphocytostatic agent. Reports of chronic
consumption of high arsenic contaminated drinking water supports the immuno-
suppressant role of arsenic. Chilean children exposed to the water displayed
histories of chronic cough and bronchitis.^ Other arsenic exposure studies
have observed the occurrence of herpes simplex, and chronic pulmonary
infections and this is evidence of arsenic as an immunosuppressant. There-
fore, it is possible, although not yet clearly defined, that long-term
exposure to airborne arsenic in the vicinity of copper smelters may contri-
bute to disease patterns within the community. Further research in this
area is needed to describe a possible association.
2.1.3 Teratogenicity/Reproductive Effects
A number of commenters expressed concern regarding the potential adverse
effects of ASARCO emissions on the fetus (IV-D-4, IV-D-593, IV-D-604, IV-F-3.37,
IV-F-3.41, IV-D-158, IV-F-3.42, IV-F-3.53, IV-F-3.57, IV-F-4.6, IV-F-4.11,
IV-F-4.68, IV-F-5.7, IV-F-5.8). One correspondent (IV-D-593) referred to a
statement made by Dr. Karle Mottet of the University of Washington at a
meeting in Tacoma in which he indicated that arsenic may cause birth defects.
Another correspondent (IV-D-604) questioned whether EPA was concerned about
2-26
-------
potential birth defects. One writer (IV-D-158) stated that it is apparent
that arsenic exposure cause increased incidence of birth defects and mis-
carriages. One individual (IV-F-5.7) stated that the more immediate risk
of the involuntarily terminated pregnancy or even a retarded child is more
threatening and of greater concern than the possibility of getting cancer
in the more distant future. A student (IV-F-4.68) from the University of
Berkely claimed that fetuses, newborns, and children are particularly
vulnerable to the effects of arsenic toxicity. She submitted a paper that
she wrote entitled ASARCO Arsenic Toxicity and the Public Health (IV-F-4.71).
Response:
Teratogenic effects of arsenic compounds have been observed in animal
studies using a variety of species.H One study observed malformations in
hamster fetuses following intravenous injection of sodium arsenate into the
pregnant female on the eighth day of gestation. Exencephaly, encephaloceles,
and skeletal defects were observed. Another hamster study of similar
design observed embryos with a delay in neural fold elevation and neural
tube closure with arsenate exposure. In experiments with mice, increased
fetal resorption, decreased fetal weights, cleft lip, fork ribs and fused
vertebrae were apparent following a single injection of sodium arsenate
(intraperitoneally). Other animal studies have exhibited clear indications
of fetal teratogenicity. Animal studies on the effects arsenic may have on
postnatal growth and development have not observed any "ef fecit.
Swedish studies (Nordstrom and co-workers) of female smelter
workers, and of females residing in the vicinity of smelters, have suggested
an increase in the rate of spontaneous abortions resulting from exposure to
smelter pollutants.12 Female smelter workers showed a prevalence of spon-
taneous abortions which was 17 percent above what was expected. Women who
worked directly on smelter processes showed a spontaneous abortion rate 28
percent higher than other female smelter workers. Women residing in the
vicinity of the smelter displayed a spontaneous abortion rate 7 to 11
percent above the expected rate, with the highest rate in the area closest
2-27
-------
to the smelter. However, many female smelter workers were reported to
reside in this area.
However, Nordstrom et al. studies are flawed for a number of reasons.
The Health Assessment Document cautions: "These studies (Nordstrom1s) were not
designed specifically to study effects of arsenic but rather to study the
effects, in general, of the smelter work. While data from these studies
suggest a low-level effect of smelter pollutants on neighboring (proximate)
populations, the diverse agents involved preclude making conclusive statements
about the specific effects of arsenic." In addition, unbeknownst to
EPA at the time of the HAD publication, the Swedish National Health Board
Expert Committee published a report in 1981 that questioned almost every
finding in the Nordstrom studies. In the Administrator's judgment, the
Nordstrom studies are not suitable for making determinations regarding the
potential human reproductive effects caused by arsenic exposure. Therefore,
it is not possible to relate arsenic exposure to the reproductive effects
observed. The risk assessment methodology employed by EPA focused on the
risk of respiratory cancer. However, EPA, in its overall evaluation of
adverse health effects, will qualitatively regard other indications of
arsenic exposure as well. Two health studies are being undertaken "by the
Washington State Department of Social and Health Services to assess the
potential impact of smelter pollutants, especially arsenic. The study
parameters will include incidences of reduced birth weight, and teratogenic
effects (oral cleft) in areas affected by smelter emissions. These
observations will be compared to areas remote from smelter emissions to
determine the effects from ASARCO Tacoma smelter pollutants (see page
2-42).
2.1.4 Systemic Effects of Arsenic Emissions/Smelter Emissions
Comment:
One writer (IV-D-41) questioned why kidney damage, a "main effect of
arsenic" exposure, was not being considered. Another writer (IV-D-404)
claimed that both arsenic and cadmium are known to cause kidney failure in
humans. One individual (IV-F-3.37) stated that arsenic accumulates in and
is excreted from the kidneys.
2-28
-------
Comment:
Four correspondents (IV-D-94, IV-D-430, IV-D-417, IV-0-592) addressed
the Inpact of ASARCO emissions on asthmatics. One correspondent (IV-D-417)
asserted that the dark cloud from ASARCO Tacoma has caused asthma sufferers
increased agony.
Comment:
Many people also testified that they had respiratory problems associated
with the ASARCO smelter emissions (IV-F-3.20, IV-F-3.24, IV-F-3.57, IV-F-4.10,
IV-F-4.36, IV-F-4.52, IV-F-5.4, IV-F-5.10, IV-F-5.11, IV-F-5.16). One person
(IV-F-5.10) testified that some people in the area (of the smelter) have
respiratory and sinus problems. Another person (IV-F-5.4) stated that there
are days when the odor in the air is so bad that they develop an asthmatic
condition and cannot breathe. Another person (IV-F-5.11), asserted that as a
youngster he lived in one of five closest houses to the ASARCO Tacoma
smokestack and he recalled playing with extensive pain in his lungs.
Comment:
In contrast, six correspondents maintained that, although they had
lived near and/or worked at ASARCO Tacoma for many years, they had never had
any respiratory problems (IV-D-265, IV-D-233, IV-D-362, IV-D-391, IV-D-465,
IV-D-215). One correspondent (IV-D-391) commented that his family had
lived within one and one-half miles of the smelter for a combined total of
over 800 years and not experienced any serious respiratory disorders.
Response:
Systemic effects other than cancer, resulting from chronic exposure to
airborne inorganic arsenic have been noted in epidemiological studies. One
study of smelter workers handling refined arsenic displayed nasal septum
perforation and rhinopharyn-golaryngitis. Workers in roaster, furnace and
converter smelter processes showed tracheobronchitis and pulmonary insuffic-
iency. Hepatic effects have been observed in arsenic ingestion studies
2-29
-------
involving chronic exposure to high arsenic contaminated water or arsenical
medicinals. These effects are cirrhosis and hypertension. Other observed
chronic systemic effects are reversible anemia, and reduced hemoglobin
production. Chronic renal effects related to arsenic ingestion or inhalation
are not well characterized. Chilean children exposed to arsenic in drinking
water showed a chronic cough and bronchitic history.
Chronic systemic effects other than cancer of either high level or low
level inhalation exposure to airborne inorganic arsenic from copper smelters
are not well understood or defined. Dose specific responses in the afore-
mentioned studies were not reported. Therefore, a determination of increased
risk of noncancerous systemic effects within the community affected by
smelter pollutants cannot be evaluated at this time.
Comment:
Comments were submitted regarding tissue levels of arsenic in other
community members (IV-F-3.67, IV-D-428, IV-D-418, IV-D-428, IV-D-604, IV-F-3.21),
One writer (IV-D-428) maintained that he and his wife had blood tests which
showed lead and arsenic contamination. Another writer stated that emissions
from the smelter had poisoned the blood of three generations in the town
of Ruston, Washington. One individual (IV-F-3.21) testified that his urine
tested positive for an arsenic contamination level of 20 micrograms per
liter. Arsenic tissue levels in animals was discussed. One correspondent
(IV-D-164) indicated that a local butcher had noted that the livers of
slaughtered animals were unusually spotted. One commenter (IV-F-3.37)
referred to a television documentary, "Green Grow the Profits," in which
arsenic was initially reported in the livers of poultry and later found in
the white meat as well.
Response:
Urinary arsenic levels have been shown to increase when arsenic is
inhaled. Arsenic may also be excreted via hair. The studies cited above
provide additional evidence for EPA's assertion that the population is
2-30
-------
being exposed through inhalation to arsenic emitted by the smelter. The
urinary arsenic studies also indicate that exposure is highest near the
plant. This is in agreement with both modeling and ambient monitoring
results.
However, EPA does not routinely use tissue levels as a measure of
public exposure to smelter emissions or lung cancer risks. The primary
reason is that arsenic concentrations in tissue reflect many factors in
addition to the inhalation of arsenic emitted by the smelter. Diet, in
particular the consumption of seafood, can occur and result in increases or
decreases in tissue levels. Individual metabolism can also cause varia-
tions in the amount of arsenic excreted. Individuals living in the same
area from which tissue samples are taken on the same day may show a range
of. arsenic levels. Therefore, arsenic levels in tissue may not be good
estimators for exposure to air emissions from ASARCO since other sources of
exposure can contribute to arsenic concentrations uncovered in the tissues.
2.1.5 Dermal Effects of Arsenic Emissions/Smelter Emissions
Comment:
Some correspondents (IV-138, IV-D-247, IV-D-613, IV-D-622, IV-D-16
(IV-D-772)) discussed the dermal effects which they believed were associated
with ASARCO Tacoma emissions. One writer attributed his wife's itching,
welting rash to the handling of objects left outside which had accumulated
dust emitted from the ASARCO smokestacks. Another writer (IV-D-217) who
worked at ASARCO Tacoma for 36 years claimed that he suffered from skin
irritation.
Comment:
Dermal effects connected with arsenic and/or employment at copper
smelters were also discussed at the hearings. One individual (IV-F-4.27)
who worked in some of the "worst areas" of the smelter stated that there
were some skin irritations, but these were taken care of with no ill effects
(IV-F-4.27). Another person (IV-F-3.18) commented that arsenic is a skin
2-31
-------
irritant if not handled properly but that it would kill bacteria, and germs.
Another individual (IV-F-4.60) testified that arsenic in significant or
toxic levels, would produce skin pigmentation of certain areas of the skin,
the neck, eyelids, nipples and arm pits and that the skin may be thickened
in these areas.
Response:
Although there are no known cases of skin disorders resulting from
arsenic inhalation in man, chronic oral exposure to arsenic induces a
sequence of changes in skin epithelium, proceeding from hyperpigmentation
to hyperkeratosis, characterized as keratin proliferation of a verrucose
nature and leading, in some cases, to late onset skin cancers.13 The U.S.
EPA is presently examining this information, along with information from
other studies, in order to determine whether quantitative does-response
relationships, similar to those seen for skin cancer, can be established
for these precancerous skin lesions. However, health effects other than
lung cancer which could result from chronic low-level exposure to arsenic
have not been sufficiently documented for EPA to quantitatively estimate or
model. This kind of health risk is considered by EPA in a qualitative
manner during the decision-making process.
2.1.6 Potential for Health Effects from the Ingestion of Arsenic Contaminated
Vegetables
Comment;
According to one resident, the King (Seattle) and Pierce (Tacoma) County
Health Departments distributed booklets which warned against the consumption
of certain vegetables grown in local gardens because of cadmium, arsenic,
and other heavy metals in the soil which had accumulated from smelter effluent
(IV-D-11). Many of the comments attested to this warning from the County
Health Departments (IV-D-9, IV-D-21, IV-D-32, IV-D-49, IV-D-76, IV-D-404,
IV-D-375, IV-D-164, IV-D-292, IV-D-434, IV-D-428, IV-F-3.51, IV-F-3.53,
IV-F-3.60, IV-F-4.15). Other individuals, with no specific reference to
this warning, indicated that they would no longer grow and/or consume local
vegetables because of soil contamination (IV-D-21, IV-D,#3, IV-D-605,
IV-D-38, IV-D-47, IV-D-71, IV-D-91, IV-D-92, IV-D-100, IV-D-591, IV-D-158,
IV-D-439, 1V-F-3.37, IV-F-3.53, IV-F-4.52, IV-D-82, IV-D-104). One
2-32
-------
correspondent (IV-D-74) stated that lives have already been altered by it
(the ASARCO smelter), and that gardening plans and dietary routines had to
be changed.
Comment:
Testimony was also given on changes that had to be made with respect
to the consumption of local fruits and vegetables because of the ASARCO Tacoma
smoke stack emissions. One individual (IV-F-5.4) testified that her fruit has
to be peeled because it has a bad taste and tackiness that can't be washed
off. Another person (IV-F-3.72) admitted giving up spinach, kale, chard,
potatoes, carrots, and other vegetables because of ASARCO (IV-F-3.72).
Comment:
Many residents reported that the foliage in their gardens and yards
were burnt or damaged in some fashion by smelter emissons (IV-D-23, IV-D-89,
IV-D-94, IV-D-138, IV-D-417, IV-D-418, IV-D-425, IV-D-427, IV-F-3.24,
IV-F-3.29, IV-F-3.38, IV-F-4.41, IV-F-5.13, IV-F-5.22).
Comment:
Comments were submitted in which concern was expressed about high
arsenic levels in the vegetation and/or soil (IV-D-34, IV-D-90, IV-D-116,
IV-D-138, IV-0-419, IV-D-364, IV-F-3.20, IV-F-3.21, IV-F-3.55, IV-F-3.57,
IV-F-3.74, IV-F-4.10, IV-F-4.15, IV-F-5.8, IV-F-5.22, IV-D-718, IV-D-739,
IV-D-515, IV-D-530, IV-D-576, IV-D-666, IV-D-676, IV-D-694, IV-D-584,
IV-D-605).
Comment:
In contrast, numerous commenters claimed that their vegetables suffered
no ill effects from the ASARCO Tacoma smelter emissions (IV-D-130, IV-D-139,
IV-D-135, IV-D-475, IV-D-210, IV-D-372, IV-D-306, IV-D-365, IV-D-273, IV-D-599,
IV-D-348, IV-D-479, IV-D-341, IV-D-253, IV-D-201, IV-D-218, IV-D-242, IV-D-249,
IV-D-451, IV-D-270, IV-D-159, IV-D-176, IV-D-166, IV-D-282, IV-D-277, IV-D-352,
IV-D-345, IV-D-327, IV-D-298, IV-D-487, IV-D-354, IV-D-355, IV-0-490, IV-D-477,
IV-D-394, IV-D-379, IV-D-385, IV-D-473, IV-D-482, IV-D-601, IV-D-476, IV-D-265,
IV-D-266, IV-D-377, IV-D-397, IV-D-472, IV-F-3.23, IV-F-3.49, IV-F-4.3,
IV-F-4.5, IV-F-4.37, IV-F-4.49, IV-F-4.69, IV-F-5.5, IV-D-625-13, IV-D-725,
2-33
-------
iV-D-784, IV-D-769, IV-D-699, IV-D-700, IV-D-705, IV-D-533, IV-D-539, IV-D-613,
IV-D-623, IV-D-624). One writer (IV-D-135) observed that the vegetation in
the "Point Defiance Park" next to the smelter was unaffected. Another
writer (IV-D-451) maintained that mixtures of arsenic and flour or cereal
spread on plants as a pesticide had been consumed by humans and livestock
with no ill effects. Another writer (IV-D-345) indicated that family
members had been eating from their gardens since 1909 and felt that no harm
had come to any of them.
Comment:
One individual (IV-F-3.13) stated that it's quite obvious that any harm-
ful effects would depend on the amount of a given vegetable eaten. Previous
calculations have shown that it would be impossible for anyone to consume
toxic amounts of any vegetable grown in the area.
Response:
The deposition of airborne inorganic arsenic emissions from inorganic
arsenic sources onto the soil surface is of concern to EPA. The EPA is
cooperating with various state agencies in a comprehensive study of smelter
emissions from the ASARCO smelter to determine the routes of exposure
responsible for the elevated urinary arsenic levels found in children
residing near the smelter. Because the Clean Air Act limits the scope of
exposure assessment to hazardous substances in the air, this study is
directed under the authority of Superfund (Comprehensive Environmental
Response, Compensation, and Liability Act, CERCLA). The multimedia approach
will include exposure assessments of inhalation of arsenic in the air or in
resuspended dust; ingestion of arsenic from vegetables, drinking water, and
ingestion of soil by children. These exposure media will be sampled
concurrently with urine, and statistical methods will be applied to determine
which exposures have caused the elevated urinary arsenic levels and what
remedial actions may be needed to reduce these exposures. Assessment will
also be made of the potential health problems associated with lead and
cadmium emissions from the smelter (see page 2-42).
2-34
-------
2.1.7 Effects of Arsenic Emissions/Smelter Emissions on Other Biological
Systems
Comment:
According to the submitted comments, bees and bottom fish are the types
of wildlife most affected by the ASARCO Tacoma smelter emissions (IV-D-422,
IV-D-418, IV-D-106, IV-D-107, IV-D-115, IV-D-404, IV-F-3.21, IV-F-3.57). One
correspondent (IV-D-412) wrote that a Seattle daily newspaper reported
scientific research linking arsenic emissions to widespread failure of
beehives north of the plant and that bottom fish have been shown to have
high levels of arsenic. Four references were made to Dr. J. Bromenshank's
study on arsenic levels in bees ranging from "12 ppm at the South end of
the Island (a world's record) to 2 ppm at the North end of the Island"
(IV-D-9, IV-D-76, IV-F-3.53, IV-F-3.60). Dr. Bromenshank (IV-F-3.17)
discussed his study at the hearings and stated that the (arsenic) levels
are certainly high enough to equal or exceed those reported to be hazardous
or lethal to honeybees and that arsenic typically acts as a stomach poison
in insects. Dr. Bromenshank submitted data sheets from his study (IV-F-3.100
IV-F-3.103). Other concerns about the effects of arsenic on the bee population
ranged from fatalities of bees due to arsenic contamination (IV-D-38, IV-D-90,
IV-D-783, IV-D-705) to the residents' inability to consume local honey
(IV-D-47).
Comment:
Comments were submitted regarding the effects of the ASARCO-Tacoma smoke-
stack emissions on other animals (IV-D-581, IV-D-20, IV-D-434, IV-D-76,
IV-D-599, IV-D-784, IV-D-137, IV-D-429). One correspondent (IV-D-20)
questioned the effect of arsenic and other heavy metal emissions on the
entire animal life chain. Another correspondent (IV-D-76) questioned the
risks posed by the consumption of fish and shellfish from the area. A
correspondent (IV-D-599) questioned the correlation between arsenic exposure
and birth defects in fish and wildlife.
Comment:
In contrast, two correspondents (IV-D-218, IV-D-160) felt that the
ASARCO-Tacoma emissions are not harmful to the wildlife. One correspondent
maintained that there is no shortage of slugs, snails, grasshoppers, tent
caterpillars, or gypsy moths around the smelter.
2-35
-------
Response:
Section 112 of the Clean Air Act specifically requires the EPA
Administrator to establish standards for hazardous airborne pollutants
which provide an ample margin of safety to protect the public health.
Therefore, consideration of ecological damage to aquatic organisms and other
biota would be secondary to evaluation of direct human health effects under
section 112 of the Clean Air Act. However, general ecological effects are
being investigated under Superfund and other statutes to determine impacts
of arsenic emissions from the smelter. Briefly summarized, these efforts
include: (1) An assessment of the effects on aquatic life of contaminated
discharges into Commencement Bay from the ASARCO smelter and other industries,
and of sediments and water that are known to be contaminated; (2) A determina-
tion of whether or not additional studies are warranted under Superfund to
investigate adverse effects of smelter emissions on other plant and animal
life. These studies may include samples of tissue levels of arsenic in
livestock (seepage 2-42).
2.1.8 ASARCO-Tacoma Smelter Emissions/Arsenic Not a Health Hazard
Several people opposed the proposed standard and/or shutdown of the
ASARCO-Tacoma smelter, claiming that the smelter was not a health hazard
(IV-D-62116.12, IV-D-621-.14.9, IV-D-621-5, IV-D-621-6, IV-D-509, IV-V-568,
IV-D-14, IV-D-621-14.17, IV-D-621-14.7, IV-D-621-6.1, IV-D-621-14.2,
IV-0-525, IV-0-536, IV-0-621-14.14, IV-D-621-15.2, IV-D-6621-15.6,
IV-D-621-15.9, IV-F-3.15, IV-D-547, IV-D-760, IV-D-695, IV-D-323, IV-D-337,
IV-0-343). Comments were received from those who had lived near the smelter,
who were employed by the smelter, and who both lived by and worked for the
smelter. Comments were also submitted by those who, although they made no
mention of living near or working for ASARCO-Tacoma, felt that the plant
didn't pose a health hazard.
2-36
-------
Comment:
Many comments were submitted by individuals who lived in the vicinity
of ASARCO (IV-D-262, IV-D-222, IV-D-146, IV-D-184, IV-D-242, IV-D-605
IV-D-367, IV-D-354, IV-D-298, IV-D-352, IV-D-377, IV-D-327, IV-D-369,
IV-D-66, IV-D-479, IV-D-341, IV-D-472, IV-D-469, IV-D-470, IV-D-508, IV-D-471,
IV-0-176, IV-D-394, IV-0-655, IV-D-360, IV-D-372, IV-D-391, IV-D-211, IV-D-318,
IV-F-3.22 IV-F-3.25, IV-F-3.26, IV-F-3.27, IV-F-3.35, IV-F-3.39, IV-F-3.45,
IV-F-3.50, IV-F-3.52, IV-F-4.3, IV-F-4.7, IV-F-4.12, IV-F-4.30, IV-F-4.33,
IV-F-4.44, IV-F-4.53, IV-D-280, IV-D-315, IV-D-393, IV-D-517, IV-D-525, IV-D-532,
IV-D-533, IV-D-534, IV-D-539, IV-D-134, IV-D-544, IV-D-547, IV-D-548,
IV-D-552, IV-D-615, IV-D-623, IV-D-624, IV-D-633, IV-D-636, IV-D-659). One
writer (IV-D-66) stated that having lived in Tacoma 55 out of 66 years, he
found no evidence of ill health from arsenic emissions including the health
of his mother, age 103, who had lived in the vicinity longer. Another
writer (IV-D-176) indicated that he had lived within a few miles of the
smelter since 1937 and had suffered no ill effects. Another writer (IV-D-360)
asserted that her family had lived in the five mile radius of the smelter
for almost thirty years, that she raised four children in that area, and
that her children were healthy adults beginning to have children of their own.
Comment:
Testimony was also given by those who lived near ASARCO-Tacoma. One
person (IV-F-4.5) testified that he had five neighbors who lived in the area of
the smelter for over 60 and 70 years and hadn't complained of any ill effects
(IV-F-4.5). Another individual (IV-F-4.32) lived within 7 blocks of the
smelter for the last 57 years and had not suffered any health effects from
the smelter nor had his children who had attended a school 3 blocks south
of the plant. One woman (IV-F-4.49) who also lived 7 blocks from the smelter
felt that the stress from making a living was more of a health hazard than
the arsenic from the plant. She also stated that studies conducted in the
U.S. and Sweden indicate no increased illness or mortality associated with
community exposure to smelters emissions and no increased rate of lung
cancer has been observed among persons exposed to ASARCO-Tacoma emissions.
2-37
-------
Comment;
Numerous commenters discussed the lack of a health hazard associated
with their past and/or present employment with ASARCO (IV-D-182, IV-D-197,
IV-D-208, IV-D-212, IV-D-218, IV-D-220, IV-D-221, IV-D-166, IV-D-225, IV-D-
226, IV-D-229, IV-D-159, IV-D-268, IV-D-223, IV-D-387, IV-D-504, IV-D-486,
IV-0-490, IV-D-656, IV-D-506, IV-D-385, IV-D-311, IV-D-289, IV-D-293, IV-D-299,
tV-F-3.59, IV-F-4.8, IV-F-4.40, IV-F-4.42, IV-F-4.58, IV-D-285, IV-D-350,
IV-D-512.IV-D-518, IV-D-522, IV-D-532, IV -D-347, IV-D-544, IV-D-547,
IV-D-558, IV-D-562, IV-D-613, IV-D-623, IV-D-636, IV-D-647, IV-D-563). One
correspondent (IV-D-221) submitted that he has been working for the Tacoma
smelter for 24 and a half years, that he carried arsenic every day for 18
years without a mask, and that he is now 75 years old and in the best of
health. Another writer (IV-D-486) commented that during fourteen years of
employment at the Tacoma smelter, he had suffered no ill effects.
In his testimony one retiree (IV-F-4.13) indicated that he had worked at
the plant for 32 years at various jobs and never felt sick throughout his
employment. Another person (IV-F-4.45) employed by ASARCO for 14 years claimed
that "in one week I breathed, inhaled and ingested more arsenic powder than
local residents would in 50 years and I can say that there have been no ill
effects to me". Another person (IV-F-5.6) testified that during his approxi-
mately thirty years of employment with ASARCO, he and his coworkers were
exposed to arsenic dust for "hours on end" and that they are all in fairly
good health.
Comment:
Thirty individuals based their opinions of the adverse health effects
caused by the smelter emissions on their experiences while both living near
and working at the plant (IV-D-199, IV-D-249, IV-D-236, IV-D-233, IV-D-202,
IV-D-165, IV-D-397, IV-D-482, IV-D-473, IV-D-492, IV-D-453, IV-D-379,
IV-D-348, IV-D-407, IV-D-306, IV-D-364, IV-D-335, IV-D-287, IV-D-196,
IV-D-297, IV-D-334, IV-F-3.18, IV-F-3.47, IV-F-4.17, IV-F-4.29, IV-F-4.36,
IV-F-4.37, IV-F-4.64, IV-F-5.17, IV-F-5.19, IV-F-5.5). One correspondent
(IV-D-407) wrote: "My home has been within 1/2 mile of the smelter for 65
2-38
-------
years. My husband has worked there for 44 years. My son worked there for
6 months. My father worked there also. None of these men including myself
have had any ill effects from the smelter". Another correspondent (IV-D-199)
maintained that he had lived and worked in the Tacoma area for 31 years and
he felt that the ASARCO-Tacoma smelter emissions had no ill effects on
anyone's health in that period of time.
Comment:
Several people generally felt that the ASARCO smelter posed no health
hazards (IV-D-304, IV-D-266, IV-D-250, IV-D-253, IV-D-160, IV-D-326, IV-D-
460, IV-D-272, IV-D-505, IV-D-279, IV-D-291, IV-D-320, IV-0-474, IV-D-371,
IV-D-399, IV-D-373, IV-D-321, IV-D-331, IV-D-339, IV-D-370, IV-D-276,
IV-D-72, IV-D-204, IV-D-227, IV-D-269, IV-D-406, IV-D-450). One correspondent
(IV-D-276) submitted that to date, no definite health problems have been
proven that will and do exist on current emissions from the plant. Another
correspondent (IV-D-460) said: "Since the EPA cannot show that the emissions
from ASARCO-Tacoma are harmful to this community, it would seem prudent to
me for you to drop your case".
Comment:
Testimony was given by one individual (IV-F-3.28) in which he referred
to arsenic's medicinal uses: "In medicine it is used in treatment of anemia
to build up red corpuscles of the blood and hemoglobin content. It has a
tonic effect on the general nervous system and it is also considered by
many authorities to have antiperiodic action, as in malaria. It is known
to be effective in various chronic skin diseases. It is used in the treat-
ment of certain forms of dyspepsia, Hodgkin's disease, neuralgia, rheumatoid
arthritis, chorea, asthma, hay fever, psoriasis, pemphigus, occasionally in
chronic eczema, tuberculosis, diabetes, leprosy, and syphilis".
Response:
A clear absence of adverse health effects, especially lung cancer, has
not been demonstrated by various community health studies. Although subject
to several shortcomings, several national community studies have indicated
2-39
-------
an increase risk of lung cancer in people residing near smelters.14
(See the summary of the community studies on page 2-9.) Other studies have
not demonstrated an excess of lung cancer mortality in communities surrounding
smelters (Rom et al. (1982); Lyon et al. (1977); Frost et al. (1983)). Due
to the inherent problems with such studies and the inconsistent findings that
they have produced, the community studies have not produced a clear under-
standing of the nature and magnitude of public risk near arsenic sources.
However, uncertain results and negative observations may not be construed
as an absence of risk to the public in view of the strong epidemiological
association between inorganic arsenic and lung cancer in smelter workers.
The EPA is taking the prudent action of reducing the risk of lung cancer
resulting from chronic community exposure to airborne arsenic emissions from
smelters. The Regulatory Council (an inter-governmental agency cancer policy
work group) has observed:
"The failure of an epidemiological study to detect an association
between the occurrence of cancer and exposure to a specific substance
should not be taken to indicate necessarily that the substance is not
carcinogenic.
Because it is unacceptable to allow exposure to potential
carcinogens to continue until human cancer actually occurs, regulatory
agencies should not wait for epidemiological evidence before taking
action to limit human exposure to chemicals considered to be carcino-
genic."15
2.1.9 Multiple Chemical Exposure: Synergistic/Additive Effects
Comment;
Several comments were submitted concerning the synergistic/additive
effects of exposure to multiple substances (IV-D-114, IV-D-322, CC, IV-F-3.37,
IV-F-3.55, IV-F-4.43, IV-F-4.50, IV-D-416, IV-D-438, IV-D-35, IV-D-6,
IV-0-718, IV-D-719, IV-D-710, IV-D-427, IV-D-541, IV-D-670, IV-D-57). One
writer stated that due to multiple contaminants from ASARCO-Tacoma smelter,
2-40
-------
single substance studies of health effects are inadequate. Another writer
(IV-D-114) asked to what degree do (arsenic, cadmium, sulfur dioxide, lead,
etc.) interact with each other and with other industrial substances to
create additional toxicity? Another writer felt that "by proposing an
arsenic standard separate from other pollutants coming from the ASARCO
smelter, the problem is divided into many segments. Each of them is less
dangerous than the sum" (IV-D-593). "One person (IV-F-3.7) cited the
conclusions of a study by Lee and Fraumeni and testified: "He know that
the ASARCO smelter emits both the sulfur dioxide and the arsenic trioxide,
which could mean that a synergistic effect is already in place in those for
us who live downwind from the emissions." Another individual (IV-F-4.31)
stated that arsenic probably becomes more toxic when it acts synergistically
with other substances so that the total exposure is greater than the sum of
the individual levels of pollution.
Response:
The EPA realizes there may exist a concomitant risk associated with
exposure to air pollutants from smelters. The Agency believes that con-
sideration of all environmental concerns associated with smelter emissions
is a necessary and important element in the risk management process.
Consequently EPA considered the impact of the proposed standard on emissions
of other pollutants, and the actions being taken under other environmental
statutes to address other environmental impacts of the smelter. The emission
of cadmium, lead, and antimony, for example, present in particulate matter
will also be controlled under the proposed arsenic standard.
The risk associated with $03 exposure have been statistically isolated
from risks associated with arsenic exposure. The data indicate that SOg
exposure does not explain the excess lung cancer rates observed. Also,
indications of excess lung cancers have been found in occupational settings
other than primary copper smelters where concomitant exposure to S02 and
other trace metals would not occur. The arsenic potency estimates (unit
risk estimates) for both types of occupational settings are approximately
of the same magnitude. These observations lead EPA to believe that excess
lung cancer risks are associated only with arsenic exposure.
2-41
-------
Although the Agency is aware that risk associated with exposure
other than inorganic arsenic inhalation may occur, the available data are
mostly inadequate as a basis for the Agency to produce meaningful additional
analyses. However, EPA and other Agencies are conducting or planning to
conduct further studies in and around Tacoma, Washington, to enhance the
available data base and to provide more insight as to the nature of other
routes of exposure and corresponding public risk. Such studies and other
activities are summarized below. These studies should provide useful
information on such impacts for all the smelters, although EPA realizes
that the ASARCO-Tacoma facility was smelting rather unique kinds of feed
material.
(1) Superfund Activities
Elevated levels of arsenic have been found in the hair and urine
of residents living near the ASARCO-Tacoma smelter. Additionally,
concentrations of arsenic are substantially above background in various
environmental media, including soil, air, household dust, and vegetation.
The Superfund law (Comprehensive Environmental Response, Compensation, and
Liability Act, CERCLA) is being used to address this multimedia arsenic
contamination. Superfund will also be used to evaluate potential problems
from cadmium which has been found in elevated levels in garden soil and
vegetables near the smelter.
Unlike most other environmental laws, Superfund can be used to correct
problems resulting from past practices and spanning all environmental
media. Based in part upon the elevated levels of arsenic in environmental
media and in urine samples of residents near the ASARCO smelter, a segment
of the Commencement Bay area (part of Commencement Bay and adjacent lands)
was designated as a Superfund (National Priority List) site in 1980. This
site is known as the Commencement Bay Near-Shore Tideflats Superfund site,
and includes parts of Tacoma/Ruston/Vashon Island, the Commencement Bay
Tideflats area, and the water adjacent to these areas.
2-42
-------
On May 2, 1983, EPA and the Washington Department of Ecology (WOOE)
signed a Cooperative Agreement making WDOE the lead agency in investigating
this Superfund site (with funds provided by the EPA Superfund program and
matching funds from the State) and in ensuring that needed remedial
actions are taken. This Cooperative Agreement is divided into two tasks,
the Ruston-Vashon Task and the Nearshore-Tideflats Task.
Ruston-Vashon Task-Investigations under this Task are focusing upon the
issues specifically related to the ASARCO smelter. An exposure assessment
study designed by the University of Washington (with assistance from the
Centers for Disease Control, WDOE, the State and local health agencies,
EPA and the Puget Sound Air Pollution Control Agency) began in January of
1985. The purpose of this study is to determine the routes of exposure
responsible for the elevated urinary arsenic levels found in children
living near the smelter. Since these exposure routes may include
inhalation of arsenic in air and in resuspended dust, ingestion of arsenic
from vegetables and drinking water and ingestion of soil and dust by children,
several of these media will be sampled concurrently with urine. Statistical
methods will then be used to determine which exposures are responsible for
the elevated urinary arsenic levels, providing information on the remedial
actions that may be needed to reduce these exposures.
Peripheral neuropathies (damage to nerves in the periphery of the body,
such as those in the arms or legs) have been found in persons exposed to
high levels of inorganic arsenic. Additionally, laboratory experiments
have shown that high levels of arsenic can affect the synthesis of
hemoglobin in exposed animals, resulting in higher than normal levels of
uroporphyrins in the urine. The investigation conducted by the University
of Washington includes urinary porphyrin analyses and peripheral
neuropathy testing to provide preliminary data on the effects of arsenic
in the smelter community.
Work is also being done as part of the Ruston-Vashon Superfund effort
to assess the potential exposures resulting from cadmium emissions from
the smelter. Cadmium levels above background have been found in the soil
2-43
-------
and vegetation near the ASARCO smelter, prompting the local health agencies
to suggest that the growth of certain vegetables (e.g., leafy) be discontinued,
Existing data on cadmium levels in garden soil and vegetables are now being
reviewed, and additional data will be collected, if necessary, to assess
what health problems, if any, may result from the levels of cadmium now in
the soil.
Several commenters expressed concern that ASARCO was damaging plant and
animal life in the vicinity of the smelter. No studies have been done on
the effects of these emissions, except for the pollutant sulfur dioxide
(SOe). The S02 plant studies done show sharp contrasts in opinion and
reflect conditions existing approximately ten years ago. However, analyses
of livestock tissue for levels of arsenic and other metals are being
considered under the Ruston/Vashon Superfund Task.
Nearshore-Tideflats TaskContamination of aquatic life in Commencement
Bay and the possible effects of this contamination on consumers of seafood
have been investigated in previous studies by NOAA (National Oceanic
and Atmospheric Administration), Tacoma-Pierce County Health Department
(TPCHD) and EPA. A NOAA report issued in 1980 reported the presence of
tumors in fish caught in Commencement Bay and higher than background
metals levels in limited areas (e.g., near ASARCO and other industries).
As a follow-up to this study, EPA analyzed additional samples of aquatic
life from the Bay in 1982. Using these data, the TPCHD concluded that
there did not appear to be short-term or long-term health risk from
consumption of fish caught in the Bay (except in Hylebos Waterway). TPCHD
recommended, however, that more data be developed for contaminants in fish
at the Point Defiance dock (near ASARCO) as well as at other areas in the
Bay.
These additional data are being collected as a part of the Superfund
investigations under the Nearshore-Tideflats Task of the EPA/WDOE
Cooperative Agreement. Under this Task, WDOE is analyzing the levels of
contaminants in Commencement Bay sediment and aquatic life and is
2-44
-------
investigating the sources of pollution that are responsible for these
contaminants. The effects of these contaminants on aquatic life and the
risk to consumers of eating seafood from the Bay are also being assessed.
In response to the initial results of this Superfund study, the TPCHD has
modified their previous advisory. They now recommend that individuals not
consume bottom fish or crab caught from the Commencement Bay Waterways and
limit consumption of fish and crabs caught in other areas of the bay. Upon
completion of the Superfund investigations, remedial actions will be designed
to control the discharge of contaminants to the Bay or remove existing
contaminants of concern (e.g., by removal of sediment).
(2) Non-Superfund Activities
Health RelatedTwo health related studies being conducted by the
Washington State Department of Social and Health Services (DSHS) to assess
potential impacts from smelter emissions are nearing completion. In these
studies, the incidences of reduced birth weight and oral cleft (a birth
defect that is easily detected) were compiled in areas near the smelter
and in control areas where exposures to smelter emissions are minimal.
The incidences in the two areas will then be compared in an attempt to
assess the smelter effects.
Dr. Tom Burbacher of the University of Washington is determining the
levels of arsenic in stillbirths and in newborns who have died soon after
birth. Samples of placenta from women living close to the smelter may
also be analyzed if funding is available. These studies will provide
information on the levels of arsenic in these various tissues and of the
potential for arsenic to be transferred to the fetus during development.
Lead Emissions - The Clean Air Act directs the Administrator to establish
air quality criteria and to propose and promulgate primary and secondary
National Ambient Air Quality Standards (NAAQS) for air pollutants emitted
from numerous and diverse sources that may reasonably be anticipated to
endanger public health or welfare. Primary standards are to be set at a level
which, in the judgment of the Administrator is required to protect public
health with an adequate margin of safety. Secondary standards must specify
a level of air quality which, in the judgment of the Administrator and
2-45
-------
based on the air quality criteria, is required to protect public welfare
from any known or anticipated adverse effects. In 1978, EPA established
the primary and secondary standards for lead at a level of 1.5 micrograms
per cubic meter (maximum arithmetic mean) averaged over a calendar quarter.
The control programs to meet the NAAQS are embodied in the State implemen-
tation plans (SIPs) which are developed by the State and local air agencies.
Lead is one of six pollutants for which the Agency has developed a NAAQS.
The SIP that WDOE developed for lead in Mashington has recently been
approved by EPA. Existing monitoring results suggested that the area
around the ASARCO smelter was meeting the NAAQS. However, to verify these
monitoring results PSAPCA and Region 10 EPA utilized smelter lead emissions
data in a dispersion model to estimate the expected maximum ambient lead
concentrations around the smelter. The results of this modeling showed
that lead emissions from the smelter would not violate the NAAQS even at
full operating capacity if ASARCO installed the controls required to reduce
the emissions of other pollutants.
In 1972, Dr. Sam Milham of DSHS studied the levels of lead in blood
and of blood enzymes expected to be affected by lead in children living
near the ASARCO smelter. These studies showed values within normal
limits for these children. However, to ensure that excessive lead exposure
is not occurring in children in the Tacoma area as a result of previous
emissions of lead from ASARCO or other environmental sources of lead,
additional testing may be done in the future by the state or local health
agencies.
Water/Solid WastePrior to ASARCO's decision to close its copper
smelter WDOE was reviewing ASARCO's NPDES permit (National Pollutant
Discharge Elimination System Permit) to determine what limits should be
included in this permit to control the discharge of arsenic and other
potentially hazardous pollutants into Commencement Bay. Final modifications
to this permit will be made after copper smelting has stopped and the
environmental impacts from remaining activities (e.g. the arsenic plant)
can be assessed. WDOE will require ASARCO to determine which sources
2-46
-------
of pollution have led to the environmental damage in the Bay off-shore of
ASARCO. Pollution resulting from run-off of contaminated water from the
smelting facility and from movement of pollutants through groundwater on
site are both being investigated and may need to be controlled.
Prior to the closure announcement several actions were also being
taken by WOOE and PSAPCA to deal with the environmental problems that may
result from the use and disposal of ASARCO slag. The potential for
emissions of arsenic into air and water at the smelter during the slag
cooling process is no longer an issue because of closure. Since slag
will no longer be produced, concerns regarding its use as sand-blasting
material have also decreased. However, ASARCO as well as several log
sort yards in the area have used ASARCO slag as fill material in the
past. Because mobilization of the metals from slag into the Bay area
from these fill areas is occurring, more extensive studies are being done
and WDOE will be working with ASARCO and the owners of these yards to
develop remedial actions (e.g. diversion of storm-water from the yards)
that can mitigate this mobilization.
Honey BeesPreliminary results from research done by Dr. Jerry
Bromenshenk in 1983 on honey bees in the Puget Sound area show elevated
levels of arsenic and cadmium in bees in the ASARCO smelter area.
Analysis of these data suggests that at least for arsenic, and possibly
cadmium, the source of this bee contamination may be current or past
emissions from the ASARCO smelter. Dr. Bromenshenk's brood survival
results as well as reports from beekeepers in the smelter area suggest
that honey bee survival may be affected by these contaminants, although
these data must be further substantiated. Dr. Bromenshenk collected
additional data in Puget Sound during 1984 using EPA research money. The
results of his study should be available by the summer of 1985.
2.2 RISK ASSESSMENT
2.2.1 Evidence for the Existence of a Threshold For Arsenic
Several commenters criticized the model used by EPA in the development
of the risk assessment. Criticism focused on the use of a linear non-threshold
2-47
-------
model. In addition flaws were pointed out in the epidemiology studies used
by EPA in the development of the unit risk estimate.
Comment:
Several commenters (IV-D-621-14.15, IV-D-621-14.17, IV-D-621-16.12,
IV-3-621-14.7, IV-F-3.3, IV-F-3.6, IV-F-3.9, IV-F-3.11, IV-F-3.15, IV-F-5.11,
IV-F-1.3, IV-F-3.39, IV-F-3.52, IV-D-294, IV-D-611) stated that a threshold
for arsenic existed below which exposure to arsenic did not pose a risk to
human health or that the risk was not substantial.
Comment:
One commenter (IV-F-3.3) said that even though EPA has taken the point of
view that there is no acceptable amount based on the continuation of the
line from the data that's available back down to zero exposure, that there
probably is a threshold.
Comment:
Other commenters (IV-F-1.6, IV-F-3.15, IV-F-5.11) addressed a specific
level at which the threshold exists and cited the study by Higgins as
evidence. One commenter (IV-D-611) cited findings of no excess lung cancer
mortality among smelter workers in a Swedish plant at levels above 200
pg/ni3 as support for the findings of Higgins Anaconda study. Another
commenter (IV-F-1.6) stated that there is no evidence of increased risk to
people who have exposures below 500 ug/m3 based on studies of smelter
workers.
Comment:
Others (IV-D-754, IV-D-708, IV-D-617, IV-D-747, IV-D-427, IV-D-530,
IV-D-580, IV-D-673) commented that there is no threshold for arsenic and
generally supported the no threshold presumption regarding dose-response
relationship for human exposure to arsenic.
2-48
-------
Comment:
A commenter (IV-F-3.11) stated that theoretical predictions suggest that
if arsenic is a carcinogen, then it acts at some epigenetic site and that
it is now widely considered that epigenetic carcinogens probably do have a
threshold. No evidence was cited.
Comment:
Another commenter stated that arsenic is not a genotoxic substance in
in-vitro tests and it has not been shown to be carcinogenic in animals
despite numerous attempts. Therefore, there is no basis in fact for the
application of a linear non-threshold model. Other commenters (IV-D-621-
14.11, IV-D-16) also remarked that arsenic does not act on DNA.
