a
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
\ WASHINGTON, D.C. 20460
OFFICE OF PREVENTION, PESTICIDES
AND TOXIC SUBSTANCES
DRAFT BIOLOGICAL OPINIONS ISSUED UNDER THE ENDANGERED SPECIES ACT, BY
THE NATIONAL MARINE FISHERIES SERVICE, RELATED TO PESTICIDES AND PACIFIC
SALMON AND STEELHEAD SPECIES
EPA has initiated formal consultation with the National Marine Fisheries Service on the potential
effects of certain pesticides to Pacific salmon and steelhead listed under the Endangered Species
Act as either threatened or endangered. As EPA receives draft Biological Opinions relative to
these consultations, they will be posted to www.epa.gov/espp and included in a public docket
EPA-HQ-OPP-2008-0654 so EPA may receive public input on any changes to a pesticide's
registration recommended by the National Marine Fisheries Service.
BACKGROUND
The Endangered species Act (ESA) requires that Federal Agencies assess their "actions" to
determine whether species listed as threatened or endangered under the ESA, may be affected
by those actions, or whether critical habitat may be adversely modified. The registered uses of a
pesticide constitute an EPA "action" under the ESA.
If EPA determines a pesticide's registered uses are likely to adversely affect a federally listed
threatened or endangered species (listed species) or modify its critical habitat, EPA initiates
"formal consultation" with the U.S. Fish and Wildlife Service or the National Marine Fisheries
Service (the Service or Services), as appropriate. In response to a Federal Agency initiating
formal consultation, the Service(s) develops a Biological Opinion (BO) in which it provides its
opinion on whether the "action" is likely to jeopardize the continued existence of a listed species or
is likely to adversely modify designated critical habitat and, if so, describes alternatives to avoid
jeopardy.
PUBLIC INPUT
In 2005, EPA published in the Federal Register (FR 70 No. 211 pp. 66392-66402), a document
titled Endangered Species Protection Program Field Implementation. That notice of how EPA
intends to implement its responsibilities under the ESA, states (p 66401):
"If EPA must formally consult with the Services, after the Services issue a draft
Biological Opinion, EPA will welcome input from State, Tribal and local
governments on draft reasonable and prudent measures and alternatives.
The purpose of this review would be to determine whether the alternatives or
measures can be reasonably implemented and whether there are different
measures that may provide adequate protection but result in less impact to
pesticide users. The Agency will consider this input in developing its response to
draft Biological Opinions."
-------
APPLICANT INPUT
Further, the Services' Consultation Handbook (pp. 2-13), supports their consultation regulations
and states:
"... the Service and the action agency meet their obligations to [the applicant or
pesticide registrant] as outlined in 50 CFR section 402 through the following:
The applicant is entitled to review draft Biological Opinions obtained through
the action agency and to provide comments through the action agency.
The Service will discuss the basis of their biological determination with the
applicant and seek the applicant's expertise in identifying reasonable and
prudent alternatives ...".
COMMENTS
Draft Biological Opinions are being included in the docket (insert Docket ID) and posted to EPA's
Web site (http://www.epa.gov/oppfead1/endanger/litstatus/effects) to seek input on the Service's
recommended reasonable and prudent measures and alternatives, as noted above. Such input
should be submitted within 30 days of the date the Biological Opinion was included in the docket
in order to be considered in EPA's response to the Draft Biological Opinion. Comments received
by EPA on other aspects of the Draft Biological Opinion, will be forwarded to the Service for their
consideration.
As stated in the Services' regulations (50 CFR 402.14(g)(5)):
"All comments on the draft Biological Opinion must be submitted to the Service
through the Federal agency, although the applicant may send a copy of its
comments directly to the Service."
SUBMITTING YOUR COMMENTS
You may submit your comments, identified by the docket identification (ID) number EPA-
HQ-OPP-2008-0654 and the pesticide to which the Biological Opinion pertains, by one of the
following methods:
Federal eRulemaking Portal: http://www.regulations.gov. Follow the on-line instructions
for submitting comments.
Mail: Office of Pesticide Programs (OPP) Regulatory Public Docket (7502P), Environmental
Protection Agency, 1200 Pennsylvania
Ave., NW, Washington, DC 20460-0001.
Delivery: OPP Regulatory Public Docket (7502P), Environmental Protection Agency, Rm.
S-4400, One Potomac Yard (South Bldg.), 2777 S. Crystal Dr., Arlington, VA. Deliveries are
only accepted during the Docket's normal hours of operation (8:30 a.m. to 4 p.m., Monday
through Friday, excluding legal holidays). Special arrangements should be made for deliveries
-------
of boxed information. The Docket Facility telephone number is (703) 305-5805.
Instructions: EPA's policy is that all comments received will be included in the docket
without change and may be made available on-line at http://www.regulations.gov, including any
personal information provided, unless the comment includes information claimed to be
Confidential Business Information (CBI) or other information whose disclosure is restricted by
statute. Do not submit information that you consider to be CBI or otherwise protected through
regulations.gov or e-mail. The regulations.gov website is an 'anonymous access" system,
which means EPA will not know your identity or contact information unless you provide it in the
body of your comment. If you send an e-mail comment directly to EPA without going through
regulations.gov, your e-mail address will be automatically captured and included as part of the
comment that is placed in the docket and made available on the Internet. If you submit an
electronic comment, EPA recommends that you include your name and other contact
information in the body of your comment and with any disk or CD-ROM you submit. If EPA
cannot read your comment due to technical difficulties and cannot contact you for clarification,
EPA may not be able to consider your comment. Electronic files should avoid the use of
special characters, any form of encryption, and be free of any defects or viruses.
Docket: All documents in the docket are listed in the docket index available in
regulations.gov. To access the electronic docket, go to http://www.regulations.gov, select
"Advanced Search," then "Docket Search." Insert the docket ID number where indicated and
select the "Submit" button. Follow the instructions on the regulations.gov website to view the
docket index or access available documents. Although, listed in the index, some information is
not publicly available, e.g., CBI or other information whose disclosure is restricted by statute.
Certain other material, such as copyrighted material, is not placed on the Internet and will be
publicly available only in hard copy form. Publicly available docket materials are available
either in the electronic docket at http:// www.regulations.gov, or, if only available in hard copy,
at the OPP Regulatory Public Docket in Rm. S-4400, One Potomac Yard (South Bldg.), 2777
S. Crystal Dr., Arlington, VA. The hours of operation of this Docket Facility are from 8:30 a.m.
to 4 p.m., Monday through Friday, excluding legal holidays. The Docket Facility telephone
number is (703)305-5805.
-------
DRAFT
National Marine Fisheries Service
Endangered Species Act Section 7 Consultation
Draft Biological Opinion
Environmental Protection Agency Registration of
Pesticides Containing Carbaryl, Carbofuran, and Methomyl
Photograph: Tom Maurer, USFWS
DRAFT
-------
DRAFT
Table of Contents
Background 21
Consultation History 24
Description of the Proposed Action 32
The Federal Action 32
Carbaryl 38
Carbofuran 43
Methomyl 48
Species 52
Approach to this Assessment 54
Overview of NMFS' Assessment Framework 54
Evidence Available for the Consultation 56
Application of Approach in this Consultation 57
General conceptual framework for assessing risk of EPA's pesticide actions to listed
resources 59
Problem Formulation 59
Risk Characterization 66
Other Considerations 67
Action Area 68
Status of Listed Resources 71
Chinook Salmon 72
Chum Salmon 110
Coho Salmon 722
Sockeye Salmon 138
Steelhead 147
Environmental Baseline 192
Natural Mortality Factors 193
Parasites and/or Disease 193
Predation 194
WildlandFire 197
DRAFT 2
-------
DRAFT
Oceanographic Features and Climatic Variability 199
Climate Change 200
Anthropogenic Mortality Factors 201
Baseline Pesticide Detections in Aquatic Environments 201
Baseline Water Temperature- Clean Water Act 207
Baseline Habitat Condition 211
Geographic Regions 213
Southwest Coast Region 273
Land Use 218
Habitat Modification 224
Mining 225
Hydromodification Projects 225
Artificial Propagation 226
Commercial and Recreational Fishing 227
Alien Species 227
Atmospheric deposition 228
Pesticide Reduction Programs 228
Pacific Northwest Region 237
Columbia River Basin 235
Land Use 236
Agriculture and Ranching 236
Urban and Industrial Development 241
Habitat Modification 242
Habitat Restoration 244
Mining 244
Hydromodification Projects 245
Artificial Propagation 248
Commercial, Recreational, and Subsistence Fishing 249
Alien Species 251
Puget Sound Region 252
Land Use 254
DRAFT 3
-------
DRAFT
Habitat Modification 259
Industrial Development 260
Habitat Restoration 262
Mining 262
Artificial Propagation 262
Hydromodification Projects 263
Commercial and Recreational Fishing 264
Atmospheric deposition 265
Oregon- Washington-Northern California Coastal Drainages 2 65
Land Use 266
Habitat Modification 266
Mining 268
Hydromodification Projects 268
Commercial and Recreational Fishing 269
Atmospheric deposition 269
Environmental Protection Programs 269
Integration of the Environmental Baseline on Listed Resources 2 77
Effects of the Proposed Action 273
Exposure Analysis 273
Summary of Chemical Fate of A.Is 274
Habitats Occupied by Listed Salmonids 279
Modeling: Estimates of Exposure to Carbaryl, Carbofuran, andMethomyl 252
Exposure estimates for non-crop pesticide applications 282
Exposure estimates for crop applications 283
Utility of EECs for consultation 284
NMFS estimates of potential exposure in shallow water habitats used by salmonids
291
Monitoring Data: Measured Concentrations of Carbaryl, Carbofuran, Methomyl, 1-
Napthol and 3-Hydroxyfuran 296
Data Described in USEPA's Biological Evaluations 296
USGS NAWQA Data for California, Idaho, Oregon, and Washington 297
DRAFT 4
-------
DRAFT
Monitoring Data from California Department of Pesticide Regulation 299
Monitoring Data from Washington State 300
Summary of National and State Monitoring Databases 304
Targeted Monitoring Studies 304
Exposure to Other Action Stressors 309
Response Analysis 377
Mode and Mechanism of Action 318
pH and toxicity 320
Temperature andtoxicity 320
Studies with mixtures of AChE inhibiting insecticides 321
Summary of Toxicity Information Presented in the Biological Evaluations 325
Summary of Toxicity Information from Open Literature 342
Summary of Response Analysis: 368
Risk Characterization 370
Exposure and Response Integration 377
Carbaryl 371
Carbofuran 373
Methomyl 374
Relationship of pesticide use to effects in the field 376
Field studies in ESA-listed salmonid habitats: Willapa Bay and Gray's Harbor
Washington 380
Field incidents reported in EPA incident database 387
Mixture Analysis of Carbaryl, Carbofuran, and Methomyl 391
Evaluation of Risk Hypotheses: Individual Salmonids 395
Effects to Salmonid Populations from the Proposed Action 406
Conclusion on population-level effects 432
Effects to Designated Critical Habitat: Evaluation of Risk Hypotheses ¥33
Risk Hypotheses: 433
Cumulative Effects 436
Urban Growth ¥37
Mining ¥37
DRAFT 5
-------
DRAFT
Agriculture 437
Recreation 438
Integration and Synthesis 439
Effects of the proposed action at the species level 440
California Coastal Chinook Salmon 443
Central Valley Spring-run Chinook Salmon 445
Lower Columbia River Chinook salmon 446
Upper Columbia River Spring-run Chinook salmon 448
Puget Sound Chinook Salmon 449
Sacramento River Winter-run Chinook salmon 451
Snake River Fall-run Chinook salmon 452
Snake River Spring/Summer-run Chinook salmon 454
Upper Willamette River Chinook Salmon 456
Columbia River Chum Salmon 457
Hood Canal Summer-run Chum Salmon 459
Central California Coast Coho Salmon 460
Lower Columbia River Coho Salmon 461
Southern Oregon/Northern California Coast coho salmon 463
Oregon Coast Coho Salmon 464
OzetteLake Sockeye Salmon 465
Snake River Sockeye Salmon 467
Central California Coast steelhead 468
California Central Valley steelhead 469
Lower Columbia River Steelhead 471
Middle Columbia River Steelhead 472
Northern California steelhead 474
Puget Sound Steelhead 475
Snake River Basin Steelhead 477
South-Central California Coast steelhead 478
Southern California Steelhead 479
Upper Columbia River Steelhead 481
DRAFT 6
-------
DRAFT
Upper Willamette River Steelhead 482
Summary of Species-Level Effects 484
Effects of the Proposed Action to Designated Critical Habitat 485
Reduction in Prey 487
Reduction in Water Quality 487
Conclusion 488
Carbaryl and Carbofuran 488
Methomyl 489
Reasonable and Prudent Alternatives 493
Specific Elements of the Reasonable and Prudent Alternative 495
Incidental Take Statement 501
Amount or Extent of Take Anticipated 507
Reasonable and Prudent Measures 504
Terms and Conditions 505
Conservation Recommendations 505
Reinitiation Notice 506
References Cited. 507
Appendix 1: Population Modeling 545
Introduction 546
Methods 549
Organismal Model 549
Acute Toxicity Model 561
Results 562
References 573
Appendix 2. Species and Population Annual Rates of Growth 582
Appendix 3: Abbreviations. 592
Appendix 4: Glossary 598
DRAFT
-------
DRAFT
Table of Figures
Figure 1. Stressors of the Action 35
Figure 2. Conceptual framework for assessing risks of EPA's action to listed resources60
Figure 3. Exposure pathways to carbaryl, carbofuran, and methomyl and general
responses of listed Pacific salmonids and habitat 61
Figure 4. Physiological systems potentially affected by acetylcholinesterase inhibition 63
Figure 5. Map showing extent of inland action area with the range of all ESU and DPS
boundaries for ESA listed salmonids highlighted in gray 70
Figure 6. CC Chinook salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7 75
Figure 7. Legend for the Land Cover Class categories found in species distribution maps.
Land cover is based on the 2001 National Land Cover Data and classifications.
http://www.mrlc.gov/index.php 76
Figure 8. CV Chinook salmon distribution 80
Figure 9. LCR Chinook salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7 84
Figure 10. UCR Spring-run Chinook salmon distribution 89
Figure 11. Puget Sound Chinook distribution. The Legend for the Land Cover Class
categories is found in Figure 7 92
Figure 12. Sacramento River Winter-run Chinook salmon distribution. The Legend for
the Land Cover Class categories is found in Figure 7 97
Figure 13. SR fall-run Chinook salmon distribution. The Legend for the Land Cover
Class categories is found in Figure 7 99
Figure 14. SR Spring/Summer-run Chinook salmon distribution 103
Figure 15. UWR Chinook salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7 108
Figure 16. Columbia River Chum salmon distribution. The Legend for the Land Cover
Class categories is found in Figure 7 114
Figure 17. Hood Canal Summer-run Chum salmon distribution. The Legend for the
Land Cover Class categories is found in Figure 7 118
DRAFT 8
-------
DRAFT
Figure 18. CCC Coho salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7 124
Figure 19 . LCR coho salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7 128
Figure 20. Southern Oregon/Northern California Coast coho salmon distribution. The
Legend for the Land Cover Class categories is found in Figure 7. Status and Trends... 132
Figure 21. Oregon Coast Coho salmon distribution. The Legend for the Land Cover
Class categories is found in Figure 7 135
Figure 22. Ozette Lake Sockeye salmon distribution. The Legend for the Land Cover
Class categories is found in Figure 7 140
Figure 23. SR Sockeye Salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7 145
Figure 24. CCC steelhead. The Legend for the Land Cover Class categories is found in
Figure 7 151
Figure 25. California CV steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7 154
Figure 26. Lower Columbia River Steelhead distribution. The Legend for the Land
Cover Class categories is found in Figure 7 159
Figure 27. MCR Steelhead distribution. The Legend for the Land Cover Class categories
is found in Figure 7 165
Figure 28. Northern California Steelhead distribution. The Legend for the Land Cover
Class categories is found in Figure 7 169
Figure 29. Puget Sound steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7 172
Figure 30. SR Basin Steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7 175
Figure 31. S-CCC steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7 179
Figure 32. Southern California steelhead distribution. The Legend for the Land Cover
Class categories is found in Figure 7 181
DRAFT
-------
DRAFT
Figure 33. UCR Steelhead distribution. The Legend for the Land Cover Class categories
is found in Figure 7 184
Figure 34. UWR Steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7 189
Figure 35. Exposure analysis 273
Figure 36. Chemical structures of carbaryl, carbofuran, and methomyl 274
Figure 37. Response analysis 318
Figure 38. Schematic of the Risk Characterization Phase 370
Figure 39. Carbaryl exposure concentrations and salmonid assessment endpoints' effect
concentrations in |ig/L 373
Figure 40. Carbofuran exposure concentrations and salmonid assessment endpoints'
effect concentrations in |ig/L 374
Figure 41. Methomyl exposure concentrations and salmonid assessment endpoints'
effect concentrations in |ig/L 376
Figure 42. Percent AChE inhibition (A.) and percent mortality (B.) expected from
exposure to carbaryl (Cl), carbofuran (Cn), and methomyl (M) as separate constituents
and as mixtures (Cl 41 |ig/L, Cn 19 |ig/L, and M 85 |ig/L) 393
Figure 43. Probability plot for each pesticide showing the distribution of the geometric
means of the ECSOs for each aquatic invertebrate species. The straight line shows the
result of a linear regression. In the regression equation, the normsinvQ function returns
the inverse of the standard normal cumulative distribution. See text for more details. A)
Plot of carbaryl ECSOs. B) Plot of carbofuran ECSOs. C) Plot of methomyl ECSOs 419
Figure 44. Percent change in lambda for Ocean-type Chinook salmon following 4 d, 21
d, and 60 d exposures to carbaryl, carbofuran, and methomyl. Open symbols denote a
percent change in lambda of less than one standard deviation from control population.
Closed symbols represent a percent change in lambda of more than one standard
deviation from control population 424
Figure 45. Percent change in lambda for Stream-type Chinook salmon following 4 d, 21
d, and 60 d exposures to carbaryl, carbofuran, and methomyl. Open symbols denote a
percent change in lambda of less than one standard deviation from control population.
DRAFT 10
-------
DRAFT
Closed symbols represent a percent change in lambda of more than one standard
deviation from control population 425
Figure 46. Percent change in lambda for Coho salmon following 4 d, 21 d, and 60 d
exposures to carbaryl, carbofuran, and methomyl. Open symbols denote a percent
change in lambda of less than one standard deviation from control population. Closed
symbols represent a percent change in lambda of more than one standard deviation from
control population 425
Figure 47. Percent change in lambda for sockeye salmon following 4 d, 21 d, and 60 d
exposures to carbaryl, carbofuran, methomyl. Open symbols denote a percent change in
lambda of less than one standard deviation from control population. Closed symbols
represent a percent change in lambda of more than one standard deviation from control
population 426
DRAFT 11
-------
DRAFT
Table of Tables
Table 1. Registered uses and application rates for carbaryl in California, Idaho, Oregon,
and Washington (EPA 2007) 42
Table 2. Registered uses and application rates for carbofuran in California, Idaho,
Oregon, and Washington (EPA 2004) 47
Table 3. Examples of registered uses and application rates for methomyl in California,
Idaho, Oregon, and Washington (EPA 2007) 51
Table 4. Examples of salmonid lifestage assessment endpoints and measures 65
Table 5. Listed Species and Critical Habitat (denoted by asterisk) in the Action Area ..71
Table 6. CC Chinook salmon—preliminary population structure, abundances, and
hatchery contributions (Good, Waples et al. 2005) 77
Table 7. CV Chinook salmon—preliminary population structure, abundances, and
hatchery contributions (Good, Waples et al. 2005) 79
Table 8. LCR Chinook salmon - preliminary population structure, abundances, and
hatchery contributions (Good, Waples et al. 2005) 85
Table 9. UCR Chinook salmon - preliminary population structure, abundances, and
hatchery contributions (Good, Waples et al. 2005) 88
Table 10. Puget Sound Chinook salmon - preliminary population structure, abundances,
and hatchery contributions (Good, Waples et al. 2005) 93
Table 11. SR Spring/Summer Chinook salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005). Note: rpm denotes redds per mile 104
Table 12. UWR Chinook salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005). Note: rpm denotes redds per mile 109
Table 13. Columbia River Chum salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005) 113
Table 14. Hood Canal summer-run Chum salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005) 119
Table 15. CCC Coho salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005) 125
DRAFT 12
-------
DRAFT
Table 16. LCR Coho salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005) 129
Table 17. Oregon Coast Coho salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005) 136
Table 18. CCC Steelhead salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005) 152
Table 19. LCR Steelhead salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005) 160
Table 20. MCR Steelhead salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005) 166
Table 21. Northern California Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005) 170
Table 22. SR Basin Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005). Note: rpm denotes redds per mile 176
Table 23. Southern California Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005) 182
Table 24. UCR Steelhead salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005) 186
Table 25. UWR Steelhead salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005). Note: rpm denotes redds per mile 190
Table 26. Use figures for AChE inhibiting pesticides in California (CDPR 2007) 205
Table 27. Washington State water temperature thresholds for salmonid habitat. These
temperatures are representative of limits set by California, Idaho, and Oregon (WSDE
2006) 209
Table 28. Number of kilometers of river, stream and estuaries included in state 303(d)
lists due to temperature that are located within each salmonid ESU. Data was taken from
the most recent GIS layers available from state water quality assessments reports* 210
Table 29. USGS Subregions and accounting units within the Northwest and Southwest
Regions, along with ESUs present within the area (Seaber, Kapinos et al. 1987) 213
Table 30. Area of land use categories within the range Chinook and Coho Salmon ESUs
in km2. Land cover image data were taken from Multi-Resolution Land Characteristics
DRAFT 13
-------
DRAFT
(MRLC) Consortium, a consortium of nine federal agencies (USGS, EPA, USFS, NOAA,
NASA, BLM, NFS, NRCS, and USFWS) (National Land Cover Data 2001). Land cover
class definitions are available at: http://www.mrlc.gov/nlcd_defmitions.php 216
Table 31. Area of Land Use Categories within the Range of Steelhead Trout DPSs (km2).
Land cover image data were taken from Multi-Resolution Land Characteristics (MRLC)
Consortium, a consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA,
BLM, NFS, NRCS, and USFWS) (National Land Cover Data 2001). Land cover class
definitions are available at: http://www.mrlc.gov/nlcd_defmitions.php 217
Table 32. Select rivers in the southwest coast region (Carter and Resh 2005) 218
Table 33. Land uses and population density in several southwest coast watersheds
(Carter and Resh 2005) 219
Table 34. California's 2006 Section 303(d) List of Water Quality Limited Segments:
segments listed for exceeding temperature and carbofuran limits (CEPA 2007) 221
Table 35. Area of land use categories within Chinook Salmon ESUs in km2. Land cover
image data were taken from Multi-Resolution Land Characteristics (MRLC) Consortium,
a consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA, BLM, NFS,
NRCS, and USFWS) (NLCD 2001). Land cover class definitions are available at:
http://www.mrlc.gov/nlcd_defmitions.php 232
Table 36. Area of land use categories within chum and coho ESUs in km2. Land cover
image data were taken from Multi-Re solution Land Characteristics (MRLC) Consortium,
a consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA, BLM, NFS,
NRCS, and USFWS) (NLCD 2001). Land cover class definitions are available at:
http://www.mrlc.gov/nlcd_defmitions.php 233
Table 37. Area of land use categories within sockeye ESUs and steelhead DPSs in km2.
Land cover image data were taken from Multi-Resolution Land Characteristics (MRLC)
Consortium, a consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA,
BLM, NFS, NRCS, and USFWS) (NLCD 2001). Land cover class definitions are
available at: http://www.mrlc.gov/nlcd_defmitions.php 234
Table 38. Select tributaries of the Columbia River (Carter and Resh 2005) 235
Table 39. Land use and population density in select tributaries of the Columbia River
(Stanford, Hauer et al. 2005) 236
DRAFT 14
-------
DRAFT
Table 40. Amount of most common a.i.s applied to crops in Hood River Basin 1990-
1996 (Jenkins etal. 2004) 240
Table 41. Examples of Water Quality Contaminants in Residential and Urban Areas. 256
Table 42. Pollutants of Concern in Puget Sound (PSAT 2005) 261
Table 43. Change in total area (acres2) of tidal wetlands (tidal marshes and swamps) due
to filling and diking between 1870 and 1970 (Good 2000) 267
Table 44. Environmental fate characteristics of carbaryl (EPA 2003) 275
Table 45. Environmental fate characteristics of carbofuran (EPA 2004) 277
Table 46. Environmental fate characteristics of methomyl (EPA 2003) 278
Table 47. General life histories of Pacific salmonids 280
Table 48. Examples of registered uses of carbaryl, carbofuran, methomyl and the
exposure method used by EPA inBEs 282
Table 49. PRZM-EXAMS exposure estimates from EPA's BEs 284
Table 50. GENEEC estimated concentrations of carbaryl, carbofuran, and methomyl in
surface water adjacent to sweet corn, potatoes, sweet corn, potatoes, and citrus 289
Table 51. Average initial concentration of any a.i. in surface water resulting from a
direct overspray of aquatic habitat 292
Table 52. Average initial pesticide concentration in 10 m wide off-channel habitat per Ib
of pesticide applied based on AgDrift simulations 295
Table 53. National Maximum Concentrations of Carbaryl, Carbofuran, and Methomyl as
Reported in EPA BEs (EPA 2003a, EPA 2003b, and EPA 2004) 297
Table 54. Summary of Occurrences of Carbaryl, Carbofuran, Methomyl, and 1-Napthol
in USGS NAWQA Database (1992-2007) for California, Idaho, Oregon, and Washington
299
Table 55. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, and 3-
Hydroxycarbofuran in CDPR Database (1991-2006) 300
Table 56. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, 1-Napthol and
3-Hydroxycarbofuran in Surface Water - Washington EIM Database (1988-2007) 302
Table 57. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, 1-Napthol and
3-Hydroxycarbofuran in recent studies by Washington Department of Ecology (2003-
2007) 303
DRAFT 15
-------
DRAFT
Table 58. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, 1-Napthol and
3-Hydroxycarbofuran in Sediments in Washington EIM Database (1988-2007) 303
Table 59. Mean concentration ( g/L) of carbaryl in the Little Missouri River following
rangeland application of Sevin-4-Oil 305
Table 60. Carbofuran Concentrations in CDPR Studies of Rice Effluents (1995-1998)
306
Table 61. Examples of listed a.i.s on pesticide products containing carbaryl and
methomyl 311
Table 62. Detection of nonionic detergent degradates in streams of the U.S. (Koplin,
Furlong et al. 2002) 311
Table 63. Concentrations ( g/L) of insecticides used in mixture exposures. ECSOs were
calculated from dose-response data of AChE activity using non-linear regression. Coho
salmon exposed to 1.0, 0.4, or 0.1 EC50 treatments had an equipotent amount of each
carbamate or OP within the treatment e.g., to attain the 1.0 EC50 treatment for diazinon
and chlorpyrifos, 1.0 g/L of chlorpyrifos (0.5 the EC50) was combined with 72.5 g/L
(0.5oftheEC50) 323
Table 64 .Assessment endpoint toxicity values ( g/L) presented in BEs and REDs for
carbaryl, Carbofuran, and methomyl 329
Table 65. Study designs and results with freshwater aquatic invertebrates 358
Table 66. Summary of assessment endpoints and effect concentrations 369
Table 67. Published field studies designed to establish a relationship between N-methyl
carbamate contamination of aquatic habitats due to agricultural practices (adapted from
Table 2 in Schulz 2004) 378
Table 68. Carbaryl concentrations in aquatic organisms on treated sites (mg/kg) 384
Table 69. Estimated dungeness crab mortalities resulting from carbaryl applications in
WillapaBay and Gray's Harbor, Washington 385
Table 70. Estimated fish mortalities resulting from carbaryl applications in Willapa Bay
and Gray's Harbor, Washington 386
Table 71. Total number of benthic crustaceans (tanaids, cumaceans, amphipods,
copepods, and ostracods; adapted from Tufts 1990) 387
DRAFT 16
-------
DRAFT
Table 72. EPA summary of field incident data with carbaryl, carbofuran, and methomyl;
highlighted incidents (yellow) are discussed 388
Table 73. Predicted AChE inhibition and mortality from estimated and measured
exposure to carbaryl, carbofuran, and methomyl 394
Table 74. Modeled output for Ocean-type Chinook salmon exposed to 4 d exposures of
carbaryl, carbofuran, and methomyl reporting the impacted factors of survival as percent
dead, lambda and standard deviation, and percent change in lambda compared to an
unexposed population 410
Table 75. Modeled output for Stream-type Chinook salmon exposed to 4 d exposures of
carbaryl, carbofuran, and methomyl reporting the impacted factors of survival as percent
dead, lambda and standard deviation, and percent change in lambda compared to an
unexposed population 411
Table 76. Modeled output for Coho salmon exposed to 4 d exposures of carbaryl,
carbofuran, and methomyl reporting the impacted factors of survival as percent dead,
lambda and standard deviation, and percent change in lambda compared to an unexposed
population 412
Table 77. Modeled output for Sockeye salmon exposed to 4 d exposures of carbaryl,
carbofuran, and methomyl reporting the impacted factors of survival as percent dead,
lambda and standard deviation, and percent change in lambda compared to an unexposed
population 413
Table 78. Modeled output for Ocean-type Chinook, Stream-type Chinook, Sockeye, and
Coho salmon exposed to 4 d exposures of carbaryl-, carbofuran-, and methomyl-
containing mixtures. The table denotes the impacted factors of survival as percent dead,
lambda and standard deviation, and percent change in lambda compared to an unexposed
population 415
Table 79. Carbaryl, carbofuran, and methomyl survival EC50 concentrations at 50th , 10th
, and 5th percentiles from probability distribution plots 420
Table 80. Multiple application scenarios for carbaryl and methomyl and predicted
percent change in lambdas for salmon populations 427
Table 81. Jeopardy and Non-Jeopardy Determinations for Listed Species 491
DRAFT 17
-------
DRAFT
Table 82. Adverse Modification Determinations for Designated Critical Habitat of Listed
Species 492
Table 83. Mandatory pesticide no application buffers for ground and aerial applications
497
Table 84. Estimated environmental concentrations of carbaryl, carbofuran, and
methomyl applied at the rate or 1 Ib per acre for ground and aerial applications 497
DRAFT 18
-------
DRAFT
National Marine Fisheries Service
Endangered Species Act Section 7 Consultation
Biological Opinion
Agency:
United States Environmental Protection Agency
Activities Considered:
Authorization of pesticide products containing the
active ingredients carbaryl, carbofuran, and
methomyl, and their formulations in the United
States and its affiliated territories
Consultation Conducted by:
Endangered Species Division of the Office of
Protected Resources, National Marine Fisheries
Service
Approved by:
Date:
Section 7(a)(2) of the Endangered Species Act of 1973, as amended (ESA; 16 U.S.C.
§1531 et seq.) requires each federal agency to insure that any action they authorize, fund,
or carry out is not likely to jeopardize the continued existence of any endangered or
threatened species or result in the destruction or adverse modification of critical habitat of
such species. When a federal agency's action "may affect" a protected species, that
agency is required to consult formally with the National Marine Fisheries Service
(NMFS) or the U.S. Fish and Wildlife Service (USFWS), depending upon the endangered
species, threatened species, or designated critical habitat that may be affected by the
action (50 CFR §402.14(a)). Federal agencies are exempt from this general requirement
if they have concluded that an action "may affect but is not likely to adversely affect"
DRAFT
19
-------
DRAFT
endangered species, threatened species or designated critical habitat (50 CFR
§420.14(b)).
The United States (U.S.) Environmental Protection Agency (EPA) initiated consultation
with NMFS on its proposal to authorize use, pursuant to the Federal Insecticide,
Fungicide, and Rodenticide Act (FIFRA), 7 U.S.C. 136 et seq., of pesticide products
containing the active ingredients (a.i.s) of carbaryl, and methomyl on April 1, 2003, and
of carbofuran on December 1, 2004. EPA authorization of pesticide uses are categorized
as FIFRA sections 3 (new product registrations), 4 (reregistrations and special review),
18 (emergency use), or 24(c) [Special Local Needs (SLN)]. At that time, EPA
determined that uses of pesticide products containing these ingredients "may affect" most
of the 26 Evolutionarily Significant Units (ESUs) of Pacific salmonids listed as
endangered or threatened and designated critical habitat for the ESUs. This document
represents NMFS' biological opinion (Opinion) on the impacts of EPA's authorization of
pesticide products containing the above-mentioned a.i.s on the listed ESUs, plus on two
newly listed salmonids. This is a partial consultation because pursuant to the court's
order, EPA sought consultation on only this group of listed species under NMFS'
jurisdiction. However, even though the court's order did not address the two more
recently listed salmonids, NMFS analyzed the impacts of EPA's action to them because
they belong to the same taxon. NMFS analysis requires consideration of the same
information. Consultation with NMFS will be completed when EPA makes effect
determinations on all remaining species and consults with NMFS as necessary.
This Opinion is prepared in accordance with section 7(a)(2) of the ESA and
implementing regulations at 50 CFR §402. However, consistent with the decision in
Gifford Pinchot Task Force v. USFWS. 378 F.3d 1059 (9th Cir. 2004). we did not apply
the regulatory definition of "destruction or adverse modification of critical habitat" at 50
CFR §402.02. Instead, we relied on the statutory provisions of the ESA to complete our
analysis of the effects of the action on designated critical habitat.
DRAFT 20
-------
DRAFT
This Opinion is based on NMFS' review of the package of information the EPA
submitted with its 2003 and 2004 requests for formal consultation on the proposed
authorization of the above a.i.s. It also includes our review of recovery plans for listed
Pacific salmonids, past and current research and population dynamics modeling efforts,
monitoring reports from prior research, Opinions on similar research, published and
unpublished scientific information on the biology and ecology of threatened and
endangered salmonids in the action area, and other sources of information gathered and
evaluated during the consultation on the proposed authorization of a.i.s for carbaryl,
carbofuran, and methomyl. NMFS also considered information and comments provided
by EPA and by the registrants identified as applicants by EPA.
On January 30, 2001, the Washington Toxics Coalition, Northwest Coalition for
Alternatives to Pesticides, Pacific Coast Federation of Fishermen's Associations, and
Institute for Fisheries Resources filed a lawsuit against EPA in the U.S. District Court for
the Western District of Washington, Civ. No. 01-132. This lawsuit alleged that EPA
violated section 7(a)(2) of the ESA by failing to consult on the effects to 26 ESUs of
listed Pacific salmonids of its continuing approval of 54 pesticide a.i.s.
On July 2, 2002, the court ruled that EPA had violated ESA section 7(a)(2) and ordered
EPA to initiate interagency consultation and make determinations about effects to the
salmonids on all 54 a.i.s by December 2004.
In December 2002, EPA and the USFWS and NMFS (referred to as the Services) began
interagency discussions for streamlining EPA's court ordered consultations.
On January 24, 2003, EPA and the Services published an Advance Notice of Proposed
Rulemaking seeking public comment on improving the process by which EPA and the
Services work together to protect listed species and critical habitat (68 FR 3785).
DRAFT 21
-------
DRAFT
Between May and December 2003, EPA and the Services reviewed EPA's ecological risk
assessment methodology and earlier drafts of EPA's "Overview of the Ecological Risk
Assessment Process in the Office of Pesticide Programs, U.S. Environmental Protection
Agency (Overview Document)". EPA and the Services also developed counterpart
regulations to streamline the consultation process.
On January 22, 2004, the court enjoined application of pesticides within 20 (for ground)
and 100 (for aerial) feet (ft) of streams supporting salmon. Washington Toxics Coalition
v. EPA. 357 F.Supp. 2d 1266 (W.D. Wash. 2004). The court imposed several additional
restrictions on pesticide use in specific settings.
On January 23, 2004, EPA finalized its Overview Document which specified EPA's
conduct of ecological risk assessment on pesticide registrations.
On January 26, 2004, the Services approved EPA's procedures and methods for
conducting ecological risk assessments and approved interagency counterpart regulations
for EPA's pesticide registration program.
On January 30, 2004, the Services published in the Federal Register (69 FR 4465)
proposed joint counterpart regulations for consultation under the ESA for regulatory
actions under the FIFRA, codified at 50 C.F.R. Part 402 Subpart D.
On August 5, 2004, the Services promulgated final joint counterpart regulations for
EPA's ESA-related actions taken pursuant to FIFRA. These regulations and the
Alternative Conservation Agreement (ACA) under the regulations allowed EPA to
conduct independent analyses of potential impacts of pesticide registration on listed
species and their designated critical habitats. The ACA outlined procedures to ensure
EPA's risk assessment approach will produce effect determinations that reliably assess
the effects of pesticides on listed species and designated critical habitat. Additionally,
EPA and the Services agreed to meet annually, or more frequently as may be deemed
appropriate. The intention of these meetings was to identify new research and other
DRAFT 22
-------
DRAFT
activities that may improve EPA's current approach for assessing the potential ecological
risks posed by use of a pesticide to listed species or designated critical habitat.
On September 23, 2004, the Washington Toxics Coalition and others challenged the
counterpart regulations in the U.S. District Court for the Western District of Washington,
Civ. No. 04-1998, alleging that the regulations were not authorized by the ESA and that
the Services had not complied with the Administrative Procedure Act and the National
Environmental Policy Act (NEPA) in promulgating these counterpart regulations.
In January 2006, EPA and the Services developed a draft joint interagency research
agenda to address several critical areas of scientific and procedural uncertainties in EPA's
current effects determination process. The jointly developed document identified eight
areas of risk assessment and research uncertainties.
On August 24, 2006, the court determined the Services did not implement NEPA
procedures properly during their promulgation of the joint counterpart regulations for
EPA actions under FIFRA. Additionally, the court determined that the "not likely to
adversely affect" and emergency consultation provisions of the counterpart regulations
were arbitrary and capricious and contrary to the substantive requirements of ESA section
7(a)(2). The court determined that EPA may conduct its own formal consultation with
the Services' involvement. Washington Toxics Coalition v. Department of the Interior,
457 F.Supp. 2d 1158 (W.D.Wash. 2006).
On November 5, 2007, the Northwest Coalition for Alternatives to Pesticides and others
filed a legal complaint in the U.S. District Court for the Western District of Washington,
Civ. No. 07-1791, against NMFS for its unreasonable delay in completing the section 7
consultations for EPA's registration of 54 pesticide a.i.s.
On July 30, 2008, NMFS and the plaintiffs entered into a settlement agreement with the
Northwest Coalition for Alternatives to Pesticides. NMFS agreed to complete
consultation within four years on 37 a.i.s. (EPA had concluded that 17 of the 54 a.i.s at
DRAFT 23
-------
DRAFT
issue in the first litigation would not affect any listed salmonid species or any of their
designated critical habitat, and so did not initiate consultation on those a.i.s.)
On November 18, 2008, NMFS issued its first Opinion for three organophosphates:
chlorpyrifos, diazinon, and malathion. This second consultation evaluates three
carbamate insecticides: carbaryl, carbofuran, and methomyl. EPA consultations on
pesticide products currently focus on their effects of listed Pacific salmonids. EPA
consultations remain incomplete until all protected species under NMFS' jurisdiction are
covered.
On April 1, 2003, the EPA sent a letter to NMFS' Office of Protected Resources (OPR)
requesting section 7 consultation for the registration of the a.i. carbaryl and detailing its
effects determinations on 26 ESUs of Pacific salmonids listed at that time. In that same
letter, EPA's Office of Pesticide Programs (OPP) determined that the use of carbaryl will
have "no effect" for 4 ESUs, "may affect but is not likely to adversely affect" 2 ESUs,
and "may affect" 20 ESUs of listed salmonids. EPA's "no effect" determinations for
carbaryl applied to Northern California steelhead, Southern Oregon/Northern California
Coast coho salmon, Hood Canal Summer-run chum salmon, and Ozette Lake sockeye
salmon.
On April 1, 2003, the EPA sent a letter to NMFS' OPR requesting section 7 consultation
for the registration of the a.i. methomyl and detailing its effects determinations on 26
ESUs of Pacific salmonids listed at that time. In that same letter, the EPA's OPP
determined that the use of methomyl will have "no effect" for 2 ESUs, and "may affect"
24 ESUs of listed salmonids. EPA's "no effect" determinations for methomyl applied to
the Northern California steelhead and California Coastal Chinook salmon ESUs.
On December 1, 2004, the EPA sent a letter to NMFS' OPR requesting section 7
consultation for the registration of the a.i. carbofuran and detailing its effects
determinations on 26 ESUs of Pacific salmonids listed at that time. In that same letter,
DRAFT 24
-------
DRAFT
EPA's OPP determined that the use of carbofuran will have "no effect" for 3 ESUs; "may
affect but is not likely to adversely affect" 18 ESUs, and "may affect" 3 ESUs of listed
salmonids. EPA's "no effect" determinations applied to the California Coastal Chinook
salmon, Central California coho salmon, and Northern California steelhead.
On June 28, 2005, NMFS listed the Lower Columbia River coho salmon ESU as
endangered. Given this recent listing, EPA's 2003 and 2004 effects determinations for
carbaryl, carbofuran, and methomyl on listed Pacific salmonids lack an effects
determination for the Lower Columbia River coho salmon.
On May 22, 2007, NMFS listed the Puget Sound Steelhead Distinct Population Segment
(DPS) as threatened. Given this recent listing, EPA's 2003 and 2004 effects
determinations for carbaryl, carbofuran, and methomyl on listed Pacific salmonids lack
an effects determination for the Puget Sound steelhead.
On December 10-12, 2007, EPA and the Services met and discussed approaches for
moving forward with ESA consultations and pesticide registrations. The agencies agreed
to develop methodologies for filling existing data gaps. In the interim, the Services will
develop approaches within their Opinions to address these gaps. The agencies identified
communication and coordination mechanisms to address technical and policy issues and
procedures for conflict resolution.
On February 11, 2008, NMFS listed the Oregon Coast coho salmon ESU as threatened.
EPA's 2003 and 2004 initiation packages for carbaryl, carbofuran, and methomyl
provided an effect determination for the Oregon Coast coho salmon ESU. This ESU was
previously listed in 1998 and its ESA status was in-flux until 2008.
On August 20, 2008, NMFS met with EPA and requested EPA to identify applicants for
this and subsequent pesticide consultations. NMFS also requested information on EPA's
cancellation of carbofuran and of existing stocks of carbofuran.
DRAFT 25
-------
DRAFT
On August 29, 2008, NMFS met with EPA and applicants for chlorpyrifos, diazinon, and
malathion. At that meeting, NMFS asked EPA to identify applicants for this and
subsequent pesticide consultations.
On September 16, 2008, NMFS requested EPA to confirm the status of EPA's
cancellation of carbofuran and for existing stocks of that same compound during a
conference call.
On September 17, 2008, NMFS requested EPA approval of Confidential Business
Information (CBI) clearance in accordance with FIFRA regulations and access to EPA's
incident database so NMFS staff may evaluate CBI materials from the applicants and
incident reports for the a.i.s under consultation. EPA conveyed to NMFS that no access
to the incident database would be authorized and the reports will be sent directly from
EPA to NMFS.
On September 23, 2008, NMFS staff received notification of CBI clearance from EPA.
On September 26, 2008, NMFS sent correspondence to EPA informing it of the roles of
the action agency and applicants during formal consultation. NMFS also requested
incident reports and label information for subsequent pesticide consultations from EPA.
The specified timeline for NMFS' receipt of incident report and label information for
carbaryl, carbofuran, and methomyl was November 3, 2008.
On October 3, 2008, NMFS received post-2002 incident reports for carbaryl, carbofuran,
and methomyl from EPA.
On November 5, 2008, NMFS sent an e-mail to EPA requesting it to identify applicants
for upcoming pesticide consultations and label and incident report information for
carbaryl, carbofuran, and methomyl. NMFS also requested information regarding
whether final cancellation of carbofuran or any if its uses had occurred.
DRAFT 26
-------
DRAFT
On November 13, 2008, EPA provided an interim e-mail response to NMFS' November
5, 2008 query. EPA stated that it was developing a process to identify applicants for
carbaryl, carbofuran, and methomyl. No applicants were identified in EPA's response.
EPA also stated that incident data for carbaryl, carbofuran, and methomyl were sent via
FedEx to NMFS on October 2, 2008. Finally, EPA confirmed that no final cancellations
for carbofuran have occurred subsequent to the Scientific Advisory Panel (SAP) meeting
required for action relative to the Notice of Intent to cancel this compound. The SAP
meeting occurred on February 5, 2008.
On December 1, 2008, NMFS repeated its request to EPA to identify applicants for
carbaryl, carbofuran, and methomyl via e-mail. NMFS also requested EPA to provide
technical staff contact information for these same chemicals so NMFS staff may request
information from them during this consultation.
On December 15, 2008, EPA informed NMFS via e-mail that it would send letters to the
technical registrants of carbaryl, carbofuran, and methomyl. EPA also stated that it
would inform the parties that they may submit information relative to the consultation
directly to NMFS with a copy to EPA. EPA also offered to include additional
information requested by NMFS pertinent to the consultation into that same letter.
On December 16, 2008, NMFS and EPA discussed each agency's notification strategy of
prospective applicants for this consultation. As the action agency, EPA indicated it
would identify and contact prospective applicants. EPA limited applicant status to those
technical registrants who have all information pertinent to the consultation.
On December 18, 2008, EPA sent formal correspondence to four technical registrants.
EPA's letter requested confirmation on their desire to have applicant status and for
parties to submit data not already provided with EPA's consultations that may inform the
outcome of the consultation. That information includes any toxicity data, field studies or
mesocosm studies not part of the consultation package, or EPA's Interim Registration
Eligibility Decision (TRED) or Registration Eligibility Decision (RED) documents for the
DRAFT 27
-------
DRAFT
pesticide a.i.; and current labels for end use products or if available, a master label that
includes all use instructions for all products containing the a.i.s. These data would be
submitted to NMFS and EPA.
On December 19, 2008, EPA identified technical staff contact information and four
applicants to NMFS for this consultation via formal correspondence. In that same letter,
EPA referred NMFS to the IRED and RED documents for any changes to the three
subject a.i.s since consultation was initiated in 2003 and 2004.
On that same date, NMFS received electronic files for EPA letters sent to the applicants.
EPA identified the following applicants: Bayer CropScience LP, Drexel Chemical
Company, E.I., duPont de Nemours and Company (DuPont), and FMC Corporation.
On January 5, 2009, NMFS requested clarification on EPA's registration action for
carbofuran via e-mail.
On January 7, 2009, EPA notified NMFS via email that it would provide a full response
to NMFS' query on carbofuran as soon as possible.
On January 8 and 9, 2009, EPA and NMFS exchanged e-mails scheduling a meeting with
identified applicants for this consultation in January.
On January 9, 2009, NMFS provided EPA with electronic draft files for the Description
of the Proposed Action and associated Appendix for this consultation. NMFS requested
EPA comments on these documents by January 23, 2009.
On January 21, 2009, the agencies agreed to meet with the applicants on January 30,
2009.
On January 22, 2009, NMFS requested the following information from EPA: the
bibliography for the carbofuran IRED (August 3, 2006), the report study cited as Table
DRAFT 28
-------
DRAFT
16 within the carbofuran IRED; and the methomyl study cited in the Science Chapter -
Master Record Identification Number (MRID) #00131255.
On that same date, DuPont informed NMFS that it would submit information to NMFS
and EPA to support the consultation for methomyl.
On that same date, EPA informed NMFS that it would provide comments to NMFS on
the draft Description of the Proposed Action by January 30, 2009.
On January 23, 2009, EPA instructed DuPont to send any data in support of the
methomyl consultation to both EPA and NMFS.
On January 26, 2009, NMFS received data on methomyl from DuPont. The package
included a cover letter, analysis of risk to methomyl to listed Pacific salmonids; copies of
four studies, including three toxicity tests with formulated material and an environmental
fate study (dissipation of methomyl in a simulated pond); and copies of methomyl
product labels held by DuPont.
On that same date, NMFS also received confidential information on sales of Lannate
(DuPont methomyl product) in Washington and Oregon from 2004-2006.
On January 27, 2009, NMFS asked EPA via e-mail to identify agenda topics and a list of
participants/applicants for the January 30, 2009, meeting. EPA provided the requested
information on that same date.
On January 29, 2009, NMFS received information from Bayer CropScience, including a
summary of Section 3 label restrictions, a "master label table", and copies of current 24
(c) and Section 3 labels.
On January 30, 2009, NMFS met with EPA, Bayer CropScience, DuPont, and FMC
Corporation. At this meeting, NMFS explained the consultation procedure and timelines
DRAFT 29
-------
DRAFT
for this consultation. Applicants presented information to NMFS and EPA. This venue
facilitated a question-answer session between the applicants and the agencies.
On that same date, NMFS received electronic files of applicant presentation materials on
carbaryl, carbofuran, and methomyl. EPA also provided NMFS with a PDF file of the
Environmental Fate and Effects Division (EFED) RED for methomyl.
On February 2, 2009, NMFS received carbofuran data from FMC Corporation. Materials
included the status on furadan registration and use in the relevant Pacific Northwest, the
proposed federal label for furadan, current special local needs label for potato use in
Oregon and spinach grown for seed use in Washington, and the Federal Register notice
for proposed voluntary cancellation of most uses.
On that same date, NMFS queried FMC Corporation via e-mail regarding when EPA's
response on the proposed federal label for furadan is expected. NMFS also requested
information whether the proposed permitted uses for foliar application of furadan on
cotton would apply in California, Idaho, Oregon, and Washington.
On February 3, 2009, NMFS received comments from EPA on the draft Description of
the Proposed Action relevant to carbofuran and methomyl. EPA disagreed with FMC's
statement that use of carbofuran has been discontinued on field corn in the Pacific
Northwest. According to EPA, carbofuran use is specified on federal labels and is used
in California, Idaho, Oregon, and Washington. Carbofuran is also used on sunflowers
grown in these states. EPA feedback for methomyl pertained to the SLN in California
and the status of some section 18c actions.
On that same date, NMFS requested clarification on registered carbofuran uses and
EPA's verification of NMFS' draft Description of the Proposed Action for this
ingredient.
DRAFT 30
-------
DRAFT
On February 5, 2009, NMFS received the two early life stage studies (MRID 131255 and
126862) from EPA. On that same date, FMC Corporation responded to NMFS'
questions raised in it February 2, 2009, e-mail.
On February 9, 2009, EPA provided responses via e-mail towards questions posed in
NMFS' February 3, 2009, e-mail. EPA provided no comments on NMFS' Description of
the Proposed Action for carbaryl and no response towards NMFS' questions on
carbofuran.
On February 11, 2009, NMFS received a study on carbaryl from Bayer CropScience.
On February 13, 2009, NMFS contacted EPA by phone and requested EPA comments on
the draft Description of the Proposed Action.
On February 17, 2009, NMFS received information from Bayer CropScience on carbaryl.
On February 20, 2009, EPA e-mailed NMFS information on carbaryl use patterns for
Idaho, Oregon, and Washington.
On March 5, 2009, NMFS received an extension from the court for this consultation from
March 31, 2009, to April 20, 2009.
On March 12, 2009, NMFS requested clarification from EPA on the carbaryl use data.
DRAFT 31
-------
DRAFT
The Federal Action
The proposed action encompasses EPA's registration of the uses (as described by product
labels) of all pesticides containing carbaryl, carbofuran, and methomyl. The purpose of
the proposed action is to provide tools for pest control that do not cause unreasonable
adverse effects to the environment throughout the U.S. and its affiliated territories.
Pursuant to FIFRA, before a pesticide product may be sold or distributed in the U.S. it
must be exempted or registered with a label identifying approved uses by EPA's OPP.
Once registered, a pesticide may not legally be used unless the use is consistent with
directions on its approved label
(http:www.epa.gov/pesticides/regulating/registering/index.htm). EPA authorization of
pesticide uses are categorized as FIFRA sections 3 (new product registrations), 4 (re-
registrations and special review), 18 (emergency use), or 24(c) SLN.
EPA's pesticide registration process involves an examination of the ingredients of a
pesticide, the site or crop on which it will be used, the amount, frequency and timing of
its use, and its storage and disposal practices. Pesticide ingredients may include active
and other ingredients, adjuvants, and surfactants (described in greater detail below). The
EPA evaluates the pesticide to ensure that it will not have unreasonable adverse effects
on humans, the environment, and non-target species. An unreasonable adverse effect on
the environment is defined in FIFRA as, "(1) any unreasonable risk to man or the
environment, taking into account the economic, social, and environmental costs and
benefits of the use of the pesticide, or (2) a human dietary risk from residues that result
from a use of a pesticide in or on any food inconsistent with the standard under section
408 of the Federal Food, Drug, and Cosmetic Act (FFDCA) (21 U.S.C. §346a)." 7
U.S.C. 136(b).
After registering a pesticide, EPA retains discretionary involvement and control over
such registration. EPA must periodically review the registration to ensure compliance
DRAFT 32
-------
DRAFT
with FIFRA and other federal laws (7 U.S.C. §136d). A pesticide registration will be
cancelled whenever "a pesticide or its labeling or other material... does not comply with
the provisions of FIFRA or, when used in accordance with widespread and commonly
recognized practice, generally causes unreasonable adverse effects on the environment."
On December 12, 2007, EPA, NMFS, and FWS agreed that the federal action for EPA's
FIFRA registration actions will be defined as the "authorization for use or uses described
in labeling of a pesticide product containing a particular pesticide ingredient." In order
to ensure that EPA's action will not jeopardize listed species or destroy or adversely
modify critical habitat, NMFS' analysis necessarily encompasses the impacts to Pacific
salmonid ESUs/DPSs of all authorized uses by EPA, regardless of whether those uses
have historically occurred.
Pesticide Labels. For this consultation, EPA's proposed action encompasses all approved
product labels containing carbaryl, carbofuran, or methomyl; their degradates,
metabolites, and formulations, including other ingredients within the formulations;
adjuvants; tank mixtures; and their individual and collective interactions when applied in
agricultural, urban, and residential landscapes throughout the U.S. and its territories.
These activities comprise the stressors of the action (Figure 1). The three biological
evaluations (BEs) indicate that carbaryl, carbofuran, and methomyl are labeled for a
variety of uses including applications to residential areas and crop lands ((EPA 2003;
EPA 2003; EPA 2004). Modifications have been made or are planned for new product
labels containing carbaryl, carbofuran, and methomyl as a result of reregi strati on
activities that have occurred since the release of the BEs.
The Food Quality Protection Act (FQPA) of 1996 required EPA to complete an
assessment of the cumulative effects on human health resulting from exposure to multiple
chemicals that have a common mechanism of toxicity. In 2001, EPA identified the N-
methyl carbamate (NMC) pesticides as a group which shares a common mechanism of
toxicity. This group includes carbaryl, carbofuran, methomyl, and seven other
cholinesterase-inhibiting pesticides
DRAFT 33
-------
DRAFT
(http://www.epa.gov/pesticides/cumulative/carbamate_risk_mgmt.htm). EPA published
a preliminary Cumulative Risk Assessment for NMC pesticides in 2005 and revised the
risk assessment in 2007 (EPA 2007). Concurrent with completing the revised
assessments, EPA completed tolerance reassessments and REDs for the NMC pesticides
(EPA 1998; EPA 2006; EPA 2007; EPA 2008). EPA has identified measures to address
cumulative risk of NMC pesticides
(http://www.epa.gov/pesticides/cumulative/carbamate risk mgmt.htm). Some of the risk
reduction measures for carbaryl, carbofuran, and methomyl follow:
Carbaryl - EPA intends to evaluate the revised worker assessment, which may require
an amendment to the RED. EPA continues to respond to petitions requesting that
carbaryl be cancelled and its tolerances revoked.
Carbofuran - EPA is pursuing cancellation of all carbofuran uses in the U.S. EPA has
received a request from the carbofuran registrant, FMC Corporation, for voluntary
cancellation of 22 crop uses of this pesticide. FMC Corporation has six uses not
proposed for voluntary cancellation that EPA indicates still present risk concerns and are
subject to future regulatory action by EPA. In July 2008, EPA initiated action to revoke
existing carbofuran tolerances (residue limits in food) due to unacceptable dietary risks,
especially to children, from consuming food or water alone or from a combination of
food and water with carbofuran residues. Following resolution of the tolerance
revocations, EPA plans to precede with cancellation of remaining carbofuran uses due to
unreasonable ecological and worker risks (Federal Register / Vol. 73, No. 245 /
December 19, 2008 /77690-77693).
Methomyl - The intent of registrants for voluntary cancellation of methomyl use on
strawberry and grapes were incorporated in the TV-methyl carbamate revised cumulative
risk assessment. With these and other mitigation measures for these individual
pesticides, EPA concluded that the cumulative risks to humans associated with the TV-
methyl carbamates are below EPA's regulatory level of concern
DRAFT 34
-------
DRAFT
(http://www.epa.gov/pesticides/cumulative/carbamate_risk_mgmt.htm). The FQPA does
not address cumulative risk of pesticides to aquatic resources.
Registration and uses of pesticide labels
containing the active ingredients (a.i.)
carbaryl, carbofuran, and methomyl
Metabolites of carbaryl, carbofuran,
methomyl
Degradates of carbaryl, carbofuran,
methomyl
Other ingredients in formulations
Label-recommended tank mixtures
Adjuvants/surfactants added to
formulations
Figure 1. Stressors of the Action
Mode of Action ofCarbamate Insecticides. NMC insecticides are neurotoxi cants to
the central and peripheral nervous systems of animals. Similar to other carbamate and
organophosphate (OP) insecticides, these a.i.s inhibit the enzyme acetylcholinesterase
(AChE) found in brain and muscle tissue of invertebrates and vertebrates. Thus, NMCs
belong to a class of insecticides known as AChE inhibitors. Inhibition of AChE results in
a build-up of the neurotransmitter, acetylcholine, which can lead to continued
stimulation. Normally, acetylcholine is broken down rapidly in the nerve synapse by
AChE. Chemical neurotransmission and communication are impaired when acetycholine
is not quickly degraded in animals which ultimately may result in a number of adverse
responses from physiological and behavioral modification to death. NMFS batched the
consultations on carbaryl, carbofuran, and methomyl into one Opinion because these
compounds have the same mechanism of action, i.e., they target the same site of action in
the exact same way. However, NMFS evaluated the effects of each a.i. independently.
DRAFT 35
-------
DRAFT
Additionally, cumulative exposure to the three a.i.s is expected given they have
overlapping uses and detections in surface water samples.
Active and Other Ingredients. Carbaryl, carbofuran, and methomyl are the a.i.s that
kill or otherwise affect targeted organisms (listed on the label). However, pesticide
products that contain these a.i.s also contain inert ingredients. Inert ingredients are
ingredients which EPA defines as not "pesticidally" active. EPA also refers to inert
ingredients as "other ingredients". The specific identification of the compounds that make
up the inert fraction of a pesticide is not required on the label. However, this does not
necessarily imply that inert ingredients are non-toxic, non-flammable, or otherwise non-
reactive. EPA authorizes the use of chemical adjuvants to make pesticide products more
efficacious. An adjuvant aides the operation or improves the effectiveness of a pesticide.
Examples include wetting agents, spreaders, emulsifiers, dispersing agents, solvents,
solubilizers, stickers, and surfactants. A surfactant is a substance that reduces surface
tension of a system, allowing oil-based and water-based substances to mix more readily.
A common group of non-ionic surfactants is the alkylphenol polyethoxylates (APEs),
which may be used in pesticides or pesticide tank mixes, and also in many common
household products. Nonylphenol (NP), one of the APEs, has been linked to endocrine-
disrupting effects to aquatic animals.
Formulations. Pesticide products come in a variety of solid and liquid formulations.
Examples of formulation types include dusts, dry flowables, emulsifiable concentrates,
granulars, solutions, soluble powders, ultra-low volume concentrates, water-soluble bags,
and powders. The formulation type can have implications for product efficacy and
exposure to humans and other non-target organisms.
Tank Mix. A tank mix is a combination by the user of two or more pesticide
formulations as well as any adjuvants or surfactants added to the same tank prior to
application. Typically, formulations are combined to reduce the number of spray
operations or to obtain better pest control than if the individual products were applied
alone. The compatibility section of a label may advise on tank-mixes known to be
DRAFT 36
-------
DRAFT
incompatible or on specific mixing instructions for use with compatible mixes. Labels
may also recommend specific tank-mixes. Pursuant to FIFRA, EPA has the discretion to
prohibit tank mixtures. Applicators are permitted to include any combination of
pesticides in a tank mix as long as each pesticide in the mixture is permitted for use on
the application site and the label does not explicitly prohibit the mix.
Pesticide Registration. The Pesticide Registration Improvement Act (PRIA) of 2003
became effective on March 23, 2004. The PRIA directed EPA to complete REDs for
pesticides with food uses/tolerances by August 3, 2006, and to complete REDs for all
remaining non-food pesticides by October 3, 2008. The goal of the reregi strati on
program is to mitigate risks associated with the use of older pesticides while preserving
their benefits. Pesticides that meet today's scientific and regulatory standards may be
declared "eligible" for reregi strati on. The results of EPA's reviews are summarized in
RED documents. EPA issued REDs for carbaryl and methomyl in 2007 and 1998,
respectively. The IRED for carbofuran was issued in August 2006. EPA considered the
registration eligibility determination for carbofuran complete upon issuance of the
cumulative assessment for the NMC pesticides in 2007
(http://www.epa.gov/pesticides/reregistration/REDs/carbofuran_red.pdf). Accordingly,
EPA treated the carbofuran IRED as its RED. The REDs for all three a.i.s include
various mitigation measures, including the cancelation of all registered uses of carbofuran
due to ecological and occupational risks. These mitigation components were considered
part of the proposed action.
Duration of the Proposed Action. EPA's goal for reassessing currently registered
pesticide active ingredients is every 15 years. Given EPA's timeframe for pesticide
registration reviews, NMFS' evaluation of the proposed action is also 15 years.
Interrelated and Interdependent Activities. No interrelated and interdependent
activities are associated with the proposed action.
DRAFT 37
-------
DRAFT
Registration Information of Pesticide Active Ingredients under Consultation. As
discussed above, the proposed action encompasses EPA's registration of the uses (as
described by product labels) of all pesticides containing carbaryl, carbofuran, or
methomyl. However, EPA did not provide copies of all product labels containing these
a.i.s. The following descriptions represent information acquired from review of a sample
of current product labels as well as information conveyed in the BEs, EPA REDs, and
other documents.
Carbaryl
Carbaryl, also known by the trade name Sevin, is an NMC insecticide which was first
registered in 1959 for use on cotton. In 2001, EPA identified the NMC insecticides as a
group which shares a common mechanism of toxicity. Therefore, EPA was required to
consider the cumulative effects on human health resulting from exposure to this group of
chemicals when considering whether to establish, modify, or revoke a tolerance for
pesticide residues in food, in accordance with the FQPA (EPA 2008).
Several regulatory documents concerning carbaryl were issued after EPA's BE of the
analysis of risk of carbaryl to threatened and endangered salmonids (EPA 2003). An
IRED for carbaryl that addressed the potential human health and ecological risks was
signed on June 30, 2003. EPA amended the IRED on October 22, 2004, to incorporate
clarifications and corrections, updated the residential risk assessment to reflect the
voluntary cancelation of the liquid broadcast use of carbaryl on residential turf to address
post-application risk to toddlers identified in the 2003 IRED, and addressed issues
regarding labeling of carbaryl formulations for mitigating potential hazards to bees. In
addition, mitigation measures required in the 2004 amended IRED included cancelation
of certain uses and application methods, reduction of application rates, application
prohibitions, personal protective equipment (PPE) and engineering control (EC)
requirements, and extension of restricted-entry intervals (REIs) for post-application
exposure (EPA 2008).
DRAFT 38
-------
DRAFT
EPA also issued generic and product-specific data call-ins (DCIs) for carbaryl in March
2005. The carbaryl generic DCI required several studies for the a.i. carbaryl, including
additional toxicology, worker exposure monitoring, and environmental fate data. The
product DCI required acute toxicity and product chemistry data for all pesticide products
containing carbaryl. In response to the 2005 DCIs, many carbaryl registrants chose to
voluntarily cancel their carbaryl products. Approximately 80% of all of carbaryl end-use
products registered at the time of the 2003 IRED have since been canceled through this
process or other voluntary cancelations (EPA 2008).
On September 26, 2007, EPA published a revised NMC cumulative risk assessment (EPA
2007), which concluded that the cumulative risks associated with the NMC pesticides
meet the safety standard set forth in the FFDCA. Concurrently, on September 26, 2007,
the RED for carbaryl was completed. The 2007 RED presents EPA's revised carbaryl
human health risk assessment under FQPA and EPA's final tolerance reassessment
decision for carbaryl. EPA amended the carbaryl RED in August of 2008. The
amendment updated the 2007 RED to reflect the Revised Occupational Exposure and
Risk Assessment, dated July 9, 2007 (EPA 2008).
The National Pesticide Information Retrieval System (NPIRS) website
(http://ppis.ceris.purdue.edu/htbin/epachem.com) suggests that there are currently 24
registrants with active registrations of 87 pesticide products containing carbaryl.
"Carbaryl is nationally registered for over 115 uses in agriculture, professional turf
management, ornamental production, and residential settings (EPA 2007). Carbaryl is
also registered for use as a mosquito adulticide.
(http://www.umass.edu/fniitadvisor/NEAPMG/145-149.pdf )(EPA 2007)."
Several product labels indicate carbaryl is commonly formulated with other a.i.s. For
example, there are active registrations of carbaryl products that also contain copper
sulfate, rotenone, malathion, captan, metaldehyde, and bifenthrin (NPIRS website).
According to EPA's BE, 26 carbaryl products are registered to individual states under
SLN provisions in Section 24(c) of FIFRA (EPA 2003). Section 24(c) registrations
DRAFT 39
-------
DRAFT
include control of shrimp in oyster beds in two tideland areas (Willapa Bay and Gray's
Harbor) in Washington and, in California, insecticidal use on fruits and nuts, prickly pear
cactus, ornamental plants, and non-food crops. Idaho and Oregon do not have any 24(c)
registrations for carbaryl (EPA 2003).
Usage Information.
The insecticide carbaryl is used in agriculture to control pests on terrestrial food crops
including fruit and nut trees, many types of fruit and vegetables, and grain crops; cut
flowers; nursery and ornamentals; turf, including production facilities; greenhouses; golf
courses; and in oyster beds. Carbaryl is also registered for use on residential sites (e.g.,
annuals, perennials, shrubs) by professional pest control operators and by homeowners on
gardens, ornamentals and turfgrass (EPA 2008). EPA estimated over 1.4 million pounds
(Ibs) of carbaryl are applied each year in agricultural crops and over 200,000 Ibs are
applied annually for turf, landscape, and horticultural uses in the U.S. (EPA 2008).
The California Department of Pesticide Regulation (CDPR) indicated approximately 150
- 250 thousand Ibs of carbaryl were applied annually in California between 2002 and
2006 based on agricultural and other "reportable uses" (CDPR 2007). The 1999 Oregon
Legislature authorized development of the Oregon Pesticide Use Reporting System
(PURS). In 2006, information on household pesticide use was collected through a
pesticide use survey. The first full year of collecting non-household pesticide use in
PURS was 2007. Over 37,000 Ibs of carbaryl were reported as applied in Oregon in 2007
(ODA 2008). Approximately 189,600 Ibs of carbaryl are used annually for agriculture in
Washington State (WSDA 2004). Similar data on pesticide use were not found for Idaho.
Examples of Registered Uses.
Agricultural Uses. Carbaryl is used on a myriad of crops. Examples of crops currently
proposed for continued carbaryl use and which are grown in areas with Pacific salmon
and steelhead include cranberries, cucumbers, beans, eggplant, grapefruit, grapes, hay,
lemons, lettuce, nectarines, olives, onions, oranges, parsley, peaches, peanuts, pears,
pecans, peppers, pistachios, plums, potatoes, prunes, pumpkins, rice, sod, spinach,
DRAFT 40
-------
DRAFT
squash, strawberries, sugar beets, sunflowers, sweet corn, sweet potatoes, tangelos,
tangerines, tomatoes, walnuts, watermelons, wheat (EPA 2008). Carbaryl is also used to
thin fruit in orchards to enhance fruit size and enhance repeat bloom.
Non-agricultural Uses. Carbaryl is used extensively by homeowners, particularly for
lawn care (EPA 2008). Examples of non-agricultural use sites include home and
commercial lawns, flowerbeds around buildings, recreation areas, golf courses, sod
farms, parks, rights-of-way, hedgerows, Christmas tree plantations, oyster beds, rural
shelter belts, and applications to control ticks, grasshoppers, and adult mosquitoes (EPA
2007). Carbaryl is also used for pet care (pet collars, powders and dip, in kennels, and on
pet sleeping quarters).
Examples of Registered Formulation Types Carbaryl products are manufactured as
granular, liquid, wettable powder, and dust formulations. All dry flowable (water
dispersible granule) products have been voluntarily cancelled. The use of dust
formulation in agriculture and backpack sprayers are not supported by Bayer
CropScience, the carbaryl technical registrant, who is amending its carbaryl registrations
to delete these uses (EPA 2008).
Methods and Rates of Application.
Methods. Groundboom, airblast, and aerial applications are typical for agricultural uses
of carbaryl. Other applications can also be made using handheld equipment, such as low
pressure hand wand sprayers, turf guns, and various ready-to-use products. Applications
by aerosol cans, hand, spoon, shaker can, and front- and back-mounted spreaders are
prohibited (EPA 2008).
Application Rates. The maximum single application rate allowed on the labels for
agricultural uses in California, Idaho, Oregon, and Washington is 16 Ib carbaryl/acre
(Table 1). Many agricultural uses allow repeated application of carbaryl at intervals of 7-
14 days. Application intervals are not specified for some uses (e.g., flower beds, home
DRAFT 41
-------
DRAFT
fruits, and vegetables). Additionally, some uses do not specify the maximum number of
applications (e.g., prickly pear, ticks, and grasshoppers).
Table 1. Registered uses and application rates for carbaryl in California, Idaho, Oregon,
and Washington (EPA 2007)
Use Site
Home Lawn
Fire ants
Flower beds around buildings
Lawns, recreation areas, golf
courses, sod farms, commercial
lawns
Parks
Citrus
Citrus
Olives
Almonds, chestnuts, pecans,
filberts, walnuts, pistachio
Flowers, shrubs
Apricot, cherries, nectarines,
peaches, plums, prunes
Apple, pear, crabapple, oriental
pear, loquat
Sweet corn
Caneberries, blueberries,
grapes, strawberries
Tomatoes, peppers, eggplant
Peanuts
Broccoli, cauliflower, cabbage,
kohlrabi, Chinese cabbage,
collards, kale, mustard greens,
brussel sprouts, hanover salad
Sweet potato
Field corn, pop corn
Leaf lettuce, head lettuce,
dandelion, endive, parsley,
spinach, swiss chard
Celery, garden beets, carrots,
horseradish, parsnip, rutabaga,
potato, salsify, root turnip, radish
Prickly pear
Rice
Fresh beans, dry beans, fresh
peas, dry peas, cowpeas, fresh
southern peas, soybeans
Sugar beets, pasture, grass for
seed
Alfalfa, birdsfoot trefoil, clover
Rangeland
Cucumber, melon, pumpkin,
squash, roses, other herbaceous
plants, woody plants
Application
Maximum Single
Application Rate
(Ib a.i./acre)
9.1
7.4
8
8
8
16
7.5
7.5
5
4.3
4 (5 dormant)
3
2
2
2
2
2
2
2
2
2
2
1.5
1.5
1.5
1.5
1
1
Maximum
Number of
Applications
2
2
4
2
1
1
8
2
4
3
3+1 dormant
8
8
5
7
5
4
8
4
5
6
As needed
2
4
2
1/cutting
1
6
Minimum
Application
Interval (days)
7
7
None
7
-
-
14
14
7
7
15
14
3
7
7
7
6
7
14
7
7
7
7
7
14
None
-
7
Maximum per
year
(Ib a.i./acre)
Not specified
Not specified
6
8
8
16
20
Not specified
15
Not specified
9 + 5 dormant
spray
Not specified
Not specified
Not specified
8
8
7
8
8
6
Not specified
6
4
6
3
Not specified
1
Not specified
DRAFT
42
-------
DRAFT
Use Site
CRP acreage, set-aside
acreage, rights-of-way,
hedgerows, ditch banks,
roadsides, wasteland
Non-urban forests, tree
plantations, Christmas trees,
parks, rangeland trees, rural
shelter belts
adult mosquitoes1"
Ticks
Grasshoppers
flax
Home fruits and vegetables
Proso millet, wheat
lentils
Oyster beds
Sunflower
Tobacco
Application
Maximum Single
Application Rate
(Ib a.i./acre)
1
1
1
2
1.5
1.5
1.95
1.5
1.5
8
1.5
2
Maximum
Number of
Applications
2
2
*
As needed
As needed
2
6
2
4
Not specified
2
4
Minimum
Application
Interval (days)
14
7
*
None
None
14
None
14
7
None
7
7
Maximum per
year
(Ib a.i./acre)
3
Not specified
Not specified
Not specified
Not specified
Not specified
12.1
Not specified
Not specified
Not specified
Not specified
Not specified
Number of applications and interval as specified for use site (pastures, rangeland, forests and wastelands,
etc.).
Metabolites and Degradates.
The major metabolite of carbaryl degradation by both abiotic and microbially mediated
processes is 1-naphthol. This degradate represented up to 67% of the applied carbaryl in
degradation studies. It is also formed in the environment by degradation of naphthalene
and other polyaromatic hydrocarbon compounds. EPA reports that only limited
information on the environmental transport and fate of 1-naphthol is available and
indicates this compound is less persistent and less mobile than the parent carbaryl (EPA
2003).
Carbofuran
Carbofuran is a NMC systemic pesticide first registered in the U.S. in 1969. The BE for
carbofuran indicates it is registered as a restricted use broad spectrum insecticide,
nematicide, and miticide for use on a wide variety of agricultural and non-agricultural
crops (EPA 2004). Carbofuran is classified as a restricted use pesticide and is formulated
into flowable, wettable powder, and granular forms. Through an agreement between
EPA and the technical registrant in 1991, granular carbofuran has been limited to the sale
DRAFT
43
-------
DRAFT
of 2,500 Ibs of a.i. per year in the U.S. since 1994, for use only on certain crops. Today
granular carbofuran is limited to use on spinach grown for seed, pine seedlings, bananas
(in Hawaii only), and cucurbits only (EPA 2006).
In the late 1990s, the technical registrant made a number of changes to labels in
order to reduce drinking water and ecological risks of concern. These included reducing
application rates and numbers of applications for alfalfa, cotton, corn, potatoes, soybeans,
sugarcane, and sunflowers. Numbers of applications were also restricted on some soils to
reduce groundwater concentrations (EPA 2006).
Several regulatory documents concerning carbofuran were issued after EPA's BE of the
analysis of risk of carbofuran to threatened and endangered salmonids (EPA 2004). An
IRED for carbofuran was published in August 2006. As previously indicated, EPA
concluded the NMC cumulative risk assessment in September 2007. All tolerance
reassessment and REDs for individual NMC pesticides were considered complete. The
carbofuran IRED, therefore is considered a completed RED.
The carbofuran IRED (EPA 2006), and draft Notice of Intent to Cancel Carbofuran
(January 2008) indicate EPA proposes cancellation of all uses of carbofuran, due to
ecological, occupational, and human dietary risks of concern from some crops.
Economic benefits are low to moderate for all of these uses, and do not outweigh the
risks (EPA 2006). There are several uses for which residues do not pose dietary risks of
concern and which have moderate benefits to growers [artichokes, chile peppers in the
Southwestern U.S., cucurbits (granular formulation only), spinach grown for seed,
sunflowers, and pine seedlings in the Southeastern U.S.]. For these uses, EPA is
allowing a four-year phase-out in order to allow time for new alternatives to become
available to growers (EPA 2006). However, EPA's effective date for final cancelation of
carbofuran is unknown at this time. For bananas, sugarcane, and coffee, however,
benefits to U.S. growers are low when compared to ecological and occupational risks
from domestic uses of these crops. Dietary risks of phase-out crops plus these imported
foods are below EPA's level of concern.
DRAFT 44
-------
DRAFT
Although EPA determined that all uses of carbofuran are ineligible for reregi strati on, use
of carbofuran will continue for an undetermined period of time. EPA has initiated
cancelation procedures for product uses of low economic benefits. The remaining uses
are subject to a four year phase-out. However, final cancellation may take several years
and the decision to cancel carbofuran registrations could be subject to legal challenges.
Additionally, EPA indicated that FMC wishes to retain registrations for six uses: corn,
potatoes, pumpkins, sunflowers, pine seedlings, and spinach grown for seed (Jones 2009).
FMC also proposed to phase out use of artichokes over two years. EPA plans to consider
FMC's proposal for the continued registration of carbofuran at a future date.
The IRED indicated there are currently one technical, two manufacturing-use, and six
end-use products registered under Section 3 of FIFRA. There are also 77 active SLN
registrations under Section 24(c) of FIFRA (EPA 2006). The NPIRS website suggests
one registrant holds nine active registrations of pesticide products containing carbofuran
(EPA registration numbers: 279-2712, 279-2862, 279-2874, 279-2876, 279-2922, 279-
3023, 279-3038, 279-3060, and 279-3310).
Usage Information
Nearly one million Ibs a.i. are applied annually from the application of liquid carbofuran
formulations (EPA 2006). The major use of liquid formulations of carbofuran is on corn,
alfalfa, and potatoes. Under the existing terms and conditions of the registration, sale of
the granular formulation is limited to 2,500 Ibs a.i. per year, and use is limited to pine
seedlings, cucurbits, bananas (in Hawaii only), and spinach grown for seed (EPA 2006).
Carbofuran use has decreased significantly in California over the last decade. CDPR
indicates agricultural uses of carbofuran exceeded 200,000 Ibs in 1996 and use declined
each preceding year with approximately 23,000 Ibs of carbofuran applied in California in
2006 (CDPR 2007). Multiyear use statistics for describing temporal trends of carbofuran
were unavailable for Idaho, Oregon, and Washington.
DRAFT 45
-------
DRAFT
Examples of Registered Uses.
Food Crops. Alfalfa, artichoke, banana, barley, coffee, corn (field, pop, and sweet),
cotton, cucurbits (cucumber, melons, and squash), grapes, oats, pepper, plantain, potato,
sorghum, soybean, sugar beet, sugarcane, sunflower, and wheat (EPA 2006).
Non-food uses. Agricultural fallow land, cotton, ornamental and/or shade trees,
ornamental herbaceous plants, ornamental non-flowering plants, ornamental woody
shrubs and vines, pine, spinach grown for seed, and tobacco (EPA 2006).
Examples of Registered Formulation Types.
Carbofuran is formulated into flowable, wettable powder, and granular forms. The
flowable formulation constitutes the vast majority of the carbofuran currently used (EPA
2004).
Examples of Approved Methods and Rates of Application.
Equipment. Carbofuran is applied by aerial equipment, chemigation systems,
groundboom sprayers, airblast sprayers, tractor-drawn spreaders, push-type granular
spreaders, and handheld equipment (EPA 2006).
Method and Rate. Carbofuran can be applied as a foliar or soil treatment. Maximum
single and seasonal application rates range from 0.002 to 10 Ibs a.i./acre, depending on
the application scenario (EPA 2006).
DRAFT 46
-------
DRAFT
Table 2. Registered uses and application rates for carbofuran in California, Idaho, Oregon,
and Washington (EPA 2004)
Use Site
Application
Maximum Single
Application Rate
(Ib a. i. /acre)
Maximum Number
of Applications
Minimum
Application
Interval (days)
Maximum per year
(Ib a. i. /acre)
Flowable Carbofuran - Section 3 registrations
Alfalfa, corn, cotton
Ornamentals
Pine seedlings
Potatoes
Small grains,
soybeans
Sugarcane
Sunflowers
Tobacco
1
0.06
0.05
1
0.25
0.75
1.4
6
1
Not specified
1
2
2
2
1
Not specified
-
Not specified
-
Not specified
Not specified
Not specified
-
Not specified
1
0.06
Prepare slurry: Add
0.05 Ibs a.L, 0.5
gallons water, and
2.0 Ibs clay; Slurry
sufficient to treat
roots of 150 to 200
seedlings.
2
0.5
1.5
1.4
6
Granular Carbofuran - Section 3 registrations
Bananas
Pine seedlings
Rice
0.006
0.002
0.5
2
Not specified
1
Not specified
Not specified
-
0.012
0.002
0.5
California- Flowable Carbofuran- Section 24C
Artichokes
Grapes
Ornamentals
1
10
10
2
1
Not specified
Not specified
-
Not specified
2
10
10
Idaho- Flowable Carbofuran- Section 24C
Potatoes
Sugar beets
3
2
2
1
Not specified
-
6
2
Oregon- Flowable Carbofuran- Section 24C
Potatoes
Nursery stock
Sugar beets
3
10
2
2
Not specified
1
Not specified
Not specified
-
6
10
2
Oregon- Granular Carbofuran- Section 24C
Watermelons
1
Not specified
Not specified
1
Washington- Flowable Carbofuran- Section 24C
Potatoes
3
2
Not specified
6
Washington- Granular Carbofuran- Section 24C
Spinach
(grown for seed)
1
1
-
1
Timing. Carbofuran is a contact insecticide applied at planting or post-planting (EPA
2006). The timing is variable among crops.
Metabolites and Degradates. The major transformation product of carbofuran in water
and aerobic aquatic metabolism is the hydrolysis product, carbofuran 7-phenol (EPA
DRAFT
47
-------
DRAFT
2006). It also appears as the transformation endpoint prior to conversion to CO2 and is
shorter lived in water than the parent. Other major expected environmental
transformation products in soils that have potential to reach the aquatic environment are
3-hydroxycarbofuran and 3-ketocarbofuran, which typically occur in small amounts (i.e.,
< 5.0 % of applied) and are relatively short lived as compared to the parent (EPA 2006).
Methomyl
Methomyl was first registered for use in the U.S. in 1968. Methomyl is currently
registered for use on a wide variety of sites including field, vegetable, and orchard crops;
turf (sod farms only); livestock quarters; commercial premises; and refuse containers
(EPA 2007). All methomyl products, except the 1% bait formulations, are classified as
restricted use pesticides (EPA 2007). A Registration Standard issued in April 1989
required additional testing, modified tolerances. It also required label modifications
related to applicator safety, re-entry intervals, and environmental hazards (EPA 2007).
Additional label modifications were required with the publication of the methomyl RED
in 1998 (EPA 1998).
EPA's BE of the analysis of risks of methomyl to threatened and endangered salmonids
indicated there were 10 end-use products registered under Section 3 of FIFRA (EPA
2003). The NPIRS website suggests that there are currently six registrants with active
registrations for nine products containing methomyl (EPA registration numbers: 270-
255, 352-342, 352-361, 352-366, 352-384, 2724-274, 7319-6, 53871-3, 5742-2).
Eighteen additional methomyl products are registered to individual states under SLN
provisions in Section 24(c) of FIFRA (EPA 2003). California has seven SLN for use to
control insects on ornamentals, beans, soybeans, radishes, sweet potatoes, Chinese
broccoli, broccoli raab, and pumpkins (EPA 2003). Idaho, Oregon, and Washington do
not have any SLNs for methomyl (EPA 2003). Methomyl also was previously registered
as a molluscicide to control snails and slugs and as a fungicide for control of blights, rots,
mildews and other fungal diseases. Those uses, as well as uses on ornamentals and in
greenhouses, have been canceled (EPA 2003).
DRAFT 48
-------
DRAFT
Usage Information
The BE indicated EPA has no recent national data on the amount of methomyl applied
annually (EPA 2003). According to the 1998 RED, an estimated 2.5 to 3.5 million Ibs of
methomyl a.i. were applied annually in the U.S. between 1987 and 1995. CDPR
indicates approximately 262 - 554 thousand Ibs of methomyl were applied annually in
California between 2000 and 2006 based on agricultural and other "reportable uses"
(CDPR 2007). Over 42,000 Ibs of methomyl were applied in Oregon in 2007 (ODA
2008). Similar data on pesticide use were not found for Idaho and Washington.
Examples of Registered Uses
Agriculture. Methomyl is used for a variety of agricultural uses including alfalfa, anise,
asparagus, barley, beans (succulent and dry), beets, Bermuda grass (pasture), blueberries,
broccoli, broccoli raab, Brussels sprouts, cabbage, carrot, cauliflower, celery, chicory,
Chinese broccoli, Chinese cabbage, collards (fresh market), corn (sweet), corn (field and
popcorn), corn (seed), cotton, cucumber, eggplant, endive, garlic, horseradish, leafy green
vegetables, lentils, lettuce (head and leaf), lupine, melons, mint, nonbearing nursery stock
(field grown), oats, onions (dry and green), peas, peppers, potato, pumpkin, radishes, rye,
sorghum, soybeans, spinach, sugar beet, summer squash, sweet potato, tomatillo, tomato,
turf (sod farms only), wheat, and orchards including apple, avocado, grapes, grapefruit,
lemon, nectarines, oranges, peaches, pomegranates, tangelo, and tangerine (EPA 2007).
Non-agriculture. Methomyl has several non-crop uses that are outside uses involving
scatter bait or bait station formulations including the following use sites: bakeries,
beverage plants, broiler houses, canneries, commercial dumpsters which are enclosed,
commercial use sites (unspecified), commissaries, dairies, dumpsters, fast food
establishments, feedlots, food processing establishments, hog houses, kennel, livestock
barns, meat processing establishments, poultry houses, poultry processing establishments,
restaurants, supermarkets, stables, and warehouses (EPA 2007).
DRAFT 49
-------
DRAFT
Examples of Registered Formulations and Types.
End-use formulations of methomyl include soluble concentrate, wettable powder,
granular, pelleted/tableted, and water soluble packaged. Products registered as fly baits
also contain (Z)-9-tricosene (0.04 to 0.26% a.i.) as an a.i.; labels note that these products
contain a sex attractant and feeding synergist (EPA 2003).
Examples of Approved Methods and Rates of Application.
Application Equipment. Methomyl can be applied by aircraft; bait box; brush; cup;
duster; glove; granule applicator; ground; high volume ground sprayer; low volume
ground sprayer; package applicator; scoop; shaker can; shaker jar; sprayer; and ultra low
volume sprayer (EPA 1998).
Application Rates. The maximum single application rate allowed on the labels for
agricultural uses in California, Idaho, Oregon, and Washington is 0.9 Ib a.i./acre. Many
agricultural uses allow repeated application of methomyl at relatively short intervals (1-5
days). For example, the application interval for methomyl for sweet corn is one day and
methomyl can be applied 28 times within a single crop of sweet corn. Additionally,
several crops of sweet corn may be grown per year in some locations within the action
area (EPA 2007). The maximum seasonal labeled application rates (indicated on the
label as maximum application rates per crop) for agricultural uses range from 0.9 Ib
a.i./acre/crop [i.e., Bermuda grass (pasture), avocado, lentils, beans (interplanted with
trees), sorghum, and soybeans (interplanted with trees)] to 7.2 Ibs a.i./acre/crop [i.e.,
cabbage, lettuce (head), cauliflower, broccoli raab, celery, and Chinese cabbage]. Several
methomyl crops can be grown more than one time per year (i.e., they have multiple crop
cycles). Therefore, for those methomyl uses that have more than one crop cycle per year,
the maximum allowable yearly application rate will be higher than the maximum
seasonal application rate. For perennial crops (e.g., alfalfa), the number of cuttings per
year was used to determine the number of crop cycles per year. Based on the labeled
application rates and information from EPA's OPP Benefits and Economic Analysis
Division (BEAD) on the number of times each crop for which methomyl is registered for
DRAFT 50
-------
DRAFT
use can be grown in California, the maximum yearly application rates for methomyl are
32.4 Ib a.i./acre/year (alfalfa) and 21.6 Ib a.i./acre/year (broccoli raab, cabbage, and
Chinese cabbage) for agricultural crops; 5.4 Ib a.i./acre/year (peaches) for orchards; and
0.22 Ib a.i./acre/application for nonagricultural uses (no maximum application/acre/year
is provided on the nonagri cultural use labels). All orchard and most agricultural uses
involve foliar application. The only granular agricultural/orchard use is for corn which
also has a foliar use (EPA 2007).
All non-agricultural outside uses for methomyl in California, Idaho, Oregon, and
Washington are limited to scatter baits and bait stations around agricultural (e.g., animal
premises) and commercial structures and commercial dumpsters, where children or
animals are not likely to contact the pesticide. The scatter bait can also be mixed with
water to form a paste which can be brushed onto walls, window sills, and support beams.
The maximum application rate for the scatter bait use is 0.22 Ib a.i./acre (0.0025 Ib
2
a.i./500 ft ). However, it is unlikely that applications would involve a full acre as the
outside use of the scatter bait is limited to areas around structures and dumpsters. No
minimum application interval or maximum application rate per year is provided on the
scatter bait labels (EPA 2007).
Table 3. Examples of registered uses and application rates for methomyl in California,
Idaho, Oregon, and Washington (EPA 2007).
Use Site
Commercial dumpsters; poultry houses;
unspecified commercial sites; outside
commercial uses: feedlots, dairies, stables,
broiler houses, hog houses, livestock barns,
meat processing establishments, poultry
processing establishments, beverage plants,
canneries, food processing establishments,
kennels, dumpsters, restaurants, supermarkets,
commissaries, and bakeries.
Alfalfa
Asparagus
Avocado
Barley
Beans
Broccoli
Cabbage
Application
Maximum Single
Application Rate
(Ib a.i./acre)
0.22
0.9
0.9
0.9
0.45
0.9
0.9
0.9
Maximum
Number of
Applications
Not specified
10x9 crops
5
2
4
10
10x3 crops
15x3 crops
Minimum
Application
Interval
(days)
1-3
5
7
5
5
5
5
2
Maximum per
year
(Ib a.i./acre)
Not specified
32.4
4.5
0.9
1.8
4.5
21.6
21.6
DRAFT
51
-------
DRAFT
Use Site
Corn
Lettuce
Onions
Spinach
Turf
Application
Maximum Single
Application Rate
(Ib a.i./acre)
0.45
0.9
0.9
0.9
0.9
Maximum
Number of
Applications
28 x 3 crops
8x2 crops
8x3 crops
8x3 crops
4x2 crops
Minimum
Application
Interval
(days)
1
2
5
5
5
Maximum per
year
(Ib a.i./acre)
18.9
14.4
16.2
10.8
7.2
Timing. The timing of application is dependent on use, but may occur throughout the
year. In most cases multiple applications are allowed to maintain pest control.
Applications occur on fruit crops during the bloom, petal fall, pre-bloom, and leaf stages
and when pest pressure is highest on a "When Needed" basis. On corn, application may
occur during the whorl/foliar stages. With other crops, application is during the foliar or
leaf stages of the crop (EPA 1998).
Metabolites and Degradates
Several degradates and metabolites have been identified for methomyl. The major
degradate in most metabolism studies was CO2. Another degradate, S-methyl-N-
hydroxythioacetamidate, which is highly mobile, appears primarily as a product of
alkaline hydrolysis. In an aquatic metabolism study, methomyl degraded with estimated
half-lives of four to five days. After seven days, acetonitrile comprised a maximum of
17% and acetamide up to 14% of the amount of methomyl applied. After 102 days,
volatilized acetonitrile totaled up to 27% of the applied and CO2up to 46% of the applied
material (EPA 2003).
Species
EPA's BEs considered the effects of carbaryl, carbofuran, and methomyl to 26 species of
listed Pacific salmonids and their designated critical habitat. EPA determined that
carbaryl, carbofuran, and methomyl may affect most of these species. Exceptions follow:
EPA concluded that the registration of carbaryl products would have no effect on
Northern California steelhead, Southern Oregon/Northern California Coastal coho, Hood
Canal summer-run chum, and Ozette Lake sockeye. EPA also concluded that the
DRAFT
52
-------
DRAFT
registration of carbaryl products would not likely adversely affect California Coastal
Chinook and Puget Sound Chinook salmon.
EPA concluded that the registration of carbofuran products would have no effect on
Northern California Steelhead, Central California Coast coho, and California Coastal
Chinook salmon. EPA also concluded that the registration of carbofuran products would
not likely adversely affect Central Valley Spring-run Chinook salmon, Lower Columbia
River Chinook salmon, Puget Sound Chinook salmon, Sacramento River winter-run
Chinook salmon, Upper Willamette River Chinook salmon, Columbia River chum, Hood
Canal summer-run chum, Oregon Coast coho, Southern Oregon/Northern California
Coast coho, Ozette Lake sockeye, Snake River sockeye, Central California Coast
steelhead, California Central Valley steelhead, Lower Columbia River steelhead, Snake
River steelhead, South-Central California steelhead, Southern California steelhead, and
Upper Willamette River steelhead.
EPA concluded that the registration of methomyl products would have no effect on
Northern California steelhead and California Coastal Chinook salmon.
Although EPA has determined that its action in registering pesticides containing the three
active ingredients is not likely to adversely affect certain ESUs/DPSs and will have no
effect on others, EPA initiated formal consultation on its action because EPA concluded
that its action may adversely affect other listed ESUs/DPSs. When an action agency
concludes that its action will not affect any listed species or critical habitat, then no
section 7 consultation is necessary (USFWS and NMFS 1998). If NMFS concurs with a
federal agency that its action is not likely to adversely affect any listed species or critical
habitat, then formal consultation is not required. Since formal consultation was triggered,
NMFS evaluated the federal action and its impacts to all listed Pacific, anadromous
salmonids and their designated critical habitat. In this Opinion, NMFS will analyze the
impacts to all ESUs/DPSs of Pacific salmonids present in the action area, including those
salmonid species identified by EPA as being unaffected or not likely to be adversely
affected including two species of salmonid listed after EPA provided its BEs.
DRAFT 53
-------
DRAFT
to
Overview of NMFS'Assessment Framework
NMFS uses a series of steps to assess the effects of federal actions on endangered and
threatened species and designated critical habitat. The first step of our analysis identifies
those physical, chemical, or biotic aspects of proposed actions that are likely to have
individual, interactive, or cumulative direct and indirect effects on the environment (we
use the term "potential stressors" for these aspects of an action). As part of this step, we
identify the spatial extent of any potential stressors and recognize that the spatial extent
of those stressors may change with time. The spatial extent of these stressors is the
"action area" for a consultation.
The second step of our analyses identifies the listed resources (endangered and threatened
species and designated critical habitat) that are likely to occur in the same space and at
the same time as these potential stressors. If we conclude that such co-occurrence is
likely, we then try to estimate the nature of co-occurrence (these represent our exposure
analyses'). In this step of our analysis, we try to identify the number, age (or life stage),
gender, and life history of the individuals that are likely to be exposed to an action's
effects and the populations or subpopulations those individuals represent.
Once we identify which listed resources are likely to be exposed to potential stressors
associated with an action and the nature of that exposure, in the third step of our analysis
we examine the scientific and commercial data available to determine whether and how
those listed resources are likely to respond given their exposure (these represent our
response analyses). We integrate the exposure and response analyses to assess the risk to
listed individuals and their habitat from the stressors of the action (these represent our
Risk characterization analyses). NMFS' analysis is ultimately a qualitative assessment
that draws on a variety of quantitative and qualitative tools and measures to address risk
to listed resources.
DRAFT 54
-------
DRAFT
In the final steps of our analyses, we establish the risks posed to listed species and to
designated critical habitat. Our jeopardy determinations for listed species must be based
on an action's effects on the continued existence of threatened or endangered species as
those "species" have been listed, which can include true biological species, subspecies, or
distinct population segments of vertebrate species. Because the continued existence of
listed species depends on the fate of the populations that comprise them, the viability
(that is, the probability of extinction or probability of persistence) of listed species
depends on the viability of the populations that comprise the species. Similarly, the
continued existence of populations are determined by the fate of the individuals that
comprise them; populations grow or decline as the individuals that comprise the
population live, die, grow, mature, migrate, and reproduce (or fail to do so).
The structure of our risk analyses reflects the relationships between listed species, the
populations that comprise each species, and the individuals that comprise each
population. Our risk analyses begin by identifying the probable risks actions pose to
listed individuals that are likely to be exposed to an action's effects. Our analyses then
integrates those individual-level effects to identify consequences to the populations those
individuals represent. Our analyses conclude by determining the consequences of those
population-level risks to the species those populations comprise.
We evaluate risks to listed individuals by measuring the individual's "fitness" defined as
changes in an individual's growth, survival, annual reproductive success, or lifetime
reproductive success. In particular, we examine the scientific and commercial data
available to determine if an individual's probable response to an action's effect on the
environment (which we identify in our Response Analyses) are likely to have
consequences for the individual's fitness.
Reductions in abundance, reproduction rates, or growth rates (or increase variance in one
or more of these rates) of the populations those individuals represent is a necessary
condition for reductions in a population's viability, which is itself a necessary condition
for reductions in a species' viability. On the other hand, when listed plants or animals
DRAFT 55
-------
DRAFT
exposed to an action's effects are not expected to experience reductions in fitness, we
would not expect that action to have adverse consequences on the viability of the
population those individuals represent or the species those populations comprise (Mills
and Beatty 1979; Stearns 1982; Anderson, Phillips et al. 2006). If we conclude that listed
species are not likely to experience reductions in their fitness, we would conclude our
assessment because an action that is not likely to affect the fitness of individuals is not
likely to jeopardize the continued existence of listed species.
If, however, we conclude that listed plants or animals are likely to experience reductions
in their fitness, our assessment determines if those fitness reductions are likely to be
sufficient to reduce the viability of the populations those individuals represent (measured
using changes in the populations' abundance, reproduction, spatial structure and
connectivity, growth rates, or variance in these measures to make inferences about the
population's extinction risks). In this step of our analyses, we use the population's base
condition (established in the Status of Listed Resources and Environmental Baseline
sections of this Opinion) as our point of reference. Finally, our assessment determines if
changes in population viability are likely to be sufficient to reduce the viability of the
species those populations comprise. In this step of our analyses, we use the species'
status (established in the Status of Listed Resources section of this Opinion) as our point
of reference.
Evidence Available for the Consultation
We search, compile and use a variety of resources to conduct our analyses including:
• EPA's BEs, REDs, IREDS, other documents developed by EPA
• Peer-reviewed literature
• Gray literature
• Books
• Available pesticide labels
• Any correspondence (with EPA or others)
• Available monitoring data and other local, county, and state information
• Pesticide registrant generated data
• Online toxicity databases (PAN, EXTOXNET, ECOTOX, USGS, NPIC)
• Pesticide exposure models run by NMFS
• Population models run by NMFS
DRAFT 56
-------
DRAFT
• Information and data provided by the registrants identified as applicants
• Comments on the draft Opinion from EPA and any applicants
• Incident reports
Collectively, this information provided the basis for our determination as to whether and
to what degree listed resources under our jurisdiction are likely to be exposed to EPA's
action and whether and to what degree the EPA can ensure that its authorization of
pesticides is not likely to jeopardize the continued existence of threatened and
endangered species or is not likely to result in the destruction or adverse modification of
designated critical habitat.
Application of Approach in this Consultation
The EPA proposes to authorize the use of over 100 pesticide formulations (pesticide
products) containing the a.i.s carbaryl, carbofuran, and methomyl through its authority to
register pesticides under the FIFRA. Registration by EPA authorizes the use of these
formulations in the U.S. and its territories, documented by EPA's approval of registrant-
derived pesticide labels. Pursuant to the court's 2002 order in Washington Toxics
Coalition v. EPA, EPA initiated consultation on registration of carbaryl, carbofuran, and
methomyl for 26 listed ESUs of Pacific salmonids. Since EPA initiated consultation,
NMFS has listed one additional Pacific coho ESU and one additional Pacific steelhead
DPS. This Opinion represents NMFS' evaluation of whether EPA's authorization of
these labels satisfies EPA's obligations to listed salmonids pursuant to section 7(a)(2) of
the ESA.
The NMFS evaluates whether endangered species, threatened species, and designated
critical habitat are likely to be exposed to the direct and indirect effects of the proposed
action. If those listed resources are not likely to be exposed to these activities, we would
conclude that EPA's action is not likely to jeopardize the continued existence of
threatened species, endangered species, or result in the destruction or adverse
modification of designated critical habitat under NMFS' jurisdiction. If, however, listed
individuals are likely to be exposed to these actions and individual fitness is reduced,
then we evaluate the potential for population level consequences.
DRAFT 57
-------
DRAFT
A Viable Salmonid Population (VSP) is an independent population of any Pacific
salmonid that has a negligible risk of extinction due to threats from demographic
variation, local environmental variation, and genetic diversity changes over a 100-year
time frame (McElhaney, Ruckleshaus et al. 2000). The independent population is the
fundamental unit of evaluation in determining the risk of extinction of salmon in an ESU.
Attributes or metrics associated with a VSP include the abundance, productivity, spatial
structure, and genetic diversity of the population. Abundance is defined as the size of the
population and can be expressed in a number of ways, e.g., the number of spawning
adults, the number of adults surviving to recruit to fisheries, or the number of emigrating
smolts. Abundance is a vital measure, as smaller populations run a greater risk of
extinction. The second VSP measure is productivity, generally defined as the growth rate
of a population. This Opinion discusses productivity in terms of lambda (X). Appendix
II contains a more detailed explanation of X in the context of our population models. The
spatial structure of a population is inherently dependant on the quantity and quality of
available habitat. A limited spatial structure can hamper the ability of the ESU to
respond to evolutionary pressures. Genetic variability within the ESU gives the species
the ability to respond to short-term stochastic events, as well as to evolve to a changing
environment in the long-term. These VSP parameters provide an indication of the
population's capacity to adapt to various environmental conditions and ability to be self-
sustaining in the natural environment (McElhaney, Ruckleshaus et al. 2000; McElhaney,
Chilcote et al. 2007).
In determining the effect of an action to populations, we first address whether individual
fitness level consequences are likely and whether those consequences affect populations.
We evaluate whether identified VSP parameters of populations such as abundance and
productivity are reduced by individual fitness effects. If populations are likely to be
adversely affected by reductions in VSP parameters, we analyze the potential effects to
the species as a whole. In parallel, if designated critical habitats are likely to be exposed
and PCEs are adversely affected, then we evaluate the potential for reductions in the
conservation value of the habitats. We devise risk hypotheses based on identified PCEs
DRAFT 58
-------
DRAFT
that are potentially affected by the stressors of the action. If the best available data
indicate that PCE-specific risk hypotheses are supported, then we discuss whether critical
habitat will remain functional to serve the intended conservation role for the species with
the Conclusion section.
General conceptual framework for assessing risk ofEPA's pesticide actions to listed
resources.
We evaluate the risk to listed species and designated critical habitat in the Effects of the
Proposed Action section by applying an ecological risk assessment framework that
organizes the available information in three phases: problem formulation, analysis, and
risk characterization (EPA 1998). We adapted the EPA framework to address ESA-
specific considerations (Figure 2). The framework follows a process for organizing,
evaluating, and synthesizing the available information on listed resources and the
stressors of the action. Below, we briefly describe each phase in the Effects of the
Proposed Action section.
Problem Formulation
The first phase of the framework is problem formulation. In this phase, we generate
conceptual models from our initial evaluation of the relationships between stressors of the
action (pesticides and identified chemical stressors and potential receptors (listed species,
habitat). We represent these relationships in conceptual models presented as diagrams
and written risk hypotheses (EPA 1998). Conceptual model diagrams are constructed to
illustrate potential pesticide exposure pathways and associated listed resources'
responses. An example of a conceptual model is presented in Figure 3 for Pacific
salmonids. In it, we illustrate where the pesticides generally reside in the environment
following application, how pesticides may co-occur with listed species and their habitats,
and how the individuals/habitat may respond upon exposure to them. In the case of
Pacific salmonids, we ascribe exposure and response to specific life stages of individuals
and then assess individual fitness endpoints sensitive to the action's stressors.
DRAFT 59
-------
DRAFT
Exposure Analysis
Stressors of the Action
(see Figure 1)
Response Analysis
Co-occurrence of pesticide
products and geographic
range of ESA-listed species
i
r
Distribution of
individuals
i
r
1
r
Analyses based
on the best
scientific and
commercial data
available on
pesticide
Distribution of products use,
habitat transport, fate,
•\
r
Exposure Profile
species ecology
Effects of pesticide products
on ESA-listed species and
their habitat
i
r
Individual
responses
/\
i
r
1
r
Habitat
responses
i
r
Response Profile
Effects on individuals
Effects on populations
Effects on species
(ESU or DPS)
Analyzed within the
context of the
Environmental Baseline
(including multiple
stressors such as
temperature and
environmental mixtures of
pesticides); the Status of
the Species; and
Cumulative Effects
I
Can EPA ensure its action is
not likely to jeopardize the
continued existence of the
species?
Effects on habitat
Effects on primary
constituent elements
Effects on conservation
value of designated
habitat
I
Can EPA ensure its action is
not likely to adversely
modify or destroy the
designated critical habitat?
Figure 2. Conceptual framework for assessing risks of EPA's action to listed resources
DRAFT
60
-------
DRAFT
Stressors of
the Action
Environmental
Matrices
carbaryl,
degradates,
metabolites
carbofuran,
degradates,
metabolites
methomyl,
degradates,
metabolites
other chemicals
in formulated
products
tank
mixtures
adjuvant
Exposure
Responses
Life stage
responses
exposure to
terrestrial
invertebrates
(salmon prey)
Interactions with
water quality
stressors in
environmental
baseline:
- other carbamate
and OP
insecticides
-temperature
-pH
Figure 3. Exposure pathways to carbaryl, carbofuran, and methomyl and general
responses of listed Pacific salmonids and habitat.
Species Risk Hypotheses
We construct risk hypotheses by identifying biological requirements or assessment
endpoints (Table 4) for listed resources in the action area. We designate assessment
endpoints as those biological properties of species and their habitat essential for
successful completion of a species life cycle. We integrate the listed resources
information with what is known about the stressors of the action, including their physical
properties, use, presence in aquatic habitats, and their toxicity. We then evaluate how
listed salmonids and their habitat are potentially affected by the stressors of the action
and integrate this information with exposure information to develop risk hypotheses.
Below are the risk hypotheses (written as affirmative statements) we evaluate in the
Effects of the Proposed Action section:
DRAFT
61
-------
DRAFT
1. Exposure to carbaryl, carbofuran, and methomyl is sufficient to:
a. Kill salmonids from direct, acute exposure;
b. Reduce salmonid survival through impacts to growth;
c. Reduce salmonid growth through impacts on the availability and quantity
of salmonid prey;
d. Impair swimming which leads to reduced growth (via reductions in
feeding), delayed and interrupted migration patterns, survival (via reduced
predator avoidance), and reproduction (reduced spawning success); and
e. Reduce olfactory-mediated behaviors resulting in consequences to
survival, migration, and reproduction.
2. Exposure to mixtures of carbaryl, carbofuran, and methomyl can act in
combination to increase adverse effects to salmonids and salmonid habitat.
3. Exposure to other stressors of the action including degradates, adjuvants, tank
mixtures, and other active and other ingredients in pesticide products containing
carbaryl, carbofuran, and methomyl cause adverse effects to salmonids and their
habitat.
4. Exposure to other pesticides present in the action area can act in combination with
carbaryl, carbofuran, and methomyl to increase effects to salmonids and their
habitat.
5. Exposure to elevated temperatures can enhance the toxicity of the stressors of the
action.
Critical Habitat Risk Hypotheses
1. Exposure to the stressors of the action is sufficient to reduce abundance of aquatic
prey items of salmonids.
2. Exposure to the stressors of the action is sufficient to degrade water quality in
designated critical habitat.
Below we discuss an example of one risk hypothesis to show the relationship between
assessment endpoints and measures with species responses. In risk hypothesis 1 (d),
aquatic exposure to carbaryl, carbofuran, and methomyl can impair a salmonid's nervous
system and consequently affect swimming ability offish. Behavioral modifications, such
as changes in swimming performance, are regularly considered in NMFS' Opinions.
Swimming performance therefore is an assessment endpoint. Measurable changes in
swimming speed are the assessment measure used to evaluate this endpoint. Reductions
in swimming performance could also affect other assessment endpoints such as migration
and predator avoidance. We may or may not have empirical data that address these
endpoints, resulting in a recognized data gap. This uncertainty would be identified
during the problem formulation phase, and discussed in the risk characterization phase.
DRAFT 62
-------
DRAFT
In the problem formulation phase, we also identify the toxic mode and mechanism of
action of chemical stressors, particularly for the pesticide a.i.s. This information helps us
understand what an organism's physiological consequences may be following exposure.
It also helps us evaluate whether mixture toxicity occurs because we identify other
pesticides that share similar modes of action and the likelihood for co-occurrence in listed
species habitats. A similar mode of action with other pesticides is a key determinant of
the likelihood of mixture toxicity. With vertebrates (fish and mammals) and
invertebrates, the three a.i.s share a common mode and mechanism of action,
acetylcholinesterase inhibition. Given this information, a range of potential adverse
responses are possible (Figure 4). We then search, compile, and review the available
toxicity information to ascertain which physiological systems are known to be affected
and to what degree.
Figure 4. Physiological systems potentially affected by acetylcholinesterase inhibition
Chemoreception
Central nervous
System
Cardiovascular
system
Growth and
metabolism
Locomotion
Autonomic
nervous system
Reproduction
Feeding and
digestion
DRAFT
63
-------
DRAFT
In Table 4, assessment endpoints and assessment measures are identified for particular
life stages. We focused on the following physiological systems identified in Figure 4:
chemoreception, locomotion, feeding, reproduction, and growth. We did not locate any
information on the remaining systems in Figure 4. Thus, they were not specifically
addressed in our analysis.
We assess the likelihood of these fitness level consequences occurring from exposure to
the action. In the exposure analysis (Figure 2), we select exposure estimates for our
listed resources derived from reviewing the available exposure data. Depending on the
chemicals being evaluated, data may or may not be available for all endpoints and
measures, and available data may vary in reliability. Thus, we use a weight-of-evidence
approach in this Opinion.
The problem formulation phase as articulated in EPA's 1998 Guidelines for Conducting
Ecological Risk Assessment concludes with the development of an analysis plan. In this
Opinion, the Approach to the Assessment section is the general analysis plan. This
section identifies how exposure will be assessed and which assessment endpoints will be
evaluated. Therefore, the Approach to the Assessment is a road map for evaluating the
effects of EPA's registration actions with carbaryl, carbofuran, and methomyl.
DRAFT 64
-------
DRAFT
Table 4. Examples of salmonid lifestage assessment endpoints and measures
Salmonid Life Stage
Assessment Endpoint
(individual fitness)
Assessment Measure
(measures of changes in individual fitness)
Egg*
* Is the egg permeable to pesticides
(measured by pesticide concentrations
in eggs)?
Development
Survival
size, hatching success, morphological deformities
viability (percent survival)
Alevin (yolk-sac fry)
Respiration
Swimming:
predator avoidance
site fidelity
Yolk-sac utilization:
growth rate
size at first feeding
Development
Survival
gas exchange, respiration rate
swimming speed, orientation, burst speed
predator avoidance assays
rate of absorption, growth
weight and length
weight and length
morphology, histology
LC50 (dose-response slope). Percent dead at a
given concentration
Fry, Juvenile, Smolt
First exogenous feeding (fry)- post yolk-
sac absorption
Survival
Growth
Feeding
Swimming:
predator avoidance behavior
migration
use of shelter
Olfaction:
kin recognition
predator avoidance
imprinting
feeding
Smoltification (smolt)
Development
time to first feeding, starvation
LC50 (dose-response slope). Percent dead at a
given concentration
weight, length
stomach contents, weight, length, starvation, prey
capture rates
swimming speed, orientation, burst swimming
speed
predator avoidance assays
swimming rate, downstream migration
fish monitoring, bioassays
electro-olfactogram measurements,
behavioral assays
behavioral assays
behavioral assays
behavioral assays
Na/K ATPase activity, sea water challenge tests
length, weight, malformations
Returning adult
Survival
Feeding
Swimming:
predator avoidance
migration
spawning
feeding
Sexual development
Olfaction:
Predator avoidance
Homing
Spawning
LC50 (dose-response slope). Percent dead at a
given concentration
stomach contents
behavioral assays
numbers of adult returns, behavioral assays
numbers of eggs fertilized
stomach contents
histological assessment of ovaries/testis
electro-olfactogram measurements,
measurements of intersex
behavioral assays
behavioral assays
behavioral assays
DRAFT
65
-------
DRAFT
Risk Characterization
We follow the framework presented in Figure 2 to conduct the analysis and risk
characterization phases. First we conduct exposure and response analyses to
estimate/determine the type, likelihood, magnitude, and frequency of adverse responses
resulting from predicted exposure based on the best available information. We evaluate
species information and pesticide information to determine when, where, and at what
concentrations listed salmonids and their habitat may be exposed. We then correlate
those exposure estimates with probable response based on available toxicity data. Once
we have conducted the analysis phase, we move to the risk characterization phase (Figure
2)
In the risk characterization phase, we revisit the risk hypotheses and apply tools to
address whether any individual fitness consequences assessed in the analysis phase would
be expected to impact populations and ultimately species. One of the tools we employ is
individual-based population models predicated on a juvenile salmonids' probability of
survival in its first year of life. We also assess interactions between the stressors of the
action and stressors in the Environmental Baseline (Figure 2). Some pesticides' toxicity
profiles are influenced by environmental parameters such as pH and temperature.
Temperature can affect pesticide metabolism in fish and is seasonally elevated in many
salmonid supporting watersheds. As described earlier in this section we translate
expected effects to identified PCEs by evaluating the available information to support
risk hypotheses. If we expect PCEs to be reduced we discuss whether the expected
reductions translate to reductions in the conservation value of designated critical habitat.
To conclude consultation, cumulative effects are described and the extent to which
species and habitat are affected is documented. Cumulative effects as defined in 50CFR
§404.2 include the effects of future, state, tribal, local, or private actions that are
reasonably certain to occur in the action area of this Opinion. Integrating the Effects of
the Proposed Action, the Status of Listed Resources, and the Environmental Baseline,
NMFS determines whether EPA's pesticide registration action jeopardizes the continued
DRAFT 66
-------
DRAFT
existence of the species. NMFS also determines whether the action results in the
destruction or adverse modification of designated critical habitat.
Other Considerations
In this Opinion, we evaluated lines of evidence constructed as species-specific risk
hypotheses to ensure relevant endpoints were addressed. Ultimately, our analysis weighs
each line of evidence by evaluating the best commercial and scientific data available that
pertain to a given risk hypothesis. Overall, the analysis is a qualitative approach that uses
some quantitative tools to provide examples of potential risks to listed salmonids and
their habitat. Multiple methods and tools currently exist for addressing contaminant-
induced risk to the environment. Hazard-based assessments, probabilistic risk assessment
techniques, combinations of the two, and deterministic approaches such as screening
level assessments have been applied to questions of risk related to human health and the
environment.
In recent pesticide risk assessments, probabilistic techniques have been used to evaluate
the probability of exceeding a "toxic" threshold for aquatic organisms by combining
pesticide monitoring data with species sensitivity distributions (Geisy, Solomon et al.
1999). There is utility in information generated by probabilistic approaches if supported
by robust data. We compared the species sensitivity distributions presented in Giddings
2009 with the probability distributions of salmonid prey acute lethality values that we
developed to highlight differences in outcomes. The assessment with carbaryl did not
address many of the species-specific risk hypotheses. We found no other probabilistic
assessments that addressed risk to salmonids affected by short-term sublethal exposures,
mixtures, or affects on growth from reduced feeding ability and reduced abundances of
prey.
NMFS considered the use of probabilistic risk assessment techniques for addressing risk
at population and species (ESU and DPS) scales for the stressors of the action. However,
we encountered significant limitations in available data regarding toxicity information
and species information, and pesticide monitoring data. Examples of these limitations
DRAFT 67
-------
DRAFT
include issues with data collection, paucity of data, non-normal distributions of data, and
quality assurance and quality control. When these types of data limitations are coupled
with the inherent complexity of EPA's proposed action (Figure 1) in California, Idaho,
Oregon, and Washington, we find that probabilistic assessments at population and species
scales introduce an unquantifiable amount of uncertainty that undermines confidence in
derived risk estimates. These same studies do not factor the status of the existing health
and baseline conditions of the environment into their assessment. At this time, the best
available data do not support such an analysis and conclusions from such an analysis
would be highly speculative. However, in the risk characterization section, probabilistic
techniques were applied to salmonid prey items to determine selection of a survival
toxicity value used in population modeling exercises. NMFS did not locate any
ecological risk assessments on carbofuran and methomyl that used probabilistic
approaches to address risk to aquatic communities.
The action area is defined as all areas to be affected directly or indirectly by the federal
action and not merely the immediate area involved in the action (50 CFR §402.02).
Given EPA's nationwide authorization of these pesticides, the action area would
encompass the entire U.S. and its territories. These same geographic areas would include
all listed species and designated critical habitat under NMFS jurisdiction.
In this instance, as a result of the 2002 order in Washington Toxics Coalition v. EPA,
EPA initiated consultation on its authorization of 37 pesticide a.i.s and their effects on
listed Pacific salmonids under NMFS' jurisdiction and associated designated critical
habitat in the states of California, Idaho, Oregon, and Washington. Consequently, for
purposes of this Opinion, the action area consists of the entire range and most life history
stages of listed salmon and steelhead and their designated critical habitat in California,
Idaho, Oregon, and Washington. The action area encompasses all freshwater, estuarine,
marsh, swamps, nearshore, and offshore marine surface waters of California, Oregon, and
Washington. The action area also includes all freshwater surface waters in Idaho (Figure
5).
DRAFT 68
-------
DRAFT
Carbaryl, carbofuran, and methomyl are the second set of three insecticides identified in
the consultation schedule established in the settlement agreement and are analyzed in this
Opinion. NMFS' analysis focuses only on the effects of EPA's action on listed Pacific
salmonids in the above-mentioned states. It includes the effects of these pesticides on the
recently listed Lower Columbia River coho salmon, Puget Sound steelhead, and Oregon
Coast coho salmon. The Lower Columbia River coho salmon was listed as endangered in
2005. The Puget Sound steelhead and the Oregon Coast coho salmon were listed as
threatened in 2007 and 2008, respectively.