Response: The Non-Threshold Hypothesis
In evaluating the public health hazards associated with exposure to
inorganic arsenic, EPA has maintained that in the absence of sound scientific
evidence to the contrary, such substances must be considered to pose some
finite risk of cancer at any exposure level above zero. Support for the
non-threshold hypothesis for carcinogenic substances is derived from sound
scientific judgment. For the most part substantiation of the non-threshold
hypothesis can be found in policy set forth by the Occupational Health and
Safety Administration (OSHA),1^ the Consumer Product Safety Commission
(CPSC), the Food and Drug Administration (FDA), the Food Safety and
Quality Service, the President's Regulatory Council,^ and the National
Academy of Science.1^
Epidemiological data support a strong association between chronic
exposure to airborne arsenic and lung cancer in humans. In the absence of
clear evidence to the contrary, EPA has assumed that if a carcinogenic
response occurs at dose levels or exposure levels in a study, then responses
at all lower doses will occur at a rate that can be determined by an
appropriate extrapolation model.
Some commenters have challenged this position by asserting that certain
studies have demonstrated no carcinogenic effect below a certain level.
2-49
-------
The threshold argument contends that there exist doses of carcinogens that
are so low that they will not cause cancer when human populations are
exposed.
It remains EPA's belief, however, that not enough is known about the
true mechanisms of initiating carcinoma in human cells and, at present,
such mechanisms can only be postulated. Unlike most clastogenic agents,
arsenic does not appear to directly damage DNA. However, arsenic does seem
to have a genetic effect through some interference with DNA synthesis.
Nordenson et al.l^ and Crossen^0 have observed that arsenic induces chromo-
somal aberrations and sister chromatid exchange (SCE) only when it is
present during DNA replication. In addition, arsenic has been known to be
a sulfhydryl reagent, and as such it can exhibit a number of thiol-dependent
enzyme systerns.21
Therefore another possible mechanism of carcinogenesis for arsenic is
the inhibition of DNA repair enzymes. Another possible mechanism for the
action of arsenic is that it may replace phosphorus within the backbone of
DNA. This may be one reason arsenic is clastogenic. At present there is no
single, well founded explanation describing how arsenic breaks chromosomes
or induces SCE. Given this evidence of interference with DNA synthesis,
especially chromosomal abberrations, SCE, and inhibition of DNA repair
systems, it is not realistic to presume a level of arsenic in the environment
that will have a zero effect on the exposed population. Genetic diversity
and individual differences in the body's capability to defend itself against
the metabolic intrusion of foreign substances greatly discounts the likeli-
hood of a level of exposure of a carcinogen that will not result in an adverse
health effect. The most extensive information on carcinogenesis is with
ionizing radiation, and certain comparisons can be made with respect to some
experimental evidence in animal bio-assays implicating thresholds in some
animal tissues, but for the most part thresholds have not been established
for most tissues.
The National Research Council of the National Academy of Sciences has
noted:
"If an effect can be caused by a single hit, a single molecule,
or a single unit of exposure, then the effect in question cannot
2-50
-------
have a threshold in the dose-response relationship, no matter how
unlikely it is that the single hit or event will produce the
effect (cancer). Mutations in prokaryotic and eukaryotic cells
can be caused by a single cluster of ion pairs which were produced
by a single beam of ionizing radiation. We would expect that
mutations can be caused by a single molecule or perhaps group of
molecules in proximity to the DNA. The necessary conclusion from
this result is that the dose-response relationship for radiation
and chemical mutagenesis cannot have a threshold and must be
linear, at least at low doses."22
Occupational exposure studies have demonstrated a strong association
between chronic exposure to airborne inorganic arsenic and lung cancer.23
Over 10000 smelter workers have been retrospectively studied spanning the
latency period of carcinogenesis. The results are that 11 of the 12
published epidemiological reports of smelter workers in the U.S., Sweden
and Japan have shown a 2-fold to 12-fold increase in lung cancer mortality
above the expected rate. The increase in lung cancer mortality is evident
even when exposure to other pollutants in the workplace was accounted for,
i.e., cigarette smoke, sulfur dioxide.
Commenters have contended that because mutagenesis has not been clearly
established, and carcinogenesis has not been clearly demonstrated in animal
studies despite varying doses and varying animal species, the assertion of
a direct acting mechanism of arsenic is unfounded. These commenters go on
to suggest an epigenetic mechanism, or possible promoting effect of inorganic
arsenic. They offer such evidence as substantiation for a level of exposure
in the community that could be tolerated, and that would not result in
cancer. The evidence, however, of smelter worker studies showing a positive
carcinogenic association to inorganic arsenic transcends the lack of animal
evidence. The Regulatory Council considers properly conducted epidemiologic
studies that show a statistically significant association between human
exposure to a substance and increased risk to cancer as good presumptive
evidence that the substance is carcinogenic.24 Known carcinogens are those
2-51
-------
substances associated with cancer in humans. Because the present state of
scientific awareness on the mechanisms of cancer are largely theoretical,
and are the subject of ongoing research, it is appropriate and prudent that
EPA not accept the argument of the existence of a threshold for human
exposure to inorganic arsenic until sound evidence in support of thresholds
for chemical carcinogens is presented. The NAS has further elaborated:
"The human population in the United States - the population we
are trying to protect - is a large, diverse, and genetically
heterogeneous group exposed to a variety of toxic agents. Genetic
variability to carcinogenesis is well-documented, and it is also
known that individuals who are deficient in immunological competence
(for genetic or environmental reasons) are particularly susceptible to
some forms of cancer. It seems, therefore, that even if we were to
postulate an average threshold for a particular cancer induced by a
particular agent, we would in practice need a series of thresholds for
different individuals. It would be difficult, in practice, to establish
a single threshold.
We (National Academy of Science) conclude from these arguments
that, despite all the complexities of chemical carcinogenesis,
thresholds in the dose-response relationships do not appear to exist
for direct-acting carcinogens. If they do exist, they are unlikely to
be detected and hence, impossible to use. This means that there can
be no totally "safe" exposure to a particular carcinogen, nor can the
term "margin of safety" have any meaning. Any dose of a carcinogen
must be considered to be associated with a risk, even if that risk is
vanishingly small; estimates must be made of that risk."25
2.2.2 The Linear, Non-Theshold Dose/Response Model
Comments were generally critical of the use by EPA's Carcinogen Assess-
ment Group (CAG) of a linear, non-threshold model to derive an arsenic unit
risk factor. These commenters (IV-F-3.12, IV-F-3.15, IV-D-189, IV-D-711,
IV-D-568, IV-D-640, IV-D-625, IV-D-621-7.1, IV-D-621-15.2, IV-F-1.6, IV-D-617,
IV-D-618, OAQPS 79-8, IV-D-27) viewed the model as extremely conservative and
2-52
-------
"deliberately designed to lead to a rough upper limit of risk that could be
considerably lower". Another commenter (IV-D-621-7.1) claimed that a zero
intercept linear absolute model may be an unsatisfactory representation of
the relationship between exposure and disease.
V-,
Comment:
Other commenters approved of EPA's method of deriving unit risk estimates
using the linear non-threshold model. One commenter (IV-D-708) noted that
use of a model which overestimates risk is consistent with public health
policy although the unit risk estimate is likely to be upperbound.
Response:
While EPA agrees that the linear, non-threshold model is conservative
in nature and would tend to provide a plausible upper bound to the risk
range, the Agency does not believe that the assumptions upon which it is
based or that the results of its use are unreasonable. The dose response
model with linearity at low dose was adopted for low dose extrapolation by
EPA because at the time of its introduction,
it had the best, albeit limited, scientific basis of any current mathematical
extrapolation model.26 The EPA described this basis most recently in a Federal
Register notice announcing the availability of Water Quality Criteria
Documents:^
"There is really no scientific basis for any mathematical
extrapolation model which relates carcinogen exposure to cancer risks
at the extremely low levels of concentration that must be delt with in
evaluating the environmental hazards. For practical reasons, such low
levels of risk cannot be measured directly either using animal experi-
ments or epidemiologic studies. We must, therefore, depend on our
current understanding of the mechanisms of carcinogenesis for guidance
as to which risk model to use. At the present time, the dominant view
of the carcinogenic process involves the concept that most agents
which cause cancer also cause irreversible damage to DNA. This position
is reflected by the fact that a very large proportion of agents which
2-53
-------
cause cancer are also mutagenic. There is reason to expect that the
quanta! type of biological response that is characteristic of mutagenesis
is associated with a linear non-threshold dose-response relationship.
Indeed, there is substantial evidence from mutagenesis studies with both
ionizing radiation and with a wide variety of chemicals that this type of
dose-response model is the appropriate one to use. This is particularly
true at the lower end of the dose-response curve; at higher doses, there
can be upward curvature, probably reflecting the effects of multistage
processes on the mutagenic response. The linear non-threshold dose-
response relationship is also consistent with the relatively few
epidemiological studies of cancer responses to specific agents that
contain enough information to make the evaluation possible (e.g.,
radiation-induced leukemia, breast and thyroid cancer, skin cancer
induced by aflatoxin in the diet). There is also some evidence from
animal experiments that is consistent with the linear non-threshold
hypothesis (e.g., liver tumors induced in mice by 2-acetylaminofluorene
in the large scale EDgi study at the National Center for Toxicological
Research, and initiation stage of the two-stage carcinogenesis model in
the rat liver and mouse skin)."
2.3 EPIDEMIOLOGIC STUDIES
2.3.1 Critique of Epidemiologic Studies
Several commenters (about 10) focused on flaws present in the
epidemioloyical studies chosen by EPA for the determination of the unit
risk estimate for lung cancer due to airborne exposure to arsenic. The
comments generally focused on the studies by Lee-Feldstein (1983), Higgins,
(1982), Enterline and Marsh (1982) and Brown and Chu (1983). An overview
of the major criticisms is presented separately for each study.
Critisims of the Lee-Feldstein Study
Comments (IV-D-711, IV-D-640, IV-F-3.15, IV-F-1.6) were received which
questioned the use of data from the 1983 Lee-Feldstein follow-up of Anaconda
smelter workers. One commenter stated that the data show poor fit for any
combination of data or models chosen. The EPA was criticized for incorporating
2-54
-------
only medium and light exposures from this study in order to fit the linear
no threshold model. Comments also claim that Lee-Feldstein did not use an
appropriate exposure classification so that exposure groups overlapped
resulting in the likelihood that someone with an exposure of 1000 ug/m3
could be in the heavy, medium or light exposure group. One commenter (IV-D-708)
stated that the lack of fit of the Lee-Fieldstein data is due to the method
of characterizing exposure rather than any inherent deviation from linearity.
Response:
The Lee-Feldstein Study (1983) has a number of features which support
its use in making quantitative risk estimates of lung cancer from exposure
to airborne arsenic.28 It was a large study involving a relatively large
number of respiratory cancer deaths. Eight thousand forty-seven male smelter
workers were observed for mortality rates from 1938 through 1977 for a
total of 192,476 person years of follow-up observations. Altogether 3550
deaths were observed of which 302 deaths were caused by lung cancer.
Expected number of cancer deaths were calculated on an age-adjusted basis
using the combined mortality of the white male population of Idaho, Wyoming
and Montana. Workers were categorized according to length of employment as
well as the level of exposure to airborne arsenic. These two factors were
correlated with lung cancer mortality. Exposure to arsenic was estimated
from 702 samples collected at 56 sampling locations at the smelter during
the years 1943 - 1958. These exposures were categorized as heavy, medium and
light, and were average levels of airborne arsenic of 11.27, 0.58, and 0.27
mg/m3 respectively. Follow-up was conducted of workers who had been
exposed for 15 years or more. Analysis of the data by EPA shows that the
risk for the high-exposure category with an exposure duration greater than
25 years does not agree with the risks for the other groups.29 Therefore,
EPA decided to use low and medium exposure groups to estimate risk.
Criticisms of Higgins Study
The findings of the Higgins study of Anaconda smelter workers was cited
by some commenters as providing evidence for the existence of a threshold
for lung cancer. One commenter (IV-F-3.15) pointed to the Higgins data to
2-55
-------
criticize EPA's assumption that the same linear relationship of risk to
exposure level is found at all levels and that there may not be levels of
exposure where the risk increases more rapidly than at other exposure
levels. Higgins data demonstrate little or no risk change between the lower
two exposure groups and a doubling of risk between the upper two exposure
groups. Strength is also given to this study due to the proper classifica-
tion of exposure categories as opposed to the methodology used by Lee-Feldstein
(IV-F-3.15, IV-F-1.6). The EPA's fit of Higgins data is questioned although an
adequate fit from both the absolute and relative risk models is demonstrated.
The criticism focuses on the point that analysis by ceiling exposure indicates
heterogeneity of data and because Higgins used an unequal sampling technique,
the heavy group dominates the analysis and thus the unit risk calculated
from these data only applies to high or very high exposure groups (IV-D-711,
IV-D-640).
With regard to the slight deficit in lung cancer mortality for persons
whose "ceiling" arsenic was below 500 ug/m3, another commenter (IV-D-708)
stated that the data with respect to low ceiling doses do not approach
statistical significance. Other criticisms include the fact that Higgins
only used 20 percent of the available cohort, problems with estimations of
exposure and the hypothesis that lung cancer risk is dependent on the
highest 30-day dose rather than cumulative exposure,
Response:
Higgins et al. studied 1800 workers at the Anaconda Smelter.30 The cohort
consisted of workers classified in Lee-Feldstein study as heavily exposed,
and a random sample of 20% of employees classified as having received medium
and light exposures to arsenic. This cohort was 22% of Anaconda workers.
Higgins et al. examined industrial hygiene records during 1943-1965 and
calculated average air concentrations of arsenic for 18 smelter departments.
For 17 other departments with no available measurements, arsenic air levels
were estimated or inferred by analogy to known measurements. Based on
duration of employment within each department, workers were assigned a time
weighted average (TWA) arsenic category, and a ceiling arsenic category.
2-56
-------
TWA values were calculated as a function of length of time a worker spent
in a given department, and the average arsenic concentration in that depart-
ment. Ceiling level was defined as the highest arsenic level a worker was
exposed to for a period of 30 days or more. In addition, workers were
assigned by cumulative arsenic exposure which was calculated as the product
of the average arsenic concentration for each department during 1943 - 1965
times the length of employment in that department; the individual's depart-
ment exposures were summed over his entire work history. Thus cumulative
exposure was an estimation of total dose of arsenic a worker received over
a lifetime. Higgins et al. grouped TWA and ceiling exposure data into four
exposure categories; low (<100.ug/m3), medium (100-499 ug/m3), high (500-4999
ug/m3), and very high (>5000 ug/m3). Cumulative exposure data was categorized
as low, medium, high, and very high with values of 500, 500-2000, 2000-1200,
and greater than 12000 ug/m3 - years, respectively. The study showed that
exposure to airborne inorganic arsenic was strongly related to increased risk
of respiratory cancer mortality. Under the TWA exposure classification
system a gradient response was observed, with SMRs ranging from 138 in the
low category to 704 in the very high exposure category. Observed increases
in lung cancer mortality were statistically significant except in the low
exposure category. Ceiling level exposures showed mortality increases to
be significant only in the high or very high categories, but a dose-response
was observed. SMRs were 129 and 116 in the low and medium categories,
respectively. Increases in lung cancer mortality were observed to be
significant for cumulative exposure groups above 2000 ug/m3 years with
lifetime ceilings above 500 ug/m3.
Commenters take the findings of no significant increase in lung cancer
mortality at ceiling exposure less than 500 ug/m3 as evidence of a threshold
for arsenic exposure. This hypothesis would represent a mechanism of
carcinogenesis suggesting a tolerable dose of arsenic exposure, or a
no-observed-effect-level. The power of Higgins et al. study to detect
increased lung cancer risk in low exposure levels considerably weakens this
hypothesis. The Occupational Safety and Health Administration (OSHA)
recently analyzed the ability of the Higgins et al. study to detect a 1.5 fold
increase in risk of lung cancer mortality to workers exposed to 150 ug/m3
2-57
-------
of arsenic for 15 years.31 The statistical power of Higgins et al. to detect
a 1.5 fold lung cancer risk for ceiling exposure categories of less than
100 ug/m3 and 100-500 ug/m3 exposure level showed a power estimate of only
37%. The study had less than 37% chance of detecting a true 50% excess
cancer risk. OSHA estimated the power of the study to detect increased
lung cancer risk in the TWA exposure category of less than 100 ug/m3 to be
only 31%. OSHA concluded that:
"Most epidemiologic investigators, when initiating a study,
attempt to choose a study cohort of sufficient size to have at least
80% power to detect a true difference in the variable of interest.
Therefore, the statistical power of Higgins et al., all of which are
less than 40%, are much lower than desirable Given the low statis-
tical power of the study by Higgins and colleagues to detect increased
respiratory cancer risk among workers in the low and medium exposure
categories, and given the dose-response gradients observed in their
study, it is appropriate to consider excesses of respiratory cancers
as evidence of potential risk, even if such excesses are not
statistically significant. Hence, the respiratory cancer SMRs of 138,
129, and 116 in the low TWA exposure category, low ceiling category,
and medium ceiling category respectively should not be disregarded."32
Therefore, in view of the low statistical power of the Higgins et
al. study to detect excess lung cancer mortality in low TWA and ceiling
exposure categories, and because the mechanism of thresholds for carcinogenic
agents is currently not supported with good scientific evidence, EPA cannot
accept the argument that Higgins et al. proves the existence of an exposure
to arsenic that will not result in an adverse health effect.
Criticisms of the Enterline and Marsh 1982
Use of the study by Enterline and Marsh was criticized for two basic
reasons. First, commenters (IV-F-1.6, IV-D-625.5, IV-F-3.15) noted that
neither duration of exposure nor time since first exposure contributed
strongly to respiratory cancer excess. The excess also held for workers
2-58
-------
with short exposure and with short latent periods as well. In other words,
exess relative risk of Tacoma smelter workers based on urinary arsenic
levels appeared to be independent of cumulative risk. The EPA was criticized
for correlating urine arsenic levels into air arsenic levels resulting in
an inadequate fit of the data (IV-F-1.6, IV-D-711, IV-D-640). A second
criticism was based on the methodology EPA used in fitting the data. One
commenter (IV-D-711) made the claim that when a "y" intercept was allowed,
the relative risk model had an excellent fit.
Response:
Enterline and Marsh studied a cohort of 2802 men employed at the
ASARCO smelter for a year or more from 1960-1964.33 Their mortality experience
was observed through 1976. During the study period, 104 deaths from lung
cancer were recorded. Respiratory cancer mortality was significantly
increased compared to U.S. males and Washington State males (SMR = 198.1
and 189.4, respectively).
To investigate dose-response, the data were assembled by dividing the
total person years of observation into 5 groups based on cumulative arsenic
exposure (0-lag), and based on cumulative arsenic exposure up to 10 years
prior to the year of observation (10-year lag). Arsenic exposure was
estimated on the basis of representative average urinary arsenic levels for
workers in a given smelter work area. The assumption was there exists a
good correlation between airborne arsenic concentrations and urinary arsenic
levels. Enterline converted urinary levels to estimated airborne levels
using a conversion factor of 0.304. Thus, a urinary level of 100 ug/1 of
arsenic was roughly equivalent to 30.4 ug/nP of arsenic in the air. In
response to the specific comment in EPA's use of'this conversion data in
risk analysis, it must be noted that the derivation of airborne arsenic
concentrations from urinary levels was the protocol of the Enterline and
Marsh cohort study. The Occupational Safety and Health Administration
(OSHA) recently reviewed this protocol in establishing rules governing
workplace exposure to inorganic arsenic and found that,"a urinary arsenic
level is a biological indicator of arsenic exposure that would reflect
protection provided by respiratory use."34 Furthermore, OSHA stated that,
2-59
-------
"Because the men studied by Pinto et al. (Enterline) were
asked not to eat seafood, which would be the major source of
urinary arsenic in the absence of air exposure, Pinto et al's
(Enter!ine's) assumption of zero urinary arsenic from zero air
arsenic exposure appears reasonable. Therefore, OSHA considers
Pinto et al's (Enterline's) correlation coefficient to be the best
available measure of the relationship between urinary arsenic
and airborne arsenic and it has been used by a number of
scientists."35
Cumulative exposure categories, expressed as micrograms of arsenic per
liter years (ug As/1-years) were: <500; 500-1500; 1500-3000; 3000-5000; and
> 7000. SMRs for lung cancer ranged from 155 to 246 in these categories.
There appeared to be no increase in SMRs with increasing dose. For workers
with less than 10 years of exposure, SMRs were highest one to two decades
after the date of hire (suggesting a short latency period). Likewise, for
workers employed 10-19 years, the SMR was highest 20-29 years after the
date of hire. These observations seem to suggest that short exposures have
a disproportionally greater effect than long exposures, and that effects of
early exposure tend to diminish with time.
However, reanalysis of the data by Enterline and Marsh in which
observations were restricted to retired workers over age 65 showed a clearer
dose-response gradient. When lung cancer mortality was analyzed by latency
from initial exposure and duration of employment, SMRs were significantly
in excess during the first 10-19 years after cessation of exposure. When
lung cancer mortality was examined by duration of employment and by average
exposure, SMRs increased both with increasing duration and increasing
average exposure.36 Enterline and Marsh concluded from this that both
duration of exposure and intensity of exposure contributed to respiratory
cancer mortality.37
The EPA considers the Enterline and Marsh study amenable to quantitative
estimation of risk to exposure of airborn arsenic. The study involved the
entire cohort of workers at the ASARCO-Tacoma smelter. Individual exposure
histories were estimated, and the exposure estimates based on a 10 year lag
2-60
-------
probably yield a more realistic dose-response than those that do not utilize
a lag. Analysis of absolute risk by group submitted before OSHA hearings
found that cumulative exposure data and 10 year lag data produced a strong
linear trend of increasing risk with increasing cumulative dose.38 Thus,
the data presented are not inconsistent with the linear non-threshold
model.
2.3.2 Negative Studies
Many comments (IV-F-1.1, IV-F-1.3, IV-F-3.2, IV-F-4.60, OAQPS 79-8,
IV-D-27, IV-F-4.62, IV-F-4.38, IV-F-4.14, IV-F-3.11, IV-F-3.12, IV-F-3.6,
IV-F-1.14, IV-0-773, IV-D-652, IV-F-4.4) were received regarding the absence
of health effects, particularly an increase in mortality due to lung cancer,
within the Tacoma community. One commenter (IV-D-621-5) provided the full
text of epidemiology statistics which demonstrate that there is no actual
support that there is increased lung cancer in communities near smelters.
Several studies were cited which demonstrated no increased risk of lung
cancer in residents residing near copper smelters. These included Polissar
et al. (1979), Hartley et al. (1982), Milham 1982, Frost (1983). Another
commenter (IV-D-710) cited problems with these studies such as small sample
sizes, lack of correction for confounding variables and flawed methodologies
as reasons for the inability to detect an increased risk of lung cancer in
the community.
Comment:
A number of commenters (IV-F-3.11, IV-D-609, IV-D-708, IV-F-4.43,
IV-D-710) questioned the extrapolation from occupational studies to determine
risks in the community. Concerns were based on the uncertainty inherent in
such extrapolations in the development of unit risk and the possibility
that such risks could be higher because of such uncertainties.
Comment:
Criticism focused on the use of occupational studies where exposures
were much higher than ambient levels found in the community (IV-D-695,
2-61
-------
IV-D-627, IV-F-3.11, IV-F-4.43, IV-D-621-15.7, IV-F-3.12, IV-F-3.57, IV-F-3.6,
IV-F-3.15, IV-F-4.60, IV-D-621-15.9) A claim was made that the statistical
data base used by EPA is weak and inadequate for determining carcinogenic
risk from low level arsenic exposure (IV-D-71, IV-D-640).
Response:
It is not unreasonable to estimate risk of respiratory cancer from
chronic airborne arsenic exposure based on observations derived from
statistically valid occupational exposure studies. A causal association
between exposure to a chemical agent and the manifestation of cancer in
humans in the context of prolonged worker exposure to that agent is a valid
and sound epidemiologic method of assuming the agent is carcinogenic in
humans. Once this has been established, as in the case of inorganic arsenic,
then exposure factors, and dose-response gradients documented in occupational
studies become a good basis of estimating risk in the general population.
A 3-fold to 11-fold increase in risk of respiratory cancer has been observed
in over ten epidemiologic studies of smelter workers exposed to airborne
arsenic.39 This strong association relating human exposure to lung cancer
has prompted the International Agency for Research and Cancer, the World
Health Organization Arsenic Working Group, the Chemical Manufacturers
Association, the Occupational Health Safety Administration,40 and the National
Toxicology Program to identify inorganic arsenic as a human carcinogen.41 Four
epidemiologic studies demonstrated a good dose-response relationship and
provided a good basis for risk assessment; they were: Brown and Chu
(1983); Lee-Feldstein (1983); Higgins et al. (1982) and Enterline and Marsh
(1982).
Dose-response curves from these studies were used to estimate unit
risk of exposure to lug/m3 of airborne arsenic. The linear non-threshold
approach in estimating risk to lung cancer was employed by EPA, because,
as the Office of Technology Assessment of the U.S. Congress has pointed
"Such linear models are conservative in that, if they err,
they overestimate the amount of disease to be expected. All govern-
ment agencies that use extrapolation employ linear models for predicting
2-62
-------
cancer incidence. Other models project risks that decrease more rapidly
than dose, and they are advanced as alternatives to the linear model.
The choice of a model is important because, if an acceptable level of
risk were decided on, almost any other model would allow higher exposures
than do linear mode Is. "42
Commenters have raised the issue of the appropriateness of extrapolating
from medium and high exposure levels discerned in occupational studies to
low level community exposure. Despite methodological differences between
smelter studies used by EPA to generate dose-response gradients, the studies
found a dose-response relationship in which increasing exposure to airborne
arsenic was correlated with increasing lung cancer risk. The World Health
Organization recently stated that,
"The use of the linear non-threshold model is recommended for
extrapolation of risks from relatively high dose levels, where cancer
responses can be measured, to relatively low doses, which are of
concern in environmental protection where such risks are too small to
be measured directly either through animal or human epidemiological
studies. The linear non-threshold model has been generally accepted
amongst regulatory bodies in the USA for chemical carcinogens and for
ionizing radiation on an international basis. The linear non-threshold
philosophy was accepted by a Task Group on Air Pollution and Cancer in
Stockholm in 1977. The scientific justification for use of a linear
non-threshold extrapolation model stems from several sources: the
similarity between carcinogenesis and mutagenesis as processes which
both have DMA as target molecules, the strong evidence of the
linearity of dose-response relationships for mutagenesis, the evidence
for the linearity of the DNA binding of chemical carcinogens in the
liver and skin, the evidence for the linearity in the dose-response
relationship in the initiation stage of the mouse 2-stage turmorigenes is
model, and the rough consistency with the linearity of the dose-response
relationships for several epidemiological studies; for example, aflatoxin
2-63
-------
and liver cancer, leukemia and radiation. This rationale for the
linear non-threshold dose-response model is strongest for the genotoxic
carcinogens."43
2.3.3 Criticism of Model-Unit Risk Estimate
Comment:
Commenters (IV-F-3.12, IV-F-3.15, IV-D-711, IV-D-640, IV-D-621-7.1)
questioned the fit of the data in the models claiming that in most cases the
fit is not adequate. Criticism focused on the Lee-Feldstein data which show
no fit when the relative risk model is used and poor fit when the absolute
risk model is used. One commenter (IV-D-621-7.1) criticized the use of
p-value to assess "goodness of fit" stating that this method is not satisfactory
because the value depends on the magnitude of discrepancies between observed
and expected and the size of the study. The same commenter stated that the
lack of fit as assessed by EPA's approach may arise from misclassification
of exposure, incomplete follow-up or misclassification of disease resulting
in errors in the data. Based on this approach, the commenter feels EPA
should exclude the Lee-Feldstein data.
Comment:
One commenter (IV-F-1.6) said that the Lee-Feldstein data should not be
used in the estimation of unit risk because the data do not fit the
linear non-threshold model and only fit it inadequately when the heavy
exposure group is removed from analysis. The Brown-Chu analysis of the same
data should not be used either because it uses out of date data and analyzes
only workers employed past age 55.
Comment:
One commenter agreed with EPA's unit risk estimate based on the linear
absolute analysis of Higgins in preference to the relative risk analysis
but claims tht EPA's value of 4.90 x 10~3 may be an error in calculation and
that the value is actually 2.67 x 10~3. The same commenter claims that the
analysis provided by EPA supports at best only 2 estimates of unit risk:
2-64
-------
1.25 x 10~3 from Brown and Chu and 4.90 x 10~3 from Higgins resulting in a
geometric mean of 2.47 x 1CT3 or 1.83 x 10~3 if the correct value using
Higgins is used.
Comment:
One person (IV-F-1.6) commented that EPA's most recent health assessment
uses the absolute risk, analyzing a method that underaccounts for the age
related incidence of lung cancer.
Comment:
One commenter (IV-F-3.45) voiced concern over the unit risk estimates.
He questioned why absolute risk shows a dose-response relationship while
relative risk does not. He offered the explanation that the groups were
exposed many years ago and therefore are older. Thus workers with highest
levels of exposure are expected to have an increased incidence of lung
cancer merely because of age and this could explain the linear relationship
between absolute risk and exposure.
Comment:
One commenter (IV-D-609) stated that use of the linear model, which is
based on an extrapolation from occupational studies that have their own
uncertainties leads to uncertainties in unit risk. Another commenter
(OAQPS 79-8/IV-D-27) felt that the Agency has failed to be clear and explicit
in its description of both the unit risk estimate and the exposure estimate
methodologies and should have explained that both are designed to overstate
the probable actual value.
Response:
The data from the various epidemiological studies used for the purpose
of deriving a unit risk estimate were statistically analyzed to assess the
appropriate fit of the data with both absolute and relative risk models.
In every case a linear model fitted the data better than the corresponding
quadratic model. In most cases, the fits of the quadratic model could be
rejected at the 0.01 level, with the exception of the two smallest data
2-65
-------
sets (Higgins et al. absolute risk, and Ott et al.). In Higgins et al. the
fit was very marginal (p=0.017). However, for each data set a linear model
provided an adequate fit. In every case, the absolute-risk linear model
fitted the data better than the relative risk model. The p-values for the
its of the absolute risk models ranged from 0.025 to 0.75.
The unit risk is defined as the lifetime cancer risk occurring in a
hypothetical population in which all individuals are exposed to an average
arsenic concentration of 1 ug/m3 throughout a 70 year lifetime. A computed
unit risk for each of the studies was used when the chi-square goodness-of-fit
p-value was greater than 0.01. The unit risks derived from linear models
ranged from 0.0013 to 0.0136. The unit risk derived from the linear absolute-
risk models are considered to be the most reliable, because although derived
from 5 different sets of data from 4 independent investigations of smelter
workers, involving 2 distinct smelter worker cohorts, these estimates were
quite consistent, ranging from 0.0013 to 0.0076. To establish a single
unit risk estimate for arsenic, first a geometric mean of the data sets
within distinct exposed populations was obtained, and then a final estimate
was made based on taking a geometric mean of those values. The final
estimate is 4.29 x ID'3.
Admittedly there are uncertainties in the unit risk process. Estimates
were made from epidemiological studies in which exposures to arsenic occurred
only after employment age was reached. It was assumed in deriving risk
estimates through either the relative or absolute risk models that the
increase in age-specific mortality rates of lung cancer was a function only
of cumulative exposures. The models did not consider how the exposures
accumulated. Thus, even though this assumption results in an adequate
description of the data, it may be in error when applied to exposures that
began early in life. In addition, risk assessment is always constrained by
the fact that it depends on original data as reported and analyzed by the
investigator who's primary objectives were to examine the incidence of
disease and not to determine quantitative risk.
2-66
-------
Comment:
A number of commenters (IV-F-4.4, IV-F-3.11, IV-F-1.6, IV-F-3.57,
IV-F-4.60, IV-F-4.62, IV-F-1.3) addressed the use of epldemiological studies
in risk assessment and their power to detect an increase in cancer incidence
above background levels. One commenter (IV-F-4.4) stated that it is not
possible using the scientific methods available today to detect 1 or 2
additional cancers over the background rate of cancer that exists in every
community with or without a copper smelter.
Comment:
One individual (IV-D-621-14.3) stated that given the fact that migration
hinders epidemiology studies, it is unlikely that it will be possible to
detect risks of 1 or 2 percent.
Comment:
One person (IV-F-4.1) said that we don't have the capacity to detect small
risks from the smelter at this time. The risk would have to be quite large
in order to really detect it in a population the size of Tacoma over a short
period of years.
Comment:
One commenter (IV-F-1.6) stated that epidemiologic study techniques are
too imprecise to measure small increases in death rates from lung cancer.
Because of the large number of people needed to measure a slightly increased
cancer rate, it may not be possible to definitely answer the question of
risk from lower levels of airborne arsenic.
Comment:
Other commenters (IV-D-741, IV-D-621-14.11, IV-D-621-14.8) felt that
additional information is needed about the health effects of arsenic and
the carcinogenic mechanism of arsenic.
2-67
-------
Response:
The EPA agrees with commenters with regard to the difficulty of
detecting an increase cancer incidence within the community. Increased
health risk to residents of the Tacoma area cannot be measured directly.
While epidemiological studies have revealed an association between
occupational exposure to ambient arsenic, such associations may not be
measurable in the general public because of the presence of many confounding
factors. These include the public's greater diversity and mobility, lack
of consolidated medical records, lack of historical exposure data over each
individual's lifetime, public exposure to many carcinogens besides arsenic,
and the long latency period of cancer. Irrefutable proof that arsenic
causes cancer in the community would require at least 95 percent certainty
about the scientific facts. -Since 95 percent certainty is unobtainable for
most conceivable cases of low level exposure to carcinogens due to the size
of the population or length of time necessary to follow a smaller population,
this requirement would preclude the promulgation of environmental standards.
Such an approach would not be in consistent with the language or the spirit
of section 112.
In the evaluation of inorganic arsenic emissions under section 112,
EPA has followed a policy in which the nature and relative magnitude of
health hazards are the primary consideration. Regulatory decisions must
be made on the basis of the best information available since perfect data
can never be obtained. In this case EPA has evaluated the potential detri-
mental effects to human health caused by pollutant exposure based on the
best scientific information currently available. For arsenic this represents
epidemiologic studies of individuals occupationally exposed to levels of
arsenic higher than are present in ambient air.
Comment:
The CEOH report submitted by several commenters (IV-D-634, IV-D-704,
IV-D711, IV-D-64Q) thought that the variable "D" used in EPA's equations
(presented in the draft Health Assessment Document) was more accurately
described as incremental exposure exposure above ambient levels rather than
exposure as measured in an environmental setting. The general population
is exposed to some background level of arsenic. The same commenters said
2-68
-------
that EPA had provided no estimate of uncertainty in the unit risk estimates
nor had EPA characterized its degree of conservatism.
Response:
The commenters have raised a valid point and their reasoning on this matter
reflects understanding of the Agency's exposure and risk assessment. Since
arsenic is a naturally-occuring element in the earth's crust, it is no surprise
that EPA has detected some arsenic at almost all arsenic monitoring sites.
Therefore, each individual probably inhales some arsenic every year over his
entire life. So, strictly speaking, the dose of exposure that is used in the
linear nonthreshold model would be that incremental exposure above the national
average ambient levels. Since the national average is quite small in relation
to the concentrations predicted around many of the sources of concern, this
correction is not meaningful (see Chapter 3 of the Health Assessment Document).
Indirectly, EPA had provided some measure of uncertainty in the unit
risk by displaying the range of values that were calculated for the human
studies with reasonable exposure/risk data. As the Health Assessment Document
indicates, the values ranged from 0.0013 to 0.0136 per microgram per cubic
meter of air. The unit risk estimate of 0.00429 was a single point "best
estimate" for the exposure/risk relationship at occupational levels of
exposure. However, the Agency has no way to quantify the uncertainty of
applying this same relationship at ambient levels. There are no studies
that are sensitive enough to detect the predicted excesses in lung cancer
in the community. Based on experience with other pollutant data, the Agency
believes that the linear, nonthreshold model produces plausible upperbound
estimates of public risk (given that the exposure is accurately known), but
how much of an upperbound estimate is not known.
Comment:
CEOH (IV-D-634, IV-D-704, IV-D-711, IV-D-640) provided an alternative analysis
for deriving a unit risk estimate. Their estimates of unit risk from the
Enterline and Marsh data were 4.49 x 10~3 for zero-lag data and 4.5 x 10~3 for
a ten year lag. These estimates represent a reduction of 34 percent to 40
percent over the unit risk calculated by EPA. The commenters used EPA's
equation, but with an intercept term (b0):
2-69
-------
i = Ei + PYRi (b0 +
where: 0-j = number of lung cancer deaths predicted by the model for the
ith exposure group,
E-j = number of expected deaths based on U.S. white male mortality
rates
= person-years of observation in the ith group, taken from
Table 5-33 of the June 1983 draft Health Document
= constants (the intercept and slope, respectively)
- cumulative exposure to arsenic in Mg/liter-yrs
They claimed an improved fit over EPA's model using this model (X^ = 0.57,
p>0.60). Next, they performed a similar analysis on the data from the
Brown and Chu and the Higgins et. al. data, and in a fashion similar to
EPA's analysis, calculated the geometric mean of the individual unit risk
estimates. In another report, CEOH derived what they termed worst-case risk
estimates by fitting linear absolute and relative risk models with intercepts
to the five data sets used by EPA. The commenter's worst case estimate
derived in this manner was 2.67 x 10~3.
Response:
The commenters desired to account for the possibility that smelter workers
were at a higher than normal lung cancer risk group, and EPA, by not accounting
for this possibility, has overstated the unit risk estimate. If the commenters
supposition was true, then one would detect greater than expected lung
cancer incidence rates in the very low exposure groups of smelter workers.
The Agency considered this possibility since the Lee-Feldstein and the
Enterline and Marsh data appear to support the commenter's hypothesis. However,
the Agency did not modify their analysis as suggested. There was not a
consistent observation of increased lung cancer in the low exposure groups
in all the studies. As a number of other commenters pointed out to the
Agency, the Higgins et. al. data indicated a less than expected lung cancer
2-70
-------
rate for the low exposure group (not statistically significant) . There
was no consistent observation of this increased cancer.risk at low exposure
from study to study. The EPA had other reasons for not modifying its
analysis. As already discussed in the earlier sections of this chapter,
the Agency believes that there are credible scientific theories for adopting
the linear nonthreshold model. Upon reviewing its previous analysis with
the commenters concept in mind (Figures 7-2 thru 7-9 in the health assessment
document), the Agency noted that the absolute linear nonthreshold model
mathematically described the data within the confidence limits of each risk
value for the low exposure groups. Thus, EPA's linear model is adequately
describing the data in this region of exposure.
Finally, EPA believes that the two approaches are producing approximately
the same results. The commenter's estimate falls within the range of unit
risk estimates that the Agency had calculated from study to study (0.0013
to 0.0136 per microgram per cubic meter of air) and so does not significantly
change the Agency's perception of arsenic's carcinogenic potency.
2-71
-------
References
2
3
8
9
10
11
12
I3
16
17
18
USEPA, Health Assessment Document for Inorganic Arsenic, Final Report,
Office of Health and Environmental Assessment, Washington, D.C.,
EPA-6UO/8-83-021F, March, 1984, pp. 7-145 to 7-147.
Ibid, pp 7-1 to 7-77.
Ibid, pp. 7-2.
Ibid, pp. 7-52
Drinking Water and Health. National Academy of Sciences, National Research
Council, Wash., D.C., 1977, pp 11-21.
Health Assessment Document, pp. 7-50 - 7-52.
Health Assessment Document, pp. 5-1 - 5-10.
Ibid. pp. 5-11.
Ibid., pp. 2-28.
Health Assessment Document, pp. 5-21.
Ibid, pp. 5-10 - 5-17.
Norstrom, Op. cit.
Health Assessment Document, pp. 9-9, 9-10.
Health Assessment Document, pp. 7-51 to 7-53
Assessment of Technologies for Determining Cancer Risks from the
Environment, Office of Technology Assessment, Congress of the United
States, Washington, D.C., June, 1981, p. 139.
U.S. Occupational Safety and Health Administration "Identification,
Classification, and Regulation of Potential Occupational Carcinogens"
45 FR 5002, Jan 22, 1980.
Regulatory Council "Statement on Regulation of Chemical Carcinogens;
Policy and Request for Public Comments"
safe Drinking Water Committee, National Research Council "Drinking Water
and Health" National Academy of Sciences, Washington, D.C., 1977.
2-72
-------
19 Nordenson, I., G. Beckman, L. Beckman, and S. Nordstrom. Occupational
and Environmental Risks In and Around a Smelter in Northern Sweden. II.
Chromosomal Aberrations in Workers Exposed to Arsenic. Hereditas 88:
47-50, 1978.
20 Crossen, P. E. Arsenic and SCE in Human Lymphocytes. Mutat. Res.
119; 415-419, 1983.
21 Leonard A., and R.R. Lauwerys. Carcinogenicity, Teratogenicity and
Mutagenicity of Arsenic. Mutat. Res. 75.: 49-62, 1980.
22 Drinking Water and Health, pp. 11-20.
23 Health Assessment Document for Inorganic Arsenic. Final Report.
EPA-600/8-83021F, March 1984, Office of Research and Development,
pp 7-1 to 7-12.
24 Regulatory Council, op. cit.
25 Drinking Water and Health, pp. 11-21.
26 Crump, K., D. Hoel, C. Langly, and R. Peto "Fundamental carcinogenic
processes and their implications for low dose risk assessment"
Cancer Res. 36:9 pp. 2973-2979, 1976.
27 U.S. EPA "Water Quality Criteria Documents; Availability" 45 FR 79319,
November 28, 1980, pp. 79359.
28 Lee-Feldstein, A. Arsenic and Respiratory Cancer in Man: Follow-up of an
Occupational Study, In Arsenic: Industrial, Biomedical and Environmental
Perspectives, W. Lederer and R. Fensterheim, eds., Van Nostrand Reinhold,
New York, 1983.
29 Health Assessment Document for Inorganic Arsenic, EPA-600/8-83-021F,
March, 1984, p. 7-99.
30 Higgins, I., K. Welch, E. Burchfield. Mortality of Anaconda Smelter
Workers in Relation to Arsenic and Other Exposures. Ann Arbor, MI, Dept of
Epidemiology, U. of Michigan, 1982.
31 48 FR 1874, January 14, 1983.
32 Ibid., pp. 1875.
33 Enterline, P.E., and 6. M. Marsh. Mortality Anaconda Smelter Workers
Exposed to Arsenic and Other Substances in a Copper Smelter.
Am. J. Epidemiol 116: 895-910,1982.
34 48 FR 1878.
35 48 FR 1879.
2-73
-------
36 48 FR 1877.
37 48 FR 1877.
38 48 FR 1878.
39 Health Assessment for Inorganic Arsenic. EPA Office of Health and
Environment. 600/8-83-021F. Wash., D.C., March, 1984, Section 7.
* 48 FR 1866.
41 Second Annual Report on Carcinogens, U.S. Dept of Health and Human
Services, Public Health Service, Dec. 1981.
42 Assessment of Technologies for Determining Cancer Risks From the
Environment Summary, Office of Technology Assessment, Congress of the
United States, Wash, D. C., June 1981, pp. 15.
43 48 FR 1887.
2-74
-------
3. LISTING OF ARSENIC
Comment:
Some commenters (IV-D-641, IV-D-622, IV-F-4.67, IV-D-708a, IV-D-741,
IV-0-747) expressed support for EPA's decision to list arsenic as a hazardous
air pollutant under section 112 of the Clean Air Act. However, another
(IV-F-3.il/IV-D-62115.6), questioned EPA's listing of arsenic as a hazardous
air pollutant, saying that this listing was based on determinations that
"there is a high probability that inorganic arsenic is carcinogenic to
humans" but that evidence to support this hypothesis is not unequivocal.
One commenter (IV-D-710) said that to remove arsenic from the list of
hazardous pollutants, EPA would have to show that it "clearly" is not
hazardous under section 112 of the Clean Air Act. The commenter said
that it cannot be shown that arsenic is safe to breathe at ambient levels.
He judged that there is substantial evidence that arsenic, is a carcinogen,
and that as such it must be regarded as posing a. cancer hazard at all
dose levels. Thus, the commenter reasoned arsenic must remain on the
list. Assertions that the risk is "small" or "acceptable" does not
provide a legally supportable basis for removing a substance from the
hazardous pollutant list.
Response:
Under section 122 of the Clean Air Act, EPA was specially directed
to list arsenic as a hazardous air pollutant if the Administrator determined
that emissions "into the ambient air will cause or contribute to air
emissions which may reasonably be anticipated to endanger public health."
Upon review of the available data, the Administrator listed inorganic
arsenic as a hazardous air pollutant under section 112. The Adminstrator's
decision to list was based on EPA findings that "there is a high probability
that inorganic arsenic is carcinogenic to humans and that there is significant
public exposure to inorganic arsenic." Evidence for this is summarized in
the Federal Register (44 FR 37886, June 5, 1980, 48 FR 33113, July 20,
1983) and EPA's Health Assessment Document for Arsenic (EPA-600/8-83-021F).