EPA's consultation remains incomplete until it analyzes the effects of its authorization of
pesticide product labels with carbaryl, carbofuran, and methomyl for all remaining
threatened and endangered species under NMFS' jurisdiction. EPA must ensure its
action does not jeopardize the continued existence or result in the destruction or adverse
modification of critical habitat for other listed species and designated critical habitat
under NMFS' jurisdiction throughout the U.S. and its territories.
DRAFT 69
-------
DRAFT
San Francisco
CA
Figure 5. Map showing extent of inland action area with the range of all ESU and DPS boundaries for ESA
listed salmonids highlighted in gray.
DRAFT
70
-------
DRAFT
Status of Listed Resources
NMFS has determined that the following species and critical habitat designations may
occur in this action area for EPA's registration of carbaryl, carbofuran, and methomyl -
containing products (Table 2). More detailed information on the status of these species
and critical habitat are found in a number of published documents including recent
recovery plans, status reviews, stock assessment reports, and technical memorandums.
Many are available on the Internet at http://www.nmfs.noaa.go/pr/species/.
Table 5. Listed Species and Critical Habitat (denoted by asterisk) in the Action Area
Common Name (Distinct Population Segment or Evolutionarily
Significant Unit)
Scientific Name
Status
Chinook salmon (California Coastal*)
Chinook salmon (Central Valley Spring-run*)
Chinook salmon (Lower Columbia River*)
Chinook salmon (Upper Columbia River Spring-run*)
Chinook salmon (Puget Sound*)
Chinook salmon (Sacramento River Winter-run*)
Chinook salmon (Snake River Fall-run*)
Chinook salmon (Snake River Spring/Summer-run*)
Chinook salmon (Upper Willamette River*)
Chum salmon (Columbia River*)
Chum salmon (Hood Canal Summer-run*)
Coho salmon (Central California Coast*)
Coho salmon (Lower Columbia River)
Coho salmon (Southern Oregon & Northern California Coast*)
Coho salmon (Oregon Coast*)
Sockeye salmon (Ozette Lake*)
Sockeye salmon (Snake River*)
Steelhead (Central California Coast*)
Steelhead (California Central Valley*)
Steelhead (Lower Columbia River*)
Steelhead (Middle Columbia River*)
Steelhead (Northern California*)
Steelhead (Puget Sound)
Steelhead (Snake River*)
Steelhead (South-Central California Coast*)
Steelhead (Southern California*)
Steelhead (Upper Columbia River*)
Steelhead (Upper Willamette River*)
Oncorhynchus tshawytscha
Oncorhynchus keta
Oncorhynchus kisutch
Oncorhynchus nerka
Oncorhynchus mykiss
Threatened
Threatened
Threatened
Endangered
Threatened
Endangered
Threatened
Threatened
Threatened
Threatened
Threatened
Endangered
Threatened
Threatened
Threatened
Threatened
Endangered
Threatened
Threatened
Threatened
Threatened
Threatened
Threatened
Threatened
Threatened
Threatened
Threatened
Threatened
DRAFT
71
-------
DRAFT
The following brief narratives summarize the biology and ecology of threatened and
endangered species in the action area that are relevant to the effects analysis in this
Opinion. Summaries of the status and trends [including (VSP) information] of each
species are presented to provide a foundation for the analysis.
One of the important factors defining a viable population is the population's long- and
short-term tendency to increase in abundance. In our status reviews of each listed
salmonid species, we calculated the median annual population growth rate (denoted as
lambda, X) from available time series of abundance for individual populations. The
lambda for each population is calculated using the rate at which four year running sums
of available abundance estimates changes through time. Several publications provide a
detailed description of the calculation of lambda (McClure, Holmes et al. 2003; Good,
Waples et al. 2005). The lambda values for salmonid VSPs presented in these papers are
summarized in Appendix 2. Unfortunately, reliable time series of abundance estimates
are not available for most Pacific salmon and steelhead populations. In those cases, we
made general inferences of long-term change based on what is known of historical and
past abundances from snapshot surveys, surveys of a population segments, harvest by
commercial and recreational fisheries, and professional judgment. We then compare
these to similar information of current populations.
Below, each species narrative is followed by a description of its critical habitat with
particular emphasis on any essential features of the habitat that may be exposed to the
proposed action, and may warrant special attention.
Chinook Salmon
Description of the Species
Chinook salmon are the largest of the Pacific salmon and historically ranged from the
Ventura River in California to Point Hope, Alaska in North America, and in northeastern
Asia from Hokkaido, Japan to the Anadyr River in Russia (Healey 1991). In addition,
Chinook salmon have been reported in the Canadian Beaufort Sea (McPhail and Lindsey
1970). We discuss the distribution, life history, diversity (when applicable), status, and
critical habitat of the nine species of endangered and threatened Chinook salmon
separately.
Of the Pacific salmon species, Chinook salmon exhibit one of the most diverse and
complex life history strategies. Chinook salmon are generally described as one of two
races, within which there is substantial variation. One form, the "stream-type" resides in
freshwater for a year or more following emergence from gravel nests. Another form, the
"ocean-type" migrates to the ocean within their first year. The ocean-type typifies
DRAFT 72
-------
DRAFT
populations north of 56°N (Healey 1991). Within each race, there is often variation in
age at seaward migration, age of maturity, timing of spawning migrations, male
precocity, and female fecundity.
Status and Trends
Over the past few decades, the size and distribution of Chinook salmon populations have
declined because of natural phenomena and human activity. Geographic features, such as
waterfalls, pose natural barriers to salmon migrating to spawning habitat. Flooding can
eliminate salmon runs and significantly alter large regions of salmon habitat. However,
these threats are not considered as serious as several anthropogenic threats. Of the
various natural phenomena that affect most populations of Pacific salmon, changes in
ocean productivity are generally considered most important. Natural variations in
freshwater and marine environments have substantial effects on the abundance of salmon
populations.
Salmon along the U.S. west coast are prey for a variety of predators, including marine
mammals, birds, sharks, and other fishes. In general, Chinook salmon are prey for
pelagic fishes, birds, and marine mammals, including harbor seals, sea lions, and killer
whales. Chinook salmon are also exposed to high rates of natural predation, during
freshwater rearing and migration stages, as well as during ocean migration. There have
been recent concerns that the increasing size of tern, seal, and sea lion populations in the
Pacific Northwest may have reduced the survival of some salmon ESUs. Human
activities include the operation of hydropower systems, over-harvest, hatcheries, and
habitat degradation including poor water quality from chemical contamination.
Chinook salmon are dependent on the quantity and quality of aquatic habitats. Juvenile
salmonids rely on a variety of non-main channel habitats that are critical to rearing. All
listed salmonids use shallow, low flow habitats at some point in their life cycle.
Examples of off-channel habitat include alcoves, channel edge sloughs, overflow
channels, backwaters, terrace tributaries, off-channel dredge ponds, and braids (Anderson
1999; Swift III 1979). Chinook salmon, like the other salmon NMFS has listed, have
declined under the combined effects of overharvests in fisheries; competition from fish
raised in hatcheries and native and non-native exotic species; dams that block their
migrations and alter river hydrology; gravel mining that impedes their migration and
alters the hydrogeomorphology of the rivers and streams that support juveniles; water
diversions that deplete water levels in rivers and streams; destruction or degradation of
riparian habitat that increase water temperatures in rivers and streams sufficient to reduce
the survival of juvenile Chinook salmon; and land use practices (logging, agriculture,
urbanization) that destroy or alter wetland and riparian ecosystems. These activities and
features introduce sediment, nutrients, biocides, metals, and other pollutants into surface
DRAFT 73
-------
DRAFT
and ground water and degrade water quality in the freshwater, estuarine, and coastal
ecosystems throughout the Pacific Northwest.
Salmonids along the west coast of the U.S. share common threats. Therefore,
anthropogenic threats for all species and stocks are summarized here (see (NMFS 2005b)
for a review). Population declines have resulted from several human-mediated causes.
However, the greatest negative influence has been the establishment of waterway
obstructions such as dams, power plants, and sluiceways for hydropower, agriculture,
flood control, and water storage. These structures have blocked salmon migration to
spawning habitat or resulted in direct mortality and have eliminated entire salmon runs.
Presently, many of these structures have been re-engineered, renovated, or removed to
allow for surviving runs to access former habitat. However, success has been limited.
Remaining freshwater habitats are threatened from development along waterways as well
as sedimentation, pollution run-off, habitat modification, and erosion. These factors can
directly cause mortality, affect salmonid health, or modify spawning habitat so as to
reduce reproductive success. Immature salmonids remain in freshwater systems and may
be exposed to these modifications for years. These conditions reduce juvenile survival.
Salmonids are also a popular commercial resource and have faced significant pressure
from fishing. Although currently protected, illegal oceanic driftnet gear is suspected of
hindering salmon survival and recovery. Despite the protection of weaker salmonid
stocks from fishing, exploitation of more populous stocks may actually harm weaker
stocks. Hatchery-reared salmon have been and are still being introduced to bolster
stocks. However, the broader effects of this action are unknown.
California Coastal Chinook Salmon
Distribution
California Coastal (CC) Chinook salmon includes all naturally-spawned coastal Chinook
salmon spawning from Redwood Creek south through the Russian River as shown in
(Figure 6).
DRAFT 74
-------
DRAFT
124 W
123 W
122 W
41 N-
40 N-
39 N-
'.» H-
Legend
W Dams
NPDES permit sites
EPA 303(d) rivers
-41 N
-40 N
-39 N
•38 N
123 W
0 15 30
60
Prepared by Dwayne Meadows
I Kilometers 15 July 2008
Figure 6. CC Chinook salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7.
DRAFT
75
-------
DRAFT
Legend
Land Cover Class
Barren Land
Cultivated Crops
Deciduous Forest
Developed, Higti Intensity
Developed, Low Intensity
Developed, Medium Intensity
[ ] Developed, Open Space
[__J Emergent Heftoaceuous Wetlands
|m Evergreen Forest
[_ J Hay/Pasture
Q^j Hertjaceuous
[_~ ] Mixed Forest
Open Water
[ ] Perennial Snow/Ice
Figure 7. Legend for the Land Cover Class categories found in species distribution
maps. Land cover is based on the 2001 National Land Cover Data and
classifications, http://www.mrlc.gov/index.php.
DRAFT
76
-------
DRAFT
CC Chinook salmon are a fall-run, ocean-type fish. Although a spring-run (river-type)
component existed historically, it is now considered extinct (Bjorkstedt, Spence et al.
2005). Table 6 identifies populations within the CC Chinook salmon ESU, their
abundances, and hatchery input.
Table 6. CC Chinook salmon—preliminary population structure, abundances, and
hatchery contributions (Good, Waples et al. 2005).
Population
Eel River
(includes * tributaries below)
Mainstem Eel River*
Van Duzen River*
Middle Fork Eel River*
South Fork Eel River*
North Fork Eel River*
Upper Eel River*
Redwood Creek
Mad River
Bear River
Mattole River
Russian River
Humbolt Bay tributaries
Tenmile to Gualala coastal
effluents
Small Humboldt County rivers
Rivers north of Mattole River
Noyo River
Historical
Abundance
17,000-55,000
13,000
2,500
13,000
27,000
Unknown
Unknown
1,000-5,000
1,000-5,000
100
1,000-5,000
50-500
40
Unknown
1,500
600
50
Most Recent
Spawner
Abundance
156-2,730
Inc. in Eel River
Inc. in Eel River
Inc. in Eel River
Inc. in Eel River
Inc. in Eel River
Inc. in Eel River
Unknown
19-103
Unknown
Unknown
200,000
120
Unknown
Unknown
Unknown
Unknown
Hatchery
Abundance
Contributions
-30%
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
0
Unknown
0
Unknown
-0%
40 (33%)
0
0
0
0
Total 20_,750-72,550 200,175_|min]i
Status and Trends
CC Chinook salmon were listed as threatened on September 16, 1999 (64 FR 50393).
Their classification was reaffirmed following a status review on June 28, 2005 (70 FR
37160). The outcome was based on the combined effect of dams that prevent individuals
from reaching spawning habitat, logging, agricultural activities, urbanization, and water
DRAFT
77
-------
DRAFT
withdrawals in the river drainages that support CC Chinook salmon. Historical estimates
of escapement, based on professional opinion and evaluation of habitat conditions,
suggest abundance was roughly 73,000 in the early 1960s with the majority offish
spawning in the Eel River [see CDFG 1965 in (Good, Waples et al. 2005)]. The species
exists as small populations with highly variable cohort sizes and discussion is underway
to split Eel River salmon into as many as five separate populations (see Table 3). The
Russian River probably contains some natural production. However, the origin of those
fish is unclear as a number of introductions of hatchery fish occurred over the last
century. The Eel River contains a substantial fraction of the remaining Chinook salmon
spawning habitat for this species.
Since the original listing and status review, little new data are available or suitable for
analyzing trends or estimating changes in the Eel River population's growth rate (Good,
Waples et al. 2005). Historical and current abundance information indicates that
independent populations of Chinook salmon are depressed in many of those basins where
they have been monitored.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52630).
The critical habitat designation for this ESU identifies PCEs that include sites necessary
to support one or more Chinook salmon life stages. Specific sites include freshwater
spawning sites, freshwater rearing sites, freshwater migration corridors, nearshore marine
habitat, and estuarine areas. The physical or biological features that characterize these
sites include water quality and quantity, natural cover, forage, adequate passage
conditions, and floodplain connectivity. Critical habitat in this ESU consists of limited
quantity and quality summer and winter rearing habitat, as well as marginal spawning
habitat. Compared to historical conditions, there are fewer pools, limited cover, and
reduced habitat complexity. The limited instream cover that does exist is provided
mainly by large cobble and overhanging vegetation. Instream large woody debris,
needed for foraging sites, cover, and velocity refuges is especially lacking in most of the
streams throughout the basin. NMFS has determined that these degraded habitat
conditions are, in part, the result of many human-induced factors affecting critical habitat.
They include dam construction, agricultural and mining activities, urbanization, stream
channelization, water diversion, and logging.
DRAFT 78
-------
DRAFT
Central Valley Spring-Run Chinook Salmon
Distribution
The Central Valley spring-run (CV) Chinook salmon includes all naturally spawned
populations of spring-run Chinook salmon in the Sacramento River and its tributaries in
California (Figure 8).
Table 7 identifies populations within the CV spring-run Chinook salmon ESU, their
abundances, and hatchery input.
Table 7. CV Chinook salmon—preliminary population structure, abundances, and
hatcherycxmtri^^
Population
Butte Creek Spring-run Chinook
Deer Creek Spring-run Chinook
Mill Creek Spring-run Chinook
Total
Historical
Abundance
-700,000 for all
populations
Most Recent
Spawner
Abundance
67-4,513
243-1,076
203-491
513-6,080
Hatchery
Abundance
Contributions
Unknown
Unknown
Unknown
Unknown
Life History
CV Chinook salmon enter the Sacramento River from March to July and spawn from late
August through early October, with a peak in September. Spring-run fish in the
Sacramento River exhibit an ocean-type life history, emigrating as fry and sub-yearlings.
Chinook salmon require cool freshwater while they mature over the summer. This
species tends to take advantage of high flows. Adult upstream migration may be blocked
by temperatures above 21°C (McCullough 1999). Temperatures below 21°C can stress
fish by increasing their susceptibility to disease (Berman 1990) and elevating their
metabolism (Brett 1979).
DRAFT 79
-------
DRAFT
41 N'
40'N'
Dams
NPDES permit sites
EPA 303(d) rivers
Migratory corridor
37 N
123 W
0 20 40
Prepared by Dwayne Meadows
Kilometers is July 2008
Figure 8. CV Chinook salmon distribution.
DRAFT
80
-------
DRAFT
Status and Trends
CV Chinook salmon were listed as threatened on September 16, 1999 (64 FR 50393).
This classification was retained following a status review on June 28, 2005 (70 FR
37160). The species was listed because dams isolated individuals from most of their
historic spawning habitat and the remaining habitat is degraded. Historically, spring-run
Chinook salmon were predominant throughout the CV. This species occupied the upper
and middle reaches (1,000 to 6,000 ft) of the San Joaquin, American, Yuba, Feather,
Sacramento, McCloud and Pit Rivers. Smaller populations occurred in most tributaries
with sufficient habitat for over-summering adults (Stone 1874; Rutter 1904; Clarke
1929).
The CV drainage as a whole is estimated to have supported spring-run Chinook salmon
runs as large as 700,000 fish between the late 1880s and the 1940s (Brown, Moyle et al.
1994). Before construction of Friant Dam, nearly 50,000 adults were counted in the San
Joaquin River alone (Fry 1961). Following the completion of Friant Dam, the native
population from the San Joaquin River and its tributaries (i.e., the Stanislaus and
Mokelumne Rivers) was extirpated. Spring-run Chinook salmon no longer exist in the
American River due to the operation of Folsom Dam. Naturally spawning populations of
CV Chinook salmon currently are restricted to accessible reaches of the upper
Sacramento River, Antelope Creek, Battle Creek, Beegum Creek, Big Chico Creek, Butte
Creek, Clear Creek, Deer Creek, Feather River, Mill Creek, and Yuba River (CDFG
1998). Since 1969, the CV Chinook salmon ESU (excluding Feather River fish) has
displayed broad fluctuations in abundance ranging from 25,890 in 1982 to 1,403 in 1993
(CDFG unpublished data).
The average abundance for the ESU was 12,499 for the period of 1969 to 1979, 12,981
for the period of 1980 to 1990, and 6,542 for the period of 1991 to 2001. In 2003 and
2004, total run size for the ESU was 8,775 and 9,872 adults, respectively. These
averages are well above the 1991 to 2001 average.
Evaluating the ESU as a whole, however, masks significant changes that are occurring
among populations that comprise the ESU. For example, the mainstem Sacramento River
population has undergone a significant decline while the abundance of many tributary
populations increased. Average abundance of Sacramento River mainstem spring-run
Chinook salmon recently declined from a high of 12,107 for the period 1980 to 1990, to a
low of 609 for the period 1991 to 2001 (Good et al. 2005). Meanwhile, the average
abundance of Sacramento River tributary populations increased from a low of 1,227 to a
high of 5,925 over the same periods.
According to Good et al. (2006), abundance time series data for Mill, Deer, Butte, and
DRAFT 81
-------
DRAFT
Big Chico creeks spring-run Chinook salmon (updated through 2001) confirm that
population increases seen in the 1990s have continued. During this period, habitat
improvements included the removal of several small dams and increases in summer flows
in the watersheds, a reduced ocean fisheries, and a favorable terrestrial and marine
climate. All three spring-run Chinook populations in the CV have long-and short-term
lambdas >1, indicating population growth. CV Chinook salmon have some of the highest
population growth rates in the CV. However, population sizes are relatively small
compared to fall-run Chinook salmon populations. Finally, Feather River hatchery and
Feather River spring-run Chinook salmon are not closely related to the Mill, Deer, and
Butte creek spring-run Chinook salmon populations.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52630).
The critical habitat designation for this ESU identifies PCEs that include sites necessary
to support one or more Chinook salmon life stages. Specific sites include freshwater
spawning sites, freshwater rearing sites, freshwater migration corridors, nearshore marine
habitat, and estuarine areas. The physical or biological features that characterize these
sites include water quality and quantity, natural cover, forage, adequate passage
conditions, and floodplain connectivity. Factors contributing to the downward trends in
this ESU include: loss of most historical spawning habitat, reduced access to
spawning/rearing habitat behind impassable dams, climatic variation, water management
activities, hybridization with fall-run Chinook salmon, predation, and harvest. Additional
factors include the degradation and modification of remaining rearing and migration
habitats in the natal stream, the Sacramento River, and the Sacramento delta. The natal
tributaries have many small hydropower dams and water diversions that in some years
have greatly reduced or eliminated in-stream flows during spring-run migration periods.
Problems in the migration corridor include unscreened or inadequately screened water
diversions, predation by nonnative species, and excessively high water temperatures.
Collectively, these factors have impacted spring-run Chinook salmon critical habitat and
population numbers (CDFG 1998). Several actions have been taken to improve and
increase the PCEs of critical habitat for spring-run Chinook salmon, including improved
management of CV water (e.g., through use of CALFED EWA and CV Project
Improvement Act (b)(2) water accounts), implementing new and improved screen and
ladder designs at major water diversions along the mainstem Sacramento River and
tributaries, removal of several small dams on important spring-run Chinook salmon
spawning streams, and changes in ocean and inland fishing regulations to minimize
harvest. Although protective measures and critical habitat restoration likely have
contributed to recent increases in spring-run Chinook salmon abundance, the ESU is still
below levels observed from the 1960s through 1990. Threats from hatchery production
(i.e., competition for food between naturally spawned and hatchery fish, and run
DRAFT 82
-------
DRAFT
hybridization and homogenization), climatic variation, reduced stream flow, high water
temperatures, predation, and large scale unscreened water diversions persist.
Lower Columbia River Chinook Salmon
Distribution
Lower Columbia River (LCR) Chinook salmon includes all naturally-spawned
populations of Chinook salmon from the Columbia River and its tributaries from its
mouth at the Pacific Ocean upstream to a transitional point between Oregon and
Washington, east of the Hood River and the White Salmon River (Figure 7). Naturally
spawned populations also occur along the Willamette River to Willamette Falls, Oregon,
exclusive of spring-run Chinook salmon in the Clackamas River (Table 5). The Cowlitz,
Kalama, Lewis, White Salmon, and Klickitat Rivers are the major river systems on the
Washington side, and the lower Willamette and Sandy Rivers are foremost on the Oregon
side. The eastern boundary for this species occurs at Celilo Falls, which corresponds to
the edge of the drier Columbia Basin Ecosystem. Historically, Celilo Falls may have
been a barrier to salmon migration at certain times of the year. Table 8 identifies
populations within the LCR Chinook salmon ESU, their abundances, and hatchery input.
DRAFT 83
-------
45 N-
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
-46 N
-45 N
Prepared by Dwayne Meadows
Kilometers 15 July 2008
Figure 9. LCR Chinook salmon distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
84
-------
DRAFT
Table 8. LCR Chinook salmon - preliminary population structure, abundances, and
^^
Population
Youngs Bay
Grays River
Big Creek
Elochoman River
Clatskanie River
Mill, Abernathy, and German Creeks
Scappoose Creek
Coweeman River
Lower Cowlitz River
Upper Cowlitz River (fall run)
Toutle River (fall run)
Kalama River (fall run)
Salmon Creek and Lewis River
Clackamas River
Washougal River
Sandy River (fall run)
Columbia Gorge-lower tributaries
Columbia Gorge-upper tributaries
Hood River (fall run)
Big White Salmon River
Sandy River (late fall run)
Lewis River-North Fork
Upper Cowlitz River (spring run)
Cispus River
Tilton River
Toutle River (spring run)
Kalama River (spring run)
Lewis River
Sandy River (spring run)
Big White Salmon River (spring run)
Hood River (spring run)
Total
Historical
Abundance
Unknown
2,477
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
4,971
Unknown
53,956
25,392
47,591
Unknown
7,518
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
2,901
4,178
Unknown
Unknown
Unknown
Unknown
Most Recent
Spawner
Abundance
Unknown
99
Unknown
676
Unknown
734
Unknown
274
1,562
5,682
Unknown
2,931
256
40
3,254
183
Unknown
Unknown
18
334
504
7,841
Unknown
1,787
Unknown
Unknown
98
347
Unknown
Unknown
51
26,273imin)
Hatchery
Abundance
Contributions
Unknown
38%
Unknown
68%
Unknown
47%
Unknown
0%
62%
Unknown
Unknown
67%
0%
Unknown
58%
Unknown
Unknown
Unknown
Unknown
21%
3%
13%
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
DRAFT
85
-------
DRAFT
Life History
LCR Chinook salmon display three life history types including early fall runs, late fall
runs, and spring-runs. Spring and fall runs have been designated as part of a LCR
Chinook salmon ESU. The predominant life history type for this species is the fall-run.
Fall Chinook salmon enter freshwater typically in August through October to spawn in
large river mainstems. The juvenile life history stage emigrates from freshwater as sub-
yearling (ocean-type). Spring Chinook salmon enter freshwater in March through June to
spawn in upstream tributaries and generally emigrate from freshwater as yearlings
(stream-type).
Status and Trends
LCR Chinook salmon were originally listed as threatened on March 24, 1999 (64 FR
14308). This status was reaffirmed on June 28, 2005 (70 FR 37160). Historical records
of Chinook salmon abundance are sparse. However, cannery records suggest a peak run
of 4.6 million fish [43 million Ibs see (Lichatowich 1999)] in 1883. Although fall-run
Chinook salmon occur throughout much of their historical range, they remain vulnerable
to large-scale hatchery production, relatively high harvest, and extensive habitat
degradation. The Lewis River late fall Chinook salmon population is the healthiest and
has a reasonable probability of being self-sustaining. Abundances largely declined
during 1998 to 2000. Trend indicators for most populations are negative, especially if
hatchery fish are assumed to have a reproductive success equivalent to that of natural-
origin fish.
New data acquired for the Good et al. (2006) report includes spawner abundance
estimates through 2001, new estimates of the fraction of hatchery spawners, and harvest
estimates. In addition, estimates of historical abundance have been provided by the
Washington Department of Fish and Wildlife (WDFW). The Willamette/Lower
Columbia River Technical Review Team (W/LCRTRT) has estimated that 8-10 historic
populations have been extirpated, most of them spring-run populations. Almost all of the
spring-run Chinook of LCR Chinook are at very high risk of extinction. Near loss of that
important life history type remains an important concern. Although some natural
production currently occurs in 20 or so populations, only one exceeds 1,000 spawners.
Most LCR Chinook salmon populations have not seen increases in recent years as
pronounced as those that have occurred in many other geographic areas.
According to Good et al. (2006), the majority of populations for which data are available
have a long-term trend of <1; indicating the population is in decline. Currently, the
spatial structures of populations in the Coastal and Cascade Fall Run major population
groups (MPGs) are similar to their respective historical conditions. The genetic diversity
of the Coastal, Cascade, and Gorge Fall Run MPGs (i.e., all except the Late Fall Run
DRAFT 86
-------
DRAFT
Chinook salmon MPG) has been eroded by large hatchery influences and periodically by
low effective population sizes. Hatchery programs for spring Chinook salmon are
preserving the genetic legacy from populations that were extirpated from blocked areas.
High hatchery production also poses genetic and ecological risks to natural populations
and masks their performance.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52630).
Designated critical habitat includes all Columbia River estuarine areas and river reaches
proceeding upstream to the confluence with the Hood Rivers as well as specific stream
reaches in a number of tributary subbasins. The critical habitat designation for this ESU
identifies PCEs that include sites necessary to support one or more Chinook salmon life
stages. Specific sites include freshwater spawning sites, freshwater rearing sites,
freshwater migration corridors, nearshore marine habitat, and estuarine areas. The
physical or biological features that characterize these sites include water quality and
quantity, natural cover, forage, adequate passage conditions, and floodplain connectivity.
Of 52 subbasins reviewed in NMFS' assessment of critical habitat for the LCR Chinook
salmon ESU, 13 subbasins were rated as having a medium conservation value, four were
rated as low, and the remaining subbasins (35), were rated as having a high conservation
value to LCR Chinook salmon. Factors contributing to the downward trends in this ESU
are hydromorphological changes resulting from hydropower development, loss of tidal
marsh and swamp habitat, and degraded freshwater and marine habitat from industrial
harbor and port development, and urban development. Limiting factors identified for this
species include: (1) Habitat degradation and loss due to extensive hydropower
development projects, urbanization, logging, and agriculture on Chinook spawning and
rearing habitat in the LCR, (2) reduced access to spawning/rearing habitat in tributaries,
(3) hatchery impacts, (4) loss of habitat diversity and channel stability in tributaries, (5)
excessive fine sediment in spawning gravels, (6) elevated water temperature in
tributaries, (7) harvest impacts, and (8) poor water quality.
Upper Columbia River Spring-run Chinook Salmon
Distribution
Endangered Upper Columbia River (UCR) spring-run Chinook salmon includes stream-
type Chinook salmon that inhabit tributaries upstream from the Yakima River to Chief
Joseph Dam (Figure 10). The UCR spring-run Chinook salmon is composed of three
major population groups (MPGs): the Wenatchee River population, the Entiat River
population, and the Methow River population. These same populations currently spawn
DRAFT 87
-------
DRAFT
in only three river basins above Rock Island Dam: the Wenatchee, Entiat, and Methow
Rivers. Several hatchery populations are also listed including those from the Chiwawa,
Methow, Twisp, Chewuch, and White rivers, and Nason Creek (Table 9). Table 9
identifies populations within the UCR Chinook salmon ESU, their abundances, and
hatchery input.
Table 9. UCR Chinook salmon - preliminary population structure, abundances,
and hatchery contributions (Good, Waples et al. 2005).
Population
Methow River
Twisp River
Chewuch River
Lost/Early River
Entiat River
Wenatchee River
Chiwawa River
Nason Creek
Upper Wenatchee River
White River
Little Wenatchee River
Total
Historical
Abundance
-2,100
Unknown
Unknown
Unknown
-380
-2,400
Unknown
Unknown
Unknown
Unknown
Unknown
~4,880Jmm)
Most Recent
Spawner
Abundance
79-9,904
10-369
6-1,105
3-164
53-444
119-4,446
34-1,046
8-374
0-215
1-104
3-74
Hatchery
Abundance
Contributions
59%
54%
41%
54%
42%
42%
47%
39%
66%
8%
21%
DRAFT
-------
124 W
121-W
120 W
119-W
118'W
•49 N
Legend
Dams
NPDES permit sites
EPA303(d) rivers
Migratory corridor
119W
new
0 30 60
120
Figure 10. UCR Spring-run Chinook salmon distribution.
DRAFT
Prepared by Dwayne Meadov/s
Kilometers is juiv 2008
89
-------
DRAFT
Life History
UCR spring Chinook salmon begin returning from the ocean in the early spring. They
enter the upper Columbia tributaries from April through July, with the run into the
Columbia River peaking in mid-May. After migration, UCR spring Chinook salmon hold
in freshwater tributaries until spawning occurs in the late summer, peaking in mid- to late
August. Juvenile spring Chinook salmon spend a year in freshwater before emigrating to
salt water in the spring of their second year.
Status and Trends
UCR spring-run Chinook salmon were listed as endangered on March 24, 1999 (64 FR
14308). This listing was reaffirmed on June 28, 2005 (70 FR 37160) based on a
reduction of UCR spring-run Chinook salmon to small populations in three watersheds.
Based on redd count data series, spawning escapements for the Wenatchee, Entiat, and
Methow rivers have declined an average of 5.6%, 4.8%, and 6.3% per year, respectively,
since 1958.
In the most recent five-year geometric mean (1997 to 2001), spawning escapements were
273 for the Wenatchee population, 65 for the Entiat population, and 282 for the Methow
population. These numbers represent only 8% to 15% of the minimum abundance
thresholds. However, escapement increased substantially in 2000 and 2001 in all three
river systems. Based on 1980-2004 returns, the average annual population growth rate,
lambda, for this ESU is estimated at 0.93 (meaning the population is not replacing itself)
(Fisher and Hinrichsen 2006). Assuming that population growth rates were to continue at
1980-2004 levels, UCR spring-run Chinook salmon populations are projected to have
very high probabilities of decline within 50 years. Population viability analyses for this
species suggest that these Chinook salmon face a significant risk of extinction: a 75 to
100% probability of extinction within 100 years (given return rates for 1980 to present).
Finally, the Interior Columbia Basin Technical Recovery Team (ICBTRT) characterizes
the diversity risk to all UCR spring Chinook populations as "high". The high risk is a
result of reduced genetic diversity from homogenization of populations that occurred
under the Grand Coulee Fish Maintenance Project in 1939-1943. Straying hatchery fish,
and a low proportion of natural-origin fish in some broodstocks and a high proportion of
hatchery fish on the spawning grounds have also contributed to the high genetic diversity
risk.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52630).
Designated critical habitat includes all Columbia River estuarine areas and river reaches
proceeding upstream to Chief Joseph Dam and several tributary subbasins. The critical
habitat designation for this ESU also identifies PCEs that include sites necessary to
DRAFT 90
-------
DRAFT
support one or more Chinook salmon life stages. Specific sites include freshwater
spawning sites, freshwater rearing sites, freshwater migration corridors, nearshore marine
habitat, and estuarine areas. The physical or biological features that characterize these
sites include water quality and quantity, natural cover, forage, adequate passage
conditions, and floodplain connectivity. The UCR spring-run Chinook salmon ESU has
31 watersheds within its range. Five watersheds received a medium rating and 26
received a high rating of conservation value to the ESU. The Columbia River
rearing/migration corridor downstream of the spawning range was rated as a high
conservation value. Factors contributing to the downward trends in this ESU include:
(1) Mainstem Columbia River hydropower system mortality, (2) tributary riparian
degradation and loss of in-river wood, (3) altered tributary floodplain and channel
morphology, (4) reduced tributary stream flow and impaired passage, (5) harvest impacts,
and (6) degraded water quality.
Puget Sound Chinook Salmon
Distribution
The boundaries of the Puget Sound ESU correspond generally with the boundaries of the
Puget Lowland Ecoregion (Figure 11). The Puget Lowland Ecoregion begins in
Washington at approximately the Dungeness River near the eastern end of the Strait of
Juan de Fuca and extends through Puget Sound to the British Columbia border and up to
the Cascade foothills. The Puget Sound ESU includes all runs of Chinook salmon in the
Puget Sound region from the North Fork Nooksack River to the Elwha River on the
Olympic Peninsula. This ESU is comprised of 31 historical populations. Of these, 22
populations are believed to be extant. Thirty-six hatchery populations were included as
part of the ESU and five were considered essential for recovery and listed. They include
spring Chinook salmon from Kendall Creek, the North Fork Stillaguamish River, White
River, and Dungeness River, and fall run fish from the Elwha River (Table 10). Table 10
identifies populations within the Puget Sound Chinook salmon ESU, their abundances,
and hatchery input.
DRAFT 91
-------
49'N-
47 N-
^ AafiW .3»
^»AJ>; r •**
NPDES permit sites
EPA 303(d) rivers
125 W
121 W
Prepared by Dwayne Meadows
Kilometers 15 July 2008
Figure 11. Puget Sound Chinook distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
92
-------
DRAFT
Table 10. Puget Sound Chinook salmon - preliminary population structure,
abundances, and hatchery contributions (Good, Waples et al. 2005).
Population
Nooksack-North Fork
Nooksack-South Fork
Lower Skagit
Upper Skagit
Upper Cascade
Lower Sauk
Upper Sauk
Suiattle
Stillaguamish-North Fork
Stillaguamish-South Fork
Skykomish
Snoqualmie
North Lake Washington
Cedar
Green
White
Puyallup
Nisqually
Skokomish
Dosewallips
Duckabush
Hamma Hamma
Mid Hood Canal
Dungeness
Elwha
Total
Historical
Abundance
26,000
13,000
22,000
35,000
1,700
7,800
4,200
830
24,000
20,000
51,000
33,000
Unknown
Unknown
Unknown
Unknown
33,000
18,000
Unknown
4,700
Unknown
Unknown
Unknown
8,100
Unknown
-690,000
Most Recent
Spawner
Abundance
1,538
338
2,527
9,489
274
601
324
365
1,154
270
4,262
2,067
331
327
8,884
844
1,653
1,195
1,392
48
43
196
311
222
688
39,343
Hatchery
Abundance
Contributions
91%
40%
0.2%
2%
0.3%
0%
0%
0%
40%
Unknown
40%
16%
Unknown
Unknown
83%
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
DRAFT
93
-------
DRAFT
Life History
Chinook salmon in this area generally have an "ocean-type" life history. Puget Sound
populations exhibit both the early-returning and late-returning Chinook salmon spawners
described by Healey (1997). However, within these two generalized behavioral forms,
substantial variation occurs in juvenile behavior and residence time in fresh water and
estuarine environments. Hayman et al. (1996) described three juvenile life histories for
Chinook salmon with varying freshwater and estuarine residency times in the Skagit
River system in northern Puget Sound. Chinook salmon use the nearshore area of Puget
Sound during all seasons of the year and can be found long distances from their natal
river systems (Brennan, Higgins et al. 2004).
Status and Trends
Puget Sound Chinook salmon were listed as threatened in 1999 (64 FR 14308). This
status was re-affirmed on June 28, 2005 (70 FR 37160). This ESU has lost 15 spawning
aggregations that were either demographically independent historical populations or
major components of the life history diversity of the remaining 22 existing independent
historical populations identified (Good, Waples et al. 2005). Nine of the 15 extinct
spawning aggregations were early-run type Chinook salmon (Good, Waples et al. 2005).
The disproportionate loss of early-run life history diversity represents a significant loss of
the evolutionary legacy of the historical ESU.
The estimated total run size of Chinook salmon in Puget Sound in the early 1990s was
240,000 fish, representing a loss of nearly 450,000 fish from historic numbers. During a
recent five-year period, the geometric mean of natural spawners in populations of Puget
Sound Chinook salmon ranged from 222 to just over 9,489 fish. Most populations had
natural spawners numbering in the hundreds (median recent natural escapement is 766).
Of the six populations with greater than 1,000 natural spawners, only two have a low
fraction of hatchery fish. Estimates of the historical equilibrium abundance, based on
pre-European settlement habitat conditions, range from 1,700 to 51,000 potential Puget
Sound Chinook salmon spawners per population. The historical estimates of spawner
capacity are several orders of magnitude higher than spawner abundances currently
observed throughout the ESU (Good, Waples et al. 2005).
Long-term trends in abundance and median population growth rates for naturally
spawning populations of Puget Sound Chinook salmon indicate that approximately half
of the populations are declining and the other half are increasing in abundance over the
length of available time series. Eight of 22 populations are declining over the short-term,
compared to 11 or 12 populations that have long-term declines (Good, Waples et al.
2005). Widespread declines and extirpations of spring- and summer-run Puget Sound
Chinook salmon populations represent a significant reduction in the life history diversity
DRAFT 94
-------
DRAFT
of this ESU (Meyers, Kope et al. 1998). The median overall populations of long-term
trend in abundance is 1, indicating that most populations are just replacing themselves.
Populations with the greatest long-term population growth rate are the North Fork
Nooksack and White rivers.
Regarding spatial structure, the populations (22) presumed to be extinct are mostly early
returning fish. Most of these are in the mid- to southern Puget Sound or Hood Canal and
the Strait of Juan de Fuca. The ESU populations with the greatest estimated fractions of
hatchery fish tend to be in the mid-to southern Puget Sound, Hood Canal, and the Strait
of Juan de Fuca. Finally, all but one of the nine extinct Chinook salmon stocks is an
early run population (or component of a population).
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52630).
The critical habitat designation for this ESU identifies PCEs that include sites necessary
to support one or more Chinook salmon life stages. Specific sites include freshwater
spawning sites, freshwater rearing sites, freshwater migration corridors, nearshore marine
habitat, and estuarine areas. The physical or biological features that characterize these
sites include water quality and quantity, natural cover, forage, adequate passage
conditions, and floodplain connectivity.
Of 49 subbasins (5th field Hydrological Units) reviewed in NMFS' assessment of critical
habitat for the Puget Sound ESUs, nine subbasins were rated as having a medium
conservation value, 12 were rated as low, and the remaining subbasins (40), where the
bulk of Federal lands occur in this ESU, were rated as having a high conservation value
to Puget Sound Chinook salmon. Factors contributing to the downward trends in this
ESU are hydromorphological changes (such as diking, revetments, loss of secondary
channels in floodplains, widespread blockages of streams, and changes in peak flows),
degraded freshwater and marine habitat affected by agricultural activities and
urbanization, and upper river tributaries widely affected by poor forest practices, and
lower tributaries. Hydroelectric development and flood control also impact Puget Sound
Chinook salmon in several basins. Changes in habitat quantity, availability, diversity,
flow, temperature, sediment load, water quality, and channel stability are common
limiting factors in areas of critical habitat.
Sacramento River Winter-Run Chinook Salmon
Distribution
Sacramento River winter-run Chinook salmon consists of a single spawning population
that enters the Sacramento River and its tributaries in California from November to June
DRAFT 95
-------
DRAFT
and spawns from late April to mid-August, with a peak from May to June (Figure 12).
Sacramento River winter Chinook salmon historically occupied cold, headwater streams,
such as the upper reaches of the Little Sacramento, McCloud, and lower Pit Rivers.
Life History
Winter-run fish spawn mainly in May and June in the upper mainstem of the Sacramento
River. Winter-run fish have characteristics of both stream- and ocean-type races. They
enter the river and migrate far upstream. Spawning is delayed for some time after river
entry. Young winter-run Chinook salmon, however migrate to sea in November and
December, after only four to seven months of river life (Burgner 1991).
Status and Trends
Sacramento River winter-run Chinook salmon were listed as endangered on January 4,
1994 (59 FR 440), and were reaffirmed as endangered on June 28, 2005 (70 FR 37160).
This was based on restricted access from dams to a small fraction of salmon historic
spawning habitat and the degraded conditions of remaining habitat. Sacramento River
winter-run Chinook salmon consist of a single self-sustaining population which is
entirely dependent upon the provision of suitably cool water from Shasta Reservoir
during periods of spawning, incubation, and rearing.
Construction of Shasta Dams in the 1940s eliminated access to historic spawning habitat
for winter-run Chinook salmon in the basin. Winter-run Chinook salmon were not
expected to survive this habitat alteration (Moffett 1949). However, cold water releases
from Shasta Dam have created conditions suitable for winter Chinook salmon for roughly
60 miles downstream from the dam. As a result the ESU has been reduced to a single
spawning population confined to the mainstem Sacramento River below Keswick Dam.
Some adult winter-run Chinook salmon were recently observed in Battle Creek, a
tributary to the upper Sacramento River.
DRAFT 96
-------
40 N'
PVf i
x1^ *•
Legend
ty Dams
*• NPDES permit sites
EPA 303(d) rivers
Migratory corridor
37'N
-41 N
•39 N
•37'N
123W
122 W
0 20 40
80
Prepared by Dwayne Meadows
Kilometers is July 2008
Figure 12. Sacramento River Winter-run Chinook salmon distribution. The Legend for the Land
Cover Class categories is found in Figure 7.
DRAFT
97
-------
DRAFT
Quantitative estimates of run-size are not available for the period before 1996, the
completion of Red Bluff Diversion Dam. However, winter-runs may have been as large
as 200,000 fish based upon commercial fishery records from the 1870s (Brown, Moyle et
al. 1994).
The CDFG estimated spawning escapement of Sacramento River winter-run Chinook
salmon at 61,300 (60,000 mainstem, 1,000 Battle Creek, and 300 in Mill Creek) in the
early 1960s. During the first three years of operation of the county facility at the Red
Bluff Diversion Dam (1967 to 1969), the spawning run of winter-run Chinook salmon
averaged 86,500 fish. From 1967 through the mid-1990s, the population declined at an
average rate of 18% per year, or roughly 50% per generation. The population reached
critically low levels during the drought of 1987 to 1992. The three-year average run size
for the period of 1989 to 1991 was 388 fish.
Based on the Red Bluff Diversion Dam counts, the population has been growing rapidly
since the 1990s. Mean run size from 1995-2000 has been 2,191, but have ranged from
364 to 65,683 (Good, Waples et al. 2005). Most recent estimates indicate that the short-
term trend is 0.26, and the population growth rate is less than one.
Critical Habitat
Critical habitat was designated for this species on June 16, 1993 (58 FR 33212). The
following areas consist of the water, waterway bottom, and adjacent riparian zones: the
Sacramento River from Keswick Dam, Shasta County (river mile 302) to Chipps Island
(river mile 0) at the westward margin of the Sacramento-San Joaquin Delta, and other
specified estuarine waters. Factors contributing to the downward trends in this ESU
include: (1) Reduced access to spawning/rearing habitat, (2) possible loss of genetic
integrity through population bottlenecks, (3) inadequately screened diversions, (4)
predation at artificial structures and by nonnative species, (5) pollution from Iron
Mountain Mine and other sources, (6) adverse flow conditions, (7) high summer water
temperatures, (8) degraded water quality, (9) unsustainable harvest rates, (10) passage
problems at various structures, and (11) vulnerability to drought (Good, Waples et al.
2005).
Snake River Fall-Run Chinook Salmon
Distribution
Historically, the primary fall-run Chinook salmon spawning areas occurred on the upper
mainstem Snake River (SR) (Connor, Sneva et al. 2005). A series of SR mainstem dams
blocks access to the upper SR, which significantly reduced spawning and rearing habitat
for SR fall-run Chinook salmon (Figure 13).
DRAFT 98
-------
124 W
122W
12DTN
118'W
116 W
114*W
46 N
44 N
Dams
'* NPDES permit sites
EPA 303(d) rivers
Migratory corridor
-48 N
—44 N
124 W
122 W
120W
118 W
0 35 70
116 W
114W
140
Prepared by Dwayne Meadows
Kilometers 15 July 2008
Figure 13. SR fall-run Chinook salmon distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
99
-------
DRAFT
The present range of spawning and rearing habitat for naturally-spawned SR fall-run
Chinook salmon is limited to the SR below Hells Canyon Dam and the lower reaches of
the Clearwater River. SR fall-run Chinook salmon spawn above Lower Granite Dam in
the mainstem SR and in the lower reaches of the larger tributaries.
As a consequence of lost access to historic spawning and rearing sites in the Upper SR,
fall-run Chinook salmon now reside in waters that are generally cooler than the majority
of historic spawning areas. Additionally, alteration of the Lower SR by hydroelectric
dams has created a series of low-velocity pools in the SR that did not exist historically.
Life History
Prior to alteration of the SR basin by dams, fall Chinook salmon exhibited a largely
ocean-type life history, where they migrated downstream and reared in the mainstem SR
during their first year. Today, fall Chinook salmon in the SR Basin exhibit one of two
life histories: ocean type and reservoir-type (Connor, Sneva et al. 2005). The reservoir-
type life history is one where juveniles overwinter in the pools created by the dams, prior
to migrating out of the SR. The reservoir-type life history is likely a response to early
development in cooler temperatures which prevents juveniles from reaching suitable size
to migrate out of the SR.
Adult SR fall-run Chinook salmon enter the Columbia River in July and August.
Spawning occurs from October through November. Juveniles emerge from gravels in
March and April of the following year, moving downstream from natal spawning and
early rearing areas from June through early fall.
Status and Trends
SR fall-run Chinook salmon were originally listed as threatened in 1992 (57 FR 14653).
Their classification was reaffirmed following a status review on June 28, 2005 (70 FR
37160). Estimated annual returns for the period 1938 to 1949 was 72,000 fish. By the
1950s, numbers had declined to an annual average of 29,000 fish (Bjornn and Horner
1980). Numbers of SR fall-run Chinook salmon continued to decline during the 1960s
and 1970s as approximately 80% of their historic habitat was eliminated or severely
degraded by the construction of the Hells Canyon complex (1958 to 1967) and the lower
SR dams (1961 to 1975). Counts of natural-origin adult SR fall-run Chinook salmon at
Lower Granite Dam were 1,000 fish in 1975, and ranged from 78 to 905 fish (with an
average of 489 fish) over the ensuing 25-year period (Good, Waples et al. 2005).
Numbers of natural-origin SR fall-run Chinook salmon have increased over the last few
years, with estimates at Lower Granite Dam of 2,652 fish in 2001, 2,095 fish in 2002, and
3,895 fish in 2003.
DRAFT 100
-------
DRAFT
SR fall-run Chinook salmon have exhibited an upward trend in returns over Lower
Granite Dam since the mid-1990s. Returns classified as natural-origin exceeded 2,600
fish in 2001, compared to a 1997-2001 geometric mean natural-origin count of 871.
Long- and short-term trends in natural returns are positive. Harvest impacts on SR fall-
run Chinook salmon declined after listing and have remained relatively constant in recent
years. There have been major reductions in fisheries impacting this stock. Mainstem
conditions for subyearling Chinook salmon migrants from the SR have generally
improved since the early 1990s. The hatchery component, derived from outside the
basin, has decreased as a percentage of the run at Lower Granite Dam from the 1998/99
status reviews (five year average of 26.2%) to 2001 (8%). This reflects an increase in the
Lyons Ferry component, systematic removal of marked hatchery fish at the Lower
Granite trap, and modifications to the Umatilla supplementation program to increase
homing of fall Chinook release groups.
Overall abundance for SR fall-run Chinook salmon is relatively low, but has been
increasing in the last decade (Good et al. 2006). The 1997 to 2001 geometric mean
natural-origin count over Lower Granite Dam approximate 35% of the proposed delisting
abundance criteria of 2,500 natural spawners averaged over 8 years. The recent
abundance is approaching the delisting criteria. However, hatchery fish are faring better
than wild fish.
Regarding productivity [population growth rate (lambda)], the long-term trend in total
returns is >1; indicating the population size is growing. Although total abundance has
dropped sharply in the past two years, it still remains at levels higher than previous
decades. Productivity is likely sustained largely by a system of small artificial rearing
facilities in the Lower SR Basin. The growth trend for natural-origin fish is close to 1,
and could either be higher or lower, depending on the number of hatchery fish that spawn
naturally.
The historic spatial structure has been reduced to one single remnant population. The
ESU occupies a relatively small amount of marginal habitat, with the vast majority of
historic habitat inaccessible. Genetic diversity is likely reduced from historic levels.
Hatcheries affect ESU genetics due to three major components: natural-origin fish
(which may be progeny of hatchery fish), returns of SR fall-run fish from the Lyons Ferry
Hatchery program, and strays from hatchery programs outside the SR. Nevertheless, the
SR fall-run Chinook salmon remains genetically distinct for similar fish in other basins.
Phenotypic characteristics have shifted in apparent response to environmental changes
from hydroelectric dams (Connor, Sneva et al. 2005).
The ICBTRT has defined only one extant population for the SR fall-run Chinook salmon,
DRAFT 101
-------
DRAFT
the lower SR mainstem population. This population occupies the SR from its confluence
with the Columbia River to Hells Canyon Dam, and the lower reaches of the Clearwater,
Imnaha, Grande Rhonde, Salmon, and Tucannonh Rivers (ICBTRT 2003).
Critical Habitat
Critical habitat for these salmon was designated on December 28, 1993 (58 FR 68543).
This critical habitat encompasses the waters, waterway bottoms, and adjacent riparian
zones of specified lakes and river reaches in the Columbia River that are or were
accessible to listed SR fall-run salmon (except reaches above impassable natural falls,
and Dworshak and Hells Canyon Dams). Adjacent riparian zones are defined as those
areas within a horizontal distance of 300 ft from the normal line of high water of a stream
channel or from the shoreline of a standing body of water. Designated critical habitat
includes the Columbia River from a straight line connecting the west end of the Clatsop
jetty (Oregon side) and the west end of the Peacock jetty (Washington side), all river
reaches from the estuary upstream to the confluence of the SR, and all SR reaches
upstream to Hells Canyon Dam. Critical habitat also includes several river reaches
presently or historically accessible to SR fall-run Chinook salmon. Limiting factors
identified for SR fall-run Chinook salmon include: (1) Mainstem lower Snake and
Columbia hydrosystem mortality, (2) degraded water quality, (3) reduced spawning and
rearing habitat due to mainstem lower SR hydropower system, (4) harvest impacts, (5)
impaired stream flows, barriers to fish passage in tributaries, excessive sediment, and (6)
altered floodplain and channel morphology (NMFS 2005b). The above activities and
features also introduce sediment, nutrients, biocides, metals, and other pollutants into
surface and ground water and degrade water quality in the freshwater, estuarine, and
coastal ecosystems throughout the Pacific Northwest.
Snake River Spring/Summer-Run Chinook Salmon
Distribution
SR spring/summer-run Chinook salmon are primarily limited to the Salmon, Grande
Ronde, Imnaha, and Tucannon Rivers in the SR basin (Figure 14). The SR basin drains
portions of southeastern Washington, northeastern Oregon, and north/central Idaho.
Environmental conditions are generally drier and warmer in these areas than in areas
occupied by other Chinook salmon species. The ICBTRT has identified 32 populations
in five MPGs (Upper Salmon River, South Fork Salmon River,
DRAFT 102
-------
124 W
122 W
I
* : '•
.>
-« , SS^&**£r"W "
i.l ,? fr. *. /^^^^^ t^. tv r
-L -48 N
-46 N
-44N
42 N'
124 W
Prepared by Dwayne Meadows
Kilometers 18 July 2008
Figure 14. SR Spring/Summer-run Chinook salmon distribution. .
DRAFT
103
-------
DRAFT
Middle Fork , Salmon River, Grande Ronde/Imnaha, Lower Snake Mainstem Tributaries)
for this species. Historic populations above Hells Canyon Dam are considered extinct
(ICBTRT 2003). This ESU includes production areas that are characterized by spring-
timed returns, summer-timed returns, and combinations from the two adult timing
patterns. Historically, the Salmon River system may have supported more than 40% of
the total run of spring and summer Chinook salmon to the Columbia system (Fulton
1968).
Some or all of the fish returning to several of the hatchery programs are also listed,
including those returning to the Tucannon River, Imnaha River, and Grande Ronde River
hatcheries, and to the Sawtooth, Pahsimeroi, and McCall hatcheries on the Salmon River.
The Salmon River system contains a range of habitats used by spring/summer Chinook.
The South Fork and Middle Fork Salmon Rivers currently support the bulk of natural
production in the drainage. Returns into the upper Salmon River tributaries have
reestablished following the opening of passage around Sunbeam Dam on the mainstem
Salmon River downstream of Stanley, Idaho. The dam was impassable to anadromous
fish from 1910 until the 1930s. Table 11 identifies populations within the SR
spring/summer Chinook salmon ESU, their abundances, and hatchery input.
Table 11. SR Spring/Summer Chinook salmon populations, abundances, and
hatchery contributions (Good, Waples et al. 2005). Note: rpm denotes redds per
mile.
Current Populations
Tucannon River
Wenaha River
Wallowa River
Lostine River
Minam River
Catherine Creek
Upper Grande Ronde River
South Fork Salmon River
Secesh River
Johnson Creek
Big Creek spring run
Big Creek summer run
Loon Creek
Marsh Creek
Bear Valley/Elk Creek
North Fork Salmon River
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Most Recent
Spawner
Abundance
128-1,012
67-586
0-29 redds
9-131 redds
96-573
13-262
3-336
277-679 redds
3 8-444 redds
49-444 redds
21-296
2-58 redds
6-255 redds
0-164
72-712
2-19 redds
Hatchery
Abundance
Contributions
76%
64%
5%
5%
5%
56%
58%
9%
4%
0%
0%
Unknown
0%
0%
0%
Unknown
DRAFT
104
-------
DRAFT
Current Populations
Lemhi River
Pahsimeroi River
East Fork Salmon spring run
East Fork Salmon summer run
Yankee Fork spring run
Yankee Fork summer run
Valley Creek spring run
Valley Creek summer run
Upper Salmon spring run
Upper Salmon summer run
Alturas Lake Creek
Imnaha River
Big Sheep Creek
Lick Creek
Total
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
~1.5 million
Most Recent
Spawner
Abundance
35-216 redds
72-1,097
0.27 rpm
1.22rpm
0
1-1 8 redds
2-28 redds
2. 14 rpm
25-357 redds
0.24 rpm
0-1 8 redds
194-3,041 redds
0.25 redds
0-29 redds
-9,700
Hatchery
Abundance
Contributions
0%
Unknown
Unknown
0%
Unknown
0%
0%
Unknown
Unknown
Unknown
Unknown
62%
97%
59%
Life History
SR spring/summer-run Chinook salmon exhibit a stream-type life history. Eggs are
deposited in late summer and early fall, incubate over the following winter, and hatch in
late winter and early spring of the following year. Juvenile fish mature in fresh water for
one year before they migrate to the ocean in the spring of their second year of life.
Depending on the tributary and the specific habitat conditions, juveniles may migrate
extensively from natal reaches into alternative summer-rearing or overwintering areas.
SR spring/summer-run Chinook salmon return from the ocean to spawn primarily as four
and five-year old fish, after two to three years in the ocean. A small fraction of the fish
return as three year-old "jacks", heavily predominated by males.
Status and Trends
SR spring/summer-run Chinook salmon were originally listed as threatened on April 22,
1992 (57 FR 14653). Their classification was reaffirmed following a review on June 28,
2005 (70 FR 37160). Although direct estimates of historical annual SR spring/summer
Chinook salmon returns are not available, returns may have declined by as much as 97%
between the late 1800s and 2000. According to Matthews and Waples (1997), total
annual SR spring/summer Chinook salmon production may have exceeded 1.5 million
adult fish in the late 1800s. Total (natural plus hatchery origin) returns fell to roughly
100,000 spawners by the late 1960s (Fulton 1968) and were below 10,000 by 1980.
Between 1981 and 2000, total returns fluctuated between extremes of 1,800 and 44,000
fish. The 2001 and 2002 total returns increased to over 185,000 and 97,184 adults,
DRAFT
105
-------
DRAFT
respectively. The 1997 to 2001 geometric mean total return for the summer run
component at Lower Granite Dam was slightly more than 6,000 fish, compared to the
geometric mean of 3,076 fish for the years 1987 to 1996. The 2002 to 2006 geometric
mean of the combined Chinook salmon runs at Lower Granite Dam was over 18,000 fish.
However, over 80% of the 2001 return and over 60% of the 2002 return originated in
hatcheries (Good, Waples et al. 2005). Good et al. (2006) reported that risks to
individual populations within the ESU may be greater than the extinction risk for the
entire ESU due to low levels of annual abundance and the extensive production areas
within the SR basin. Year-to-year abundance has high variability and is most pronounced
in natural-origin fish. Although the average abundance in the most recent decade is more
abundant than the previous decade, there is no obvious long-term trend. Additionally,
hatchery fish are faring better than wild fish, which comprise roughly 40% of the total
returns in the past decade. Overall, most populations are far below their respective
interim recovery targets.
Regarding population growth rate (lambda), long-term trends are <1; indicating the
population size is shrinking. However, recent trends, buoyed by last 5 years, are
approaching 1. Nevertheless, many spawning aggregates have been extirpated, which has
increased the spatial separation of some populations. Populations are widely distributed
in a diversity of habitats although roughly one-half of historic habitats are inaccessible.
There is no evidence of wide-scale genetic introgression by hatchery populations. The
high variability in life history traits indicates sufficient genetic variability within the DPS
to maintain distinct subpopulations adapted to local environments. Despite the recent
increases in total spring/summer-run Chinook salmon returns to the basin, natural-origin
abundance and productivity remain below their targets. SR spring/summer Chinook
salmon remains likely to become endangered (Good, Waples et al. 2005).
Critical Habitat
Critical habitat for these salmon was designated on October 25, 1999 (64 FR 57399).
This critical habitat encompasses the waters, waterway bottoms, and adjacent riparian
zones of specified lakes and river reaches in the Columbia River that are or were
accessible to listed SR salmon (except reaches above impassable natural falls, and
Dworshak and Hells Canyon Dams). Adjacent riparian zones are defined as those areas
within a horizontal distance of 300 ft from the normal line of high water of a stream
channel or from the shoreline of a standing body of water. Designated critical habitat
includes the Columbia River from a straight line connecting the west end of the Clatsop
jetty (Oregon side) and the west end of the Peacock jetty (Washington side). Critical
habitat also includes all river reaches from the estuary upstream to the confluence of the
SR, and all SR reaches upstream to Hells Canyon Dam; the Palouse River from its
confluence with the SR upstream to Palouse Falls, the Clearwater River from its
DRAFT 106
-------
DRAFT
confluence with the SR upstream to its confluence with Lolo Creek; the North Fork
Clearwater River from its confluence with the Clearwater river upstream to Dworshak
Dam.
Limiting factors identified for this species include: (1) Hydrosystem mortality, (2)
reduced stream flow, (3) altered channel morphology and floodplain, (4) excessive fine
sediment, and (5) degraded water quality (Myers, Busack et al. 2006). The above
activities and features also introduce sediment, nutrients, biocides, metals, and other
pollutants into surface and ground water and degrade water quality in the freshwater,
estuarine, and coastal ecosystems throughout the Pacific Northwest.
Upper Willamette River Chinook Salmon
Distribution
Upper Willamette River (UWR) Chinook salmon occupy the Willamette River and
tributaries upstream of Willamette Falls (Figure 15). In the past, this ESU included
sizable numbers of spawning salmon in the Santiam River, the middle fork of the
Willamette River, and the McKenzie River, as well as smaller numbers in the Molalla
River, Calapooia River, and Albiqua Creek. Historically, access above Willamette Falls
was restricted to the spring when flows were high. In autumn, low flows prevented fish
from ascending past the falls. The UWR Chinook salmon are one of the most genetically
distinct Chinook salmon groups in the Columbia River Basin. Fall-run Chinook salmon
spawn in the Upper Willamette but are not considered part of the species because they are
not native. None of the hatchery populations in the Willamette River were listed
although five spring-run hatchery stocks were included in the species' listing. UWR
Chinook salmon migrate far north and are caught incidentally in ocean fisheries,
particularly off southeast Alaska and northern Canada, and in spring season fisheries in
the mainstem Columbia and Willamette Rivers. Table 12 identifies populations within
the UWR Chinook salmon ESU, their abundances, and hatchery input.
DRAFT 107
-------
123'W
122 W
4S N
4: II-
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
Migratory corridor
80 Prepared by Dwayne Meadows
Kilometers 15 July 2008
-43 N
Figure 15. UWR Chinook salmon distribution. The Legend for the Land Cover
Class categories is found in Figure 7.
DRAFT
108
-------
DRAFT
Table 12. UWR Chinook salmon populations, abundances, and hatchery
j^ontributions (GoojdjJtjV^
Most Recent
Historical Hatchery
Current Populations Spawner Abundance
Abundance „ „.,.,.
Abundance Contributions
Clackamas River
Molalla River
North Santiam River
South Santiam River
Calapooia River
McKenzie River
Middle Fork Willamette River
Upper Fork Willamette River
Total
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
>70,000
2,910
52 redds
-7.1 rpm
982 redds
16 redds
-2,470
235 redds
Unknown
-9,700
64%
>93%
>95%
>84%
100%
26%
>39%
Unknown
Mostly^atehery^
Life History
UWR Chinook salmon exhibit an earlier time of entry into the Columbia River and
estuary than other spring Chinook salmon ESUs (Meyers, Kope et al. 1998). Although
juveniles from interior spring Chinook salmon populations reach the mainstem migration
corridor as yearling, some juvenile Chinook salmon in the lower Willamette River are
subyearlings (Friesen, Vile et al. 2004).
Status and Trends
UWR Chinook salmon were listed as threatened on March 24, 1999 (64 FR 14308), and
reaffirmed as threatened on June 28, 2005 (70 FR 37160). The total abundance of adult
spring-run Chinook salmon (hatchery-origin + natural-origin fish) passing Willamette
Falls has remained relatively steady over the past 50 years (ranging from approximately
20,000 to 70,000 fish). However, it is an order of magnitude below the peak abundance
levels observed in the 1920s (approximately 300,000 adults). Until recent years,
interpretation of abundance levels has been confounded by a high but uncertain fraction
of hatchery-produced fish.
Most natural spring Chinook salmon populations is likely extirpated or nearly so. Only
one remaining naturally reproducing population is identified in this ESU: the spring
Chinook salmon in the McKenzie River. Unfortunately, recent short-term declines in
abundance suggest that this population may not be self-sustaining (Meyers, Kope et al.
DRAFT
109
-------
DRAFT
1998; Good, Waples et al. 2005). Most of the natural-origin populations in this ESU
have very low current abundances (less than a few hundred fish) and many largely have
been replaced by hatchery production. Long- and short-term trends for population
growth rate are approximately 1 or are negative, depending on the metric examined (i.e.,
long-term trend [regression of log-transformed spawner abundance] or lambda [median
population growth rate]). Although the population increased substantially in 2000-2003,
it was probably due to increased survival in the ocean. Future survival rates in the ocean
are unpredictable, and the likelihood of long-term sustainability for this population has
not been determined. Although the number of adult spring-run Chinook salmon crossing
Willamette Falls is in the same range (about 20,000 to 70,000 adults) it has been for the
last 50 years, a large fraction of these are hatchery produced. Of concern is that a
majority of the spawning habitat and approximately 30 to 40% of total historical habitat
are no longer accessible because of dams (Good, Waples et al. 2005).
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52630).
Designated critical habitat includes all Columbia River estuarine areas and river reaches
proceeding upstream to the confluence with the Willamette River as well as specific
stream reaches in a number of subbasins. The critical habitat designation for this ESU
also identifies PCEs that include sites necessary to support one or more Chinook salmon
life stages. Specific sites include freshwater spawning and rearing sites, freshwater
migration corridors. The physical or biological features that characterize these sites
include water quality and quantity, natural cover, forage, adequate passage conditions,
and floodplain connectivity. Of 65 subbasins reviewed in NMFS' assessment of critical
habitat for the UWR Chinook salmon ESU, 19 subbasins were rated as having a medium
conservation value, 19 were rated as low, and the remaining subbasins (27), were rated as
having a high conservation value to UWR Chinook salmon. Federal lands were generally
rated as having high conservation value to the species' spawning and rearing. Factors
contributing to the downward trends in this ESU include: (1) Reduced access to
spawning/rearing habitat in tributaries, (2) hatchery impacts, (3) altered water quality and
temperature in tributaries, (4) altered stream flow in tributaries, and (5) lost/degraded
floodplain connectivity and lowland stream habitat.
Chum Salmon
Description of the Species
Chum salmon has the widest natural geographic and spawning distribution of any Pacific
salmonid because its range extends farther along the shores of the Arctic Ocean than
other salmonids. Chum salmon have been documented to spawn from Korea and the
Japanese island of Honshu, east around the rim of the North Pacific Ocean to Monterey
DRAFT 110
-------
DRAFT
Bay, California. Historically, chum salmon were distributed throughout the coastal
regions of western Canada and the U.S. Presently, major spawning populations are found
only as far south as Tillamook Bay on the northern Oregon coast. We discuss the
distribution, life history diversity, status, and critical habitat of the two species of
threatened chum salmon separately.
Chum salmon are semelparous, spawn primarily in freshwater, and exhibit obligatory
anadromy (there are no recorded landlocked or naturalized freshwater populations).
Chum salmon spend two to five years in feeding areas in the northeast Pacific Ocean,
which is a greater proportion of their life history than other Pacific salmonids. Chum
salmon distribute throughout the North Pacific Ocean and Bering Sea. North American
chum salmon (as opposed to chum salmon originating in Asia) rarely occur west of 175°
E longitude.
North American chum salmon migrate north along the coast in a narrow coastal band that
broadens in southeastern Alaska. However, some data suggest that Puget Sound chum,
including Hood Canal summer run chum, may not make extended migrations into
northern British Columbian and Alaskan waters. Instead, they may travel directly
offshore into the north Pacific Ocean.
Chum salmon, like pink salmon, usually spawn in the lower reaches of rivers, with redds
usually dug in the mainstem or in side channels of rivers from just above tidal influence
to nearly 100 km from the sea. Juveniles outmigrate to seawater almost immediately
after emerging from the gravel that covers their redds (Salo 1991). The immature salmon
distribute themselves widely over the North Pacific Ocean. The maturing adults return to
the home streams at various ages, usually at two through five years, and at some cases up
to seven years (Bigler 1985). This ocean-type migratory behavior contrasts with the
stream-type behavior of some other species in the genus Oncorhynchus (e.g., coastal
cutthroat trout, steelhead, coho salmon, and most types of Chinook and sockeye salmon),
which usually migrate to sea at a larger size, after months or years of freshwater rearing.
This means that survival and growth in juvenile chum salmon depend less on freshwater
conditions (unlike stream-type salmonids which depend heavily on freshwater habitats)
than on favorable estuarine conditions. Another behavioral difference between chum
salmon and species that rear extensively in freshwater is that chum salmon form schools.
Presumably, this behavior reduces predation (Pitcher 1986), especially if fish movements
are synchronized to swamp predators (Miller and Brannon 1982).
The duration of estuarine residence for chum salmon juveniles are known for only a few
estuaries. Observed residence times range from 4 to 32 days; with a period of about 24
days being the most common (Johnson, Grant et al. 1997). Juvenile salmonids rely on a
DRAFT 111
-------
DRAFT
variety of non-main channel habitats that are critical to rearing. All listed salmonids use
shallow, low flow habitats at some point in their life cycle. Examples of off-channel
habitat include alcoves, channel edge sloughs, overflow channels, backwaters, terrace
tributaries, off-channel dredge ponds, and braids (Anderson 1999; Swift III 1979).
Status and Trends
Chum salmon have been threatened by overharvests in commercial and recreational
fisheries, adult and juvenile mortalities associated with hydropower systems, habitat
degradation from forestry and urban expansion, and shifts in climatic conditions that
changed patterns and intensity of precipitation.
Chum salmon, like the other salmon NMFS has listed, have declined under the combined
effects of overharvests in fisheries; competition from fish raised in hatcheries and native
and non-native exotic species; dams that block their migrations and alter river hydrology;
gravel mining that impedes their migration and alters the dynamics of the rivers and
streams that support juveniles; water diversions that deplete water levels in rivers and
streams; destruction or degradation of riparian habitat that increase water temperatures in
rivers and streams sufficient to reduce the survival of juvenile chum salmon; and land use
practices (logging, agriculture, urbanization) that destroy or alter wetland and riparian
ecosystems. The above activities and features also introduce sediment, nutrients,
biocides, metals, and other pollutants into surface and ground water and degrade water
quality in the freshwater, estuarine, and coastal ecosystems throughout the Pacific
Northwest.
Columbia River Chum Salmon
Distribution
Columbia River chum salmon includes all natural-origin chum salmon in the Columbia
River and its tributaries in Oregon and Washington. The species consists of three
populations: Gray's River, Hardy, and Hamilton Creek in Washington State (Figure 16).
This ESU also includes three artificial hatchery programs. There were 16 historical
populations in three MPGs in Oregon and Washington between the mouth of the
Columbia River and the Cascade crest. Significant spawning now occurs for two of the
historical populations. About 88% of the historical populations are extirpated. Table 10
identifies populations within the Columbia River Chum salmon ESU, their abundances,
and hatchery input.
DRAFT H2
-------
DRAFT
Table 13. Columbia River Chum salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005).
Most Recent
Current Populations
Youngs Bay
Gray's River
Big Creek
Elochoman River
Clatskanie River
Mill, Abernathy, and German
Creeks
Scappoose Creek
Cowlitz River
Kalama River
Lewis River
Salmon Creek
Clackamus River
Sandy River
Washougal River
Lower gorge tributaries
Upper gorge tributaries
Total
Historical
Abundance
Unknown
7,511
Unknown
Unknown
Unknown
Unknown
Unknown
141,582
9,953
89,671
Unknown
Unknown
Unknown
15,140
>3,141
>8,912
>283,421
Spawner
Abundance
0
331-704
0
0
0
0
0
0
0
0
0
0
0
0
425
0
756-1,129
Hatchery
Abundance
Contributions
0
Unknown
0
0
0
0
0
0
0
0
0
0
0
0
0
0
DRAFT
113
-------
47 N-
45 N-
Legend
Dams
NPDES permit sites
EPA 303{d) rivers
-47 N
—46:N
-45 N
124 W
123 W
20 40
122 W
80
121 W
Prepared by Dwayne Meadows
I Kilometers 15ju|v200e
Figure 16. Columbia River Chum salmon distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
114
-------
DRAFT
Life History
Chum salmon return to the Columbia River in late fall (mid-October to December). They
primarily spawn in the lower reaches of rivers, digging redds along the edges of the
mainstem and in tributaries or side channels. Some spawning sites are located in areas
where geothermally-warmed groundwater or mainstem flow upwells through the gravel.
Chum salmon fry emigrate from March through May shortly after emergence in contrast
to other salmonids (e.g., steelhead, coho salmon, and most Chinook salmon), which
usually migrate to sea at a larger size after months or years of freshwater rearing.
Juvenile chum salmon reside in estuaries to feed before beginning a long-distance
oceanic migration. Chum salmon may choose either the upper or lower estuaries
depending on the relative productivity of each. The timing of entry of juvenile chum
salmon into sea water is commonly correlated with the warming of the nearshore waters
and the accompanying plankton blooms (Burgner 1991). The movement offshore
generally coincides with the decline of inshore prey resources and is normally at the time
when the fish has grown to a size that allows them to feed upon neritic organisms and and
avoid predators (Burgner 1991).
Although most juvenile chum salmon migrate rapidly from freshwater to shallow
nearshore marine habitats after emergence from gravel beds, some may remain up to a
year in fresh water in large northern rivers. The period of estuarine residence appears to
be a critical life history phase and may play a major role in determining the size of the
subsequent adult run back to freshwater.
Status and Trends
Columbia River chum salmon were listed as threatened on March 25, 1999, and their
threatened status was reaffirmed on June 28, 2005 (71 FR 37160). Chum salmon in the
Columbia River once numbered in the hundreds of thousands of adults and were reported
in almost every river in the LCR basin. However, by the 1950s most runs disappeared
(Rich 1942; Marr 1943; Fulton 1968). The total number of chum salmon returning to the
Columbia River in the last 50 years has averaged a few thousand per year, with returns
limited to a very restricted portion of the historical range. Significant spawning occurs in
only two of the 16 historical populations. Nearly 88% of the historical populations are
extirpated. The two remaining populations are the Gray's River and the Lower Gorge
(Good, Waples et al. 2005). Chum salmon appear to be extirpated from the Oregon
portion of this ESU. In 2000, the Oregon Department of Fish and Wildlife (ODFW)
conducted surveys to determine the abundance and distribution of chum salmon in the
Columbia River. Of 30 sites surveyed, only one chum salmon was observed.
DRAFT H5
-------
DRAFT
Historically, the Columbia River chum salmon supported a large commercial fishery in
the first half of this century which landed more than 500,000 fish per year as recently as
1942. Commercial catches declined beginning in the mid-1950s, and in later years rarely
exceeded 2,000 per year. During the 1980s and 1990s, the combined abundance of
natural spawners for the Lower Gorge, Washougal, and Grays River populations was
below 4,000 adults. In 2002, however, the abundance of natural spawners exhibited a
substantial increase at several locations (estimate of natural spawners is approximately
20,000 adults). The cause of this dramatic increase in abundance is unknown.
Estimates of abundance and trends are available only for the Gray's River and Lower
Gorge populations. The 10-year trend was negative for the Gray's River population and
just over 1.0 for the Lower Gorge. The Upper Gorge population, and all four of the
populations on the Oregon side of the river in the Coastal MPG, are extirpated or nearly
so (McElhaney, Chilcote et al. 2007). However, long- and short-term productivity trends
for populations are at or below replacement. Regarding spatial structure, few Columbia
River chum salmon have been observed in tributaries between The Dalles and Bonneville
dams. Surveys of the White Salmon River in 2002 found one male and one female
carcass and the latter had not spawned (Ehlke and Keller 2003). Chum salmon were not
observed in any of the upper gorge tributaries, including the White Salmon River, during
the 2003 and 2004 spawning ground surveys. Finally, most Columbia River chum
populations have been functionally extirpated or are presently at very low abundance
levels. However in the Cascade MPG, chum sampled from each tributary recently were
shown to be the remnants of genetically distinct populations (Greco, Capri et al. 2007).
The loss of off-channel habitat and the extirpation of approximately 17 historical
populations increase this species' vulnerability to environmental variability and
catastrophic events. Overall, the populations that remain have low abundance, limited
distribution, and poor connectivity (Good, Waples et al. 2005).
Critical Habitat
Critical habitat was originally designated for this species on February 16, 2000 (65 FR
7764) and was re-designated on September 2, 2005 (70 FR 52630). The critical habitat
designation for this ESU identifies PCEs that include sites necessary to support one or
more chum salmon life stages. Columbia River chum salmon have PCEs of: (1)
Freshwater spawning, (2) freshwater rearing, (3) freshwater migration, (4) estuarine areas
free of obstruction, (5) nearshore marine areas free of obstructions, and (6) offshore
marine areas with good water quality. The physical or biological features that
characterize these sites include water quality and quantity, natural cover, forage, adequate
passage conditions, and floodplain connectivity.
Of 21 subbasins reviewed in NMFS' assessment of critical habitat for the Columbia
River chum salmon ESU, three subbasins were rated as having a medium conservation
DRAFT 116
-------
DRAFT
value, no subbasins were rated as low, and the majority of subbasins (18), were rated as
having a high conservation value to Columbia River chum salmon. Washington's federal
lands were rated as having high conservation value to the species. The major factors
limiting recovery for Columbia River chum salmon are altered channel form and stability
in tributaries, excessive sediment in tributary spawning gravels, altered stream flow in
tributaries and the mainstem Columbia River, loss of some tributary habitat types, and
harassment of spawners in the tributaries and mainstem. The above activities and
features also introduce sediment, nutrients, biocides, metals, and other pollutants into
surface and ground water and degrade water quality in the freshwater, estuarine, and
coastal ecosystems throughout the Pacific Northwest.
Hood Canal Summer-Run Chum Salmon
Distribution
This ESU includes all naturally spawned populations of summer-run chum salmon in
Hood Canal and its tributaries as well as populations in Olympic Peninsula rivers
between Hood Canal and Dungeness Bay, Washington (64 FR 14508, Figure 17). Eight
artificial propagation programs are considered as part of the ESU: the Quilcene National
Fish Hatchery, Hamma Hamma Fish Hatchery, Lilliwaup Creek Fish Hatchery, Union
River/Tahuya, Big Beef Creek Fish Hatchery, Salmon Creek Fish Hatchery, Chimacum
Creek Fish Hatchery, and the Jimmycomelately Creek Fish Hatchery summer-run chum
hatchery programs. NMFS determined that these artificially propagated stocks are no
more divergent relative to the local natural population(s) than what would be expected
between closely related natural populations within the species. Table 14 identifies
populations within the Hood Canal summer-run Chum salmon ESU, their abundances,
and hatchery input.
DRAFT H7
-------
124 W
A
NPDES permit sites
EPA 303(d) rivers
i
124 W
0 10 20
40 Prepared by Dwayne Meadovre
I Kilometers 15 July 2008
Figure 17. Hood Canal Summer-run Chum salmon distribution. The Legend for the Land Cover Class categories is found in
Figure 7.
DRAFT
118
-------
DRAFT
Table 14. Hood Canal summer-run Chum salmon populations, abundances, and
hatchery contributions (Good, Waples et al. 2005).
Current Populations
Jimmycomelately Creek
Salmon/Snow creeks
Big/Little Quilcene rivers
Lilliwaup Creek
Hamma Hamma River
Duckabush River
Dosewallips River
Union River
Chimacum Creek
Big Beef Creek
Dewetto Creek
Total
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Most Recent
Spawner
Abundance
-60
-2,200
-4,240
-164
-758
Unknown
-900
-690
0
0
0
-9,012
Hatchery
Abundance
Contributions
Unknown
0-69%
5-51%
Unknown
Unknown
Unknown
Unknown
Unknown
100
100
Unknown
Life History
The Hood Canal summer-run Chum salmon are defined in the Salmon and Steelhead
Stock Inventory (WDF, WDW et al. 1993) as fish that spawn from mid-September to
mid-October. However, summer chum have been known to enter natal rivers in late
August. Fall-run chum salmon are defined as fish that spawn from November through
December or January. Run-timing data for as early as 1913 indicated temporal separation
between summer and fall chum salmon in Hood Canal (Johnson, Grant et al. 1997).
Hood Canal summer Chum salmon are genetically distinct from healthy populations of
Hood Canal fall Chum salmon originating within this area. Hood Canal summer Chum
salmon return to natal rivers to spawn during the August through early October period.
The fall Chum salmon spawn between November and December, when streams are
higher and water temperature is lower.
The time to hatching varies among populations and among individuals within a
population (Salo 1991). Fry tend to emerge when they had their best chances of
surviving in streams and estuaries (Koski 1975). A variety of factors may influence the
DRAFT
119
-------
DRAFT
time to hatching, emergence from the gravel, or both. They include dissolved oxygen,
gravel size, salinity, nutritional conditions, behaviour of alevins in the gravel and
incubation temperature [reviewed in (Bakkala 1970; Schroder, Koski et al. 1974;
Schroder 1977; Salo 1991)]. The average residence time in estuaries for Hood canal
chum salmon is 23 days. Fry in Hood Canal have not been observed to display daily tidal
migrations (Bax 1983). Fry movement is associated with prey availability. Summer-run
chum salmon migrate up the Hood Canal and into the main body of Puget Sound. Fish
may emerge from streams over an extended period or juveniles may also remain in
Quilcene Bay for several weeks.
Status and Trends
Hood Canal summer-run Chum salmon were listed as threatened on March 25, 1999, and
reaffirmed as threatened on June 28, 2005 (70 FR 37160). Adult returns for some
populations in the Hood Canal summer-run Chum salmon species showed modest
improvements in 2000, with upward trends continuing in 2001 and 2002. The recent
five-year mean abundance is variable among populations in the species, ranging from one
fish to nearly 4,500 fish. Hood Canal summer-run chum salmon are the focus of an
extensive rebuilding program developed and implemented since 1992 by the state and
tribal co-managers. Two populations (the combined Quilcene and Union River
populations) are above the conservation thresholds established by the rebuilding plan.
However, most populations remain depressed. Estimates of the fraction of naturally
spawning hatchery fish exceed 60% for some populations. This indicates that
reintroduction programs are supplementing the numbers of total fish spawning naturally
in streams. Long-term trends in productivity are above replacement for only the Quilcene
and Union River populations. Buoyed by recent increases, seven populations are
exhibiting short-term productivity trends above replacement.
Of an estimated 16 historical populations in the ESU, seven populations are believed to
have been extirpated or nearly extirpated. Most of these extirpations have occurred in
populations on the eastern side of Hood Canal, generating additional concern for ESU
spatial structure. The widespread loss of estuary and lower floodplain habitat was noted
by the BRT as a continuing threat to ESU spatial structure and connectivity. There is
some concern that the Quilcene hatchery stock is exhibiting high rates of straying, and
may represent a risk to historical population structure and diversity. However, with the
extirpation of many local populations, much of this historical structure has been lost, and
the use of Quilcene hatchery fish may represent one of a few remaining options for Hood
Canal summer-run Chum salmon conservation.
Of the eight programs releasing summer chum salmon that are considered to be part of
this ESU, six of the programs are supplementation programs implemented to preserve
and increase the abundance of native populations in their natal watersheds. NMFS'
DRAFT 120
-------
DRAFT
assessment of the effects of artificial propagation on ESU extinction risk concluded that
these hatchery programs collectively do not substantially reduce the extinction risk of the
ESU. The hatchery programs are reducing risks to ESU abundance by increasing total
ESU abundance as well as the number of naturally spawning summer-run chum salmon.
Several of the programs have likely prevented further population extirpations in the ESU.
The contribution of ESU hatchery programs to the productivity of the ESU in-total is
uncertain. The hatchery programs are benefiting ESU spatial structure by increasing the
spawning area utilized in several watersheds and by increasing the geographic range of
the ESU through reintroductions. These programs also provide benefits to ESU diversity.
By bolstering total population sizes, the hatchery programs have likely stemmed adverse
genetic effects for populations at critically low levels. Additionally, measures have been
implemented to maintain current genetic diversity, including the use of native broodstock
and the termination of the programs after 12 years of operation to guard against long-term
domestication effects. Collectively, artificial propagation programs in the ESU presently
provide a slight beneficial effect to ESU abundance, spatial structure, and diversity.
However, artificial propagation programs also provide uncertain effects to ESU
productivity.
Critical Habitat
Critical habitat for this species was designated on September 2, 2005 (70 FR 52630).
Hood Canal summer-run chum salmon have PCEs of: (1) Freshwater spawning, (2)
freshwater rearing, (3) freshwater migration, (4) estuarine areas free of obstruction, (5)
nearshore marine areas free of obstructions, and (6) offshore marine areas with good
water quality. The physical or biological features that characterize these sites include
water quality and quantity, natural cover, forage, adequate passage conditions, and
floodplain connectivity.
Of 17 subbasins reviewed in NMFS' assessment of critical habitat for the Hood Canal
chum salmon ESU, 14 subbasins were rated as having a high conservation value, while
only three were rated as having a medium value to the conservation. Limiting factors
identified for this species include: (1) Degraded floodplain and mainstem river channel
structure, (2) degraded estuarine water quality conditions and loss of estuarine habitat, (3)
riparian area degradation and loss of in-river wood in mainstem, (4) excessive sediment
in spawning gravels, and (5) reduced stream flow in migration areas. These conditions
also introduce sediment, nutrients, biocides, metals, and other pollutants into surface and
ground water and degrade water quality in the freshwater, estuarine, and coastal
ecosystems throughout the Pacific Northwest.
DRAFT 121
-------
DRAFT
Coho Salmon
Description of the Species
Coho salmon occur naturally in most major river basins around the North Pacific Ocean
from central California to northern Japan (Laufle, Pauley et al. 1986). We discuss the
distribution, life history diversity, status, and critical habitat of the four endangered and
threatened coho species separately.
After entering the ocean, immature coho salmon initially remain in nearshore waters
close to the parent stream. Most coho salmon adults are three-year-olds, having spent
approximately 18 months rearing in freshwater and 18 months in salt water. Most coho
salmon enter rivers between September and February. However, entry is influenced by
discharge and other factors. In many systems, coho salmon and other Pacific salmon are
unable to enter the rivers until sufficiently strong flows open passages and provide
sufficient depth. Wild female coho salmon return to spawn almost exclusively at age
three. Coho salmon spawn from November to January, and occasionally into February
and March. Spawning occurs in a few third-order streams. Most spawning activity
occurs in fourth- and fifth-order streams. Spawning generally occurs in tributaries with
gradients of 3% or less.
Eggs incubate for about 35 to 50 days, and start emerging from the gravel within two to
three weeks after hatching. Following emergence, fry move to shallow areas near the
stream banks. As fry grow, they disperse upstream and downstream to establish and
defend territories. Juvenile rearing usually occurs in tributaries with gradients of 3% or
less, although they may move to streams with gradients of 4 to 5%. Juvenile coho
salmon are often found in small streams less than five ft wide, and may migrate
considerable distances to rear in lakes and off-channel ponds. During the summer, fry
prefer pools featuring adequate cover such as large woody debris, undercut banks, and
overhanging vegetation. Overwintering tends to occur in larger pools and backwater
areas.
North American coho salmon will migrate north along the coast in a narrow coastal band
that broadens in southeastern Alaska. During this migration, juvenile coho salmon tend
to occur in both coastal and offshore waters. During spring and summer, coho salmon
will forage in waters between 46°N, the Gulf of Alaska, and along Alaska's Aleutian
Islands.
Status and Trends
Coho salmon survive only in aquatic ecosystems and depend on the quantity and quality
of those aquatic systems. Coho salmon, like the other salmon NMFS has listed, have
DRAFT 122
-------
DRAFT
declined under the combined effects of overharvests in fisheries; competition from fish
raised in hatcheries and native and non-native exotic species; dams that block their
migrations and alter river hydrology; gravel mining that impedes their migration and
alters the dynamics of the rivers and streams that support juveniles; water diversions that
deplete water levels in rivers and streams; destruction or degradation of riparian habitat
that increase water temperatures in rivers and streams sufficient to reduce the survival of
juvenile chum salmon; and land use practices (logging, agriculture, urbanization) that
destroy wetland and riparian ecosystems. The above activities and features introduce
sediment, nutrients, biocides, metals, and other pollutants into surface and ground water
and degrade water quality in the freshwater, estuarine, and coastal ecosystems throughout
the Pacific Northwest.
Central California Coast Coho Salmon
Distribution
The CCC coho salmon ESU extends from Punta Gorda in northern California south to
and including the San Lorenzo River in central California (Weitkamp, Wainwright et al.
1995). Table 15 identifies populations within the CCC Coho salmon ESU, their
abundances, and hatchery input (Figure 18).
DRAFT 123
-------
124 W
123 W
I
122 W
38' N-
37 N-
'40 N
'38 N
Legend
W Dams
«• NPDES permit sites
- EPA 303(d) rivers
'37 N
I
122 W
0 15 30
60
Prepared by Dwayne Meadows
Kilometers 16 July 2008
Figure 18. CCC Coho salmon distribution. The Legend for the Land Cover Class
categories is found in Figure 7.
DRAFT
124
-------
DRAFT
Table 15. CCC Coho salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005).
Historical 1987-1991
Hatchery
River/Region Escapement Escapement Abundance
(1963) Abundance Contributions
Ten Mile River
Noyo River
Big River
Navarro River
Garcia River
Other Mendacino County rivers
Gualala River
Russian River
Other Sonoma County rivers
Marin County
San Mateo County
Santa Cruz County
San Lorenzo River
Total
6,000
6,000
6,000
7,000
2,000
10,000
4,000
5,000
1,000
5,000
1,000
1,500
1,600
200,000-
500,000
160
3,740
280
300
500(1984-1985)
470
200
255
180
435
Unknown
50(1984-1985)
Unknown
6,570 (min)
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Life History
Both run and spawn timing of coho salmon in this region are very late (both peaking in
January), with little time spent in freshwater between river entry and spawning. This
compressed adult freshwater residency appears to coincide with the single, brief peak of
river flow characteristic of this area.
Status and Trends
The CCC coho salmon ESU was originally listed as threatened under the ESA on October
31, 1996 (61 FR 56138) and later revised to endangered status on June 28, 2005 (70 FR
37160). The ESU includes all naturally spawned populations of coho salmon from Punta
Gorda in northern California south to and including the San Lorenzo River in central
California, as well as populations in tributaries to San Francisco Bay, excluding the
Sacramento-San Joaquin River system. The ESU also includes four artificial propagation
DRAFT
125
-------
DRAFT
programs: the Don Clausen Fish Hatchery Captive Broodstock Program, Scott
Creek/King Fisher Flats Conservation Program, Scott Creek Captive Broodstock
Program, and the Noyo River Fish Station egg-take Program coho hatchery programs.
Information on the abundance and productivity trends for the naturally spawning
component of the CCC coho salmon ESU is extremely limited. There are no long-term
time series of spawner abundance for individual river systems. Analyses of juvenile coho
presence-absence information, juvenile density surveys, and irregular adult counts for the
South Fork Noyo River indicate low abundance and long-term downward trends for the
naturally spawning populations throughout the ESU. Improved ocean conditions coupled
with favorable stream flows and harvest restrictions have contributed to increased returns
in 2001 in streams in the northern portion of the ESU, as indicated by an increase in the
observed presence offish in historically occupied streams. Data are lacking for many
river basins in the southern two thirds of the ESU where naturally spawning populations
are considered at the greatest risk. The extirpation or near extirpation of natural coho
salmon populations in several major river basins, and across most of the southern
historical range of the ESU, represents a significant risk to ESU spatial structure and
diversity. Artificial propagation of coho salmon within the CCC ESU has declined since
the ESU was listed in 1996 though it continues at the Noyo River and Scott Creek
facilities, and two captive broodstock populations have recently been established.
Genetic diversity risk associated with out-of-basin transfers appears to be minimal.
However, diversity risk from domestication selection and low effective population sizes
in the remaining hatchery programs remains a concern. An out-of-ESU artificial
propagation program for coho was operated at the Don Clausen hatchery on the Russian
River through the mid-1990s. However, the program was terminated in 1996.
Termination of this program was considered by the Biological Review Team (BRT) as a
positive development for naturally produced coho salmon in this ESU.
CCC coho salmon populations continue to be depressed relative to historical numbers.
Strong indications show that breeding groups have been lost from a significant
percentage of streams in their historical range. A number of coho salmon populations in
the southern portion of the range appear to be either extinct or nearly so. They include
those in Gualala, Garcia, and Russian rivers, as well as smaller coastal streams in and
south of San Francisco Bay (Good, Waples et al. 2005). For the naturally spawning
component of the ESU, the BRT found very high risk (of extinction) for the abundance,
productivity, and spatial structure VSP parameters and comparatively moderate risk with
respect to the diversity VSP parameter. The lack of direct estimates of the performance
of the naturally spawned populations in this ESU, and the associated uncertainty this
generates, was of specific concern to the BRT. Informed by the VSP risk assessment and
the associated uncertainty, the strong majority opinion of the BRT was that the naturally
spawned component of the CCC coho salmon ESU was "in danger of extinction." The
DRAFT 126
-------
DRAFT
minority opinion was that this ESU is "likely to become endangered within the
foreseeable future" (70 FR 37160). Based on these conclusions, NMFS granted
endangered status for this ESU on June 28, 2005 (70 FR 37160).
Critical Habitat
Critical habitat for the CCC coho salmon ESU was designated on May 5, 1999 (64 FR
24049). Designated critical habitat encompasses accessible reaches of all rivers
(including estuarine areas and tributaries) between Punta Gorda and the San Lorenzo
River (inclusive) in California. Critical habitat for this species also includes two streams
entering San Francisco Bay: Arroyo Corte Madera Del Presidio and Corte Madera
Creek.
Lower Columbia River Coho Salmon
Distribution
LCR coho salmon include all naturally spawned populations of coho salmon in the
Columbia River and its tributaries in Oregon and Washington, from the mouth of the
Columbia up to and including the Big White Salmon and Hood Rivers, and includes the
Willamette River to Willamette Falls, Oregon (Figure 19). This ESU also includes 25
artificial propagation programs: the Gray's River, Sea Resources Hatchery, Peterson
Coho Project, Big Creek Hatchery, Astoria High School Coho Program, Warrenton High
School Coho Program, Elochoman Type-S Coho Program, Elochoman Type-N Coho
Program, Cathlamet High School FFA Type-N Coho Program, Cowlitz Type-N Coho
Program in the Upper and Lower Cowlitz Rivers, Cowlitz Game and Anglers Coho
Program, Friends of the Cowlitz Coho Program, North Fork Toutle River Hatchery,
Kalama River Type-N Coho Program, Kalama River Type-S Coho Program, Washougal
Hatchery Type-N Coho Program, Lewis River Type-N Coho Program, Lewis River
Type-S Coho Program, Fish First Wild Coho Program, Fish First Type-N Coho Program,
Syverson Project Type-N Coho Program, Eagle Creek National Fish Hatchery, Sandy
Hatchery, and the Bonneville/Cascade/Oxbow complex coho hatchery programs.
DRAFT 127
-------
124 W
123 W
I
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
124 W
Kilometers
Prepared by Dwayne Meadows
16 July 2008
Figure 19 . LCR coho salmon distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
128
-------
DRAFT
Table 16 identifies populations within the LCR Coho salmon ESU, their abundances, and
hatchery input.
Table 16. LCR Coho salmon populations, abundances, and hatchery contributions
(Good, Waples et al. 2005).
River/Region
Youngs Bay and Big Creek
Gray's River
Elochoman River
Clatskanie River
Mill, Germany, and Abernathy
creeks
Scappoose Rivers
Cispus River
Tilton River
Upper Cowlitz River
Lower Cowlitz River
North Fork Toutle River
South Fork Toutle River
Coweeman River
Kalama River
North Fork Lewis River
East Fork Lewis River
Upper Clackamas River
Lower Clackamas River
Salmon Creek
Upper Sandy River
Lower Sandy River
Washougal River
LCR gorge tributaries
White Salmon
Upper Columbia River gorge
tributaries
Hood River
Total
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
2002
Spawner
Abundance
4,473
Unknown
Unknown
229
Unknown
458
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
1,001
2,402
Unknown
310
271
Unknown
Unknown
Unknown
1,317
Unknown
10,461 (min)
Hatchery
Abundance
Contributions
91%
Unknown
Unknown
60%
Unknown
0%
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
12%
78%
Unknown
0%
97%
Unknown
Unknown
Unknown
>65%
Unknown
Life History
Although run time variation is inherent to coho salmon life history, the ESU includes two
distinct runs: early returning (Type S) and late returning (Type N). Type S coho salmon
DRAFT
129
-------
DRAFT
generally migrate south of the Columbia once they reach the ocean, returning to
freshwater in mid-August and to the spawning tributaries in early September. Spawning
peaks from mid-October to early November. Type N coho salmon have a northern
distribution in the ocean, return to the Columbia River from late September through
December and enter the tributaries from October through January. Most Type N
spawning occurs from November through January. However some spawning occurs in
February and as late as March (LCFRB 2004). Almost all LCR ESU coho salmon
females and most males spawn at three years of age.
Status and Trends
LCR coho salmon were listed as endangered on June 28, 2005 (70 FR 37160). The vast
majority (over 90%) of the historic population in the LCR coho salmon ESU appear to be
either extirpated or nearly so. The two populations with any significant natural
production (Sandy and Clackamas) are at appreciable risk because of low abundance,
declining trends, and failure to respond after a dramatic reduction in harvest. Most of the
other populations are believed to have very little, if any, natural production.
The Sandy population had a recent mean abundance of 342 spawners and a very low
fraction of hatchery-origin spawners. Trends in the Sandy are similar to the Clackamas.
The long-term trends and growth rate estimates over the period 1977 to 2001 have been
slightly positive and the short-term trends have been slightly negative. Other populations
in this ESU are dominated by hatchery production. There is very little, if any, natural
production in Oregon beyond the Clackamas and Sandy rivers. The Washington side of
the ESU is also dominated by hatchery production. There are no populations with
appreciable natural production. The most serious threat facing this ESU is the scarcity of
naturally-produced spawners, with attendant risks associated with small population, loss
of diversity, and fragmentation and isolation of the remaining naturally-produced fish. In
the only two populations with significant natural production (Sandy and Clackamas),
short- and long-term trends are negative and productivity (as gauged by pre-harvest
recruits) is down sharply from recent (1980s) levels.
The Federal Columbia River Power System Opinion (FCRPS) (2008) describes this ESU
as consisting of three MPGs. Each is comprised of three to 14 populations. In many
cases, populations have low abundance and natural runs have been extensively replaced
by hatchery production. Abundance estimates are available for only five populations and
trend estimates for only two. Time series are not available for Washington coho
populations. The 100-year risk of extinction was derived qualitatively, based on risk
categories and criteria identified by the W/LCTRT in 2004. Most of the population of
LCR had high or very high extinction risk probabilities. Spatial structure has been
substantially reduced by the loss of access to the upper portions of some basins from
DRAFT 130
-------
DRAFT
tributary hydro development (i.e., Condit Dam on the Big White Salmon River and
Powerdale Dam on the Hood River). Finally, the diversity of populations in all three
MPGs has been eroded by large hatchery influences and periodically, low effective
population sizes. Nevertheless, the genetic legacy of the Lewis and Cowlitz River coho
salmon populations is preserved in ongoing hatchery programs.
Critical Habitat
NMFS has not designated critical habitat for LCR coho salmon.
Southern Oregon/Northern California Coast Coho Salmon
Distribution
Southern Oregon/Northern California Coast coho salmon consists of all naturally
spawning populations of coho salmon that reside below long-term, naturally impassible
barriers in streams between Punta Gorda, California and Cape Blanco, Oregon (Figure
20).
This ESU also includes three artificial propagation programs: the Cole Rivers Hatchery
(ODFW stock #52), Trinity River Hatchery, and Iron Gate Hatchery coho hatchery
programs. The three major river systems supporting Southern Oregon / Northern Coastal
California coast coho are the Rogue, Klamath (including the Trinity), and Eel rivers.
Life History
Southern Oregon/Northern California Coast coho salmon enter rivers in September or
October. River entry is much later south of the Klamath River Basin, occurring in
November and December, in basins south of the Klamath River to the Mattole River,
California. River entry occurs from mid-Decmeber to mid-February in rivers farther
south. Because coho salmon enter rivers late and spawn late south of the Mattole River,
they spend much less time in the river prior to spawning. Coho salmon adults spawn at
age three, spending just over a year in freshwater and a year and a half in the ocean.
DRAFT 131
-------
124 W
I
123 W
J_
42 N-
Legend
rf Dams
*• NPDES permit sites
EPA 303(d) rivers
43 N
Prepared by Dwayne Meadows
0 15 30 60
Figure 20. Southern Oregon/Northern California Coast coho salmon distribution.
The Legend for the Land Cover Class categories is found in Figure 7.
Status and Trends
DRAFT
132
-------
DRAFT
Southern Oregon/Northern California Coast coho salmon were listed as threatened on
May 7, 1997 (62 FR 24588). This species retained its original classification when its
status was reviewed on June 28, 2005 (70 FR 37160). The status of coho salmon coast
wide, including the Southern Oregon/Northern California Coast coho salmon ESU, was
formally assessed in 1995 (Weitkamp, Wainwright et al. 1995). Two subsequent status
review updates have been published by NMFS. One review update addressed all West
Coast coho salmon ESUs (Busby, Wainwright et al. 1996). The second update
specifically addressed the Oregon Coast and Southern Oregon/Northern California Coast
coho salmon ESUs (Gustafson, Wainwright et al. 1997). In the 1997 status update,
estimates of natural population abundance were based on very limited information. New
data on presence/absence in northern California streams that historically supported coho
salmon were even more disturbing than earlier results. Data indicated that a smaller
percentage of streams contained coho salmon compared to the percentage presence in an
earlier study. However, it was unclear whether these new data represented actual trends
in local extinctions, or were biased by sampling effort.
Data on population abundance and trends are limited for the California portion of this
ESU. No regular estimates of natural spawner escapement are available. Historical point
estimates of coho salmon abundance for the early 1960s and mid-1980s suggest that
statewide coho spawning escapement in the 1940s ranged between 200,000 and 500,000
fish. Numbers declined to about 100,000 fish by the mid-1960s with about 43%
originating from this ESU. Brown et al. (1994) estimated that the California portion of
this ESU was represented by about 7,000 wild and naturalized coho salmon (Good,
Waples et al. 2005). In the Klamath River, the estimated escapement has dropped from
approximately 15,400 in the mid-1960s to about 3,000 in the mid-1980s, and more
recently to about 2,000 (Good, Waples et al. 2005). The second largest producing river
in this ESU, the Eel River, dropped from 14,000, to 4,000 to about 2,000 during the same
period. Historical estimates are considered "best guesses" made using a combination of
limited catch statistics, hatchery records, and the personal observations of biologists and
managers.
Most recently, Williams et al. (2006) described the structure of historic populations of
Southern Oregon/Northern California Coast coho salmon. They described three
categories of populations: functionally independent populations, potentially independent
populations, and dependent populations. Functionally independent populations are
populations capable of existing in isolation with a minimal risk of extinction. Potentially
independent populations are similar but rely on some interchange with adjacent
populations to maintain a low probability of extinction. Dependent populations have a
high risk of extinction in isolation over a 100-year timeframe and rely on exchange of
individuals from adjacent populations to maintain themselves.
DRAFT 133
-------
DRAFT
Critical Habitat
Critical habitat was designated for the Southern Oregon/Northern California Coast coho
salmon on November 25, 1997, and re-designated on May 5, 1999. Species critical
habitat encompasses all accessible river reaches between Cape Blanco, Oregon, and
Punta Gorda, California and consists of the water, substrate, and river reaches (including
off-channel habitats) in specified areas. Accessible reaches are those within the historical
range of the ESU that can still be occupied by any life stage of coho salmon. Of 155
historical streams for which data are available, 63% likely still support coho salmon.
Limiting factors identified for this species include: (1) Loss of channel complexity,
connectivity and sinuosity, (2) loss of floodplain and estuarine habitats, (3) loss of
riparian habitats and large in-river wood, (4) reduced streamflow, (5) poor water quality,
temperature and excessive sedimentation, and (6) unscreened diversions and fish passage
structures.
Oregon Coast Coho Salmon
Distribution
The Oregon Coast (OC) coho salmon ESU includes all naturally spawned populations of
coho salmon in Oregon coastal streams south of the Columbia River and north of Cape
Blanco (63 FR 42587; August 10, 1998; Figure 21). One hatchery stock, the Cow Creek
(ODFW stock # 37) hatchery coho, is considered part of the ESU. Table 17 identifies
populations within the OC coho salmon ESU, their abundances, and hatchery input.
DRAFT 134
-------
46 N-
45 N-
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
Prepared by Dv/ayne Meadows
•46 N
•45 N
•43 N
Figure 21. Oregon Coast Coho salmon distribution. The Legend for the Land
Cover Class categories is found in Figure 7.
DRAFT
135
-------
DRAFT
Table 17. Oregon Coast Coho salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005).
Basin
Necanicum
Nehalem
Tillamook
Nestucca
Siletz
Yaquima
Alsea
Siuslaw
Umpqua
Coos
Coquille
Total
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
924,000
Recent
Spawner
Abundance
1,889
18,741
3,949
3,846
2,295
3,665
3,621
16,213
24,351
20,136
8,847
107,553
Hatchery
Abundance
Contributions
35-40%
40-75%
30-35%
-5%
-50%
-25%
-40%
-40%
<10%
<5%
<5%
Status and Trends
The OC coho salmon ESU was listed as a threatened species on February 11, 2008 (73
FR 7816). The most recent NMFS status review for the OC coho salmon ESU was
conducted by the BRT in 2003, which assessed data through 2002. The abundance and
productivity of OC coho salmon since the previous status review (Gustafson, Wainwright
et al. 1997) represented some of the best and worst years on record. Yearly adult returns
for this ESU were in excess of 160,000 natural spawners in 2001 and 2002, far exceeding
the abundance observed for the past several decades. These encouraging increases in
spawner abundance in 2000-2002 were preceded, however, by three consecutive brood
years (the 1994-1996 brood years returning in 1997-1999, respectively) exhibiting
recruitment failure. Recruitment failure is when a given year class of natural spawners
fails to replace itself when its offspring return to the spawning grounds three years later.
These three years of recruitment failure were the only such instances observed thus far in
the entire 55-year abundance time series for OC coho salmon (although comprehensive
population-level survey data have only been available since 1980). The encouraging
DRAFT
136
-------
DRAFT
2000-2002 increases in natural spawner abundance occurred in many populations in the
northern portion of the ESU, which were the most depressed at the time of the last review
(Gustafson, Wainwright et al. 1997). Although encouraged by the increase in spawner
abundance in 2000-2002, the BRT noted that the long-term trends in ESU productivity
were still negative due to the low abundances observed during the 1990s (73 FR 7816).
Since the BRT convened, the total abundance of natural spawners in the OC coho salmon
ESU has declined each year (i.e., 2003-2006). The abundance of total natural spawners
in 2006 (111,025 spawners) was approximately 43% of the recent peak abundance in
2002 (255,372 spawners). In 2003, ESU-level productivity (evaluated in terms of the
number of spawning recruits resulting from spawners three years earlier) was above
replacement, and in 2004, productivity was approximately at replacement level.
However, productivity was below replacement in 2005 and 2006, and dropped to the
lowest level since 1991 in 2006.
Preliminary spawner survey data for 2007 (the average peak number of spawners per
mile observed during random coho spawning surveys in 41 streams) suggest that the
2007-2008 return of Oregon Coast coho salmon is either: (1) much reduced from
abundance levels in 2006, or (2) exhibiting delayed run timing from previous years. As
of December 13, 2007, the average peak number of spawners per mile was below 2006
levels in 38 of 41 surveyed streams (ODFW 2007 in 73 FR 7816). It is possible that the
timing of peak spawner abundance is delayed relative to previous years, and that
increased spawner abundance in late December and January 2008 will compensate for the
low levels observed thus far.
The recent five year geometric mean abundance (2002-2006) of approximately 152, 960
total natural spawners remains well above that of a decade ago (approximately 52,845
from 1992-1996). However, the decline in productivity from 2003 to 2006, despite
generally favorable marine survival conditions and low harvest rates, is of concern. (73
FR 7816). The long-term trends in productivity in this ESU remain strongly negative.
Critical Habitat
Critical habitat was proposed for Oregon Coast coho salmon on December 14, 2004 (69
FR 74578). The final designation of critical habitat is included in the final rule published
on February 11, 2008 (73 FR 7816). Approximately 6,568 stream miles (10,570 km) and
15 square miles (38.8 sq km) of lake habitat are designated critical habitat. Refer to the
final rule for a detailed description of the watersheds included in the critical habitat, and a
map for each subbasin.
DRAFT 137
-------
DRAFT
Sockeye Salmon
Description of the Species
Sockeye salmon occur in the North Pacific and Arctic oceans and associated freshwater
systems. This species ranges south as far as the Klamath River in California and northern
Hokkaido in Japan, to as far north as far as Bathurst Inlet in the Canadian Arctic and the
Anadyr River in Siberia. We discuss the distribution, life history diversity, status, and
critical habitat of the two endangered and threatened sockeye species separately.
The species exhibits riverine and lake life history strategies, the latter of which may be
either freshwater resident forms or anadromous forms. The vast majority of sockeye
salmon spawn in outlet streams of lakes or in the lakes themselves. These "lake-type"
sockeye use the lake environment for rearing for up to three years and then migrate to
sea, returning to their natal lake to spawn after one to four years at sea. Some sockeye
spawn in rivers, however, without lake habitat for juvenile rearing. Offspring of these
riverine spawners tend to use the lower velocity sections of rivers as the juvenile rearing
environment for one to two years, or may migrate to sea in their first year.
Certain populations of O. nerka become resident in the lake environment over long
periods of time and are called kokanee or little redfish (Burgner 1991). Kokanee and
sockeye often co-occur in many interior lakes, where access to the sea is possible but
energetically costly. On the other hand, coastal lakes where the migration to sea is
relatively short and energetic costs are minimal, rarely support kokanee populations.
Spawning generally occurs in late summer and autumn, but the precise time can vary
greatly among populations. Males often arrive earlier than females on the spawning
grounds, and will persist longer during the spawning period. Average fecundity ranges
from about 2,000 to 2,400 eggs per female to 5,000 eggs, depending upon the population
and average age of the female. Fecundity in kokanee is much lower and may range from
about 300 to less than 2,000 eggs.
Incubation is a function of water temperatures, but generally lasts between 100 and
roughly 200 days (Burgner 1991). After emergence, fry move rapidly downstream or
upstream along the banks to the lake rearing area. Fry emerging from lakeshore or island
spawning grounds may simply move along the shoreline of the lake (Burgner 1991).
Juvenile salmonids rely on a variety of non-main channel habitats that are critical to
DRAFT 138
-------
DRAFT
rearing. All listed salmonids use shallow, low flow habitats at some point in their life
cycle. Examples of off-channel habitat include alcoves, channel edge sloughs, overflow
channels, backwaters, terrace tributaries, off-channel dredge ponds, and braids (Anderson
1999; Swift III 1979).
Sockeye salmon survive only in aquatic ecosystems and depend on the quantity and
quality of those aquatic systems. Sockeye salmon, like the other salmon NMFS has
listed, have declined under the combined effects of overharvests in fisheries; competition
from fish raised in hatcheries and native and non-native exotic species; dams that block
their migrations and alter river hydrology; gravel mining that impedes their migration and
alters the hydrogeomorphology of the rivers and streams that support juveniles; water
diversions that deplete water levels in rivers and streams; destruction or degradation of
riparian habitat that increase water temperatures in rivers and streams sufficient to reduce
the survival of juvenile chum salmon; and land use practices (logging, agriculture,
urbanization) that destroy wetland and riparian ecosystems. These activities and features
introduce sediment, nutrients, biocides, metals, and other pollutants into surface and
ground water and degrade water quality in the freshwater, estuarine, and coastal
ecosystems throughout the Pacific Northwest.
Ozette Lake Sockeye Salmon
Distribution
This ESU includes all naturally spawned populations of sockeye salmon in Ozette Lake,
Ozette River, Coal Creek, and other tributaries flowing into Ozette Lake, Washington.
This ESU is composed of one historical population, with substantial sub structuring of
individuals into multiple spawning aggregations (Figure 22). The primary spawning
aggregations occur in two beach locations - Allen's and Olsen's beaches, and in two
tributaries Umbrella Creek and Big River (both tributary-spawning groups were initiated
through a hatchery introduction program).
Sockeye salmon stock reared at the Makah Tribe's Umbrella Creek Hatchery were
considered part of the ESU, but were not considered essential for recovery of the ESU.
NMFS determined that it is presently not necessary to consider the progeny of intentional
hatchery-wild or wild-wild crosses produced through the Makah Tribal hatchery program
as listed under the ESA (March 25, 1999, 64 FR 14528). However, once the hatchery
fish return and spawn in the wild, their progeny are considered listed.
DRAFT 139
-------
124
NPDES permit sites
EPA 303(d) rivers
124 45'W
124 30'W
0 2.5 5 10 Prepared by Dwayne Meadows
I Kilometers f 6 Jury 2008
Figure 22. Ozette Lake Sockeye salmon distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
140
-------
DRAFT
Life History
The sockeye salmon life history is one of the most complex of any Pacific salmon species
because of its variable freshwater residency (one to three years in freshwater), and
because the species has several different forms: fish that go to the ocean and back, fish
that remain in freshwater, and fish that do both.
Adult Ozette Lake sockeye salmon enter Ozette Lake through the Ozette River from
April to early August. Adults remain in the lake for an extended period of time (return
April - August; spawn late October-February) before spawning on beaches or in the
tributaries. Sockeye salmon spawn primarily in lakeshore upwelling areas in Ozette Lake
(at Allen's Bay and Olsen's Beach). Minor spawning may occur below Ozette Lake in
the Ozette River or in Coal Creek, a tributary of the Ozette River. Sockeye salmon do
not presently spawn in tributary streams to Ozette Lake. However, they may have
spawned there historically. Eggs and alevins remain in gravel redds until the fish emerge
as fry in spring. Fry then migrate immediately to the limnetic zone in Ozette Lake, where
the fish rear. After one year of rearing, in late spring, Ozette Lake sockeye salmon
emigrate seaward as one + smolts. The majority of Ozette Lake sockeye salmon return to
spawn as four year old adult fish, having spent one winter in fresh water and two winters
at sea (NMFS 2005b). As prespawning mortality is unknown, it is unclear what
escapement levels to the spawning aggregations may be.
In Ozette Lake, naturally high water temperatures and low summer flows in the Ozette
River may affect migration by altering timing of the runs (La Riviere 1991). Declines in
abundance have been attributed to a combination of introduced species, predation, loss of
tributary populations, decline in quality of beach spawning habitat, temporarily
unfavorable ocean conditions, habitat degradation, and excessive historical harvests
(Jacobs, Larson et al. 1996).
Status and Trends
The Ozette Lake sockeye salmon ESU was originally listed as a threatened species in
1999 (64 FR 14528). This classification was retained following a species status review
on June 28, 2005 (70 FR 37160).
The historical abundance of Ozette Lake sockeye salmon is poorly documented, but may
have been as high as 50,000 individuals (Blum 1988). Nevertheless, the overall
abundance of naturally-produced Ozette Lake sockeye salmon is believed to have
declined substantially from historical levels. In the first study of lake escapement of
Ozette Lake sockeye salmon (Kemmerich 1945), the run size entering the lake was
estimated at a level of several thousand fish. These counts appear to be roughly double
DRAFT 141
-------
DRAFT
the current mean lake abundance, considering that they were likely conducted upstream
from fisheries in or near to the Ozette River. Makah Fisheries Management (2006)
concluded that there appears to be a substantial decline in the Tribal catch of Ozette Lake
sockeye salmon beginning in the 1950s and a similar decline in the run size since the
1920s weir counts reported by Kemmerich (1945).
An updated NMFS analysis of total annual Ozette Lake sockeye salmon abundance
(based on adult run size data presented in Jacobs et al. (1996) indicates a trend in
abundance averaging minus 2% per year over the period 1977 through 1998 (Meyers,
Kope et al. 1998). The current tributary-based hatchery program was planned and
initiated in response to the declining population trend identified for the Ozette Lake
sockeye salmon population. The updated analysis also indicated that the most recent ten
year (1989-98) trend for the population is plus 2% per year (Meyers, Kope et al. 1998),
improving from the minus 9.9% annual trend reported in Gustafson et al. (1999).
Data from the early 1900s indicate the spawning population was as large as 10,000 to
20,000 fish in large run years. Recent information on abundance of Ozette Lake sockeye
salmon ESU comes from visual counts at a weir across the lake outlet. Therefore, the
counts represent total run size. The estimates of total run size were revised upward after
the 1997 status review due to resampling of data using new video counting technology.
The Makah Fisheries biologists estimate that previous counts of adult sockeye salmon
returning to the lake were underestimates, and they have attempted to correct run-size
estimates based on their assessments of human error and variations in interannual run
timing (Makah Fisheries Management 2000) in (Good, Waples et al. 2005).
The most recent (1996-2003) run-size estimates range from a low of 1,609 in 1997 to a
high of 5,075 in 2003, averaging approximately 3,600 sockeye per year (Hard, Jones et
al. 1992; Haggerty, Ritchie et al. 2007). For return years 2000 to 2003, the four-year
average abundance estimate was slightly over 4,600 sockeye (Haggerty, Ritchie et al.
2007). Because run-size estimates before 1998 are likely to be even more unreliable than
recent counts, and new counting technology has resulted in an increase in estimated run
sizes, no statistical estimation of trends is reported. The current trends in abundance are
unknown for the beach spawning aggregations. Although overall abundance appears to
have declined from historical levels, whether this resulted in fewer spawning
aggregations, lower abundances at each aggregation, or both, is unknown (Good, Waples
et al. 2005). It is estimated that between 35,500 and 121,000 spawners could be normally
carried after full recovery (Hard, Jones et al. 1992).
There has been no harvest of Ozette Lake sockeye salmon for the past four brood cycle
years (since 1982). Prior to that time, ceremonial and subsistence harvests by the Makah
DRAFT 142
-------
DRAFT
Tribe were low, ranging from 0 to 84 fish per year. Harvest has not been an important
mortality factor for the population in over 35 years. In addition, due to the early river
entry timing of returning Ozette Lake sockeye salmon (beginning in late April, with the
peak returns prior to late-May to mid-June), the fish are not intercepted in Canadian and
U.S. marine area fisheries directed at Fraser River sockeye salmon. There are currently
no known marine area harvest impacts on Ozette Lake sockeye salmon.
According to Good et al. (2006) it appears that overall abundance is low for this
population, which represents an entire ESU, and may be substantially below historical
levels. The number of returning adults in the last few years has increased. However, a
substantial (but uncertain) fraction of these appear to be of hatchery origin. This
condition leads to uncertainty regarding growth rate and productivity of the natural
component of the ESU. Genetic integrity may have been compromised due to the
artificial supplementation that has occurred in this population. Approximately one
million sockeye have been released into the Ozette watershed from the late 1930s to
present (Kemmerich 1945; Boomer 1995; Good, Waples et al. 2005).
Critical Habitat
On September 2, 2005, NMFS designated critical habitat for the Ozette Lake sockeye
salmon ESU (70 FR 52630), and encompasses areas within the Hoh/Quillayute subbasin.