3-1
-------
The data and documents supporting the listing are filed under Docket
Number OAQPS-79-8 and are available for public inspection and copying at
EPA's Central Docket section in Washington, D.C.
The Administrator stated at the time of proposal, and many commenters
agreed, that there are uncertainties in the health data base and that a
significant public health risk in the general community has not been
absolutely proven. But, neither the language of the Act nor prudent
public health protection policy requires absolute proof of health risks
before the Agency invokes its authority to act under section 112.
When the decision to propose inorganic standards was made, the
Administrator was aware, via an updated draft document entitled "Health
Assessment Document for Inorganic Arsenic" (EPA-600/8-83-021) of issues
and the data subsequently presented by the dissenting commenters and was
considered when the Agency proposed the inorganic arsenic standards. On
balance, however, this draft document presented a strong case for inorganic
arsenic being a human carcinogen. In November, 1983, the Science Advisory
Board, an advisory group of nationally prominent scientists from outside
EPA, concurred with the report's conclusion that the weight of evidence
places inorganic arsenic in a group of pollutants that are characterized
as "carcinogenic to humans." This conclusion is based on two general
observations. First, associations between cancer and inorganic arsenic
exposure have been demonstrated in occupational groups, such as in copper
smelters, pesticide manufacturing and agricultural work, and in non-occupa-
tional populations using arsenical drugs or consuming arsenic-contaminated
drinking water and/or food. Second, the results from several independent
human studies have consistently demonstrated the same study findings,
high relative risks, and specificity of tumor sites (skin and lungs).
The EPA has now published these conclusions in the final health document
(EPA-600/8-83-021F). ,
Others have made similar findings regarding inorganic arsenic's
carcinogenicity. Widely-respected scientific groups such as the National
Cancer Institute and the National Academy of Sciences have concluded there
is substantial evidence that inorganic arsenic is carcinogenic to humans
and the International Agency for Research on Cancer (IARC) has stated there
is sufficient evidence that inorganic arsenic is carcinogenic to humans.
3-2
-------
In addition, the Occupational Safety and Health Administration, also recently
reviewed the substantial body of evidence and concluded that inorganic arsenic
"is clearly a human carcinogen" (45 FR 19584).
After a substance is listed as a hazardous air pollutant, section 11?. of
the Act requires the Administrator to subject the listing decision to public
review during the proposal of the hazardous emission standards for that
pollutant and to continue with the promulgation of standards unless the
Administrator finds, on the basis of information presented by commenters,
"that such pollutant is clearly not a hazardous air pollutant" (section
112 (b)(l)(B)). Thus, in the July 20, 1983 proposal, the Agency specifically
requested comments on the listing decision and the Administrator's findings.
After reviewing all the public comments and considering the available
human health data, the Administrator has affirmed his judgment that
inorganic arsenic is a probable human carcinogen and is appropriately
listed as a hazardous air pollutant under section 112.
Comment:
One commenter (IV-D-625) requested that EPA identify specific inorganic
arsenic compounds as hazardous rather than just grouping them all in the
category of "inorganic arsenic." The commenter felt that a further breakdown
was appropriate since not all inorganic arsenic compounds were of toxicolog-
ical concern and cited the June, 1983 draft health assessment, in which EPA
stated that elemental arsenic was of "little toxicological interest."
According to the commenter, EPA showed evidence that trivalent and penta-
valent oxides have adverse health effects but did not establish whether
other inorganic forms are hazardous. The commenter said solubility was
not considered in the hazard determination. Therefore, the commenter
felt EPA should identify which compounds produce the risks estimated, and
which are expected to produce greater or lesser risks.
Response:
The keystone of the inorganic arsenic listing decision is the relatively
large human health data base that has successfully linked excess lung cancer
and total arsenic exposure arsenic exposure in the workplace. Because of the
3-3
-------
known chemical conposition of the plant products or by-products, the Agency
believes that, although total arsenic was measured for the occupational
studies, the particular arsenic compounds involved primary were 1) inorganic
pentavalent arsenic in the pesticides manufacturing workplace, and 2)
inorganic trivalent arsenic in the smelter workplace. In reviewing this
health data base in EPA's health assessment document (OAQPS 79-8, II-A-13,
EPA-600/8-83-021F), it is apparent to the Agency that exposure to both
forms of inorganic arsenic, i.e., the inorganic trivalent arsenic and the
inorganic pentavalent arsenic are linked with increased risk cancer risks
and the potencies of each form of inorganic arsenic are approximately the
same magnitude. Thus, based on the health effects data, it makes little
sense to separate arsenic compounds by valency or by specific compound.
In addition, identifying and quantifying the various arsenic compounds
present in an unknown matrix is not a routine analytical matter. The Agency
has worked with several analytical researchers, Dr. Edwin Woolson of the
U.S. Department of Agriculture and Dr. Kurt Irgolic of Texas A&M University,
who have much experience in speciating various forms of arsenic in matrices.
The Agency realizes that arsenic speciation techniques are in the develop-
mental stage and not readily adapted to a regulatory program. Thus, the
Administrator has determined that separate regulation of several forms of
inorganic arsenic is unnecessary and impractical.
Comment;
One commenter (IV-D-617) felt that in the application of section 112 to
hazardous air pollutants, more explicit provisions should be made for a
decision not to list a pollutant if EPA is unable to determine that a
significant health risk exists. The commenter endorsed EPA's conditioning
of its decision to list arsenic on an intention to establish standards for
some source categories and not for those deemed to pose insignificant risks.
The commenter noted that exposure should be considered at the time of
listing to determine if a significant section 112 health risk exists and that
EPA should not list (or should delist) a pollutant when public exposure
to that pollutant does not create a significant section 112 health risk.
3-4
-------
Response:
Exposure was considered in the decision to list arsenic as a hazardous
pollutant under the Clean Air Act (44 FR 37886, June 5, 1980). The arsenic
emissions from primary copper smelters and glass plants were determined to
pose significant public exposure. The evidence for significant exposure
at the time of listing is contained in the listing docket [Docket No. OAQPS
79-8 II-A-6], If the Administrator were to determine that there is clearly
no significant risk, section 112 regulations would not be promulgated.
3-5
-------
-------
4.0 EXPOSURE AND RISK DETERMINATION
This chapter is divided into three sections. The first (section 4.1)
contains comments on the exposure and risk determination models for which a
fairly detailed technical description of the model is required in response.
The second part of the chapter (section 4.2) contains the response to
the comments in section 4.1. The response section contains an overview
of the exposure and risk models and explains the assumptions and
uncertainties questioned by the commenters.
The third, section of the chapter (section 4.3) contains additional
comments and responses. Many of the comments in this third section are on
health risk management and policy issues. The responses to these comments
generally do not require the detailed description nf the model (given in
the BIDs). But they may require a description of the chemistry and fate of
arsenic, or of policy under section 112 of the Clean Air Act, or of court
cases which may have a bearing on risk management policy. Because of the
different nature of responses to these comments, they are located in a
separate section (section 4.3).
4.1 COMMENT SUMMARIES
The comments in section 4.1 are divided into 9 subcategories, which
include:
- factors not considered in the exposure/risk estimation,
- degree of conservatism of estimates,
- criticisms of input data and general modeling assumptions,
- reasons for use of the dispersion model versus ambient air
monitoring data,
- criticism of the exposure estimation model,
- criticism of the unit risk estimate,
- miscellaneous criticisms of the model,
- numerical estimates of risk and exposure, and
- uses of the model and risk estimates.
Responses to the comments in this section are given in section 4.2.
4-1
-------
4.1.1 Factors Not Considered in the Exposure/Risk Estimation
Comment:
Several commenters (IV-0-677, IV-0-575, IV-D-571, IV-F-4.11,IV-D-67Q,
IV-F-9) maintained that EPA did not model or consider health risks other
than lung cancer, an omission which results in underestimation of risk.
Other potential health effects cited by the commenters include non-fatal
cancers, general health, and birth defects.
Two commenters (IV-F-4.11, IV-D-710) thought EPA should consider
workplace exposure to arsenic. Since they did not, exposure and risk were
underestimated, according to the commenters.
Some commenters (IV-F-4.11, IV-F-3.57, IV-D-710, IV-D-632, IV-D-757)
felt that the risks to sensitive subpopulations were not considered, causing
the model to underestimate risk. Others (IV-F-3.20, IV-F-3.55, IV-F-4.50,
IV-D-754, IV-F-9, IV-F-10) said EPA should consider sensitive groups in the
rulemaking. Two (IV-F-4.15, IV-D-757) said sensitive groups may include
infants, pregnant women, and people with respiratory problems. Commenter
IV-D-757 pointed out that that studies used as a basis for the unit risk
reflected healthy male worker exposure risks. Commenter IV-H-604
asked if there is a statistical distribution for susceptibility to cancer
which could be incorporated in a risk estimation procedure.
4.1.2 Degree of Conservatism of Estimates
Comment:
Several commenters said EPA's risk estimate was not conservative, since
there are many uncertainties and a variety of factors were not taken into
account. The commenters judged that risk may, in fact, be underestimated.
(These commenters include IV-F-3.42, IV-D-710, IV-D-698, IV-D-608, and
IV-D-579.) One commenter (IV-F-3.42) called the model a "middle-of-the-
road" approach, neither excessively conservative nor reckless. Commenter
TV-D-710 said EPA cannot know if the model over- or under-predicts. Two
others (IV-F-3.55, IV-0-731) thought the exposure estimate in particular
was an underestimate. One commenter (IV-D-708a) said that there is a high
degree of uncertainty in EPA's estimates.
4-2
-------
Comment:
Many commenters said EPA's model was excessively conservative or "worst
case," and that risk has been overstated. (These include IV-F-3.15,
IV-F-3.12, IV-F-1.6, IV-F-3.9, IV-D-198, IV-D-210, IV-D-321, IV-D-330,
IV-D-356, IV-0-362, IV-D-484, IV-D-486, IV-D-4999 IV-F-4.4, IV-D-120, Air
Products, IV-F-1.2, IV-D-617," IV-D-621-7, IV-0-621-16.10, IV-D-621-14.7,
IV-D-621-15.9, IV-D-621-15.7, IV-D-621-15.2, and IV-D-621-5, IV-D-708a,
OAQPS 79-8/IV-D-27, IV-F-9.) Specific criticisms of the model given by
these commenters are listed in sections 4.1.3 through 4.1.7.
Comment:
Some commenters (IV-F-5.6, IV-D-330, OAQPS 79-8, IV-D-27, IV-F-4.4,
IV-D-522, IV-D-708a, IV-F-10) said the model was just guesswork or
speculation without any real scientific basis. Others (IV-D-210, IV-D-342,
IV-D-489, IV-D-504, IV-F-4.17, IV-D-529, IV-D-731) said the risk estimates
were questionable or wrong, but did not give specific criticisms. ASARCO
(IV-F-3.9, IV-D-621-15.2) did not agree with the model.
4.1.3 Criticism of Model - Input Data and General Assumptions
Comment:
Several commenters said the results of the risk determinations are in
error because they are based on poor or inaccurate data. (These commenters
include IV-D-167, IV-D-168, IV-D-215, IV-D-222, IV-D-232, IV-D-254,
IV-D-267, IV-D-276, IV-D-238, IV-D-316, IV-D-330, IV-D-362, IV-D-499,
IV-F-3.45. IV-D-157, IV-D-579, IV-D-621-16.4, IV-D-645, IV-D-538, and
IV-D-568.) Some commented specifically on the inaccuracy of the
epidemiology data and unit risk estimate. Such comments are addressed in
section 2.0.
Several commenters said that, in particular, the results of the exposure
modeling portion of the risk determination were in error because they were
based on inaccurate data. (These commenters included IV-D-165, IV-D-169, IV-D-232
IV-D-330, IV-F-1.7, IV-F-1.8, IV-F-3.9, IV-F-4.38, IV-F-11.) Commenters
(IV-F-3.9, IV-D-591, IV-D-14.16, IV-D-710, IV-D-579, IV-F-5.7, IV-D-741,
IV-D-793) specifically mentioned emissions data. Others thought ambient
4-3
-------
arsenic concentration data rather than a dispersion model should have been
used and some commenters disagreed with the dispersion model results. These
comments are included in the section on "Criticisms of the Exposure Model."
One commenter (OAQPS-79-8/IV-D-27) said EPA failed to consider
available data or check the model against such data.
Comment:
One commenter (IV-D-238) requested that EPA make a new risk
determination based on accurate data.
Comment:
Two commenters (IV-D-330, OAQPS-79-8/IV-D-27) said there is no solid
scientific basis for the kind of mathematical modeling EPA has done to
estimate health risk.
Comment:
Several commenters (IV-D-164, IV-D-167, IV-D-232, IV-n-330, IV-0-499,
IV-D-504) said that use of unjustifiable assumptions and faulty reasoning
were major defects in EPA's model. One commenter (IV-F-5.14) said EPA's
risk determination model treats assumptions as facts.
Comment:
Another commenter (IV-D-600) heard that EPA's statistical model was
based on data from another plant.
4.1.4 Criticism of Model - Use of Dispersion Model Versus Ambient
Monitoring Data
Comment:
Several commenters thought monitored ambient concentrations rather
than concentrations estimated by a dispersion model should be used in the
exposure analysis. They thought this would make results more accurate.
(These commenters include IV-F-3.15, IV-F-4.11, IV-F-3.45, IV-D-125,
IV-D-20, IV-n-67, IV-D-342, IV-F-4.71, IV-D-621-16.10, IV-D-621-16.4,
IV-D-622, IV-D-608, and IV-0-621-10, IV-D-708a, IV-D-703, IV-D-793.)
4-4
-------
One commenter (IV-D-609) said EPA did not attempt to measure actual
levels of arsenic in the local community even though technology and cost
were not prohibitive. Therefore, the dispersion model cannot be validated.
Another commenter (IV-F-9) sugested that ambient arsenic data be collected
by a disinterested party (not ASARCO).
One commenter (IV-F-3.15/IV-D-621-15.9) said that EPA should estimate
risk for persons exposed to average environmental levels of 0.05 to
1.0 ug/nP. The commenters pointed out the fact that EPA has not followed
this approach but has instead estimated the upper limits of risk still
marginally consistent with the data.
4-1.5 Criticism of Model - Exposure Estimation
Comment:
Four commenters (IV-D-698, IV-0-608, IV-D-729, IV-n-749) said using
a 20 km radius was unrealistic and could underestimate exposure. Commenter
IV-D-698 said the 20 km radius might be conservative for fugitive emissions,
but was not adequate for stack emissions. On the other hand, one commenter
(IV-F-9) felt that exposure from fugitive emissions may not decrease with
distance as rapidly as EPA expects.
Two other commenters (IV-F-3.37, IV-F-3.17) said EPA's exposure
estimates should not be confined to the immediate vicinity of the sources,
since stack emissions can travel great distances and affect other communities,
Comment:
Several commenters (IV-D-710, IV-0-617, IV-D-618, IV-D-120) said the
assumption that a person would be exposed continuously over 70 years was
conservative and unrealistic. Another (IV-D-618) said annual individual
risk should be used rather than "maximum lifetime risk" since a person would
probably not be exposed to the same level for 70 years. Another commenter
(IV-D-622) also called fora more realistic appraisal of the length of time
an individual lives in one area.
Some commenters (IV-D-617, IV-D-621-16.10, IV-D-621-14.8, IV-F-1.6,
IV-F-11) said exposure should be measured under existing conditions
rather than estimated. Two commenters (IV-D-621-16.10, IV-D-608)
said that public exposure should be measured with the source
4-5
-------
operating and with it not operating to determine arsenic exposure due to the
smelter.
Comment;
One commenter (IV-D-621-16.10) said that since Enumeration District/
Block Group (ED/BG) data and maps were not given out by EPA, there is no way
to judge the accuracy of the exposure assessments.
One commenter (IV-D-621-16.10) said getting a map of the area, counting
homes, and applying a factor for the average number of people per home would
better estimate population than using national census data.
One commenter (IV-D-621-16.10) said 1980 census data should be used
rather than 1970 data.
Comment:
Two commenters (IV-D-297, IV-F-4.17) said they did not question EPA's
exposure estimates since EPA has better knowledge of that area than they do.
Comment:
One commenter (OAOPS-79-8/IV-D-27) said the Agency had failed to be
clear and explicit in its description of the exposure estimation methodology.
He also said the most probable way for exposure estimates to differ is
downwa rd.
t
4.1.6 Criticism of Model - Unit Risk Estimate
NOTE: Detailed comments on the derivation of the unit risk estimate are
included in section 2.0; however, some general comments on the effects of
the unit risk estimate on model outcome are included here.
Comment:
Several commenters (IV-F-3.15, IV-F-1.2, IV-F-4.4, IV-F-3.42, IV-D-617,
IV-D-621-15.9) felt the unit risk estimate was an extremely conservative or
upper limit estimate. One commenter (I-V-F-3.15) said it was the maximum
estimated possible risk per unit of exposure.
4-6
-------
One commenter (IV-D-710) said EPA's use of the linear no-threshold
model is generally regarded as conservative, but it may not be conservative
given the uncertainties in the unit risk estimation procedure. He noted
the unit risk estimate has increased by a factor of 1.45 since proposal as
new data were considered.
Comment:
Several commenters (IV-F-3.15, IV-F-1.2, IV-F-4.4, IV-F-3.57, IV-D-617,
IV-D-621-15.9, IV-D-604) questioned the use of the linear no-threshold
model, while another commenter (IV-D-622) said the linear model should be
used. '
Comment:
Some commenters (IV-F-4.4, IV-D-627, IV-D-604) questioned the validity
of using data on workers (exposed to high levels of arsenic) to estimate
risk to the general public (exposed to low levels of arsenic).
Comment:
Two commenters (IV-D-297, IV-F-4.17) said they did not question EPA's
exposure estimates, but they did disagree with the way the risk estimates
were derived from the exposure estimates.
4.1.7 Miscellaneous Criticisms of the Model
Comment;
Several commenters (IV-D-164, IV-D-322, IV-D-339, IV-D-362, IV-D-398,
IV-D-427) said that EPA's health risk analysis is based on an unrealistic
computer model.
Comment:
Two commenters (IV-F-4.4, IV-D-708a) said EPA's final results are
an upper bound, and the risk may, in fact, be zero. Another commenter
(IV-D-621-15.9) said based on his judgment and review of the data, zero
is the most likely estimate of increased risk. Another (IV-D-621-15.7)
felt that the range given about the risk estimate should include 0. The
commenter said that statistics from the epidemiology studies may follow a
4-7
-------
normal rather than log-normal distribution. This assumption would give a
range of -7 to +16. Another commenter (IV-F-9) suggested a statistical
analysis of validity of the model and the confidence limits.
One commenter (OAQPS-79-8/IV-D-27) said EPA's explanation of the risk
determination process was unclear.
One commenter (IV-D-609) said that, according to the model, most
cancers will occur a large distance f,rom the smelter due to the magnitude of
the population being exposed.
One commenter (IV-D-525) asked if health impacts consider whether
victims smoke, their age, family history of cancer, or other environmental
circumstances.
Comment:
Two commenters (IV-0-600, IV-D-525) asked if there had actually been
increased cancer deaths in areas near sources. One commenter(IV-D-652)
believed lung cancer increases might be detectable in the Tacoma, Washington
area by community studies. Data from such studies should be used instead
of a risk model. One commenter (IV-D-773) said studies had not shown any
increased risk of lung cancer and that the model's predictions are unlikely.
Comment:
One commenter (IV-F-3.57) objected to current risk assessment practices
which start with the assumption that a substance is harmless and then try to
prove this assumption. He stated it would be more appropriate to start with
the assumption of harmfulness and try to disprove it.
4.1.8 Uses of Model and Risk Estimates
Comment:
One commenter (IV-F-3.12) stated that according to EPA, the purpose of
its risk assessment is to "give a rough estimate of the potential cancer
hazard that can be used to guide regulatory decisions." He felt that this
regulatory purpose would be difficult to achieve, considering that the rough
4-8
-------
estimate of risk was obtained by EPA in a series of steps, each of which was
deliberately and consistently designed to inflate the risk.
Comment:
Other commenters (IV-F-3.4, IV-D-622) said that the risk model
(composed of the health and exposure models) does not necessarily mirror
real life situations; they are used to monitor change in variables such as
emission levels and resulting or ambient air concentrations. The commenters
continued by saying that the risk model is used to predict the effects of
reduced emissions on the potential reductions in risk to public health.
Comment:
One commenter (IV-D-622) said the dispersion model and health risk
model do not necessarily mirror real life situations. He said they are used
to model changes in variables such as emission levels and resulting.changes
in ambient air concentrations, and can be used to predict potential
reductions in public health risk.
Comment:
One commenter (IV-D-617) said health risk is too uncertain to be used
to estimate actual health effects posed by a source. However, he said that
since the unit risk estimate is the same for all sources of arsenic,
exposure estimates can be used to rank the relative severity of the sources,
without assigning actual risk values.
One commenter (IV-D-617) stated that realistic estimates of risk should
be used in evaluating residual risk and that measurement of actual public
risk should be used whenever possible. Two other commenters (IV-D-20,
IV-D-67) agreed that actual health tests of the public should be used in
setting standards.
One commenter (IV-F-3.42) supported EPA's approach to this problem of
making a risk assessment that is as quantitative as possible and considers
both health and economics. Another (IV-D-741) said risk analysis is a
necessary part of rulemaking and is the only available way to set standards
4-9
-------
with a margin of safety. Others (IV-D-708a, IV-D-735) supported the use of
risk assessment as a factor in decision-making, but results should he
used with caution.
Comment:
One commenter (IV-D-621-14.7) noted that, risk assessment is used by all
regulatory agencies, industries, and environmental groups. The commenter
said it is mandated under TSCA and FIFRA. Uses are as follows:
- Target levels of risk are needed to take action.
- Risk estimates can aid in setting agency priorities.
- Risk assessment can help analyze the effects of a proposed action.
- Going through the risk assessment procedure can help establish
what facts are known and unknown.
- The results of a risk assessment cannot be compared directly with
countable cases in the community.
Comment:
One commenter (IV-D-621-14.9) said average risk will vary depending on
the radius from the plant considered and the size of the population used.
Comment:
One commenter (IV-D-621-15.9) said that the health assessment gives a
high estimate of risk, hut does not indicate the likelihood that this risk
actually exists.
Comment;
One commenter (IV-D-641) felt that the range given around the risk
factors is inadequate.
4-10
-------
4.2 RESPONSE TO COMMENTS ON THE EXPOSURE AND RISK ESTIMATION PROCEDURE
Several commenters were critical of the mathematical models EPA used
to estimate human exposure to arsenic and cancer risk in the vicinity of
inorganic arsenic sources. This section deals with the preceding comments
on the exposure model which was used during proposal of the standards.
This model has been previously explained in the Federal Register notice
of proposal (48 FR 33112, July 20, 1983) and in Appendix E of the back-
ground document for the proposed standards EPA-450/3-83-010a. Supplemental
information on the dispersion and human exposure models in can be found
in previous EPA studies of the models [Docket No. (A-80-40), II-A-69,
II-A-42, and II-A-72],
The Agency was aware of the shortcomings of the proposal analysis.
Since the proposal, EPA has completed extensive site-specific air quality
modeling analyses and has compared the predicted concentrations to the
monitored air quality data collected near several sources. These analyses,
the key assumptions and a discussion of the uncertainties are described
in detail in the Appendices of the Background Information Documents as
listed in the Introduction.
4.2.1 Need for a Model to Estimate Exposure and Risk
The Human Exposure Model (HEM) is used to make quantitative estimates
of public exposure, current risk, and risk reductions associated with proposed
or final NESHAP. These quantitative estimates are considered by EPA in
its decision-making process. Although there are underlying uncertainties
in the model, EPA considers this methodology a reasonable approach to the
estimation of health risks and the best tool available to EPA for predicting
the probable effects of a standard.
It is not feasible to measure exposure to ambient arsenic in the
nearby area directly. It would require a large number of monitors to establish
concentrations to which all persons living near urban sources are exposed.
4-11
-------
Exposure will vary with distance and direction from the plant. Furthermore,
there is no way that ambient air quality monitors can predict that future
ambient concentrations may be if arsenic emissions are reduced as a result
of a promulgated standard. However, atmospheric dispersion models can be
used to estimate these directional variations in exposure and to predict
exposure under different emissions control scenarios. Also, existing
monitored data can be used to check or validate the model predictions.
Increased health risk to nearby residents cannot be readily measured
either. Epidemiological studies have revealed an association between
occupational exposure to ambient arsenic and lung cancer (EPA-600/8-83-021f),
but such associations are not readily measurable in the general public
because of the presence of many confounding factors. These include the
public's greater diversity and mobility, lack of consolidated medical
records, lack of historical exposure data over each individual's lifetime,
public exposure to many carcinogens besides arsenic, and the long latency
period of cancer. Because of such factors, increases in cancer observed
in the public can rarely be assigned to a specific chemical or emissions
source.
In addition, the increased risk estimates are a fraction of the
average lung cancer rates and make such predictions difficult to detect
(see chapter 2). Therefore, in the case bf inorganic arsenic, public risk
is estimated by using air dispersion models and site-specific population
data to estimate exposure. Next, the Agency and then applies the exposure/
risk relationship as derived from the occupational studies to estimate
public risks. Although plagued with uncertainty, quantitative estimates
of risk are desirable for decisionmaking and risk assessment methods used
in the arsenic analysis are the best tools currently available to EPA to
obtain such estimates.
4.2.2. Uses of the Human Exposure Model
While it is true that risk estimates obtained from the HEM are
considered in decision-making, they are not used as precise predictions
There are many uncertainties in the model, so the numbers obtained may
4-12
-------
over- or underestimate actual risk, as several commenters correctly pointed
out to the Agency (see Appendix C of Background Information Document, EPA-
450/3-83-OlOb)
The model results can be used in the decision-making process for making
relative comparisons. For example, modeled risk estimates can be used to
compare the relative severity of risks from different sources of arsenic
(i.e., different plants or industries). And they can be used to compare the
relative risk reductions which could be achieved by two or more emissions
control options.
4.2.3 Specific Purposes and Scope of the Model
The HEM estimates public exposure to ambient arsenic under baseline
conditions (i.e., no NESHAP) and under the proposed NESHAP and other
regulatory alternatives. The model also predicts lung cancer risks
associated with these exposures.
Risk is expressed in two ways - the "maximum lifetime risk" and "annual
incidence." The maximum lifetime risk is the lifetime risk of developing
cancer for the individual or individuals estimated to live in the area of
highest ambient arsenic concentrations as determined by the exposure
model. The aggregate risk is the summation of the risks to people living
around a source. It is expressed as incidence of cancer among the total
population after 70 years of exposure. For statistical convenience, the
aggregate risk is often divided by 70 and expressed as cancer incidence
per year.
Both measures of risk are based on lung cancer incidence. Risks from
other potential arsenic-related health affects were not modeled. In the
judgement of the Agency epidemiologic studies show a strong dose-response
correlation between lung cancer and inhalation of of arsenic by smelter
workers; the smelter studies did not show a correlation between arsenic
exposure and other health effects such as skin cancer (EPA-600/8-83-021F).
Thus, increased lung cancer incidence is the likely effect of public inhalation
of arsenic.
Health effects other than cancer which could result from chronic
low-level exposure to arsenic have not been well documented. For exanple,
cardiovascular effects have been noted. However, the data are limited and
4-13
-------
health effects are not consistently demonstrated to the point where a
dose-response relationship can be developed (EPA-600/8-83-021F). For this
reason, health risks other than cancer cannot be quantitatively estimated
or modeled. These other potential risks are considered by EPA in a
qualitative manner during.the decision-making procedure; the Administrator,
upon reviewing the risk assessment results and the associated uncertainties,
recognizes that the lung cancer risk estimates may not represent the entire
spectrum of public health effects associated with inorganic arsenic exposure,
The arsenic lung cancer dose/response relationship was derived from
generally healthy male smelter worker's records. Thus, the distribution of
the public's susceptibility to cancer due to arsenic exposure is unknown,
so risk to sensitive subpopulations or individuals could not be considered
quantitatively in EPA's model. As stated in the background document for
the proposed standards, this is one of the uncertainties in risk assessment.
The EPA in its decision-making is aware of the possible risk to sensitive
individuals and considers this by realizing that the risk estimates are not
true measures of actual risks but may vary considerably from the Agency's
estimates.
It shoul'd also be noted that risks from exposure to arsenic in other
media besides air (e.g., water, food) and risks from chemicals other than
arsenic were not modeled. These issues are discussed in the chapter on the
piecemeal approach (section 5).
4.2.4 Elements of the Exposure/Risk Model
4.2.4.1 Overview. Risk estimates are calculated in a series of steps
and require several types of data. An overview of the procedure is given
here. First, emissions data or emissions estimates and meteorological data
are entered into a dispersion model which calculates the expected long-term
ambient arsenic concentrations at various distances and directions from the
plant. Census data are used to estimate the number and location of people
living near the plant. Then the modeled concentrations are matched to this
population distribution using an exposure model. "Exposure", as determined
4-14
-------
by the model, is the number of people multiplied by the ambient concentra-
tions to which they are exposed. (The units are people - gg/m3).
Once the exposure is calculated, lifetime aggregate risk is estimated by
multiplying the exposure results by a "unit risk estimate". The unit risk
factor is defined as the lifetime cancer risk occurring in a hypothetical
population in which all individuals are exposed throughout their lifetimes
to an average concentration of 1 gg/m3 in the air they breath. The unit
risk is calculated from dose-response curves which are developed from
epidemiology studies.
The resulting lifetime incidence is, therefore, the aggreggate risk
expected in the exposed population over 70 years. Annual incidence is
calculated by dividing the lifetime incidence by 70. The "maximum lifetime
risk", or the lifetime risk of developing cancer for those people exposed
to the highest concentration determined by the model, can also be calculated
by multiplying the highest concentration to which people are exposed times
the unit risk estimate. '- .
4.2.4.2 Details of Each Element of the Model and Associated
Uncertainties. The various elements of the exposure and risk estimation
model, the various assumptions and uncertainties in the modeling procedure
and the modifications which have been made to the model since proposal are
discussed in detail in the referenced BIDs, e.g. Appendix C of the Background
Information Document-EPA450-3/83-010b.
Uncertainties. The method of matching concentrations and populations
within about 3 km of the plant was criticized by commenters because people
within about 3 km were assigned by the model to live at and be exposed to
concentrations at receptors sites located over water or other unlikely
spots. Commenters felt that this was obviously unrealistic. When estimating
risks for arsenic NESHAP promugation documentation, EPA checked the location
of the most exposed individual on small-scale U.S.6.S. maps to insure that
such location was either accurate or realistic. However, when calculating
annual incidence, EPA's experience is that corrections in the exposure
model to more closely account for unrealistic placement of people tend
4-15
-------
to make insignificant changes to the estimate. The corrected locations,
rather than being at points over 'water, will be located somewhere else
where the arsenic concentrations may be higher, lower, or about the same.
When several hundred to several thousand poeple are exposed, the "corrected"
exposures seem to be about the same. It is EPA's judgment that given all
the uncertainties in the exposure modeling corrections are not normally
warranted for calculating annual incidence.
The exposure model also assumes that people are continuously exposed to
the average ambient arsenic concentration at their residence. In reality,
people travel within and beyond the local area. They are exposed to
different concentrations at their workplaces, schools, shopping centers,
etc. It would be extremely difficult to model local travel and exposures,
and any result would be uncertain. Even if the Agency were to collect
detailed information on the public at large near a source, these data
would not necessarily reflect mobility and migration patterns of past or
future generations. Therefore, exposure is modeled using the concentration
at the population centroid nearest their residence, where it is likely
people spend the majority of their time. It is not known if this over-
or underestimates actual exposure.
For the exposure model and unit risk factor, it is also assumed people
stay at the same locatio.n and are exposed to the same concentration for
70 years. Human mobility and variable lifespans make this assumption
unrealistic. However, long-term individual mobility cannot be modeled
for the same reasons as given for modeling individual daily mobility.
Another problem is that sources do not emit at a constant annual level for
70 years. Since many sources have been reducing emissions over the past
decade, the use of current figures may underestimate risk from previous
exposure. Predicted future emission rates under various control scenarios
are also uncertain. If they are too high or low, the lifetime risk may be
over-or underestimated. Similarly, if the population grows in the future,
4-16
-------
aggregate risk would increase. This is not currently modeled. The intent
of the model is to present risk estimates to a hypothetical population
under a "snapshot" of emissions and population distribution scenario.
As a result of the combined effect of all the above assumptions and
uncertainties, the model may or may not underestimate exposure depending on
the actual circumstances. On balance, however, EPA believes that the
methodology represents a reasonable approach given the inherent uncertainties
of exposure modeling.
In the proposal analysis, the unit risk estimate for arsenic was
extrapolated from workplace epidemiology studies using a linear non-threshold
model to estimate risks for the general populations exposed to the arsenic
levels characteristic of the ambient air. This was a weighted average of
values obtained from 3 epidemiologic studies [Docket No. (A-80-4Q)
III-B-1 and (OAQPS-79-8) II-A-7]. A 95% confidence limit around this
estimate produced a range of 7.5 x 1Q-4 to 1.2 x lO"2 for the risk factor.
The proposal risk estimates were given as ranges because of the range
ascribed to the unit risk factor. The range reflected only the uncertainty
around the unit risk factor resulting from the combination of results
from 3 epidemiologic studies to produce one unit risk estimate. The
range did not reflect uncertainties in the use of a linear vs. other type
of dose-response model or uncertainties in the exposure estimates in the
epidemiology studies. It also did not consider uncertainties in the dispersion
and public exposure models. Therefore, the range cannot be used as a
"statistical confidence interval" around the risk estimates predicted by
the modeling procedure as a whole. Actual risk could lie outside of the
range. A range was presented to give the reader an idea of the wide
margin of uncertainty in the risk assessment.
Since proposal, EPA's unit risk factor for arsenic has changed
from 2.95 x 10"3 to 4.29 x 10~3. The new value is based on 5 epidemiologic
studies (EPA-600/8-83-021F). A detailed summary of the derivation of the
current unit risk factor is given in each of the BIDs. Specific procedures used
to calculate unit risk and comments on the unit risk estimate are addressed
4-17
-------
in section 2.0. Uncertainties in the unit risk factor contribute uncertainty
to the aggregate risk and maximum lifetime risk as predicted by the HEM.
The unit risk factor, which defines the relationship between exposure
and lung cancer risks in the linear, non-threshold model was derived from
workplace epidemiological studies. The Agency has assumed that the same
dose/response relationsip calculated at the higher exposures characteristic
of the workplace holds at the lower public exposure levels. There are no
arsenic data available to confirm EPA's assumption. As mentioned in the
Federal Register notice, data on other human carcinogens have indicated
that the linear, non-threshold model provides a plausible, upper-bound
limit on public risk at lower exposure levels if the exposure is accurately
quantified. Thus, as a matter of prudent public health protection policy,
and based on EPA's understanding of the health effects data, the Agency has
selected the linear, non-threshold model to estimate cancer risks.
When using the risk model for decision-making purposes, it is important
to recognize the sources of uncertainty in the final output. Some issues
raised which the model did not consider quantitatively are:
- effects of exposure on sensitive subpopulations,
- effects of exposure to other carcinogens on a person's probability
of contracting cancer when exposed to arsenic,
- workplace exposure and exposures at locations other than the
population centroid of the census area where people currently
live,
- probability of arsenic-related health effects other than lung
cancer.
To the extent possible, the Administrator considers these factors qualitatively
in his decision-making process, along with the estimates made using the
exposure and risk models. He understands that the lung cancer risks may not
present the total health risk picture for the arsenic sources.
4-18
-------
4.2.6 Summary
The EPA has estimated public exposure to ambient arsenic and associated
risks using the Human Exposure Model. While EPA recognizes that there
are uncertainties in the risk estimates, the Agency believes that the
methods used represent a reasonable approach and the results reflect
the best estimates that the Agency can produce within the available resources
Where possible, EPA has confirmed the predicted concentration profile by
obtaining available arsenic data and comparing these data to the predictions.
In several cases, EPA conducted site-specific air dispersion analysis,
considering on-site or local meteorology and terrain features to improve
its risk estimates.
4.3 ADDITIONAL COMMENTS AND RESPONSES ON RISK DETERMINATION AND MANAGEMENT
ISSUES
4.3.1 Estimation of Exposure Through Measurement of Urinary Arsenic
Concentrations
Comment:
One commenter (IV-D-593) suggested that exposure could be measured by
measuring arsenic in the hair and urine of children. Another commenter
(IV-F-3.2/IV-D-621-14.2) has studied urinary arsenic concentrations in the
vicinity of the ASARCO-Tacoma smelter. His results are as follows:
(1) Urinary arsenic levels show a linear decrease with distance from
the stack up to a distance of 2-1/2 to 3 miles. After that point,
levels are what he considers normal.
(2) Part of the exposure appears to be due to inhalation because
urinary arsenic levels in children near the smelter correlated
with wind direction.
4-19
-------
(3) These findings indicate that low-level emissions contribute more
to exposure than stack emissions. Stack emissions would blow over
people near the plant where the highest urinary arsenic concentra-
tion is found. No basic trend in urinary arsenic levels over time
was observed in spite of the fact that stack emissions had been
reduced.
(4) Urinary arsenic levels correlate well with concentrations at
ASARCO's monitoring stations.
(5) Vacuum cleaner samples taken at homes where urinary arsenic was
sampled show a linear decrease to 2-1/2 miles, and then level off.
(6) Relatively few samples were taken due to the expense involved.
ASARCO (IV-D-621-13) included a description of a urinary arsenic study
done by the State government in 1976. The study was designed, to investigate
arsenic exposure to Vashon Island residents. Samples from 22 residents and
110 school children were collected on a day when winds were blowing toward
Vashon Island and exposure levels were estimated to be at a maximum. For
the northern end of the Island, the average urinary arsenic concentration
found was 0.03 ppm, and for the southern end (which is closer to the
shelter) it was 0.02 ppm. About 0.014 ppm was considered normal. Arsenic
concentration did not appear to correlate well with age, and the averages by
age group ranged from 0.018 to 0.036 ppm. The mean arsenic level for people
who had recently eaten seafood was 0.09 ppm. Another sampling was planned,
hut ASARCO felt these results indicated that Vashon Island residents were
not being exposed to detrimental levels of arsenic.
One commenter (IV-F-5.7/IV-D-621-15.9) said it is not clear that total
excess arsenic exposure in local children is due to current smelter
emissions. Even in the absence of current emissions, children may have
4-20
-------
increased arsenic levels. Another commenter (IV-D-630) said children near
the smelter have high urinary arsenic levels indicating high exposure. He
did not provide any specific data.
Response:
Urinary arsenic levels have been shown to increase when arsenic is
inhaled (EPA-600/8-83-021F). Several of the urinary arsenic studies cited
above provide additional evidence for EPA's assertion that the population
is being exposed through inhalation to arsenic emitted by the smelter.
The urinary arsenic studies also indicate that exposure is highest near
the plant. This is in agreement with both modeling and ambient monitoring
results.
However, EPA has not used urinary arsenic concentration as a measure of
public exposure to smelter emissions or lung cancer risks. The primary
reason is that urinary arsenic levels reflect many factors in addition to
the inhalation of arsenic emitted by the smelter. Diet, in particular the
consumption of seafood, can account for increases and decreases in urinary
arsenic concentrations. Also, the particular species of arsenic compound
will affect the way the body reacts to and the rate in which the blood stream
absorbs the contaminant. Individual metabolism and age may also cause varia-
tions in the amount of arsenic excreted. As shown in the study ASARCO
refers to, individuals living in the same area from which urine samples
were taken on the same day showed a range of arsenic levels. Thus,
urinary arsenic levels cannot be used to estimate exposure to air emissions
from ASARCO only, because other sources of exposure can contribute to
arsenic concentrations measured in urine. To get a good "map" of exposure,
one would have to measure urinary arsenic levels in many individuals
living at many different locations at different times of the year under a
variety of wind conditions. Dispersion and exposure modeling is a much
more practical approach.
There are no data to support commenters1 contention that there is no
risk if arsenic levels are measured as "normal." According to modeling
and ASARCO and EPA monitoring data, people beyond 2-1/2 km from the
4-21
-------
smelter are inhaling measurable levels of arsenic. Under EPA's presunption
there is no threshold for arsenic and some risk must be recognized to
exist even at low doses (see section 2.0, on health effects). Modeling
is the only approach available to EPA to quantify these risks due to low
level exposure. These risks are then considered in setting standards.
4.3.2. Miscellaneous Comments on Risk Determination
Comment:
One commenter (IV-D-625) suggested EPA provide separate risk estimates
for each arsenic compound or valence identified as hazardous.
Response:
For reasons described in section 39 EPA has decided to regulate
inorganic arsenic compounds as a class rather than establishing separate
standards for individual compounds. Several problems would arise if EPA
tried to do a separate risk analysis for each compound. First, there are
monitoring, sampling, and analysis problems which would make it difficult to
estimate emissions of each. Second, reactions in the atmosphere may convert
trivalent arsenic to the pentavalent form; some of each form may continue
to exist in the atmosphere. Thus the chemical compounds emitted may not
be the same compounds to which people are exposed. Atmospheric reactions
are uncertain and cannot be adequately modeled at this time. This uncerta-
inty in the form of arsenic received by receptors makes exposure analyses
for individual compounds extremely difficult. Third, several workplace
epidemiologic studies link exposure to both trivalent and pentavalent
inorganic arsenic with lung cancer. Analysis of the data indicate that
the potency of arsenic in either valence state is approximately the same.
These studies provide an adequate basis for risk analysis for inorganic
arsenic as a category without further separation of compounds.
Comment:
Two commenters (IV-F-1.6, IV-D-621-16.10) said that the unit risk
estimate is based on lung cancer mortality data rather than incidence data,
4-22
-------
so "cancer incidence per year" is a misnomer. It should be cancer
"mortality."
Response:
This statement is accurate. The term "cancer incidence per year"
should be read to mean "incidence of fatal lung cancers per year." However,
with lung cancer, there is little difference between incidence and mortality
since about 90 percent of those who contract lung cancer die within 5 years.
Comment:
One commenter (IV-D-621-14.8) questioned whether the annual average,
geometric mean, or maximum 24-hour concentration would be used in the risk
estimation if good ambient data were available.
Response:
Long term (at least annual average) concentrations are the basis for
EPA's risk model. Cancer risks are proportional to long-term exposure to
arsenic (under EPA's modeling assumptions). Therefore, long-term average
ambient air concentrations of inorganic arsenic should be used in the
exposure model.
Comment:
One commenter (OAQPS-79-8/IV-D-27) said EPA failed to consider
available data or check the model against it. He said a similar flaw led
the 5th Circuit Court to set aside the CPSC formaldehyde insulation rule.
Response:
The case cited by the commenter, Gulf South Insulation v. Consumer
Products Safety Commission, 701 F.2d 1137 (5th Cir. 1983) deals with the
Consumer Product Safety Act.. The EPA notes that the legal analysis in
this case cannot be directly applied to actions under Section 112 of the
Clean Air Act, because Section 112 imposes defferent requirements from
the CPS Act. For example, under Section 2058 (f)(3)(A) of the Consumer
Product Safety Act, the Commission is expressly prohibited from promulgating
a safety rule unless it finds that the product that will be subject to
the rule poses an unreasonable risk of injury. Section 2060 (c) provides
4-23
-------
that a consumer product safety rule shall not be affirmed "unless the
Commission's findings ....are supported by substantial evidence on the
record taken as a whole. "The court read this as justification for a
stricter acrutinization of the Commission's actions than the "arbitrary
and capricious" standard would allow. 701 F.2d at 1142.
It is believed EPA's data and risk estimation procedure satisfy the
requirements of Section 112 and show that arsenic emitted by copper
smelters and glass plants may reasonably be anticipated to result in an
increase in lung cancer. Therefore, EPA is proceeding with promulgation
of emission standards.
The commenter cites inadequacy of the data base as a reason the
formaldehyde rule was set aside and as a possible court challenge to EPA's
arsenic NESHAP. The data base CPSC used to perform its formaldehyde
insulation risk analysis was deemed inadequate by the court. This was
largely because emissions and exposure to formaldehyde were determined by
testing of unrepresentative buildings by a variety of test methods, some of
which have not been approved. Secondly, in estimating health risk from
exposure, CPSC extrapolated from the results of one rat study rather than
using human epidemiologic data.