Refer to the final rule for additional information on the watersheds within this subbasin,
including a map of the area. Limiting factors for this species include siltation of beach-
spawning habitat and logging.
Snake River Sockeye Salmon
Distribution
The SR sockeye salmon ESU includes all anadromous and residual sockeye from the SR
basin Idaho, as well as artificially propagated sockeye salmon from the Redfish Lake
Captive Broodstock Program (Figure 23).
Life History
SR sockeye salmon are unique compared to other sockeye salmon populations. Sockeye
salmon returning to Redfish Lake in Idaho's Stanley Basin travel a greater distance from
the sea (approximately 900 miles) to a higher elevation (6,500 ft) than any other sockeye
salmon population and are the southern-most population of sockeye salmon in the world
(Bjornn, Craddock et al. 1968). Stanley Basin sockeye salmon are separated by 700 or
more river miles from two other extant upper Columbia River populations in the
DRAFT 143
-------
DRAFT
Wenatchee River and Okanogan River drainages. These latter populations return to lakes
at substantially lower elevations (Wenatchee at 1,870 ft, Okanagon at 912 ft) and occupy
different ecoregions.
DRAFT 144
-------
119W
I
^•~^v
fc«fif • d^^sf: «^* t
i* «rifi* ii-^
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
Migratory corridor
114 W
Prepared by Dwayne Meadows
Kilometers 15 J«lv 2008
Figure 23. SR Sockeye Salmon distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
145
-------
DRAFT
Status and Trends
SR sockeye salmon were originally listed as endangered in 1991. Their classification
was retained following a status review on June 28, 2005 (70 FR 37160). The only extant
sockeye salmon population in the SR basin at the time of listing was that in Redfish Lake,
in the Stanley Basin (upper Salmon River drainage) of Idaho. Other lakes in the SR basin
historically supported sockeye salmon populations, including Wallowa Lake (Grande
Ronde River drainage, Oregon), Payette Lake (Payette River drainage, Idaho) and Warm
Lake (South Fork Salmon River drainage, Idaho) (Gustafson, Wainwright et al. 1997).
These populations are now considered extinct. Although kokanee, a resident form of O.
nerka, occur in numerous lakes in the SR basin, resident O. nerka were not considered
part of the species at the time of listing in 1991. Subsequent to the 1991 listing, a
residual form of sockeye residing in Redfish Lake was identified. The residuals are non-
anadromous. They complete their entire life cycle in freshwater, but spawn at the same
time and in the same location as anadromous sockeye salmon. In 1993, NMFS
determined that residual sockeye salmon in Redfish Lake were part of the SR sockeye
salmon. Also, artificially propagated sockeye salmon from the Redfish Lake Captive
Propagation program are considered part of this species (June 28, 2005, 70 FR 37160).
NMFS has determined that this artificially propagated stock is genetically no more than
moderately divergent from the natural population (Good, Waples et al. 2005). Five lakes
in the Stanley Basin historically contained sockeye salmon: Alturas, Pettit, Redfish,
Stanley and Yellowbelly (Bjornn, Craddock et al. 1968). It is generally believed that
adults were prevented from returning to the Sawtooth Valley from 1910 to 1934 by
Sunbeam Dam. Sunbeam Dam was constructed on the Salmon River approximately 20
miles downstream of Redfish Lake. Whether or not Sunbeam Dam was a complete
barrier to adult migration remains unknown. It has been hypothesized that some passage
occurred while the dam was in place, allowing the Stanley Basin population or
populations to persist (Bjornn, Craddock et al. 1968; Matthews and Waples 1991).
Adult returns to Redfish Lake during the period 1954 through 1966 ranged from 11 to
4,361 fish (Bjornn, Craddock et al. 1968). Sockeye salmon in Alturas Lake were
extirpated in the early 1900s as a result of irrigation diversions, although residual sockeye
may still exist in the lake (Chapman and Witty 1993). From 1955 to 1965, the Idaho
Department of Fish and Game eradicated sockeye salmon from Pettit, Stanley, and
Yellowbelly lakes, and built permanent structures on each of the lake outlets that
prevented re-entry of anadromous sockeye salmon (Chapman and Witty 1993). In 1985,
1986, and 1987, 11, 29, and 16 sockeye, respectively, were counted at the Redfish Lake
weir (Good, Waples et al. 2005). Only 18 natural origin sockeye salmon have returned to
the Stanley Basin since 1987. The first adult returns from the captive brood stock
program returned to the Stanley Basin in 1999. From 1999 through 2005, a total of 345
DRAFT 146
-------
DRAFT
captive brood program adults that had migrated to the ocean returned to the Stanley
Basin.
Recent annual abundances of natural origin sockeye salmon in the Stanley Basin have
been extremely low. No natural origin anadromous adults have returned since 1998 and
the abundance of residual sockeye salmon in Redfish Lake is unknown. This species is
entirely supported by adults produced through the captive propagation program at the
present time. Current smolt-to-adult survival of sockeye originating from the Stanley
Basin lakes is rarely greater than 0.3% (Hebdon, Kline et al. 2004). Based on current
abundance and productivity information, the SR sockeye salmon ESU does not meet the
ESU-level viability criteria (non-negligible risk of extinction over a 100-year time
period).
Critical Habitat
Critical habitat for these salmon was designated on December 28, 1993 (58 FR 68543).
Designated habitats encompasses the waters, waterway bottoms, and adjacent riparian
zones of specified lakes and river reaches in the Columbia River that are or were
accessible to listed SR salmon (except reaches above impassable natural falls, and
Dworshak and Hells Canyon Dams). Adjacent riparian zones are defined as those areas
within a horizontal distance of 300 ft from the normal line of high water of a stream
channel or from the shoreline of a standing body of water. Designated critical habitat
areas include the Columbia River from a straight line connecting the west end of the
Clatsop jetty (Oregon side) and the west end of the Peacock jetty (Washington side), all
river reaches from the estuary upstream to the confluence of the SR, and all SR reaches
upstream to the confluence of the Salmon River; all Salmon River reaches to Alturas
Lake Creek; Stanley, Redfish, yellow Belly, Pettit, and Alturas Lakes (including their
inlet and outlet creeks); Alturas Lake Creek and that portion of Valley Creek between
Stanley Lake Creek; and the Salmon River. Limiting factors identified for SR sockeye
include: (1) Reduced tributary stream flow, (2) impaired tributary passage and blocks to
migration, (3) degraded water quality; and (4) mainstem Columbia River hydropower
system mortality.
Steelhead
Description of the Species
Steelhead are native to Pacific Coast streams extending from Alaska south to
DRAFT 147
-------
DRAFT
northwestern Mexico (Moyle 1976; Gustafson, Wainwright et al. 1997; Good, Waples et
al. 2005). We discuss the distribution, life history diversity, status, and critical habitat of
the 11 endangered and threatened steelhead species separately.
Steelhead can be divided into two basic run-types: the stream-maturing type, or summer
steelhead and the ocean-maturing type, or winter steelhead. The stream-maturing type or
summer steelhead enters fresh water in a sexually immature condition. It requires several
months in freshwater to mature and spawn. The ocean-maturing type or winter steelhead
enters freshwater with well-developed gonads and spawns shortly after river entry.
Variations in migration timing exist between populations. Some river basins have both
summer and winter steelhead, while others only have one run-type.
Summer steelhead enter freshwater between May and October in the Pacific Northwest
(Nickelsen, Nicholas et al. 1992; Busby, Wainwright et al. 1996). They require cool,
deep holding pools during summer and fall, prior to spawning (Nickelsen, Nicholas et al.
1992). They migrate inland toward spawning areas, overwinter in the larger rivers,
resume migration in early spring to natal streams, and then spawn (Meehan and Bjornn
1991; Nickelsen, Nicholas et al. 1992) in January and February (Barnhart 1986). Winter
steelhead enter freshwater between November and April in the Pacific Northwest
(Nickelsen, Nicholas et al. 1992; Busby, Wainwright et al. 1996), migrate to spawning
areas, and then spawn, generally in April and May (Barnhart 1986). Some adults,
however, do not enter some coastal streams until spring, just before spawning (Meehan
and Bjornn 1991).
There is a high degree of overlap in spawn timing between populations regardless of run
type (Busby, Wainwright et al. 1996). Difficult field conditions at that time of year and
the remoteness of spawning grounds contribute to the relative lack of specific information
on steelhead spawning. Unlike Pacific salmon, steelhead are iteroparous, or capable of
spawning more than once before death (Busby, Wainwright et al. 1996), although
steelhead rarely spawn more than twice before dying; most that do so are females
(Nickelsen, Nicholas et al. 1992). Iteroparity is more common among southern steelhead
populations than northern populations (Busby, Wainwright et al. 1996).
After two to three weeks, in late spring, and following yolk sac absorption, alevins
emerge from the gravel and begin actively feeding. After emerging from the gravel, fry
usually inhabit shallow water along banks of perennial streams. Fry occupy stream
margins (Nickelsen, Nicholas et al. 1992). Summer rearing takes place primarily in the
faster parts of pools, although young-of-the-year are abundant in glides and riffles.
Winter rearing occurs more uniformly at lower densities across a wide range of fast and
slow habitat types. Some older juveniles move downstream to rear in larger tributaries
and mainstem rivers (Nickelsen, Nicholas et al. 1992).
DRAFT 148
-------
DRAFT
Juvenile steelhead migrate little during their first summer and occupy a range of habitats
featuring moderate to high water velocity and variable depths (Bisson, Sullivan et al.
1988). Juvenile steelhead feed on a wide variety of aquatic and terrestrial insects
(Chapman and Bjornn 1969), and older juveniles sometimes prey on emerging fry.
Steelhead hold territories close to the substratum where flows are lower and sometimes
counter to the main stream; from these, they can make forays up into surface currents to
take drifting food (Kalleberg 1958). Juveniles rear in freshwater from one to four years,
then smolt and migrate to the ocean in March and April (Barnhart 1986). Winter
steelhead juveniles generally smolt after two years in freshwater (Busby, Wainwright et
al. 1996). Juvenile steelhead tend to migrate directly offshore during their first summer
from whatever point they enter the ocean rather than migrating along the coastal belt as
salmon do. During the fall and winter, juveniles move southward and eastward (Hartt
and Dell 1986) op. cit. (Nickelsen, Nicholas et al. 1992). Steelhead typically reside in
marine waters for two or three years prior to returning to their natal stream to spawn as
four or five year olds. Juvenile salmonids rely on a variety of non-main channel habitats
that are critical to rearing. All listed salmonids use shallow, low flow habitats at some
point in their life cycle. Examples of off-channel habitat include alcoves, channel edge
sloughs, overflow channels, backwaters, terrace tributaries, off-channel dredge ponds,
and braids (Anderson 1999; Swift III 1979).
Status and Trends
Steelhead, like the other salmon discussed previously, survive only in aquatic ecosystems
and, therefore, depend on the quantity and quality of those aquatic systems. Steelhead,
like the other salmon NMFS has listed, have declined under the combined effects of
overharvests in fisheries; competition from fish raised in hatcheries and native and non-
native exotic species; dams that block their migrations and alter river hydrology; gravel
mining that impedes their migration and alters the hydrogeomorphology of the rivers and
streams that support juveniles; water diversions that deplete water levels in rivers and
streams; destruction or degradation of riparian habitat that increase water temperatures in
rivers and streams sufficient to reduce the survival of juvenile chum salmon; and land use
practices (logging, agriculture, urbanization) that destroy wetland and riparian
ecosystems. These same activities and features introduce sediment, nutrients, biocides,
metals, and other pollutants into surface and ground water and degrade water quality in
the freshwater, estuarine, and coastal ecosystems throughout the Pacific Northwest.
Central California Coast Steelhead
Distribution
DRAFT 149
-------
DRAFT
The CCC steelhead DPS includes all naturally spawned anadromous O. mykiss
(steelhead) populations below natural and manmade impassable barriers in California
streams from the Russian River (inclusive) to Aptos Creek (inclusive), and the drainages
of San Francisco, San Pablo, and Suisun Bays eastward to Chipps Island at the
confluence of the Sacramento and San Joaquin Rivers (Figure 24). Tributary streams to
Suisun Marsh including Suisun Creek, Green Valley Creek, and an unnamed tributary to
Cordelia Slough (commonly referred to as Red Top Creek), excluding the Sacramento-
San Joaquin River Basin, as well as two artificial propagation programs: the Don
Clausen Fish Hatchery, and Kingfisher Flat Hatchery/ Scott Creek (Monterey Bay
Salmon and Trout Project) steelhead hatchery programs. Table 18 identifies populations
within the CCC Steelhead salmon ESU, their abundances, and hatchery input.
DRAFT 150
-------
39 N-
NPDES permit sites
EPA 303(d) rivers
0 15 30
I
122W
60
Prepared by Dwayne Meadows
Kilometers 16 Jury 2008
Figure 24. CCC steelhead. The Legend for the Land Cover Class categories is
found in Figure 7.
DRAFT
151
-------
DRAFT
Table 18. CCC Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005).
Basin
Russian River
Lagunitas
San Oregon o
Waddell Creek
Scott Creek
San Vicente Creek
San Lorenzo River
Soquel Creek
Aptos Creek
Total
Historical
Abundance
65,000 (1970)
Unknown
1,000 (1973)
481
Unknown
150(1982)
20,000
500-800
f -1 /~\O O\
(1982)
200 (1982)
94,000
Most Recent
Spawner
Abundance
1,750-7,000
(1994)
400-500 (1990s)
Unknown
150(1994)
<100(1991)
50 (1994)
<150(1994)
<100(1991)
50-75 (1994)
2,400^8,125
Hatchery
Abundance
Contributions
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Life History
Only winter steelhead are found in this DPS and those to the south. Migration and spawn
timing are similar to adjacent steelhead populations. There is little other life history
information for steelhead in this DPS.
Status and Trends
The CCC steelhead DPS was listed as a threatened species on August 18, 1997(62 FR
43937). Its threatened status was reaffirmed on January 5, 2006 (71 FR 834). Busby et
al. (1996) reported one estimate of historical (pre-1960s) abundance. Shapovalov and
Taft (1954) described an average of about 500 adults in Waddell Creek (Santa Cruz
County) for the 1930s and early 1940s. Johnson (Johnson 1964) estimated a run size of
20,000 steelhead in the San Lorenzo River before 1965. The CDFG (1965) estimated an
average run size of 94,000 steelhead for the entire DPS, for the period 1959-1963. The
analysis by CDFG (1965) was compromised for many basins, as the data did not exist for
the full 5-year analytical period. The authors of CDFG (1965) state that "estimates given
DRAFT 152
-------
DRAFT
here which are based on little or no data should be used only in outlining the major and
critical factors of the resource."
Recent data for the Russian and San Lorenzo rivers (Reavis 1991; CDFG 1994; Shumann
1994) suggested that these basins had populations smaller than 15% of their size 30 years
earlier. These two basins were thought to have originally contained the two largest
steelhead populations in the CCC steelhead ESU.
A status review update in 1997 (Gustafson, Wainwright et al. 1997) concluded that slight
increases in abundance occurred in the three years following the status review. However,
the analyses on which these conclusions were based had various problems. They include
the inability to distinguish hatchery and wild fish, unjustified expansion factors, and
variance in sampling efficiency on the San Lorenzo River. Presence-absence data
indicated that most (82%) sampled streams (a subset of all historical steelhead streams)
had extant populations of juvenile O. mykiss (Adams 2000; Good, Waples et al. 2005).
The majority (69%) of BRT votes were for "likely to become endangered," and another
25% were for "in danger of extinction". Abundance and productivity were of relatively
high concern (as a contributing factor to risk of extinction), and spatial structure was also
of concern. Predation by pinnipeds at river mouths and during the ocean phase was noted
as a recent development posing significant risk. There were no time-series data for the
CCC steelhead DPS. A variety of evidence suggested the ESU's largest run (the Russian
River winter steelhead run) has been, and continues to be, reduced in size. Concern was
also expressed about populations in the southern part of the DPS's range—notably those
in Santa Cruz County and the South Bay area (Good, Waples et al. 2005).
Critical Habitat
Critical habitat was designated for the CCC steelhead DPS on September 2, 2005 (70 FR
52488), and includes areas within the following hydrologic units: Russian River,
Bodega, Marin Coastal, San Mateo, Bay Bridges, Santa Clara, San Pablo, Big Basin.
Refer to the final rule for a more detailed description of critical habitat, including a map
for each hydrologic unit.
California Central Valley Steelhead
Distribution
California CV steelhead occupy the Sacramento and San Joaquin Rivers and its
tributaries (Figure 25).
DRAFT 153
-------
Legend
to* Dams
* NPDES permit sites
EPA 303(d) rivers
Migratory corridor
37 N-
123 W
Prepared by Dwayne Meadows
Kilometers 16 July 2008
Figure 25. California CV steelhead distribution. The Legend for the Land Cover
DRAFT
154
-------
DRAFT
Class categories is found in Figure 7.
DRAFT 155
-------
DRAFT
Life History
California CV steelhead are considered winter steelhead by the CDFG. Although "three
distinct runs," including summer steelhead, may have occurred there as recently as
1947(CDFG 1995; McEwan and Jackson 1996). Steelhead within this DPS have the
longest freshwater migration of any population of winter steelhead. There is essentially a
single continuous run of steelhead in the upper Sacramento River. River entry ranges
from July through May, with peaks in September and February. Spawning begins in late
December and can extend into April (McEwan and Jackson 1996).
Status and Trends
California CV steelhead were listed as threatened on March 19, 1998. Their
classification was retained following a status review on January 5, 2006 (71 FR 834).
This DPS consists of steelhead populations in the Sacramento and San Joaquin River
(inclusive of and downstream of the Merced River) basins in California's CV. Steelhead
historically were well distributed throughout the Sacramento and San Joaquin Rivers
(Busby, Wainwright et al. 1996). Steelhead were found from the upper Sacramento and
Pit River systems (now inaccessible due to Shasta and Keswick Dams), south to the
Kings and possibly the Kern River systems (now inaccessible due to extensive alteration
from water diversion projects), and in both east- and west-side Sacramento River
tributaries (Yoshiyama, Gerstung et al. 1996). The present distribution has been greatly
reduced (McEwan and Jackson 1996). The California Advisory Committee on Salmon
and Steelhead (1988) reported a reduction of steelhead habitat from 6,000 miles
historically to 300 miles today. Historically, steelhead probably ascended Clear Creek
past the French Gulch area, but access to the upper basin was blocked by Whiskeytown
Dam in 1964 (Yoshiyama, Gerstung et al. 1996). Steelhead also occurred in the upper
drainages of the Feather, American, Yuba, and Stanislaus Rivers which are now
inaccessible (McEwan and Jackson 1996; Yoshiyama, Gerstung et al. 1996).
Historic CV steelhead run size is difficult to estimate given limited data, but may have
approached one to two million adults annually (McEwan 2001). By the early 1960s, the
steelhead run size had declined to about 40,000 adults (McEwan 2001). Over the past 30
years, the naturally spawned steelhead populations in the upper Sacramento River have
declined substantially. Hallock et al. (1961) estimated an average of 20,540 adult
steelhead in the Sacramento River, upstream of the Feather River, through the 1960s.
Steelhead counts at Red Bluff Diversion Dam declined from an average of 11,187 for the
period of 1967 to 1977, to an average of approximately 2,000 through the early 1990s,
with an estimated total annual run size for the entire Sacramento-San Joaquin system,
based on Red Bluff Diversion Dam counts, to be no more than 10,000 adults (McEwan
and Jackson 1996; McEwan 2001). Steelhead escapement surveys at Red Bluff
DRAFT 156
-------
DRAFT
Diversion Dam ended in 1993 due to changes in dam operations.
The only consistent data available on steelhead numbers in the San Joaquin River basin
come from CDFG mid-water trawling samples collected on the lower San Joaquin River
at Mossdale. These data indicate a decline in steelhead numbers in the early 1990s,
which have remained low through 2002 (CDFG 2003). In 2004, a total of 12 steelhead
smolts were collected at Mossdale (CDFG unpublished data).
Existing wild steelhead stocks in the CV are mostly confined to the upper Sacramento
River and its tributaries, including Antelope, Deer, and Mill Creeks and the Yuba River.
Populations may exist in Big Chico and Butte Creeks. A few wild steelhead are produced
in the American and Feather Rivers (McEwan and Jackson 1996).
Snorkel surveys from 1999 to 2002 indicate that steelhead are present in Clear Creek (J.
Newton, FWS, pers. comm. 2002, as reported in Good et al. (2006). Because of the large
resident O. mykiss population in Clear Creek, steelhead spawner abundance has not been
estimated.
Until recently, steelhead were thought to be extirpated from the San Joaquin River
system. Recent monitoring has detected small self-sustaining populations of steelhead in
the Stanislaus, Mokelumne, Calaveras, and other streams previously thought to be void of
steelhead (McEwan 2001). On the Stanislaus River, steelhead smolts have been captured
in rotary screw traps at Caswell State Park and Oakdale each year since 1995 (Demko
and Cramer 2000). It is possible that naturally spawning populations exist in many other
streams. However, these populations are undetected due to lack of monitoring programs
(IEPSPWT 1999).
The majority (66%) of BRT votes was for "in danger of extinction," and the remainder
was for "likely to become endangered". Abundance, productivity, and spatial structure
were of highest concern. Diversity considerations were of significant concern. The BRT
was concerned with what little new information was available and indicated that the
monotonic decline in total abundance and in the proportion of wild fish in the California
CV steelhead ESU was continuing.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005. The critical habitat
designation for this DPS identifies PCEs that include sites necessary to support one or
more life stages of steelhead. Specific sites include: (1) Freshwater spawning, (2)
freshwater rearing, (3) freshwater migration, (4) estuarine areas free of obstruction, (5)
nearshore marine areas free of obstructions, and (6) offshore marine areas with good
DRAFT 157
-------
DRAFT
water quality. The physical or biological features that characterize these sites include
water quality and quantity, natural cover, and adequate forage.
Lower Columbia River Steelhead
Distribution
LCR steelhead DPS includes 23 historical anadromous populations in four MPGs. This
DPS includes naturally-produced steelhead returning to Columbia River tributaries on the
Washington side between the Cowlitz and Wind rivers in Washington and on the Oregon
side between the Willamette and Hood rivers, inclusive (Figure 26). In the Willamette
River, the upstream boundary of this species is at Willamette Falls. This species includes
both winter and summer steelhead. Two hatchery populations are included in this
species, the Cowlitz Trout Hatchery winter-run stock and the Clackamas River stock.
However, neither hatchery population was listed as threatened.
Table 19 identifies populations within the LCR Steelhead salmon DPS, their abundances,
and hatchery input.
DRAFT 158
-------
123 W
I
45 N-
Legend
to* Dams
* NPDES permit sites
- EPA 303(d) rivers
I Migratory corridor
-45 N
n
123 W
0 15 30
60
Prepared by Dwayne Meadows
Kilometers 15 Jury 2008
Figure 26. Lower Columbia River Steelhead distribution. The Legend for the Land
Cover Class categories is found in Figure 7.
DRAFT
159
-------
DRAFT
Table 19. LCR Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005).
Population
Cispus River
Tilton River
Upper Cowlitz River
Lower Cowlitz River
Coweeman River
South Fork Toutle River
North Fork Toutle River
Kalama River-winter run
Kalama River- summer run
North Fork Lewis River- winter run
North Fork Lewis River-summer
run
East Fork Lewis River-winter run
East Fork Lewis River-summer run
Salmon Creek
Washougal River-winter run
Washougal River-summer run
Clackamas River
Sandy River
Lower Columbia gorge tributaries
Upper Columbia gorge tributaries
Hood River-winter run
Hood River-summer run
Historical
Abundance
Unknown
Unknown
Unknown
1,672
2,243
2,627
3,770
554
3,165
713
Unknown
3,131
422
Unknown
2,497
1,419
Unknown
Unknown
793
243
Unknown
Unknown
Most Recent
Spawner
Abundance
Unknown
2,787
Unknown
Unknown
466
504
196
726
474
Unknown
Unknown
Unknown
434
Unknown
323
264
560
977
Unknown
Unknown
756
931
Hatchery
Abundance
Contributions
Unknown
-73%
Unknown
Unknown
-50%
-2%
0%
0%
-32%
Unknown
Unknown
Unknown
-25%
Unknown
0%
-8%
41%
42%
Unknown
Unknown
-52%
-83%
DRAFT
160
-------
DRAFT
Wind River
Total
2,288
25,537 (min)
472
9,870 (min)
-5%
Life History
Summer steelhead return to freshwater from May to November, entering the Columbia
River in a sexually immature condition and requiring several months in freshwater before
spawning. Winter steelhead enter freshwater from November to April. They are close to
sexual maturation and spawn shortly after arrival in their natal stream. Where both races
spawn in the same stream, summer steelhead tend to spawn at higher elevations than the
winter forms. Juveniles rear in freshwater (stream-type life history).
Status and Trends
LCR steelhead were listed as threatened on March 19, 1998 (63 FR 13347), and
reaffirmed as threatened on January 5, 2006 (71 FR 834). The 1998 status review noted
that this ESU is characterized by populations at low abundance relative to historical
levels, significant population declines since the mid-1980s, and widespread occurrence of
hatchery fish in naturally-spawning steelhead populations. During this review NMFS
was unable to identify any natural populations that would be considered at low risk.
All populations declined from 1980 to 2000, with sharp declines beginning in 1995.
Historical counts in some of the larger tributaries (Cowlitz, Kalama, and Sandy Rivers)
suggest the population probably exceeded 20,000 fish. During the 1990s, fish abundance
dropped to 1,000 to 2,000 fish. Recent abundance estimates of natural-origin spawners
range from completely extirpated for some populations above impassable barriers to over
700 for the Kalama and Sandy winter-run populations. A number of the populations have
a substantial fraction of hatchery-origin spawners in spawning areas. These populations
are hypothesized to be sustained largely by hatchery production. Exceptions are the
Kalama, the Toutle, and East Fork Lewis winter-run populations. These populations
have relatively low recent mean abundance estimates with the largest being the Kalama
(geometric mean of 728 spawners).
According to Good et al. (2006), most populations are at relatively low abundance.
Those with adequate data for modeling are estimated to have a relatively high extinction
probability. Some populations, particularly summer run, have shown higher return in the
last two to three years. Many of the long-and short-term trends in abundance of
individual populations are negative, some severely so. The trend in natural spawners is
<1; indicating the population is not replacing itself and in decline. Spatial structure has
been substantially reduced by the loss of access to the upper portions of some basins due
to tributary hydro development. Finally, a number of the populations have a substantial
fraction of hatchery-origin spawners. Exceptions are the Kalama, North and South Fork
DRAFT 161
-------
DRAFT
Toutle, and East Fork Lewis winter-run populations, which have few hatchery fish
spawning in natural spawning areas.
Over 73% of the BRT votes for this species fell in the "likely to become endangered"
category. There were small minorities falling in the "danger of extinction" and "not
likely to become endangered" categories. The BRT found moderate risks in all VSP
categories, with mean risk matrix scores ranging from moderately low for spatial
structure to moderately high for abundance and productivity (population growth rate).
DRAFT 162
-------
DRAFT
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52488).
The critical habitat designation for this DPS identifies PCEs that include sites necessary
to support one or more steelhead life stages. Specific sites include: (1) Freshwater
spawning, (2) freshwater rearing, (3) freshwater migration, (4) estuarine areas free of
obstruction, (5) nearshore marine areas free of obstructions, and (6) offshore marine areas
with good water quality. The physical or biological features that characterize these sites
include water quality and quantity, natural cover, forage, adequate passage conditions,
and floodplain connectivity.
Of 47 subbasins reviewed in NMFS' assessment of critical habitat for the LCR steelhead,
34 subbasins were rated as having a high conservation value. Eleven subbasins were
rated as having a medium value and two were rated as having a low value to the
conservation of the DPS. Limiting factors identified for LCR steelhead include: (1)
Degraded floodplain and steam channel structure and function, (2) reduced access to
spawning/rearing habitat, (3) altered streamflow in tributaries, (4) excessive sediment and
elevated water temperatures in tributaries, and (5) hatchery impacts (NMFS 2005b). The
above conditions also introduce sediment, nutrients, biocides, metals, and other pollutants
into surface and ground water and degrade water quality in the freshwater, estuarine, and
coastal ecosystems throughout the Pacific Northwest.
Middle Columbia River Steelhead
Distribution
Middle Columbia River (MCR) steelhead DPS includes anadromous populations in
Oregon and Washington subbasins upstream of the Hood and Wind River systems to and
including the Yakima River (Figure 27). There are four MPGs with 17 populations in
this DPS. Steelhead from the SR Basin (described elsewhere) are excluded. This
species includes the only populations of inland winter steelhead in the U.S., in the
Klickitat River and Fifteenmile Creek (Busby, Wainwright et al. 1996).
Two hatchery populations are considered part of this species, the Deschutes River stock
and the Umatilla River stock. Listing for neither of these stocks was considered
warranted. MCR steelhead occupy the intermontane region which includes some of the
driest areas of the Pacific Northwest, generally receiving less than 15.7 inches of rainfall
annually. Vegetation is of the shrub-steppe province, reflecting the dry climate and harsh
temperature extremes. Because of this habitat, occupied by the species, factors
contributing to the decline include agricultural practices, especially grazing, and water
diversions and withdrawals. In addition, hydropower development has impacted the
DRAFT 163
-------
DRAFT
species by preventing these steelhead from migrating to habitat above dams, and by
killing some of them when they try to migrate through the Columbia River hydroelectric
system. Table 20 identifies populations within the MCR Steelhead salmon DPS, their
abundances, and hatchery input.
DRAFT 164
-------
124 W
123W
122 W
121 W
-A", n
47 N-
45 N-
<>,- <•
*
4-
120 W
0 25 50
119 W 118 W
1 nn
Prepared by Dwayne Meadov/s
I Kilometers 16 July 2008
Figure 27. MCR Steelhead distribution. The Legend for
over Class categories is found in Figure 7.
165
-------
DRAFT
Table 20. MCR Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005).
Population
Klickitat River
Yakima River
Fifteenmile Creek
Deschutes River
John Day upper main stream
John Day lower main stream
John Day upper north fork
John Day lower north fork
John Day middle fork
John Day south fork
Umatilla River
Touchet River
Total
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Most Recent
Spawner
Abundance
97-261 reds
1,058-4,061
2.87rpm
10,026-21,457
926-4,168
1.4 rpm
2.57rpm
.52 rpm
3.7 rpm
2. 52 rpm
1,480-5,157
273-527
Hatchery
Abundance
Contributions
Unknown
97%
100%
38%
96%
0%
0%
0%
0%
0%
60%
84%
Life History
Most MCR steelhead smolt at two years and spend one to two years in saltwater prior to
re-entering freshwater. Here they may remain up to a year prior to spawning (Howell,
Jones et al. 1985). Within this ESU, the Klickitat River is unusual as it produces both
summer and winter steelhead. The summer steelhead are dominated by age two ocean
steelhead. Most other rivers in this region produce about equal numbers of both age one
and two ocean steelhead.
DRAFT
166
-------
DRAFT
Status and Trends
MCR steelhead were listed as threatened in 1999 (64 FR 14517), and their status was
reaffirmed on January 5, 2006 (71 FR 834). The ICBTRT (2003) identified 15
populations in four MPGs (Cascades Eastern Slopes Tributaries, John Day River, the
Walla Walla and Umatilla Rivers, and the Yakima River) and one unaffiliated
independent population (Rock Creek) in this species. There are two extinct populations
in the Cascades Eastern Slope MPG: the White Salmon River and Deschutes Crooked
River above the Pelton/Round Butte Dam complex.
Seven hatchery steelhead programs are considered part of the MCR steelhead species.
These programs propagate steelhead in three of 16 populations and improve kelt survival
in one population. No artificial programs produce the winter-run life history in the
Klickitat River and Fifteenmile Creek populations. All of the MCR steelhead hatchery
programs are designed to produce fish for harvest. However, two hatchery programs are
also implemented to augment the naturally spawning populations in the basins where the
fish are released. The NMFS assessment of the effects of artificial propagation on MCR
steelhead extinction risk concluded that these hatchery programs collectively do not
substantially reduce the extinction risk. Artificial propagation increases total species
abundance, principally in the Umatilla and Deschutes Rivers. The kelt reconditioning
efforts in the Yakima River do not augment natural abundance and benefit the survival of
the natural populations. The Touchet River Hatchery program has only recently been
established, and its contribution to species viability is uncertain. The hatchery programs
affect a small proportion of the species. Collectively, artificial propagation programs
provide a slight beneficial effect to species abundance and have neutral or uncertain
effects on species productivity, spatial structure, and diversity.
The precise pre-1960 abundance of this species is unknown. However, historic run
estimates for the Yakima River imply that annual species abundance may have exceeded
300,000 returning adults (Busby, Wainwright et al. 1996). MCR steelhead run estimates
between 1982 and 2004 were calculated by subtracting adult counts for Lower Granite
and Priest Rapids Dams from those at Bonneville Dam. The five year average (geometric
mean) return of natural MCR steelhead for 1997 to 2001 was up from previous years'
basin estimates. Returns to the Yakima River, the Deschutes River, and sections of the
John Day River system were substantially higher compared to 1992 to 1997 (Good,
Waples et al. 2005). Yakima River returns are still substantially below interim target
levels of 8,900 (the current five year average is 1,747 fish) and estimated historical return
levels, with the majority of spawning occurring in one tributary, Satus Creek (Berg
2001). The recent five year geometric mean return of the natural-origin component of the
Deschutes River run exceeded interim target levels (Good, Waples et al. 2005). Recent
five year geometric mean annual returns to the John Day River basin are generally below
the corresponding mean returns reported in previous status reviews. However, each
DRAFT 167
-------
DRAFT
major production area in the John Day system has shown upward trends since the 1999
return year (Good, Waples et al. 2005). The Touchet and Umatilla are below their
interim abundance targets of 900 and 2,300, respectively. The five year average for these
basins is 298 and 1,492 fish, respectively (Good, Waples et al. 2005).
As per the FCRPS (2008), during the most recent 10-year period (for which trends in
abundance could be estimated), trends were positive for approximately half of the
populations and negative for the remainder. On average, when only natural production is
considered, most of the MCR steelhead populations have replaced themselves. The
ICBTRT characterizes the diversity risk to all but one MCR steelhead population as
"low" to "moderate". The Upper Yakima is rated as having "high" diversity risk because
of introgression with resident O. mykiss and the loss of presmolt migration pathways.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52488).
The critical habitat designation for this DPS identifies PCEs that include sites necessary
to support one or more life stages of steelhead. MCR steelhead have PCEs of: (1)
freshwater spawning, (2) freshwater rearing, (3) freshwater migration, (4) estuarine areas
free of obstruction, (5) nearshore marine areas free of obstructions, and (6) offshore
marine areas with good water quality. The physical or biological features that
characterize these sites include water quality and quantity, natural cover, forage, and
adequate passage conditions. Although pristine habitat conditions are still present in
some wilderness, roadless, and undeveloped areas, habitat complexity has been greatly
reduced in many areas of designated critical habitat for MCR steelhead. Limiting factors
identified for MCR steelhead include: (1) Hydropower system mortality; (2) reduced
stream flow; (3) impaired passage; (4) excessive sediment; (5) degraded water quality;
and (6) altered channel morphology and floodplain.
Northern California Steelhead
Distribution
Northern California steelhead includes steelhead in CC river basins from Redwood Creek
south to the Gualala River, inclusive (Figure 28). Table 21 identifies populations within
the Northern California Steelhead salmon ESU, their abundances, and hatchery input.
DRAFT 168
-------
124 W
I
123 W
I
41 N-
40 N-
39'N-
Legend
tr/ Dams
* NPDES permit sites
EPA 303(d) rivers
-41 N
-40 N
I
124 W
0 15 30
60
Prepared by Dwayne Meadows
Kilometers 16 July 2008
Figure 28. Northern California Steelhead distribution. The Legend for the Land
Cover Class categories is found in Figure 7.
DRAFT
169
-------
DRAFT
Table 21. Northern California Steelhead salmon populations, abundances, and
hatchery contributions (Good, Waples et al. 2005).
Most Recent
River
Redwood Creek
Mad River
Eel River
Mattel e River
Ten Mile River
Noyo River
Big River
Navarro River
Garcia River
Gualala River
Other Humboldt County streams
Other Mendocino County streams
Total
Historical
Abundance
10,000
6,000
82,000
12,000
9,000
8,000
12,000
16,000
4,000
16,000
3,000
20,000
198,000
Spawner
Abundance
Unknown
162-384
3,127-21,903
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Hatchery
Abundance
Contributions
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Unknown
Life History
Steelhead within this DPS include winter and summer steelhead. Half-pounder juveniles
occur in the Mad and Eel Rivers. Half-pounders are immature steelhead that returns to
freshwater after only two to four months in the ocean, and generally overwinter in
freshwater. These juveniles then outmigrate in the following spring.
Status and Trends
NC steelhead were listed as threatened on June 7, 2000 (65 FR 36074). They retained
that classification following a status review on January 5, 2006 (71 FR 834). Long-term
data sets are limited for this NC steelhead. Before 1960, estimates of abundance specific
to this DPS were available from dam counts in the upper Eel River (Cape Horn Dam-
annual avg. no. adults was 4,400 in the 1930s), the South Fork Eel River (Benbow Dam-
annual avg. no. adults was 19,000 in the 1940s), and the Mad River (Sweasey Dam-
annual avg. no. adults was 3,800 in the 1940s). Estimates of steelhead spawning
populations for many rivers in this DPS totaled 198,000 by the mid-1960s.
DRAFT
170
-------
DRAFT
During the first status review on this population, adult escapement trends could be
computed on seven populations. Five of the seven populations exhibited declines while
two exhibited increases with a range of almost 6% annual decline to a 3.5% increase. At
the time little information was available on the actual contribution of hatchery fish to
natural spawning, and on present total run sizes for the DPS (Busby, Wainwright et al.
1996).
More recent time series data are from snorkel counts conducted on summer-run steelhead
in the Middle Fork Eel River. An estimate of lambda over the interval 1966 to 2002 was
made and a random-walk with drift model fitted using Bayesian assumptions. Good et al.
(2006) estimated lambda at 0.98 with a 95% confidence interval of 0.93 and 1.04. The
result is an overall downward trend in both the long- and short- term. Juvenile data were
also recently examined. Both upward and downward trends were apparent (Good,
Waples et al. 2005). The majority (74%) of BRT votes were for "likely to become
endangered," with the remaining votes split equally between "in danger of extinction"
and "not warranted".
Critical Habitat
Critical habitat was designated for NC steelhead on September 2, 2005 (70 FR 52488).
The critical habitat designation for this DPS identifies PCEs that include sites necessary
to support one or more life stages of steelhead. Specific sites include: (1) freshwater
spawning, (2) freshwater rearing, (3) freshwater migration, (4) estuarine areas free of
obstruction, (5) nearshore marine areas free of obstructions, and (6) offshore marine areas
with good water quality. The physical or biological features that characterize these sites
include water quality and quantity, natural cover, and adequate forage.
Puget Sound Steelhead
Distribution
Puget Sound steelhead occupy river basins of the Strait of Juan de Fuca, Puget Sound,
and Hood Canal, Washington. Included are river basins as far west as the Elwha River
and as far north as the Nooksack River (Figure 29). Puget Sound's fjord-like structure
may affect steelhead migration patterns. For example, some populations of coho and
Chinook salmon, at least historically, remained within Puget Sound and did not migrate
DRAFT 171
-------
NPDES permit sites
EPA 303(d) rivers
121 W
Prepared by Dwayne Meadows
Kilometers 16 July 2008
Figure 29. Puget Sound steelhead distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
172
-------
DRAFT
to the Pacific Ocean. Even when Puget Sound steelhead migrate to the high seas, they
may spend considerable time as juveniles or adults in the protected marine environment
of Puget Sound. This is a feature not readily accessible to steelhead from other areas of
the Pacific Northwest. The species is primarily composed of winter steelhead but
includes several stocks of summer steelhead, usually in subbasins of large river systems
and above seasonal hydrologic barriers.
Life History
Life history attributes of Puget Sound steelhead (migration and spawn timing, smolt age,
ocean age, and total age at first spawning) appear similar to those of other west coast
steelhead. Ocean age for Puget Sound summer steelhead varies among populations.
Status and Trends
Puget Sound steelhead were listed as a threatened species on May 11, 2007 (72 FR
26722). Run size for this DPS, was calculated in the early 1980s at about 100,000
winter-run fish and 20,000 summer-run fish. It is unclear what portion were hatchery
fish. However, a combined estimate with coastal steelhead suggested that roughly 70%
of steelhead in ocean runs were of hatchery origin. The percentage in escapement to
spawning grounds would be substantially lower due to differential harvest and hatchery
rack returns. By the 1990s, total run size for four major stocks exceeded 45,000, roughly
half of which was natural escapement.
Nehlsen et al. (1997) identified nine Puget Sound steelhead stocks at some degree of risk
or concern. The WDFW et al. (1993) estimated that 31 of 53 stocks were of native origin
and predominantly natural production. The WDFW assessment of the status of these 31
stocks was 11 healthy, three depressed, one critical, and 16 of unknown status. Their
assessment of the status of the remaining (not native/natural) stocks was three healthy, 11
depressed, and eight of unknown status.
Of the 21 populations in the Puget Sound ESU reviewed by Busby et al. (1996), 17 had
declining and four had increasing trends, with a range from 18% annual decline (Lake
Washington winter-run steelhead) to 7% annual increase (Skykomish River winter-run
steelhead). Eleven of these trends (nine negative, two positive) were significantly
different from zero. These trends were for the late-run naturally produced component of
winter-run steelhead populations. No adult trend data were available for summer-run
steelhead. Most of these trends were based on relatively short data series. The Skagit
and Snohomish River winter-run populations have been approximately three to five times
larger than the other populations in the DPS, with average annual spawning of
approximately 5,000 and 3,000 total adult spawners, respectively. These two basins
exhibited modest overall upward trends at the time of the Busby et al. (1996) report.
DRAFT 173
-------
DRAFT
Busby et al. (1996) estimated five-year average natural escapements for streams with
adequate data range from less than 100 to 7,200, with corresponding total run sizes of
550 to 19,800.
Critical Habitat
Critical habitat is not currently designated for Puget Sound steelhead. However, factors
for essential habitat are under evaluation to designate future critical habitat.
Snake River Steelhead
Distribution
SR Basin steelhead is an inland species that occupies the SR basin of Idaho, northeast
Oregon, and southeast Washington. The SR Basin steelhead species includes all
naturally spawned populations of steelhead (and their progeny) in streams in the SR
Basin of Idaho, northeast Oregon, and southeast Washington SR Basin steelhead do not
include resident forms of O. mykiss (rainbow trout) co-occurring with these steelhead.
The historic spawning range of this species included the Salmon, Pahsimeroi, Lemhi,
Selway, Clearwater, Wallowa, Grande Ronde, Imnaha, and Tucannon Rivers.
Managers classify up-river summer steelhead runs into two groups based on ocean age
and adult size upon return to the Columbia River. A-run steelhead are predominately
age-one-ocean fish. B-run steelhead are larger, predominated by age-two-ocean fish. A-
run populations are found in the tributaries to the lower Clearwater River, the upper
Salmon River and its tributaries, the lower Salmon River and its tributaries, the Grand
Ronde River, Imnaha River, and possibly the SR's mainstem tributaries below Hells
Canyon Dam. B-run steelhead occupy four major subbasins. They include two on the
Clearwater River (Lochsa and Selway) and two on the Salmon River (Middle Fork and
South Fork Salmon); areas not occupied by A-run steelhead. Some natural B-run
steelhead are also produced in parts of the mainstem Clearwater and its major tributaries.
There are alternative escapement objectives of 10,000 (Columbia River Fisheries
Management Plan) and 31,400 (Idaho) for B-run steelhead. B-run steelhead represent at
least one-third and as much as three-fifths of the production capacity of the DPS.
DRAFT 174
-------
124 W
122W
120 W
I
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
Migratory corridor
12 M
Prepared by Owayne Meadows
Kilometers 16 July 2008
Figure 30. SR Basin Steelhead distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
175
-------
DRAFT
Table 22 identifies populations within the SR Basin Steelhead salmon ESU, their
abundances, and hatchery input.
Table 22. SR Basin Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005). Note: rpm denotes redds per mile.
Most Recent
River
Tucannon River
Lower Granite run
Snake A run
Snake B run
Asotin Creek
Upper Grande Ronde River
Joseph Creek
Imnaha River
Camp Creek
Total
Historical
Abundance
3,000
Unknown
Unknown
Unknown
Unknown
15,000
Unknown
4,000
Unknown
Spawner
Abundance
257-628
70,721-259,145
50,974-25,950
9,736-33,195
0-543 redds
1.54 rpm
1,077-2,385
3.7 rpm
55-307
9
Hatchery
Abundance
Contributions
26%
86%
85%
89%
Unknown
23%
0%
20%
0%
Life History
SR Basin Steelhead occupy habitat that is considerably warmer and drier (on an annual
basis) than other Steelhead DPSs. SR Basin Steelhead are generally classified as summer
run, based on their adult run timing pattern. Sexually immature adult SR Basin summer
steelheads enter the Columbia River from late June to October. SR Basin Steelhead
returns consist of A-run fish that spend one year in the ocean, and larger B-run fish that
spend two years at sea. Adults typically migrate upriver until they reach tributaries from
1,000 to 2,000 m above sea level where they spawn between March and May of the
following year. Unlike other anadromous members of the Oncorhynchus genus, some
adult Steelhead survive spawning, return to the sea, and later return to spawn a second
time. After hatching, juvenile SR Basin Steelhead typically spend two to three years in
fresh water before they smolt and migrate to the ocean.
DRAFT
176
-------
DRAFT
Status and Trends
SR Basin steelhead were listed as threatened in 1997 (62 FR 43937). Their classification
status was reaffirmed following a status review on January 5, 2006 (71 FR 834). The
ICBTRT (2003) identified 23 populations in the following six MPGs: Clearwater River,
Grande Ronde River, Hells Canyon, Imnaha River, Lower SR, and Salmon River. SR
Basin steelhead remain spatially well distributed in each of the six major geographic
areas in the SR basin (Good, Waples et al. 2005). Environmental conditions are
generally drier and warmer in these areas than in areas occupied by other steelhead
species in the Pacific Northwest. SR Basin steelhead were blocked from portions of the
upper SR beginning in the late 1800s and culminating with the construction of Hells
Canyon Dam in the 1960s. The SR Basin steelhead "B run" population levels remain
particularly depressed. The ICBTRT has not completed a viability assessment for SR
Basin steelhead.
Limited information on adult spawning escapement for specific tributary production areas
for SR Basin steelhead made a quantitative assessment of viability difficult. Annual
return estimates are limited to counts of the aggregate return over Lower Granite Dam,
and spawner estimates for the Tucannon, Grande Ronde, and Imnaha Rivers. The 2001
return over Lower Granite Dam was substantially higher relative to the low levels seen in
the 1990s; the recent five-year mean abundance (14,768 natural returns) was
approximately 28% of the interim recovery target level. The 10-year average for natural-
origin steelhead passing Lower Granite Dam between 1996 and 2005 is 28,303 adults.
Parr densities in natural production areas, which are another indicator of population
status, have been substantially below estimated capacity for several decades. The SR
supports approximately 63% of the total natural-origin production of steelhead in the
Columbia River Basin. The current condition of SR Basin steelhead (Good, Waples et al.
2005) is summarized below:
There is uncertainty for wild populations given limited data for adult spawners in
individual populations. Dam counts are currently 28% of interim recovery target for the
SR Basin (52,000 natural spawners). Only the Joseph Creek population exceeds the
interim recovery target. Regarding population growth rate, there are mixed long- and
short-term trends in abundance and productivity. Regarding spatial structure, the SR
Basin steelhead are well distributed with populations remaining in six major areas.
However, the core area for B-run steelhead, once located in the North Fork of the
Clearwater River, is now inaccessible to steelhead. Finally, genetic diversity is affected
by the displacement of natural fish by hatchery fish (declining proportion of natural-
origin spawners). Homogenization of hatchery stocks occurs within basins, and some
stocks exhibit high stray rates.
DRAFT 177
-------
DRAFT
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52488).
The critical habitat designation for this ESU identifies PCEs that include sites necessary
to support one or more steelhead life stages. Specific sites include: (1) Freshwater
spawning, (2) freshwater rearing, (3) freshwater migration, (4) estuarine areas free of
obstruction, (5) nearshore marine areas free of obstructions, and (6) offshore marine areas
with good water quality.
Of the 291 fifth order streams reviewed in this DPS, 220 were rated as high, 44 were
rated as medium, and 27 were rated as low conservation value. The physical or
biological features that characterize these sites include water quality and quantity, natural
cover, and adequate forage. Limiting factors identified for SR Basin steelhead include:
(1) Hydrosystem mortality, (2) reduced stream flow, (3) altered channel morphology and
floodplain, (4) excessive sediment, (5) degraded water quality, (6) harvest impacts, and
(7) hatchery impacts (Myers, Busack et al. 2006).
South-Central California Coast Steelhead
Distribution
The South-Central California Coast (S-CCC) steelhead DPS includes all naturally
spawned populations of steelhead (and their progeny) in streams from the Pajaro River
(inclusive) to, but not including the Santa Maria River, California (Error! Reference
source not found.)
Life History
Only winter steelhead are found in this DPS. Migration and spawn timing are similar to
adjacent steelhead populations. There is little other life history information for steelhead
in this DPS.
DRAFT 178
-------
122'W
120 W
37 N
35 N-
Legend
Dams
«* NPDES permit sites
EPA 303(d) rivers
120 W
Prepared by Dwayne Meadows
Kilometers 16 Jury 2008
Figure 31. S-CCC steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7
DRAFT
179
-------
DRAFT
Status and Trends
S-CCC steelhead were listed as threatened in 1997. Their classification was retained
following a status review on January 5, 2006 (71 FR 834). Historical data on the S-C CC
steelhead DPS are limited. In the mid-1960s, the CDFG estimated the adult population at
about 18,000. We know of no recent estimates of the total DPS. However, five river
systems, the Pajaro, Salinas, Carmel, Little Sur, and Big Sur, indicate that runs are
currently less than 500 adults. Past estimates for these basins were almost 5,000 fish.
Carmel River time series data indicate that the population declined by about 22% per year
between 1963 and 1993 (Good, Waples et al. 2005). From 1991 the population increased
from one adult, to 775 adults at San Clemente Dam. Good et al. (2006) thought that this
recent increase seemed too great to attribute simply to improved reproduction and
survival of the local steelhead population. Other possibilities were considered including
that the substantial immigration or transplantation occurred, or that resident trout
production increased as a result of improved environmental conditions within the basin.
Nevertheless, the majority (68%) of BRT votes were for "likely to become endangered,"
and another 25% were for "in danger of extinction".
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52488).
The critical habitat designation for this DPS identifies PCEs that include sites necessary
to support one or more steelhead life stages. Specific sites include: (1) freshwater
spawning, (2) freshwater rearing, (3) freshwater migration, (4) estuarine areas free of
obstruction, (5) nearshore marine areas free of obstructions, and (6) offshore marine areas
with good water quality. The physical or biological features that characterize these sites
include water quality and quantity, natural cover, and adequate forage.
Southern California Steelhead
Distribution
Southern California (SC) steelhead occupy rivers from the Santa Maria River to the U.S.
-Mexico border (Figure 32).
DRAFT 180
-------
new
I
N
A
Legend
w Dams
c* NPDES permit sites
EPA 303(d) rivers
120 W
119-W
0 15 30
60
I Kilometers -|g ju|y 2008
Prepared by Dwayne Meadows
Figure 32. Southern California steelhead distribution. The Legend for the Land Cover Class categories is found in Figure 7.
DRAFT
181
-------
DRAFT
Table 23 identifies populations within the Southern California Steelhead salmon ESU,
their abundances, and hatchery input.
Table 23. Southern California Steelhead salmon populations, abundances, and
hatchery contributions (Good, Waples et al. 2005).
River
Santa Ynez River
Ventura River
MatilijaRiver
Creek River
Santa Clara River
Total
Historical
Abundance
12,995-30,000
4,000-6,000
2,000-2,500
Unknown
7,000-9,000
32,000-46,000
Most Recent
Spawner
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
<500
Hatchery
Abundance
Contributions
Unknown
Unknown
Unknown
Unknown
Unknown
Life History
Migration and life history patterns of SC Steelhead are dependent on rainfall and
streamflow (Moore 1980). Steelhead within this DPS can withstand higher temperatures
than populations to the north. The relatively warm and productive waters of the Ventura
River have resulted in more rapid growth of juvenile Steelhead than occurs in more
northerly populations (Moore 1980). There is little life history information for Steelhead
in this DPS.
Status and Trends
SC Steelhead were listed as endangered in 1997 (62 FR 43937). Their classification was
retained following a status review on January 5, 2006 (71 FR 834). In many watersheds
throughout Southern California, dams isolate Steelhead from historical spawning and
rearing habitats. Dams also alter the hydrology of the basin (e.g., Twitchell Reservoir
within the Santa Maria River watershed, Bradbury Dam within the Santa Ynez River
watershed, Matilija and Casitas dams within the Ventura River watershed, Rindge Dam
within the Malibu Creek watershed). Based on combined estimates for the Santa Ynez,
Ventura, and Santa Clara rivers, and Malibu Creek, an estimated 32,000 to 46,000 adult
Steelhead occupied this DPS. In contrast, less than 500 adults are estimated to occupy the
same four waterways presently. The last estimated run size for Steelhead in the Ventura
River, which has its headwaters in Los Padres National Forest, is 200 adults (Busby,
Wainwright et al. 1996). The majority (81%) of the BRT votes were for "in danger of
DRAFT
182
-------
DRAFT
extinction," with the remaining 19% of votes for "likely to become endangered. This was
based on extremely strong concern for abundance, productivity, and spatial concern (as
per the risk matrix); diversity was also of concern. The BRT also expressed concern
about the lack of data on the SC steelhead DPS, including uncertainly on the
metapopulation dynamics in the southern part of the DPS's range and the fish's nearly
complete extirpation from the southern part of the range.
Critical Habitat
Critical habitat was designated for this species on September 2, 2005. The designation
identifies PCEs that include sites necessary to support one or more steelhead life stages.
These sites contain the physical or biological features essential for the species
conservation. Specific sites include freshwater spawning sites, freshwater rearing sites,
freshwater migration corridors, and estuarine areas. The physical or biological features
that characterize these sites include water quantity, depth, and velocity, shelter, cover,
living space and passage conditions.
Upper Columbia River Steelhead
Distribution
UCR steelhead occupy the Columbia River Basin upstream from the Yakima River,
Washington, to the border between the U.S. and Canada (Figure 33). This area includes
the Wenatchee, Entiat, and Okanogan Rivers. All UCR steelhead are summer steelhead.
Steelhead primarily use streams of this region that drain the northern Cascade Mountains
of Washington State. This species includes hatchery populations of summer steelhead
from the Wells Hatchery because it probably retains the genetic resources of steelhead
populations that once occurred above the Grand Coulee Dam. This species does not
include the Skamania Hatchery stock because of its non-native genetic heritage.
Abundance estimates of returning naturally produced UCR steelhead have been based on
extrapolations from mainstem dam counts and associated sampling information (e.g.,
hatchery/wild fraction, age composition). The natural component of the annual steelhead
run over Priest Rapids Dam increased from an average of 1,040 (1992-1996),
representing about 10% of the total adult count, to 2,200 (1997-2001), representing about
17% of the adult count during this period of time (ICBTRT 2003). Table 24 identifies
populations within the UCR Steelhead salmon DPS, their abundances, and hatchery
input.
DRAFT 183
-------
124 W
123 W
I
11S W
us w
49 H-
Legend
tor Dams mL
NPDES permit sites
EPA 303(d) rivers
Migratory corridor
I Kilometers 18 July 2008
Figure 33. UCR Steelhead distribution. The Legend for the Land Cover Class categories is found in Figure 7.
Prepared by Dvvayne Meadows
DRAFT
184
-------
DRAFT
DRAFT 185
-------
DRAFT
Table 24. UCR Steelhead salmon populations, abundances, and hatchery
Most Recent
Historical Hatchery
Population Spawner Abundance
Abundance „ , ., ,.
Abundance Contributions
Wenatchee/Entiat rivers
Methow/Okanogan rivers
Total
Unknown
Unknown
Unknown
1,899-8,036
1,879-12,801
3,778-20,837
71%
91%
Life History
The life history patterns of UCR Steelhead are complex. Adults return to the Columbia
River in the late summer and early fall. Most migrate relatively quickly up the mainstem
to their natal tributaries. A portion of the returning run overwinters in the mainstem
reservoirs, passing over the upper-mid-Columbia dams in April and May of the following
year. Spawning occurs in the late spring of the calendar year following entry into the
river. Juvenile Steelhead spend one to seven years rearing in freshwater before migrating
to sea. Smolt outmigrations are predominantly age-two and age-three juveniles. Most
adult Steelhead return after one or two years at sea, starting the cycle again.
Status and Trends
UCR Steelhead were originally listed as endangered in 1997 (62 FR 43937). Following a
status review, they were reclassified to threatened on January 5, 2006 and then reinstated
to endangered status per U.S. District Court decision in June 2007 (62 FR 43937). This
DPS includes all naturally spawned anadromous Steelhead populations below natural and
manmade impassable barriers in streams in the Columbia River Basin upstream from the
Yakima River, Washington, to the U.S.-Canada border, as well six artificial propagation
programs: the Wenatchee River, Wells Hatchery (in the Methow and Okanogan Rivers),
Winthrop NFH, Omak Creek, and the Ringold Steelhead hatchery programs. The
ICBTRT has identified five populations within this DPS: the Wenatchee River, Entiat
River, Methow River, Okanogan Basin, and Crab Creek.
Returns of both hatchery and naturally produced Steelhead to the UCR have increased in
recent years. The average 1997 to 2001 return counted through the Priest Rapids fish
ladder was approximately 12,900 fish. The average for the previous five years (1992 to
1996) was 7,800 fish. Abundance estimates of returning naturally produced UCR
Steelhead have been based on extrapolations from mainstem dam counts and associated
sampling information (e.g., hatchery/wild fraction, age composition). The natural
component of the annual Steelhead run over Priest Rapids Dam increased from an
DRAFT 186
-------
DRAFT
average of 1,040 (1992-1996), representing about 10% of the total adult count, to 2,200
(1997-2001), representing about 17% of the adult count during this period of time
(ICBTRT 2003).
In terms of natural production, recent population abundances for both the Wenatchee and
Entiat aggregate population and the Methow population remain well below the minimum
abundance thresholds developed for these populations (ICBTRT 2005). A five-year
geometric mean (1997 to 2001) of approximately 900 naturally produced steelhead
returned to the Wenatchee and Entiat rivers (combined). Although this is well below the
minimum abundance thresholds, it represents an improvement over the past (an
increasing trend of 3.4% per year). However, the average percentage of natural fish for
the recent five-year period dropped from 35% to 29%, compared to the previous status
review. For the Methow population, the five-year geometric mean of natural returns over
Wells Dam was 358. Although this is well below the minimum abundance thresholds, it
is an improvement over the recent past (an increasing trend of 5.9% per year). In
addition, the 2001 return (1,380 naturally produced spawners) was the highest single
annual return in the 25-year data series. However, the average percentage of wild origin
spawners dropped from 19% for the period prior to the 1998 status review to 9% for the
1997 to 2001 returns.
Regarding the population growth rate of natural production, on average, over the last 20
full brood year returns (1980/81 through 1999/2000 brood years), including adult returns
through 2004-2005, UCR steelhead populations have not replaced themselves. The
ICBTRT has characterized the spatial structure risk to UCR steelhead populations as
"low" for the Wenatchee and Methow, "moderate" for the Entiat, and "high" for the
Okanogan. Overall adult returns are dominated by hatchery fish, and detailed
information is lacking on the productivity of the natural population. All UCR steelhead
populations have reduced genetic diversity from homogenization of populations that
occurred during the Grand Coulee Fish Maintenance project from 1939-1943, from 1960,
and 1981 (Chapman, Hillman et al. 1994).
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52488).
The critical habitat designation for this DPS identifies PCEs that include sites necessary
to support one or more steelhead life stages. They include all Columbia River estuarine
areas and river reaches upstream to Chief Joseph Dam and several tributary subbasins.
Specific sites include freshwater spawning and rearing sites, freshwater migration
corridors, estuarine areas free of obstruction, and offshore marine areas. The physical or
biological features that characterize these sites include water quality and quantity, natural
cover, forage, and adequate passage conditions.
DRAFT 187
-------
DRAFT
The UCR steelhead DPS has 42 watersheds within its range. Three watersheds received
a low rating, eight received a medium rating, and 31 rated a high conservation value to
the DPS. In addition, the Columbia River rearing/migration corridor downstream of the
spawning range was rated as a high conservation value. Limiting factors identified for
the UCR steelhead include: (1) Mainstem Columbia River hydropower system mortality,
(2) reduced tributary streamflow, (3) tributary riparian degradation and loss of in-river
wood, (4) altered tributary floodplain and channel morphology, and (5) excessive fine
sediment and degraded tributary water quality. The above activities and features also
introduce sediment, nutrients, biocides, metals, and other pollutants into surface and
ground water and degrade water quality in the freshwater, estuarine, and coastal
ecosystems throughout the Pacific Northwest.
Upper Willamette River Steelhead
Distribution
UWR steelhead occupy the Willamette River and its tributaries upstream of Willamette
Falls (Figure 34). This is a late-migrating winter group that enters freshwater in March
and April (Howell, Jones et al. 1985). Only the late run was included in the listing of this
species, which is the largest remaining population in the Santiam River system. Table 25
identifies populations within the UWR Steelhead salmon ESU, their abundances, and
hatchery input.
DRAFT 188
-------
Legend
Dams
NPDES permit sites
EPA 303(d) rivers
Migratory corridor
-46 N
-45 N
123 W
0 10 20
40
Prepared by Dwayne Meadows
Kilometers 18 July 2008
Figure 34. UWR Steelhead distribution. The Legend for the Land Cover Class
categories is found in Figure 7.
DRAFT
189
-------
DRAFT
Table 25. UWR Steelhead salmon populations, abundances, and hatchery
contributions (Good, Waples et al. 2005). Note: rpm denotes redds per mile.
Population
Mollala Rivers
North Santiam River
South Santiam River
Calapooia River
Total
Historical
Abundance
Unknown
Unknown
Unknown
Unknown
Unknown
Most Recent
Spawner
Abundance
0.972 rpm
0.963 rpm
0.917 rpm
1.053 rpm
5,819
Hatchery
Abundance
Contributions
Unknown
Unknown
Unknown
Unknown
Life History
Winter steelhead enter the Willamette River beginning in January and February. They do
not ascend to their spawning areas until late March or (Dimick and Merryfield
1945). Spawning occurs April to June 1st and redd counts are conducted in May.
The smolt migration past Willamette Falls also begins in early April and extends through
early June (Howell, Jones et al. 1985) Migration peaks in early- to mid-May. Steelhead
smolts generally migrate away from the shoreline and enter the Columbia via Multnomah
Channel rather than the mouth of the Willamette. Most spend two years in the ocean
before re-entering fresh water to span (Busby, Wainwright et al. 1996). Steelhead in the
UWR DPS generally spawn once or twice. A few fish may spawn three times based on
patterns found in the LCR steelhead DPS. Repeat spawners are predominantly female
and generally account for less than 10% of the total run size (Busby, Wainwright et al.
1996).
Status and Trends
UWR steelhead were listed as threatened in 1999 (64 FR 14517). Their classification
was retained following a status review on January 5, 2006 (71 FR 834). A major threat to
Willamette River steelhead results from artificial production practices. Fishways built at
Willamette Falls in 1885 have allowed Skamania-stock summer steelhead and early-
migrating winter steelhead of Big Creek stock to enter the range of UWR steelhead. The
population of summer steelhead is almost entirely maintained by hatchery salmon,
although natural-origin, Big Creek-stock winter steelhead occur in the basin (Howell,
Jones et al. 1985). In recent years, releases of winter steelhead are primarily of native
DRAFT
190
-------
DRAFT
stock from the Santiam River system.
Steelhead in this DPS are depressed from historical levels, but to a much lesser extent
than are spring Chinook in the Willamette basin (McElhaney, Chilcote et al. 2007). All
of the historical populations remain extant and moderate numbers of wild steelhead are
produced each year. The population growth rate data indicate long-term trends are <1;
short-term trends are 1 or higher (McElhaney, Chilcote et al. 2007). Spatial structure for
the North and South Santiam populations has been substantially reduced by the loss of
access to the upper North Santiam basin and the Quartzville Creek watershed in the South
Santiam subbasin due to construction of the dams owned and operated by the U.S. Army
Corps of Engineers without passage facilities (McElhaney, Chilcote et al. 2007).
Additionally, the spatial structure in the Molalla subbasin has been reduced significantly
by habitat degradation and in the Calapooia by habitat degradation and passage barriers.
Finally, the diversity of some populations have been eroded by small population size, the
loss of access to historical habitat, legacy effects of past winter-run hatchery releases, and
the ongoing release of summer steelhead (McElhaney, Chilcote et al. 2007).
Critical Habitat
Critical habitat was designated for this species on September 2, 2005 (70 FR 52488). It
includes all Columbia River estuarine areas and river reaches proceeding upstream to the
confluence with the Willamette River as well as specific steam reaches in the following
subbasins: Upper Willamette, North Santiam, South Santiam, Middle Willamette,
Molalla/Pudding, Yamhill, Tualatin, and Lower Willamette (NMFS 2005b). The critical
habitat designation for this DPS identifies PCEs that include sites necessary to support
one or more steelhead life stages. Specific sites include: (1) Freshwater spawning, (2)
freshwater rearing, (3) freshwater migration, (4) estuarine areas free of obstruction, (5)
nearshore marine areas free of obstructions, and (6) offshore marine areas with good
water quality. The physical or biological features that characterize these sites include
water quality and quantity, natural cover, forage, adequate passage conditions, and
floodplain connectivity. Anthropogenic land uses introduce sediment, nutrients, biocides,
metals, and other pollutants into surface and ground water and degrade water quality in
the freshwater, estuarine, and coastal ecosystems throughout the Pacific Northwest.
These human impacts affect the essential feature requirements for this DPS.
Of 43 subbasins reviewed in NMFS' assessment of critical habitat for the UWR
steelhead, 20 subbasins were rated as having a high conservation value, while six were
rated as having a medium value and 17 were rated as having a low value to the
conservation of the DPS.
DRAFT 191
-------
DRAFT
''.,.• •
By regulation, environmental baselines for Opinions include the past and present impacts
of all state, federal or private actions and other human activities in the action area, the
anticipated impacts of all proposed federal projects in the action area that have already
undergone formal or early section 7 consultation, and the impact of state or private
actions which are contemporaneous with the consultation in process (50 CFR §402.02).
The environmental baseline for this Opinion includes a general description of the natural
and anthropogenic factors influencing the current status of listed Pacific salmonids and
the environment within the action area.
Our summary of the environmental baseline complements the information provided in the
Status of Listed Resources section of this Opinion, and provides the background
necessary to understand information presented in the Effects of the Action, and
Cumulative Effects sections of this Opinion. We then evaluate these consequences in
combination with the environmental baseline to determine the likelihood of jeopardy or
adverse modification of designated critical habitat.
The proposed action under consultation is geographically focused on the aquatic
ecosystems in the states of California, Idaho, Oregon, and Washington. Accordingly, the
environmental baseline for this consultation focuses on the general status and trends of
the aquatic ecosystems in these four states and the consequences of that status for listed
resources under NMFS' jurisdiction. We describe the overall principal natural
phenomena affecting all listed Pacific salmonids under NMFS jurisdiction in the action
area.
We further describe anthropogenic factors through the predominant land and water uses
within a region, as land use patterns vary by region. Background information on
pesticides in the aquatic environment is also provided. This context illustrates how the
physical and chemical health of regional waters and the impact of human activities have
contributed to the current status of listed resources in the action area.
DRAFT 192
-------
DRAFT
Natural Mortality Factors
Available data indicate high natural mortality rates for salmonids, especially in the open
ocean/marine environment. According to Bradford (1995), salmonid mortality rates
range from 90 to 99%, depending on the species, the size at ocean entry, and the length of
time spent in the ocean. Predation, inter- and intraspecific competition, food availability,
smolt quality and health, and physical ocean conditions likely influence the survival of
salmon in the marine environment (Brodeur, Fisher et al. 2004). In freshwater rearing
habitats, the natural mortality rate averages about 70% for all salmonid species (Bradford
1995b). Past studies in the Pacific Northwest suggest that the average freshwater survival
rate (from egg to smolt) is 2 to 3% throughout the region (Marshall and Britton 1990;
Bradford 1995b). A number of suspected causes contributing to natural mortality include
parasites and/or disease, predation, water temperature, low water flow, wildland fire, and
oceanographic features and climatic variability.
Parasites and/or Disease
Most young fish are highly susceptible to disease during the first two months of life. The
cumulative mortality in young animals can reach 90 to 95%. Although fish disease
organisms occur naturally in the water, native fish have co-evolved with them. Fish can
carry these diseases at less than lethal levels (Kier Associates 1991; Walker and Foott
1993; Foott, Harmon et al. 2003). However, disease outbreaks may occur when water
quality is diminished and fish are stressed from crowding and diminished flows (Spence,
Lomnicky et al. 1996; Guillen 2003). Young coho salmon or other salmonid species may
become stressed and lose their resistance in higher temperatures (Spence, Lomnicky et al.
1996). Consequently, diseased fish become more susceptible to predation and are less
able to perform essential functions, such as feeding, swimming, and defending territories
(McCullough 1999). Examples of parasites and disease for salmonids include whirling
disease, infectious hematopoietic necrosis (IHN), sea-lice (Lepeophtheirus salmonis),
Henneguya salminicola, Ichthyopthirius multifiliis or Ich, and Columnar! s
(Flavobacterium columnare).
Whirling disease is a parasitic infection caused by the microscopic parasite Myxobolus
DRAFT 193
-------
DRAFT
cerebrali. Infected fish continually swim in circular motions and eventually expire from
exhaustion. The disease occurs in the wild and in hatcheries and results in losses to fry
and fmgeiiing salmonids, especially rainbow trout. The disease is transmitted by infected
fish and fish parts and birds.
IHN is a viral disease in many wild and farmed salmonid stocks in the Pacific Northwest.
This disease affects rainbow/steelhead trout, cutthroat trout (Salmo clarki), brown trout
(Salmo tmtta), Atlantic salmon (Salmo salar\ and Pacific salmon including Chinook,
sockeye, chum, and coho. The virus is triggered by low water temperatures and is shed
in the feces, urine, sexual fluids, and external mucus of salmonids. Transmission is
mainly from fish to fish, primarily by direct contact and through the water.
Sea lice also cause deadly infestations of wild and farm-grown salmon. Henneguya
salminicola, a protozoan parasite, is commonly found in the flesh of salmonids. The fish
responds by walling off the parasitic infection into a number of cysts that contain milky
fluid. This fluid is an accumulation of a large number of parasites. Fish with the longest
freshwater residence time as juveniles have the most noticeable infection. The order of
prevalence for infection is coho followed by sockeye, Chinook, chum, and pink salmon.
Additionally, ich (a protozoan) and Columnaris (a bacterium) are two common fish
diseases that were implicated in the massive kill of adult salmon in the Lower Klamath
River in September 2002 (CDFG 2003; Guillen 2003).
Predation
Salmonids are exposed to high rates of natural predation, during freshwater rearing and
migration stages, as well as during ocean migration. Salmon along the U.S. west coast
are prey for marine mammals, birds, sharks, and other fishes. Concentrations of juvenile
salmon in the coastal zone experience high rates of predation. In the Pacific Northwest,
the increasing size of tern, seal, and sea lion populations may have reduced the survival
of some salmon ESUs.
DRAFT 194
-------
DRAFT
Marine Mammal Predation
Marine mammals are known to attack and eat salmonids. Harbor seals (Phoca vitulind),
California sea lions (Zalophus californianus)., and killer whales (Orcinus orca) prey on
juvenile or adult salmon. Killer whales have a strong preference for Chinook salmon (up
to 78% of identified prey) during late spring to fall (Hard, Jones et al. 1992; Hanson,
Baird et al. 2005; Ford and Ellis 2006). Generally, harbor seals do not feed on salmonids
as frequently as California sea lions (Pearcy 1997). California sea lions from the Ballard
Locks in Seattle, Washington have been estimated to consume about 40% of the
steelhead runs since 1985/1986 (Gustafson, Wainwright et al. 1997). In the Columbia
River, salmonids may contribute substantially to sea lion diet at specific times and
locations (Pearcy 1997). Spring Chinook salmon and steelhead are subject to pinniped
predation when they return to the estuary as adults ((NMFS) 2006). Adult Chinook
salmon in the Columbia River immediately downstream of Bonneville Dam have also
experienced increased predation by California sea lions. In recent years, sea lion
predation of adult LCR winter steelhead in the Bonneville tailrace has increased. This
prompted ongoing actions to reduce predation effects. They include the exclusion,
hazing, and in some cases, lethal take of marine mammals near Bonneville Dam (FCRPS
2008).
NOAA Fisheries has granted permits to the states of Idaho, Oregon, and Washington for
the lethal removal of individual California sea lions that prey on adult spring-run
Chinook salmon in the tail race of Bonneville Dam under section 120 of the Marine
Mammal Protection Act ((NMFS) 2006). This action may increase the survival of adult
Chinook salmon and steelhead. The Humane Society of the U.S. unsuccessfully
challenged the issuance of these permits. The case is now on appeal.
Avian Predation
Large numbers of fry and juveniles are eaten by birds such as mergansers (Mergus spp.),
common murre (Uria aalage), gulls (Larus spp.), and belted kingfishers (Megaceryle
alcyori). Avian predators of adult salmonids include bald eagles (Haliaeetus
DRAFT 195
-------
DRAFT
leucocephalus) and osprey (Pandion haliaetus) (Pearcy 1997). Caspian terns {Sterna
caspia) and cormorants (Phalacrocorax spp.) also take significant numbers of juvenile or
adult salmon. Stream-type juveniles, especially yearling smolts from spring-run
populations, are vulnerable to bird predation in the estuary. This vulnerability is due to
salmonid use of the deeper, less turbid water over the channel, which is located near
habitat preferred by piscivorous birds (Binelli, Ricciardi et al. 2005). Recent research
shows that subyearlings from the LCR Chinook salmon ESU are also subject to tern
predation. This may be due to the long estuarine residence time of the LCR Chinook
salmon (Ryan, Carper et al. 2006). Caspian terns and cormorants may be responsible for
the mortality of up to 6% of the outmigrating stream-type juveniles in the Columbia
River basin (Roby, Collis et al. 2006), (Collis 2007).
Antolos et al. (2006) quantified predation on juvenile salmonids by Caspian terns nesting
on Crescent Island in the mid-Columbia reach. Between 1,000 and 1,300 adult terns
were associated with the colony during 2000 and 2001, respectively. These birds
consumed about 465,000 juvenile salmonids in the first and approximately 679,000
salmonids in the second year. However, Caspian tern predation in the estuary was
reduced from a total of 13,790,000 smolts to 8,201,000 smolts after relocation of the
colony from Rice to East Sand Island in 1999. Based on PIT-tag recoveries at the colony,
these were primarily steelhead for Upper Columbia River stocks. Less than 0.1% of the
inriver migrating yearling Chinook salmon from the Snake River and less than 1% of the
yearling Chinook salmon from the Upper Columbia were consumed. PIT-tagged coho
smolts (originating above Bonnevile Dam) were second only to steelhead in predation
rates at the East Sand Island colony in 2007 (Roby, Colis et al. 2008). There are few
quantitative data on avian predation rates on Snake River sockeye salmon. Based on the
above, avian predators are assumed to have a minimal effect on the long-term survival of
Pacific salmon (FCRPS 2008).
Fish Predation
Pikeminnows (Ptychocheilus oregonensis) are significant predators of yearling juvenile
migrants (Friesen and Ward 1999). Chinook salmon were 29% of the prey of northern
DRAFT 196
-------
DRAFT
pikeminnows in lower Columbia reservoirs, 49% in the lower Snake River, and 64%
downstream of Bonneville Dam. Sockeye smolts comprise a very small fraction of the
overall number of migrating smolts (Ferguson 2006) in any given year. The significance
offish predation on juvenile chum is unknown. There is little direct evidence that
piscivorous fish in the Columbia River consume juvenile sockeye salmon. Nevertheless,
predation of juvenile sockeye likely occurs. The ongoing Northern Pikeminnow
Management Program (NPMP) has reduced predation-related juvenile salmonid mortality
since 1990. Benefits of recent northern pikeminnow management activities to chum
salmon are unknown. However, it may be comparable to those for other salmon species
with a subyearling juvenile life history (Friesen and Ward 1999).
The primary fish predators in estuaries are probably adult salmonids or juvenile
salmonids which emigrate at older and larger sizes than others. They include cutthroat
trout (O. clarkf) or steelhead smolts preying on chum or pink salmon smolts. Outside
estuaries, many large fish population reside just offshore and may consume large
numbers of smolts. These fishes include Pacific hake (Merluccius productus), Pacific
mackerel (Scomber japonicus), lingcod (Ophiodon elongates)., spiny dogfish (Squalus
acanthias\ various rock fish, and lamprey (Beamish, Thomson et al. 1992; Pearcy 1992;
Beamish and Neville 1995).
Wildland Fire
Wildland fires that are allowed to burn naturally in riparian or upland areas may benefit
or harm aquatic species, depending on the degree of departure from natural fire regimes.
Although most fires are small in size, large size fires increase the chances of adverse
effects on aquatic species. Large fires that burn near the shores of streams and rivers can
have biologically significant short-term effects. They include increased water
temperatures, ash, nutrients, pH, sediment, toxic chemicals, and large woody debris
(Buchwalter, Sandahl et al. 2004; Rinne 2004). Nevertheless, fire is also one of the
dominant habitat-forming processes in mountain streams (Bisson, Rieman et al. 2003).
As a result, many large fires burning near streams can result in fish kills with the
survivors actively moving downstream to avoid poor water quality conditions (Greswell
DRAFT 197
-------
DRAFT
1999; Rinne 2004). The patchy, mosaic pattern burned by fires provides a refuge for
those fish and invertebrates that leave a burning area or simply spares some fish that were
in a different location at the time of the fire (USFS 2000). Small fires or fires that burn
entirely in upland areas also cause ash to enter rivers and increase smoke in the
atmosphere, contributing to ammonia concentrations in rivers as the smoke adsorbs into
the water (Greswell 1999).
The presence of ash also has indirect effects on aquatic species depending on the amount
of ash entry into the water. All ESA-listed fishes rely on macroinvertebrates as a food
source for at least a portion of their life histories. When small amounts of ash enter the
water, there are usually no noticeable changes to the macroinvertebrate community or the
water quality (Bowman and Minshall 2000). When significant amounts of ash are
deposited into rivers, the macroinvertebrate community density and composition may be
moderately to drastically reduced for a full year with long-term effects lasting 10 years or
more (Buchwalter, Jenkins et al. 2003), (Minshall, Royer et al. 2001; Buchwalter,
Sandahl et al. 2004). Larger fires can also indirectly affect fish by altering water quality.
Ash and smoke contribute to elevated ammonium, nitrate, phosphorous, potassium, and
pH, which can remain elevated for up to four months after forest fires (Buchwalter,
Jenkins et al. 2003).
Many species have evolved in the presence of regular fires and have developed
population-level mechanisms to withstand even the most intense fires (Greswell 1999).
These same species have come to rely on fire's disturbance to provide habitat
heterogeneity. In the past century, the human population has increased dramatically,
resulting in urban sprawl and the development of formerly remote locations. This
condition has increased the urban/wildland interface. As a result, the threat of fires to
personal property and people has increased, including the demand for protection of their
safety and belongings. We expect listed fish species will be exposed to an increasing
number of fires and fire fighting techniques over time. Currently, federal, state, and local
resource agencies lack long-term monitoring data on the effects of wildland fire on listed
Pacific salmonids and their habitats. Thus, we are unable to quantify the overall effects
DRAFT 198
-------
DRAFT
of wildland fire on the long-term survival of listed Pacific salmonids at this time.
Oceanographic Features and Climatic Variability
Oceanographic features of the action area may influence prey availability and habitat for
Pacific salmonids. The action area includes important spawning and rearing grounds and
physical and biological features essential to the conservation of listed Pacific salmonids -
i.e., water quality, prey, and passage conditions. Ocean conditions and climatic
variability may affect salmonids in the action area.
The primary effects of the ocean on salmon productivity involve growth and survival of
salmon. All salmon growth is completed in the ocean. According to Welch (1996), fish
growth will not reach its maximum potential if food density (food available divided by
ocean volume) is insufficient to provide the maximum daily ration. If this critical level of
food is not exceeded, then the potential for the ocean to limit salmon growth exists.
The decline in salmon survival in Oregon and Washington since 1977 may be caused by
poorly understood processes in the marine (as opposed to freshwater) environment
(Welch 1996). Current findings also indicate that the primary control on salmon
distribution is temperature. However, the upper thermal limit varies throughout the year
(Welch 1996).
Naturally occurring climatic patterns, such as the Pacific Decadal Oscillation and the El
Nino and La Nina events, are major causes of changing marine productivity. Recent
studies have shown that long-term changes in climate affect oceanic structure and
produce abrupt differences in salmon marine survival and returns (Mantua, Hare et al.
1997; Hare, Mantua et al. 1999). A major regime shift in the subarctic and California
Current ecosystems during the late 1970s may have been a factor in reducing ocean
survival of salmon in the Pacific Northwest and in increasing the marine survival in
Alaska (Hare, Mantua et al. 1999). Fluctuations in mortality of salmon in the freshwater
and marine environment have been shown to be almost equally significant sources of
annual recruitment variability (Bradford 1995b). These events and changes in ocean
DRAFT 199
-------
DRAFT
temperature may also influence salmonid abundance in the action area. In years when
ocean conditions are cooler than usual, the majority of sockeye salmon returning to the
Fraser River do so via this route. However, when warmer conditions prevail, migration
patterns shift to the north through the Johnstone Strait (Groot and Quinn 1987).
Climate Change
Anthropogenic climate change, caused by factors such as the continuing build-up of
human-produced atmospheric carbon dioxide, is predicted to have major environmental
impacts along the west coast of North America during the 21st century and beyond (IPCC
2001; CIG 2004). Warming trends continue in both water and air temperatures.
Projections of the consequences of climate change include disruption of annual cycles of
rain and snow, alteration of prevailing patterns of winds and ocean currents, and
increases in sea levels (Glick 2005; Snover, Mote et al. 2005). Oceanographic models
project a weakening of the therm ohaline circulation resulting in a reduction of heat
transport into high latitudes of Europe, an increase in the mass of the Antarctic ice sheet,
and a decrease in the Greenland ice sheet (IPCC 2001). These changes, coupled with
increased acidification of ocean waters, are expected to have substantial effects on marine
productivity and food webs, including populations of salmon and other salmonid prey
(Hard, Jones et al. 1992).
Climate change poses significant hazards to the survival and recovery of salmonids along
the west coast. Changes in water temperature can alter migration timing, reduce growth,
reduce the supply of available oxygen in the water, reduce insect availability as prey, and
increase the susceptibility offish to toxicants, parasites, and disease (Fresh, Casillas et
al. 2005; NMFS 2007). Earlier spring runoff and lower summer flows make it difficult
for returning adult salmon to negotiate obstacles (NMFS 2007). Excessively high levels
of winter flooding can scour eggs from their nests in the stream beds and increase
mortalities among overwintering juvenile salmon. The predicted increased winter
flooding, decreased summer and fall stream flows, and elevated warm season
temperatures in the streams and estuaries may further degrade conditions for salmon that
are already stressed from habitat degradation. Although the impacts of global climate
DRAFT 200
-------
DRAFT
change are less clear in the ocean environment, early modeling efforts suggest that
increased temperatures will likely increase ocean stratification. This stratification
coincides with relatively poor ocean habitat for most Pacific Northwest salmon
populations (IPCC 2001; CIG 2004).
We expect changing weather and oceanographic conditions may affect prey availability,
temperature and water flow in habitat conditions, and growth for all 28 ESUs.
Consequently, we expect the long-term survival and reproductive success for listed
salmonids to be greatly affected by global climate change.
Anthropogenic Mortality Factors
In this section we address anthropogenic threats in the geographic regions across the
action area. Among the threats discussed are the "four Hs": hatcheries, harvest,
hydropower, and habitat. Prior to discussion of each geographic region, three major
issues are highlighted: pesticide contamination, elevated water temperature, and loss of
habitat/habitat connectivity. These three factors are the most relevant to the current
analysis. To address these issues, we provide information on pesticide detections in the
aquatic environment and highlight their background levels from past and ongoing
anthropogenic activities. This information is pertinent to EPA's proposed registration of
carbaryl, carbofuran, and methomyl in the U.S. and its territories. As water temperature
plays such a strong role in salmonid distribution, we also provide a general discussion of
anthropogenic temperature changes. Finally, we discuss the health of riparian systems
and floodplain connectivity, as this habitat is vital to salmonid survival.
Baseline Pesticide Detections in Aquatic Environments
In the environmental baseline, we address pesticide detections reported as part of the U.S.
Geological Survey (USGS) National Water-Quality Assessment Program's (NAWQA)
national assessment (Gilliom, Barbash et al. 2006). We chose this approach as the
NAWQA studies present the same level of analysis for each area. Further, given the lack
of reporting standards, we are unable to present a comprehensive basin-specific analysis
DRAFT 201
-------
DRAFT
of detections from other sources.
In the exposure section of the Effects of the Proposed Action we also present more recent
unpublished data on the chemicals and degradates addressed in this Opinion from the
NAWQA program and state databases maintained by California and Washington. As far
as NMFS was able to ascertain, neither Oregon nor Idaho maintain publically available
state-wide water quality databases. The California and Washington databases include
some data from the NAWQA, but mostly the data are from more localized studies.
Overall, data from those databases are relatively consistent, with carbaryl generally being
the most frequently quantifiable parent compound. Carbaryl and carbofuran were
measured in concentrations ranging from 0.0001-33.5 |J,g/L. Methomyl generally was
measured at slightly lower concentrations, ranging from 0.004-5.4 |J,g/L. Methomyl is
also detected less frequently in some monitoring datasets, as it dissipates rapidly in
aquatic systems, and non-targeted monitoring does not necessarily coincide with
applications. Both 1-napthol and 3-hydroxycarbofuran were detected in slightly lower
concentrations, ranging from 0.0007-0.64 |J,g/L, than any of the parent compounds.
According to Gilliom et al. (2006), the distributions of the most prevalent pesticides in
streams and ground water correlate with land use patterns and associated present or past
pesticide use. When pesticides are released into the environment, they frequently end up
as contaminants in aquatic environments. Depending on their physical properties some
are rapidly transformed via chemical, photochemical, and biologically mediated reactions
into other compounds, known as degradates. These degradates may become as prevalent
as the parent pesticides depending on their rate of formation and their relative persistence.
National Water-Quality Assessment Program.
From 1992-2001, the USGS sampled water from 186 stream sites within 51 study units;
bed-sediment samples from 1,052 stream sites, and fish from 700 stream sites across the
continental U.S. Concentrations of pesticides were detected in streams and groundwater
within most areas sampled with substantial agricultural or urban land uses. NAWQA
results further detected at least one pesticide or degradate more than 90% of the time in
DRAFT 202
-------
DRAFT
water, in more than 80% in fish samples, and greater than 50% of bed-sediment samples
from streams in watersheds with agricultural, urban, and mixed land use (Gilliom,
Barbash et al. 2006).
About 40 pesticide compounds accounted for most detections in water, fish, or bed
sediment. Twenty-four pesticides and one degradate were each detected in more than
10% of streams in agricultural, urban, or mixed land use settings. These 25 pesticide
compounds include 11 herbicides used most heavily in agriculture during the study
period (plus the atrazine degradate, deethylatrazine); 7 herbicides used extensively for
non-agricultural purposes; and 6 insecticides used in both agricultural and urban settings.
Three of those insecticides were chlorpyrifos, diazinon, and malathion. Thirteen
organochlorine pesticide compounds, including historically used parent pesticides and
their degradates and by-products, were each found in more than 10% offish or bed-
sediment samples from streams draining watersheds with either agricultural, urban, or
mixed land use (Gilliom, Barbash et al. 2006).
Additionally, more frequent detections and higher concentrations of insecticides occur in
sampled urban streams (Gilliom, Barbash et al. 2006). Diazinon, chlorpyrifos, carbaryl,
and malathion nationally ranked 2nd, 4th, 8th, and 15th among pesticides in frequencies of
outdoor applications for home- and garden use in 1992 (Whitmore, Kelly et al. 1992).
These same insecticides accounted for the most insecticide detections in urban streams.
Diazinon and carbaryl were the most frequently detected and were found at frequencies
and levels comparable to those for the common herbicides. Historically used insecticides
were also found most frequently in fish and bed sediment from urban streams. The
highest detection frequencies were for chlordane compounds, dichloro diphenyl
trichloroethane (DDT) compounds, and dieldrin. Urban streams also had the highest
concentrations of total chlordane and dieldrin in both sediment and fish tissue. Chlordane
and aldrin were widely used for termite control until the mid-to-late 1980s. Their
agricultural uses were restricted during the 1970s.
Chlorpyrifos and diazinon were commonly used in agricultural and urban areas from
DRAFT 203
-------
DRAFT
1992-2001 and prior to the sampling period. About 13 million Ibs of chlorpyrifos and
about 1 million Ibs of diazinon were applied for agricultural use. Non-agricultural uses
of chlorpyrifos and diazinon totaled about 5 million and 4 million Ibs per year in 2001,
respectively (Gilliom, Barbash et al. 2006). For both insecticides, concentrations in most
urban streams were higher than in most agricultural streams, and were similar to those
found in agricultural areas with the greatest intensities of use. Diazinon and chlorpyrifos
were detected about 75% and 30% of the time in urban streams, respectively (Gilliom,
Barbash et al. 2006). NMFS (2008) determined that current use of chlorpyrifos,
diazinon, and malathion is likely to jeopardize the continued existence of 27 listed
salmonid ESUs. NMFS provided EPA with reasonable and prudent alternatives (RPAs),
including buffers and vegetative strips, to reduce pesticide exposure to listed salmon.
Until the EPA implements the RPAs, we must assume current exposure will continue.
Another dimension of pesticides and their degradates in the aquatic environment is their
simultaneous occurrence as mixtures (Gilliom, Barbash et al. 2006). Mixtures result
from the use of different pesticides for multiple purposes within a watershed or
groundwater recharge area. Pesticides generally occur more often as mixtures than as
individual compounds. Mixtures of pesticides were detected more often in streams than
in ground water and at relatively similar frequencies in streams draining areas of
agricultural, urban, and mixed land use. More than 90% of the time, water from streams
in these developed land use settings had detections of two or more pesticides or
degradates. About 70% and 20% of the time, streams had five or more and ten or more
pesticides or degradates, respectively (Gilliom, Barbash et al. 2006). Fish experiencing
coincident exposure to multiple pesticides also have additive and synergistic effects. If
the effects on a biological endpoint from concurrent exposure to multiple pesticides can
be predicted by adding the potency of the pesticides involved, the effects are said to be
additive. If, however, the response to a mixture lead to a greater than expected effect on
the endpoint, and the pesticides within the mixture enhance the toxicity of one another,
the effects are characterized as synergistic. These effects are of particular concern when
the pesticides share a mode of action. Carbaryl, carbofuran, and methomyl are all AChE
inhibitors. In California, there are 61 pesticides that inhibit AChE approved for use
DRAFT 204
-------
DRAFT
(CDPR 2007). According to CDPR, the amount of these chemicals used has decreased
(Table 26). However, some AChE a.i.s - such as bensulide and naled - are increasing in
use (CDPR 2007). While the trend indicates decreased reliance on these products, we
note that their current use remains significant.
Table 26. Use figures for AChE inhibiting pesticides in California (CDPR 2007)
Ibs a.i. applied
Acres treated (agriculture use only)
1996
15,473,843
11,720,058
2006
6,857,530
5,729,958
Mixtures of organochlorine pesticide compounds were also common in fish-tissue
samples from most streams. About 90% offish samples collected from urban steams
contained two or more pesticide compounds and 33% contained 10 or more pesticides.
Similarly, 75% offish samples from streams draining watersheds with agricultural and
mixed land use contained 2 or more pesticide compounds and 10% had 10 or more
compounds (Gilliom, Barbash et al. 2006).
NAWQA analysis of all detections indicaes that more than 6,000 unique mixtures of 5
pesticides were detected in agricultural streams (Gilliom, Barbash et al. 2006). The
number of unique mixtures varied with land use. Mixtures of the most often detected
individual pesticides include the herbicides atrazine (and its degradate deethylatrazine),
metolachlor, simazine, and prometon. Each herbicide occurred in more than 30% of all
mixtures found in agricultural and urban uses in streams. Also present in more than 30%
of the mixtures were cyanazine, alachlor, metribuzin, and trifluralin in agricultural
streams. Dacthal and the insecticides diazinon, chlorpyrifos, carbaryl, and malathion
were also present in urban streams. Carbaryl occurred in at least 50% of urban streams.
In 15% of urban streams carbaryl concentration was over O.lug/L (Gilliom, Barbash et al.
2006). Insecticides are typical constituents in mixtures and are commonly found in urban
streams.
The numbers of unique mixtures of organochlorine pesticide compounds found in whole-
fish tissue samples were greater in urban streams than in streams from agricultural or
DRAFT
205
-------
DRAFT
mixed land use watersheds. About 1,400 unique 5-compound mixtures were found in
fish from urban steams compared to fewer than 800 unique 5-compound mixtures
detected in fish from agricultural and mixed land use steams. The relative contributions
of most organochlorine compounds to mixtures in fish were about the same for urban and
agricultural streams.
More than half of all agricultural streams sampled and more than three-quarters of all
urban streams had concentrations of pesticides in water that exceeded one or more
benchmarks for aquatic life. Aquatic life criteria are EPA water-quality guidelines for
protection of aquatic life. Exceedance of an aquatic life benchmark level indicates a
strong probability that aquatic species are being adversely affected. However, aquatic
species may also be affected at levels below criteria. Finally, organochlorine pesticides
that were discontinued 15 to 30 years ago still exceeded benchmarks for aquatic life and
fish-eating wildlife in bed sediment or fish-tissue samples from many streams.
National Pollutant Discharge Elimination System
Pollution originating from a discrete location such as a pipe discharge or wastewater
treatment outfall is known as a point source. Point sources of pollution require a National
Pollutant Discharge Elimination System (NPDES) permit. These permits are issued for
aquaculture, concentrated animal feeding operations, industrial wastewater treatment
plants, biosolids (sewer/sludge), pre-treatment and stormwater overflows. The EPA
administers the NPDES permit program and states certify that NPDES permit holders
comply with state water quality standards. Nonpoint source discharges do not originate
from discrete points; thus, nonpoint sources are difficult to identify, quantify, and are not
regulated. Examples of nonpoint source pollution include, but are not limited to, urban
runoff from impervious surfaces, areas of fertilizer and pesticide application, and manure.
According to EPA's database of NPDES permits, about 243 NPDES permits are co-
located with listed Pacific salmonids in California. Collectively, the total number of
EPA-recorded NPDES permits in Idaho, Oregon, and Washington, that are co-located
with listed Pacific salmonids is 1,978. See ESU Figures in the Status of Listed Resources
DRAFT 206
-------
DRAFT
section for NPDES permits co-located within listed salmonid ESUs within the states of
California, Idaho, Oregon, and Washington.
On November 27, 2006, EPA issued a final rule which exempted pesticides from the
NPDES permit process, provided that application was approved under FIFRA. The
NPDES permits, then, do not include any point source application of pesticides to
waterways in accordance with FIFRA labels. This rule was vacated by the courts on
January 7, 2009 (National Cotton Council v. EPA. 553 F.3d 927 (6th Cir. 2009)).
Baseline Water Temperature- Clean Water Act
Elevated temperature is considered a pollutant in most states with approved Water
Quality Standards under the federal Clean Water Act (CWA) of 1972. As per the CWA,
states periodically prepare a list of all surface waters in the state for which beneficial uses
- such as drinking, recreation, aquatic habitat, and industrial use - are impaired by
pollutants. These are water quality limited estuaries, lakes, and streams that do not meet
state surface water quality standards, and are not expected to improve within the next two
years. This process is in accordance with section 303(d) of the CWA. Water bodies
listed under 303(d) are those that are considered impaired or threatened by pollution.
Each state has separate and different 303(d) listing criteria and processes. Generally a
water body is listed separately for each standard it exceeds, so it may appear on the list
more than once. If a water body is not on the 303(d) list, it is not necessarily
contaminant-free; rather it may not have been tested. Therefore, the 303(d) list is a
minimum list for the each state regarding polluted water bodies by parameter.
After states develop their lists of impaired waters, they are required to prioritize and
submit their lists to EPA for review and approval. Each state establishes a priority
ranking for such waters, considering the severity of the pollution and the uses to be made
of such waters. States are expected to identify high priority waters targeted for Total
Maximum Daily Load (TMDL) development within two years of the 303(d) listing
process.
DRAFT 207
-------
DRAFT
Temperature is significant for the health of aquatic life. Water temperatures affect the
distribution, health, and survival of native cold-blooded salmonids in the Pacific
Northwest. These fish will experience adverse health effects when exposed to
temperatures outside their optimal range. For listed Pacific salmonids, water temperature
tolerance varies between species and life stages. Optimal temperatures for rearing
salmonids range from 10°C and 16°C. In general, the increased exposure to stressful
water temperatures and the reduction of suitable habitat caused by drought conditions
reduce the abundance of salmon. Warm temperatures can reduce fecundity, increase egg
survival, retard growth of fry and smolts, reduce rearing densities, increase susceptibility
to disease, decrease the ability of young salmon and trout to compete with other species
for food, and to avoid predation (Spence, Lomnicky et al. 1996; McCullough 1999).
Migrating adult salmonids and upstream migration can be delayed by excessively warm
stream temperatures. Excessive stream temperatures may also negatively affect
incubating and rearing salmonids (Gregory and Bisson 1997).
Sublethal temperatures (above 24°C) could be detrimental to salmon by increasing
susceptibility to disease (Colgrove and Wood 1966) or elevating metabolic demand (Brett
1995). Substantial research demonstrates that many fish diseases become more virulent
at temperatures over 15.6°C (McCullough 1999). Due to the sensitivity of salmonids to
temperature, states have established lower temperature thresholds for salmonid habitat as
part of their water quality standards. A water body is listed for temperature on the 303(d)
list if the 7-day average of the daily maximum temperatures (7-DADMax) exceeds the
temperature threshold (Table 27).
DRAFT 208
-------
DRAFT
Table 27. Washington State water temperature thresholds for salmonid habitat. These
temperatures are representative of limits set by California, Idaho, and Oregon (WSDE
2006).
Category
Salmon and Trout Spawning
Core Summer Salmonid Habitat
Salmonid Spawning, Rearing, and Migration
Salmonid Rearing and Migration Only
Highest 7-DADMax
13°C(55.4°F)
16°C(60.8°F)
17.5°C(63.5°F)
17.5°C(63.5°F)
Water bodies that are not designated salmonid habitat are also listed if they have a one-
day maximum over a given background temperature. Using publicly available GIS
layers, we determined the number of km on the 303(d) list for exceeding temperature
thresholds within the boundaries of each ESU (Table 28). Because the 303(d) list is
limited to the subset of rivers tested, the chart values should be regarded as
underestimates.
DRAFT
209
-------
DRAFT
Table 28. Number of kilometers of river, stream and estuaries included in state 303(d) lists
due to temperature that are located within each salmonid ESU. Data was taken from the
most recent GIS layers available from state water quality assessments reports*
Species
Chinook
Salmon
Chum
Salmon
Coho
Salmon
Sockeye
Salmon
Steelhead
Trout
ESU
California Coastal
Central Valley Spring - Run
Lower Columbia River
Upper Columbia River
Spring - Run
Puget Sound
Sacramento River Winter -
Run
Snake River Fall - Run
Snake River Spring /
Summer - Run
Upper Williamette River
Columbia River
Hood Canal Summer - Run
Central California Coast
Lower Columbia River
Southern Oregon and
Northern California Coast
Oregon Coast
Ozette Lake
Snake River
Central California Coast
California Central Valley
Lower Columbia River
Middle Columbia River
Northern California
Puget Sound
Snake River
South-Central California
Coast
Southern California
Upper Columbia River
Upper Williamette River
California
39.3
0.0
—
-
-
0.0
-
-
-
-
-
39.3
-
1,416.2
-
-
-
0.0
0.0
-
-
39.3
-
-
0.0
0.0
-
-
Oregon
-
-
56.6
-
-
-
610.1
809.3
2,468.0
56.6
-
-
291.9
1,833.0
3,715.8
-
-
-
-
201.2
3,518.5
-
-
990.7
-
—
-
1,668.0
Washington Idaho
- -
- -
229.8
254.6
705.0
- -
246.6 400.2
243.2 543.8
- -
225.0
90.1
- -
233.5
- -
- -
4.8
0.0
- -
- -
169.3
386.2
- -
704.9
246.6 737.6
- -
— —
282.3
- -
Total
39.3
0.0
286.4
254.6
705.0
0.0
1,256.9
1,596.3
2,468.0
281.6
90.1
39.3
525.4
3,249.2
3,715.8
4.8
0.0
0.0
0.0
370.5
3,904.7
39.3
704.9
1,974.9
0.0
0.0
282.3
1,668.0
*CA 2006, Oregon 2004/2006, Washington 2004, and Idaho 1998. (California EPA TMDL Program 2007b, Oregon Department of
Environmental Quality 2007, Washington State Department of Ecology 2005, Idaho Department of Environmental Quality 2001).
DRAFT
210
-------
DRAFT
While some ESU ranges do not contain any 303(d) rivers listed for temperature, others
show considerable overlap. These comparisons demonstrate the relative significance of
elevated temperature among ESUs. Increased water temperature may result in
wastewater discharge, decreased water flow, minimal shading by riparian areas, and
climatic variation.
Baseline Habitat Condition
Riparian zones are the areas of land adjacent to rivers and streams. These systems serve
as the interface between the aquatic and terrestrial environments. Riparian vegetation is
characterized by emergent aquatic plants and species that thrive on close proximity to
water, such as willows. This vegetation maintains a healthy river system by reducing
erosion, stabilizing main channels, and providing shade. Leaf litter that enters the river
becomes an important source of nutrients for invertebrates (Bisson and Bilby 2001).
Riparian zones are also the major source of large woody debris (LWD). When trees fall
and enter the water, they become an important part of the ecosystem. The LWD alters
the flow, creating the pools of slower moving water preferred by salmon (Bilby, Fransen
et al. 2001). While not necessary for pool formation, LWD is associated with around
80% of pools in northern California, Washington, and the Idaho pan-handle (Bilby and
Bisson 2001).
Bilby and Bisson (2001) discuss several studies that associate increased LWD with
increased pools, and both pools and LWD with salmonid productivity. Their review also
includes documented decreases in salmonid productivity following the removal of LWD.
Other benefits of LWD include deeper pools, increased sediment retention, and channel
stabilization.
Floodplains are relatively flat areas adjacent to larger streams and rivers. They allow for
the lateral movement of the main channel and provide storage for floodwaters during
periods of high flow. Water stored in the floodplain is later released during periods of
low flow. This process ensures adequate flows for salmonids during the summer months,
DRAFT 211
-------
DRAFT
and reduces the possibility of high-energy flood events destroying salmonid redds (Smith
2005).
Periodic flooding of these areas creates habitat used by salmonids. Storms also wash
sediment and LWD into the main stem river, often resulting in blockages. These
blockages force the water to take an alternate path and result in the formation of side
channels and sloughs (Benda et al. 2001). Side channels and sloughs are important
spawning and rearing habitat for salmonids. The degree to which these off-channel
habitats are linked to the main channel via surface water connections is referred to as
connectivity (PNERC 2002). As river height increases with heavier flows, more side
channels form and connectivity increases.
Healthy riparian habitat and floodplain connectivity are vital for supporting a salmonid
population. Once the area has been disturbed, it can take decades to recover (Smith
2005). Consequently, most land use practices cause some degree of impairment.
Development leads to construction of levees and dikes, which isolate the mainstem river
from the floodplain. Agricultural development and grazing in riparian areas also
significantly change the landscape. Riparian areas managed for logging, or logged in the
past, are often impaired by a change in species composition. Most areas in the northwest
were historically dominated by conifers. Logging results in recruitment of deciduous
trees, decreasing the quality of LWD in the rivers. Deciduous trees have smaller
diameters than conifers; they decompose faster and are more likely to be displaced
(Smith 2005).
Without a properly functioning riparian zone, salmonids contend with a number of
limiting factors. They face reductions in both off-channel and pool habitats. Also, when
seasonal flows are not moderated, both higher and lower flow conditions exist. Higher
flows can displace fish and destroy redds, while lower flows cut off access to parts of
their habitat. Finally, decreased vegetation limits the available shade and cover, exposing
individuals to higher temperatures and increased predation.
DRAFT 212
-------
DRAFT
Geographic Regions
For a more fine scale analysis, we divided the action area into geographic regions: the
Southwest Coast Region (California) and the Pacific Northwest Region (Idaho, Oregon,
and Washington). The Pacific Northwest Region was further subdivided according to
ecoregions or other natural features important to NMFS trust resources. Use of these
geographic regions is consistent with previous NMFS consultations conducted at the
national level (NMFS 2007d). We summarize the principal anthropogenic factors
occurring in the environment that influence the current status of listed species within each
region. Table 29 provides a breakdown of these regions and includes the USGS
subregions and accounting units for each region. It also provides a list of ESUs found in
each accounting unit, as indicated by Federal Register listing notices.
Southwest Coast Region
The basins in this section occur in the State of California and the southern parts of the
State of Oregon. Table 30 and Table 31 show land area in km2 for each ESU /DPS
located in the Southwest Coast Region.
Table 29. USGS Subregions and accounting units within the Northwest and Southwest
Regions, along with ESUs present within the area (Seaber, Kapinos et al. 1987).
Region
Pacific
Northwest:
Columbia River
Basin
USGS
Subregion
Upper
Columbia River
Basin
Yakima River
Basin
Lower Snake
River Basin
Accounting
Unit
—
—
Lower
Snake River
Basin
Salmon
River Basin
State
WA
WA
ID,
OR,
WA
ID
HUC
no.
170200
170300
170601
170602
ESU
Upper Columbia Spring-run
Chinook; Upper Columbia
Steelhead; Middle Columbia
Steelhead
Middle Columbia Steelhead
Snake River Steelhead;
Snake River Spring/Summer-
run Chinook; Snake River
Fall-run Chinook; Snake
River Sockeye
Snake River Steelhead;
Snake River Spring/Summer
- Run Chinook; Snake River
Fall - Run Chinook; Snake
River Sockeye
DRAFT
213
-------
DRAFT
Region
Pacific
Northwest:
Coastal
Drainages
Pacific
Northwest: Puget
Sound
Southwest Coast
USGS
Subregion
Middle
Columbia River
Basin
Lower
Columbia River
Basin
Willamette
River Basin
Oregon-
Washington
Coastal Basin
Puget Sound
Klamath-
Northern
California
Coastal
Sacramento
River Basin
Accounting
Unit
Clearwater
River Basin
Middle
Columbia
River Basin
John Day
River Basin
Deschutes
River Basin
—
—
Washington
Coastal
Northern
Oregon
Coastal
Southern
Oregon
Coastal
—
Northern
California
Coastal
Klamath
River Basin
Lower
Sacramento
River Basin
State
ID,
WA
OR,
WA
OR
OR
OR,
WA
OR
WA
OR
OR
WA
CA
CA,
OR
CA
HUC
no.
170603
170701
170702
170703
170800
170900
171001
171002
171003
171100
180101
180102
180201
ESU
Snake River Steelhead;
Snake River Fall - Run
Chinook
Middle Columbia Steelhead;
Lower Columbia Chinook;
Columbia Chum; Lower
Columbia Coho
Middle Columbia Steelhead
Middle Columbia Steelhead
Lower Columbia Chinook;
Columbia Chum; Lower
Columbia Steelhead; Lower
Columbia Coho
Upper Willamette Chinook;
Upper Willamette Steelhead;
Lower Columbia Chinook;
Lower Columbia Steelhead;
Lower Columbia Coho
Ozette Lake Sockeye
Oregon Coast Coho
Oregon Coast Coho;
Southern Oregon and
Northern California Coast
Coho
Puget Sound Chinook; Hood
Canal Summer - Run Chum;
Puget Sound Steelhead
Southern Oregon and
Northern California Coast
Coho; California Coastal
Chinook; Northern California
Steelhead; Central California
Coast Steelhead; Central
California Coast Coho
Southern Oregon and
Northern California Coast
Coho
Central Valley Spring-run
Chinook; California Central
Valley Steelhead;
Sacramento River Winter-run
Chinook
DRAFT
214
-------
DRAFT
Region
USGS
Subregion
San Joaquin
River Basin
San Francisco
Bay
Central
California
Coastal
Southern
California
Coastal
Accounting
Unit
—
—
—
Ventura-
San Gabriel
Coastal
Laguna- San
Diego
Coastal
State
CA
CA
CA
CA
CA
HUC
no.
180400
180500
180600
180701
180703
ESU
California Central Valley
Steelhead
Central California Coast
Steelhead; Southern Oregon
and Northern California
Coast Coho; Central
California Coast Coho;
Sacramento River Winter-run
Chinook
Central California Coast
Steelhead; Southern Oregon
and Northern California
Coast Coho; South-Central
California Coast Steelhead;
Southern California
Steelhead; Central California
Coast Coho; Sacramento
River Winter-run Chinook
Southern California
Steelhead
Southern California
Steelhead
DRAFT
215
-------
DRAFT
Table 30. Area of land use categories within the range Chinook and Coho Salmon ESUs in
km2. Land cover image data were taken from Multi-Resolution Land Characteristics
(MRLC) Consortium, a consortium of nine federal agencies (USGS, EPA, USFS, NOAA,
NASA, BLM, NFS, NRCS, and USFWS) (National Land Cover Data 2001). Land cover class
definitions are available at: http://www.mrlc.gov/nlcd definitions.php
Landcover Type
code
Open Water
Perennial
Snow/Ice
Developed,
Open Space
Developed,
Low Intensity
Developed,
Medium
Intensity
Developed,
High Intensity
Barren Land
Deciduous
Forest
Evergreen
Forest
Mixed Forest
Shrub/Scrub
Herbaceous
Hay/Pasture
Cultivated
Crops
Woody
Wetlands
Emergent
Herbaceous
Wetlands
TOTAL (inc.
open water)
TOTAL (w/o
open water)
11
12
21
22
23
24
31
41
42
43
52
71
81
82
90
95
Chinook Salmon
CA
Coastal
128
0
826
137
95
10
70
850
10,700
1,554
3,801
2,114
183
212
42
18
20,740
20,612
Central
Valley
346
0
1,150
578
567
135
158
664
3,761
479
3,203
6,317
769
5,110
191
553
23,982
23,636
Sacramento
River
0
12
16
313
0
313
40
7
1
51
0
12
11
0
0
18
792
792
Coho Salmon
Central CA
Coast
157
0
629
171
138
30
23
208
4,752
922
1,620
1,646
6
233
25
13
10,572
10,415
So. Oregon
and No. CA
197
11
1,384
225
92
23
261
1,057
28,080
2,426
8,864
2,708
736
454
130
50
46,697
46,499
DRAFT
216
-------
DRAFT
Table 31. Area of Land Use Categories within the Range of Steelhead Trout DPSs (km2).
Land cover image data were taken from Multi-Resolution Land Characteristics (MRLC)
Consortium, a consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA,
BLM, NFS, NRCS, and USFWS) (National Land Cover Data 2001). Land cover class
definitions are available at: http://www.mrlc.gov/nlcd definitions.php
LandcoverType code
Open Water
Perennial Snow/Ice
Developed, Open
Space
Developed, Low
Intensity
Developed,
Medium Intensity
Developed, High
Intensity
Barren Land
Deciduous Forest
Evergreen Forest
Mixed Forest
Shrub/Scrub
Herbacous
Hay/Pasture
Cultivated Crops
Woody Wetlands
Emergent
Herbacous
Wetlands
TOTAL (inc. open
water)
TOTAL (w/o open
water)
11
12
21
22
23
24
31
41
42
43
52
71
81
82
90
95
Steelhead
Central
CA Coast
1,406
0
1,224
876
1,223
327
26
179
2,506
2,086
2,253
3,588
36
486
36
392
16,645
15,240
CA Central
Valley
409
0
1,431
693
744
181
202
751
3,990
598
3,745
9,435
1,671
9,054
248
450
33,601
33,193
Northern
CA
106
0
610
50
32
3
63
763
9,790
1,159
2,878
1,478
179
14
32
17
17,173
17,067
South-
Central CA
Coast
127
0
1,019
247
168
23
303
1
1,721
1,925
4,952
6,194
203
1,297
93
73
18,345
18,218
Southern
CA
86
0
685
364
262
12
62
0
835
897
4,370
1,516
141
653
35
35
9,954
9,868
Select watersheds described herein characterize the past, present, and future human
activities and their impacts on the area. The Southwest Coast region encompasses all
Pacific Coast rivers south of Cape Blanco, Oregon through southern California. NMFS
has identified the Cape Blanco area as an ESU/DPS biogeographic boundary for Chinook
and coho salmon, and Steelhead based on strong genetic, life history, ecological and
habitat differences north and south of this landmark. Major rivers contained in this
DRAFT
217
-------
DRAFT
grouping of watersheds are the Sacramento, San Joaquin, Salinas, Klamath, Russian,
Santa Ana, and Santa Margarita Rivers (Table 32).
Table 32. Select rivers in the southwest coast region (Carter and Resh 2005).
Watershed
Rogue River
Klamath River
Eel River
Russian River
Sacramento
River
San Joaquin
River
Salinas River
Santa Ana River
Santa Margarita
River
Approx
Length
(mi)
211
287
200
110
400
348
179
110
27
Basin
Size
(mi2)
5,154
15,679
3,651
1,439
27,850
83,409
4,241
2,438
1,896
Physiographic
Provinces*
CS, PB
PB, B/R, CS
PB
PB
PB, CS, B/R
PB, CS
PB
PB
LC, PB
Mean
Annual
Precipitation
(in)
38
33
52
41
35
49
14
13
49.5
Mean
Discharge
(cfs)
10,065
17,693
7,416
2,331
23,202
4,662
448
60
42
No.
Fish
Species
(native)
23(14)
48 (30)
25(15)
41 (20)
69 (29)
63
36(16)
45(9)
17(6)
No.
Endangered
Species
11
41
12
43
>50T&Espp.
>50T&Espp.
42 T & E spp.
54
52
* Physiographic Provinces: PB = Pacific Border, CS = Cascades-Sierra Nevada Range, B/R =
Basin & Range.
Land Use
Forest and vacant land are the dominant land uses in the northern basins. Grass,
shrubland, and urban uses are the dominant land uses in the southern basins (Table 33).
Overall, the most developed watersheds are the Santa Ana, Russian, and Santa Margarita
rivers. The Santa Ana watershed encompasses portions of San Bernardino, Los Angeles,
Riverside, and Orange counties. About 50% of the coastal subbasin in the Santa Ana
watershed is dominated by urban land uses and the population density is about 1,500
people per square mile. When steep and undevelopable lands are excluded from this
area, the population density in the watershed is about 3,000 people per square mile.
However, the most densely populated portion of the basin is near the City of Santa Ana.
Here, the population density reaches 20,000 people per square mile (Burton, Izbicki et al.
1998; Belitz, Hamlin et al. 2004). The basin is home to nearly 5 million people.
However, this population is projected to increase two-fold in the next 50 years (Burton,
Izbicki et al. 1998; Belitz, Hamlin et al. 2004).
DRAFT
218
-------
DRAFT
Table 33. Land uses and population density in several southwest coast watersheds
(Carter and Resh 2005).
Watershed
Rogue River
Klamath River
Eel River
Russian River
Sacramento River
San Joaquin River
Salinas River
Santa Ana River
Santa Margarita River
Land Use Categories (Percent)
Agriculture
6
6
2
14
15
30
13
11
12
Forest
83
66
65
50
49
27
17
57
11
Urban
<1
<1
<1
3
2
2
1
32
3
Other
9 grass & shrub
24 grass, shrub,
wetland
31 grass & shrub
31 (23
grassland)
30 grass & shrub
36 grass & shrub
65(49
grassland)
—
71 grass & shrub
Density
(people/mi2)
32
5
9
162
61
76
26
865
135
As a watershed becomes urbanized, population increases and changes occur in stream
habitat, water chemistry, and the biota (plants and animals) that live there. The most
obvious effect of urbanization is the loss of natural vegetation which results in an
increase in impervious cover and dramatic changes to the natural hydrology of urban
streams. Urbanization generally results in land clearing, soil compaction, modification
and/or loss of riparian buffers, and modifications to natural drainage features (Richter
2002). The increased impervious cover in urban areas leads to increased volumes of
runoff, increased peak flows and flow duration, and greater stream velocity during storm
events.
Runoff from urban areas also contains all the chemical pollutants from automobile traffic
and roads as well as those from industrial sources and residential use. Urban runoff is
also typically warmer than receiving waters and can significantly increase temperatures
in small urban streams. Warm stream water is detrimental to native aquatic life resident
fish and the rearing and spawning needs of anadromous fish. Wastewater treatment
plants replace septic systems to treat greater quantities of human waste and combined
sewer /stormwater overflows (CSOs). Wastewater treatment plant outfalls often
discharge directly into the rivers containing salmonids. These urban nonpoint and point
source discharges affect the water quality and quantity in basin surface waters.
DRAFT
219
-------
DRAFT
In many basins, agriculture is the major water user and the major source of water
pollution to surface waters. In 1990, nearly 95% of the water diverted from the San
Joaquin River was diverted for agriculture. Additionally, 1.5% of the water was diverted
for livestock (Carter and Resh 2005). The amount and extent of water withdrawals or
diversions for agriculture impact streams and their inhabitants via reduced water
flow/velocity and dissolved oxygen levels. For example, adequate water flow is required
for migrating salmon along freshwater, estuarine, and marine environments in order to
complete their life cycle. Low flow events may delay salmonid migration or lengthen
fish presence in a particular water body until favorable flow conditions permit fish
migration along the migratory corridor or into the open ocean.