The EPA's arsenic data base is believed to be adequate to support a
risk analysis which can meet the "reasonably anticipated" criteria of
section 112. The EPA has used test data obtained by approved methods
from various sources in generating emissions data. Pollutant dispersion
and public exposure have been calculated using reasonable, widely accepted
dispersion and exposure models. Since the proposal, EPA used ambient
monitoring data collected by the States and the companies and compared
this data to the air dispersion results. Health risk has been estimated
from 5 human epidemiologic studies which link ambient arsenic exposure to
increased lung cancer mortality (EPA-600/8-83-021F). While there are
uncertainties in the data bases and modeling procedure, EPA has sufficient
evidence that arsenic emissions from primary copper smelters and glass
manufacturing plants may reasonably be anticipated to pose a significant
public health risk and should be regulated under section 112 of the Clean
Air Act.
4-24
-------
4.3.3 Uses of Risk Analyses in the Decision-Making Process
Comment;
One commenter (OAQPS-79-8/IV-D-27) cited several court decisions which
he said had a bearing on the use of risk assessment by regulatory agencies.
He cited a Supreme Court decision made in Industrial Union Department of
the AFL-CIO v. American Petroleum Institute, 448 US 607 (1980), saying
that a significant risk at expected exposures must be demonstrated and
that the proposed remedy must provide a significant reduction in that
risk before a rule can be promulgated. In Monsanto v. Kennedy 613 F.2d
947 (D.C.Cir. 1979) he said the courts stated that risks cannot be
inferred or assumed but must be found by a reliable scientific process. In
Marshall Minerals v. FDA 0661 F.2d 409 (5th Cir. 1981) he said the Court
specifically rejected the position that all neoplastic response may be
deemed cancer and said that such substances must be evaluated for risk
under the intended conditions of use.
Response:
In response to the first comment EPA judges that inorganic arsenic
does pose a significant risk, and that these standards significantly
reduce that risk.
The Monsanto Co. v. Kennedy cas§ centers on the definition of a food
additive and the determination if a substance (acrylonitrile) used in
food packaging materials should be considered a food additive under the
Federal Food, Drug, and Cosmetic Act. It appeared the FDA Commissioner
relied on the general principle of diffusion in establishing that acryloni-
trile present in beverage cans in very low quantities would migrate into
food in significant amounts. Tests showing migration had been done on
cans with higher levels of acrylonitrile. But migration into food from
the cans in question was too low to detect. The court ruled that in such
a case the Commissioner had the latitude to consider migration insignificant
and was not mandated to regulate low levels of the substance in packaging
as a food additive. They remanded the decision to the FDA Commissioner
for further consideration 613 F.2d. at 955-956 (1979).
4-25
-------
In this rulemaking, risk has not just been "assumed" as claimed by
the commenter. Evidence is presented in the Health Assessment Document
and in the listing decision published in the Federal Register (EPA-600/8-
83-021f and 44 FR 37886, June 5, 1980). The calculation of risk for the
arsenic sources is presented in Section 4.2. Reliable scientific evidence
has been used where available in making this risk determination.
In Marshall Minerals v. FDA, FDA, supra denied a request for a public
hearing on a petition for food additive regulation. The issues referred to
by the commenter were considered by the court because under-section 348
(c)(3)(A) the FDA is to make an evaluation of whether the food additive
"under the conditions of use" induces cancer when ingested by man or animal.
The commenter implies EPA has no evidence arsenic causes cancer in
humans and has not considered conditions of public exposure. The EPA has
evidence that inhalation of arsenic by smelter workers causes increased lung
cancer incidence and mortality (EPA-600/8-83-021F). Lung cancer, not
just neoplastic responses, is associated with arsenic exposure.
Comment:
One commenter (IV-D-698) said section 112 of the Clean Air Act does not
authorize reliance on risk analysis to identify the level to which emissions
must be controlled. Another commenter (IV-D-590) saw the use of risk
assessment in standards development as inconsistent with the section 112
directive to provide an ample margin of safety. Another commenter
(IV-D-710) said risk estimates were too uncertain to be used to justify
non-application of available controls, and that available technology should
be applied to all sources of emissions of a plant before risk assessment is
used in decision-making.
The Office of Management and Budget (OMB), on the other hand,
(IV-D-618) believes risk assessment should be used by EPA at all stages of
its standards-setting procedure. Currently, OMB said, EPA uses it to
estimate residual risk after BAT and to determine if further control is
necessary. OMB felt that risk assessment should be used in determining
whether BAT need be applied. The commenter stated that the best estimate of
likely effects rather than conservative estimates should be used.
4-26
-------
Response:
The EPA believes that the quantitative estimation of health risk is a
reasonable and necessary part of the decision to regulate sources under
section 112 of the Clean Air Act. For carcinogenic pollutants such as
arsenic for which a health effect threshold has not been conclusively
demonstrated, any level of control short of an absolute ban may pose finite
health risks. The EPA believes Congress did not intend that emissions
standards of zero must be set for such pollutants. Such an intent could
cause wide-scale industrial shut-down and considerable economic disruption.
Thus, standards which permit some level of residual risk must be considered
to provide the "ample margin of safety" to protect public health specified
in section 112 of the Clean Air Act.
Risk assessment models represent the best, and often the only, tool
available to EPA to determine the health impacts associated with various
control alternatives. Therefore, EPA believes that it is appropriate that
risk assessment play a role, along with other criteria, in the section
112 regulatory process. In setting standards for a plant or source
category, EPA will examine current controls and each regulatory option
including application of various technologies, substituting feedstock
materials, and closing the plant. The control efficiency, technical feasi-
bility, cost, and reductions in risk as estimated by the risk assessment
will be among the impacts considered for each option. In choosing the
control option, the Administrator considers whether the estimated risks
remaining after each successively more stringent option are unreasonable.
This is a judgemental evaluation of the estimated maximum lifetime risk and
cancer incidences per year remaining after each control option, the impacts
(including economic impacts) of further reducing those risks, and the
benefits of the substance or activity producing the risk.
In all cases where risks and other parameters are estimated, the
significant uncertainties associated with these numbers will be weighted
carefully in reaching the final decision.
4-27
-------
Comment:
Several commenters expressed opinions on which measure of risk should
be used by EPA in regulatory development. These follow:
OMB (IV-D-618) felt that aggregate (population) risk is a better
measure of public health than individual risk. According to OMB, individual
risks should only be considered if they are unusually high. Commenter
IV-D-673 believed the size of the exposed population and aggregate risk
should be used in determining whether a source should be allowed to continue
operation. Commenters IV-F-1.17 and IV-D-401 also favored using cancer
incidence rate to set priorities, and controlling situations where many
people are exposed before those in which few people are exposed.
One commenter (IV-D-618) said that EPA needs to decide what weighting
to give the estimated risk for the most exposed individual in comparison to
the estimated aggregate population risk. He stressed that a decision to
give more weight to the most exposed individuals would likely result in
a more extensive regulatory intervention without commensurate public
health gains.
One commenter (IV-D-617) pointed out that no consideration is given to
the number of people exposed to the maximum lifetime risk. On the other
hand, one commenter (IV-D-641) felt that people living in low population
density areas should not be subjected to higher individual risks than those
living in high density areas.
Response:
The EPA represents cancer risk in two ways. The first is individual
risk or "maximum lifetime risk", and the second is annual incidence, a measure
of population or aggregate risk. These are described in section 4.2.
Such measures aid EPA in estimating if the emissions standard will protect
both the highly exposed individual and the public at large. Therefore,
EPA considers both expressions of risk in their regulatory development
procedure.
4-28
-------
5.0 PIECEMEAL APPROACH
5.1 EXPOSURE TO CHEMICALS OTHER THAN ARSENIC
Comment:
Several commenters (IV-0-61, IV-D-120, IV-D-142, IV-D-164, IV-D-443,
IV-D-541, IV-D-592, IV-D-731, IV-F-3.37, IV-F-3.51, IV-F-3.55,IV-F-11)
maintained that the environmental and health problems due to smelters
(mostly made in reference to ASARCO-Tacoma) are not confined to current
emissions of arsenic to the atmosphere. They cited environmental problems
in other media and problems caused by other pollutants from historical
emissions, as well as current and future emissions. These commenters
urged EPA to find a way to look at the problem as a whole. They objected
to EPA's piecemeal approach.
Several commenters (IV-D-20, IV-D-61, IV-D-69, IV-D-85, IV-D-87,
IV-D-104, IV-D-106, IV-D-111, IV-D-112, IV-D-115, IV-0-137, IV-D-164,
IV-D-416, IV-D-417, IV-D-427, IV-D-429, IV-D-541, IV-D-551, IV-D-557,
IV-D-597, IV-0-666, IV-D-677, IV-D-698, IV-D-710, IV-D-719, IV-D-731,
IV-F-3.31, IV-F-3.37, IV-F-3.73, IV-F-3.103, IV-F-4.11, IV-F-4.31, IV-F-4.43,
IV-F-9, IV-F-11) said EPA should take into account public exposure to
hazardous chemicals other than arsenic. Many (IV-D-6, IV-D-11, IV-D-13,
IV-D-21, IV-D-35, IV-D-36, IV-D-38, IV-D-39, IV-D-41, IV-D-43, IV-D-61,
IV-D-71, IV-D-76, IV-D-114, IV-D-144, IV-0-164, IV-D-404, IV-D-438,
IV-D-443, IV-n-554, IV-D-558, IV-D-592, IV-D-593, IV-D-644, IV-D-660,
IV-D-666, IV-D-670, IV-D-677, IV-D-705, IV-D-710, IV-F-3.17, IV-F-3.31,
IV-F-3.37, IV-F-3.40, IV-F-3.55, IV-F-4.6, IV-F-4.11, IV-F-4.12, IV-F-4.19,
IV-F-4.50, IV-F-5.1, IV-F-5.9, IV-F-5.13, IV-F-9, IV-F-11) specifically
mentioned that risk from cadmium, S02, lead, copper and antimony should
also be considered. Another (IV-F-4.31) noted that EPA should consider
the possible synergistic effects among various pollutants. Some (IV-D-593,
IV-F-4.43) said that by dividing the problem up into many segments and
considering only arsenic, total risk could be underestimated since small
segments appear less dangerous than the total picture. Another (IV-D-20)
asked that EPA advise the public of the risks of cigarette smoking coupled
with arsenic exposure.
5-1
-------
One commenter (IV-F-4.50) said that while EPA may be able by law to
consider the effects of multiple chemicals separately, those who live with
the problem of multiple exposure cannot. Others (IV-F-3.73, IV-D-754)
said that the same system that requires EPA to investigate the smelter
ties the hands of the investigators by forcing a piecemeal approach. He
continued by stating that the studies are too narrow in scope, taking
into account only part of the pollutant source and only some of the
health effects.
Three commenters (IV-D-70, IV-D-164, IV-F-4.43) said that regulation
would be more efficient if other pollutants were considered. One
(IV-F-4.43) suggested tradeoffs between different pollutants could be
allowed to reduce total risk to a target number. One commenter (IV-D-164)
felt that when the piecemeal approach was used, regulations causing a
significant change do not result. Another commenter (IV-D-70) suggested it
might be simpler and more efficient for some industries to regulate
hazardous air pollutants by category rather than by specific pollutant.
One commenter (IV-D-120) said that section 112 of the Clean Air Act is
limited only to airborne emissions, resulting in a dependence on Superfund
and other statutes for significant reduction in health risk. He continued
by stating that many people have been frustrated about the piecemeal
approach of the statutes to addressing pollution. Other commenters
(IV-D-571, IV-D-783) stated that under the Clean Air Act, EPA has the
authority and responsibility to take into account such factors as "non-air
quality health and environmental impact" when formulating regulations
(section 119 (2)(C)(3)(b)(3)).
Several commenters (IV-D-427, IV-D-592, IV-D-593, IV-D-698, IV-F-3.55,
IV-F-4.31, IV-F-9, IV-F-10, IV-F-11) said that arsenic has accumulated
in the environment over the years, and can reenter the air, water, or
food chain and expose people. They noted that considering current emissions
without regard to past history results in underestimation of exposure and
risk. One commenter (IV-D-571) requested that information concerning the
additional load of hazardous waste be included in the determination of
the level of the standard.
5-2
-------
Some commenters (IV-D-116, IV-D-433, IV-D-515, IV-D-520, IV-D-571,
IV-D-579, IV-D-591, IV-D-592, IV-D-698, IV-D-710, IV-F-3.42, IV-F-4.10,
IV-F-5.7, IV-F-9, IV-F-10, IV-F-11) suggested that EPA consider more
complex routes of exposure from arsenic emissions. In addition to ambient
arsenic levels, the commenter said that reentrainment of dust and secondary
ingestion via food, water, and soil should be considered. Two commenters
(IV-F-3.42, IV-F-4.10) contended that these routes are especially important
for children. However, another commenter (IV-0-621-15.6) said that
arsenic concentrations in soil and water were generally low, and arsenic
in the soil generally forms insoluble complexes with amorphous aluminum
or iron oxide. He noted that this makes arsenic less hazardous, even if
dust is reentrained. Therefore, he concluded that EPA and the public
should not be overly concerned about these exposure pathways.
Response:
The EPA's estimates of the health risks posed by arsenic emissions from
the primary copper smelters do not include those risks attributable to
emissions of other regulated air pollutants or pollutants that are candidates
for regulation. For this reason, EPA's approach has been criticized as a
partial response to the total health hazard and one which perpetuates an
interactive strategy that is less efficient and less desirable to all
parties concerned.
Although this rulemaking addresses inorganic arsenic emissions, and
is being conducted under section 112, the Administrator believes that it
is unreasonable to fail to consider the other pollutants emitted by a source
and other potential environmental impacts. Because new control technologies
and smelter processes that affect arsenic emissions also affect other
pollutants, the Agency believes that consideration of all environmental
concerns is a necessary and important element in the risk management
process. Consequently, EPA considered the impact of the inorganic arsenic
copper smelter standards on emissions of other pollutants and the actions
5-3
-------
being taken under other environmental statutes to address other environmental
impacts of the smelter. Specific actions considered included actions being
taken by various state agencies and EPA to reduce PM and S02 emissions from
smelters and actions being taken to reduce occupational exposures to arsenic.
Although the primary copper smelter standards are directed toward
reducing inorganic arsenic emissions, they regulate particulate matter
emissions. Consequently, emissions of other pollutants (e.g., cadmium,
lead or antimony) which are also present in particulate matter will also
be reduced under the standards. Emissions of gaseous pollutants, such as
S02, are not limited by today's standards; however, there are other
regulations that limit emissions of these pollutants from the primary
copper smelters. The Administrator has considered in the development of
standards the control actions (and the compliance schedule) that are
being taken by the various smelters.
As indicated in a previous section, EPA's assessment of human exposure
to arsenic presently being emitted from primary copper smelters includes
the use of available monitored air concentrations. At the same time, EPA
recognizes that some portion of the arsenic to which nearby residents are
exposed may not be subject to prediction by dispersion models that consider
only current emissions from the source. There is a growing body of
scientific data indicating that historical emissions of arsenic from
smelters may, through accumulation in soil and dust, deposition on or
incorporation into food, and via re-entrainment into the air, contribute
to human exposure.
The EPA has attempted to evaluate and consider such risks explicitly
in the regulatory decision process through the use of monitored data, which
should include the effects of re-entrained arsenic, to estimate health
risks. The smaller the estimated ambient concentrations in comparison to
those measured by the air quality monitors, the greater the concern that
5-4
-------
sources of arsenic other than current, direct air emissions are contributing
to ambient levels. However, as previously mentioned, EPA's predicted
arsenic concentrations are about the same or sometimes even higher than
the measured values collected near the stet at El Paso, air dispersion model
consistently underestimated measured concentrations. Although the available
data do not allow the Agency to accurately quantify the impact of other
exposure pathways, the Administrator has considered this potential in the
regulatory decision.
Other environmental impacts of the smelter are being studied by EPA
and other agencies and efforts are underway to assess the several problems
identified by public comments. The 1980 Comprehensive Environmental
Response Compensation and Liability Act (Superfund) designed for EPA to
take actions needed to protect public health from exposure to hazardous
substances in all environmental media, is being used to investigate other
pollutants, such as cadmium and lead, and to remedy the problems resulting
from multimedia exposure to these pollutants in the vicinity of the ASARCO-
Tacoma smelter (see Chapter 2). Several investigations funded in part or
entirely by the Superfund program are underway or being developed to
study the potential health problems resulting from the historical accumulation
of arsenic, lead, and cadmium. The EPA believes that this work will aid in
the characterization and resolution of the environmental problems associated
with the ASARCO-Tacoma smelter's operations as well as those problems
associated with the other primary copper smelters.
The Administrator also recognizes that even at the control levels
required by these standards that some degree of accumulation of arsenic
and heavy metals in the soil may occur. The EPA believes, however, that the
present levels of these materials in other environmental media are largely
the result of the much higher emissions from the smelter before effective
control equipment was installed. Emissions have decreased significantly
over the past 20 to 30 years. Although the standards will not eliminate
arsenic and heavy metal deposition, EPA believes that the controls will
further reduce emissions significantly and will reduce the rate of
accumulation in the environment.
5-5
-------
-------
6.0 ERA'S STATUTORY OBLIGATION UNDER SECTION 112
This section contains comments and responses concerning EPA's statutory
obligation under section 112. These comments have been divided into four
major subsections:
6.1 Acceptable Risk/Ample Margin of Safety
6.2 BAT Approach
6.3 Economics/Costs As A Decision-Making Criterion
6.4 Recommended Action in the Face of Uncertainty
6.1 ACCEPTABLE RISK/AMPLE MARGIN OF SAFETY
6.1.1 EPA's Responsibility Under the Clean Air Act
Comment:
Some commenters (IV-D-53, IV-D-106, IV-D-107, IV-D-137, IV-D-144,
IV-D-158, IV-0-164, IV-D-411, IV-D-439, IV-D-627, IV-D-718, IV-D-724,
IV-D-731, IV-D-747, IV-F-1.18, IV-F-3.31, IV-F-3.40, IV-F-3.57, IV-F-4.3, .
IV-F-4.6, IV-F-4.11, IV-F-4.15, IV-F-4.24, IV-F-4.50, IV-F-4.55, IV-F-4.59,
IV-F-4.66) pointed out to EPA that the Clean Air Act of 1970 and 1977
requires that the public must be protected with an "ample margin of
safety" from hazardous pollutants. Some said that EPA should implement
the Clean Air Act as passed by Congress (IV-D-66, IV-D-107, IV-D-224,
IV-D-411, IV-D-662, IV-F-1.18). One commenter (IV-D-45) said that it
should not be up to the discretion of either the public or EPA to waive
the law as established by the Clean Air Act to provide an ample margin of
safety for toxic emissions.
Some commenters (IV-D-25, IV-D-39, IV-D-53, IV-D-104, IV-0-106,
IV-0-107, IV-D-111, IV-D-112, IV-D-115, IV-D-137, IV-D-144, IV-D-158,
IV-D-161, IV-D-414, IV-D-420, IV-D-422, IV-D-429, IV-D-437, IV-D-443,
IV-D-731, IV-D-747, IV-D-530, IV-D-541, IV-D-580, IV-D-632, IV-D-662,
IV-D-663, IV-D-666, IV-D-673, IV-D-677.3, IV-D-677.5, IV-D-677.7,
IV-D-677.8, IV-F-1.16, IV-F-1.17, IV-F-3.31, IV-F-3.60, IV-F-3.103,
IV-F-4.6, IV-F-4.43, IV-F-4.66, IV-F-9, IV-F-10) said that EPA's
6-1
-------
proposed arsenic standards are entirely inadequate because the standards
would not provide an "ample margin of safety" from toxic emissions. Others
(IV-D-111, IV-D-115) said that the proposal does not comply with EPA's
obligation to protect the public. Others (IV-D-698, IV-F-1.18, IV-D-732)
said that EPA had failed to provide the protection of public health
required by section 112 of the Clean Air Act. Still another commenter
(IV-D-660) urged EPA to issue standards requiring a reduction in emissions
to levels which permit the public to live not only safely, but pleasantly.
The United Steelworkers of America (IV-D-708) said that the standards
might not assure an adequate margin of safety to the exposed public.
One commenter (IV-F-1.103), who felt the proposed standards fell short
of providing an ample margin of protection, was concerned that in proposing
a standard which would allow such residual risks the present EPA administra-
tion is attempting to establish a precedent for a weaker risk exposure
criterion than has been used as the basis for other environmental protection
standards. Another commenter (IV-F-1.18) said that the current level of
risk that EPA has proposed under the "margin of safety" mandate, makes one
shudder to speculate the level of protection the Agency might provide where
the statutory mandate is not as explicit.
Some commenters (IV-D-137, IV-D-259, IV-D-310, IV-D-545/IV-n-621-16.6/
IV-F-24) said that the standards set by EPA are adequate to protect the
health of the citizens living in the local communities. One commenter
(IV-D-545/IV-D-621-16.6/IV-F-24) qualified his position by stating that if
the proposed standard is not adopted, further delays in the reduction of
arsenic emissions would result. Others (IV-D-567, IV-D-568, IV-D-621-6,
IV-D-621-7, IV-D-628, IV-F-3.12, IV-F-3.18) said that the current emission
standards provide "an ample margin of safety" to protect public health.
Others (IV-D-621-15.1, IV-D-621-15.7, IV-F-3.15) said that in their
judgement current exposure levels provide a vast margin of safety with
respect to pulmonary carcinogenic risk from airborne arsenic exposure in the
environs of the ASARCO smelter.
6-2
-------
One commenter (IV-D-621-15.9/IV-F-3.15) said that the proposed
regulations will most likely reduce the total amount of arsenic emissions
from the smelter but the airborne levels of arsenic may or may not be
reduced proportionally. The commenter continued saying that since lung
cancer risk at current levels of exposure is not expected, there would not
be any reduction in health risk as a result of new regulation, rather only
an increase in the margin of safety would be attained.
Response:
Section 112 does require that EPA set standards that provide an "ample
margin of safety." Where a health effects threshold can be determined, this
requirement can be met by establishing the standard at a level that insures
that the exposure threshold is highly unlikely to be exceeded. Where
identifiable thresholds do not exist or are indeterminate, as with
carcinogens, any level of control selected short of an absolute ban on
emissions, may pose a finite carcinogenic risk.
In establishing the appropriate level of control for carcinogens,
therefore, the Administrator views the objective as a judgement of the
extent to which the estimated risk of cancer must be reduced before the
degree of control can be considered amply protective. Two choices are
available: either the emission standards must be set at zero to eliminate
the risk of cancer altogether, or some residual risk must be permitted.
Neither the language nor the legislative history of section 112 reveals
any specific Congressional intent on how to apply the phrase "provides an
ample margin of safety to protect the public health" to non-threshold
pollutants like inorganic arsenic that present cancer risks at any level
of exposure (48 FR 33116, July 20, 1983).
In the absence of specific direction from section 112, in recognition
of the drastic economic consequences that could follow a requirement to
eliminate all risk from carcinogenic emissions (see Zero Risk section), EPA
believes that it is not the intent of this section to totally eliminate
6-3
-------
all risks. Therefore, EPA believes that the final inorganic arsenic
standards which permits some level of residual risk provides that is not
unreasonable in light of the impacts associated with requiring further control
6.1.2 What is an Acceptable Risk Level
Comment;
Some commenters (IV-D-164, IV-F-4.59) defined acceptable risk as that
which provides an ample margin of safety. One commenter (IV-D-144) agreed
with a newspaper article which stated that "ample" means what any
intelligent Congress would have been aiming at all along; it denotes the
point where we begin to prefer savings over greater safety. Another
commenter (IV-D-25) referred to Webster's dictionary for the meaning of the
word "ample." He said that it means "of large size, extent, capacity,
volume, or scope" and "more than adequate."
One commenter (IV-F-1.18) stated that the determination of an
acceptable risk level is a societal decision and each pollutant must be
looked at separately. Another commenter (IV-D-621-14.7) felt that risk
management, unlike risk assessment, is not a scientific decision; it depends
on politics, economics, technology, and public perception of the extent of
the risk. Still another (IV-F-4.11) felt that determination of ample margin
of safety is a policy issue that involves weighing properly identified risks
and benefits.
Response:
The EPA agrees with the commenters who stated that the determination
of an acceptable risk level depends not only upon health considerations but
upon economics and technology. However, neither the language nor the
legislative history of section 112 reveals any specific Congressional
intent on how to apply the phrase "provides an ample margin of safety to
protect public health" to nonthreshold pollutants like inorganic arsenic
that may present cancer risks at any level of exposure. (See previous
response for further details on EPA's risk management approach.)
6-4
-------
Comment:
Several opinions were expressed concerning what should constitute an
acceptable level of risk. Some commenters (IV-D-618, IV-D-627, IV-F-3.55)
stated that it is very difficult to determine an acceptable level of risk
because it is not easy to prove where the line between danger and safety
lies. However, the Chemical Manufacturers Association (IV-D-617) noted that
in the past EPA has identified levels of increased risk that are so
negligible that they can be deemed fully consistent with the protection of
public health.
One commenter (IV-F-3.3) stated that acceptable risk levels have varied
from 0.10 for OSHA standards to zero. He did not think that anyone knows what
level of health risk is acceptable. Another (IV-F-4.62) felt that one microgram
per cubic meter should provide an ample margin of safety for the community.
One commenter (IV-0-657) believed that the small health risk that currently
exists is acceptable at this point in time, while others (IV-D-630, IV-D-622,
IV-D-722) thought it is not acceptable. One commenter said traces of
arsenic in children's urine is unacceptable. Another said that closure
of the plant may create greater overall public risk than exists presently.
One commenter (IV-F-3.3) felt that while the acceptable level is being
determined, ASARCO-Tacoma should continue to reduce arsenic emissions. In
contrast, the Chemical Manufacturers Association (IV-D-617) felt that the
question of what level of risk is significant must be answered by EPA before
establishing regulations. Similarly, another commenter (IVrD-621-14.7)
stated that it is necessary for EPA to make a judgment about what level of
risk would be unacceptable, in order for EPA to set priorities. Finally,
one commenter (IV-D-621-14.11) felt that effective regulation requires a
determination of what levels of residual risk, if any, should be tolerated.
He asked whether the level of arsenic is low enough now or can be made low
enough that it is judged to be a risk not worth further action.
One commenter (IV-D-609) stated that by not promulgating an ambient air
quality standard, EPA has allowed any measured concentration levels in the
community to be considered acceptable. In contrast, other commenters
6-5
-------
(IV-D-621-16.10, IV-F-1.6) felt that the arsenic proposal appears to contain
an unannounced policy definition of acceptable risk for environmental
exposure.
Some commenters (IV-D-621-16.10, IV-F-1.6) said that EPA's arsenic
proposal appears to be based on a policy decision that "unacceptable risk"
exists when the maximum lifetime risk to some exposed population is greater
than 10-4. Another commenter (IV-D-621-14.9) pointed out that EPA's normal
acceptable risk level is 1 in 100,000. Another commenter (IV-D-718)
stated that EPA has drifted away from the previous acceptable lifetime
risk levels of 1 in 1,000,000 and 1 in 100,000. She states that EPA's
proposed standards will set the level down to 2 in 100, and that this is
not fulfilling the responsibilities of EPA.
One commenter (IV-D-241) said that the controversy over the proposed
standards centers around the question, "What is acceptable risk?" He
continued saying that one cannot determine a risk/benefit ratio until an
agreement is reached concerning how many arsenic-related deaths are
"acceptable." The Clean Air Act as it now stands, he said, sets the risk
factor (for all toxins) at 1 in a million, an unhappy compromise as it is,
since it is obvious that there is no safe threshold for toxic emissions.
However, according to the commenter, the newly proposed arsenic emissions
regulations would condone an estimated national ratio of 9.4 to 150 deaths
per 10,000 lifetime exposures to such emissions. Where is the concern for
the public in such a ratio, and who benefits, he asked?
The New Jersey Department of Environmental Protection (IV-D-641)
suggested that EPA initially concentrate on controlling existing sources to
below a risk of 1,000 in one million. LAER could be required for sources
which exceed that risk level. BAT would be required for existing sources
with risks between 1,000 in a million and 1 in a million. NESHAPS regula-
tion would not be required for sources with risks of less than 1 in a
mi 11i on.
The Natural Resources Defense Council (NRDC) (IV-D-710b) stated that
they could accept an Agency policy of technology-forcing which stops at the
6-6
-------
point where the residual emissions are predicted to create an additional
risk level of one in a million. At that point, NRDC suggested that EPA
should move on to the next unregulated hazardous pollutant. In the future,
NRDC said, it may be possible to further reduce the risk, but if EPA were
working seriously on the backlog of unregulated hazardous pollutants, then
for the present the protection of public health would be better served by
moving on to the next substance.
Response:
Many of these commenters, in effect, are advocating that EPA establish
a target risk level for setting standards under section 112. Under this
approach, a fixed numerical risk or expected cancer incidence rate target
could be used in determining the degree of control required for carcinogens.
Although EPA finds the concept of an established "acceptable" risk level
appealing, it suffers from several drawbacks. First, the Agency perceives
substantial difficulty in determining such levels. This perception was
borne out by the wide range of opinions of what constituted acceptability
in the minds of the commenters. Second, although current quantitative
risk assessment techniques for chemical carcinogens are useful decision-
making tools, considerable uncertainties are associated with the techniques
at their current stage of development. Consequently, the Administrator
believes that in using quantitative risk assessments, he should generally
be free to consider that actual cancer risks may be significantly above
or below those predicted by the estimated procedures, and not be bound by
a fixed target. Third, a fixed target level fails to provide the flexibiity
necessary for an appropriate response. For example, where risks could be
reduced beyond the target without significant costs, that should be
permitted. Likewise, where attainment of the risk-based goal would
eliminate a highly beneficial or necessary activity, the decision-maker
should be able to consider less stringent standards. The EPA agrees with
those commenters who perceived that specific acceptable risk levels are very
6-7
-------
difficult to set and are not reasonable as a basis for regulation. After
reflecting on the various points presented, the Administrator supports
the concept of reducing public risks to the extent possible considering
the uncertainty, technical feasibility, environmental, economic, energy,
and other impacts on society and industry.
6.1.3 Zero Risk/Zero Exposure
Comment:
Many commenters (IV-D-8, IV-D-9, IV-D-33, IV-D-69, IV-D-71, IV-D-73,
IV-D-88, IV-D-102, IV-D-105, IV-D-116, IV-D-144, IV-D-152, IV-D-161,
IV-D-301, IV-D-302, IV-D-329, IV-D-401, IV-0-420, IV-D-424, IV-D-433,
IV-D-440, IV-D-575, IV-D-583, IV-D-590, IV-D-596, IV-D-598, IV-D-610,
IV-D-644, IV-D-661, IV-D-664, IV-D-676.1, IV-D-676.3, IV-D-676.4, IV-D-721,
IV-D-725, IV-D-727, IV-D-730, IV-D-783, IV-D-709, IV-D-734, IV-D-744,
IV-D-752, IV-D-753, IV-D-768, IV-D-778, IV-D-781, IV-D-784, IV-F-3.7,
IV-F-3.29, IV-F-3.37, IV-F-3.51, IV-F-3.65, IV-F-3.66, IV-F-3.70, IV-F-3.74,
IV-F-4.3, IV-F-4.28, IV-F-4.59, IV-F-5.13, IV-F-5.15, IV-F-5.22, (IV-D-1,
IV-D-57, IV-D-62, IV-D-67, IV-D-72, IV-D-75, IV-D-98, IV-0-144, IV-D-163,
IV-D-301, IV-D-400, IV-D-427, IV-D-524, IV-D-556, IV-D-557, IV-D-582,
IV-D-598, IV-D-660, IV-D-677.1, IV-D-686, IV-D-689, IV-D-710, IV-F-3.38,
IV-F-3.40, IV-F-3.51, IV-F-4.4, IV-F-3.103, IV-F-9, IV-F-10, IV-F-11)
(IV-D-109, IV-D-161, IV-D-292, IV-D-329, IV-D-424, IV-D-587, IV-F-1.17,
IV-F-3.65, IV-F-3.66, IV-F-3.70, IV-F-4.28, IV-F-5.13) IV-D-1, IV-D-57,
IV-D-62, IV-D-67, IV-D-72, IV-D-75, IV-D-98, thought the emission standard
should be set at a zero level for arsenic and other pollutants. The
commenters reasoned that a zero level would protect the health and welfare
of the community. Some commenters (IV-D-710, IV-F-4.66) felt that, using
currently available information, it is not feasible for EPA to determine
that any exposure to arsenic greater than zero will provide an ample
margin of safety. Another commenter (IV-F-4.28) felt that if EPA permits
6-8
-------
any emissions of a known carcinogen, a decision has been made by the
Agency to use human subjects in research.
The Natural Resources Defense Council (IV-D-710) elaborated upon this
thought by stating that the "ample margin of safety" requirement signifies a
firmly held goal that no one should lose his or her life or health on
account of toxic air pollution. NRDC continued by saying that the absence
of identifiable thresholds does not permit the Agency to deem some rates of
death or serious illness "insignificant." If one person living near a plant
contracts cancer and dies, there has been a health effect of the most
serious and final nature in NRDC's view. The fact that only one person died
does not make the effect insignificant according to NRDC. The commenter
representing NRDC said expressing the effect in terms of an individual risk
only means that the death is acceptable because the cancer strikes randomly
from a pool of people and the victim may not even be identified.
The Attorney General's Office of the State of New York (IV-D-698)
stated that the courts have recognized that EPA may legitimately require
zero emissions of hazardous air pollutants in order to meet section 112's
mandate. They cited the case of United States v. Borden, Inc., 572 F.
Supp. 684, 688 (D. Mass. 1983).
Another commenter (IV-D-25) said that EPA must establish standards for
completely safe operations, free from any possibility of causing illness
and/or impaired health through arsenic emissions to comply with the ample
margin of safety requirement.
Some commenters (IV-D-621-15.2, IV-F-3.9) stated that there is a built
in conflict suggested by an overall EPA policy based upon a zero-risk (no-
threshold) level for hazardous pollutants, and the use of a risk assessment
and risk management approach to determine what levels of risk are accept-
able. Another commenter (IV-D-621-14.11) pointed out that since EPA uses a
linear model, there will always be some residual risk unless all sources are
cut to zero (i.e., the plant is closed and the contaminated soil is removed).
One commenter (IV-F-3.3) interpreted the use of the linear no-threshold
model to mean that EPA thinks there is no acceptable exposure level for
arsenic.
6-9
-------
Some commenters (IV-F-4.6, IV-F-3.103, IV-F-9) stated that it is not
now possible to completely eliminate arsenic emissions from certain
industrial processes, but EPA should have zero or near zero emission
levels as an ultimate goal.
Some commenters (IV-D-621-16.9, IV-F-3.18, IV-F-4.15) stated that zero
risk from arsenic can never be achieved. One commenter (IV-F-4.15) noted
that even if the ASARCO plant closed, some residual risk would still remain
from the build up of arsenic in the soil.
Some commenters (IV-D-146, IV-D-545, IV-D-621-16.6, IV-F-4.24) said
that it is neither possible nor desirable to eliminate all risks and that
life is filled with risks. Therefore, they did not see a necessity for a
standard based on zero risk. In support of this view, one commenter
(IV-D-621-16.2/IV-F-3.18) noted that even if manmade pollution is totally
controlled, pollution caused by natural disasters (fires, storms, earth-
quakes, volcanic eruptions, etc.) cannot be controlled.
Several commenters (IV-D-125, IV-D-180, IV-D-621-15.2, IV-D-724,
IV-F-4.2, IV-F-3.78) thought that a zero emission level is not possible
or needed. Economic infeasibility was cited by some commenters (IV-D-125,
IV-F-3.78) as a reason. One commenter (IV-D-154) did not believe one
life was worth unlimited cost. Some commenters (IV-F-4.2, IV-F-3.78)
stated that as a standard approaches the zero risk level, compliance
costs increase rapidly, while the benefits are hard to quantify. Another
(IV-D-125) went on to say that disasterous economic consequences would
result if zero risk was required. He called the concept economic suicide
and alarmism. Another (IV-D-724) said a zero risk level for arsenic and
other carcinogens would prove socially catastrophic given the pervasiveness
of at least minimal levels of carcinogenic emissions from American
industries. In support of this view, some commenters (IV-D-617, IV-D-622,
IV-F-3.4) felt that it was not the intent of EPA or Congress to interpret
section 112 as requiring a zero emission level for arsenic. They concluded
that such a requirement would shut down major segments of American industry.
[See section entitled "Economics as a Decision Making Criterion Under
6-10
-------
Section 112."] The Chemical Manufacturers Association (CMA) (IV-D-617)
continued by saying that the agency's rejection of a zero-risk interpretation
of section 112 is amply supported by legislative materials and a variety
of administrative and judicial decisions in the health and safety area.
Citing the legislative history of the 1977 Clean Air Act Amendments, CMA
stated that Congress specifically rejected the suggestion that an ample
margin of safety for no-threshold pollutants requires zero emission
standards. CMA cited 1977 Legislative History at 1030-31, 2577-79, 2594,
House Report, 1978; Comm. Print No. 16, Senate Comm. on Environment and
Public Works.
Response:
The EPA and other public health agencies and groups have, as a matter of
prudent health policy, taken the position that in the absence of identifi-
able effect thresholds, carcinogens may pose some risk of cancer at any
exposure level above zero. In establishing margins of safety for carcinogens,
therefore, the task is to determine how low the risk of the occurrence of
cancer in an exposed persons or the projected incidence in an exposed
population must be driven before a margin of safety can be considered
ample to protect the public health. Only two approaches are available
for performing this task: either the emission standards must be set at
zero to eliminate the risk of cancer incidence altogether, or some residual
risk must be permitted. The Administrator does not believe that section
112 expresses an intent to eliminate totally all risks from emissions of
airborne carcinogens. Section 112 standards which permit some residual
risk can, in the Administrator's judgment, therefore, provide an ample
margin of safety to protect the public health.
This view is based on several additional factors. Foremost among these
is the belief that if Congress had intended the drastic results that would
flow from a requirement to eliminate all risk from emissions of carcinogens,
it would have spoken with much greater clarity.
6-11
-------
A requirement that the risk from atmospheric carcinogenic emissions he
reduced to zero would produce massive social dislocations, given the
pervasiveness of at least minimal levels of carcinogenic emissions in key
American industries. Since few such industries could soon operate in
compliance with zero-emission standards, closure would be the only legal
alternative. Among the important activities affected would be the genera-
tion of electricity from either coal-burning or nuclear energy; the
manufacturing of steel; the mining, smelting, or refining of virtually any
mineral (e.g., copper, iron, lead, zinc, and limestone); the manufacture of
synthetic organic chemicals; and the refining, storage, or dispensing of any
petroleum product. That Congress had no intention of mandating such results
seems self-evident.
The conclusion that Congress did not contemplate closure of the
nation's basic industries, or even widespread industry closures, is also
supported by the history and language of section 112. First, Congress in
1970 gave the subject of plant closures only brief consideration in connec-
tion with section 112. While the legislative history makes clear that the
Administrator is empowered to set standards under section 112 that result in
plant or industry closures where appropriate, it is by no means clear that
Congress intended that result for all non-threshold hazardous pollutants, or
even that Congress really focused on the problem. Indeed, the very limited
nature of the legislative history itself compels the conclusion that closure
of the nation's basic industries, irrespective of the actual levels of risk
involved, could not have been contemplated. That conclusion becomes even
more inescapable in light of the 1977 Amendments, which added radioactive
substances - long regarded as confirmed carcinogens and emitted from a wide
variety of sources - to the coverage of the Act, with no mention anywhere of
industry closures as the inevitable consequence.
The language of section 112 is also consistent with this arsenic
standard. In using the phrase "margin of safety," Congress was borrowing a
concept from the field of engineering, where it had previously employed the
term. By prescribing the use of a margin of safety for the load factors of
6-12
.
-------
underground mine hoist cables in the 1969 Mine Safety Act, for example,
Congress surely did not intend to suggest that the safety factor must
guarantee a failure risk of zero. Indeed, no reputable engineer would say
that even with a margin of safety an "adequately strong" hoist cable
presents a failure risk of absolutely zero.
Nor does the use of the term "safety" necessarily imply a zero-risk
concept. Where Congress has intended to require safety from the risk of
cancer to be absolute, it has known how to express that intention clearly,
as it did in the Delaney Clause of the Food and Drug Act, prohibiting the
use of any food additive found to induce cancer in man or animal at any
level of exposure. This provision was enacted years before section 112, and
the absence of comparable specificity in section 112 suggests that "an ample
margin of safety to protect the public health" need not be interpreted as
requiring the complete elimination of all risks.
In interpreting the margin of safety concept in section 112 of the
Clean Air Act, moreover, there is no reason to believe that Congress
intended to make air pollution practically the sole facet of American life
from which the government would attempt to eliminate risk entirely.
Not only is there no indication, as noted above, that Congress
considered the inevitable consequences of such a decision, but such an
interpretation would also be quite incongruous in view of the provisions of
numerous other public health statutes enacted during or since 1970. These
statutes deal with, among other things, environmental carcinogens to which
people are equally or more exposed, and they all permit consideration of
factors other than risk in setting standards or taking comparable actions.
In particular, the recent enactment of the Toxic Substances Control
Act, which was intended to address the problem of toxic substances compre-
hensively, supports the view that where Congress has specifically considered
the problem of reducing risks posed by environmental exposure to carcino-
gens, it has not required complete elimination of those risks. Lastly, as
several commenters pointed out, closing down the copper smelter and other
inorganic arsenic sources will not completely eliminate the risks associated
6-13
-------
with exposure. Since arsenic is a naturally occurring element in the
earth's crust, airborne arsenic has been detected in the air almost
everywhere the Agency has sampled for it. However, the measured levels
of arsenic in other areas are generally several orders of magnitude below
the levels measured in Tacoma. Thus, the Agency suspects that, even with
plant shutdown and soil cleanup, Tacoma residents will always be exposed
to some inorganic arsenic.
Taken together, the Administrator believes that these statutes
provide strong evidence that the complete elimination of risk from environ-
mental exposure to carcinogens is 1) a virtually impossible assignment, and
2) not the task with which he has been charged by Congress.
6.1.4 Comparative Risk
Comment:
Commenters sought a framework for analysis of risk. Many suggested
that comparisons of risk levels to those associated with other societal and
environmental factors might be appropriate. Both voluntary and involuntary
risks were used as a basis of comparison. One commenter (IV-D-668) stated
that the comparison of voluntary vs. involuntary risks is an unfair one.
One commenter (IV-0-721) said comparing cancer risk caused by the smelter
to background cancer incidence is inappropriate. He likened it to justifying
a murder by comparing it with the background incidence of accidental death.
Some commenters (IV-F-3.57, IV-F-4.68, IV-F-4.71, IV-F-5.18) stated that it
was unfair for the residents of the Tacoma area to be subjected to the same
risks as smelter employees, because the employees accepted the risks
associated with arsenic exposure when they decided to work for ASARCO. One
commenter (IV-D-164/IV-D-666) said that as an adult, she might be willing to
tolerate the kind of risk associated with EPA's current proposal for arsenic
emissions at ASARCO, but children should not have to accept such risks.
Another commenter (IV-F-3.6) stated that while individuals need to take
responsibility for personal health practices, such as smoking, society must
take responsibility for public health measures which decrease involuntary
exposure to know harmful substances.
6-14
-------
The Office of Management and Budget (IV-D-618) compared EPA's estimate
of annual cancer risks from various source categories to death risk from
accidents, homicide, and natural background radiation. They concluded
that further risk reductions are warranted when the annual risk to
the most exposed individual is greater than other risks routinely encountered
in daily life.
Similarly, the Pacific Gas and Electric Company (IV-D-625) argued that
significance is a function of relative risk because, as other risks are
reduced, previously insignificant risks become significant. They continued
by saying that a risk that is relatively high (significant) in one area
might be totally insignificant in another area, since current total every
day health risks vary considerably in different areas due to differences in:
traffic hazards, crime rates, earthquakes, floods, storms, landslides,
subsidence hazards, occupational hazards, lifestyle choices, economic well
being, and environmental pollution hazards. They point out that although
EPA is only authorized to regulate environmental pollution hazards, require-
ments to reduce environmental hazards could preempt public and private
resources that might otherwise have been used to reduce far greater hazards
of another nature. Therefore, all sources or risks should be considered
when determining significance. On the other hand, one commenter (IV-D-9)
felt that other risks had no bearing on the arsenic decision and did not
need to be considered.
In general, the selection of comparative risks was dependent upon
whether the commenter felt that the risk associated with proposed standard
was acceptable or unacceptable.
6.1.4.1 Risk Associated With Proposed Standard is Acceptable. The
following comparisons were made to illustrate that the risks associated with
the arsenic emissions from ASARCO-Tacoma are much less than those associated
with other voluntary and involuntary risks.
6-15
-------
Voluntary Risks
- risk associated with cigarette smoking and the use of tobacco
products (IV-D-15, IV-D-120, IV-D-130, IV-D-141, IV-D-145,
IV-D-187, IV-D-227, IV-D-246, IV-D-265, IV-D-267, IV-D-313,
IV-D-322, IV-D-355, IV-D-359, IV-D-361, IV-D-364, IV-D-382,
IV-D-453, IV-D-548, IV-D-607, IV-D-613, IV-D-621-12.1,
IV-D-621-12.5, IV-D-621-12.10, IV-D-621-12.11, IV-D-621-
12.13, IV-D-621-12.22, IV-D-623, IV-D-645, IV-D-647,
IV-D-657, IV-F-3.18, IV-F-4.25, IV-F-4.30, IV-F-4.32,
IV-F-4.54, IV-F-4.60).
- risk associated with drinking alcohol (IV-D-128, IV-D-245,
IV-D-452, IV-D-453, IV-D-621-12.5, IV-D-645, IV-D-647,
IV-F-4.8, IV-F-4.25, IV-F-4.30).
- risk associated with drug abuse (IV-D-246, IV-D-645,
IV-F-4.8, IV-F-4.25, IV-F-4.49).
- risk associated with driving a car (IV-D-146, IV-D-472,
IV-D-613, IV-D-760, IV-F-4.53) and with auto traffic
(IV-F-4.49).
- risk associated with the use of lawn fertilizers (IV-D-613).
- risk due to radiation associated with living in a brick or
stone building (IV-D-128).
- risk associated with different life styles (IV-D-377,
IV-D-647).
- risk associated with eating processed food (IV-D-132).
6-16
-------
Involuntary Risks
- health risk associated with mobile source pollution (IV-D-31,
IV-D-230, IV-D-247, IV-D-254, IV-D-265, IV-D-313, IV-D-322,
IV-D-359, IV-D-382, IV-D-453, IV-D-472, IV-D-621-12.5,
IV-D-621-12.12, IV-D-621-12.13, IV-D-621-12.24, IV-D-623,
IV-D-636, IV-D-657, IV-F-4.30).
- risk of accidents caused by drunk drivers (IV-D-621-12.24).
- risk associated with air pollution from wood burning
heaters (IV-D-364, IV-D-535, IV-D-613, IV-D-657).
- risk associated with COg pollution (IV-F-4.30).
- risk associated with breathing dust (IV-D-322).
- risk associated with emissions from other local industries
(IV-D-271, IV-D-272, IV-D-280, IV-D-313).
- risk associated with living in polluted cities (IV-D-525,
IV-0-616, IV-F-5.14).
- risk associated with living with a smoker (IV-D-128,
IV-F-3.47).
- risk associated with DDT (IV-D-322).
- risk associated with toxic contaminants in medicines
(IV-D-453).
- risk associated with contaminants in flour (IV-D-621-12.5,
IV-F-5.12).
6-17
-------
- risk associated with arsenic residues in seafood (IV-D-537).
risk associated with toxic residues in vegetables and meat
(IV-D-323, IV-D-472).
- risk associated with pesticide residues in vegetables
(IV-D-452), and grains (IV-F-4.30).
- risk associated with eating vegetables that contain natural
pesticides made by plants to protect themselves from insects,
fungus and animals (IV-F-4.44).
- risk associated with artificial food coloring and food
preservatives (IV-D-278, IV-D-322, IV-D-406, IV-D-452,
IV-D-472).
- risk associated with fats present in such foods as meat and
buttermilk; these fats can be broken down in the body to
mutagenic substances (IV-F-4.44).
- risk associated with burned and brown foods, including
everything from carmelized sugar to toast, that contain
mutagenic substances (IV-F-4.44).
- risk associated with living (IV-D-187, IV-D-353, IV-D-472).
- risk associated with heart disease (IV-D-621-12.16).
- risk of being kidnapped (IV-F-4.49).
- risk of being murdered (IV-D-621-12.24).
- risk associated with war (IV-D-384).
6-18
-------
- risk of drowning (IV-D-760).
- risk associated with natural disasters: floods
(IV-0-621-12.24) and volcanic eruptions (IV-D-621-12.13).
- risk associated with untraviolet radiation from sunlight
(IV-D-621-12.13).
- risks that existed in past years when automobile and
smelter emissions were higher than they are today (IV-D-695,
IV-D-366).
6.1.4.2 Risk Associated With Proposed Standard is Unacceptable. The
following comparisons were made by other commenters to illustrate that the
risks associated with the arsenic emissions from ASARCO-Tacoma are much
more than those associated with other involuntary risks. They also
provided guidelines for what the commenters believe is an acceptable
level of risk.
Involuntary Risks
- risk of botulism from canned foods (IV-D-71).
- risk associated with nuclear radiation and fallout is at
least as great as the risk associated with exposure to heavy
metals (IV-D-41).
- cancer risk to the general population (IV-D-120, IV-0-164,
IV-D-590, IV-D-666, IV-F-1.17, IV-F-4.71).
- risk levels in other industrial geographic areas (IV-D-114,
IV-D-142, IV-D-582).
6-19
-------
risk levels comparable to natural background levels of
arsenic in other communities (IV-D-721, IV-D-771, IV-F-3.43,
IV-F-3.54, IV-F-4.6, IV-F-3.103). Three commenters (IV-D-734,
, IV-F-3.43, IV-F-3.54) went on to state that the body or urinary
levels of arsenic in persons living around ASARCO-Tacoma should
not be greater than are normally found elsewhere in the
country. Some commenters (IV-D-593, IV-D-643) specifically
mentioned that children's urinary arsenic levels should be
within a normal range.
risk levels in other communities affected by arsenic
emissions (IV-D-114, IV-D-438, IV-D-443, IV-F-3.20,
IV-F-3.53, IV-F-3.58, IV-F-4.43, IV-F-4.52, IV-F-5.18,
IV-F-3.103, IV-F-11). One commenter (IV-F-4.43) said
that the risk to Tacoma is ten times the combined total
risk to 14 other communities that have copper smelters.
risk levels associated with hazardous compounds that are
regulated by other agencies, such as FDA (IV-D-621-14.9,
IV-F-3.43). One commenter (IV-D-621-14.9) suggested that EPA
adopt FDA's stringent action level for carcinogens (1 in
1,000,000) because the aggregate risk for arsenic, $03,
cadmium, etc. could be quite large.
risk levels associated with other environmental standards
(IV-D-541, IV-D-580, IV-F-11). One commenter (IV-F-1.18)
stated that arsenic is about as carcinogenic as DDT, EDB,
chlordane and heptachlor, all of which have been banned by
EPA regulation under FIFRA. He also added that FIFRA does
not impose as stringent a requirement for the maintenance
of an ample margin of safety as does the Clean Air Act.
6-20
-------
Another commenter (IV-F-11) said that limits on such pollutants
as benzene, vinyl chloride, and dioxin are more restrictive
than the proposed arsenic standard.
- risk levels associated with other EPA regulated hazardous air
pollutants (IV-D-710, IV-F-1.18, IV-F-3.43).
- risk levels associated with other EPA regulated air-borne
carcinogens (IV-D-142, IV-D-147, IV-D-314). Other commenters
mentioned benzene (IV-D-443, IV-D-621-14.9), dioxin
(IV-D-443) and vinyl chloride (IV-D-120, IV-D-443,
IV-0-621.14.9, IV-F-1.18). Elaborating further, one
commenter (IV-F-1.18) stated that EPA regulations for vinyl
chloride reduced the lifetime risk of cancer to 1 in a
million. He continued by stating that the emission standards
for many major vinyl chloride sources were even set at zero.
However, he noted that according to the Health Assessment
Document for arsenic (p. 5-145), arsenic is three times more
carcinogenic than vinyl chloride. Another commenter
(IV-D-621-14.9) stated that the unit risk estimates for both
vinyl chloride and benzene are on the order of 1:100,000,
whereas, the unit risk estimate for arsenic is about 400
times higher.
Response:
Comparing the risks associated with arsenic exposure to risks
associated with activities such as cigarette smoking and drinking alcohol is
inappropriate, because risks due to arsenic exposure are largely involuntary.
That is people who live near the smelter may be unaware of their inorganic
arsenic exposure or, because of thei r ci rcumstances, cannot relocate in
some other more acceptable area.
6-21
-------
Many commenters mentioned involuntary risks that they perceived to be
either less or more than the risks associated with the proposed standard.
Comparing the risk levels associated with these involuntary risks (particularly
those associated with environmental hazards or contaminated food products)
to the risk levels associated with the inorganic arsenic emissions, may not
be appropriate because different risk methodologies and different assumptions
may have been used to calculate them. (See the discussion of the uncertainties
associated with the risk determination model, see Section 4.2.) However, EPA
understands the desire of the public to seek a reference.for relating to
the estimated risk levels associated with inorganic arsenic source categories.
The EPA believes that comparing the estimated increased lung cancer
risk associated with inorganic arsenic source categories to national lung
cancer rates provides a useful perspective (see Table 6-1).
Table 6-1. National Annual Cancer and Lung Cancer Rates - All Ages (1982)a
Annual Deaths'30 Percent ofc
Per 100,000 Total Deaths
Malignant neoplasms of respiratory
and intrathoracic organs
50.2
5.8
Malignant neoplasms, including
neoplasms of lymphatic and
hematopoietic tissues (cancer-
all forms)
188.1
21.9
a Source: "Monthly Vital Statistics Report," National Center for Health
Statistics, Vol 31, No 13, October 5, 1983.
b Based on a ten percent sample of deaths
c Rates are not age-adjusted.
6-22
-------
6.2 BAT APPROACH
Overview;
At the time of proposal EPA used a series of steps in deciding what the
level of Section 112 standards should be. This series of steps included one
step in which a determination was made concerning what level of control
constitutes best available technology (BAT). As expressed in the preamble
to the proposed standard (48 FR 33116), EPA's policy for implementation of
Section 112 was as follows:
1. Source categories are identified on the basis
of estimates of their potential to result in significant
risk because risk to public health is the dominant theme
of Section 112. A significant risk is considered to be
associated with a source category when the weight of the
health evidence indicates a strong likelihood that the
substance emitted by the source category is a human
carcinogen and either individuals or larger population
groups are significantly exposed to the substance as
emitted from the source category.
2. All source categories that are estimated to
result in significant risks are evaluated and the
current level of control ascertained. That control may
result voluntarily or from State, local or other Federal
6-22a
-------
regulations. Whether the level of control meets the
definition of BAT (considering cost and other inpacts)
then is determined. The BAT determination in this case
can take into account such factors as the potential for
improved control, the economic impacts of improved
control on the source category., and the age and
remaining useful life of the facilities.
3. The use of risk estimates generally has been
confined to areas of broad comparisons, e.g., in
selecting source categories to evaluate, and in
assessing the incremental change in risk that results
from application of various control options. The use of
risk estimates in an absolute sense is avoided because
of the many uncertainties of the estimates. These
uncertainties are compounded as the focus is narrowed.
In other words, in evaluating specific sources, as
opposed to source categories, the uncertainties
associated with the risk estimates increase dramatically.
4. Cost-effectiveness is one of the criteria
used in selecting BAT. However, the use of cost-effec-
tiveness in the BAT selection may result in some
apparent disparities in risk improvement at some
sources. Risk estimates are highly uncertain
while technology and cost are generally well understood
and provide an objective means of determining reason-
ableness of control.
5. If in the judgment of the Administrator, if the
residual risks after BAT are unreasonable, then the source
category must be controlled to a more stringent level.
Whether the estimated risks remaining after the application
of BAT are unreasonable will be decided in light of a
judgmental evaluation of the estimated residual risks (and
6-23
-------
their uncertainties), the economic, energy and environmental
impacts of further reducing those risks, the readily available
benefits of the substance or activity producing the risks and
the availability of substitutes and possible health effects
resulting from their use.
The public comments indicated that the risk management approach, as
described in the July 20, 1983, Federal Register notice of proposal, did
not give sufficient consideration to the protection of public health.
Evidently, some commenters saw the selection of BAT as the final step in
the decision-making process. Also, there seemed to be some level of
misunderstanding as to what BAT represented and some confusion between
similar terms used in other EPA programs such as "best available control
technology" (BACT) found in the Prevention of Significant Deterioration
program and "best available technology" (BAT) in the water program.
Based on consideration of the public comments (as described on the following
pages), the above concerns of possible public misinterpretation, and the
recent experiences that the Agency has had with other pollutants, the
Administrator has decided to refine the risk management process described
in the proposal.
The EPA's refined strategy for risk management under section 112 provides
for the comprehensive assessment of candidate source categories to evaluate
current control levels and associated health risks as well as options for
further control, the health risks reduction obtainable and the associated
costs and economic impacts. Based on this assessment, EPA selects a level of
control which in the judgment of the Administrator reduces health risks to
the greatest extent possible, cognizant of the other impacts of regulation.
The EPA believes this approach is both rational and consistent with the
requirements of section 112. The major steps in EPA's procedures are
outlined below.
1. Source categories are identified on the basis of estimates of
their potential to result in significant risk. A significant risk is
considered to be associated with a source category when the weight of the
6-24
-------
health evidence indicates a strong likelihood that the substance emitted
by the source category is a carcinogen and either individuals or larger
population groups are significantly exposed to the substance as emitted
from the source category.
2. All source categories that are estimated to result in significant
risks are evaluated and the current level of control ascertained. The EPA
examines the various options available to reduce emissions from these
sources, including controls similar to those imposed under Section 111 of
the Clean Air Act (New Source Performance Standards), the use of substitute
feedstock materials, and closing a plant. Options are examined in terms
of control efficiency, technical feasibility, costs, and the reductions in
risk that they achieve. If a source category is not already required to
apply the selected emissions reduction option, EPA will set the Section 112
standard which reflects the level of control of the selected option. If a
category is already controlled (for example, by other EPA standards, other
Federal, State, or local requirements, or standard industry practice) to the
selected level, and EPA expects that the level of control will be required
for these and new sources (EPA will continue to monitor this), a Section 112
standard would be redundant and need not be established. The level of
control selected by the Administrator may be different for new and existing
sources within a source category because of higher costs associated with
retrofitting controls on existing sources. When selecting the control
option, the Administrator considers whether the estimated risks remaining
after application of each level of control are unreasonable. This is of
a judgmental evaluation of the estimated maximum lifetime risk and cancer
incidences per year remaining after application of each control option,
the impacts, including economic impacts, of further reducing those risks,
and the readily available benefits of the substance or activity producing
the risks. In all cases where risks and other parameters are estimated,
the significant uncertainties associated with those numbers will also be
considered in reaching the final decision.
6-25
-------
As can be seen when comparing the current risk management approach to
the one given in the proposal, the term "BAT" has been removed. This change
reflects something more than just a revision in terms: this is a refined
approach used selecting the final control option as a basis for the Section
i
112 regulation. Instead of the previous multi-step process, this approach
incorporates an amalgam of elements of the BAT residual risk approach combined
with the elements of the two risk-based alternatives set forth in the proposal.
With the refined approach there is no separate step to determine the appropriate
level of control and then to examine the reasonableness of the residual risks.
Rather, these two steps are combined into a single selection process which
involves considering simultaneously the possible control options and the
technical, economic, public health, and other implications of each option.
This refinement, the Administrator believes, is both rational and more
consistent with the language of section 112, and, as seen by reading the
following comment summaries, it responds to many concerns of the commenters
on this proposal.
Comment:
Many commenters (IV-D-61, IV-D-74, IV-D-114, IV-D-142, IV-D-147,
IV-D-301, IV-D-345, IV-D-401, IV-D-438, IV-D-443, IV-D-524, IV-0-541,
IV-D-557, IV-D-593, IV-D-604, IV-D-608, IV-D-609, IV-D-618, IV-D-660,
IV-D-662, IV-D-663, IV-D-677.3, IV-D-747, IV-F-1.1, IV-F-1.18, IV-F-3.31,
IV-F-3.58, IV-F-3.60, IV-F-3.103, IV-F-4.15, IV-F-9, IV-F-11) objected to
what they saw as EPA's setting standards for ASARCO based on "best available
technology" (BAT). Commenters felt that basing a standard on BAT placed
primary emphasis on issues other than health, such as affordability, technology
and economics. The commenters felt that health concerns were the appropriate
primary emphasis.
Other commenters (IV-D-154, IV-D-231, IV-0-237, IV-D-271, IV-D-288,
IV-D-399, IV-D-464, IV-D-480, IV-D-519, IV-D-622, IV-F-3.7, IV-F-3.8,
IV-F-3.50, IV-F-4.49, IV-F-9, IV-F-11) favored basing a standard on BAT,
6-26
-------
calling it a reasonable, logical approach. Some felt it reasonable in the
face of uncertainty concerning health risk. Others felt it an equitable
approach until other smelters were required to install similar levels of
controls. One (IV-D-622) said that anything less than BAT would be unaccept-
able.
Response;
It is important to keep in mind the fact that the application of BAT
was only one step within the risk management approach as described at
proposal in deciding what level of control should be applied to a source
category. Standards for hazardous air pollutants were not to be based on
BAT unless, in the Administrator's judgment, the residual health risk levels
after the application of BAT were not unreasonable. Within the context of
the risk management approach discussed at proposal, EPA judged that all
source categories of a hazardous air pollutant which are estimated to result
in significant risk should be at a minimum controlled to a level which
reflects BAT. Each such source category would then be controlled to a
greater degree if, in the judgment of the Administrator, it was necessary
to prevent unreasonable risks.
Thus, the proposed decision-making process begins and ends with the
consideration of risks because the Agency views its primary mission under
the section 112 as the reduction of public risk.
As a practical matter, there is a certain portion of the Agency's time
spent on evaluating factors that are not directly related to reduction of air
pollution risks but are important in the overall selection of the appropriate
control option. For instance, has the technology been demonstrated at
other installations as a means to reduce emissions? If required, can the
control device actually be used safely on the process or the stack gases?
Will the control technology create a pollution problem in another medium such
as the water or land? Is the control technology so expensive that its
application will surely shut the plant down? The EPA agrees with those
commenters who felt that answers to these kinds of questions must be part
of the control option selection process. Such analyses are part of the
Agency's refined risk management approach.
6-27
-------
The Agency agrees with those commenters who perceived that the protection
of public health did not weigh very heavily in the selection of BAT; yet the
Agency did not disregard the reduction of public risks. The effectiveness of
the control equipment to reduce emissions (and risks) was weighed against the
costs to install and operate that control equipment. Also, to the extent
possible EPA considered the impacts from the pollution controls on other
environmental media such as soil and water. After BAT was selected, the
Agency reviewed the level of residual risks (and thei r uncertainties),
determined if they were unreasonable, and considered requiring controls
beyond BAT. Thus, the risk management policy, as outlined in the proposal,
considered the protection of public health in each of the three steps for
selecting the controls to be used as a basis for regulation.
Comment:
Several commenters (IV-D-386, IV-D-399, IV-D-466, IV-F-3.44, IV-F-4.51)
noted that basing standards on BAT allowed for continued improvement. As
new technology becomes available and economically feasible, they saw it as
appropriate to require that technology for control of emissions. Several
commenters (IV-D-269. IV-D-271, IV-D-372, IV-D-373, IV-D-386, IV-D-403) said
that as long as ASARCO is making and is willing to make noticeable improve-
ments, the plant should be allowed to continue operation. One commenter
(IV-D-710) said EPA must revise standards periodically to continue to
approach the statutory goal of complete public health protection as rapidly
as possible. Another (IV-D-483) viewed EPA's role as one of helping
industry work out its environmental problems by giving them technology,
giving them easily attainable standards to meet over reasonable periods of
time, then setting tougher standards as time and technology move forward.
Comment:
Some commenters (IV-F-3.29, IV-F-3.31, IV-F-3.103, IV-F-3.31, IV-F-9)
objected to basing a standard on BAT because they felt it eliminated any
incentive on the part of the smelter to develop improved control technology.
One commenter (IV-F-1.17) complained that EPA needs to be pushed to. require
6-28
-------
even today's state-of-the-art controls, let alone any technological innovations.
Still other commenters (IV-D-73, IV-D-302, IV-D-575) advocated requiring
emissions to be essentially zero. The commenters said this would force
ASARCO to design and build totally effective anti-pollution equipment.
Another (IV-D-580) said that it seems that EPA is allowing existing technology
and its costs to call the shots rather than forcing technology under the
Clean Air Act. He called this an undesirable precedent. Another (IV-D-698)
rejected standards based on BAT. He advocated technology-forcing standards
under section 112 of the Clean Air Act. Another commenter (IV-D-710) said
that the law recognizes that a standard may be set at a level which reflects
a projection of what can be achieved by sources in the foreseeable future.
Response;
The EPA agrees that continued improvement in arsenic emissions control
is a desirable goal. However, the agency must be reasonable. The EPA cannot
set a rapidly moving target because it would be difficult if not impossible
for industry to comply.
For example, since it may take industry up to several years to design,
purchase and install control equipment, the controls could be outdated before
they are in operation. Once again, the company would have to begin designing
and purchasing the latest controls. The costs to the company would be very
high and, most importantly, the hazardous emissions are not being effectively
reduced; the company never gains expertise in the operation of the controls.
In general, the Agency does not plan to use such an approach. It could prove
infeasible for the Agency to implement.
In some circumstances, EPA's risk management approach will be forcing
technology. Where source owners (or source category) are required to apply
all the controls they can afford and the residual risks remain unacceptable,
the owners have only two operations available. Either the source must devise
some control option that is more effective than state-of-the-art controls
(technology-forcing) or close down their operations which are posing the
health risk problems. In this scenario, the owners must take some serious
steps to reduce further or eliminate their emissions.
6-29
-------
In addition, EPA periodically reviews the final emissions standards.
The review considers the availability of irtproved emission control tech-
nologies, process modifications or substitute materials. The Agency then
determines if there is a need for a change in the standards. The decision
process will focus on the amount of risk reduction that may result from
developing a more stringent standard. If the risk reduction is significant,
the Agency will carefully consider the possibility of more restrictive
standards.
Comment:
Some commenters (IV-F-1.7, IV-F-1.10) objected to basing a standard on
BAT because they saw it as penalizing those smelters which had installed
controls in the past. They said the cost-effectiveness analysis which
underlies BAT determinations makes it appear more costly for those who have
made improvements in the past and rewards those who have postponed
installing controls. In particular, one commenter (IV-F-1.10) saw it as
unfair to ASARCO-Tacoma because ASARCO had previously installed controls.
Another (IV-F-1.1) said that any approach where those who can afford it
pay for it, and those that cannot may have a lesser degree of control, creates
artificial competitive disadvantages.
Response:
It is hard to understand how cost effectiveness analysis could work
against a facility which had installed controls previously if the
previously-installed controls are effective ones. Cost effectiveness ratios
are expressed as the cost in dollars of adding additional controls divided
by the additional emission reductions which could be achieved in megagrams
per year. If a facility has installed effective control technology,
additional emission reductions achievable by additional control would
probably be small. A small emission reduction would cause the cost
effectiveness ratio to be a large number and, therefore, less attractive
as a control alternative.
6-30
-------
If a facility has installed ineffective control technology, additional
control technology may offer significant additional emission reductions. A
large emission reduction would cause the cost effectiveness ratio to be
small. A low cost effectiveness ratio indicates that the control technology
under consideration may be reasonable if the cost to the industry is
affordable.
Comment:
Some commenters (IV-D-489, IV-D-515, IV-F-3.7) expressed strong support
for continued efforts to develop control technology for arsenic emissions.
Comment:
One commenter (IV-D-710) said BAT is not the best technology in a
technical sense; rather, it is the best control available considering
economic, energy, and environmental impacts. The commenter inferred that
the term "BAT" was borrowed from the Clean Water Act in which it establishes
the test for toxic water pollutant standards. But, he said that in the
Clean Water Act, BAT denotes more stringent standards than BAT as articulated
for hazardous air pollutants regulated under the Clean Air Act. He
explained that although EPA takes costs and other factors into account when
establishing BAT for water pollutants, the best performing facilities
provide a floor below which BAT may not slip. The commenter cited a 1982
notice regarding water pollutant effluent limitations: "BAT limitations, in
general, represent the best existing performance of technology in the
industrial category or subcategory" (47 FR 46435, October 18, 1982). The
commenter objected to EPA's failure to recognize such a floor for hazardous
air pollutants. He felt that standards set for hazardous air pollutants
based on the BAT approach have in the past fallen short of requiring
technology even as good as the best already in use.
Response:
The EPA does not agree that the term BAT, as defined within the Clean Water
Act and its subsequent regulations, provides a floor below which BAT may not
6-31
-------
slip. The Federal Register notice which this commenter cites continues with
the following:
"In arriving at BAT, the Agency considers the age
of the equipment and facilities involved, the process
employed, the engineering aspects of control technologies,
process changes, the cost of achieving such effluent
reduction, and non-water quality environmental impacts.
The Administrator retains considerable discretion in
assigning the weight to be accorded these factors" (47 FR 46435,
October 18, 1982).
Thus, in the Clean Water Program, BAT may not necessarily reflect in all
cases the best performing control technologies because of case specific
differences which for some sources make this level of performance impossible
to attain.
Comment:
A commenter (IV-D-710) found BAT as defined and implemented for
hazardous air pollutants indistinguishable from the test applicable to New
Source Performance Standards under section 111. He said that this
implementation was contrary to Congress' intent that EPA set more stringent
requirements under Section 112 than under section 111.
The commenter (IV-D-710) provided a recommended alternative to EPA's
BAT approach. He argued that at a minimum, Section 112 must be interpreted
to mandate standards which require technology at least as good as the best
in use now or available in the foreseeable future. The commenter said that
the required technology should include all design, operational, and mainte-
nance improvements that can be installed at present or within reasonable
lead times. Another commenter (IV-D-572) agreed with the first commenter's
(IV-D-710) reasoning, saying that ASARCO must set pollution control levels
at the lowest possible levels, not on achievable levels which are claimed to
be affordable. Another (IV-D-778) felt EPA had based its standard on
"Best Affordable Technology" rather than "Best Available Techhnology", and
6-32
-------
he objected to this. Similarly, another commenter (IV-D-721) objected to
EPA's policy of allowing a conpany to install only the available technology
it says is affordable. He said this allows the conpany to resist development
or installation of further control technologies. The first commenter
(IV-D-710) went on to say that it was only after this stringent minimum
has been applied that risk assessments should be used.
The commenter referred to a settlement agreement for litigation over
the vinyl chloride standard as evidence that EPA had once embraced his
recommended approach. He further urged EPA to return to the approach taken
with vinyl chloride. Another (IV-D-731) said EPA's analysis of available
technology and selection of BAT is less stringent than existing regulations
for other hazardous air pollutants.
Response:
The commenters are arguing for a minimum requirement for all sources of
hazardous air pollutants. The minimum requirement, being fostered by the
commenters appears to be the best technology in use now or available in the
foreseeable future, regardless of cost of current emission levels or current
risk estimates. After this minimum level of control has been applied, the
commenters would favor examination of residual risk.
This approach is similar to the one discussed at proposal but differs
in the way in which the minimum level of control would be chosen as a first
step. The EPA's implementation policy at proposal included requiring best
available technology considering economic, energy, and environmental impacts.
The commenters would apply a more stringent minimum requirement before
examining residual risk.
As previously discussed in this section, EPA has refined the approach
described at proposal to one in which the Administrator considers all
factors and impacts together in making his decision. Based on EPA's
experience to date, using the approach suggested by the commenters would
have serious economic consequences on certain source categories (and the
surrounding communities) such as the primary copper smelters. In addition,
6-33
-------
applying the best technically available controls may provide little or no
reduction in risks if the existing sources are already applying some
control measures or if the application of further control creates a
hazard in another medium. The Administrator feels that such information
must be considered in the process of selecting the appropriate level of
controls and in certain situations, applying the most stringent level of
control, regardless of its economic and environmental impacts, does not
constitute sound public policy. The above concerns are part of the
reason why EPA has refined the decision-making process.
Comment:
When danger from emissions remaining after application of this minimum
level of control remains great, a commenter (IV-D-710) stressed that EPA
must set more stringent standards. One commenter (IV-D-718) said that if
any doubt remains that additional controls might be warranted, it is
EPA's legal and moral obligation to go beyond BAT. In some cases the
commenter foresaw no alternatives to closing a plant.
Another commenter (IV-D-571) offered some suggestions for going beyond
BAT. He favored using technology-forcing criteria or emission taxes to make
going beyond BAT possible.
Response:
The Agency agrees with the commenters. As stated earlier in EPA's
risk management approach, the Administrator considers the health risk
estimated to remain after implementation of a control option. If, in
his judgment, the residual risk is unreasonable, he will require a more
stringent control option which may include plant closure.
Comment;
One commenter (IV-D-617) felt that alternatives considered in selecting
BAT should be limited to technologies that have been demonstrated to be
6-34
-------
feasible and effective for the source category under consideration. The
commenter was basically supportive of the BAT approach, noting several
advantages. He said it allows EPA to determine whether the costs and other
impacts of a control requirement are disproportionate to the resulting
emission reduction benefits. Another virtue he cited is that it allows
recognition that BAT may already exist in certain source categories and that
no standards need to be established for those categories. The commenter
felt that this approach agreed with his opinion that controls should only be
as stringent as needed to eliminate a section 112 risk and that controls
should not be required when they are not necessary.
The commenter predicted that when BAT is selected with attention to
cost effectiveness and is applied to sources which have been pre-sorted on
the basis of population exposure, it is unlikely that residual risks will be
unreasonable.
Response:
As discussed in previous responses in this section, EPA has refined
its decision-making procedures. However, the selected control option's
feasibility and effectiveness, costs, current control levels and risk
estimates are still considered in the decision-making process.
Comment;
One commenter (IV-D-710) said that there is no legal basis for the
determination of whether the risk remaining after application of BAT is
reasonable. He called the judgmental evaluation of risk remaining, the
impacts (including economic) of further reducing the risk, and the benefits
of the substance producing the risk a cost-benefit analysis. The commenter
contended that there is no basis in the Clean Air Act for applying a cost-
benefit analysis under section 112.
The commenter said that in practice, the analysis of residual risk is
nothing more than a repetition of the analysis of BAT because EPA had never
come to a conclusion that a standard should go beyond BAT.
6-35
-------
The commenter further argued that risk estimates being as uncertain as
they are, in implementing a statute requiring a precautionary, preventative
approach, EPA cannot rationally use risk assessments as a basis for not
requiring the use of available emission controls on all sources of arsenic
emissions.
Response:
As EPA pointed out at proposal, the statute requirements does not
accommodate air carcinogens that may pose health risks at any levels of
exposure. Therefore, EPA has adopted a pragmatic approach to regulate such
pollutants after considering residual risks, costs and other factors.
Rejecting the idea of zero risk and massive plant closures, EPA does consider
costs and risk reductions achievable in selecting the control option for
the standard. The concept presented by the commenter of not examining
residual risk and costs and other impacts of reducing risk still further is
one which EPA rejects.
The EPA has not had to make a decision to go beyond BAT. The EPA
and industry have been able to find solutions which have allowed for
continued operation. The option is still there, however. If the
Administrator should determine that measures more stringent than BAT
(including plant closure) are required to protect the public health, he
would act to require those more stringent measures.
Comment:
One commenter (IV-D-618) saw large differences across source categories
in the level of costs EPA found reasonable in determining BAT. The
commenter said he could find no clear criteria applied in a consistent
fashion which would differentiate among the controls considered to be BAT.
Response:
BAT determinations included consideration of feasibility and economic,
energy, and environmental impacts. Cost was only one part of the
consideration.
6-36
-------
Comment:
One commenter (IV-D-641) said that BAT based on economics should be
used only as a baseline. Where health risk is significantly higher than for
other plants, the commenter felt that LAER should apply.
Response;
The EPA agrees in concept with the commenter. Where health risk is
unreasonably high, stringent control measures must be applied. The level
of control selected may be even more stringent than what might be considered
lowest achievable emission reduction (LAER). The EPA is not using pre-
determined levels of control in the standard selection process.
Comment:
One commenter (IV-D-698) said that EPA's analysis of available control
technology does not meet the Clean Air Act requirements. The commenter said
that EPA purports to have established BAT for each source category, yet
technology-based standards are only allowable when emissions standards are
not feasible. The commenter said EPA had not demonstrated the infeasibility
of emissions standards. Another commenter (IV-D-609) said that adoption of
BAT as an approach to emission control would require additional
Congressional legislation.
Two commenters (IV-D-621-5, IV-D-621-15.1) favored the BAT approach
over the approach of setting an ambient air standard. Specifically, one
(IV-D-621-15.1) felt that the proper way to deal with control of fugitive
arsenic emissions is by means of identifying the source and determining
what, if any action can be taken to control or reduce the emissions.
Response:
The commenters appear to be confusing the decision on the level of the
standard with the decision on the format of the standard. The Clean Air Act
6-37
-------
specifies that a section 112 standard must be expressed as an emission limit
unless it is infeasible to do so. In that case, section 112 states that the
Administrator may instead promulgate a design, equipment, operational, work
practice standard.
Under both the past and the current approach, when selecting a control
option EPA considers available technology which could be used to meet the
standard, associated costs of that control technology and the level of
residual risks.
Comment:
One commenter (IV-D-604) saw the BAT approach as creating a de facto
air quality standard. The commenter further reasoned that since the same
proposed BAT applies to both low and high arsenic feed copper smelters, EPA
has created two margins of safety for the public.
Response:
The commenter appears to have misunderstood the concept of BAT as it
was presented at proposal. BAT, because it depended on economic, energy,
and environmental factors, could be set at different control levels for
different source categories. As mentioned earlier, determining BAT was
not the final step of the decision-making process. BAT was the selected
control option if the residual risks were not unreasonable in light of
the impacts of requiring controls beyond BAT.
The commenter appears to have observed two levels of residual risk,
surmising that the Administrator has defined two different numbers for
risk levels that are not unreasonable. This concept is incorrect. A
range of residual risks could be considered not unreasonable depending
on the outcome of the evaluation of the factors used in the decision-making
process.
6-38
-------
Comment:
One commenter (IV-D-144) stated that EPA's approach to determining
acceptable risk requires that EPA estimate the cancer risk remaining for the
population after controls are in place. Then EPA determines if the
remaining cancer risk is acceptable taking into account the costs and
technical feasibility of reducing the risk further. The commenter suggested
that the degree of risk be defined first. Then, the economic and social
costs of reducing this risk need to be assessed. Finally an acceptable
level of risk can be determined.
Response;
All the factors the commenter suggested for inclusion in the rulemaking
process are included in the Administrator's considerations. However, the
order of consideration may differ. The commenter is suggesting that changing
the order of factor consideration will affect the final decision. The
refined policy calls for the simultaneous consideration of all relevant
factors and so accommodates the commenter's concerns.
Comment;
One commenter (IV-F-4.59) said that an ample margin of safety is
related to what is technologically feasible.
One commenter (IV-F-1.18) felt that the Clean Air Act should require
the best available technology (BAT), even if BAT drives a company out of
business. Congress should then decide if plant closure is an unacceptable
tradeoff between risk reduction and the cost of compliance.
Response:
The commenters are expressing opinions about the role that technology,
its capabilities and its costs, should play in determining the level of a
standard. As defined in the proposal preamble, the requirement of BAT would
not drive a source category out of business; however, individual sources
within a large source category might be impacted in this manner. However,
6-39
-------
both the BAT and the current risk managmement approaches focus on the
reasonableness of the residual risks and does not stop at determining
technologically feasible control options.
Comment:
One commenter (IV-D-609/IV-F-4.15) stated that it does appear that
Congress intended Section 112 to cause shutdown of any industry that either
cannot or will not comply with air quality standards protective of public
health. He noted that the Senate committee that enacted the Clean Air Act
denied that the concept of technical feasibility could be used as the basis
for establishing ambient air standards, saying that the public health is
more important than the question of whether the early achievement of
ambient air quality standards protective of health is technically feasible.
The commenter quoted the Senate Committee report (S. Rep. N. 1196):
"In the Committee discussions, considerable concern
was expressed regarding the use of the concept of
technical feasibility as the basis of ambient air
standards. The Committee determined that (1) the health
of the people is more important than the question of
whether the early achievement of (ambient air quality)
standards protective of public health is technically
feasible; and (2) the growth of the pollution load in
many areas even with the application of available
technology, would still be deleterious to public health.
"Therefore, the Committee determined that existing
sources of pollutants either should meet the standard of
the law or be closed down ..."
Response:
The commenter is referring to national ambient air quality standards
instead of the hazardous emission standards presented in this package.
Unlike the criteria pollutants, the health effects associated with arsenic
public exposure levels are not documented and the protection of public
6-40
-------
health presents a more difficult determination. However, EPA agrees that
public health is the primary concern for section 112 standards as it is
when setting ambient air quality standards. Technical feasibility is
also of concern but only one factor considered when the Administrator has
determined that the public health risks are not unreasonable.
6.3 ECONOMICS AS A DECISION-MAKING CRITERION UNDER SECTION 112
Comment;
Some commenters (IV-D-439, IV-D-541, IV-D-557, IV-D-630, IV-D-710,
IV-D-724, IV-D-754, IV-D-778, IV-F-3.7, IV-F-3.31, IV-F-4.6, IV-F-4.15)
stated that EPA is required to place the protection of public health and
the environment, not costs or the availability of technology, as the
primary consideration in developing standards. Other commenters (IV-D-25,
IV-D-106, IV-D-112, IV-D-137, IV-D-698, IV-F-3.1) said that the proposed
arsenic standard was based on economic feasibility, an action which is
against the legal mandate of the Clean Air Act to provide an ample margin
of safety.
One commenter (IV-D-224) said the foundation of EPA and the Clean Air
Act is to protect people's health and the environment, not to attack their
health and well-being for the sake of the financial well-being of the copper
industry. Another commenter (IV-D-641) said that EPA should not avoid its
responsibility to protect health through case-by-case acceptance of high
risks by locality to avoid closure of a major local industry. According to
the commenter, such a policy could result in the location of inadequately
controlled facilities in economically depressed areas.
Some commenters (IV-D-710, IV-F-4.6) stated that Congress had no
intention of authorizing EPA to perform cost-benefit analyses when setting
hazardous air pollutant standards. Another (IV-D-747) thought cost-benefit
analysis was one of several factors which might be considered, but should
never be the basis of standards for hazardous pollutants. He felt in
this case EPA has overemphasized costs to industry and underenphasized health
6-41
-------
costs and benefits to individuals and society. One commenter (IV-D-609/
IV-F-4.15) stated that it does appear that Congress intended section 112
to cause shutdown of any industry that either cannot or will not comply
with air quality standards protective of public health.
Some commenters (IV-F-4.59, IV-D-718, IV-D-724, IV-D-731) felt that
EPA's proposed regulation implies that economic risks can be considered
under section 112 of the Clean Air Act although section 112 itself does
not allow this. One commenter (IV-F-4.59) also said that it was not the
intent of Congress to include economic analyses in the decision-making
process. He continued by pointing out that the regulatory process starts
with BAT, and if that is not stringent enough9 more controls must be added
until an ample margin of safety is reached without'regard to economic
factors.
Similarly, the Attorney General's Office of the State of New York
(IV-D-698) stated that section 112 has been violated by establishing BAT
based on costs. They continued by saying that EPA must either set emission
limits or establish technology-forcing performance standards. They
contended that EPA has not demonstrated that emission standards are not
feasible as required in the Clean Air Act.
Both the Natural Resources Defense Council (IV-D-710) and the Attorney
General's Office of the State of New York (IV-D-698) stated that the New
Source Performance Standards were clearly intended by Congress to apply to
less dangerous pollutants. They continued by saying that Congress
explicitly provided authority in section 111 to consider costs. However,
the commenters said, the Act does not mention that economics should be
factored into section 112. Congress intended, and the law requires, EPA to
set more stringent, more protective standards for hazardous air pollutants
regulated under section 112. New York State (IV-D-698) went on to say that
EPA has simply abandoned section 112's requirements and followed the easier
path laid by section 111.
One commenter (IV-F-3.7) pointed out that while section 317 of the
Clean Air Act mandates that an economic impact assessment be made, it also
6-42
-------
states that this information does not affect or alter the final decision in
setting standards.
Another commenter (IV-F-4.71) noted that the Toxic Substances Control
Act states that, while economic impacts of EPA decisions must be considered,
they must not prevent implementation of the strictest standards necessary to
protect public health.
One commenter (IV-D-466) asked EPA to use sound scientific basis with
plenty of weight on economic effects in regulating arsenic emissions.
Others (IV-F-3.4, IV-D-728 ) said that economic data for the local area
are important and must be considered in setting standards or policies.
The Chemical Manufacturers Association (IV-D-617) stated that the
margin of safety concept embodied in the Clean Air Act Amendments was
intended by Congress as a means of providing a "reasonable degree of
protection" for public health, not as an instrument for eliminating environ-
mental health risks entirely. Citing legislative history, this commenter
continued by stating that Congress was well aware that equating the term
"margin of safety" with absence of risk would be "an illusion" that "ignores
all economic and social consequences." (1977 Legislative History at 2578,
House Report).
Response:
Section 112 of the Clean Air Act is a potentially powerful tool which
does not provide explicitly, either in language or legislative history, for
the weighing of the benefits of control against the control costs. At face
value, section 112 could be construed to require regulation even when the
costs clearly exceed any measurable benefit. A total disregard for
economics would result in a zero risk philosophy. However, this philosophy
has been dismissed by EPA as being impractical (see Section 6.1.3 on Zero
Risk). In view of this, EPA has sought to construct an approach to the
implementation of section 112 which will not necessitate the establishment
of regulations which would impose costs unreasonably disproportionate to the
benefits obtainable.
6-43
-------
This approach considers current control levels and associated health
risks as well as options for further control, the health risk reductions
obtainable and the associated costs and economic impacts. Based on this
assessment, EPA selects a level of control which in the judgment of the
Administrator reduces health risks to the greatest extent possible,
cognizant of the significance of the residual risks and the societal impacts
of the regulations. The EPA believes this approach is both rational and
consistent with the requirements of section 112.
6.4 RECOMMENDED ACTION IN FACE OF UNCERTAINTY
Comment;
Many commenters (IV-D-162, IV-D-170, IV-D-177, IV-D-179, IV-0-181,
IV-D-185, IV-D-193, IV-D-196, IV-D-212, IV-D-221, IV-D-229, IV-D-230,
IV-D-250, IV-D-281, IV-D-298, IV-D-299, IV-D-312, IV-D-326, IV-D-333,
IV-D-316, IV-D-339, IV-D-349, IV-D-367, IV-D-370, IV-D-371, IV-0-372,
IV-D-373, IV-D-382, IV-D-383, IV-D-456, IV-D-460, IV-D-465, IV-D-474,
IV-D-485, IV-D-486, IV-D-508, IV-D-516, IV-D-545, IV-D-633, IV-D-659,
IV-D-735) felt that the ASARCO/Tacoma smelter should not be put in economic
jeopardy. Commenters felt that the smelter should remain open and smelter
workers should remain secure in their jobs because there is no proven
link between smelter emissions and lung cancer.
One commenter (IV-D-322) said that until better information is
available, EPA should remove the risk portions from its standards. Another
(IV-D-376) said that EPA should not impose any more stringent regulations on
arsenic emissions until a link between arsenic emissions and cancer is
established. Similarly, another commenter (IV-F-3.50) suggested that it
would he prudent to proceed with the proposed standards and the best
available technology until more is known about the risk associated with
arsenic. Some commenters (IV-D-588, IV-D-621-16.3, IV-F-3.39, IV-F-5.17)
felt that actions to control emissions and provide an ample margin of safety
must be based on irrefutable proof. Another (IV-D-542) did not support
implementation of strict ambient standards for arsenic when there are no
6-44
-------
scientific data to indicate its necessity for achievement of an ample margin
of safety.
Commenters (IV-D-274, IV-D-466, IV-D-569) said that EPA should base its
regulations on facts and not on theory or supposition. Others (IV-D-332,
IV-D-388, IV-D-391, IV-D-763) said that actions to reduce industrial pollution
must be based on objective studies and analysis, not emotion. One commenter
(IV-D-278) stated that EPA should take time to run the necessary tests and
give ASARCO time to prove what they are doing is working. Another (IV-D-489)
said that community decisions which attempt to balance impacts of potential
health hazards with jobs and other benefits should be based on fact, not
perception. Another (IV-D-512) said the smelter should not be shut down
because of false information or emotions.
Another (IV-D-538) stressed that it is essential that EPA make decisions
based on accurate, empirical scientific data rather than misinformed, vague
public concern. Another (IV-D-621-16.9) supported the use of the very best
scientific methods for examining the problem and appraising the risks. The
commenter felt the information should be verified and subject to peer revoew.
Another commenter (IV-D-735) supported continuing research on health risks.