Water diversions may also increase nutrient load, sediments (from bank erosion), and
temperature. Flow management and climate changes have decreased the delivery of
suspended parti culate matter and fine sediment to the estuary. The conditions of the
habitat (shade, woody debris, overhanging vegetation) whereby salmonids are
constrained by low flows also may make them more or less vulnerable to predation,
elevated temperatures, crowding, and disease. Water flow effects on salmonids may
seriously impact adult migration and water quality conditions for spawning and rearing
salmonids. High temperature may also result from the loss of vegetation along streams
that used to shade the water and from new land uses (buildings and pavement) whereby
rainfall picks up heat before it runs off into the stream.
Currently, California has over 500 water bodies on its 303(d) list (Wu 2000). The 2006
list includes 779 stream segments, rivers, lakes, and estuaries and 12 pollutant categories
(CEPA 2007). Pollutants represented on the list include pesticides, metals, sediments,
nutrients or low dissolved oxygen, temperature, bacteria and pathogens, and trash or
debris. There are 2,237 water body/pollutant listings; a water body is listed separately for
each pollutant detected (CEPA 2007). The 2006 303(d) list identifies water bodies listed
due to the presence of specific pollutants, including carbofuran and elevated temperature
(Table 34). See species ESU/DPS maps for NPDES permits and 303(d) waters co-
located within listed salmonid ESUs/DPSs in California.
DRAFT 220
-------
DRAFT
Table 34. California's 2006 Section 303(d) List of Water Quality Limited Segments:
segments listed for exceeding temperature and carbofuran limits (CEPA 2007).
Pollutant
Temperature
Carbofuran
Estuary Acres Affected
-
-
River/ Stream Miles Affected
16,907.2
49
# Water Bodies
41
1
Estuary systems of the region are consistently exposed to anthropogenic pressures
stemming from high human density sources. For example, the largest west coast estuary
is the San Francisco Estuary. This water body provides drinking water to 23 million
people, irrigates 4.5 million acres of farmland, and drains roughly 40% of California's
land area. As a result of high use, many environmental measures of the San Francisco
Estuary are poor. Water quality suffers from high phosphorus and nitrogen loads,
primarily from agricultural, sewage, and storm water runoff. Water clarity is also
compromised. Sediments from urban runoff and historical activities contain high levels
of contaminants. They include polychlorinated biphenols (PCBs), nickel, selenium,
cadmium, pesticides, mercury, copper, and silver. Specific pesticides include
pyrethroids, malathion, carbaryl, and diazinon. Other pollutants include DDT and
polynuclear aromatic hydrocarbons (PAHs).
Other wastes are also discharged into San Francisco Bay. Approximately 150 industries
discharge wastewater into the bay. Discharge of hot water from power plants and
industrial sources may elevate temperatures and negatively affect aquatic life.
Additionally, about 60 sewage treatment plants discharge treated effluent into the bay and
elevate nutrient loads. However, since 1993, many of the point sources of pollution have
been greatly reduced. Pollution from oil spills also occur due to refineries in the bay
area. As these stressors persist in the marine environment, the estuary system will likely
carry loads for future years, even with strict regulation. Gold mining has also reduced
estuary depths in much of the region, causing drastic changes to habitat.
Large urban centers are foci for contaminants. Contaminant levels in surface waters near
San Francisco, Oakland, and San Jose are highest. These areas are also where water
clarity is at its worst. Some of the most persistent contaminants (PCBs, dioxins, DDT,
DRAFT 221
-------
DRAFT
etc.) are bioaccumulated by aquatic biota and can biomagnify in the food chain. Fish
tissues contain high levels of PCB and mercury. Concentrations of PCB were 10 times
above human health guidelines for consumption. Birds, some of which are endangered
(clapper rail and least tern), have also concentrated these toxins.
Santa Ana Basin: NAWQA assessment
The Santa Ana watershed is the most heavily populated study site out of more than 50
assessment sites studied across the nation by the NAWQA Program. According to Belitz
et al. (2004), treated wastewater effluent is the primary source of baseflow to the Santa
Ana River. Secondary sources that influence peak river flows include stormwater runoff
from urban, agricultural, and undeveloped lands (Belitz, Hamlin et al. 2004). Stormwater
and agricultural runoff frequently contain pesticides, fertilizers, sediments, nutrients,
pathogenic bacteria, and other chemical pollutants to waterways and degrade water
quality. The above inputs have resulted in elevated concentrations of nitrates and
pesticides in surface waters of the basin. Nitrates and pesticides were more frequently
detected here than in other national NAWQA sites (Belitz, Hamlin et al. 2004).
Additionally, Belitz et al. (2004) found that pesticides and volatile organic compounds
(VOCs) were frequently detected in surface and ground water in the Santa Ana Basin. Of
the 103 pesticides and degradates routinely analyzed for in surface and ground water, 58
were detected. Pesticides included diuron, diazinon, carbaryl, chlorpyrifos, lindane,
malathion, and chlorothalonil. Carbaryl was detected in 42% of urban samples, though it
generally did not exceed the standard for protection of aquatic life (Belitz etal. 2004).
Carbofuran was also detected, but did not exceed any water quality standards. Methomyl
was tested for but not detected. Of the 85 VOCs routinely analyzed for, 49 were
detected. VOCs included methyl fert-butyl ether (MTBE), chloroform, and
trichloroethylene (TCE). Organochlorine compounds were also detected in bed sediment
and fish tissue. Organochlorine concentrations were also higher at urban sites than at
undeveloped sites in the Santa Ana Basin. Organochlorine compounds include DDT and
its breakdown product diphenyl dicloroethylene (DDE), and chlordane. Other
contaminants detected at high levels included trace elements such as lead, zinc, and
DRAFT 222
-------
DRAFT
arsenic. According to Belitz et al. (2004), the biological community in the basin is
heavily altered as a result from these pollutants.
San Joaquin-Tulare Basin: NAWQA assessment
A study was conducted by the USGS in the mid-1990s on water quality within the San
Joaquin-Tulare basins. USGS detected 49 of the 83 pesticides it tested for in the
mainstem and three subbasins. Pesticides were detected in all but one of the 143
samples. The most common detections were of the herbicides simazine, dacthal,
metolachlor, and EPTC (Eptam), and the insecticides diazinon and chlorpyrifos. Twenty-
two pesticides were detected in 20% of the samples (Dubrovsky, Kratzer et al. 1998).
Carbaryl and methomyl were detected in all three subbasins, despite land use differences.
Carbaryl was detected in roughly 20% of samples from each subbasin, while methomyl
detections ranged from 5% to 25%. Further, most samples contained mixtures of
between 7 and 22 pesticides. Criteria for the protection of aquatic life were exceeded in
37% of samples of streams (Dubrovsky, Kratzer et al. 1998). Only seven pesticides
exceeded their criteria: diuron, trifluralin, azinphos-methyl, carbaryl, chlorpyrifos,
diazinon, and malathion. Forty percent of these exceedances were attributed solely to
diazinon. However, criteria do not exist yet for over half of the detected compounds
(Dubrovsky etal. 1998).
Organochlorine insecticides in bed sediment and tissues offish or clams were also
detected. They include DDT and toxaphene. Levels at some sites were among the
highest in the nation. Concentrations of trace elements in bed sediment generally were
higher than concentrations found in other NAWQA study units (Dubrovsky, Kratzer et al.
1998).
Sacramento River Basin: NAWQA analysis
Another study conducted by the USGS from 1996-1998 within the Sacramento River
Basin detected up to 24 out of 47 pesticides in surface waters (Domagalski 2000).
Pesticides included thiobencarb, carbofuran, molinate, simazine, metolachlor, dacthal,
chlorpyrifos, carbaryl, and diazinon. Land use differences between sites are reflected in
DRAFT 223
-------
DRAFT
pesticide detections. Carbofuran was detected in 100% of samples from the agricultural
site, but only 6.7% of urban samples (Domagalski 2000). Carbaryl, however, was
detected in 100% of urban samples and 42.9% of agricultural samples. Some pesticides
were detected at concentrations higher than criteria for the protection of aquatic life in the
smaller streams, but were diluted to safer levels in the mainstem river. Intensive
agricultural activities also impact water chemistry. In the Salinas River and in areas with
intense agriculture use, water hardness, alkalinity, nutrients, and conductivity are also
high.
Habita t Modiflca tion
The Central Valley area, including San Francisco Bay and the Sacramento and San
Joaquin River Basins, has been drastically changed by development. Salmonid habitat
has been reduced to 300 miles from historic estimates of 6,000 miles (CDFG 1993). In
the San Joaquin Basin alone, the historic floodplain covered 1.5 million acres with 2
million acres of riparian vegetation (CDFG 1993). Roughly 5% of the Sacramento River
Basin's riparian forests remain. Impacts of development include loss of LWD, increased
bank erosion and bed scour, changes in sediment loadings, elevated stream temperature,
and decreased base flow. Thus, lower quantity and quality of LWD and modified
hydrology reduce and degrade salmonid rearing habitat.
The Klamath Basin in Northern California has been heavily modified as well. Water
diversions have reduced spring flows to 10% of historical rates in the Shasta River, and
dams block access to 22% of historical salmonid habitat. The Scott and Trinity Rivers
have similar histories. Agricultural development has reduced riparian cover and diverted
water for irrigation (NRC 2003). Riparian habitat has decreased due to extensive logging
and grazing. Dams and water diversions are also common. These physical changes
resulted in water temperatures too high to sustain salmonid populations. The Salmon
River, however, is comparatively pristine; some reaches are designated as Wild and
Scenic Rivers. The main cause of riparian loss in the Salmon River basin is likely wild
fires - the effects of which have been exacerbated by salvage logging (NRC 2003).
DRAFT 224
-------
DRAFT
Mining
Famous for the gold rush of the mid-1800s, California has a long history of mining.
Extraction methods such as suction dredging, hydraulic mining, strip mining may cause
water pollution problems. In 2004, California ranked top in the nation for non-fuel
mineral production with 8.23% of total production (NMA 2007). Today, gold, silver, and
iron ore comprise only 1% of the production value. Primary minerals include
construction sand, gravel, cement, boron, and crushed stone. California is the only state
to produce boron, rare-earth metals, and asbestos (NMA 2007).
California contains some 1,500 abandoned mines. Roughly 1% of these mines are
suspected of discharging metal-rich waters into the basins. The Iron Metal Mine in the
Sacramento Basin releases more than 1,100 Ibs of copper and more than 770 Ibs of zinc
to the Keswick Reservoir below Shasta Dam. The Iron Metal Mine also released
elevated levels of lead (Cain et al. 2000 in Carter and Resh 2005). Metal contamination
reduces the biological productivity within a basin. Metal contamination can result in fish
kills at high levels or sublethal effects at low levels. Sublethal effects include a reduction
in feeding, overall activity levels, and growth. The Sacramento Basin and the San
Francisco Bay watershed are two of the most heavily impacted basins within the state
from mining activities. The basin drains some of the most productive mineral deposits in
the region. Methylmercury contamination within San Francisco Bay, the result of 19th
century mining practices using mercury to amalgamate gold in the Sierra Nevada
Mountains, remains a persistent problem today. Based on sediment cores, pre-mining
concentrations were about five times lower than concentrations detected within San
Francisco Bay today (Conaway, Squire et al. 2003).
Hydromodification Projects
Several of the rivers within the area have been modified by dams, water diversions,
drainage systems for agriculture and drinking water, and some of the most drastic
channelization projects in the nation (see species distribution maps). In all, there are
about 1,400 dams within the State of California, more than 5,000 miles of levees, and
more than 140 aqueducts (Mount 1995). In general, the southern basins have a warmer
DRAFT 225
-------
DRAFT
and drier climate and the more northern, coastal-influenced basins are cooler and wetter.
About 75% of the runoff occurs in basins in the northern half of California, while 80% of
the water demand is in the southern half. Two water diversion projects meet these
demands—the federal Central Valley Project (CVP) and the California State Water
Project (CSWP). The CVP is one of the world's largest water storage and transport
systems. The CVP has more than 20 reservoirs and delivers about 7 million acre-ft per
year to southern California. The CSWP has 20 major reservoirs and holds nearly 6
million acre-ft of water. The CSWP delivers about 3 million acre-ft of water for human
use. Together, both diversions irrigate about 4 million acres of farmland and deliver
drinking water to roughly 22 million residents.
Both the Sacramento and San Joaquin rivers are heavily modified, each with hundreds of
dams. The Rogue, Russian, and Santa Ana rivers each have more than 50 dams, and the
Eel, Salinas, and the Klamath Rivers have between 14 and 24 dams each. The Santa
Margarita is considered one of the last free flowing rivers in coastal southern California.
Nine dams occur in this watershed. All major tributaries of the San Joaquin River are
impounded at least once and most have multiple dams or diversions. The Stanislaus
River, a tributary of the San Joaquin River, has over 40 dams. As a result, the
hydrograph of the San Joaquin River is seriously altered from its natural state. Alteration
of the temperature and sediment transport regimes had profound influences on the
biological community within the basin. These modifications generally result in a
reduction of suitable habitat for native species and frequent increases in suitable habitat
for nonnative species. The Friant Dam on the San Joaquin River is attributed with the
extirpation of spring-run Chinook salmon within the basin. A run of the spring-run
Chinook salmon once produced about 300,000 to 500,000 fish (Carter and Resh 2005).
Artificial Propagation
Anadromous fish hatcheries have existed in California since establishment of the
McCloud River hatchery in 1872. There are nine state hatcheries: the Iron Gate
(Klamath River), Mad River, Trinity (Trinity River), Feather (Feather River), Warm
Springs (Russian River), Nimbus (American River), Mokelumne (Mokulumne River),
DRAFT 226
-------
DRAFT
and Merced (Merced River). The California Department of Fish and Game (CDFG) also
manages artificial production programs on the Noyo and Eel rivers. The Coleman
National Fish Hatchery, located on Battle Creek in the upper Sacramento River, is a
federal hatchery operated by the USFWS. The USFWS also operates an artificial
propagation program for Sacramento River winter run Chinook.
Of these, the Feather River, Nimbus, Mokelumne, and Merced River facilities comprise
the Central Valley Hatcheries. Over the last ten years, the Central Valley Hatcheries
have released over 30 million young salmon. State and the federal (Coleman hatchery)
hatcheries work together to meet overall goals. State hatcheries are expected to release
18.6 million smolts in 2008 and Coleman is aiming for more than 12 million. There has
been no significant change in hatchery practices over the year that would adversely affect
the current year class offish. A new program marking 25% of the 32 million Sacramento
Fall-run Chinook smolts may provide data on hatchery fish contributions to the fisheries
in the near future.
Commercial and Recreational Fishing
The region is home to many commercial fisheries. The largest in terms of total landings
in 2006 were northern anchovy, Pacific sardine, Chinook salmon, sablefish, Dover sole,
Pacific whiting, squid, red sea urchin, and Dungeness crab (CDFG 2007). Red abalone is
also harvested. The commercial landings report does not include information on bycatch
of listed salmonids (CDFG 2007). The first salmon cannery established along the west
coast was located in the Sacramento River watershed in 1864. However, this cannery
only operated for about two years because the sediment from hydraulic mining decimated
the salmon runs in the basin (NRC 1996).
Alien Species
Plants and animals that are introduced into habitats in which they do not naturally occur
are called non-native species. They are also known as non-indigenous, exotic,
introduced, or invasive species, and have been known to affect ecosystems. Non-native
species are introduced through infested stock for aquaculture and fishery enhancement,
DRAFT 227
-------
DRAFT
through ballast water discharge and from the pet and recreational fishing industries
(http://biology.usgs.gov/s+t/noframe/xl91.htm.). The Aquatic Nuisance Species (ANS)
Task Force suggests that it is inevitable that cultured species will eventually escape
confinement and enter U.S. waterways. Non-native species were cited as a contributing
cause in the extinction of 27 species and 13 subspecies of North American fishes over the
past 100 years (Miller, Williams et al. 1989). Wilcove, Rothstein et al. (1998) note that
25% of ESA listed fish are threatened by alien species. By competing with native species
for food and habitat as well as preying on them, non-native species can reduce or
eliminate populations of native species.
Surveys performed by CDFG state that at least 607 alien species are found in California
coastal waterways (Foss et al. 2007). The majority of these species are representatives of
four phyla: annelids (33%), arthropods (22%), chordates (13%), and mollusks (10%).
Non-native chordate species are primarily fish and tunicates which inhabit fresh and
brackish water habitats such as the Sacramento-San Joaquin Delta (Foss, Ode et al.
2007). The California Aquatic Invasive Species Management Plan (CAISMP) includes
goals and strategies for reducing the introduction rate of new invasive species as well as
removing those with established populations.
Atmospheric deposition
In 2002, chlopyrifos, diazinon, trifluralin, and other pesticides were detected in air
samples collected from Sacramento, California (Majewski and Baston 2002).
Pesticide Reduction Programs
There are several measures in place in California that may reduce the levels of pesticides
found in the aquatic environment beyond FIFRA label requirements. Monitoring of
water resources is handled by California Environmental Protection Agency's Regional
Water Boards. Each Regional Board makes water quality decisions for its region
including setting standands and determining waste discharge requirements. The Central
Valley Regional Water Quality Control Board (CVRWQCB) addresses issues in the
Sacramento and San Joaquin River Basins. These river basins are characterized by crop
DRAFT 228
-------
DRAFT
land, specifically orchards, which historically rely heavily on organophosphates for pest
control.
In 2003, the CVRWQCB adopted the Irrigated Lands Waiver Program (ILWP).
Participation was required for all growers with irrigated lands that discharge waste which
may degrade water quality. However, the ILWP allowed growers to select one of three
methods for regulatory coverage (Markle, Kalman et al. 2005). These options included:
1) join a Coalition Group approved by the CVRWQCB, 2) file for an Individual
Discharger Conditional Waiver, and 3) comply with zero discharge regulation (Markle,
Kalman et al. 2005). Many growers opted to join a Coalition as the other options were
more costly. Coalition Groups were charged with completing two reports - a Watershed
Evaluation Report and a Monitoring and Reporting Plan. The Watershed Evaluation
Report had to include information on crop patterns and pesticide/nutrient use, as well as
mitigation measures that would prevent orchard run-off from impairing water quality.
Similar programs are in development in other agricultural areas of California.
As a part of the Waiver program, the Central Valley Coalitions undertook monitoring of
"agriculture dominated waterways". Some of the monitored waterways are small
agricultural streams and sloughs that carry farm drainage to larger waterways. The
coalition was also required to develop a management plan to address exceedance of State
water quality standards. Currently, the Coalitions monitor toxcity to test organisms,
stream parameters (e.g., flow, temperature, etc.), nutrient levels, and pesticides used in
the region, including diazinon and chlorpyrifos. Sampling diazinon exceedances within
the Sacramento and Feather Rivers resulted in the development of a TMDL. The
Coalitions were charged with developing and implementing management and monitoring
plans to address the TMDL and reduce diazinon run-off.
The Coalition for Urban/Rural Environmental Stewardship (CURES) is a non-profit
organization that was founded in 1997 to support educational efforts for agricultural and
urban communities focusing on the proper and judicious use of pest control products.
CURES educates growers on methods to decrease diazinon surface water contamination
DRAFT 229
-------
DRAFT
in the Sacramento River Basin. The organization has developed best-practice literature
for pesticide use in both urban and agricultural settings (www.curesworks.org). CURES
also works with California's Watershed Coalitions to standardize their Watershed
Evaluation Reports and to keep the Coalitions informed. The organization has worked
with local organizations, such as the California Dried Plum Board and the Almond Board
of California, to address concerns about diazinon, pyrethroids, and sulfur. The CURES
site discusses alternatives to oprganophosphate dormant spray applications. It lists
pyrethroids and carbaryl as alternatives, but cautions that these compounds may impact
non-target organisms. For example, carbaryl is highly toxic to honeybees, so bees must
be removed from the area prior to application
In 2006, CDPR put limitations on dormant spay application of most insecticides in
orchards, in part to adequately protect aquatic life in the Central Valley region. While the
legislation was prompted by organophosphate use, limitations also apply to pyrethroids
and carbamates.
The CDPR publishes voluntary interim measures for mitigating the potential impacts of
pesticide useage to listed species. These measures are available online as county
bulletins (http://www.cdpr.ca.gov/docs/endspec/colist.htm). Measures that apply to
carbaryl, carbofuran, and methomyl use in salmonid habitat are:
• Do not use in currently occupied habitat
• Provide a 20 ft minimum strip of vegetation (on which pesticides should not
be applied) along rivers, creeks, streams, wetlands, vernal pools and stock
ponds, or on the downhill side of fields where runoff could occur. Prepare
land around fields to contain runoff by proper leveling, etc. Contain as much
water "on-site" as possible. The planting of legumes, or other cover crops for
several rows adjacent to off-target water sites is recommended. Mix
pesticides in areas not prone to runoff such as concrete mixing/loading pads,
disked soil in flat terrain or graveled mix pads, or use a suitable method to
contain spills and/or rinsate. Properly empty and triple-rinse pesticide
containers at time of use.
• Conduct irrigations efficiently to prevent excessive loss of irrigation waters
through runoff. Schedule irrigations and pesticide applications to maximize
the interval of time between the pesticide application and the first subsequent
irrigation. Allow at least 24 hours between application of pesticides listed in
this bulletin and any irrigation that results in surface runoff into natural
DRAFT 230
-------
DRAFT
waters. Time applications to allow sprays to dry prior to rain or sprinkler
irrigations. Do not make aerial applications while irrigation water is on the
field unless surface runoff is contained for 72 hours following the application.
• For sprayable or dust formulations: when the air is calm or moving away
from habitat, commence applications on the side nearest the habitat and
proceed away from the habitat. When air currents are moving toward habitat,
do not make applications within 200 yards by air or 40 yards by ground
upwind from occupied habitat. The county agricultural commissioner may
reduce or waive buffer zones following a site inspection, if there is an
adequate hedgerow, windbreak, riparian corridor or other physical barrier that
substantially reduces the probability of drift.
Pacific Northwest Region
This region encompasses Idaho, Oregon, and Washington and includes parts of Nevada,
Montana, Wyoming, and British Columbia. In this section we discuss three major areas
that support salmonid populations within the action area. They include the Columbia
River Basin and its tributaries, the Puget Sound Region, and the coastal drainages north
of the Columbia River. Table 35, Table 36, and Table 37 show the types and areas of
land use within each salmonid ESU/DPS.
DRAFT 231
-------
DRAFT
Table 35. Area of land use categories within Chinook Salmon ESUs in km2. Land cover
image data were taken from Multi-Resolution Land Characteristics (MRLC) Consortium, a
consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA, BLM, NFS, NRCS,
and USFWS) (NLCD 2001). Land cover class definitions are available at:
http://www.mrlc.gov/nlcd definitions.php
Landcover Type
code
Open Water
Perennial
Snow/Ice
Developed,
Open Space
Developed,
Low Intensity
Developed,
Medium
Intensity
Developed,
High Intensity
Barren Land
Deciduous
Forest
Evergreen
Forest
Mixed Forest
Shrub/Scrub
Herbaceous
Hay/Pasture
Cultivated
Crops
Woody
Wetlands
Emergent
Herbaceous
Wetlands
TOTAL (inc.
open water)
TOTAL (w/o
open water)
11
12
21
22
23
24
31
41
42
43
52
71
81
82
90
95
Chinook Salmon
Lower
Columbia
River
641
12
649
517
290
118
287
551
6,497
927
1,598
520
547
278
377
223
14,031
13,390
Upper
Columbia
River
Spring Run
188
16
203
218
55
11
360
21
8,138
7
6,100
1,737
327
636
92
59
18,168
17,981
Puget
Sound
6,172
313
1,601
1,694
668
266
1,042
999
14,443
2,526
2,415
957
1,188
258
648
492
35,683
29,511
Snake
River
Fall
Run
6,172
313
1,601
1,694
668
266
1,042
999
14,443
2,526
2,415
957
1,188
258
648
492
35,683
29,511
Snake
River
Spring/
Summer
Run
253
40
328
113
30
2
500
10
27,701
4
13,618
11,053
456
3,860
96
92
58,157
57,904
Upper
Willamette
River
124
7
632
722
322
112
220
248
9,531
1,130
1,940
801
3,617
2,355
431
78
22,269
22,146
DRAFT
232
-------
DRAFT
Table 36. Area of land use categories within chum and coho ESUs in km2. Land cover
image data were taken from Multi-Resolution Land Characteristics (MRLC) Consortium, a
consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA, BLM, NFS, NRCS,
and USFWS) (NLCD 2001). Land cover class definitions are available at:
http://www.mrlc.gov/nlcd definitions.php
LandcoverType
code
Open Water
Perennial
Snow/Ice
Developed,
Open Space
Developed,
Low
Intensity
Developed,
Medium
Intensity
Developed,
High
Intensity
Barren Land
Deciduous
Forest
Evergreen
Forest
Mixed
Forest
Shrub/Scrub
Herbaceous
Hay/Pasture
Cultivated
Crops
Woody
Wetlands
Emergent
Herbaceous
Wetlands
TOTAL (inc.
open water)
TOTAL (w/o
open water)
11
12
21
22
23
24
31
41
42
43
52
71
81
82
90
95
Chum Salmon
Columbia
River
655
1
605
463
258
110
247
548
4,294
892
1,353
526
533
213
363
222
1 1 ,284
10,628
Hood
Canal
Summer
Run
704
51
134
77
20
6
166
97
2,477
200
299
133
64
2
61
56
4,548
3,843
Coho Salmon
Lower
Columbia
River
675
12
708
563
305
124
290
575
8,487
999
1,982
600
680
348
386
225
16,959
16,284
Oregon
Coast
200
0
1,107
163
49
20
467
418
14,943
4,126
3,134
1,478
860
64
263
226
27,520
27,320
DRAFT
233
-------
DRAFT
Table 37. Area of land use categories within sockeye ESUs and steelhead DPSs in km2.
Land cover image data were taken from Multi-Resolution Land Characteristics (MRLC)
Consortium, a consortium of nine federal agencies (USGS, EPA, USFS, NOAA, NASA,
BLM, NFS, NRCS, and USFWS) (NLCD 2001). Land cover class definitions are available at:
http://www.mrlc.gov/nlcd definitions.php
LandcoverType
code
Open Water
Perennial
Snow/Ice
Developed,
Open Space
Developed,
Low
Intensity
Developed,
Medium
Intensity
Developed,
High
Intensity
Barren Land
Deciduous
Forest
Evergreen
Forest
Mixed
Forest
Shrub/Scrub
Herbaceous
Hay/Pasture
Cultivated
Crops
Woody
Wetlands
Emergent
Herbaceous
Wetlands
TOTAL (inc.
open water)
TOTAL (w/o
open water)
11
12
21
22
23
24
31
41
42
43
52
71
81
82
90
95
Sockeye
Salmon
Ozette
Lake
30
0
1
0
0
0
2
3
158
3
14
8
0
0
8
1
228
199
Snake
River
19
18
3
2
0
0
9
0
755
0
185
269
12
1
16
34
1,323
1,304
Steelhead
Lower
Columbia
River
250
12
518
506
287
116
174
382
7,023
611
1,589
398
605
322
244
93
13,128
12,878
Middle
Columbia
River
575
13
1,276
627
192
25
183
54
18,347
41
32,089
2,752
863
11,908
217
291
69,453
68,878
Puget
Sound
6,172
313
1,601
1,694
668
266
1,042
999
14,443
2,526
2,415
957
1,188
258
648
492
35,683
29,511
Snake
River
285
42
515
144
40
3
504
35
39,556
17
15,644
12,361
463
6,227
116
111
76,061
75,777
Upper
Columbia
River
359
16
343
294
80
13
361
25
8,223
7
9,351
1,823
448
3,236
109
81
24,771
24,411
Upper
Willamette
River
62
0
382
513
231
75
77
171
4,133
791
994
519
2,529
1,844
292
43
12,655
12,593
DRAFT
234
-------
DRAFT
Columbia River Basin
The most notable basin within the region is the Columbia River. The Columbia River is
the largest river in the Pacific Northwest and the fourth largest river in terms of average
discharge in the U.S. The Columbia River drains over 258,000 square miles, and is the
sixth largest in terms of drainage area. Major tributaries include the Snake, Willamette,
Salmon, Flathead, and Yakima rivers. Smaller rivers include the Owyhee, Grande
Ronde, Clearwater, Spokane, Methow, Cowlitz, and the John Day Rivers (see Table 38
for a description of select Columbia River tributaries). The Snake River is the largest
tributary at more than 1,000 miles long. The headwaters of the Snake River originate in
Yellowstone National Park, Wyoming. The second largest tributary is the Willamette
River in Oregon (Kammerer 1990; Hinck, Schmitt et al. 2004). The Willamette River is
also the 19th largest river in the nation in terms of average annual discharge (Kammerer
1990). The basins drain portions of the Rocky Mountains, Bitteroot Range, and the
Cascade Range.
Table 38. Select tributaries of the Columbia River (Carter and Resh 2005)
Watershed
Snake/Salmon
rivers
Yakima River
Willamette River
Approx
Length
(mi)
870
214
143
Basin
Size (mi2)
108,495
6,139
11,478
Physiographic
Provinces*
CU, NR, MR,
B/R
CS, CU
CS, PB
Mean
Annual
Precipitation
(in)
14
7
60
Mean
Discharge
(cfs)
55,267
3,602
32,384
No.
Fish
Species
(native)
39(19)
50
61
(~31)
No. Endangered
Species
5 fish (4 T, 1 E), 6
(1 T, 5 E) snails,
1 plant (T)
2 fish (T)
5 fish (4 T, 1 E),
* Physiographic Provinces: CU = Columbia-Snake River Plateaus, NR = Northern Rocky
Mountains, MR = Middle Rocky Mountains, B/R = Basin & Range, CS = Cascade-Sierra
Mountains, PB = Pacific Border
The Columbia river and estuary were once home to more than 200 distinct runs of Pacific
salmon and steelhead with unique adaptations to local environments within a tributary
(Stanford, Hauer et al. 2005). Salmonids within the basin include Chinook salmon, chum
salmon, coho salmon, sockeye salmon, steelhead, redband trout, bull trout, and cutthroat
trout.
DRAFT
235
-------
DRAFT
Land Use
More than 50% of the U.S. portion of the Columbia River Basin is in federal ownership
(most of which occurs in high desert and mountain areas). Approximately 39% is in
private land ownership (most of which occurs in river valleys and plateaus). The
remaining 11% is divided among the tribes, state, and local governments (Hinck, Schmitt
et al. 2004). See Table 39 for a summary of land uses and population densities in several
subbasins within the Columbia River watershed (data from (Stanford, Hauer et al. 2005).
Table 39. Land use and population density in select tributaries of the Columbia River
(Stanford, Hauer et al. 2005).
Watershed
Snake/Salmon rivers
Yakima River
Willamette River
Land Use Categories (Percent)
Agriculture
30
16
19
Forest
10-15
36
68
Urban
1
1
5
Other
54
scrub/rangeland/barren
47 shrub
-
Density
(people/mi2)
39
80
171
The interior Columbia Basin has been altered substantially by humans causing dramatic
changes and declines in native fish populations. In general, the basin supports a variety
of mixed uses. Predominant human uses include logging, agriculture, ranching,
hydroelectric power generation, mining, fishing, a variety of recreational activities, and
urban uses. The decline of salmon runs in the Columbia River is attributed to loss of
habitat, blocked migratory corridors, altered river flows, pollution, overharvest, and
competition from hatchery fish. In the Yakima River, 72 stream and river segments are
listed as impaired by the Washington Department of Ecology (DOE) and 83% exceed
temperature standards. In the Willamette River, riparian vegetation was greatly reduced
by land conversion. By 1990, only 37% of the riparian area within 120 m was forested,
30% was agricultural fields, and 16% was urban or suburban lands. In the Yakima River,
non-native grasses and other plants are commonly found along the lower reaches of the
river (Stanford, Hauer et al. 2005).
Agriculture and Ranching
Agriculture, ranching, and related services in the Pacific Northwest employ more than
nine times the national average [19% of the households within the basin (NRC 2004)].
DRAFT
236
-------
DRAFT
Ranching practices have led to increased soil erosion and sediment loads within adjacent
tributaries. The worst of these effects may have occurred in the late 1800s and early
1900s from deliberate burning to increase grass production (NRC 2004). Several
measures are currently in place to reduce the impacts of grazing. Measures include
restricted grazing in degraded areas, reduced grazing allotments, and lowered stocking
rates. Today, the agricultural industry impacts water quality within the basin.
Agriculture is second only to the large-scale influences of hydromodification projects
regarding power generation and irrigation. Water quality impacts from agricultural
activities include alteration of the natural temperature regime, insecticide and herbicide
contamination, and increased suspended sediments.
Roughly 6% of the annual flow from the Columbia River is diverted for the irrigation of
7.3 million acres of croplands within the basin. The vast majority of these agricultural
lands are located along the lower Columbia River, the Willamette, Yakima, Hood, and
Snake rivers, and the Columbia Plateau (Hinck, Schmitt et al. 2004).
Agriculture and ranching increased steadily within the Columbia River basin from the
mid- to late-1800s. By the early 1900s, agricultural opportunities began increasing at a
much more rapid pace with the creation of more irrigation canals and the passage of the
Reclamation Act of 1902 (NRC 2004). Today, agriculture represents the largest water
user within the basin (>90%).
The USGS has a number of fixed water quality sampling sites throughout various
tributaries of the Columbia River. Many of the water quality sampling sites have been in
place for decades. Water volumes, crop rotation patterns, croptype, and basin location
are some of the variables that influence the distribution and frequency of pesticides
within a tributary. Detection frequencies for a particular pesticide can vary widely. One
study conducted by the USGS between May 1999 and January 2000 in the surface waters
of Yakima Basin detected 25 pesticide compounds (Ebbert and Embry 2001) Atrazine
was the most widely detected herbicide and azinphos-methyl was the most widely
detected insecticide. Other detected compounds include simazine, terbacil, trifluralin;
DRAFT 237
-------
DRAFT
deethylatrazine, carbaryl, diazinon, malathion, and DDE. In addition to current use-
chemicals legacy chemicals continue to pose a serious problem to water quality and fish
communities despite their ban in the 1970s and 1980s (Hinck, Schmitt et al. 2004).
Fish and macroinvertebrate communities exhibit an almost linear decline in condition as
the level of agriculture intensity increases within a basin (Cuffney, Meador et al. 1997;
Fuhrer, Morace et al. 2004). A study conducted in the late 1990s examined 11 species of
fish, including anadromous and resident fish collected throughout the basin, for a suite of
132 contaminants. They included 51 semi-volatile chemicals, 26 pesticides, 18 metals, 7
PCBs, 20 dioxins, and 10 furans. Sampled fish tissues revealed PCBs, metals,
chlorinated dioxins and furans (products of wood pulp bleaching operations), and other
contaminants.
Yakima River Basin: NAWQA analysis
The Yakima River Basin is one of the most agriculturally productive areas in the U.S.
(Fuhrer, Morace et al. 2004). Croplands within the Yakima Basin account for about 16%
of the total basin area of which 77% is irrigated. The extensive irrigation-water delivery
and drainage system in the Yakima River Basin greatly controls water quality conditions
and aquatic health in agricultural streams, drains, and the Yakima River (Fuhrer, Morace
et al. 2004). From 1999 to 2000, the USGS conducted a NAWQA study in the Yakima
River Basin. Fuhrer et al. (2004) reported that nitrate and orthophosphate were the
dominant forms of nitrogen and phosphorus found in the Yakima River and its
agricultural tributaries. Arsenic, a known human carcinogen, was also detected in
agricultural drains at elevated concentrations during the nonirrigation season when
ground water is the primary source of streamflow.
The USGS also detected 76 pesticide compounds in the Yakima River Basin. They
include 38 herbicides (including metribuzin), 17 insecticides (such as carbaryl, diazinon,
and malathion), 15 breakdown products, and 6 others. Ninety-one percent of the samples
collected from the small agricultural watersheds contained at least two pesticides or
pesticide breakdown products. Carbaryl was detected in 29% of tributary samples and
DRAFT 238
-------
DRAFT
17% of mainstem Yakima River samples by a screening level 21 nanograms/liter (Fuhrer
et al. 2004). Carbofuran was screened for, but not detected. The assessment did not
screen for methomyl. The median and maximum number of chemicals in a mixture was
8 and 26, respectively (Fuhrer, Morace et al. 2004). The herbicide 2,4-D, occurred most
often in the mixtures, along with azinphos-methyl, the most heavily applied pesticide,
and atrazine, one of the most aquatic mobile pesticides (Fuhrer, Morace et al. 2004).
However, the most frequently detected pesticides in the Yakima River Basin are total
DDT, and its breakdown products DDE, dichloro diphenyl dichloroethane (ODD), and
dieldrin (Johnson and Newman 1983; Joy 2002; Fuhrer, Morace et al. 2004).
Nevertheless, concentrations of total DDT in water have decreased since 1991. These
reductions are attributed to erosion-controlling best management practices (BMPs).
Williamette Basin: NAWQA analysis
From 1991 to 1995, the USGS also sampled surface waters in the Willamette Basin,
Oregon. Wentz et al. (1998) reported that, of the 86 tested for, 50 pesticides and
pesticide degradates were detected in streams. Ten of the pesticides exceeded criteria
established by the EPA for the protection of freshwater aquatic life from chronic toxicity.
Carbaryl exceeded protective criteria in 17 of its 46 detections, while carbofuran
exceeded limits in three of 51 detections (Wentz et al. 1998). Atrazine, simazine,
metolachlor, deethylatrazine, diuron, and diazinon were detected in more than one-half of
stream samples. Methomyl was tested for but not detected. Forty-nine pesticides were
detected in streams draining predominantly agricultural land. About 25 pesticides were
detected in streams draining mostly urban areas. The highest pesticide concentrations
generally occurred in streams draining predominately agricultural land.
Snake River Basin: NAWQA assessment
The USGS conducted a water quality study from 1992-1995 in the upper Snake River
basin, Idaho and Wyoming (Clark et al. 1998). In basin wide stream sampling in May
and June 1994, Eptam [EPTC] (used on potatoes, beans, and sugar beets), atrazine and its
breakdown product desethylatrazine (used on corn), metolachlor (used on potatoes and
beans), and alachlor (used on beans and corn) were the most commonly detected
DRAFT 239
-------
DRAFT
pesticides. These same compounds accounted for 75% of all detections. Seventeen
different pesticides were detected downstream from American Falls Reservoir. Carbaryl
and carbofuran were each detected in only 1% of samples; methomyl was screened for
but not detected (Clark et al. 1998).
Hood River Basin
The Hood River Basin ranks fourth in the state of Oregon in total agricultural pesticide
usage (Jenkins, Jepson et al. 2004). The land in Hood River basin is used to grow five
crops: alfalfa, apples, cherries, grapes, and pears. About 61 a.i.s, totaling 1.1 million Ibs,
are applied annually to roughly 21,000 acres. Of the top nine, three are carbamates and
three are organophosphate insecticides (Table 40). These compounds will have a similar
mode of action, though different toxicities, as carbaryl, carbofuran, and methomyl.
Table 40. Amount of most common a.i.s applied to crops in Hood River Basin 1990-1996
(Jenkins et al. 2004).
Active Ingredient
Oil
Lime Sulfur
Mancozeb
Sulfur
Ziram
Azinphos-methyl
Metam-Sodium
Phosmet
Chlorpyrifos
Class
-
-
Carbamate
-
Carbamate
Organo-phosphate
Carbamate
Organo-phosphate
Organo-phosphate
Lbs applied
624,392
121,703
86,872
60,552
45,965
22,294
17,114
15,919
14,833
The Hood River basin contains approximately 400 miles of perennial stream channel, of
which an estimated 100 miles is accessible to anadromous fish. These channels are
important rearing and spawning habitat for salmonids, making pesticide drift a major
concern for the area.
Central Columbia Plateau: NAWQA Assessment
The USGS sampled 31 surface-water sites representing agricultural land use, with
different crops, irrigation methods, and other agricultural practices for pesticides in Idaho
and Washington from 1992-1995 (Williamson, Munn et al. 1998). Pesticides were
detected in samples from all sites, except for the Palouse River at Laird Park (a
DRAFT
240
-------
DRAFT
headwaters site in a forested area). Many pesticides were detected in surface water at
very low concentrations. Concentrations of six pesticides exceeded freshwater-chronic
criteria for the protection of aquatic life in one or more surface-water samples. They
include the herbicide triallate and five insecticides (azinphos-methyl, chlorpyrifos,
diazinon, gamma-HCH, and parathion). Carbaryl and carbofuran were detected in 6%
and 5% of samples, respectively. Methomyl was screened for, but not detected in any
samples (Williamson et al. 1998).
Detections at four sites were high, ranging from 12 to 45 pesticides. The two sites with
the highest detection frequencies are in the Quincy-Pasco subunit, where irrigation and
high chemical use combine to increase transport of pesticides to surface waters. Pesticide
detection frequencies at sites in the dryland farming (non-irrigated) areas of the North-
Central and Palouse subunits are below the national median for NAWQA sites. All four
of the sites had at least one pesticide concentration that exceeded a water-quality standard
or guideline.
Concentrations of organochlorine pesticides and PCBs are higher than the national
median (50th percentile) at seven of 11 sites; four sites were in the upper 25% of all
NAWQA sites. Although most of these compounds have been banned, they still persist
in the environment. Elevated concentrations were observed in dryland farming areas as
well as in irrigated areas.
Urban and Industrial Development
The largest urban area in the basin is the greater Portland metropolitan area, located at the
mouth of the Willamette River. Portland's population exceeds 500,000 (Hinck, Schmitt
et al. 2004). Although the basin's land cover is about 8% of the U.S. total land mass, its
human population is one-third the national average (about 1.2% of the U.S. population)
(Hinck, Schmitt et al. 2004).
Discharges from sewage treatment plants, paper manufacturing, and chemical and metal
production represent the top three permitted sources of contaminants within the lower
DRAFT 241
-------
DRAFT
basin according to discharge volumes and concentrations (Rosetta and Borys 1996).
Rosetta and Borys (1996) review of 1993 data indicate that 52% of the point source waste
water discharge volume is from sewage treatment plants, 39% from paper and allied
products, 5% from chemical and allied products, and 3% from primary metals. However,
the paper and allied products industry are the primary sources of the suspended sediment
load (71%). Additionally, 26% comes from sewage treatment plants and 1% is from the
chemical and allied products industry. Nonpoint source discharges (urban stormwater
runoff) account for significant pollutant loading to the lower basin, including most
organics and over half of the metals. Although rural nonpoint sources contributions were
not calculated, Rosetta and Borys (1996) surmised that in some areas and for some
contaminants, rural areas may contribute a large portion of the load. This is particularly
true for pesticide contamination in the upper river basin where agriculture is the
predominant land use.
Water quality has been reduced by phosphorus loads and decreased water clarity,
primarily along the lower and middle sections of the Columbia River Estuary. Although
sediment quality is generally very good, benthic indices have not been established within
the estuary. Fish tissue contaminant loads (PCBs, DDT, ODD, DDE, and mercury) are
high and present a persistent and long lasting effect on estuary biology. Health advisories
have been recently issued for people eating fish in the area that contain high levels of
dioxins, PCBs, and pesticides.
Habita t Modified tion
Basin wide, critical ecological connectivity (mainstem to tributaries and riparian
floodplains) has been disconnected by dams and associated activities such as floodplain
deforestation and urbanization. Dams have flooded historical spawning and rearing
habitat with the creation of massive water storage reservoirs. More than 55% of the
Columbia River Basin that was accessible to salmon and steelhead before 1939 has been
blocked by large dams (NWPPC 1986). Construction of the Grand Coulee Dam blocked
1,000 miles (1,609 km) of habitat from migrating salmon and steelhead (Wydoski and
Whitney 1979). Similarly, over one third ( 2,000 km) of coho salmon habitat is no longer
DRAFT 242
-------
DRAFT
accessible (Good, Waples et al. 2005). The mainstem habitats of the lower Columbia and
Willamette rivers have been reduced primarily to a single channel. As a result, floodplain
area is reduced, off-channel habitat features have been eliminated or disconnected from
the main channel, and the amount of LWD in the mainstem has been reduced.
Remaining areas are affected by flow fluctuations associated with reservoir management
for power generation, flood control, and irrigation. Overbank flow events, important to
habitat diversity, have become rare as a result of controlling peak flows and associated
revetments. Portions of the basin are also subject to impacts from cattle grazing and
irrigation withdrawals. Consequently, estuary dynamics have changed substantially.
Stream habitat degradation in Columbia Central Plateau is relatively high (Williamson et
al. 1998). In the most recent NAWQA survey, a total of 16 sites were evaluated - all of
which showed signs of degradation (Williamson et al. 1998). Streams in this area have
an average of 20% canopy cover and 70% bank erosion. These factors have severely
affected the quality of habitat available to salmonids. The Palouse subunit of the Lower
Snake River exceeds temperature levels for the protection of aquatic life (Williamson et
al. 1998).
Habitat loss has fragmented habitat and human density increase has created additional
loads of pollutants and contaminants within the Columbia River Estuary (Anderson,
Dugger et al. 2007). About 77% of swamps, 57% of marshes, and over 20% of tree cover
have been lost to development and industry. Twenty four threatened and endangered
species occur in the estuary, some of which are recovering and others (i.e., Chinook
salmon) are not.
The Willamette Basin Valley has been dramatically changed by modern settlement. The
complexity of the mainstem river and extent of riparian forest have both been reduced by
80% (PNERC 2002). About 75% of what was formerly prairie and 60% of what was
wetland has been converted to agricultural purposes. These actions, combined with urban
development, extensive (96 miles) bank stabilization, and in-river and near-shore gravel
DRAFT 243
-------
DRAFT
mining, have resulted in a loss of floodplain connectivity and off-channel habitat
(PNERC 2002).
Habita t Restora tion
Since 2000, land management practices included improving access by replacing culverts
and fish habitat restoration activities at Federal Energy Regulatory Commission (FERC)-
licensed dams. Habitat restoration in the upper (reducing excess sediment loads) and
lower Gray's River watersheds may benefit the Gray's River chum salmon population as
it has a subyearling juvenile life history type and rears in such habitats. Short-term daily
flow fluctuations at Bonneville Dam sometimes create a barrier (i.e., entrapment on
shallow sand flats) for fry moving into the mainstem rearing and migration corridor.
Some chum fry have been stranded on shallow water flats on Pierce Island from daily
flow fluctuations. Coho salmon are likely to be affected by flow and sediment delivery
changes in the Columbia River plume. Steelhead may be affected by flow and sediment
delivery changes in the plume (Casillas 1999).
In 2006, NOAA Fisheries completed consultation on issuance of a 50-year incidental take
permit to the State of Washington for its Washington State Forest Practices Habitat
Conservation Plan (HCP). The HCP is expected to improve habitat conditions on state
forest lands within the action area. Improvements include removing barriers to
migration, restoring hydrologic processes, increasing the number of large trees in riparian
zones, improving stream bank integrity, and reducing fine sediment inputs (FCRPS
2008).
Mining
Most of the mining in the basin is focused on minerals such as phosphate, limestone,
dolomite, perlite, or metals such as gold, silver, copper, iron, and zinc. Mining in the
region is conducted in a variety of methods and places within the basin. Alluvial or
glacial deposits are often mined for gold or aggregate. Ores are often excavated from the
hard bedrocks of the Idaho batholiths. Eleven percent of the nation's output of gold has
come from mining operations in Washington, Montana, and Idaho. More than half of the
DRAFT 244
-------
DRAFT
nation's silver output has come from a few select silver deposits.
Many of the streams and river reaches in the basin are impaired from mining. Several
abandoned and former mining sites are also designated as superfund cleanup areas
(Stanford, Hauer et al. 2005; Anderson, Dugger et al. 2007). According to the U.S.
Bureau of Mines, there are about 14,000 inactive or abandoned mines within the
Columbia River Basin. Of these, nearly 200 pose a potential hazard to the environment
(Quigley, Arbelbide et al. 1997 in Hincke et al. 2004). Contaminants detected in the
water include lead and other trace metals.
Hydromodification Projects
More than 400 dams exist in the basin, ranging from mega dams that store large amounts
of water to small diversion dams for irrigation. Every major tributary of the Columbia
River except the Salmon River is totally or partially regulated by dams and diversions.
More than 150 dams are major hydroelectric projects. Of these, 18 dams are located on
the mainstem Columbia River and its major tributary, the Snake River. The FCRPS
encompasses the operations of 14 major dams and reservoirs on the Columbia and Snake
rivers. These dams and reservoirs operate as a coordinated system. The Corps operates 9
of 10 major federal projects on the Columbia and Snake rivers, and the Dworshak, Libby
and Albeni Falls dams. The BOR operates the Grand Coulee and Hungry Horse dams.
These federal projects are a major source of power in the region. These same projects
provide flood control, navigation, recreation, fish and wildlife, municipal and industrial
water supply, and irrigation benefits.
BOR has operated irrigation projects within the basin since 1904. The irrigation system
delivers water to about 2.9 million acres of agricultural lands. About 1.1 million acres of
land are irrigated using water delivered by two structures, the Columbia River Project
(Grand Coulee Dam) and the Yakima Project. The Grand Coulee Dam delivers water for
the irrigation of over 670,000 acres of croplands and the Yakima Project delivers water to
nearly 500,000 acres of croplands (Bouldin, Farris et al. 2007).
DRAFT 245
-------
DRAFT
The Bonneville Power Administration (BPA), an agency of the U.S. Department of
Energy, wholesales electric power produced at 31 federal dams (67% of its production)
and non-hydropower facilities in the Columbia-Snake Basin. The BPA sells about half
the electric power consumed in the Pacific Northwest. The federal dams were developed
over a 37-year period starting in 1938 with Bonneville Dam and Grand Coulee in 1941,
and ending with construction of Libby Dam in 1973 and Lower Granite Dam in 1975.
Development of the Pacific Northwest regional hydroelectric power system, dating to the
early 20th century, has had profound effects on the ecosystems of the Columbia River
Basin (ISG 1996). These effects have been especially adverse to the survival of
anadromous salmonids. The construction of the FCRPS modified migratory habitat of
adult and juvenile salmonids. In many cases, the FCRPS presented a complete barrier to
habitat access for salmonids. Approximately 80% of historical spawning and rearing
habitat of Snake River fall-run Chinook salmon is now inaccessible due to dams. The
Snake River spring/summer run has been limited to the Salmon, Grande Ronde, Imnaha,
and Tuscanon rivers. Damming has cut off access to the majority of Snake River
Chinook salmon spawning habitat. The Sunbeam Dam on the Salmon River is believed
to have limited the range of Snake River sockeye salmon as well.
Both upstream and downstream migrating fish are impeded by the dams. Additionally, a
substantial number of juvenile salmonids are killed and injured during downstream
migrations. Physical injury and direct mortality occurs as juveniles pass through
turbines, bypasses, and spillways. Indirect effects of passage through all routes may
include disorientation, stress, delays in passage, exposure to high concentrations of
dissolved gases, warm water, and increased predation. Non-federal hydropower facilities
on Columbia River tributaries have also partially or completely blocked higher elevation
spawning.
Qualitatively, several hydromodification projects have improved the productivity of
naturally produced Snake River fall Chinook salmon. Improvements include flow
augmentation to enhance water flows through the lower Snake and Columbia Rivers
DRAFT 246
-------
DRAFT
(USER 1998 in (FCRPS 2008); providing stable outflows at Hells Canyon Dam during
the fall Chinook salmon spawning season and maintaining these flows as minimums
throughout the incubation period to enhance survival of incubating fall-run Chinook
salmon; and reduced summer temperatures and enhanced summer flow in the lower
Snake River (see Corps et al. 2007b, Appendix 1 in (FCRPS 2008). Providing suitable
water temperatures for over-summer rearing within the Snake River reservoirs allows the
expression of productive "yearling" life history strategy that was previously unavailable
to Snake River fall-run Chinook salmon.
The mainstem FCRPS corridor has also improved safe passage through the hydrosystem
for juvenile steelhead and yearling Chinook salmon with the construction and operation
of surface bypass routes at Lower Granite, Ice Harbor, and Bonneville dams and other
configuration improvements (Corps et al. 2007a)
For salmon, with a stream-type juvenile life history, projects that have protected or
restored riparian areas and breached or lowered dikes and levees in the tidally influenced
zone of the estuary have improved the function of the juvenile migration corridor. The
FCRPS action agencies recently implemented 18 estuary habitat projects that removed
passage barriers. These activities provide fish access to good quality habitat.
The Corps et al. (2007) estimated that hydropower configuration and operational
improvements implemented from 2000 to 2006 have resulted in an 11.3% increase in
survival for yearling juvenile LCR Chinook salmon from populations that pass
Bonneville Dam. Improvements during this period included the installation of a corner
collector at Powerhouse II (PH2) and the partial installation of minimum gap runners at
Powerhouse 1 (PHI) and of structures that improve fish guidance efficiency at PH2.
Spill operations have been improved and PH2 is used as the first priority powerhouse for
power production because bypass survival is higher than at PHI. Additionally, drawing
water towards PH2 moves fish toward the corner collector. The bypass system screen
was removed from PHI because tests showed that turbine survival was higher than
through the bypass system at that location.
DRAFT 247
-------
DRAFT
Artificial Propagation
There are several artificial propagation programs for salmon production within the
Columbia River Basin. These programs were instituted under federal law to lessen the
effects of lost natural salmon production within the basin from the dams. The hatcheries
are operated by federal, state, and tribal managers. For more than 100 years, hatcheries
in the Pacific Northwest have been used to produce fish for harvest and replace natural
production lost to dam construction. Hatcheries have only minimally been used to
protect and rebuild naturally produced salmonid population (e.g., Redfish Lake sockeye
salmon). In 1987, 95% of the coho salmon, 70% of the spring Chinook salmon, 80% of
the summer Chinook salmon, 50% of the fall-run Chinook salmon, and 70% of the
steelhead returning to the Columbia River Basin originated in hatcheries (CBFWA 1990).
More recent estimates suggest that almost half of the total number of smolts produced in
the basin come from hatcheries (Beechie, Liermann et al. 2005).
The impact of artificial propagation on the total production of Pacific salmon and
steelhead has been extensive (Hard, Jones et al. 1992). Hatchery practices, among other
factors, are a contributing factor to the 90% reduction in natural coho salmon runs in the
lower Columbia River over the past 30 years (Flagg, Waknitz et al. 1995). Past hatchery
and stocking practices have resulted in the transplantation of salmon and steelhead from
non-native basins. The impacts of these hatchery practices are largely unknown.
Adverse effects of these practices likely included: loss of genetic variability within and
among populations (Busack 1990; Riggs 1990; Hard, Jones et al. 1992; Reisenbichler
1997), disease transfer, increased competition for food, habitat, or mates, increased
predation, altered migration, and the displacement of natural fish (Steward and Bjornn
1990; Hard, Jones et al. 1992; Fresh 1997). Species with extended freshwater residence
may face higher risk of domestication, predation, or altered migration than species that
spend only a brief time in freshwater (Hard, Jones et al. 1992). Nonetheless, artificial
propagation may also contribute to the conservation of listed salmon and steelhead.
However, it is unclear whether or how much artificial propagation during the recovery
process will compromise the distinctiveness of natural populations (Hard, Jones et al.
1992).
DRAFT 248
-------
DRAFT
The States of Oregon and Wasington and other fisheries co-managers are engaged in a
substantial review of hatchery management practices through the Hatchery Scientific
Review Group (HSRG). The HSRG was established and funded by Congress to provide
an independent review of current hatchery program in the Columbia River Basin. The
HSRG has completed its work on LCR populations and provided its recommendations.
A general conclusion is that the current production programs are inconsistent with
practices that reduce impacts on naturally-spawning populations, and will have to be
modified to reduce adverse effects on key natural populations identified in the Interim
Recovery Plan. The adverse effects are caused by hatchery-origin adults spawning with
natural-origin fish or competing with natural-origin fish for spawning sites (FCRPS
2008). Oregon and Washington initiated a comprehensive program of hatchery and
associated harvest reforms (WDFW 2005; ODFW 2007). The program is designed to
achieve HSRG objectives related to controlling the number of hatchery-origin fish on the
spawning grounds and in the hatchery broodstock.
Coho salmon hatchery programs in the lower Columbia have been tasked to compensate
for impacts of fisheries. However, hatchery programs in the LCR have not operated
specifically to conserve LCR coho salmon. These programs threaten the viability of
natural populations. The long-term domestication of hatchery fish has eroded the fitness
of these fish in the wild and has reduced the productivity of wild stocks where significant
numbers of hatchery fish spawn with wild fish. Large numbers of hatchery fish have also
contributed to more intensive mixed stock fisheries. These programs largely
overexploited wild populations weakened by habitat degradation. Most LCR coho
salmon populations have been heavily influenced by hatchery production over the years.
Commercial Recreational, and Subsistence Fishing
Archeological records indicate that indigenous people caught salmon in the Columbia
River more than 7,000 years ago. One of the most well known tribal fishing sites within
the basin was located near Celilo Falls, an area in the lower river that has been occupied
by Dalles Dam since 1957. Salmon fishing increased with better fishing methods and
DRAFT 249
-------
DRAFT
preservation techniques, such as drying and smoking. Salmon harvest substantially
increased in the mid-1800s with canning techniques. Harvest techniques also changed
over time, from early use of hand-held spears and dip nets, to riverboats using seines and
gill-nets. Harvest techniques eventually transitioned to large ocean-going vessels with
trolling gear and nets and the harvest of Columbia River salmon and steelhead from
California to Alaska (Beechie, Liermann et al. 2005).
During the mid-1800s, an estimated 10 to 16 million adult salmon of all species entered
the Columbia River each year. Large annual harvests of returning adult salmon during
the late 1800s ranging from 20 million to 40 million Ibs of salmon and steelhead
significantly reduced population productivity (Beechie, Liermann et al. 2005). The
largest known harvest of Chinook salmon occurred in 1883 when Columbia River
canneries processed 43 million Ibs of salmon (Lichatowich 1999). Commercial landings
declined steadily from the 1920s to a low in 1993. At that time, just over one million Ibs
of Chinook salmon were harvested (Beechie, Liermann et al. 2005).
Harvested and spawning adults reached 2.8 million in the early 2000s, of which almost
half are hatchery produced (Beechie, Liermann et al. 2005). Most of the fish caught in
the river are steelhead and spring/summer Chinook salmon. Ocean harvest consists
largely of coho and fall Chinook salmon. Most ocean catches are made north of Cape
Falcon, Oregon. Over the past five years, the number of spring and fall salmon
commercially harvested in tribal fisheries has averaged between 25,000 and 110,000 fish
(Beechie, Liermann et al. 2005). Recreational catch in both ocean and in-river fisheries
varies from 140,000 to 150,000 individuals (Beechie, Liermann et al. 2005).
Non-Indian fisheries in the lower Columbia River are limited to a harvest rate of 1%.
Treaty Indian fisheries are limited to a harvest rate of 5 to 7%, depending on the run size
of upriver Snake River sockeye stocks. Actual harvest rates over the last 10 years have
ranged from 0 to 0.9%, and 2.8 to 6.1%, respectively (see TAC 2008, Table 15 in FCRPS
(2008).