Another (IV-D-729) felt that more information about the effects of arsenic
exposure could and should have been obtained by EPA. He suggested epidemi-
ologic studies and more accurate estimation of exposure and controls.
Another (IV-D-203) asked that EPA not let inaccurate data and faulty
assumptions of a few panic the Agency into choosing to regulate this
industry.
One commenter (IV-D-154) said that requiring ASARCO to spend millions
because someone may get cancer (or maybe not) is unreasonable because one
possible death (one maybe) is not significant.
Others (IV-D-155, IV-F-3.39) said that elimination of a source of
income to the Pacific Northwest without reasonable documentation of health
risk is unwise and inappropriate. One (IV-F-3.39) felt that to destroy a
6-45
-------
major source of financial support to the community by demanding an operation
so clean it cannot financially exist is a tough pill to swallow, especially
when the current arsenic levels have not been proven to be a health hazard.
One commenter (IV-D-625) recommended that EPA should not contemplate
the projected risks do, in fact, exist. He continued by stating that if the
existing data fail to resolve this uncertainty, consideration should be
given to conducting a more thorough epidemiological study. Another
commenter (IV-D-545/IV-D-621-16.6/IV-F-4.24) felt that since the available
scientific data are uncertain, more studies should be conducted to determine
if arsenic is a no-threshold pollutant. He continued by saying that
considering that more health data will be available and new control
technologies will be developed, there should be a periodic review of the
standard.
Some commenters (IV-D-338, IV-D-342) said that EPA needs to back off
and look more closely at actual data over a longer time period. Two
commenters (IV-D-144, IV-D-460) said that EPA cannot show emissions from
ASARCO-Tacoma are harmful to the community. One (IV-D-460) concluded that EPA
should drop its case against ASARCO. Two others (IV-F-1.1, IV-F-1.3) said
if there are no demonstrated community health effects, regulation is not
warranted. One commenter (IV-D-256) resented what he called EPA's attempt
at baffling the public with unsupportable statements. He said that EPA should
either find the true facts and be able to support them, or they should just
keep quiet.
Response:
Many commenters believe that EPA should wait to regulate arsenic
emissions until (1) the link between arsenic and lung cancer is established
beyond any doubt, and (2) the effect of low ambient concentrations of arsenic
on public health has been determined.,
The current status of inorganic arsenic as a human and experimental
animal carcinogen has been extensively and critically reviewed by public
agencies such as the National Institutes of Occupational Safety and Health,
6-46
-------
scientific bodies such as the National Academy of Science and the Inter-
national Agency for Research on Cancer, and in a number of individual
assessments. In addition, EPA's inorganic arsenic Health Assessment
Document has been reviewed by the Science Advisory Board, a group of
scientific experts from outside the Agency. At present, the collective
evidence for an etiological role of inorganic arsenic in human cancers is
strongest for cancers of the skin and lung. Cancer (and possible pre-
cancerous lesion) producing inorganic arsenic exposures have been
demonstrated in both occupational populations, such as copper smelters,
pesticide manufacturers and agricultural workers, and in non-occupational
populations using arsenical drugs or consuming arsenic contaminated
drinking water and/or food. (For further information see Chapter 2.)
However, the effect of low ambient concentrations of arsenic on public
health has not been adequately determined. Very little information exists
that can be used to extrapolate from high-exposure occupational studies to
low environmental levels. For several practical reasons as mentioned
earlier in this document, such low levels of risk cannot be readily measured
either by animal experiments or by epidemiologic studies. The linear non-
threshold model is used as the primary basis for risk extrapolation at low
levels of exposure. The EPA considers this model to be a viable possibility
for the true dose-response relationship (Health Assessment Document p. 7-89-
90).
The Agency is not required to wait until irrefutable proof that arsenic
causes cancer at low ambient concentrations is produced. It must be noted
that section 112 of the Clean Air Act defines a toxic air pollutant as that
which may reasonably be anticipated to result in an increase in mortality
or an increase in serious irreversible, or incapacitating reversible illness.
Comment;
Many other commenters urged an approach which would err on the side of
overprotection in the face of uncertain health results. Commenters urged
immediate mitigation of potential health risks. One commenter (IV-D-150)
6-47
-------
said that 1n the absence of clear knowledge of effects of arsenic on health
and environment, EPA should close the smelter. Others (IV-D-420, IV-D-717,
IV-D-722) advised that EPA take the same attitude and accused EPA of not
acting until a serious amount of damage has occurred. One (IV-D-420)
urged that EPA not take the gamble when cancer Is at stake. One commenter
(IV-D-13) asked how many cases of lung cancer are necessary to show that
a health hazard exists? Similarly, another (IV-F-3.29) asked why a health
hazard must be proven before any action Is taken, when It would be obvious
that a health hazard exists. One commenter (IV-F-4.11) reminded EPA
that they are under a court order to proceed in a determination before
all the facts are In. Another (IV-D-431) said that it makes no sense to
risk public health with unknown consequences just for profit. Another
(IV-D-11) asked EPA not to wait a generation to stop carcinogenic effects
from arsenic emissions after the effects have materialized. He felt that
enough is known right now to demonstrate the ill effects of AsarcO's
emissions on public health. Another (IV-F<-5.15) said that until more is
known about cancer, we cannot allow known carcinogens to be present in
our environment at possibly hazardous levels.
Another commenter (IV-F-4.67) stated that the procrastination on this
issue has gone too far. He did not. intend to accept the absence of evidence
for health hazards to be conclusive evidence of the safety of the emissions.
Two commenter (IV-F-3.53, IV-D-733) said that until there 1s proof
that elevated urinary arsenics are safe, ASARCO should be required to
control emissions to the point where local children's urinary arsenic
levels are normal.
Another (IV-0-8) advised EPA to proceed with caution 1n the face of
uncertainty. He said that EPA (society) cannot take the chance of being
wrong. He argued that EPA must assume arsenic is creating a health hazard
and must be prevented from entering the environment. Other commenters
(IV-F-3.41, IV-F-4.4, IV-F-4.9, IV-F-4.31, IV-D-708a, IV-D-747) said that
a safe exposure level may be difficult to determine, but EPA should err
on the side of safety and adopt as stringent a standard as possible.
6-43
-------
Two commenters (IV-F-3.55, IV-D-733), noting that nobody knows what
the combined effect of arsenic is with other pollutants, stated that the
only prudent thing to do is to reduce unnecessary exposure to all pollutants.
One (IV-D-628) recognized that certain actions must be taken to protect
the environment and the public without adequate scientific support data.
The commenter urged adoption of the proposed standards.
Another (IV-D-677.4) said fu rther studies should be conducted, but they
in no way should delay full enforcement of the proposed standards. Another
(IV-D-621-16.1) said it would be prudent to minimize human exposure to
arsenic by reducing arsenical emissions, especially low-level or fugitive
emissions. Others (IV-D-545, IV-D-621-16.6, IV-F-4.24) urged timely
adoption of the standards. They argued that given the doubts about the
presently available data, further delay would only delay the time when
reduction of present emissions could be accomplished.
One commenter (IV-F-3.21, IV-D-718) asked that EPA's proposed standards
serve as interim controls until more conclusive results are in. Another
(IV-F-4.24) recommended timely adoption of the proposed standards, stating
that delay in doing so will only further delay the reduction of current
emission levels. He also urged EPA to continue its research on the
health effects of arsenic and suggested that there be periodic review of
the standards as new evidence and/or technologies are developed. Another
commenter (IV-F-4.3) said that the delay in setting standards is uncalled
for- ASARCO should be allowed to install hoods immediately to eliminate
unnecessary health risks. After these standards are met, EPA should go
forward with further research. In the meantime, he said, EPA must assure
those who perceive a health risk that everything that can be done is
being done. Another commenter (IV-D-657, IV-D-733) urged adoption of the
proposed standards until the arsenic question is fu rther evaluated.
Another (IV-D-368) felt that further study is advisable and that ASARCO
should in the meantime be encouraged to cut down further on emissions if
possible.
6-49
-------
Response:
The EPA is following the prudent person policy advocated by many of
these commenters, erring on the side of protecting public health. Section
112 requires that standards be set at levels which, in the Administrator's
judgment, provide an ample margin of safety to protect public health.
Thus, one factor EPA considers is the nature and relative magnitude of
health hazards. Unfortunately., agencies can never obtain perfect data but
have to make regulatory decisions on the basis of the best information
available. So, EPA evaluates the potential detrimental effects to human
health.caused by pollutant exposure based on the best scientific information
currently available.
The scientific uncertainties not resolved to date include the
establishment of toxicity to humans based on extrapolation, using uncertain
mathematical models from occupational exposure to low-dose public exposure at
ambient air concentrations, and the identification of the appropriate level
of emission controls for pollutants for which health effects thresholds have
not been demonstrated. There also is uncertainty with exposure estimates
because of difficulty in obtaining precise data on emission rates, atmospheric
dispersion patterns and population concentrations around individual sources,
and because of the lack of information on short-term and long-term movement
(migration) of people and indoor versus outdoor toxic air pollutant concen-
tration patterns (see exposure and risk determination section). Finally,
there are uncertainties concerning possible additive effects of multiple
sources or pollutants, synergistic or antagonistic health effects, and
heightened susceptibilities to some cancers by some population groups.
These factors make it difficult, if not impossible, to determine the absolute
magnitude of the risk to human health based on the available data.
Comment:
One commenter (IV-F-4.6) said that in the Senate Report on Amendments
to the Clean Air Act, the Senate stated: "Margins of safety are essential
to any health-related environmental standards if a reasonable degree of
protection is to be provided against hazards which research has not yet
identified." Another commenter (IV-F-4.11) stated that a margin of safety
6-50
-------
must be incorporated in the permissible dose to compensate for the degree of
uncertainty in determining that dose. The less precise the determination of
hazard, the larger must be the margin of safety.
Response:
Again, with cancer-causing agents, there appears to be no level at
which an exposed individual is entirely safe (non-threshold pollutant).
The question surrounding the decision is the uncertainty and acceptability
of the risks which remain after the application of the selected control
option. The term "margin of safety" are more commonly used with threshold
pollutants and is not readily applied to inorganic arsenic.
6.5 JOBS VS. HEALTH
Comment:
Many commenters (IV-D-1, IV-D-32, IV-D-694, IV-D-43, IV-D-53, IV-D-61,
IV-D-62, IV-D-94, IV-D-107, IV-D-116, IV-D-138, IV-D-144, IV-D-151,
IV-D-158, IV-D-163, IV-D-224, IV-D-301, IV-D-375, IV-D-400, IV-D-426,
IV-0-431, IV-D-632, IV-D-672, IV-D-582, IV-D-643, IV-D-644, IV-D-637,
IV-D-670, IV-D-690, IV-D-241, IV-D-435, IV-D-437, IV-D-346, IV-D-674,
IV-D-435, IV-D-710, IV-D-677-1,JV-D-677-6, IV-F-4.50, IV-F-4.68, IV-F-5.10,
IV-F-5.15, IV-D-720, IV-D-730, IV-D-734, IV-D-783, IV-D-753, IV-F-9, IV-F-11,
OAQPS 79-8/IV-D-4, IV-D-705, IV-F-10) felt that health concerns were more
important than jobs or economic advantage. Several (IV-D-720, IV-D-778,
IV-F-10, IV-D-20, IV-F-3.30, IV-D-551, IV-D-661, IV-D-56, IV-D-10, IV-D-55,
IV-D-638, IV-D-224, IV-D-412, IV-D-421, IV-D-378, IV-D-87, IV-F-3.60,
IV-F-4.43, IV-D-688, IV-D-415, IV-D-66, IV-F-5.18) said it is EPA's job to
protect people's health from pollution, not to promote jobs and corporate
p rof i ts.
One commenter (IV-F-4.50) said that health, in keeping with the intent
of Congress, must be the one non-negotiable component. Another commenter
(IV-F-4.15) said EPA should not defer to industry and wait for cancer deaths
to occur. Another (IV-D-72) said that it is not the right of smelter
workers to choose jobs over public health. Commenters (IV-D-671, IV-D-660,
6-51
-------
IV-D-4, IV-F-4.35, IV-F-3.60) said that 600 jobs cannot justify even one
additional death per year.
One commenter (IV-D-444) said that human life is of the greatest value
and should not be compared to other things. He reasoned that a smelter
could not be of greater value than the people who operate it or whose
material needs brought it into existence in the first place. He further
argued that personal income could not be of greater value than a person.
Similarly, another (IV-F-4.13) stated that without life there can be no
jobs. Another (IV-F-3.67) said that when comparing loss of jobs to loss of
life, it must be taken into account that people can always get other jobs.
A commenter (IV-D-595) said that even if it were a choice between
health and jobs, given the seriousness of the health risk, the number of
residents adversely affected, and the number of jobs arguably at risk, the
decision should clearly be to protect the public health. Another
(IV-F-3.38) understood the profit motive of ASARCO, but did not think it
should infringe on the public.
Some commenters (IV-D-78, IV-D-9, IV-D-710) objected to EPA weighing
jobs v. health. The two commenters said that this was not allowable under
the Clean Air Act.
Comment:
Some commenters (IV-D-361, IV-D-464, IV-D-489, IV-D-507, IV-D-508,
IV-D-217) said that people of Tacoma need the jobs and money the smelter
puts into their economy. Another (IV-D-467) asked EPA not to consider
regulations that would put people out of business. Two (IV-D-278, IV-D-735)
said the smelter should not be closed or cause loss of jobs unless there is
solid proof of health risks. Still another (IV-D-162) said that jobs and
economic considerations are more important than a health risk he perceived
as minimal or nonexistent. Another (IV-D-395) said that the value of the
smelter to the community far outweighs any danger to the health and well-being
of people living in the area. Another (IV-D-550) said the economic significance
6-52
-------
of the smelter far outweighs the potential community health hazards.
Another (IV-D-349) said no more stringent standard should be proposed since
this would create unreasonable economic conditions for ASARCO and the State
of Washington. Other commenters (IV-0-483, IV-F-3.25) asked that EPA set
reasonable standards and provide more than a reasonable length of time for
ASARCO to reach those standards, especially in these tough economic conditions
One commenter (IV-D-713) attached a newspaper article to his letter which
stated that "Residents and governmental organizations have told the Federal
Environmental Protection Agency they will accept a small amount of cancer-
causing arsenic in the air so a copper smelter can continue to operate".
Comment:
Several commenters (IV-F-4.3, IV-F-4.6, IV-F-4.68, IV-F-5.1, IV-F-5.2,
IV-F-5.3, IV-F-5.4, IV-F-5.11, IV-D-623, IV-D-713, IV-D-737, IV-D-779,
IV-D-756) said that the jobs v. health question should not exist - the
community can have both. Another commenter (IV-F-4.4, IV-D-708a) stated
that jobs v. health is not the issue - the real issue is jobs and health
v. neither. One commenter saw the proposed standards as a way of protecting
the public health and keeping an economic contributor to the community.
Another (IV-D-75) said EPA should keep jobs and protect health. Others
(IV-D-210, (IV-D-473, IV-D-168) said that people want to keep ASARCO in
operation and at the same time keep the environment clean. One commenter
(IV-F-5.1) said the public hearings gave residents the false impression
they had to choose between jobs and health. Some commenters (IV-D-125,
IV-D-514, IV-D-4.66, IV-D-5.2) said that to balance health and jobs is an
unfai r choice.
Another (IV-F-1.17) accused EPA of setting one part of the community
against the other in asking them to choose between health risks and jobs.
He said the people of Tacoma were being presented a false tradeoff.
Comment:
Some commenters (IV-D-4.61, IV-D-721, IV-D-781, IV-D-783, IV-D-753)
said that the question of potential job loss probably does not exist,
6-53
-------
because light industry will move into the area as the effects of the smelter
diminish. This will more than offset any potential job loss from the
smelter.
Comment;
One commenter (IV-U-144) said that balancing health risks and jobs is
an appropriate tradeoff only when the people who bear the risks are the same
as those who stand to receive the benefits. Another commenter (IV-D-74)
said it is less dangerous to be unemployed than to be exposed to arsenic,
cadmium, S02 and other hazardous pollutants. Another commenter (IV-D-4.4)
stated that, although you cannot trade lives for dollars, you can compare
the health risks associated with smelter emissions to health risks associated
with unemployment.
Comment:
Several comments (IV-D-339, IV-D-252, IV-F-3.1, IV-F-3.31, IV-F-3.48,
IV-F-4.39) urged that EPA proceed with reason, suggesting compromise between
environmental and economic interests. One (IV-D-359) requested that EPA
allow improvements to be made at a pace that ASARCO can afford. The
commenter cited progress which has been made and is continuing to be made by
ASARCO. One commenter (IV-D-386) said the smelter should be allowed to
continue its operation with the understanding that it continue to improve
control equipment as it becomes technologically and economically available,
especially considering the state of the economy and the copper industry.
One (IV-D-731) suggested as a general regulatory principle that if a company
could prove that it cannot afford controls, it should be required to phase
in such controls over a reasonable period of time, but should not be totally
exempted from regulation.
Some commenters (IV-D-187, IV-D-255, IV-D-274, IV-D-395, IV-D-396,
IV-D-344) urged both improvement of air quality and continued smelter
at further reductions coupled with a decision to retain the ASARCO facility.
6-54
-------
One commenter (IV-D-384) asked that EPA place viable limits on emissions but
that the Agency not kill industry while trying to find what is viable.
One commenter (IV-D-279) said that we must as a nation be concerned and
protective of our natural resources, but excluding the cost of controls,
jobs, and standard of living from rulemaking is just as foolish as ignoring
the consequences of contamination. Others (IV-F-4.2, IV-F-3.78) supported
the use of the best scientific methods for examining the problem and
assessing the risks, but stated that this information must then be weighed
in reference to the economic burden of closing the smelter. Another
commenter (IV-D-388) said EPA's decision must be based on both economics and
the morality of placing others lives in potential danger.
Others, however (IV-F-3.7, IV-F-3.7), said that it is difficult to
equate the costs of abatement with human life, as the health impact is
somewhat immeasurable in terms of dollars.
Response:
The EPA recognizes as the principal objective under section 112 of the
establishment of regulations to protect public health. The EPA does not
interpret this objective as a requirement that risks must be totally
eliminated. It is EPA's view that the intent of section 112 is to insure
that health risks from significant sources of hazardous air pollutants are
reduced to the maximum extent practicable.
The EPA's regulatory analysis includes evaluation of all major impacts
of selected control alternatives, focusing on health impacts but including
consideration of energy, environmental, and economic impacts. In its arsenic
rulemaking, EPA has sought to reduce health risks to an acceptable level
while minimizing adverse economic impacts.
6-55
-------
6.5 Other
This section contains comments which could not be classified under any
of the major categories presented in this document.
Comment:
One commenter (IV-F-1.18) requested that EPA create a mechanism by
which funds would be set aside, so when bad years come for cyclical
industries they cannot claim that they cannot afford pollution control
requirements that are needed to protect public health.
When asked what legal authority EPA could use to establish such a
mechanism, the commenter agreed to give some thought to the question and
submit his suggestions to EPA.
Response;
The EPA can find no basis under the Clean Air Act or other legal ground
which would permit them to establish such a mechanism.
Comment;
One commenter (IV-D-28) said that we desperately need a legal
definition of what constitutes arsenic-induced cancer.
Response;
It would simplify liability claims procedures and regulatory
development if a definition of arsenic-induced cancer could be developed.
However, there is not currently, and is not likely to be in the future, any
way to precisely define arsenic-induced cancer. Lung cancer can be caused
by smoking, genetic predisposition, or exposure to numerious environmental
pollutants. Therefore it is extremely difficult to establish that a
particular case of cancer was caused by arsenic exposure, even though
epidemiologic studies show that, statistically, increased arsenic exposure
leads to increased lung cancer risk.
6-56
-------
Comment:
One commenter (IV-F-3.38) conpared statistical deaths to a lottery
where, for instance, there is one death out of 400,000 people. One unlucky
person will lose this lottery, while 399,999 people will win it. He added
that no one knows who that unlucky person will be. Two commenters (IV-D-56,
IV-D-784) objected to EPA's dealing with human life in statistical terms.
Another (IV-D-144) felt that if victims could be identified it would be
more likely that standards would be set.
Response:
The EPA must estimate risk using a mathematical model because, for
reasons discussed in the above response and section 2.2, it is not possible
to measure risk directly or to predict if a specific individual will contract
arsenic-induced lung cancer. Statistical risk is the only estimate of risk
available for use in setting standards. The EPA realizes that the risk
model is not precise enough for use in predicting the actual number of
deaths which may occur in Tacoma as a result of arsenic exposure.
Comment:
One commenter (IV-F-4.43) suggested regulatory options that may be used
to reduce the health risk from the arsenic emissions associated with the
ASARCO-Tacoma smelter. He lists these options in order of preference:
(i) The zero risk option.
(ii) Impose no greater risk than imposed on 14 other copper
smelting communities.
(iii) Restrict arsenic emissions to a level no greater than that
emitted by the 14 other copper smelters and apply.the
standard at all times.
6-57
-------
(iv) Require state-of-the-art technology at ASARCO and consider
costs to the community when developing best available
technology.
Response:
These approaches were considered by EPA in regulating arsenic emissions.
The zero risk approach is considered impractical because if this were
applied to all hazardous pollutants the result would be wide-scale economic
disruption. The zero risk approach is discussed in section 6.1.4. The
commenter's second and third alternatives were also rejected as a regulatory
policy. The reasons are contained in section 6.1.5 on comparative risks.
The commenters last suggestion is similar to the BAT approach, which is
discussed in section 6.2. As explained in that section, EPA's regulatory
approach is to evaluate each control option in terms of health risk
reduction and residual risk, as well as technical feasibility and economic
impact. Considering these factors, the Administrator will select a
control option. [The ASARCO-Tacoma plant has ceased copper smelting operations
The EPA has-promulgated standards for the arsenic plant which may remain in
operation.]
Comment;
One commenter (IV-D-710) said that an ambient air quality standard for
a carcinogen would be inappropriate as a public health policy matter and
unauthorized under the Clean Air Act. Others (IV-D-621-5, IV-D-621-15.5,
IV-D-621-15.1) commented that EPA lacks authority under § 112 of the Clean
Air Act to adopt an ambient standard for a hazardous air pollutant. One
(IV-D-621-15.5) thought it an improper technique for reducing fugitive
emissions. Commenter IV-F-3.5 said that setting an ambient standard for a
carcinogen is tricky, since there will always be a health risk with any
non-zero exposure. Others (IV-F-3.8, IV-D-708a) wanted to know how EPA
would propose to set an acceptable ambient air standard in the absence of
6-58
-------
established medical criteria for setting risk levels. He said that the
Clean Air Act does not provide for adoption of an ambient air standard for
arsenic in the circumstances which now exist.
Another commenter (IV-D-24) said that EPA should base its arsenic
emission standards on ambient concentration. He said it was his under-
standing that EPA's position is one of not being able to set ambient-air
quality standards since there is no safe level, therefore no margin for
protection of public health. He and another commenter (IV-D-708a) wanted
EPA to set "action levels".
Response:
Since an enforceable ambient standard is not being established in
the copper smelter standard, the comment that section 112 of the Clean
Air Act does not give EPA the authority to set enforceable ambient
standards is not pertinent to this rulemaking. The EPA agrees that an
ambient standard cannot be established for inorganic arsenic based solely
on health effects or risk estimates. The EPA does believe, however, that
an enforceable ambient limit, which is an indicator of proper operation
and maintenance of emission control systems is consistent with the goal
of section 112 and may consider establishing such a limit at a later
date.
Comment:
One commenter (IV-F-4.59) felt that, rather than attesting to control
the methodology for emission reduction, EPA should just set an emission
level and let industry decide how it wants to achieve that level
economically and technologically.
Response:
In response to the first commenter, EPA's equipment standard is not
intended to preclude the use of other secondary inorganic arsenic capture
systems which may be as effective as an air curtain secondary hood. As
specified in the Federal Register (48 FR 33134, July 20, 1983):
6-59
-------
"Upon written application to EPA, the use of an alternative
secondary inorganic arsenic capture system which has been
demonstrated to EPA's satisfaction to be equivalent in terms
of capture efficiency for inorganic arsenic systems may be
approved."
Therefore, industry may decide what methods it wants to use to comply with
the emission standard.
Comment:
One commenter (IV-F-1.13) said that funds available for environmental
control measures are not unlimited. For this reason he said it is important
to put the available money where it gives the maximum benefit. The
commenter said that the most benefit appears to be gained by reducing the
overall pollutant emissions and by improving conditions in plants rather
than by spending money to bring down an already very low level (0.05 ug/rn-^)
of a single element (arsenic). Another (IV-D-616) emphasized that funds
were not unlimited and stressed the need to put available money where it can
do the most good. With this in mind, he saw little support for imposing
strict regulations on low levels of a single corrpound based on lung cancer
found in groups of people exposed to high levels of a large number of
compounds.
Response:
The EPA realizes priorities must be set in controlling various pollutants
in order to utilize the available resources efficiently. However, it is
unclear how the first commenter derived the arsenic level of 0.05 ug/m^.
Furthermore, it is not the concentration of a pollutant per se, but the
health risks associated with ambient concentrations which must be used in
deciding whether to regulate a pollutant.
Comment:
Two commenters (IV-D-301, IV-D-778) said that because U. S. citizens
are entitled under the 1st Amendment of the U.S. Constitution to life
6-60
-------
Itself, they are all entitled to a clean, safe environment. Others (IV-F-3.103,
IV-D-732, IV-D-779) felt that it is the duty of government in our society
to protect the rights of individuals from those who would infringe upon
them, and the right to a healthy environment is fundamental and inalienable.
Another (IV-D-598) said the laws, guidelines, and standards of EPA must
comply with the principles set forth in the Constitution. The commenter
saw only one remedy: elimination of the hazard immediately. The commenter
found no other remedy suggested in the Constitution.
Response:
The EPA must rely on the directives of the Clean Air Act in setting
standards for air pollutants. The EPA has followed the provisions of
section 112 of the Act in setting arsenic standards.
Comment:
One commenter (IV-F-4.68/IV-F-4.71) said that 40 CFR (along with the
Geneva and Nuremburg Codes) states that human subjects must be protected
from involuntary or uninformed exposure. She further stated that if EPA
does not set stringent standards, the Agency's failure to act could be
construed as sanctioning an epidemiological project in which people are
exposed involuntarily to uncertain risks. Such projects, the commenter
said, would be a violation of these codes. In a situation such as this,
true voluntary consent is not possible because people are afraid of the
economic consequences of strict standards.
One commenter (IV-F-4.28) said that EPA was using the people of Tacoma
as human guinea pigs; yet they have not received informed consent from the
residents, nor have they guaranteed that the health care needs of people who
are exposed to arsenic will be taken care of. Another (IV-F-3.58) stated
that citizens living near polluters should not bear the burden of proving
poisons are harmful by being used as unwilling research subjects. Another
commenter (IV-D-661) said that EPA cannot permit continued emissions in a
6-61
-------
large populous area where many do not consent and many more are not even
aware of the risk.
Response:
The EPA is not using the citizens of Tacoma to conduct a research
experiment. Citizens were informed of the potential risk and asked for
their opinions on whether any further regulations should be applied. The
EPA had considered these comments before ASARCO announced the shutdown of
the copper smelter operations.
Comment:
One commenter (IV-D-51) said that the cost-benefit criterion where
human life is at stake is of doubtful morality. Another (IV-D-67) said that
placing the environment and jobs on a balancing scale is neither morally nor
rationally defensible. Another commenter (IV-D-241) felt that someone up at
the "top" playing God is saying, "We can afford to let between 10 and 150
people die so that industry can operate at a larger profit." This kind of
attitude he found insensitive.
Response;
Section 112 of the Clean Air Act gives the Agency the authority to
impose controls on hazardous air pollutant emissions. The Administrator
believes that section 112 decisions that may or may not require further
emissions controls must consider not only the potential health effects
associated with such hazardous air pollutant emissions, but also the
costs and other impacts (e.g., loss of jobs) on society. Admittedly,
they may be at times difficult but the Clean Air Act requires that these
decisions be made.
Comment:
Two commenters (IV-F-4.4, IV-D-708a) said that if the smelter closes,
foreign smelters will get their business. This would mean, in effect, that
6-62
-------
we would sinply be transferring our health risks to another country which
might have little or no pollution control, and he questioned the morality
of this action.
Response:
Due to the lack of information, EPA did not consider the pollution
impacts on foreign populations if the U.S. smelters close down as a result
of the promulgation of section 112 air emissions standards.
6-63
-------
7.0 QUALITY OF LIFE
Comment:
Several commenters urged EPA to consider the effects of emissions from
ASARCO on the quality of life in the Tacoma area. One commenter
(IV-D-164/IV-D-666) said that the public was entitled to "peace of mind,"
knowing that their families were located in safe places to live. The
commenter complained that protecting quality of life did not seem to be a
part of protecting public health. Others (IV-D-375, IV-D-732, IV-D-751,
IV-F-11) said that their quality of life has been altered by the smelter.
Still another (IV-F-3.7) said that health and welfare was a paramount
concern, as well as the ability of people to enjoy life productively.
Another (IV-D-639) said that EPA should put quality of life as its top
priority. Another (IV-D-524) urged EPA to consider the effect of the
emissions on the environment and its being a pleasant, comfortable, and
safe place to live.
Response;
The EPA is aware that quality of life is an important concern. Some
sections of the Clean Air Act, such as sections 108 and 109, incorporate
quality of life considerations. Under these sections primary national
ambient air quality standards for criteria pollutants are established to
protect health, and secondary standards to protect "welfare." However,
section 112 of the Act deals with hazardous air pollutants and is a
completely health-based section. Welfare, or quality of life, is not
mentioned. Hazardous pollutants are defined in section 112(a)(l) as those
which may cause ". . .an increase in mortality or an increase in serious
irreversible, or incapacitating reversible, illness." In section 112
(b)(l)(B) EPA is authorized to set standards which provide ". . .an ample
margin of safety to protect public health for such hazardous air pollutant."
7-1
-------
The Administrator realizes that reduction in inorganic arsenic emissions
may have other beneficial effects on the "quality of life" for the community.
On the other hand, commenters have also reminded EPA that, in general,
plant closures will have some potentially severe adverse effects on the
"quality of life" for the community, such as unemployment and loss of tax
revenues. In making his decision, the Administrator considered these
public comments and was mindful that the selection of the control option
would have other effects, both positive and negative, on the surrounding
communities.
';
7-2
-------
8.0 VICTIM COMPENSATION
Comment:
Several commenters felt that ASARCO and/or EPA should assume some
liability for their impact on the community. Commenters (IV-D-621-12.2,
IV-F-3.43, IV-F-4.28, IV-D-28, IV-D-721, IV-F-3.103) said ASARCO should be
required to post a bond, establish a health fund, or buy insurance to cover
future claims against the company and compensate victims. One commenter
(IV-D-520) said ASARCO should periodically replace residents' topsoil and
be accountable for financial loss and health risks incurred from their
emissions. Two commenters (IV-D-520, IV-F-4.66) said costs of victim
compensation should be included in the economic assessment. One commenter
(IV-F-5.16) said ASARCO should pay for health testing and monitoring of
residents and their employees. Three commenters (IV-F-5.16, IV-F-3.46) said
ASARCO or the government should buy people's homes near the smelter for
fair market value. Three commenters (IV-F-4.28, IV-F-4.43, IV-D-710, IV-F-11)
said EPA should give aid to ASARCO in bearing the burden of compensation,
relocation of potential victims, or adjustment assistance for displaced
employees and the city. Another (IV-D-530) said if jobs were lost, money
should be spent on retraining and placing v/orkers. He said workers should
not have to bear the burden of ASARCO's polluting effects.
On the other hand, two commenters (IV-0-481, IV-0-28) expressed concern
that some people would misuse victims' rights in the hope af getting large
settlements to which they are not entitled.
Two commenters (IV-D-28, IV-D-719) said the problem of determining who
the victims are and if or how they should be compensated has not been
addressed very well. The one commenter (IV-D-28) presented positive and
negative factors associated with a private insurance approach for conpensating
victims. He favored this approach over a Federal program for compensating
victims.
8-1
-------
Response:
The Clean Air Act has no provision that allows EPA to compensate
victims or to require ASARCO to set up a program to do this. However,
section 304 of the Act does provide for citizen suits against emission
sources, governmental agencies, and the EPA Administrator. As detailed in
section 304, suits can be filed if sources are in violation of emissions
standards or permits or if the Administrator fails to perform his duty under
the Clean Ai r Act.
8-2
-------
8.0 VICTIM COMPENSATION
Comment:
Several commenters felt that ASARCO and/or EPA should assume some
liability for their impact on the community. Commenters (IV-D-621-12.2,
IV-F-3.43, IV-F-4.28, IV-D-28, IV-D-721, IV-F-3.103) said ASARCO should be
required to post a bond, establish a health fund, or buy insurance to cover
future claims against the company and compensate victims. One commenter
(IV-D-520) said ASARCO should periodically replace residents' topsoil and
be accountable for financial loss and health risks incurred from their
emissions. Two commenters (IV-D-520, IV-F-4.66) said costs of victim
compensation should be included in the economic assessment. One commenter
(IV-F-5.16) said ASARCO should pay for health testing and monitoring of
residents and their employees. Three commenters (IV-F-5.16, IV-F-3.46) said
ASARCO or the government should buy people's homes near the smelter for
fair market value. Three commenters (IV-F-4.28, IV-F-4.43, IV-D-710, IV-F-11)
said EPA should give aid to ASARCO in bearing the burden of compensation,
relocation of potential victims, or adjustment assistance for displaced
employees and the city. Another (IV-D-530) said if jobs were lost, money
should be spent on retraining and placing workers. He said workers should
not have to bear the burden of ASARCO's polluting effects.
On the other hand, two commenters (IV-D-481, IV-D-28) expressed concern
that some people would misuse victims' rights in the hope of getting large
settlements to which they are not entitled.
Two commenters (IV-D-28, IV-D-719) said the problem of determining who
the victims are and if or how they should be compensated has not been
addressed very well. The one commenter (IV-D-28) presented positive and
negative factors associated with a private insurance approach for compensating
victims. He favored this approach over a Federal program for compensating
victims.
8-1
-------
Response:
(;';.,
The Clean Air Act has no provision that allows EPA to compensate
victims or to require ASARCO to set up a program to do this. However,
section 304 of the Act does provide for citizen suits against emission
sources, governmental agencies, and the EPA Administrator. As detailed in
section 304, suits can be filed if sources are in violation of emissions
standards or permits or if the Administrator fails to perform his duty under
the Clean Air Act.
8-2
-------
-------
5 CO
^ nT
f?
f
-------
&EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/5-85-002
April 1985
Air
Inorganic Arsenic
Risk Assessment
For Primary and
Secondary Lead
Smelters, Primary
Zinc Smelters, Zinc
Oxide Plants,
Cotton Gins, and
Arsenic Chemical
Plants
-------
-------
NOTICE
Thisdocument has not been formally released by EPAand should not now be construed to represent Agency
policy. It is being circulated for comment on its technical accuracy and policy implications.
EPA-450/5-85-002
Inorganic Arsenic Risk Assessment for
Primary and Secondary Lead Smelters,
Primary Zinc Smelters, Zinc Oxide Plants,
Cotton Gins, and Arsenic Chemical Plants
Strategies and Air Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
April 1985
-------
This report has been reviewed by the Strategies and Air Standards Division of the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names orcommercial products
is not intended to constitute endorsement or recommendation for use. Copies of this report are available
through the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle
Park, N.C. 27711, or from National Technical Information Services, 5285 Port Royal Road, Springfield
Virginia 22161.
-------
TABLE OF CONTENTS
Title Page
1 INTRODUCTION 1
1.1 Overview 1
1.2 The Relationship of Exposure to Cancer Risk 1
1.3 Public Exposure 4
1.4 Public Cancer Risks 5
2 THE UNIT RISK ESTIMATE FOR INORGANIC ARSENIC 6
2.1 The Linear No-Threshold Model for Estimation of
Unit Risk Based on Human Data (General) 6
2.2 Unit Risk Estimates Derived from Epidemiologic Studies . 9
3 QUANTITATIVE EXPRESSIONS OF PUBLIC EXPOSURE TO INORGANIC ARSENIC
EMISSIONS 13
3.1 EPA's Human Exposure Model (HEM) (General) . . 13
3.1.1 Pollutant Concentrations Near A Source 13
3.1.2 Expansion of Analysis Area ..... 14
3.2 Methodology for Reviewing Pollutant Concentrations . * . 15
3.2.1 Use of Ambient Data 19
3.2.2 The People Living Near A Source 19
3.2.3 Exposure 20
3.3 ASARCO-East Helena 22
3.3.1 Public Exposure to Inorganic Arsenic Emissions from
Primary Lead Smelters ...........24
3.3.1.1 Source Data 24
3.3.1.2 Exposure Data 24
3.4 Murph Metals-Dallas and Quemetco-Seattle 29
3.4.1 Public Exposure to Inorganic Arsenic Emissions from
Secondary Lead Smelters 32
3.4.1.1 Source Data 32
3.4.1.2 Exposure Data 32
3.5 Public Exposure to Inorganic Arsenic Emissions from
Primary Zinc Smelters ......... 39
3.5.1 Source Data 39
3.5.2 Exposure Data 39
3.6 Public Exposure to Inorganic Arsenic Emissions from
Zinc Oxide Plants 45
3.6.1 Source Data 45
3.6.2 Exposure Data 45
3.7 Methodology for Reviewing Pollutant Concentrations -
Cotton Gins 51
3.7.1 Public Exposure to Inorganic Arsenic Emissions from
Cotton Gins 53
3.7.1.1 Source Data 53
3.7.1.2 Exposure Data 53
iii
-------
Title
3.8 Public Exposure to Inorganic Arsenic Emissions from
Arsenic Plants 68
3.8.1 Source Data 68
3.8.2 Exposure Data 68
4 QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM INORGANIC
ARSENIC EMISSIONS 74
4.1 Methodology (General) 74
4.1.1 The Two Basic Types of Risk 74
4.1.2 The Calculation of Aggregate Risk 74
4.1.3 The Calculation of Individual Risk 76
4.2 Risks Calculated for Emissions of Inorganic Arsenic ... 76
5 ANALYTICAL UNCERTAINTIES APPLICABLE TO THE CALCULATION OF PUBLIC
HEALTH RISKS CONTAINED IN THIS DOCUMENT 85
5.2 Public Exposure 86
5.2.1 General 86
5.2.2 The Public 87
5.2.3 The Ambient Air Concentrations 88
6 REFERENCES 90
IV
-------
LIST OF TABLES
Table
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Page
Summary of Quantitative Risk Analyses 10
Combined Unit Risk Estimates for Absolute Risk Linear Models. . 12
Arsenic Concentrations Near ASARCO-East Helena Primary
Lead Smelter 1(5
Identification of Primary Lead Smelters 25
Input Data to Exposure Model Primary Lead Smelting Industry
(Assuming Baseline Controls) 26
Total Exposure and Number of People Exposed Primary Lead
Smelting Industry 27
Public Exposure for Primary Lead Smelting Industry as
Produced by the Human Exposure Model (Assuming Baseline
Controls) 28
Arsenic Concentrations Near Select Secondary Lead Smelters . . 35
Identification of Secondary Lead Smelters 34
Secondary Lead Industry Inputs to HEM Model
(Assuming Baseline Controls) 35
Total Exposure and Number of People Exposed Secondary Lead
Smelting Industry
37
Public Exposure for Secondary Lead Smelters as Produced
by the Human Exposure Model (Assuming Baseline Controls) ... 38
Arsenic Concentrations Near Select Primary Zinc Smelters ... 40
Identification of Primary Zinc Smelters . . 41
Input Data to Exposure Model Primary Zinc Smelting Industry
(Assuming Baseline Controls) 42
Total Exposure and Number of People Exposed Primary
Zinc Smelter ....
43
Public Exposure for Primary Zinc Smelters as Produced by the
Human Exposure Model (Assuming Baseline Controls ) ...... 44
Arsenic Concentrations Near Select Zinc Oxide Plants
Identification of Zinc Oxide Plants
v
46
-------
Table Page
20 Input Data to Exposure Model Zinc Oxide Plants
(Assuming Baseline Controls) . 48
21 Total Exposure and Number of People Exposed (Zinc Oxide
Plants) 49
22 Public Exposure for Zinc Oxide Plants as Produced by the
Human Exposure Model (Assuming Baseline Controls) 50
23 Arsenic Concentrations Near Two Texas Cotton Gins 52
24 Identification of Model Cotton Gins 54
25 Input Data to Exposure Model Cotton Gins (Assuming Baseline
Controls) 55
26 Public Exposure for 4 Bales/Hour Model Cotton Gin (Hutto,TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 56
27 Public Exposure for 7 Bales/Hour Model Cotton Gin (Hutto.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 57
28 Public Exposure for 12 Bales/Hour Model Cotton Gin (Hutto,TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 58
29 Public Exposure for 20 Bales/Hour Model Cotton Gin (Hutto,TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 59
30 Public Exposure for 4 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 60
31 Public Exposure for 7 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 61
32 Public Exposure for 12 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 62
33 Public Exposure for 20 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 63
vi
-------
34
35
36
37
38
39
40
41
.42
43
44
45
46
47
48
49
Public Exposure for 4 Bales/Hour Model Cotton Gin (Itasca.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 64
Public Exposure for 7 Bales/Hour Model Cotton Gin (Itasca,TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) . 65
Public Exposure for 12 Bales/Hour Model Cotton Gin (Itasca,TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 66
Public Exposure for 20 Bales/Hour Model Cotton Gin (Itasca,TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 67
Arsenic Concentrations Near Select Arsenic Chemical Plants . . 69
Identification of Arsenic Chemical Plants 70
Input Data to Exposure Model Arsenic Chemical Plants
(Assuming Baseline Controls) 71
Total Exposure and Number of People Exposed (Arsenic
Chemical Plants) 72
Public Exposure for Arsenic Chemical Plants as Produced by
the Human Exposure Model (Assuming Baseline Controls) . ,
73
Maximum Lifetime Risk and Cancer Incidence for Primary Lead
Smelters (Assuming Baseline Controls) 78
Maximum Lifetime Risk and Cancer Incidence for Secondary
Lead Smelters (Assuming Baseline Controls) 79
Maximum Lifetime Risk and Cancer Incidence for Primary
Zinc Smelters (Assuming Baseline Controls) ,
80
Maximum Lifetime Risk and Cancer Incidence for Zinc Oxide
Plants (Assuming Baseline Controls) 81
Maximum Lifetime Risk and Cancer Incidence for Model Cotton
Gins (Assuming Baseline Controls) .... 82
Lifetime Risk for Two Texas Cotton Gins (Assuming Baseline
Controls) 83
Maximum Lifetime Risk and Cancer Incidence for Arsenic
Chemical Plants (Assuming Baseline Controls) ........ 84
vii
-------
LIST OF FIGURES
Figure
1
2
3
4
Page
Group 2 BG/ED Interpolation ...... 17
Predicted Versus Measured Inorganic Arsenic Ambient
Concentrations (ASARCO-East Helena, MT) ..... 23
Predicted Versus Measured Inorganic Arsenic Ambient
Concentrations (Murph Metals-Dallas, TX)
30
Predicted Versus Measured Inorganic Arsenic Ambient
Concentrations (Quemetco, Seattle, WA) .... 31
vm
-------
INORGANIC ARSENIC RISK ASSESSMENT FOR PRIMARY AND SECONDARY LEAD SMELTERS,
PRIMARY ZINC SMELTERS AND ZINC OXIDE PLANTS, COTTON GINS AND ARSENIC CHB1ICAL
PLANTS
1 INTRODUCTION
1.1 Overview
The quantitative expressions of public cancer risks presented in this
document are based on ( 1) a dose-response model that numerically relates
the degree of exposure to airborne inorganic arsenic to the risk of getting
lung cancer, and (2) numerical expressions of public exposure to ambient
air concentrations of inorganic arsenic estimated to be caused by emissions
from stationary sources. Each of these factors is discussed briefly below
and details are provided in the following sections of this document.
1.2 The Relationship of Exposure to Cancer Risk
The relationship of exposure to the risk of contracting lung cancer is
derived from epidemiological studies in occupational settings rather than
from studies of excess cancer incidence among the public. The epidemiological
methods that have successfully revealed associations between occupational
exposure and cancer for substances such as asbestos, benzene, vinyl chloride,
and ionizing radiation, as well as for inorganic arsenic, are not readily
applied to the public sector, with its increased number of confounding
variables, much more diverse and mobile exposed population, lack of consoli-
dated medical records, and almost total absence of historical exposure
data. Given such uncertainties, EPA considers it improbable that any
association, short of very large increases in cancer, can be verified in
the general population with any reasonable certainty by an epidemiological
study. Furthermore, as noted by the National Academy of Sciences (NAS)l,
"...when there is exposure to a material, we are not starting at an origin
-------
of zero cancers. Nor are we starting at an origin of zero carcinogenic
agents in our environment. Thus, it is likely that any carcinogenic agent
added to the environment will act by a particular mechanism on a particular
cell population that is already being acted on by the same mechanism to
induce cancers." In discussing experimental dose-response curves, the NAS
observed that most information on carcinogenesis is derived from studies of
ionizing radiation with experimental animals and with humans which indicate
a linear no-threshold dose-response relationship at low doses. They added
that although some evidence exists for thresholds in some animal tissues,
by and large, thresholds have not been established for most tissues. NAS
concluded that establishing such low-dose thresholds "...would require
massive, expensive, and impractical experiments ..." and recognized that
the U.S. population "...is a large, diverse, and genetically heterogeneous
group exposed to a large variety of toxic agents." This fact, coupled with
the known genetic variability to carcinogenesis and the predisposition of
some individuals to some form of cancer, makes it extremely difficult, if
not impossible, to identify a threshold.
For these reasons, EPA has taken the position, shared by other Federal
regulatory agencies, that in the absence of sound scientific evidence to
the contrary, carcinogens should be considered to pose some cancer risk
at any exposure level. This no-threshold presumption is based on the view
that as little as one molecule of a carcinogenic substance may be sufficient
to transform a normal cell into a cancer cell. Evidence is available from
both the human and animal health literature that cancers may arise from a
single transformed cell. Mutation research with ionizing radiation in cell
cultures indicates that such a transformation can occur as the result of
interaction with as little as a single cluster of ion pairs. In reviewing
the available data regarding carcinogenicity, EPA found no compelling
scientific reason to abandon the no-threshold presumption for inorganic
arsenic.
-------
In developing the exposure-risk relationship for inorganic arsenic, EPA
has assumed that a linear no-threshold relationship exists at and below the
levels of exposure reported in the epidemic!ogical studies of occupational
exposure. This means that any exposure to inorganic arsenic is assumed
to pose some risk of lung cancer and that the linear relationship between
cancer risks and levels of public exposure is the same as that between cancer
risks and levels of occupational exposure. EPA believes that this assumption
is reasonable for public health protection in light of presently available
information. However, it should be recognized that the case for the linear
no-threshold dose-response relationship model for inorganic arsenic is not
quite as strong as that for carcinogens which interact directly or in
metabolic form with DNA. Nevertheless, there is no adequate basis for
dismissing the linear no-threshold model for inorganic arsenic. Assuming
that exposure has been accurately quantified, it is the Agency's belief
that the exposure-risk relationship used by EPA at low concentrations
represents only a plausible upper-limit risk estimate in the sense that the
risk is probably not higher than the calculated level and could be much
lower.
The numerical constant that defines the exposure-risk relationship
used by EPA in its analysis of carcinogens is called the unit risk estimate.
The unit risk estimate for an air pollutant is defined as the lifetime cancer
risk occurring in a hypothetical population in which all individuals are
exposed throughout their lifetimes (about 70 years) to an average concentration
of 1 Mg/m3 of the agent in the air which they breathe. Unit risk estimates
are used for two purposes: (1) to compare the carcinogenic potency of several
agents with each other, and (2) to give a crude indication of the public
health risk which might be associated with estimated air exposure to these
agents.
-------
The unit risk estimate for inorganic arsenic that is used in this
appendix was prepared by combining the five different exposure-risk numerical
constants developed from four occupational studies.2 The methodology used
to develop the unit risk estimate from the four studies is described in
Section 2 below.
1. 3 Public Exposure
The unit risk estimate is only one of the factors needed to produce
quantitative expressions of public health risks. Another factor needed
is a numerical expression of public exposure, i.e., the numbers of
people exposed to the various concentrations of inorganic arsenic. The
difficulty of defining public exposure was noted by the National Task
Force on Environmental Cancer and Health and Lung Disease in their 5th
Annual Report to Congress, in 1982. 3 They reported that "...a large
proportion of the American population works some distance away from their
homes and experience different types of pollution in their homes, on the
way to and from work, and in the workplace. Also, the American population
is quite mobile, and many people move every few years." They also noted the
necessity and difficulty of dealing with long-term exposures because of
"...the long latent period required for the development and expression
of neoplasia [cancer]..." The reader should note that the unit risk estimate
has been changed from that value used in the inorganic NESHAP proposal as a
result of EPA's analysis of several occupational epidemic!ogical studies that
have recently been completed.
EPA's numerical expression of public exposure is based on two estimates.
The first is an estimate of the magnitude and location of long-term average
ambient air concentrations of inorganic arsenic in the vicinity of emitting
sources based on dispersion modeling using long-term estimates of source
emissions and meteorological conditions. The second is an estimate of the
number and distribution of people living in the vicinity of emitting sources
based on 1980 Bureau of Census data which "locates" people by population
-------
centroids in census tract areas. The people and concentrations are combined
to produce numerical expressions of public exposure by an approximating
technique contained in a computerized model. The methodology is described
in Section 3 below.
1.4 Public Cancer Risks
By combining numerical expressions of public exposure with the unit
risk estimate, two types of numerical expressions of public cancer risks are
produced. The first, called individual risk, relates to the person or
persons estimated to live in the area of highest concentration as estimated
by the computer model. Individual risk is expressed as "maximum lifetime
risk." As used here, the work "maximum" does not mean the greatest possible
risk of cancer to the public. It is based only on the maximum annual average
exposure estimated by the procedure used. The second, called aggregate risk,
is a summation of all the risks to people estimated to be living within the
vicinity (usually within 50 kilometers) of a source and is customarily summed
f for all the sources in a particular category. The aggregate risk is expressed
as incidences of cancer among all of the exposed population after 70 years of
exposure; for convenience, it is often divided by 70 and expressed as cancer
incidences per year. These calculations are described in more detail in
Section 4 below.
There are also risks of nonfatal cancer and other potential health effects,
depending on which organs receive the exposure. No numerical expressions
of such risks have been developed.
-------
2. THE UNIT RISK ESTIMATE FOR INORGANIC ARSENIC2
The following discussion is summarized from a more detailed description
of the Agency's derivation of the inorganic arsenic unit risk estimate as
found in EPA's "Health Assessment Document for Inorganic Arsenic" (EPA-600/
8-83-021F).
2.1 The Linear No-Threshold Model for Estimation of Unit Risk Based on
Human Data (General)
The methodologies used to arrive at quantitative estimates of risk
must be capable of being implemented using the data available in existing
epfdemiologic studies of exposure to airborne arsenic. This requires
extrapolation from the exposure levels and temporal exposure patterns in
these studies to those for which risk estimates are required. It is assumed
that the age-specific mortality rate of respiratory cancer per year per
100,000 persons for a particular 5-year age interval, i, can be
represented using the following linear absolute or additive risk model:
a-j(D) = ai + 100,000 a'D
U)
With this model, a-,- is the age-specific mortality rate per year of
respiratory cancer in a control population not exposed to arsenic, a' is
a parameter representing the potential of airborne arsenic to cause
respiratory cancer, and D is some measure of the exposure to arsenic up
to the ith age interval. For example, D might be the cumulative dose
in years-ug/m3, the cumulative dose neglecting exposure during the last
10 years prior to the ith age interval, or the average dose in ug/m^
over some time period prior to the ith age interval. The forms to be used
for D are constrained by the manner in which dose was treated in each
individual epidemic!ogic study. At low exposures the extra lifetime
probability of respiratory cancer mortality will very correspondingly
( e.g. , linearly).
-------
The dose-response data available in the epidemiologic studies for esti-
mating the parameters in these models consists primarily of a dose measure
Dj for the jth exposure group, the person-years of observation Yj, the observed
number of respiratory cancer deaths Oj, and the number Ej of these deaths
expected in a control population with the same sex and age distribution as
the exposure group. The expected number Ej is calculated as
= s
i
i7100,000
(2}
here YJJ is the number of person-years of observation in the ith age cate-
gory and the jth exposure group (Yj = Z Y j,). This is actually a simplified
i
representation, because the calculation also takes account of the change in
the age-specific incidence rates with absolute time. The expected number
of respiratory cancer deaths for the ith exposure group is
E(0j) = 2 YJ-J (a-j + 100,OOOa'Dj)/100sOOO
i
EJ +a'YJDJ
(3)
under the linear absolute risk model. Consequently, E(0j) can be expressed
in terms of quantities typically available from the published epidemiologic
studies.
Making the reasonable assumption that Oj has a Poisson distribution,
the parameter a' can be estimated from the above equation using the method
of maximum likelihood. Once this parameter is estimated, the age-specific
mortality rates for respiratory cancer can be estimated for any desired ex-
posure pattern.
-------
To estimate the corresponding additional lifetime probability of res-
piratory cancer mortality, let bi,...,bis be the mortality rates, in the
absence of exposure, for all cases per year per 100,000 persons for the age
intervals 0-4, 5-9,..., 80-84, and 85+, respectively; let ai,...,a]_8 represent
the corresponding rates for malignant neoplasms of the respiratory system.
The probability of survival to the beginning of the ith 5-year age interval
is estimated as
n [1 - 5bj 7100,000]
(4)
Given survival to the beginning of age interval i, the probability of dying
of respiratory cancer during this 5-year interval is estimated as
531/100,000
(5)
The probability of dying of respiratory cancer given survival to age
85 is estimated as ais/big. Therefore, the probability of dying of respir-
atory cancer in the absence of exposure to arsenic can be estimated as:
17 i-1
P0 = £ [5a-j 7100,000) n (l-5bj/100,000)3
17
- 5bj/100,000)
Here the mortality rates a-,- apply to the target population for which risk
estimates are desired, and consequently will be different from those in
-------
(l)-(5), which applied to the epidemiologic study cohort. If the 1976 U.S.
mortality rates I male, female, white, and non-white combined) are used in
this expression, then PQ = 0.0451.
To estimate the probability PEP of respiratory cancer mortality when
exposed to a particular exposure pattern EP, the formula (6) is again used,
but a-,- and b, are replaced by a-j(D-j) and b-j(Dj), where 0-,- is the exposure
measure calculated for the ith age interval from the exposure pattern EP.
For example, if the dose measure used in (1) is cumulative dose to the be-
ginning of the ith age interval in ug/m3-years, and the exposure pattern
EP is a lifetime exposure to a constant level of 10 ug/m3, then 0^ =
(1-1H5H10), where the-5 accounts for the fact that each age interval has
a width of 5 years. The additional risk of respiratory cancer mortality is
estimated as
PEP - PO
(7)
If the exposure pattern EP is constant exposure to 1 ug/m , then PEP - P0 is
called the "unit risk."
This approach can easily be modified to estimate the extra probability of
respiratory cancer mortality by a particular age due to any specified
exposure pattern.
2.2 Unit Risk Estimates Derived from Epidemiologic Studies
Prospective studies of the relationship between mortality and exposure
to airborne arsenic have been conducted for the Anaconda, Montana smelter
and the Tacoma, Washington smelter. Table 1 summarizes the fit of the
absolute linear model to dose-response data from 4 different studies at the
two smelters. (See the "Health Assessment Document for Inorganic Arsenic",
-------
1-
h- c=
r-
U.
,.* ^
0 3
1 r
in ro
in >
a °-
o
o
o T
in .
P -a
3 *"
in CM
0) X
a
o
Si
a>
o
1± ^
10 <^
o ra
in «
O 3
ex a.
X 0
UI Q-
ja
r** *.»*
~ 2
in in
"E T!
a* a>
r r
O 0 '
ex +J
in x a)
^a in
a> >> a> c
A
o-j: o>
O J= ^ >r-
_J ^^ O 3Z
t.
c e o
«C in ^
PO
i
tn
CM
<
5
o
f-
o
r-
in
"C
' 01
O
in
10
j2
o
1
03
ro"
i
t-i
CO
U3
cti
0
un
tn
in
"j~
5
3
"o
in
a
in
i.
y
oO___
* p^ x^
00 ll °
CO
1
5
10
I
«*
CJ
.
o
r-^
in
"J^
ai
o
in
J3
i
5
L.
U
y
J
U
-
~
1
i
J
i
5
^
5
>
3
3
X
in
i
t- CO
O 1
o
i
r- X
O
> CO
i c in
i- ai c
r- in (O
t- ai
>a e
ico
C E CO
i en t
- 1 O
: -n c\j
-------
11
Chapter 7, EPA-600/8-83-021F for detailed description of occupational studies.)
Table 1 also displays the carcinogenic potencies a'. It should be noted
that the potencies estimated from different models are in different units,
and are therefore not comparable.
The estimated unit risk is presented for each fit for which the chi-
square goodness-of-fit p-value is greater than 0.01. The unit risks derived
from linear models8 in allrange from 0.0013 to 0.0136. The largest of
these is from the Ott et al. study, which probably is the least reliable
for developing quantitative estimates, and which also involved exposures to
pentavalent arsenic, whereas the other studies involved trivalent arsenic.
The unit risks derived from the linear absolute-risk models are considered
to be the most reliable; although derived from 5 sets of data involving 4
sets of investigators and 2 distinct exposed populations, these estimates
are quite consistent, ranging from 0.0013 to 0.0076.
To establish a single point estimate, the geometric mean for data sets
is obtained within distinct exposed populations, and the final estimate is
taken to be the geometric mean of those values. This process is illustrated
in Table 2.
-------
12
Table 2
Combined Unit Risk Estimates for Absolute-Risk Linear Models
Exposure Source
Anaconda smelter
ASARCO smelter
Study
Brown & Chu
Lee-Feldstein
Higgins et al.
Enterl ine &
Marsh
Unit Risk
1.25 x 10-3
2.80 x lO-3
4.90 x 10-3
6.81 x 10-3
7.60 x 10-3
Geometric Final
Mean Unit Estimated
Risk Unit Risk
2.56 x 10-3
4.29 x lO-3
7.19 x lO-3
-------
13
3. QUANTITATIVE EXPRESSIONS OF PUBLIC EXPOSURE TO INORGANIC ARSENIC
EMISSIONS
3.1 EPA's Human Exposure Model (HEM) (General)
EPA's Human Exposure Model is a general model capable of producing
quantitative expressions of public exposure to ambient air concentrations
of pollutants emitted from stationary sources. HEM contains (1) an atmospheric
dispersion model, with included meteorological data, and (2) a population
distribution estimate based on Bureau of Census data. The input data needed
to operate this model are source data, e.g., plant location, height of the
emission release point, and volumetric rate of release temperature of the
off-gases. Based on the source data, the model estimates the magnitude and
distribution of ambient air concentrations of the pollutant in the vicinity
of the source. The model is programmed to estimate these concentrations
for a specific set of points within a radial distance of 50 kilometers from
the source. If the user wishes to use a dispersion model other than the
one contained in HEM to estimate ambient air concentrations in the vicinity
of a source, HEM can accept the concentrations if they are put into an
appropriate format.
Based on the radial distance specified, HEM numerically combines the
distributions of pollutant concentrations and people to produce quantitative
expressions of public exposure to the pollutant.
3.1.1 Pollutant Concentrations Near a Source
The HEM dispersion model is a climatological model which is a sector-
averaged gaussian dispersion algorithm that has been simplified to improve
computational efficiency. ^
-------
14
Stability array (STAR) summaries are the principal meteorological input to
the HEM dispersion model. STAR data are standard climatological frequency-
of-occurence summaries formulated for use in EPA models and available for
major U.S. meteorological monitoring sites from the National Climatic Center,
Asheville, N.C. A STAR summary is a joint frequency-of-occurence of wind
speed, atmospheric stability, and wind direction, classified according to
Pasquill's categories. The STAR summaries in HEM usually reflect five years
of meteorological data for each of 314 sites nationwide. The model produces
polar coordinate receptor grid points consisting of 10 downwind distances
located along each of 16 radials which represent wind directions. Concen-
trations are estimated by the dispersion model for each of the 160 receptors
located on this yrid. The radials are separated by 22.5-degree intervals
beginning with 0.0 degrees and proceeding clockwise to 337.5 degrees. The
10 downwind distances for each radial are 0.2, 0.5, 1.0, 2.0, 5.0, 10.0,
20.0, 30.0, 40.0, and 50.0 kilometers. The center of the receptor grid for
each plant is assumed to be the plant center. Concentrations at other
points were calculated by using a log-linear scheme as illustrated in
Figure 1.
3.1.2 Expansion of Analysis Area
At proposal, exposure and risk were estimated for people residing
within 20 kilometers of the smelter. Some commenters pointed out that
since people beyond 20 kilometers are exposed to some level of arsenic due
to a source's emissions, EPA's proposal analysis underestimates the total
exposure and risk. EPA agreed with the commenters and expanded its analysis
out to 50 kilometers. When applying air dispersion models, the EPA's
modeling guidelines recommend that, because of the increasing uncertainty
of estimates with distance from the modeled source and because of the
paucity of validation studies at larger distances, the impact may extend
-------
15
out to 50 kilometers but the analysis should generally be limited to this
distance from the source.4 Such site-specific factors as terrain features
(complex or flat), the objectives of the modeling exercise, and distance to
which the model has been validated will determine the appropriate distance
(whether greater than or less than the guideline distance) for which the
Agency should apply the model.
3.2 Methodology for Reviewing Pollutant Concentrations
Before making HEM computer runs, EPA reviewed small-scale U.S. Geological
Survey topographical maps (scale 1:24000) to verify locational data for each
arsenic source. Plants were given accurate latitude and longitude values which
were then incorporated into the HEM program.
After completing the HEM runs, nearby monitoring sites with ambient
air quality data were identified by a computer search of EPA's National
Aerometric Data Bank (MADE) (Table 3). At some sites, data collected over
several years along with annual averages (based on different numbers of
sample sizes for the years monitored) for each year were available. In
these instances, weighted multi-year averages were calculated to provide an
overall mean for each monitoring site. For purposes of annual mean calculations,
values measured below mimimum detection limits were considered by EPA to be
equal to one-half the detection limit. These ambient arsenic data were
then compared to HEM predicted values in order to gauge the accuracy of the
'air dispersion model's estimates. As noted above, HEM predicted values
were based on concentrations at 160 polar coordinate receptor grid points
consisting of 10 downwind distances located along each of 16 radials which
represented wind directions. Because the actual monitoring site locations
identified in the NADB retrieval usually did not correspond to exact grid
point locations, a log-linear interpolation scheme (Figure 1) was used to
calculate an estimated concentration at the site.
-------
16
Table 3
Arsenic Concentrations Near ASARCO-East Helena
Primary Lead Smelter
Plant # Obs.
ASARCO-East
Helena 27
41
137
25
31
36
81
23
20
Company Data 1460
1460
1460
638
1460
274
Distance1
(km)
.5
.7
.8
.9
1.4
1.5
3.9
4.7
7.2
1.1
1.3
1.3
2.1
6.1
7.2
Bearing
119.6
11.5
20.4
343.9
45.3
156.9
176.5
270.4
273.4
275
5
145
92
275
162
Predicted2
(yg/m3)
0.230
0.078
0.056
0.050
0.076
0.047
0.0159
0.005
0.003
0.024
0.050
0.077
0.071
0.0037
0.0084
Measured3
(Mg/m3)
0.108
0.151
0.242
0.161
0.078
0.109
0.031
0.025
0.030
0.059
0.24
0.078
0.074
0.024
0.028
MDL4 Percent! le5
0.02 30< % <50
0.02 <10
0.02 <10
0.02 <10
0.02 30< % <50
0.02 <10
0.02 70< % <90*
0.02 50< % <70*
0.0055 30< % <50*
__
~
.-
__
__
* Indicates data point was disregarded; see Section 3.2.1
1 Distance from source to monitor (km)
2 Concentration predicted by Human Exposure Model (HEM). See Section 3.1.
3 The measured values are weighted averages. When the sampled arsenic
concentrations were below the HDL, a value of 1/2 MDL was assumed for
purposes of calculating the annual averages
4 Minimum detection limit
5 Percentile indicates percentage of data falling below minimum detectable
levels.
-------
17
Figure 1 Group 2 BG/ED Interpolation
A
Gi ven:
A -
Al -
A2 -
R
Rl -
R2 -
Cl -
C2 -
The angle in radians subtended clockwise about the source from due
south to the BG/ED centroid;
The angle from due south to the radial line immediately counter-
clockwise of A, or passing through A if there is an exact match;
The angle from due south to the radial line immediately clockwise of
Al (A2 is 0 if it is due south); ; -
The distance in km from the source to the BG/ED centroid;
The distance from the source to the largest circular arc of radius
less than R;
The distance from the source to the smallest circular arc of
radius greater than or equal to R;
The natural logarithm of the concentration value at (Al, Rl);
The natural logarithm of the concentration value at (Al, R2);
-------
18
C3 - The natural logarithm of the concentration value at (A2, Rl);
C4 - The natural logarithm of the concentration value at (A2, R2);
then:
RTEMP - ln(R/Rl)/ln(R2/Rl);
ATEMP - (A-A1)/(A2-A1);
CA1 - exp(Cl + (C2-Cl)xRTEMP);
CA2 - exp(C3 + (C4-C3)xRTEMP); and
CX - CA1 + (CA2-CAl)xATEMP,
where CX is the interpolated concentration at the B6/ED centroid.
-------
19
3.2.1 Use of Ambient Data
Certain criteria were considered in review of ambient levels. Mean
concentration values derived from sample sizes of less than 25 data points
were disregarded. When reviewing the available monitoring data, it appeared
that monitors situated at distances greater than 15 km from the arsenic
source were considered too far from the source to gauge air dispersion
results without interference from other arsenic sources. Furthermore, at
distances greater than 15 km from the source, plant impacts were often
predicted to be significantly lower than minimum detection limits. These
data were not incorporated in the analyses. A third consideration in
reviewing ambient data concerned the percentage of monitored data which
fell below minimum detection limits. Although some monitoring sites
registered data with over 90 percent of the values above minimum detection
levels, many had about half the data points or more below such levels.
Instances where more than 50 percent of the data were below MDL were dis-
regarded. It should be noted that the various tables in subsequent sections
display, in addition to company-collected data, all ambient monitoring data
that were collected at sites within 15 kilometers of the source as identified
by EPA's computer search although not all the data were used in the final
analysis.
3.2.2 The People Living Near A Source
To estimate the number and distribution of people residing within 50
kilometers of the source, the HEM model uses the 1980 Master Area Reference
File (MARF) from the U.S. Bureau of Census. This data base consists of
enumeration district/block group (ED/BG) values. MARF contains the population
centroid coordinates (latitude and longitude) and the 1980 population of each
ED/BG (approximately 300,000) in the United States (50 states plus the District
of Columbia). HEM identifies the population around each plant, by using the
-------
20
geographical coordinates of the plant, and identifies, selects, and stores
for later use those ED/BGs with coordinates falling within 50 kilometers of
plant center.
3.2.3 Exposure5
The Human Exposure Model (HEM) uses the estimated ground level
concentrations of a pollutant together with population data to calculate
public exposure. For each of 160 receptors located around a plant, the
concentration of the pollutant and the number of people estimated by the
HEM to be exposed to that particular concentration are identified. The HEM
multiplies these two numbers to produce exposure estimates and sums these
products for each plant.
A two-level scheme has been adopted in order to pair concentrations
and populations prior to the computation of exposure. The two level approach
is used because the concentrations are defined on a radius-azimuth (polar)
grid pattern with non-uniform spacing. At small radii, the grid cells are
usually smaller than ED/861s; at large radii, the grid cells are usually
larger than ED/BG's. The area surrounding the source is divided into two
regions, and each ED/BG is classified by the region in which its centroid
lies. Population exposure is calculated differently for the ED/BG's located
within each region. For ED/BG centroids located between 0.2 and 3.5 km
from the emission source, populations are divided between neighboring
concentration grid points. There are 64 (4 x 16) polar grid points within
this range. Each ED/BG can be paired with one or many concentration points.
The population associated with the ED/BG centroid is then divided among all
concentration grid points assigned to it. The land area within each polar
sector is considered in the apportionment.
-------
21
For population centroids between 3.5 and 50 km from the source, a
concentration grid cell, the area approximating a rectangular shape bounded
by four receptors, is much larger than the area of a typical EO/BG. Since
there is an approximate linear relationship between the logarithm of
concentration and the logarithm of distance for receptors more than 2 km
'from the source, the entire population of the ED/BG is assumed to be exposed
to the concentration that is logarithmically interpolated radially and
arithmetically interpolated azimuthally from the four receptors bounding
the grid cell. Concentration estimates for 96 (6 x 16) grid cell receptors
at 5.0, 10.0, 20.0, 30.0, 40.0, and 50.0 km from the source along each of
16 wind directions are used as reference points for this interpolation.
In summary, two approaches are used to arrive at coincident concentration/
population data points. For the 64 concentration points within 3.5 km of the
source, the pairing occurs at the polar grid points using an apportionment
of ED/BG population by land area. For the remaining portions of the grid,
pairing occurs at the ED/BG centroids themselves through the use of log-log
and linear interpolation. (For a more detailed discussion of the model used
to estimate exposure, see Reference 5.)
-------
22
3.3 ASARCO-East Helena
Predicted (HEM) versus measured data were plotted (Figure 2) and a
least squares weighted linear regression analysis was run based on thirteen
data points (see Table 3). The least squares regression line (solid line)
was determined on the basis of a comparison of National Aerometric Data Bank
monitoring data (circumscribed dots) and ASARCO monitoring data (circumscribed
Xs) with ambient concentrations predicted by the Human Exposure Model.
The reader should note that a perfect fit for the least squares regression
analysis results in a line running through the origin at a 45° angle (dotted
line on Figure 2). This means that if the HEM model predicts the measured
data perfectly, then the data points would fall on the dotted line. In cases
where the HEM model underpredicts concentrations, data points will be located
above the 45° perfect fit line. Likewise, when the HEM model overpredicts
concentrations, data points will be located below the perfect fit line.
The regression line resulting from our comparison of predicted and monitored
data runs nearly parallel to the perfect fit line but intersects the ordinate
axis at a value of approximately 0.05 ug/m^. This result is consistent with
the expectation that air dispersion modeling would underpredict ambient con-
centrations. The air dispersion modeling did not consider other local
sources of arsenic such as naturally-occurring arsenic in the windblown
dust and reentrained arsenic particulate matter that had settled to the
earth from past smelter emissions.
A study to determine source apportionment for particulate lead and total
suspended particulates (TSP) in East Helena was completed in 1982. High
volume TSP, low volume TSP, and dichotomous samplers were co-collected
(same time period and same site) to permit differences in sample collection
mass and chemistry to be understood. Analysis of hi-vol samples was carried
out by the State of Montana and lo-vol and dichotomous samples were analyzed
-------
23
FIGURE 2 Predicted Versus Measured
Inorganic Arsenic Ambient Concentrations
(ASARCO - East Helena, MT)
MODEL
UNDERPREDICTION
MODEL
OVERPREDICTIONl
Perfect Fit
Linear Regression t
EPA Data
Company Data
0.1 0.2 0.3
Predicted Concentration (>ug/m3)
-------
24
by NEA, Inc. In addition to participate lead and TSP, sanples were also
measured in some cases for arsenic.^
At six locations where arsenic concentrations were measured using both
lo-vol and hi-vol samplers, the ratio of lo-vol to hi-vol in percent ranged
from 104 to 133 with a mean of 118%. This loss of arsenic compounds could
have occurred in two areas: (1) the volatilization of the arsenic conpounds
from the hi-vol filter itself during sanpling, and (2) the loss of volatile
arsenic conpounds during digestion and storage of samples prior to analysis.
However, based on the data from the study, EPA concluded that the loss of
arsenic on hi-vol filters was relatively minor in nature and within the over-
all accuracy goal of +_ 15-20% considered adequate for most ambient air quality
measurements.
3.3.1 Public Exposure to Inorganic Arsenic Emissions from Primary Lead
Smelters
3.3.1.1 Source Data
Five primary lead smelters are included in the analysis. Table 4
lists the names and addresses of the plants considered, and Table 5 lists
the plant data used as input to the Human Exposure Model (HEM).
3.3.1.2 Exposure Data
Table 6 lists, on a plant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total exposure
is the sum of the products of number of people times the ambient air concentration
to which they are exposed, as calculated by HEM. Table 7 sums, for the
entire source category (5 plants), the numbers of people exposed to various
ambient concentrations, as calculated by HEM. (Source-by-source exposure
results are provided in the EPA docket numbered A-83-23.)
-------
25
TABLE 4
IDENTIFICATION OF PR MARY LEAD SMELTERS
Plant Number Code
1
2
3
4
5
1
Plant Name and Address
ASARCO East Helena,
ASARCO El Paso, TX
St. Joe Herculaneum
ASARCO Glover, MO
Amax Boss, MO
MT
, MD
-------
c
O 4->
i- c ai
in Q.
ut o >^
I0-1-
UJ
in
C Id
o co
1- Q-
in 4-> t= i£
I/I C 01 O
UJ 0.
t- in
M c: id >,
in o CD -u o
i - +J -r- 0)
w in +-> ?- o in
er in c x o .
» 1 «r- » UI E
E 0 0)
O) til <» ^>
c
4-> * i ^-
* C IdCM
0) O C E
B T- j-> in o
co in c in 'i
in -i- o 4-> id
a *- o t_ u o)
a E a. o 01 c-
01 ui co *
CM Q Ol
£ in c:
id eo. .
0) CO o j-> >- in
«- T- e <-> (-
3 Ol in 'r- rtj 01
in c= m o > 4-1
o -j; f- a. 01 a>
ex E E i z:
X 3 Ul UI
ui in
°s.
O "^
*J in 4J >,
id in id ^
o i- cc en
4J UI ~^
3
fl) ^^
"O in in in
in 3 a> a) T;
+-> OJ 4-> C
O) !- C_ 3 O
r- C31 01 S o
ja c 01 -i- 01
CZ
*-> t- 3 O
r- Ol C O
4J Ol -i- Ol
re o :c co
"*
ai
u
Id
3
4-> U.
C w^
(d
a.
O)
^
o o o -
4-> 4-> 4-> 3
CO CO CO U.
CM o in co
in co r^ en
CO PO CO CM
in * r^
i i
10 cr> I-H i
i-( i-l r-<
o o o o
0000
l-l t-H ,-H C3
C3
r-H
r- o « i
i
CM CO CO
CO C3 CM O
CM CO CM
(~H t-H r~(
C3 CO CO ^
r-; 10 vo o
3* t-H I-H in
t-H
CM
I-H
1
in
in
i
I-H
t-H
CM
in
4
CO
1
4-> I
in 3£
id
UI *
1 Id
O C
t_J Ol
a:
a: 01
co a:
+J 3
CO CO U.
CO CO CM
O 0
1
1 CO 1
o o o
CZ) O O
t-H t-H CZ)
o
t-H
CO Ol 1
1
U3 t-H O
eo en
I-H
IO IO O
*-H I-H in
CO
CM
1
1
CO
IO
o
t 1
vo
0
1
1
rH
CO
X
1
ft
o
1 in
O id
O Q.
a:
CO UJ
3
CO U.
CO CO
in en
CO CM
in
CO' !
T 1
o o
0 0
i-H O
O
tl
i-H |
IO
2°
OJ «*
r*- co
C\J
IT)
I
CM
CVJ
1
o
CTi
^
in
i i
t
00
CO
^
3
l id
o "3
T) U
c_
Ol
U 3C
CO
ai
.^
O !-
Id O>
4-> 3
CO U.
» CO
o< en
OJ CM
CO
i
co i
0 0
o o
i-l O
o
1 1
m i
t i
* o
CM
O CM
CO "*
IO
t-H
CO
CM
I
"3'
, ,
ai
«
CTt
CVJ
I
vo
CO
M
t t-
^3 QJ
CJ >
a: o
*f
l/l C3
01
O -r-
Id Ol
in co
in 01
CO CM
o
i
O 1
0 0
0 0
o
'
VO 1
a-' '
s°
,3.
S °
co m
to
CO
i
r 1
r- 1
r-H
CT>
ro
i
CO
CO
1
CO
sr
w
i t/i
X trt
ro O
«^
-------
27
TABLE 6 TOTAL EXPOSURE AND NUMBER OF PEOPLE EXPOSED
PRIMARY LEAD SMELTING INDUSTRY*
Plant
Total
Number of
People Exposed
Total
Exposure
(People - ug/m3)
1
2
3
4
5
48,600
497,000
1,510,000
97,300
42,700
215
715
186
27
7
* A 50-kilometer radius was used for the analysis of primary lead
smelting industry.
-------
28
TABLE 7
PUBLIC EXPOSURE FOR PRIMARY LEAD SMELTING INDUSTRY
AS PRODUCED BY THE HUMAN EXPOSURE MODEL
(ASSUMING BASELINE CONTROLS)
Concentration
Level (ug/m3)
Population
Exposed
(Persons)*
Exposure
(Persons - ug/m3)**
U.437
0.25
a.l
0.05
0.025
0.01
0 .005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.0000269
1
40
441
1240
7700
15900
71700
340000
545000
657000
1470000
2080000
2190000
0
0
6
33
62
144
199
398
801
945
983
1100
1150
1150
*Column 2 displays the computed value, rounded to the nearest whole number, of the
cumulative number of people exposed to the matching and higher concentration levels
found in column 1. For example, 0.5 people would be rounded to 0 and 0.51 people
would be rounded to 1.
**Column 3 displays the computed value of the cumulative exposure to the matching
and higher concentation levels found in column 1.
-------
29
3.4 Murph Metals-Dallas and Quemetco-Seattle
Predicted (HEM) versus measured data for Murph Metals-Dallas and
Quemetco-Seattle were plotted (Figures 3 and 4) and a least squares weighted
linear regression analysis was run based on a number of data points. The
least squares regression line (solid line) was determined on the basis of a
comparison of National Aerometric Data Bank monitoring data (circumscribed
dots) and State agency monitoring data (circumscribed Xs) with ambient con-
centrations predicted by the Human Exposure Model.
The reader should note that a perfect fit for the least squares
regression analysis results in a line running through the origin at a 45°
angle (dotted lines in Figures 3 and 4). This means that if the HEM model
predicts perfectly, then the data points would fall on the 45° line. In
cases where the HEM model underpredicts concentrations, data points will be
located above the 45° perfect fit line. Likewise, when the HEM model
overpredicts concentrations, data points will be located below the perfect
fit line. The regression line resulting from our comparison of predicted
and monitored data lies above the perfect fit line, intersecting the
ordinate axis at values of approximately 0.011 ug/m3 and 0.026 ug/m3 for Murph
Metals and Quemetco respectively. This result is consistent with the
expectation that air dispersion modeling would underpredict ambient con-
centrations. The air dispersion modeling did not consider other local sources
of arsenic such as naturally-occurring arsenic in the windblown dust and re-
entrained arsenic particulate matter that had settled to the earth from past
smelter emissions.
-------
30
FIGURE 3 Predicted Versus Measured
Inorganic Arsenic Ambient Concentrations
(Murph Metals - Dallas, TX)
stSiriT rtn}H:r£r:;)jirr: mtilit:
MODEL
UNDERPREDICTION
OVERPREDICTION OSES
Perfect Fit
Linear Regression Ii
Municipal Data
0.02 0.04
Predicted Concentration (/ig/m3)
0.06
-------
31
FIGURE 4 Predicted Versus Measured
Inorganic Arsenic Ambient Concentrations
(QUEMETCO - Seattle, WA)
0.1
OVERPREDICTIONg
UNDERPREDICTION
Perfect Fit
Linear Regression IS
EPA Data
State Data
0.02 0.04
Predicted Concentration
0.06
-------
32
3.4.1 Public Exposure to Inorganic Arsenic Emissions from Secondary Lead
Smelters
3.4.1.1 Source Data
Thirty-five secondary lead smelters are included in the analysis.
Table 8 lists arsenic concentrations near select secondary lead smelters.
Table 9 lists the names and addresses of the plants considered, and Table 10
lists the plant data used as input to the Human Exposure Model (HEM).
3.4.1.2 Exposure Data
Table 11 lists, on a plant-by-plant basis, the total number of
people encompassed by the exposure analysis and the total exposure. Total
exposure is the sum of the products of number of people times the ambient
air concentration to which they are exposed, as calculated by HEM. Table
12 sums, for the entire source category (35 plants), the number of people
exposed to various ambient concentrations, as calculated by HEM. (Source-
by-source exposure results are provided in the EPA docket numbered A-83-9.)
-------
33
Table 8
Arsenic Concentrations Near Select
Secondary Lead Smelters
Plant t
General Battery,
Reading, PA
Murph Metals-
Dallas, TX
Murph Metals-Dallas
Texas Air Control
Board Data
Quemetco-City of
Industry, CA
Quemetco-
Indianapolis, IN
Quemetco-Seattle,
WA
Quemetco-Seattle
Washington State Dept.
of Ecology Data
F Obs.
29
86
21
93
57
28
31
31
29
25
81
85
27
47
30
121
29
64
80
60
72
60
60
Di stance^ Bearing
(km)
5.1
3.6
3.7
7.6
9.0
0.2
0.2
0.5
17.8
22.8
23.6
24.0
31.2
32.2
35.1
36.1
38.1
12.5
1.9
3.2
13.5
0.2
1.4
189.5
181.6
181.5
311.6
256.2
0
337.5
157.5
314.0
281.9
164.2
275.2
218.2
161.0
300.0
84.8
235.8
78.2
150.8
30.3
2.6
157.5
180
Predicted2
(Pg/m3)
0.00104
0.0024
0.0024
0.00095
0.0004
0.062
0.042
0.014
0.00037
0.000133
0.00018
0.00023
0.000083
0.000112
0.000106
0.000113
0.000066
0.00040
0.0036
0.00183
0.00047
0.031
0.0075
Measured3 MDL4
(pg/m3)
0.009
0.028
0.010
0.029
0.025
0.085
0.077
0.025
0.005
0.003
0.003
0.006
0.003
0.003
0.003
0.005
0.004
0.005
0.041
0.038
0.020
0.09
0.03
(pg/m3)
0.0055
0.05
0.0055
0.05
0.05
«B
~
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
_
Percenti le5
30< %
90< %
30< %
90, %
^
~
70< %
90< %
70< %
70< %
70< 2
70< %
50< %
30< %
30< %
_
<50
<95*
<50*
<95*
>99*
<90*
>99*
<95*
<90*
>99*
<90*
>99*
<90*
<90*
<70*
<10
<50
<50
*iiui\*hai«^..j VIM UM f/w t 11 if nud \j i ^ i ^^u luics* ^cc *JC\* U 1 Vll u>o
-------
I
34
Table 9
Identification of Secondary Lead Smelters
Plant Number Code
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
" 30'"
31
32
33
34
35
Plant Name and Address
Alco Pacific Gardena, CA
Bergsoe St rfelens, OR
Chloride Metals Columbus, GA
Chloride Metals Tampa, FL
Dixie Metals Dallas, tX
tast Penn Lyons Station
Federated Metals San 1-ran, CA
General Battery Reading, PA
General Smelting College Grov, fl\f
Gopher Eugene, Minn
Gould Frisco, TX
Gould Vernon, CA
Gulf Coast Tampa, Fl
Hyman Viener Richmond, VA
Interstate Lead Leeds, AL
Lancaster Lancaster, PA
Master Metals Cleveland, OH
Murph Metals Dallas, TX
National Smelting Atlanta, GA
National Smelting Pedricktown, NJ
Quemetco City of5 Industry, CA
Quemetco Indianapolis, IN
Quemetco Seattle, VA
Refined Metals Beach Grove, IN
Refined Metals Memphis, TN
Revere Wall, NY
Ross Metals Rossville, TN
Sanders Lead troy, Al
SchuyTOrnBaton Rouge, LA
Schuylkill Forest City, MO
Standard San Antonio, TX
Taracorp Atlanta, GA
Taracorp Granite City, IL
Tonolli Nesquehoning, PA
USS Lead El. Chicago, IF
-------
8
o
O tn
*J r
o
tn t-
4-> +J
II
CU
U
C3
o.
t/1
01
o
e
O-
2 E
c
o
tn tn -u >,
E ^
UJ
4-> Q.
X 0)0
Ul 1
"o^ cu
(o cu en
o E
a to
u i- E
co a
« -u E
4-> HZ
t/i
c
o
en c_
W) -U >)
^ O£ "**^
E 01
LtJ -V*
_4J 0.
X ^-
co e
«
o E
10 1C
OO
O C-
VI V) +* ^
1 Q" x^
E
t/1
T3
CU C
a o
4-> CU
r- CO
01 1
o cu
^^, J
a> c
13 i-
3 s:
*-> i
f~ tn
+^ cu
>a cu
"^ CT
CU
o
*i
^S
0.
CTV CO OO
CM tn uo
o oo
P-H
CO CO CO
O CO CO
CM CM CM
LO CO CO
LT> CO
CM tO
-O
CM
-" O O
CO ^ ^
o> on
CM CM CM
CM CM CM
OOO
CO CO CO
to to cn
r^ to *^"
CO CO
0 CO 0
1 1 O
oo en i
i i in
CO CM 1
t-H CM *3"
0 CO 0
CM CO O
1 1 1
p o to
1 1 t
CO * CO
U
£
U CU
0 Si 0
O t. i
<=C CD CJ
s
CO
00
1 t
CO
CO
CO
CM
CO
CO
o
o
^
°I
CM
CM
tn
o
CO
cr.
«j-
CO
OJ
f »
4
CM
CM
CO
4
1
CM
to,
*S 0
CU t
o
CO CM CM CM CM CM CO OO Uj> CM CM CO CM CO CM,CO OJ GO CM 00
oc oo oo cn r^crtcooOf co en co cr* co r*. oo t
CM CO CO CO CO CO CM OOCMCOCOCOCOCMCOCMCOOOCO OO
OO OO C71OG OOcHCMOO CT\ i-< ^
i co .-i r-~
i . . if)
. t-1 CM OO <-H
co co co co co co co cococococococococococo'co1 co
co co co »* to<3-cococoCOr-tlO
IT) O
J-,4,
oo
r,
"!fo
to S ' ' '
01 i-i oo t~ oo
tnocrico5*'^'ooo
i i i i i i i i
CMCQOOOOr^t-ttI
IIII
OLf>oco
^CO^COCOCMCOCO
I I
VO >t t
r^ CO <
«-l CM I
I I
co co ;
§5!!
en
££
CU 4J
C "O O CO GO Ul d) 4-1
cuccu~~. x 4->
^o in
r^^CU^OICU^r-^-M-COCU
I -r-r O_(0r
OUJLU
-------
CJ
t_
f
c
o
i- £Ti£
X CU °
LU I
±£ O
O r CO
CO CO CM CM CM
CO CO to to to
r~ r^ co co co
> * M * M
co co in to 10
CM CM f1 to
CT* en to r*--
to to ^
en I-H cocncotocoi-Hcoco
CM O Cn
o co i
CM >-i
co co en 11
iI CM to IO
a
CJ
s
o
s:
E
Ul
a:
o'tn
*J r
o
u> t,
4-> 4->
3 C
P
CU
£
o
10 +J E
-H 3C
IO
co co co co co co co 01 3 £ co1
cocococo
totocoro co to * co to co * co to to co
2222 2 2222222222
CMCMCMCM CM CM CM CM CM CM CM CM CM CM CM
"* ** *
in LO in'
c^ cy> cyi to
*
^ a. o
<-> cfo
c
o
CJ
+ <i
t/1 CU
3 u>
o ra
c cc
en
o
u
CU
+j a.
I ET ^
X 0)0
-3£ O
o « a)
ra 2
01
lJ
W) tf) 4J *^
CO tO CO "3- CO CM * 00 CO
««....
1-1 "-a-
to en CD o
I 10 I 10 ,-1
CO I i-l I i
10 (^ CM CO «3"
I i-H I O O
r-~ i CM i i
-1 tO CM tO CD
I-H co -H co en
CO -! ^"CO 2
I I _L I I
"* iQ «T CM IO
O 1" CO "» o
to C3 to o cn co
co co 1-1 >a-10 co
,i 4 co 4 eo cf,
CM CO IO rH i-l CM
"=1* cn 10 1-1 vo co CM
^- 00 co cn en cn CM
co >* CM o to o co
i i i i i i ~-.
r- CM r--. co 1-1 o i--
CM o «* 10 o CM «*
00 CM r
O to CM
I I I
co
en
i
co co
I I I I
i o o cr
o co sr oj
CO (
CM CO (
to co co
a o 10
i i i
-------
37
Plant
Table 11
Total Exposure and Number of People Exposed
Secondary Lead Smelting Industry*
Total
Number of
People Exposed
Total
Exposure
(People - ug/m3)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
8,450,000
1,120,000
314,000
1,670,000
2,350,000
1,310,000
3,370,000
1,260,000
679,000
58,200
1,800,000
8,900,000
1,690,000
766,000
844,000
1,160,000
2,530,000
2,560,000
1,920,000
4,210,000
8,860,000
1,150,000
2,060,000
1,180,000
927 ,000
948,000
902,000
92,000
143,000
149,000
1,050,000
1,920,000
2,190,000
934,000
5,280,000
32
15
17
19
228
16
22
250
3
1
17
189
32
15
41
6
59
668
51
197
2300
460
576
67
115
249
4
11
4
2
24
83
159
21
103
* A 50-kilometer radius was used for the analysis of secondary lead
smel ters.
-------
I
38
Table 12
Public Exposure for Secondary Lead Smelters
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
0.101
0.1
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
Population
Exposed
(Persons)*
<1
<1
256
2880
16000
53100
152000
743000
1940000
4510000
12800000
21700000
31100000
44000000
55700000
64100000
70800000
73800000
7 4700000
Exposure
(Persons - ug/m3)**
0
0
16
104
300
543
878
1770
2590
3480
4750
5390
5720
5930
6010
6050
6060
6060
6060
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
39
3.5 Public Exposure to Inorganic Arsenic Emissions from Primary Zinc
Smelters
3. 5.1 Source Data
Five primary zinc smelters are included in the analysis. Table 13
lists ambient arsenic concentrations near select primary zinc smelters.
Table 14 lists the names and addresses of the plants considered, and Table
15 lists the plant data used as input to the Human Exposure Model (HEM).
3. 5. 2 Exposure Data
Table 16 lists, on a pi ant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total exposure
is the sum of the products of number of people times the ambient air
concentration to which they are exposed, as calculated by HEM. Table 17
sums, for the entire source category (5 plants), the numbers of people
exposed to various ambient concentrations, as calculated by HEM. ( Source-
by-source exposure results are provided in the EPA docket numbered A-83-23. )
-------
t- E
c-s
g
r~ I OJ
.. O «- p.
XI - a. a) (T
a) S x > 3
_
a; o E 4-> n-
t- !-> 3 J=
O 4-3 n3 O E
C -r- O .£1 rtJ -r- 2: a-
-------
41
Table 14
Identification of Primary Zinc Smelters
Plant Number Code
Plant Name and Address
1
2
3
4
5
St. Joe - Monaca, PA
ASARCO - Corpus Christi, TX
Amax - Sauget, IL
Jersey Miniere Zinc Co - Clarksville, TN
National Zinc - Bartlesville, OK
-------
cz
O !->
r- C. CU
i
»- 0- 1
E
UJ
in
c to
O CD
in j-> c= ii
tn c: cu o
,-,_(_ ^
E 0
uj a.
in
tz 10 >>
o CD <-> u
i 4J ! CU
in *J i- o tn
tn c x o
£ -r- Ul r- E
E o cu
HI f\ ^»
C ICCM
0 CZ E
>, -i- *j in o
r c. -^^ tn cz tn *t
cu4->tn tn -i o +J 10
am. T-ot-ocu
030 EO.OCUC-
E "O C- Ul CO <_3 C C_
Ul C O CU "-"*
a.4-> c vi cr cu u.
>T* T- « *r- e r~ flJ
o c/) to uj a si
+j nj ^
5.^ o> cz
IO M C CZ O ^~
O *^" O 4-> « tn
>> E i- c t> t-
4-» t. 3 in *- to cu
zs Q tn tn o > 4->
&E in « o_ cu cu
r- >i
E v^
UJ *
cu
"O tn tn in
Z) eu eu T3
4-^ CU 4-* CZ
r- C_ ZJ O
C7) cn cz o
c: cu -t- cu
O Q E CO
cu tn tn tn
T3 CU CU T3
ZI CU 4-> C
«-> t_ 3 O
r- D> CZ O
4-J CU t- Q)
_J
CZ
,!«
Q-
\^ J^ vx vx
U U U U
4-1 4-> 4J 4-»
CO CO CO CO
to tn to ir>
CM CM CO CM
CO CO CO CO
O CO r^ r-<
r~* r * o » i
f-H
80 o o
o o o
^H ^-1 t 1 t-H
"* co .-i o
CO ^-" CM OJ
»-H i-H r^ CM
to co co co
CTV OO t-~ OO
o to
0
1 t
1
o
CM
1
o
CO
CM
.-H
1
^f
1
Q_
t
CU IO
o u
-3 IO
cz
o
4-» E
CO
vj*
o
10
CO
cn
CO
CO
p^
CM
g
« 1
o
CM
8!
00
vo
CO
OJ
1
CTi
o
o
*3r
1
CM M
in
p-
f
CJ
1
O 3
0 a.
«: o x
CO CO t
"*
SX
OJ
un
OJ
^.
r-H
4
CM
t
r^.
00
s
Jj
CO
to
CO Z
1
t
cu
a.1
cu o >
t- o in
>, CU J^
cu - o c-
tn cz c
CM
CO
r-
U5
C5
o
tt
CO
t 1
CO
CO
t-H
01
CO
tr>
1
ir>
o-.
CM
4
^~
1
to
CO
u cu
C r-
r^i >
to
(O i
c cu
O -U
4-> 10 i^
(O CD O
z.
-------
43
Table 16
Total Exposure and Number of People
Primary Zinc Smelter*
Total
Number of
Plant People Exposed
1 2,000,000
2 336,000
3 2,200,000
4 235,000
5 120,000
Exposed
Total
Exposure
(People - pg/m^)
47
2
16
3
2
* A 50-kilometer radius was used for the analysis of primary
zinc smelters.
-------
44
Table 17
Public Exposure for Primary Zinc Smelters
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
0.00182
0.001
0.0005
0.00025
0.0001
0 .00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.0000001
0.00000005
Population
Exposed
(Persons)*
7
109
2350
19400
91700
212000
450000
1870000
3170000
3960000
4550000
4800000
4850000
4870000
4890000
Exposure
(Persons - ug/m3**
0
0
2
8
18
27
35
55
65
68
69
69
69
69
69
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
**
Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
45
3.6 Public Exposure to Inorganic Arsenic Emissions from Zinc Oxide
Plants
3.6.1 Source Data
Two zinc oxide plants are included in the analysis. Table 18 lists
ambient arsenic concentrations near select zinc oxide plants. Table 19
lists the names and addresses of the plants considered, and Table 20 lists
the plant data used as input to the Human Exposure Model (HEM).
3.6. 2 Exposure Data
Table 21 lists on a pi ant-by-pi ant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total
exposure is tHe sum of tHe products of number of people times the ambient
air concentration to which they are exposed, as calculated by HEM.
Table 22 sums, for the entire source category (2 plants), the numbers of
people exposed to various ambient concentrations, as calculated by HEM.
(Source-by-source exposure results are provided in the EPA docket numbered
A-83-11.)
-------
46
Table 18
Arsenic Concentrations Near
Select Zinc Oxide Plants
Plant
ASARCO-
Columbus,
New Jersey
Palme rton
t Obs
127
OH
Zinc-
, PA
Distance*-
(km)
3.8
Bearing
206.0
Predicts
(gg/m;
if
0.0000124
Measured3
(ug/m3)
0.006
MDL4
(ug/m3)
0.0055
Percentile^
70< % <90*
No data within 15 km
* Indicates data point was disregarded; see Section 3.5.1.1.
* Distance from source to monitor (km).
2 Concentration predicted by Human Exposure Model (HEM).
3 The measured values are weighted averages. When the sampled arsenic concentrations
were below the MDL, a value of 1/2 MDL was assumed for purposes of calculating the
annual averages.
* Minimum detection limit.
5 Percentila indicates percentage of data falling below minimum detectable levels.
-------
47
TABLE 19
Identification of Zinc Oxide Plants
Plant Number Code
Plant Name and Address
ASARCO-Columbus, OH
New Jersey Zinc
Palmerton, PA.
-------
in 4->
££,
E O
UJ Cu
o
CM
to
a.
cu
542
o
u t-
C 4->
r- c
M O
O
t- «
3 CO
S =>
E
3
2?,
in 4-1 V- o in
in c x o ^s.
E o
UJ a.
o c E
i 4-1 in o *
in c: in T-
tn -i- o 4-1 10
r- o t- o tu
E o. o cu t-
Ul 00
in » 10
tn o >
go. ^a;
ui ui
in 4J
in (O
r rv
O)
CU
t_
cn
*
in in in
Cu Cu ~o
cu 4-> c
4-> t- 3 O
en 01 c u
c cu !- cu
o a E to
cu 4-1 c
t- 3 O
01 c o
flj "r- QJ
a z: to
u
IO
4-1
to
CO
CO
CO
in
I
CM
CO
in
I
en
o o o
o o o
f-H r-1 r-i
"* CO CTi
CM t-4
in 10 **
in in in
ii 10 r^
CO CM CM
s
-------
49
Table 21
Total Exposure and Number of People Exposed
(Zinc Oxide Plants)*
Plant Total Number of
People Exposed
1 1,210,000
2 907,000
Total Exposure
(People - ug/rn^)
8
1260
* A 50 kilometer radius was used for the analysis of zinc oxide plants,
-------
50
Concentration
Level (ug/m3)
Table 22
Public Exposure for Zinc Oxide Plants
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Population Exposed
(Persons)*
Exposure
(Persons - ug/m3)**
0.269
0.25
0.1
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
2
2
138
1160
3990
11700
21300
54000
392000
732000
883000
908000
913000
976000
1160000
1360000
1610000
1840000
1960000
2110000
0
0
17
79
180
300
366
474
921
1200
1250
1260
1260
1260
1260
1260
1270
1270
1270
1270
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
51
3.7 Methodology for Reviewing Pollutant Concentrations - Cotton Gins
A total of 320 cotton gins were identified as processors of arsenic
desiccated cotton. Due to the large number of gins, EPA determined that it
was impractical to obtain the location data necessary for arsenic risk
assessment. Based on information regarding the range of processing rates
possible, four model plants operating at 4, 7, 12 and 20 bales/hour were
designed that are representative of the operations and emissions of the gin
population.7 These were located at each of three sites typical of the areas
in which the gins are located. Of the 320 gins, it was assumed that 32
processed 4 bales per hour, 96 processed 7 bales per hour, 160 processed 12
bales per hour, and 32 processed 20 bales per hour. The Human Exposure
Model was run for each scenario to establish a range of exposure and risk
estimates for individual sources. To provide data for validating the model
plant exposure estimates, two operating gins in south central Texas were
chosen for test sites over a one year period. Monitors were arranged in a
fan-like array of sites positioned at distances of 100, 200 and 400 meters
downwind of the gin. Upwind sites were placed at 400m (one gin only) and
100m. This configuration provided a total of 13 sampling sites. The study
was conducted over a period of one year with intense sampling (4 hour
intervals) for 15 days during the short ginning season followed by 6 day
interval sampling for the remainder of the year.
Data from these two gins were compared to Human Exposure Model
calculated values (Table 23). The comparison was hampered somewhat by
the large number of monitored values which fell below minimum detection
limits -- only 298 measurements out of 708 were above the MDL of 0.05
ug/m3. To circumvent this problem, a range of mean measured values was
developed. At one end, all values below MDL were considered as zero
values, and at the other end, all such values were considered equal to the
MDL of 0.05 ug/m3.
-------
52
Table 23
Arsenic Concentrations Near Two Texas Cotton Gins
Plant
A ( *9 bales/hr)
B ( =12 bales/hr)
Distance ( km)l
0.1
0.2
0.2
0.1
0.1
0.2
0.1
Predicted (HEM)
( uq/m3)
0.011
0.011
~.
0.011
Measured^
1 ug/m3)
0.083-0.088
__
0.051-0.060
0.12 -0.12
0.015-0.024
0.013-0.022
0.013-0.022
Distance from source to monitor.
Weighted mean concentrations for one calendar year. Lesser value
represents weighted mean concentration calculated with values less
than minimum detection limit set equal to zero. Greater value
represents weighted mean concentration calculated with values less
than minimum detection limit set equal to MDL (0.0065 gg/m3).
-------
53
When conparing the measured arsenic values to the predicted con-
centrations from the appropriate model gin exposure analysis, EPA found
that the predicted values were reasonably close to concentrations measured
very near the gins. The monitoring study data also showed that the
arsenic concentrations fell off very rapidly with distance from the gins.
This result suggests that people living at some distance from the gins
are not being significantly exposed to the gins' emissions. Such a
result, coupled with the observation that many gins are in rural areas
supports the Agency's conclusion that the aggregate risks for this source
category are low.
3.7.1 Public Exposure to Inorganic Arsenic Emissions from Cotton Gins
3.7.1.1 Source Data
Four model cotton gins at each of three geographic locations are
included in the analysis. Table 24 lists the names and addresses of the
plants considered, and Table 25 lists the plant data used as input to the
Human Exposure Model (HEM).
3.7.1.2 Exposure Data
Tables 26 - 37 sum, for the entire source category (12 plants), the
numbers of people exposed to various ambient concentrations, as calculated
by HEM. (Model plant-by-model plant exposure results are provided in the
EPA docket numbered A-83-10.)
-------
54
Table 24
Identification of Model Cotton Gins
Model Plant Location
Model Plant Production
Hutto, TX
4 Bales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales/Hour
Buckholtz, TX
4 Bales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales/Hour
Itasca, TX
4 Bales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales/Hour
-------
0 4J
t- C >
o cs 4-> u
in 4-> >- o in
in c x o ~~
'i 'o ^ "a! E
in
f
O C IOCM
0 C= E
c '-^ T- 4-> in o
O in tn cz m 'r-
4-> ' in -r- O 4-> re
u o >- o t- u UJ OO «C
c
03 O
-a c c.
O (1) O 0)
in z c -i- 4J 40 m
CM -r- in C 01 t-
QJ r tn '^- E QJ
in E a. M- 01
JQ tn re uj o 3E
re o oo ^--
1 M. ____«»_ ^^__
x en
r- = O
o E o 4J -i- in
4J 3 -r- C 4-> «-
w in » re 01
re in tn o > 4->
4-> t-
in 4-* >j
r Q£ 01
E ^
UJ -
ai
"O in in in
3 a> 4J =
*- t- 3 0
en en c o
C 03 'r- QJ
o a z oo
C
!-> t- 3 0
- en c o
4J 0) -I-
*-> 3
00 U.
ro co
CT> en
CM CM
a-
s :
in CM
CM rH
CO
O 1
en in
o o
«-H r-H
O
o
1
CO
CO
en
0
o
1
CO
CO
1
o
CO
£_
g
3:
X ~~
1 in
3
C/} U_
ro ro
en en
CM CM
^
0 1
CM 1
in r
CM CM
^"
i
O 1
en in
10 m
« «=!
t_
|
in
0)
CO
CM
g
^ 4-
0 !-
cio u_
ro ro
en en
CM CM
,3-
O 1
CM 1
in i.
CM CM
^t*
0 1
o m
r-<
CM CM
0 O
t t (*-(
3
O
IE
in
3
oo Li-
ra ro
en en
CM CM
3-
g !
un CM
CM 1-1
CO
1
O 1
en tn
o o
f i ? i
o
o
1
ro
o
i
en
0
0
CM
in
o
CO
X C.
-g
in
4-> in
i a>
0
\s t*r*.
O _^
3 -3-
00
g
^ 4-
O 1-
re cr
oo u_
ro ro
en en
CM CM
Q.
S !
in CM
CM .-H
^*
1
O 1
en in
r-. r^
CM CM
|
fi
u
re
CO,
1
\s ^_>
O 1-
re cr
oo u,
ro co
en en
CM CM
q.
C 1
CM 1
in r-.
CM CM
^>
0 1
en in
10 10
<3-«3-
c
in
V
re
OQ
CM
i-H
a.
3t T
U *r
4-> 3
OO U.
» ro
en en
CM CM
^.
0 1
CM 1
in f-
CM CM
^f
1
O 1
o m
(_H
CM CM
0 O
f4 f 1
c
tn
11
«
CQ
O
CM
(U
u* 4->
O 'I-
re cr
4-> 3
00 U.
ro ro
en en
CM CM
3-
O 1
CM 1
CM
CM <-l
CO
O 1
en in
o o
I I rH
o
o
en
o
en
0
0
o
i
CM
m
3
O
X IE
tn
0}
re i
o re
tn oo
re
>-H
S
T
U *r-
4-» 3
OO U-
ro ro
en en
CM CM
.3.
NJ 1
in CM
CM .-H
«3-
1
O 1
en in
r~ r^
CM CM
|
C
Ul
OJ
Q
CO
§;
"^
O "r-
4-> 3
00 U.
ro ro
en en
CM CM
^.
0 1
CM 1
in r»
CM CM
^«
1
0 1
en in
IO 1C
=t «3-
3
3
in
U
re
CO
CM
a.
'^
o *»~
re en
00 Cu
co co
en en
CM CM
^.
O 1
CM 1
CVJ CM
"3-
» |
0 1
Oin
CM CM
00
*"*
<_
1
1)
n3
CO
o
CM
-------
56
Table 26
Public Exposure for 4 Bales/Hour Model Cotton Gin
(Hutto,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.00263
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0 .00000025
1
1
1
6
23
112
177
433
1810
1810
3390
46800
285000
506000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the conputed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
57
Table 27
Public Exposure for 7 Bales/Hour Model Cotton Gin
(Hutto,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m^)
Population Exposed
(Persons )*
Exposure
(Persons-ug/m^)**
0.00682
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0 100005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.000000528
1
1
1
8
28
112
282
799
1810
2420
7220
55400
448000
506000
0
0
0
0
0
0
0
0
0
0
0
0
1
1
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
58
Table 28
Public Exposure for 12 Bales/Hour Model Cotton Gin
(Hutto.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (jjg/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/irr)**
0.011
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
1
. 1
1
5
25
102
161
523
1810
1810
3820
39500
232000
506000
0
0
0
0
0
0
0
0
0
0
0
1
1
2
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
59
Table 29
Public Exposure for 20 Bales/Hour Model Cotton Gin
(Hutto.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level
Population Exposed
(Persons)*
Exposure
(Persons-pg/m3)**
0.0234
0.01
0.005
0.0025
0.001
0 .0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
1
1
5
23
102
169
433
1810
1810
3390
46800
300000
506000
0
0
0
0
0
0
0
1
1
1
1
3
4
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
60
Table 30
Public Exposure for 4 Bales/Hour Model Cotton Gin
(Buckholts,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (gg/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.00263
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.000000196
<1
<1
<1
2
10
49
77
190
1050
1050
4020
10600
86700
129000
131000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
61
Table 31
Public Exposure for 7 Bales/Hour Model Cotton Gin
(Buckholts,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.00682
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.000000528
<1
<1
<1
3
12
49
124
269
1050
1050
6020
15500
121000
131000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
62
Table 32
Public Exposure for 12 Bales/Hour Model Cotton Gin
(Buckholts,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (jjg/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.011
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
<1
<1
<1
2
11
45
71
230
1050
1050
6020
10100
81500
131000
0
0
0
0
0
0
0
0
0
0
0
0
0
1
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
63
Table 33
Public Exposure for 20 Bales/Hour Model Cotton Gin
(Buckholts,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/nr)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.0234
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000002
<1
<1
2
10
45
74
190
1050
1050
4020
11300
89500
129000
131000
0
0
0
0
0
0
0
0
0
0
1
1
1
1
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
64
Table 34
Public Exposure for 4 Bales/Hour Model Cotton Gin
(Itasca,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (tig/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.0011
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.0000001
0.0000000634
1
1
5
19
57
153
489
1280
2140
2660
6520
38900
107000
156000
162000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,.
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
65
Table 35
Public Exposure for 7 Bales/Hour Model Cotton Gin
(Itasca,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.00285
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
.0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.000000171
1
1
7
23
70
167
587
1280
2140
3870
6520
65200
120000
159000
162000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For exanple, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
66
Table 36
Public Exposure for 12 Bales/Hour Model Cotton Gin
(Itasca,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.00461
0.0025
0.001
0 .0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.000000293
1
4
22
42
118
489
948
1980
2660
4970
20300
114000
153000
162000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
67
Table 37
Public Exposure for 20 Bales/Hour Model Cotton Gin
(Itasca.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/nr)
Population Exposed
(Persons)*
Exposure
(Persons-ug/m3)**
0.0097
0.005
'0.0025
0.001
0.0005
0.00025
0.0001
0.00005
.0.000025
0.00001
0.000005
0.0000025
0.000001
:0. 0000006 49
1
4
14
46
146
489
1280
2140
2660
6520
40100
109000
156000
162000
0
0
0
0
0
0
0
0
0
1
1
1
1
1
* Column 2 displays the computed value, rounded to the nearest whole number,
; of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
68
3.8 Public Exposure to Inorganic Arsenic Emissions from Arsenic
Plants
3.8.1 Source Data
Eight arsenic chemical plants are included in the analysis. Table 38
lists ambient arsenic concentrations near select arsenic chemical plants.
Table 39 lists the names and addresses of the plants considered, and Table
40 lists the plant data used as input to the Human Exposure Model (HEM).
3.8.2 Exposure Data
Table 41 lists, on a plant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total exposure
is the sum of the products- of numbers of people times the ambient air con-
centration to which they are exposed, as calculated by HEM. Table 42 sums,
for the entire source category (8 plants), the numbers of people exposed to
various ambient concentrations, as calculated by HEM. (Source-by-source
exposure results are provided in the EPA docket numbered A-83-23.)
-------
oij^ co ro t*-*
z^"o 3 < 1 o o
ro ro ' cu 3- o -j.
<-3 3 ro 3 in
ro o -. c o ft-
jro 3 co 3 to 01
vi 3 c ro 3 3
rt- 3 o o» rt- O ci
-" -hin -, ct> cu
' Q. c: cu wi
ro ro t '-3 rt- *
-" ro ro Q. o o i
30 3 3 rt
-j. _j. o cu "o in
o o r^ ~3 o ^
CU 3 C to C O
rt- £ ro a. -t -
ro >cu in -j. o 3
in _..w o ro rt
3 cu rt-
a ... a, -, ro rt- s
ro rt- in to a, o cu
-3 in tn
2 c X ,cr 3
ro 3 ro < o a.
rt- a.ia 3; -i. in
cu 3" c rt T
ia -
o- o"3 --^ §
ro c: <-* re
J » 3" rri t*
o cu ro 3
^ C*" *^^- «.
-^. V)
3 3 Cu i >
c-f '
g ro a.
CU Ql
s- =» ~>
ro 3 w
ro cu 3
O t i.
rt O
Cu CU
cr < o
' ro o
ro -33
Cu O
ro 3
in c+
->
fa
rt
O
in
ro
ro
ro
f
rt-
3"
ro
t i
i
cu
CD
-1
O
c:
XI
i
CO -3
o v:
3
3- -^
CU C
3 -
^ !
X 1
_
I]
o
o.
c
o
£
^
)_J
tn
^>
c
i.
3
ro
cu ;
§.°
«* 1
=2 <
~
1
cu
c
Q.
o
CU
^
Z
_
t *
7^*
s.
Cu ;
V* _
1
X »-
i
*
C
=
"*
1\
CTI
45.
p
O
ro
0
o
tn
tn
A
se
A
10
*
rf?
f i
in < ;
3'
^IQ Q
~_
^
ro f\.
\o 01
CO UD
ro ro
<__> ^s
co^i
o o
o
CD O
0 0
O O
O O
tn vo
o o
O B
O O
o o
.p, tn
o o
o o
0 C
tn tn
tji -^i
o o
A A
s< s«
A A
S 0
iinerai t
Concord.
o
o
o
^
~
J1
_
o"
CD
t/1
O
*~l
c
c
£t
5
~
H-
2
I:
CD
""
5
>-
CO
o
CO
c
^
0
,0000000079
o
o
o
00
o
o
c:
w
^^
o
A
S<
A
*
tn
-j
rD
Z3
DO
O =
C
«
* 7^
X
ro t i
O O^ CJl t ' CO "^J O 1
t 1 »
Ccn^jcoro^otnci
co cji r^ c^ "*j cr> vo f~*
Ovorocococo-»-ji *
oooooooo
l__i <^« ^3 C3 C3 C3 ^D C3
go o o o o o o
o o o c o o o
oooocooo
oooooooo
oooooooo
OOOOOOO_ts,
^o ro C3 i * VD ^~* ro
O i-* en co
OOOOOOOC3
oooooooo
i\>rorororooror\
oooooooo
oooooooo
en
O O VO vO tji 10
^ O (ji tn o c
/^ A A A
*^ 3r«»« **
V A A V A
X-^X-X-^sf^S4?
-o
a»
3
=tfe
0
3-
C3
^- rt-
3 cu
O
ro
ro
cw -? o
T in
- ro o
3 3 O
tO -* 3
O O
ro
03 1
3" rt- Cu
ro -3 cr
- * ro o ~j.
T= a, cu o co
to 13 oo
~**^ O in
toa> ' 2:
" a. a* n>
ro zj D*
rt- ~i
en
CO
(D
fD
3 O
z"cu
. c
T
corD
'O.
CO
2
' r-
tn
-------
70
Table 39
Identification of Arsenic Chemical Plants
Plant Number Code
Plant Name and Address
1
2
3
4
5
6
7
8
Diamond Shamrock - Greens Bayou, TX
Koppers Co. - Conley, 6A
Koppers Co., - Valparaiso, IN
Mineral Research & Development Co. -
Concord, NC
Osmose Wood Preserving Co., -Memphis, TN
Pennwalt Inc. - Bryan, TX
Vine!and Chemical - Vine!and, NJ
Voluntary Purchasing Group - Bonham, TX
-------
5
OJ
IO
t
I/I
£
IO
r
Ou
«
o
01
- c
c o
OJ O
in
i- OJ
"* .c
r r
OJ OJ
0 tn
o to
S CO
OJ cn
3 *r~
tn E
O 3
a. tn
X V)
LU «C
^ ^
o
IO
IO
a
1
c
o +J
.- C OJ
in ! ex
in o 5?
r- 0. (
E
UJ
C 10
O CD
1- p. *~.
in +J t= ^
in c: oj o
r 'r V ^-^*
E 0
UJ O.
in
C IO >s
O CD 4-> O
r- 4_> ! OJ
in +J o in
in c x o "^
i- -r- UJ i E
E O OJ
UJ a. >
c: IOCM
0 c E
r- 4J m O
O £- O OJ
E Q. o 0) I-
UJ 00 cC
C L.
O OJ -^
r- -M +J in
in 1= oj t-
in T- E OJ
'i S. -2 a!
LU a s
CO.
O -M 'f in
'r- C -M t-
in -i 10 OJ
in o > 4->
> o- ai a>
E i s:
LU UJ
C
o *»
"- OJ t-
in 10 *
r; ce cn
UJ ^ '
"O in in in
3 ai co -a
W OJ 4J C
r- S- 3 0
cn cn c u
C OJ 'r OJ
o a s: oo
O) in in tn
"o aj OJ ~o
3 CO 4J C:
£ fei S
4-> 0) » OJ
IO O S OO
C
IO
a.
sx
u
IO
CM
Ol
<-
O
0
OO
CO
o
00
T-H
8
0
o
CM
CM
1
CM
f-H
1
Ol
03
in
I
ct
CM
1
O
||
00 CO
"O tn
e c:
§0!
O)
10 C-
\s
U
IO
co
Ol
CM
f-4
O
OO
f I
0
f-H
CO
CM
O
0
a-
CO
Ol
t i
4
CO
1
7
CO
OO
1
O CD
O
M
OJ r
0. C
&O
0
^s
U
IO
OO
co
Ol
CM
VO
T-H
0
OO
IO
o
Ol
o
o
4
o
1
00
OO
1
co
OJ
5?
z.
1 >
Koppers Co.
Valparaiso
^_y
U
IO
00
co
Ol
CM
co
00°
o
CO
o
in
o
f-H
CM
CM
0
e
O
a-
4
CO
o
CO
Ol
CM
4-
CM
1
in
CO
I
O
o
o:
IO O
0! C
c o
^
u
to
00
co
Ol
CM
OO
<£
O
0
o
CO
OO
o
m
CO
e
f-H
in
Ol
4
o
1
0
Ol
OO
f-H
in
o
i
in
00
i
o
Osmose Wood
Preserving
Menphis,TN
u
IO
45
co
Ol
CM
cn
co
o
o
o
CO
OO
tn
o
CD
f-H
cn
f-H
O
0
CM
f-H
1
CM
CM
1
IO
Ol
o
OO
1
1
o
OO
1
c
I
u
r **
-------
Plant
1
2
3
4
5
6
7
8
72
Table 41
Total Exposure and Number of People
(Arsenic Chemical Plants)*
Total Number of
People Exposed
2,680,000
1,900,000
1,190,000
813,000
927,000
138,000
4,230,000
152,000
Exposed
Total Exposure
( People - M9/m^)
0
0
0
0
68
0
0
0
* A 50-kilometer radius was used for the analysis of arsenic chemical
plants.
-------
73
Table 42
Public Exposure for Arsenic Chemical Plants
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level dJ9/m^)
0.0541
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0 .000005
0.0000025
0.000001
0.0000005
0.00000025
0.0000001
0.00000005
«0. 000000025
,0 .00000001
0.000000005
0.0000000025
0.000000001
0 .0000000005
0.00000000025
0.0000000001
0.00000000005
0 .0000000000395
Population Exposed
(Persons)*
31
31
62
472
1440
2720
8130
16700
30900
105000
210000
374000
653000
852000
908000
935000
952000
984000
1040000
1100000
1320000
2230000
3670000
5880000
7750000
7890000
8320000
9910000
11900000
12000000
Exposure
(Persons-ug/m^)**
2
2
3
10
15
19
28
34
38
49
56
61
66
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
*Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
**Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
74
4 QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM INORGANIC ARSENIC
EMISSIONS
4.1 M ethodology (General)
4.1.1 The Two Basic Types of Risk
Two basic types of risk are dealt with in the analysis. "Aggregate
risk" applies to all of the people encompassed by the particular analysis.
Aggregate risk can be related to a single source, to all of the sources in
a source category, or to all of the source categories analyzed. Aggregate
risk is expressed as incidences of cancer among all of the people included
in the analysis, after 70 years of exposure. For statistical convenience,
it is often divided by 70 and expressed as cancer incidences per year.
"Individual risk" applies to the person or persons estimated to live in the
area of the highest ambient air concentrations and it applies to the single
source associated with this estimate as estimated by the dispersion model.
Individual risk is expressed as "maximum lifetime risk" and reflects the
probability of getting cancer if one were continuously exposed to the
estimated maximum ambient air concentration for 70 years.
4.1.2 The Calculation of Aggregate Risk
Aggregate risk is calculated by multiplying the total exposure produced
by HEM (for a single source, a category of sources, or all categories of
sources) by the unit risk estimate. The product is cancer incidences among
the included population after 70 years of exposure. The total exposure,
as calculated by HEM, is illustrated by the following equation:
Total Exposure =
-------
75
where
£ = summation over all grid points where exposure is calculated
Pi = population associated with grid point i,
Cj = long-term average inorganic arsenic concentration at grid point i,
N = number of grid points to 2.8 kilometers and number of ED/B6
centroids between 2.8 and 50 kilometers of each source.
To more clearly represent the concept of calculating aggregate risk, a
simplified example illustrating the concept follows:
EXWPLE
This example uses assumptions rather than actual data and uses only
three levels of exposure rather than the large number produced by HEM. The
assumed unit risk estimate is 4.29 x 10~3 at 1 ug/m3 and the assumed
exposures are:
ambient air number of people exposed
concentrations to given concentration
ug/m-
1 ug/m3
1,000
10,000
0.5 ug/m3 100,000
The probability of getting cancer if continuously exposed to the assumed
concentrations for 70 years is given by:
concentration unit risk probability of cancer
2 ug/m3
1 Mg/m3
x 4. 29 x lO'3
4.29 x lO"3
9 x 10-3
4 x 10
-3
0.5 ug/m3
x 4. 29 x 10-3
2 x 10-3
-------
76
9 x 10-3
4 x lO-3
2 x 10-3
x
X
X
1,000
10,000
100,000
The 70 year cancer incidence among the people exposed to these concentrations
is given by:
probability of cancer number of people at after 70 years
at each exposure level each exposure level of exposure
9
40
200
TOTAL = 249
The aggregate risk, or total cancer incidence, is 249 and, expressed
as cancer incidence per year, is 249 * 70, or 3.6 cancers per year. The
total cancer incidence and cancers per year apply to the total of 111,000
people assumed to be exposed to the given concentrations.
4.1. 3 The Calculation of Individual Risk
Individual risk, expressed as "maximum lifetime risk," is calculated
by multiplying the highest concentration to which the public is exposed, as
reported by HEM, by the unit risk estimate. The product, a probability of
getting cancer, applies to the number of people which HEM reports as being
exposed to the highest listed concentration. The concept involved is a
simple proportioning from the 1 ug/m3 on which the unit risk estimate is
based to the highest listed concentration. In other words:
maximum lifetime risk the unit risk estimate
highest concentration to = 1 pg/m^
which people are exposed
4.2 Risks Calculated for Emissions of Inorganic Arsenic
The explained methodologies for calculating maximum lifetime risk and
cancer incidences were applied to each plant, assuming a baseline level of
emissions. A baseline level of emissions means the level of emissions after
-------
77
the application of controls either currently in place or required to be in
place to conply with current state or Federal regulations but before application
of controls that would be required by a NESHAP.
Tables 43-49 summarize the calculated risks for each source category.
To understand the relevance of these numbers, one shoulqi refer to the
analytical uncertainties discussed in section 5 below. Note that the annual
incidence is not calculated for cotton gins. As mentioned earlier in this
document, it was impractical to identify and locate all the gins handling
arsenic-acid-desiccated cotton ( ~ 300 gins). The Agency does not have enough
available data to provide an estimate of annual cancer incidence that would
be comparable in accuracy to the other source category estimates. As outlined
in Section 3.7, three model gins operating at each of four production rates
were used to establish a range of exposure and risk estimates for individual
sources. Likewise, two operating gins in south central Texas were chosen
for ambient air monitoring in order to validate the model plant exposure
estimates. Maximum lifetime risk estimates were calculated for each of the
three model plants (Table 47) and for the two operating gins (Table 48).
-------
78
Table 43
Maximum Lifetime Risk and Cancer Incidence for Primary Lead Smelters
(Assuming Baseline Control s)
Plant
1
2
3
4
5
Maximum
Lifetime
Risk
2 x 10-3
4 x 10-5
2 x lO-5
1. 3 x 1C-5
4 x lO-5
Cancer Incidences
Per Year
0.013
0.044
0.011
0.0016
0.0004
-------
79
Table 44
Maximum Lifetime Risk and Cancer Incidence for Secondary Lead Smelters
(Assuming Baseline Controls)
Plant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Maximum
Lifetime
Risk
5 x 10-6
4 x 10-6
8 x 10-6
8 x 10-6
8 x 10-5
1.1 x lO-5
5 x ID'6
1. 1 x 10-4
2 x 10-7
1.6 x ID'5
9 x 10-6
4 x KT5
1.1 x ID'5
4 x ID'6
6 x 10-5
9 x ID'6
4 x ID"6
3 x 10-4
8 x ID'6
4 x 10-5
4 x 10-^
2 x 10-4
4 x 10-4
2 x ID'5
7 x 10-5
3 x 10-4
2 x 10-5
1.2 x 10-5
1.7 x 10-6
1. 1 x 10-6
1.5 x lO-5
6 x 10-6
1.5 x ID'5
3 x 10-5
2 x 10-5
Cancer Incidences
Per Year
0.0019
0. 0009
0.0010
0.0011
0.014
0.0010
0.0013
0.015
0.0002
<0. 0001
0.0010
0.011
0.0019
0. 0009
0.0024
0. 0004
0.0035
0.040
0. 00 31
0.012
0. 14
0.028
0.035
0. 0040
0.0069
0.015
0.0002
0. 0007
0.0002
0.0001
0.0015
0.0050
0. 0095
0.0013
0.0062
-------
80
Table 45
Maximum Lifetime Risk and Cancer Incidence for Primary Zinc Smelters
(Assuming Baseline Controls)
Plant
1
2
3
4
5
Maximum
Lifetime
Risk
8 x lO-6
1.9 x 10-7
1.1 x 10-6
9 x 10-7
3 x 10-6
Cancer Incidences
Per Year
0.0029
0.0001
0.0010
0.0002
0.0001
-------
81
Table 46
Maximum Lifetime Risk and Cancer Incidence for Zinc Oxide Plants
(Assuming Baseline Controls)
Plant
Maximum
Lifetime
Risk
4 x 10-7
Cancer Incidences
_ Per Year
0.005
1.2 x
0.077
-------
82
Table 47
Maximum Lifetime Risk and Cancer Incidence for Model Cotton Gins
(Assuming Baseline Controls)
Model
Plant
Hutto, TX
4 Bales/Hr
7 Bales/Hr
12 Bales/Hr
20 Bales/Hr
Buckholts.TX
4 Bales/Hr
7 Bales/Hr
12 Bales/Hr
20 Bales/Hr
Itasca,TX
4 Bales/Hr
7 Bales/Hr
12 Bales/Hr
20 Bales/Hr
Maximum
Lifetime
Risk
1.1 x lO-5
3 x lO-5
5 x lO-5
1.0 x lO-4
1. 1 x ID"5
3 x ID'5
5 x 10-5
1.0 x 10-4
5 x 10-6
1.2 x lO-5
2 x lO-5
4 x lO-5
Cancer Incidences
Per Year
<0. 0001
0. 0001
0.0001
0.0002
O.0001
O.OOOl
O.0001
0.0001
0.0001
<0.0001
<0.0001
0.0001
-------
83
Table 48
Lifetime Risk for Two Texas Cotton Gins
(Assuming Baseline Controls)
Plant
Maximum Lifetime Risk
5 x 10-4*
1.0 x 10-4
* Represents final risk estimate as incorporated
by EPA.
-------
84
Table 49
Maximum Lifetime Risk and Cancer Incidence for Arsenic Chemical Plants
(Assuming Baseline Controls)
Plant
1
2
3
4
5
6
7
8
Maximum
Lifetime
Risk
4 x 10~8
7 x 10-9
3 x 1CT8
3 x 10-8
2 x 1CT4
3 x 10-8
9 x ID'10
3 x 10-8
Cancer Incidences
Per Year
<0.0001
<0.0001
<0.0001
<0.0001
0.0042
<0.0001
<0.0001
<0.0001
-------
85
5 ANALYTICAL UNCERTAINTIES APPLICABLE TO THE CALCULATIONS OF PUBLIC
HEALTH RISKS CONTAINED IN THIS DOCUMENT
5.1 The Unit Risk Estimate
The procedure used to develop the unit risk estimate is described in
reference 2. The model used and its application to epidemiological data
have been the subjects of substantial comment by health scientists. The
uncertainties are too complex to be summarized sensibly in this appendix.
Readers who wish to go beyond the information presented in the reference
should see the following Federal Register notices: (1) OSHA's "Supplemental
Statement of Reasons for the Final Rule", 48 FR 1864 (January 14, 1983);
and (2) EPA's "Water Quality Documents Availability" 45 FR 79318 (November
28, 1980).
The unit risk estimate used in this analysis applies only to lung
cancer. Other health effects are possible; these include skin cancer,
hyperkeratosis, peripheral neuropathy, growth retardation and brain
dysfunction among children, and increase in adverse birth outcomes. No
numerical expressions of risks relevant to these health effects is included
in this analysis.
Although the estimates derived from the various studies are quite
consistent, there are a number of uncertainties associated with them. The
estimates were made from occupational studies that involved exposures only
after employment age was reached. In estimating risks from environmental
exposures throughout life, it was assumed through the absolute-risk model
that the increase in the age-specific mortality rates of lung cancer was a
function only of cumulative exposures, irrespective of how the exposure was
accumulated. Although this assumption provides an adequate description of
-------
86
all of the data, it may be in error when applied to exposures that begin
very early in life. Similarly, the linear models possibly are inaccurate
at low exposures, even though they provide reasonable descriptions of the
experimental data.
The risk assessment methods employed were severely constrained by the
fact that they were based only upon the analyses performed and reported by
the original authorsanalyses that had been performed for purposes other
than quantitative risk assessment. For example, although other measures of
exposure might be more appropriate, the analyses were necessarily based
upon cumulative dose, since that was the only usable measure reported. Given
greater access to the data from these studies, other dose measures, as well
as models other than the simple absolute-risk model, could be studied. It
is possible that such wide analyses would indicate that other approaches
are more appropriate than the ones applied here.
5.2 Public Exposure
5.2.1 General
The basic assumptions implicit in the methodology are that all exposure
occurs at people's residences, that people stay at the same location for 70
years, that the ambient air concentrations and the emissions which cause
these concentrations persist for 70 years, and that the concentrations are
the same inside and outside the residences. From this it can be seen that
public exposure is based on a hypothetical premise. It is not known whether
this results in an over-estimation or an underestimation of public exposure.
-------
87
5.2.2 The Public
The following are relevant to the public as dealt with in this analysis:
1. Studies show that all people are not equally susceptible to cancer.
There is no numerical recognition of the "most susceptible" subset of the
population exposed.
2. Studies indicate that whether or not exposure to a particular
carcinogen results in cancer may be affected by the person's exposure to
other substances. .The public's exposure to other substances is not
numerically considered.
3. Some members of the public included in this analysis are likely to
be exposed to inorganic arsenic in the air in the workplace, and workplace
air concentrations of a pollutant are customarily much higher than the
concentrations found in the ambient, or public air. Workplace exposures
are not numerically approximated.
4. Studies show that there is normally a long latent period between
exposure and the onset of lung cancer. This has not been numerically
recognized.
5. The people dealt with in the analysis are not located by actual
residences. As explained previously, people are grouped by census districts
and these groups are located at single points called the population centroids
The effect is that the actual locations of residences with respect to the
estimated ambient air concentrations are not known and that the relative
locations used in the exposure model may have changed since the 1980 census.
However, for the population sectors estimated to be at highest risk, U.S.
-------
88
Geological Survey topographical maps were checked to verify that people did
live or could live in locations near the sources as modeled predictions
estimated. Maps in certain instances were old and the possibility could
not be excluded that additional areas near sources have been developed
since publication of the maps.
6. Many people dealt with in this analysis are subject to exposure to
ambient air concentrations of inorganic arsenic where they travel and shop
(as in downtown areas and suburban shopping centers), where they congregate
(as in public parks, sports stadiums, and schoolyards), and where they work
outside (as mailmen, milkmen, and construction workers). These types of
exposures are not numerically dealt with.
5.2.3. The Ambient Air Concentrations
The following are relevant to the estimated ambient air concentrations
of inorganic arsenic used in this analysis:
1. Flat terrain was assumed in the dispersion model. Concentrations
much higher than those estimated would result if emissions impact on elevated
terrain or tall buildings near a plant.
2. The estimated concentrations do not account for the additive impact
of emissions from plants located close to one another.
3. The increase in concentrations that could result from re-entrainment
of arsenic-bearing dust from, e.g., city streets, dirt roads, and vacant
lots, is not considered.
-------
89
4. Meteorological data specific to plant sites are not used in the
dispersion model. As explained, HEM uses the meteorological data from the
STAR station nearest the plant site. Site-specific meteorological data
could result in significantly different estimates, e.g., the estimated
location of the highest concentrations.
5. In some cases, the arsenic emission rates are estimates that are based
on assumptions rather than on measured data.
-------
90
6 REFERENCES
1. National Academy of Sciences, "Arsenic," Committee on Medical and
Biological Effects of Environmental Pollutants, Washington, D.C., 1977.
Docket Number (OAQPS 79-8) II-A-3.
2. Health Assessment Document for Inorganic Arsenic - Final Report EPA-600/
8-83-021F March 1984, OAQPS Docket Number OAQPS 79-8, II-A-13.
3. U.S. EPA, et.al., "Environmental Cancer and Heart and Lung Disease,"
Fifth Annual Report to Congress by the Task Force on Environmental Cancer
and Health and Lung Disease, August, 1982.
4. OAQPS Guideline Series, "Guidelines on Air Quality Models". Publication
Number EPA-450/2-78-027, (OAQPS Guideline No. 1.2-080).
5. Systems Application, Inc., "Human Exposure to Atmospheric Concentrations
of Selected Chemicals." (Prepared for the U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina). Volume I, Publication
Number EPA-2/250-1, and Volume II, Publication Number EPA-1/250-2.
6.- NEA, Inc., "East Helena Source Apportionment Study Particulate Source
Apportionment Analysis Using the Chemical Mass Balance Receptor Model."
(Prepared for th,e Department of Health and Environmental Sciences, State
of Montana.) Volume I, September, 1982.
7. RADIAN Corporation, "Preliminary Study of Sources of Inorganic Arsenic."
(Prepared for the U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina.) Publication Number EPA-450/5-82-005, August 1982.
------- |