DRAFT 250
-------
DRAFT
Columbia River chum salmon are not caught incidentally in tribal fisheries above
Bonneville Dam. However, Columbia River chum salmon are incidentally caught
occasionally in non-Indian fall season fisheries below Bonneville Dam. There are no
fisheries in the Columbia River that target hatchery or natural-origin chum salmon. The
species' later fall return timing make them vulnerable to relatively little potential harvest
in fisheries that target Chinook salmon and coho salmon. Columbia River chum salmon
rarely take the sport gear used to target other species. Incidental catch of chum amounts
to a few tens of fish per year (TAG 2008). The harvest rate of Columbia River chum
salmon in proposed state fisheries in the lower river is estimated to be 1.6% per year and
is less than 5%.
LCR coho salmon are harvested in the ocean and in the Columbia River and tributary
freshwater fisheries of Oregon and Washington. Incidental take of coho salmon prior to
the 1990s fluctuated from approximately 60 to 90%. However, this number has been
reduced since its listing to 15 to 25% (LCFRB 2004). The exploitation of hatchery coho
salmon has remained approximately 50% through the use of selective fisheries.
LCR steelhead are harvested in Columbia River and tributary freshwater fisheries of
Oregon and Washington. Fishery impacts of LCR steelhead have been limited to less
than 10% since implementation of mark-selective fisheries during the 1980s. Recent
harvest rates on UCR steelhead in non-Treaty and treaty Indian fisheries ranged from 1%
to 2%, and 4.1% to 12.4%, respectively (FCRPS 2008).
Alien Species
Many non-native species have been introduced to the Columbia River Basin since the
1880s. At least 81 invasive species have currently been identified, composing one-fifth
of all species in some areas. New non-native species are discovered in the basin
regularly; a new aquatic invertebrate is discovered approximately every 5 months
(Sytsma, Cordell et al. 2004). It is clear that the introduction of non-native species has
changed the environment, though whether these changes will impact salmonid
populations is uncertain (Sytsma, Cordell et al. 2004).
DRAFT 251
-------
DRAFT
Puget Sound Region
Puget Sound is the second largest estuary in the U.S. It has about 1,330 miles of
shoreline and extends from the mouth of the Strait of Juan de Fuca east. Puget Sound
includes the San Juan Islands and south to Olympia, and is fed by more than 10,000
rivers and streams.
Puget Sound is generally divided into four major geographic marine basins: Hood Canal,
South Sound, Whidbey Basin, and the Main Basin. The Main Basin has been further
subdivided into two subbasins: Admiralty Inlet and Central Basin. About 43% of the
Puget Sound's tideland is located in the Whidbey Island Basin. This reflects the large
influence of the Skagit River, which is the largest river in the Puget Sound system and
whose sediments are responsible for the extensive mudflats and tidelands of Skagit Bay.
Habitat types that occur within the nearshore environment include eelgrass meadows,
kelp forest, mud flats, tidal marshes, sub-estuaries (tidally influenced portions of river
and stream mouths), sand spits, beaches and backshore, banks and bluffs, and marine
riparian vegetation. These habitats provide critical functions such as primary food
production and support habitat for invertebrates, fish, birds, and other wildlife.
Major rivers draining to Puget Sound from the Cascade Mountains include the Skagit,
Snohomish, Nooksack, Puyallup, and Green rivers, as well as the Lake
Washington/Cedar River watershed. Major rivers from the Olympic Mountains include
the Hamma Hamma, the Duckabush, the Quilcene, and the Skokomish rivers. Numerous
other smaller rivers drain to the Sound, many of which are significant salmonid
production areas despite their small size.
The Puget Sound basin is home to more than 200 fish and 140 mammalian species.
Salmonids within the region include coho, Chinook, sockeye, chum, and pink salmon,
kokanee, steelhead, rainbow, cutthroat, and bull trout (Wydoski and Whitney 1979;
Kruckeberg 1991). Important commercial fishes include the five Pacific salmon and
several rockfish species. A number of introduced species occur within the region,
DRAFT 252
-------
DRAFT
including brown and brook trout, Atlantic salmon, bass, tunicates (sea squirts), and a
saltmarsh grass (Spartina spp.). Estimates suggest that over 90 species have been
intentionally or accidentally introduced in the region (Ruckelshaus and McClure 2007).
At present, over 40 species in the region are listed as threatened and endangered under
the ESA.
Puget Sound is unique among the nation's estuaries as it is a deep fjord-like structure that
contains many urban areas within its drainage basin (Collier, O'Neill et al. 2006).
Because of the several sills that limit entry of oceanic water into Puget Sound, it is
relatively poorly flushed compared to other urbanized estuaries of North America. Thus,
toxic chemicals that enter Puget Sound have longer residence times within the system.
This entrainment of toxics can result in biota exposure to increased levels of contaminant
for a given input, compared to other large estuaries. This hydrologic isolation puts the
Puget Sound ecosystem at higher risk from other types of populations that enter the
system, such as nutrients and pathogens.
Because Puget Sound is a deep, almost oceanic habitat, the tendency of a number of
species to migrate outside of Puget Sound is limited relative to similar species in other
large urban estuaries. This high degree of residency for many marine species, combined
with the poor flushing of Puget Sound, results in a more protracted exposure to
contaminants. The combination of hydrologic and biological isolation makes the Puget
Sound ecosystem highly susceptible to inputs of toxic chemicals compared to other major
estuarine ecosystems (Collier, O'Neill et al. 2006).
An indication of this sensitivity occurs in Pacific herring, one of Puget Sound's keystone
forage fish species (Collier, O'Neill et al. 2006). These fish spend almost all of their lives
in pelagic waters and feed at the lower end of the food chain. Pacific herring should be
among the least contaminated offish species. However, monitoring has shown that
herring from the main basins of Puget Sound have higher body burdens of persistent
chemicals (e.g., PCBs) compared to herring from the severely contaminated Baltic Sea.
Thus, the pelagic food web of Puget Sound appears to be more seriously contaminated
DRAFT 253
-------
DRAFT
than previously anticipated.
Chinook salmon that are resident in Puget Sound (a result of hatchery practices and
natural migration patterns) are several times more contaminated with persistent
bioaccumulative contaminants than other salmon populations along the West Coast
(Collier, O'Neill et al. 2006). Because of associated human health concerns, fish
consumption guidelines for Puget Sound salmon are under review by the Washington
State Department of Health.
Extremely high levels of chemical contaminants are also found in Puget Sound's top
predators, including harbor seals and ESA-listed southern resident killer whales (Collier,
O'Neill et al. 2006). In addition to carrying elevated loads of toxic chemicals in their
tissues, Puget Sound's biota also show a wide range of adverse health outcomes
associated with exposure to chemical contaminants. They include widespread cancer and
reproductive impairment in bottom fish, increased susceptibility to disease in juvenile
salmon, acute die-offs of adult salmon returning to spawn in urban watersheds, and egg
and larval mortality in a variety offish. Given current regional projections for population
growth and coastal development, the loadings of chemical contaminants into Puget Sound
will increase dramatically in future years.
Land Use
The Puget Sound Lowland contains the most densely populated area of Washington. The
regional population in 2003 was an estimated 3.8 million people, with 86% residing in
King, Pierce, and Snohomish counties (Snohomish, Cedar-Sammamish Basin, Green-
Duwamish, and Puyallup River watersheds). The area is expected to attract 4 to 6 million
new human residents in the next 20 years (Ruckelshaus and McClure 2007). The
Snohomish River watershed, one of the fastest growing watersheds in the region,
increased about 16% in the same period.
Land use in the Puget Sound lowland is composed of agricultural areas (including forests
for timber production), urban areas (industrial and residential use), and rural areas (low
DRAFT 254
-------
DRAFT
density residential with some agricultural activity). Pesticides are regularly applied to
agricultural and non-agricultural lands and are found virtually in every land use area.
Pesticides and other contaminants drain into ditches in agricultural areas and eventually
to stream systems. Roads bring surface water runoff to stream systems from industrial,
residential, and landscaped areas in the urban environment. Pesticides are also typically
found in the right-of-ways of infrastructure that connect the major landscape types.
Right-of-ways are associated with roads, railways, utility lines, and pipelines.
In the 1930s, all of western Washington contained about 15.5 million acres of
"harvestable" forestland. By 2004, the total acreage was nearly half that originally
surveyed (PSAT 2007). Forest cover in Puget Sound alone was about 5.4 million acres in
the early 1990s. About a decade later, the region had lost another 200,000 acres of forest
cover with some watersheds losing more than half the total forested acreage. The most
intensive loss of forest cover occurred in the Urban Growth Boundary, which
encompasses specific parts of the Puget Lowland. In this area, forest cover declined by
11% between 1991 and 1999 (Ruckelshaus and McClure 2007). Projected land cover
changes indicate that trends are likely to continue over the next several decades with
population changes (Ruckelshaus and McClure 2007). Coniferous forests are also
projected to decline at an alarming rate as urban uses increase.
According to the 2001 State of the Sound report (PS AT 2007), impervious surfaces
covered 3.3% of the region, with 7.3% of lowland areas (below 1,000 ft elevation)
covered by impervious surfaces. From 1991 to 2001, the amount of impervious surfaces
increased 10.4% region wide. Consequently, changes in rainfall delivery to streams alter
stream flow regimes. Peak flows are increased and subsequent base flows are decreased
and alter in-stream habitat. Stream channels are widened and deepened and riparian
vegetation is typically removed which can cause increases in water temperature and will
reduce the amounts of woody debris and organic matter to the stream system.
Pollutants carried into streams from urban runoff include pesticides, heavy metals, PCBs,
polybrominated diphenyl ethers (PBDEs) compounds, PAHs, pharmaceuticals, nutrients
DRAFT 255
-------
DRAFT
(phosphorus and nitrogen), and sediment (Table 41). Other ions generally elevated in
urban streams include calcium, sodium, potassium, magnesium, and chloride ions where
sodium chloride is used as the principal road deicing salt (Paul and Meyer 2001). The
combined effect of increased concentrations of ions in streams is the elevated
conductivity observed in most urban streams.
Table 41. Examples of Water Quality Contaminants in Residential and Urban Areas
Contaminant groups
Fertilizers
Heavy Metals
Pesticides including-
Insecticides (I)
Herbicides (H)
Fungicides (F)
Wood Treatment
chemicals (WT)
Legacy Pesticides (LP)
Other ingredients in
pesticide formulations
(Ol)
Pharmaceuticals and
personal care products
Polyaromatic
hydrocarbons (PAHs)
Industrial chemicals
Select constituents
Nutrients
Pb, Zn, Cr, Cu, Cd, Ni, Hg, Mg
Organophosphates (I)
Carbamates (I)
Organochlorines (I)
Pyrethroids (I)
Triazines (H)
Chloroacetanilides (H)
Chlorophenoxy acids (H)
Triazoles (F)
Copper containing fungicides (F)
Organochlorines (LP)
Surfactants/adjuvants (Ol)
Natural and synthetic hormones
soaps and detergents
Tricyclic PAHs
PCBs
PBDEs
Dioxins
Select example(s)
Phosphorus
Nitrogen
Cu
Chlorpyrifos (I)
Diazinon (I)
Carbaryl (I)
Atrazine (H)
Esfenvalerate (I)
Creosote (WT)
DDT (LP)
Copper sulfate (F)
Metalaxyl (F)
Nonylphenol (Ol)
Ethinyl estradiol
Nonylphenol
Phenanthrene
Penta-PBDE
Source and Use
Information
lawns, golf courses,
urban landscaping
brake pad dust,
highway and parking
lot runoff, rooftops
golf courses, right of
ways, lawn and plant
care products, pilings,
bulkheads, fences
hospitals, dental
facilities, residences,
municipal and
industrial waste water
discharges
fossil fuel combustion,
oil and gasoline leaks,
highway runoff,
creosote-treated wood
utility infrastructure,
flame retardants,
electronic equipment
Many other metals have been found in elevated concentrations in urban stream sediments
including arsenic, iron, boron, cobalt, silver, strontium, rubidium, antimony, scandium,
molybdenum, lithium, and tin (Wheeler, Angermeier et al. 2005). The concentration,
storage, and transport of metals in urban streams are connected to particulate organic
matter content and sediment characteristics. Organic matter has a high binding capacity
for metals and both bed and suspended sediments with high organic matter content
frequently exhibit 50-7,500 times higher concentrations of zinc, lead, chromium, copper,
mercury, and cadmium than sediments with lower organic matter content.
DRAFT
256
-------
DRAFT
Although urban areas occupy only 2% of the Pacific Northwest land base, the impacts of
urbanization on aquatic ecosystems are severe and long lasting (Spence, Lomnicky et al.
1996). O'Neill et al. (2006) found that Chinook salmon returning to Puget Sound had
significantly higher concentrations of PCBs and PBDEs compared to other Pacific coast
salmon populations. Furthermore, Chinook salmon that resided in Puget Sound in the
winter rather than migrate to the Pacific Ocean (residents) had the highest concentrations
of POPs, followed by Puget Sound fish populations believed to be more ocean-reared.
Fall Chinook salmon from Puget Sound have a more localized marine distribution in
Puget Sound and the Georgia Basin than other populations of Chinook salmon from the
west coast of North America. This ESU is more contaminated with PCBs (2 to 6 times)
and PBDEs (5 to 17 times). O'Neill et al. (2006) concluded that regional body burdens
of contaminants in Pacific salmon, and Chinook salmon in particular, could contribute to
the higher levels of contaminants in federally-listed endangered southern resident killer
whales.
In addition to POPs, endocrine disrupters are chemicals that mimic natural hormones,
inhibit the action of hormones and/or alter normal regulatory functions of the immune,
nervous and endocrine systems and are discharged with treated effluent (King County
2002d). Endocrine disruption has been attributed to DDT and other organochlorine
pesticides, dioxins, PAHs, alkylphenolic compounds, phthalate plasticizers, naturally
occurring compounds, synthetic hormones and metals. Natural mammalian hormones
such as l?p-estradiol, are also classified as endocrine disrupters. Both natural and
synthetic mammalian hormones are excreted through the urine and are known to be
present in wastewater discharges.
Jobling et al. (1995) reported that ten chemicals known to occur in sewage effluent
interacted with the fish estrogen receptor by reducing binding of l?p-estradiol to its
receptor, stimulating transcript onal activity of the estrogen receptor or inhibiting
transcription activity. Binding of the ten chemicals with the fish endocrine receptor
indicates that the chemicals could be endocrine disrupters and forms the basis of concern
DRAFT 257
-------
DRAFT
about WWTP effluent and fish endocrine disruption.
Fish communities are impacted by urbanization (Wheeler, Angermeier et al. 2005).
Urban stream fish communities have lower overall abundance, diversity, taxa richness
and are dominated by pollution tolerant species. Lead content in fish tissue is higher in
urban areas. Furthermore, the proximity of urban streams to humans increases the risk of
non-native species introduction and establishment. Thirty-nine non-native species were
collected in Puget Sound during the 1998 Puget Sound Expedition Rapid Assessment
Survey (Brennan, Higgins et al. 2004). Lake Washington, located within a highly urban
area, has 15 non-native species identified (Ajawani 1956).
PAH compounds also have distinct and specific effects on fish at early life history stages
(Incardona, Collier et al. 2004). PAHs tend to adsorb to organic or inorganic matter in
sediments, where they can be trapped in long-term reservoirs (Johnson, Collier et al.
2002). Only a portion of sediment-adsorbed PAHs are readily bioavailable to marine
organisms, but there is substantial uptake of these compounds by resident benthic fish
through the diet, through exposure to contaminated water in the benthic boundary layer,
and through direct contact with sediment. Benthic invertebrate prey are a particularly
important source of PAH exposure for marine fishes, as PAHs are bioaccumulated in
many invertebrate species (Varanasi, Stein et al. 1989; Varanasi, Stein et al. 1992;
Meador, Stein et al. 1995).
PAHs and their metabolites in invertebrate prey are passed on to consuming fish species,
PAHs are metabolized extensively in vertebrates, including fishes (Johnson, Collier et al.
2002). Although PAHs do not bioaccumulate in vertebrate tissues, PAHs cause a variety
of deleterious effects in exposed animals. Some PAHs are known to be immunotoxic and
to have adverse effects on reproduction and development. Studies show that PAHs
exhibit many of the same toxic effects in fish as they do in mammals (Johnson, Collier et
al. 2002).
DRAFT 258
-------
DRAFT
Habita t Modiflca tion
Much of the region's estuarine wetlands have been heavily modified, primarily from
agricultural land conversion and urban development (NRC 1996). Although most
estuarine wetland losses result from conversions to agricultural land by ditching,
draining, or diking, these wetlands also experience increasing effects from industrial and
urban causes. By 1980, an estimated 27,180 acres of intertidal or shore wetlands had
been lost at 11 deltas in Puget Sound (Bortleson, Chrzastowski et al. 1980). Tidal
wetlands in Puget Sound amount to roughly 18% of their historical extent (Collins and
Sheikh 2005). Coastal marshes close to seaports and population centers have been
especially vulnerable to conversion with losses of 50-90%. By 1980, an estimated 27,180
acres of intertidal or shore wetlands had been lost at eleven deltas in Puget Sound
(Bortleson, Chrzastowski et al. 1980). More recently, tidal wetlands in Puget Sound
amount to about 17-19% of their historical extent (Collins and Sheikh 2005). Coastal
marshes close to seaports and population centers have been especially vulnerable to
conversion with losses of 50-90% common for individual estuaries. Salmon use
freshwater and estuarine wetlands for physiological transition to and from saltwater and
rearing habitat. The land conversions and losses of Pacific Northwest wetlands constitute
a major impact. Salmon use marine nearshore areas for rearing and migration, with
juveniles using shallow shoreline habitats (Brennan, Higgins et al. 2004).
About 800 miles of Puget Sound's shorelines are hardened or dredged (PSAT 2004;
Ruckelshaus and McClure 2007). The area most intensely modified is the urban corridor
(eastern shores of Puget Sound from Mukilteo to Tacoma). Here, nearly 80% has been
altered, mostly from shoreline armoring associated with the Burlington Northern Railroad
tracks (Ruckelshaus and McClure 2007). Levee development within the rivers and their
deltas has isolated significant portions of former floodplain habitat that was historically
used by salmon and trout during rising flood waters.
Urbanization has caused direct loss of riparian vegetation and soils and significantly
altered hydrologic and erosion rates. Watershed development and associated
urbanization throughout the Puget Sound, Hood Canal, and Strait of Juan de Fuca regions
DRAFT 259
-------
DRAFT
have increased sedimentation, raised water temperatures, decreased LWD recruitment,
decreased gravel recruitment, reduced river pools and spawning areas, and dredged and
filled estuarine rearing areas (Bishop and Morgan 1996 in NMFS 2008b). Large areas of
the lower rivers have been channelized and diked for flood control and to protect
agricultural, industrial, and residential development.
The NMFS' 2005 Report to Congress on implementation of the Pacific Coastal Salmon
Recovery Fund listed habitat-related factors as the leading limits to Puget Sound Chinook
Salmon and Hood-Canal Summer Run Chum recovery (PCSRF 2006). Similarly, the
principal factor for decline of Puget Sound steelhead is the destruction, modification, and
curtailment of its habitat and range. Barriers to fish passage and adverse effects on water
quality and quantity resulting from dams, the loss of wetland and riparian habitats, and
agricultural and urban development activities have contributed and continue to contribute
to the loss and degradation of steelhead habitats in Puget Sound (NMFS 2008b).
Industrial Development
More than 100 years of industrial pollution and urban development have affected water
quality and sediments in Puget Sound. Many different kinds of activities and substances
release contamination into Puget Sound and the contributing waters. According to the
State of the Sound Report (PS AT 2007) in 2004, more than 1,400 fresh and marine
waters in the region were listed as "impaired." Almost two-thirds of these water bodies
were listed as impaired due to contaminants, such as toxics, pathogens, and low dissolved
oxygen or high temperatures, and less than one-third had established cleanup plans.
More than 5,000 acres of submerged lands (primarily in urban areas; 1% of the study
area) are contaminated with high levels of toxic substances, including polybrominated
diphenyl ethers (PBDEs; flame retardants), and roughly one-third (180,000 acres) of
submerged lands within Puget Sound are considered moderately contaminated. In 2005
the Puget Sound Action Team (PS AT) identified the primary pollutants of concern in
Puget Sound and their sources listed below in Table 42.
DRAFT 260
-------
DRAFT
Table 42. Pollutants of Concern in Puget Sound (PSAT 2005)
Pollutant
Heavy Metals: Pb, Hg, Cu, and others
Organic Compounds:
Polycyclic aromatic hydrocarbons (PAHs)
Polychlorinated biphenyls (PCBs)
Dioxins, Furans
Dichloro-diphenyl-trichloroethane (DDTs)
Phthalates
Polybrominated diphenyl ethers (PBDEs)
Sources
vehicles, batteries, paints, dyes, stormwater
runoff, spills, pipes.
Burning of petroleum, coal, oil spills, leaking
underground fuel tanks, creosote, asphalt.
Solvents electrical coolants and lubricants,
pesticides, herbicides, treated wood.
Byproducts of industrial processes.
Chlorinated pesticides.
Plastic materials, soaps, and other personal
care products. Many of these compounds are
in wastewaterfrom sewage treatment plants.
PBDEs are added to a wide range of textiles
and plastics as a flame retardant. They easily
leach from these materials and have been
found throughout the environment and in
human breast milk.
Puget Sound Basin: NAWQA analysis
The USGS sampled waters in the Puget Sound Basin between 1996 and 1998. (Ebbert,
Embrey et al. 2000) reported that 26 of 47 analyzed pesticides were detected. A total of
74 manmade organic chemicals were detected in streams and rivers, with different
mixtures of chemicals linked to agricultural and urban settings. NAWQA results
reported that the herbicides atrazine, prometon, simazine and tebuthiuron were the most
frequently detected herbicides in surface and ground water (Bortleson and Ebbert 2000).
Herbicides were the most common type of pesticide found in an agricultural stream
(Fishtrap Creek) and the only type of pesticide found in shallow ground water underlying
agricultural land (Bortleson and Ebbert 2000). The most commonly detected VOC in the
agricultural landuse study area was associated with the application of fumigants to soils
prior to planting (Bortleson and Ebbert 2000). One or more fumigant-related compound
(1,2-dichloropropane, 1,2,2-trichloropropane, and 1,2,3-trichloropropane) were detected
in over half of the samples. Insecticides, in addition to herbicides, were detected
frequently in urban streams (Bortleson and Ebbert 2000). Sampled urban streams
showed the highest detection rate for the three insecticides carbaryl, diazinon, and
malathion. Carbaryl was detected at over 60% of urban sample sites (Ebbert et al. 2000).
The insecticide diazinon was also frequently detected in urban streams at concentrations
that exceeded EPA guidelines for protecting aquatic life (Bortleson and Ebbert 2000).
DRAFT
261
-------
DRAFT
Insecticides screened for included both carbofuran and methomyl. Carbofuran was
detected while methomyl was not. No insecticides were found in shallow ground water
below urban residential land (Bortleson and Ebbert 2000).
Habita t Restora tion
Positive changes in water quality in the region are evident. One of the most notable
improvements was the elimination of sewage effluent to Lake Washington in the mid-
1960s. This significantly reduced problems within the lake from phosphorus pollution
and triggered a concomitant reduction in cyanobacteria (Ruckelshaus and McClure
2007). Even so, as the population and industry has risen in the region a number of new
and legacy pollutants are of concern.
Mining
Mining has a long history in Washington. In 2004, the state was ranked 13th nationally in
total nonfuel mineral production value and 17th in coal production (Palmisano, Ellis et al.
1993; NMA 2007). Metal mining for all metals (zinc, copper, lead, silver, and gold)
peaked between 1940 and 1970 (Palmisano, Ellis et al. 1993). Today, construction sand
and gravel, Portland cement, and crushed stone are the predominant materials mined.
Where sand and gravel is mined from riverbeds (gravel bars and floodplains) it may
result in changes in channel elevations and patterns, instream sediment loads, and
seriously alter instream habitat. In some cases, instream or floodplain mining has
resulted in large scale river avulsions. The effect of mining in a stream or reach depends
upon the rate of harvest and the natural rate of replenishment, as well as flood and
precipitation conditions during or after the mining operations.
Artificial Propagation
The artificial propagation of late-returning Chinook salmon is widespread throughout
Puget Sound (Good, Waples et al. 2005). Summer/fall Chinook salmon transfers
between watersheds within and outside the region have been commonplace throughout
this century. Therefore, the purity of naturally spawning stocks varies from river to river.
Nearly 2 billion Chinook salmon have been released into Puget Sound tributaries since
the 1950s. The vast majority of these have been derived from local late-returning adults.
DRAFT 262
-------
DRAFT
Returns to hatcheries have accounted for 57% of the total spawning escapement.
However, the hatchery contribution to spawner escapement is probably much higher than
that due to hatchery-derived strays on the spawning grounds. The genetic similarity
between Green River late-returning Chinook salmon and several other late-returning
Chinook salmon in Puget Sound suggests that there may have been a significant and
lasting effect from some hatchery transplants (Marshall etal. 1995).
Overall, the use of Green River stock throughout much of the extensive hatchery network
in this ESU may reduce the genetic diversity and fitness of naturally spawning
populations (Good, Waples et al. 2005).
Hydromodification Projects
More than 20 dams occur within the region's rivers and overlap with the distribution of
salmonids. A number of basins contain water withdrawal projects or small
impoundments that can impede migrating salmon. The resultant impact of these and land
use changes (forest cover loss and impervious surface increases) has been a significant
modification in the seasonal flow patterns of area rivers and streams, and the volume and
quality of water delivered to Puget Sound waters. Several rivers have been
hydromodified by other means including levees and revetments, bank hardening for
erosion control, and agriculture uses. Since the first dike on the Skagit River delta was
built in 1863 for agricultural development (Ruckelshaus and McClure 2007), other basins
like the Snohomish River are diked and have active drainage systems to drain water after
high flows that top the dikes. Dams were also built on the Cedar, Nisqually, White,
Elwha, Skokomish, Skagit, and several other rivers in the early 1900s to supply urban
areas with water, prevent downstream flooding, allow for floodplain activities (like
agriculture or development), and to power local timber mills (Ruckelshaus and McClure
2007).
Over the next few years, however, a highly publicized and long discussed dam removal
project is expected to begin in the Elwha River. The removal of two dams in the Elwha
DRAFT 263
-------
DRAFT
River, a short but formerly very productive salmon river, is expected to open up more
than 70 miles of high quality salmon habitat (Wunderlich, Winter et al. 1994;
Ruckelshaus and McClure 2007). Estimates suggest that nearly 400,000 salmon could
begin using the basin within 30 years after the dams are removed (PSAT 2007).
In 1990, only one-third of the water withdrawn in the Pacific Northwest was returned to
the streams and lakes (NRC 1996). Water that returns to a stream from an agricultural
irrigation is often substantially degraded. Problems associated with return flows include
increased water temperature, which can alter patterns of adult and smolt migration;
increased toxicant concentrations associated with pesticides and fertilizers; increased
salinity; increased pathogen populations; decreased dissolved oxygen concentration; and
increased sedimentation (NRC 1996). Water-level fluctuations and flow alterations due
to water storage and withdrawal can affect substrate availability and quality, temperature,
and other habitat requirements of salmon. Indirect effects include reduction of food
sources; loss of spawning, rearing, and adult habitat; increased susceptibility of juveniles
to predation; delay in adult spawning migration; increased egg and alevin mortalities;
stranding of fry; and delays in downstream migration of smolts (NRC 1996).
Commercial and Recreational Fishing
Most of the commercial landings in the region are groundfish, Dungeness crab, shrimp,
and salmon. Many of the same species are sought by Tribal fisheries and by charter and
recreational anglers. Nets and trolling are used in commercial and Tribal fisheries.
Recreational anglers typically use hook and line, and may fish from boat, river bank, or
docks. Entanglement of marine mammals in fishing gear is not uncommon and can lead
to mortality or serious injury.
Pesticides used in commercial oyster-producing areas in Willapa Bay and Gray's Harbor
may have an impact on salmonids residing in these locations. Currently, under a
statewide SLN or Section 24 (c) registration, carbaryl can be applied to aerially applied to
intertidal areas when exposed at low tide. Pesticide use is intended to control ghost
shrimp (Neotrypaea californiensis) and mud shrimp (Upogebiapugettensis). An NPDES
DRAFT 264
-------
DRAFT
permit is required for application, and although the registration allows application
anywhere in the state, these locations are the only ones covered under active permits.
There has been some discussion of phasing out carbaryl use and replacing it with
alternative pesticides.
Harvest impacts on Puget Sound Chinook salmon populations average 75% in the earliest
five years of data availability and have dropped to an average of 44% in the most recent
five-year period (Good, Waples et al. 2005). Populations in Puget Sound have not
experienced the strong increases in numbers seen in the late 1990s in many other ESUs.
Although more populations have increased than decreased since the last BRT assessment,
after adjusting for changes in harvest rates, trends in productivity are less favorable.
Most populations are relatively small, and recent abundance within the ESU is only a
small fraction of estimated historic run size.
Atmospheric deposition
Pesticides were detected in wet deposition (rain) (Capel et al. 1998), and snow samples
from Mount Rainier National Park, Washington (Hageman et al. 2006). Three of the four
most frequently detected pesticides were found in the Mount Rainier snow (dacthal,
chlorpyrifos, and endosulfan).
Oregon-Washington-Northern California Coastal Drainages
This region encompasses drainages originating in the Klamath Mountains, the Oregon
Coast Mountains, and the Olympic Mountains. More than 15 watersheds drain the
region's steep slopes including the Umpqua, Alsea, Yaquina, Nehalem, Chehalis,
Quillayute, Queets, and Hoh rivers. Numerous other small to moderately sized streams
dot the coastline. Many of the basins in this region are relatively small. The Umpqua
River drains a basin of 4,685 square miles and is slightly over 110 miles long. The
Nehalem River drains a basin of 855 square miles and is almost 120 miles long.
However, systems here represent some of the most biologically diverse basins in the
Pacific Northwest (Kagan, Hak et al. 1999; Belitz, Hamlin et al. 2004; Carter and Resh
2005).
DRAFT 265
-------
DRAFT
Land Use
The rugged topography of the western Olympic Peninsula and the Oregon Coastal Range
has limited the development of dense population centers. For instance, the Nehalem
River and the Umpqua River basins consist of less than 1% urban land uses. Most basins
in this region have long been exploited for timber production, and are still dominated by
forest lands. In Washington State, roughly 90% of the coastal region is forested
(Palmisano, Ellis et al. 1993). Roughly 80% of the Oregon Coastal Range is forested as
well (Gregory 2000). Approximately 92% of the Nehalem River basin is forested, with
only 4% considered agricultural (Belitz, Hamlin et al. 2004). Similarly, in the Umpqua
River basin, about 86% is forested land, 5% agriculture, and 0.5% is considered urban
lands. Roughly half the basin is under federal management (Carter and Resh 2005).
Habitat Modification
While much of the coastal region is forested, it has still been impacted by land use
practices. Less than 3% of the Oregon coastal forest is old growth conifers (Gregory
2000). The lack of mature conifers indicates high levels of habitat modification. As
such, overall salmonid habitat quality is poor, though it varies by watershed. The amount
of remaining high quality habitat ranges from 0% in the Sixes to 74% in the Siltcoos
(ODFW 2005). Approximately 14% of freshwater winter habitat available to juvenile
coho is of high quality. Much of the winter habitat is unsuitable due to high
temperatures. For example, 77% of coho salmon habitat in the Umpqua basin exceeds
temperature standards.
Reduction in stream complexity is the most significant limiting factor in the Oregon
coastal region. An analysis of the Oregon coastal range determined the primary and
secondary lifecycle bottlenecks for the 21 populations of coastal coho salmon (Nicholas
et al 2005). Nicholas et al. (2005) determined that stream complexity is either the
primary (13) or secondary (7) bottleneck for every population. Stream complexity has
been reduced through past practices such as splash damming, removing riparian
vegetation, removing LWD, diking tidelands, filling floodplains, and channelizing rivers.
DRAFT 266
-------
DRAFT
Habitat loss through wetland fills is also a significant factor. Table 43 summarizes the
change in area of tidal wetlands for several Oregon estuaries (Good 2000).
Table 43. Change in total area (acres ) of tidal wetlands (tidal marshes and swamps) due
to filling and diking between 1870 and 1970 (Good 2000).
Estuary
Necanicum
Nehalem
Tillamook
Netarts
Sand Lake
Nestucca
Salmon
Siletz
Yaquina
Alsea
Siuslaw
Umpqua
Coos Bay
Coquille
Rogue
Chetco
Total
Diked or
Filled Tidal
Wetland
15
1,571
3,274
16
9
2,160
313
401
1,493
665
1,256
1,218
3,360
4,600
30
5
20,386
Percent of
1870 Habitat
Lost
10
75
79
7
2
91
57
59
71
59
63
50
66
94
41
56
72%
The only listed salmonid population in coastal Washington is the Ozette Lake Sockeye.
The range of this ESU is small, including only one lake (31 km2) and 71 km of stream.
Like the Oregon Coastal drainages, the Ozette Lake area has been heavily managed for
logging. Logging resulted in road building and the removal of LWD, which affected the
nearshore ecosystem (NMFS 2008c). LWD along the shore offered both shelter from
predators and a barrier to encroaching vegetation (NMFS 2008c). Aerial photograph
analysis shows near-shore vegetation has increased significantly over the past 50 years
(Ritchie 2005). Further, there is strong evidence that water levels in Ozette Lake have
dropped between 1.5 and 3.3 ft from historic levels (Herrera 2005 in NMFS 2008c). The
impact of this water level drop is unknown. Possible effects include increased
desiccation of sockeye redds and loss of spawning habitat. Loss of LWD has also
DRAFT
267
-------
DRAFT
contributed to an increase in silt deposition, which impairs the quality and quantity of
spawning habitat.
Very little is known about the relative health of the Ozette Lake tributaries and their
impact on the sockeye salmon population.
Mining
Oregon is ranked 35th nationally in total nonfuel mineral production value in 2004. In
that same year, Washington was ranked 13th nationally in total nonfuel mineral
production value and 17th in coal production (Palmisano, Ellis et al. 1993; NMA 2007).
Metal mining for all metals (e.g., zinc, copper, lead, silver, and gold) peaked in
Washington between 1940 and 1970 (Palmisano, Ellis et al. 1993). Today, construction
sand, gravel, Portland cement, and crushed stone are the predominant materials mined in
both Oregon and Washington. Where sand and gravel is mined from riverbeds (gravel
bars and floodplains) changes in channel elevations and patterns, instream sediment
loads, may result and alter instream habitat. In some cases, instream or floodplain mining
has resulted in large scale river avulsions. The effect of mining in a stream or reach
depends upon the rate of harvest and the natural rate of replenishment. Additionally, the
severity of the effects is influenced by flood and precipitation conditions during or after
the mining operations.
Hydromodification Projects
Compared to other areas in the greater Northwest Region, the coastal region has fewer
dams and several rivers remain free flowing (e.g., Clearwater River). The Umpqua River
is fragmented by 64 dams, the fewest number of dams on any large river basin in Oregon
(Carter and Resh 2005). According to Palmisano et al. (1993) dams in the coastal
streams of Washington permanently block only about 30 miles of salmon habitat. In the
past, temporary splash dams were constructed throughout the region to transport logs out
of mountainous reaches. The general practice involved building a temporary dam in the
creek adjacent to the area being logged, and filling the pond with logs. When the dam
broke the floodwater would carry the logs to downstream reaches where they could be
DRAFT 268
-------
DRAFT
rafted and moved to market or downstream mills. Thousands of splash dams were
constructed across the Northwest in the late 1800s and early 1900s. While the dams
typically only temporarily blocked salmon habitat, in some cases dams remained long
enough to wipe out entire salmon runs. The effects of the channel scouring and loss of
channel complexity resulted in the long-term loss of salmon habitat (NRC 1996).
Commercial and Recreational Fishing
Most commercial landings in the region are groundfish, Dungeness crab, shrimp, and
salmon. Many of the same species are sought by Tribal fisheries, as well as by charter,
and recreational anglers. Nets and trolling are used in commercial and Tribal fisheries.
Recreational anglers typically use hook and line and may fish from boat, river bank, or
docks.
Atmospheric deposition
Pesticides and other chemicals may be transported through the air and later deposited on
land and into waterways. For example, orthophosphate insecticides were detected in two
Oregon streams, Hood River and Mill Creek (tributaries of the Columbia River).
Detection occurred following periods of chemical applications on orchard crops, and may
be related to atmospheric drift, mixing operations, or other aspects of pesticide use.
Environmental Protection Programs
When using carbaryl, carbofuran, and methomyl, growers must adhere to the court-
ordered injunctive relief, requiring buffers of 20 yards for ground application and 100
yards for any aerial application. These measures are mandatory in all four states, pending
completion of consultation.
California and Oregon both have Pesticide Use Reporting System (PURS) legislation.
California PURS requires all agricultural uses of registered pesticides be reported. In this
case "agricultural" use includes applications to parks, golf courses, and most livestock
uses. Oregon requires reporting if application is part of a business, for a government
agency, or in a public place. However, the Governor of Oregon has suggested
suspending the PURS program for 2009 - 2011 due to budget shortages. A final decision
DRAFT 269
-------
DRAFT
will be made during the summer. If suspension occurs, PURS will resume for the 2012
growing season.
Washington State has a Surface Water Monitoring Program that looks at pesticide
concentrations in some salmonid bearing streams and rivers. The program was initiated
in 2003 and now monitors four areas. Three of these were chosen due to high overlap
with agriculture: the Skagit-Samish watershed, the Lower Yakima Watershed, and the
Wenatchee and Entiat watersheds. The final area, in the Cedar-Sammamish watershed, is
an urban location, intended to look at runoff in a non-agriculture setting. It was chosen
due to detection of pesticides coincident with pre-spawning mortality in Coho salmon.
The Surface Water Monitoring program is relatively new and will continue to add
watersheds and testing for additional pesticides over time.
Washington State also has a voluntary program that assists growers in addressing water
rights issues within a watershed. Several watersheds have elected to participate, forming
Comprehensive Irrigation District Management Plans (CIDMPs). The CIDMP is a
collaborative process between government and landowners and growers; the parties
determine how they will ensure growers get the necessary volume of water while also
guarding water quality. This structure allows for greater flexibility in implementing
mitigation measures to comply with both the CWA and the ESA.
Oregon has also implemented a voluntary program. The Pesticide Stewardship
Partnerships (PSP) program began in 1999 through the Oregon Department of
Environmental Quality. Like the CIDMP program, the goal is to involve growers and
other stakeholders in water quality management at a local level. Effectiveness
monitoring is used to provide feedback on the success of mitigation measures. As of
2006, there were six pilot PSPs planned or in place. Early results from the first PSPs in
the Columbia Gorge Hood River and in Mill Creek demonstrate reductions in
chlorpyrifos and diazinon levels and detection frequencies. DEQ's pilot programs
suggest that PSPs can help reduce contamination of surface waters.
DRAFT 270
-------
DRAFT
Oregon is in the process of developing a Pesticide Management Plan for Water Quality
Protection, as required under FIFRA. This plan describes how government agencies and
stakeholders will collaboratively reduce pesticides in Oregon water supplies. The PSP
program is a component of this Plan, and will provide information on the effectiveness of
mitigation measures.
The Columbia Gorge Fruit Growers Association is a non-profit organization dedicated to
the needs of growers in the mid-Columbia area. The association brings together over 440
growers and 20 shippers of fruit from Oregon and Washington. It has issued a BMP
handbook for OPs, including information on alternative methods of pest control.
However, their website does not mention carbamate pesticides. The mid-Columbia area
is of particular concern, as many orchards are in close proximity to streams.
Idaho State Department of Agriculture has published a BMP guide for pesticide use. The
BMPs include eight "core" voluntary measures that will prevent pesticides from leaching
into soil and groundwater. These measures include applying pest-specific controls, being
aware of the depth to ground water, and developing an Irrigation Water Management
Plan.
Integration of the Environmental Baseline on Listed Resources
Collectively, the components of the environmental baseline for the action area include
sources of natural mortality as well as influences from natural oceanographic and climatic
features in the action area. Climatic variability may affect the growth, reproductive
success, and survival of listed Pacific salmonids in the action area. Temperature and
water level changes may lead to: (1) Reduced summer and fall stream flow, leading to
loss of spawning habitat and difficulty reaching spawning beds; (2) increased winter
flooding and disturbance of eggs; (3) changes in peak stream flow timing affecting
juvenile migration; and (4) rising water temperature may exceed the upper temperature
limit for salmonids at 64°F (18°C) (JISAO 2007). Additional indirect impacts include
changes in the distribution and abundance of the prey and the distribution and abundance
DRAFT 271
-------
DRAFT
of competitors or predators for salmonids. These conditions will influence the population
structure and abundance for all listed Pacific salmonids.
The baseline also includes human activities resulting in disturbance, injury, or mortality
of individual salmon. These activities include hydropower, hatcheries, harvest, and
habitat degradation, including poor water quality and reduced availability of spawning
and rearing habitat for all 28 ESUs/DPSs. As such, these activities degrade salmonid
habitat, including all designated critical habitat and their PCEs. While each area is
affected by a unique combination of stressors, the two major impacts to listed Pacific
salmonid critical habitat are habitat loss and decreased prey abundance. Although habitat
restoration and hydropower modification measures are ongoing, the long-term beneficial
effects of these actions on Pacific salmonids, although anticipated, remain to be realized.
Thus, we are unable to quantify these potential beneficial effects at this time.
Listed Pacific salmonids and designated critical habitat may be affected by the proposed
registration of carbaryl, carbofuran, and methomyl in California, Idaho, Oregon, and
Washington. These salmonids are and have been exposed to the components of the
environmental baseline for decades. The activities discussed above have some level of
effect on all 28 ESUs/DPSs in the proposed action area. They have also eroded the
quality and quantity of salmonid habitat - including designated critical habitat. We
expect the combined consequences of those effects, including impaired water quality,
temperature, and reduced prey abundance, may increase the vulnerability and
susceptibility of overall fish health to disease, predation, and competition for available
suitable habitat and prey items. The continued trend of anthropogenic impairment of
water quality and quantity on Pacific salmonids and their habitats may further compound
the declining status and trends of listed salmonids, unless measures are implemented to
reverse this trend.
DRAFT 272
-------
DRAFT
o,
The analysis includes three primary components: exposure, response, and risk
characterization. We analyze exposure and response, and integrate the two in the risk
characterization phase where we address support for risk hypotheses. These risk
hypotheses are predicted on effects to salmonids and designated critical habitats' PCEs.
The combined analysis evaluated effects to listed Pacific salmonids and their designated
critical habitat (see Approach to the Assessment).
Exposure Analysis
In this section, we identify and evaluate exposure information from the stressors of the
action (Figure 1). We begin by presenting a general discussion of the physical and
chemical properties of carbaryl, carbofuran, and methomyl that influence the distribution
and persistence of action stressors in the environment and exposure of listed species and
designated critical habitat. Next we present general life history information of Pacific
salmon and steelhead and evaluate the likely co-occurrence of action stressors with the
listed Pacific salmonids. We then summarize exposure estimates presented in the three
BEs and present other sources of information, including other modeling estimates and
monitoring data to further characterize exposure to listed species and designated critical
habitat. Finally, we conclude with a summary of expected ranges of exposure and the
uncertainty contained in the exposure analysis. Because the ESA section 7 consultation
process is intended to ensure that the agency action is not likely to jeopardize listed
species or destroy or adversely modify critical habitat, NMFS considers a variety of
exposure scenarios in addition to those presented in EPA's BEs. These scenarios provide
exposure estimates for the range of habitats utilized by listed salmonids.
Co-occurrence of action stressors
and listed species
Distribution of
individuals
Distribution of
habitat
Exposure Profile
Figure 35. Exposure analysis
DRAFT
273
-------
Summary of Chemical Fate ofA.I.s
DRAFT
Carbaryl
Carbofuran
o
Methomyl
..X*
Figure 36. Chemical structures of carbaryl, carbofuran, and methomyl.
Carbaryl
"Carbaryl is a widely used pesticide that is commonly detected in the environment from
its application in agricultural and non-agricultural settings (EPA 2003)." Carbaryl is
primarily applied to terrestrial habitats, although a 24(c) registration in Washington State
allows for application to commercial oyster beds to control native ghost shrimp and mud
shrimp. Carbaryl can contaminate surface waters via runoff, erosion, leaching, and spray
drift from application at terrestrial sites, or direct application to aquatic habitats.
Carbaryl and its primary degradate, 1-naphthol, are fairly mobile and slightly persistent in
the environment. Although they are not likely to persist or accumulate under most
conditions, they may do so under acidic conditions with limited microbial activity.
Carbaryl dissipates in the environment by abiotic and microbially mediated degradation.
The environmental fate characteristics for carbaryl are listed below (Table 44).
DRAFT
274
-------
DRAFT
Table 44. Environmental fate characteristics of carbaryl (EPA 2003).
Parameter
Water solubility
Vapor pressure
Henry's law constant
Octanol/Water partition
Hydrolysis (ti/2) pH 5, pH 7, and pH 9
Aqueous photolysis (r/2)
Soil photolysis(t'/2)
Aerobic soil metabolism (r/2)
Anaerobic soil metabolism (r/2)
Aerobic aquatic metabolism (ti/2)
Anaerobic aquatic metabolism (ti/2)
Koc
Value
32 mg/L (ppm) at 20 degrees C
1.36107mmHg(25degC)
1 .28 x 1 0-" atm msmol1
Kow=229
Stable, 12 days, 3.2 hours
21 days
assumed stable
4 days - sandy loam soil
72 days
4.9 days
72 days
177-249ml/g
Fish are most likely to be exposed to carbaryl through direct uptake of the chemical from
the water column and across the gills, although other routes of exposure may also be
important. Potential exposure routes for aquatic organisms include direct uptake from the
water column or pore water of sediment, incidental ingestion of the chemical in sediment,
or ingestion of the chemical in food items. Accumulation of carbaryl can occur in fish,
invertebrates, algae and plants. Residue levels in fish can be 140 fold greater than the
concentration of carbaryl in water (http ://pmep. cce. Cornell .edu/profiles/extoxnet/carbaryl-
dicrotophos/carbaryl-ext.html#14). In general, due to its rapid metabolism and rapid
degradation, carbaryl should not pose a significant bioaccumulation risk in alkaline
waters. However, under conditions below neutrality accumulation of carbaryl may be
significant, (http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryl-dicrotophos/carbaryl-
ext.html#14)
The major degradation products of carbaryl are CC>2 and 1-naphthol, which is further
degraded to CO2. Carbaryl is stable to hydrolysis in acidic conditions but hydrolyzes in
neutral environments. Hydrolysis rates increase with increasing alkalinity. Carbaryl is
degraded by photolysis in water with a half-life of 21 days (d). Under aerobic conditions,
it degrades rapidly by microbial metabolism in soil and aquatic environments.
Metabolism is much slower in anaerobic environments, with half-lives on the order of 2
to 3 months. Carbaryl is mobile in the environment. Sorption onto soils is positively
correlated with increasing soil organic content. Because of its low octanol/water partition
DRAFT
275
-------
DRAFT
coefficient (Kow values range from 65 to 229), carbaryl is not expected to significantly
bioaccumulate (EPA 2003).
The major metabolite of carbaryl degradation by both abiotic and microbially mediated
processes is 1-naphthol. This degradate represented up to 67% of the applied carbaryl in
degradation studies. It is also formed in the environment by degradation of naphthalene
and other PAH compounds. Data suggest 1-naphthol is less persistent and less mobile
than its, parent carbaryl (EPA 2003).
In a field dissipation study, carbaryl was applied on 3- to 8- ft tall pine trees in an
Oregon forest. Maximum measured concentrations were 264 mg/kg on foliage at 2 d
post-treatment, 28.7 mg/kg in leaf litter after 92 days, 0.16 mg/kg in the upper 15 cm of
litter-covered soil at 62 d, and 1.14 mg/kg in the upper 15 cm of exposed soil at 2 d.
Carbaryl was detected in the leaf litter up to 365 d after treatment and in litter-covered
soil up to 302 d after treatment. Half-lives were 21 d on foliage, 75 d in leaf litter, and 65
days in soil. Carbaryl was detected at <0.003 mg/L in water and <0.003 mg/kg in
sediment from a pond and stream located approximately 50 ft from the treated area (EPA
2003).
Carbofuran
Carbofuran is an TV-methyl carbamate and is used as a broad spectrum insecticide and
nematicide. Direct application of carbofuran to aquatic habitats is not permitted.
Carbofuran can contaminate surface waters via runoff, erosion, leaching, and spray drift
from application at upland sites.
Potential exposure routes for aquatic organisms include direct uptake of the chemical
from the water column or pore water of sediment, incidental ingestion of the chemical in
sediment, or ingestion of the chemical in food items. Considering that carbofuran has
high water solubility, low Kow, and a relatively low bioconcentration factor (2-12), fish
are most likely to be exposed to carbofuran through direct uptake from the water column.
DRAFT 276
-------
DRAFT
Exposure via the food chain, pore water, and sediment pathways are also possible but are
less likely to be relevant for most life stages offish (EPA 2004).
Table 45. Environmental fate characteristics of carbofuran (EPA 2004).
Parameter
Water solubility
Vapor pressure
Henry's law constant
Octanol/Water partition
Hydrolysis (ti/2) pH 6, pH 7, pH 7, & pH 9
Aqueous photolysis (r/2)
Soil photolysis (r/2)
Aerobic soil metabolism (r/2)
Anaerobic soil metabolism (r/2)
Aerobic aquatic metabolism (ti/2)
Anaerobic aquatic metabolism (ti/2)
Koc
Value
700 mg/liter
6x10"' torrs
No data
Kow = 30 mg/g
Stable, 28 days, 3 days, and 0.8-15 hrs
6 days
78 days
321 days
624 days
No data
No data
9 to 62 ml/g
Carbofuran is highly mobile and can leach to ground water in many soils or reach surface
waters via runoff. The median KOC of carbofuran is 30 and the Freundlich coefficient (Kf)
ranges from 0.10 to 30.3 (Table 45)(EPA 2004). Major factors influencing the fate and
persistence of carbofuran are water and soil pH. Carbofuran is very mobile and persistent
in acidic environments, but dissipates more rapidly in pHs that are basic. Carbofuran is
stable to hydrolysis at pHs < 6, but becomes increasingly susceptible to hydrolysis as the
pH increases, hydrolyzing rapidly in alkaline aquatic environments (EPA 2004). A study
evaluating its persistence in natural surface waters found it took 3 weeks to degrade by
50% (Sharom, Miles et al. 1980). Carbofuran phenol (7-phenol) was the only degradate
detected in hydrolysis studies (EPA 2004). The rate of carbofuran degradation in soils is
also pH dependent. In an acidic soil (pH 5.7), carbofuran dissipated with a half-life of
321 d, but when the soil was limed to a pH of 7.7, the half-life dropped to 149 d. The
major identified degradate was 3-keto carbofuran, which peaked at 12% of the amount
applied after 181 d. The other degradation products formed during photolysis, soil, and
aquatic metabolism studies are 3-hydroxycarbofuran, 3-hydroxy-7-phenol, and 3-keto-7-
phenol (EPA 2004).
In an aqueous photolysis study carbofuran photodegraded in neutral water (buffered [pH
7] solution) at 25°C with a half-life of 6 d. In a soil photolysis study, carbofuran
DRAFT
277
-------
DRAFT
photodegraded with a half-life of 78 d on a sandy loam soil. Carbofuran is moderately
persistent to microbial degradation, with half-lives on the order of a year. Near-surface
photolysis is significant under laboratory conditions in aqueous solution, with a half-life
on the order of days (EPA 2004).
Methomyl
Methomyl is a broad spectrum insecticide approved for a variety of terrestrial use sites.
Use in aquatic habitats is not permitted. Methomyl is moderately persistent and highly
mobile and may contaminate surface waters through runoff, erosion, leaching, and spray
drift from application at upland sites.
Potential exposure routes for fish and other aquatic organisms include direct uptake of the
chemical from the water column or pore water of sediment, incidental ingestion of the
chemical in sediment, or ingestion of the chemical in food items. Methomyl
bioaccumulation has not been studied in fish, although it is expected to be relatively low
given low octanol/water partition coefficients (Kow ranges from 1.29 to 1.33, Table 46).
Methomyl degradation occurs through metabolism and photolysis. Site-specific factors
affecting the persistence of methomyl include aerobicity, organic matter and soil moisture
content, exposure to sunlight, and pH. Climate, crop management, soil type and other
site-specific factors also influence leaching and runoff.
Table 46. Environmental fate characteristics of methomyl (EPA 2003).
Parameter
Water solubility
Vapor pressure
Henry's law constant
Octa no I/Water partition1
Hydrolysis (ti/2) pH 5, pH 7, & pH 9
Aqueous photolysis (r/2)
Soil photolysis (r/2)
Aerobic soil metabolism (r/2)
Anaerobic soil metabolism (r/2)
Aerobic aquatic metabolism (ti/2)
Anaerobic aquatic metabolism (ti/2)
Koc
Value
58,000 ppm
1x10-smMHg
1.8x1010atm-m3/mol
Kow =1.29 -1.33
Stable, stable, & 30 days
1 day
36 days
1 1 to 45 days
No data
3.5-4.5 days1
<7 to 14 days
24
1 Reported in (EPA 2007)
DRAFT 278
-------
DRAFT
Methomyl photolysis occurs relatively quickly in water but slowly in soils. It is
moderately stable to aerobic soil metabolism but degrades more rapidly under anaerobic
conditions. While methomyl becomes more susceptible to hydrolysis as the pH increases
above neutral, this is not expected to be a major route of dissipation under most
circumstances. Laboratory studies show that methomyl does not readily adsorb to soil
and has the potential to be very mobile. Dissipation from the soil surface occurs by a
combination of chemical breakdown and movement. Field studies show that the varying
dissipation rates for methomyl were related primarily to differences in soil moisture
content, which may affect the microbial activity, and rainfall/irrigation, which could
influence leaching (EPA 2003).
Several degradates of methomyl have been identified. CC>2 is a major degradate of
methomyl. Another degradate, S-methyl-N-hydroxythioacetamidate, which is highly
mobile, appears to be primarily a product of alkaline hydrolysis. In an aquatic
metabolism study, methomyl degraded with estimated half-lives of 4-5 d. After 7 d,
acetonitrile comprised a maximum of 17% and acetamide up to 14% of the amount of
methomyl applied. After 102 d, volatilized acetonitrile totaled up to 27% of the parent
methomyl applied and CChup to 46% of the applied material (EPA 2003).
Habitats Occupied by Listed Salmonids
Listed salmonids occupy habitats that range from shallow, low flow freshwaters to open
reaches of the Pacific Ocean. All listed Pacific salmonid species use freshwater,
estuarine, and marine habitats. The temporal and spatial use of habitats by salmonids
depends on the species and the individuals' life history and life stage (Table 47). Many
migrate hundreds or thousands of miles during their lifetime, increasing the likelihood
that they will come in contact with aquatic habitats contaminated with pesticides. .
Given that all listed Pacific salmonid ESUs/DPSs use watersheds where the use of
carbaryl, carbofuran, and methomyl products is authorized, and these compounds are
frequently detected in watersheds where they are used (Gilliom, Barbash et al. 2006), we
DRAFT 279
-------
DRAFT
expect all listed Pacific salmonid ESUs/DPSs will continue to be exposed to these
compounds and other stressors of the action.
Table 47. General life histories of Pacific salmonids.
Species
General Life History Descriptions
(number of
listed
ESUs or
DPSs)
Spawning Migration
Spawning Habitat
Juvenile Rearing and Migration
Chinook
(9)
Mature adults (usually
four to five years old)
enter rivers (spring
through fall, depending on
run). Adults migrate and
spawn in river reaches
extending from above the
tidewater to as far as
1,200 miles from the sea.
Chinook salmon migrate
and spawn in four distinct
runs (spring, fall, summer,
and winter). Chinook
salmon are semelparous
(can spawn only once).
Generally spawn in
the middle and
upper reaches of
main stem rivers
and larger tributary
streams.
The alevin life stage primarily
resides just below the gravel
surface until they approach or
reach the fry stage. Immediately
after leaving the gravel, fry swim-
up and distribute to habitats that
provide refuge from fast currents
and predators. Juveniles exhibit
two general life history types:
Ocean-type fish migrate to sea in
their first year, usually within six
months of hatching. Ocean-type
juveniles may rear in the estuary
for extended periods. Stream-
type fish migrate to the sea in the
spring of their second year.
Coho
(4)
Mature adults (usually two
to four years old) enter
the rivers in the fall. The
timing varies depending
on location and other
variables. Coho salmon
are semelparous (can
spawn only once).
Spawn throughout
smaller coastal
tributaries, usually
penetrating to the
upper reaches to
spawn. Spawning
takes place from
October to March.
Following emergence, fry move
to shallow areas near stream
banks. As fry grow they
distribute up and downstream
and establish territories in small
streams, lakes, and off-channel
ponds. Here they rear for about
18 months. In the spring of their
second year juveniles rapidly
migrate to sea. Initially, they
remain in nearshore waters of
the estuary close to the natal
stream following downstream
migration.
Chum
(2)
Mature adults (usually
three to four years old)
enter rivers as early as
July, with arrival on the
spawning grounds
occurring from September
to January. Chum salmon
are semelparous (can
spawn only once).
Generally spawn
from just above
tidewater in the
lower reaches of
mainstem rivers,
tributary stream, or
side channels to
100 km upstream.
The alevin life stage primarily
resides just below the gravel
surface until they approach or
reach the fry stage. Immediately
after leaving the gravel, swim-up
fry migrate downstream to
estuarine areas. They reside in
estuaries near the shoreline for
one or more weeks before
migrating for extended distances,
usually in a narrow band along
the Pacific Ocean's coast.
DRAFT
280
-------
DRAFT
General life histories of Pacific salmonids (continued)
(number of
listed
ESUs)
Spawning Migration
Spawning Habitat
Juvenile Rearing and Migration
Sockeye Mature adults (usually
(2) four to five years old)
begin entering rivers from
May to October. Sockeye
are semelparous (can
spawn only once).
Spawn along
lakeshores where
springs occur and
in outlet or inlet
streams to lakes.
The alevin life stage primarily
resides just below the gravel
surface until they approach or
reach the fry stage. Immediately
after leaving the gravel, swim-up
fry migrate to nursery lakes or
intermediate feeding areas along
the banks of rivers. Populations
that migrate directly to nursery
lakes typically occupy shallow
beach areas of the lake's littoral
zone; a few cm in depth. As they
grow larger they disperse into
deeper habitats. Juveniles
usually reside in the lakes for
one to three years before
migrating to offshore habitats in
the ocean. Some are residual,
and complete their entire
lifecycle in freshwater.
Steelhead Mature adults (three to
(11) five years old) may enter
rivers any month of the
year, and spawn in late
winter or spring.
Migration in the Columbia
River system extends up
to 900 miles from the
ocean in the Snake River.
Steelhead are iteroparous
(can spawn more than
once).
Usually spawn in
fine gravel in a
riffle above a pool.
The alevin life stage primarily
resides just below the gravel
surface until they approach or
reach the fry stage. Immediately
after leaving the gravel, swim-up
fry usually inhabit shallow water
along banks of stream or aquatic
habitats on streams margins.
Steelhead rear in a wide variety
of freshwater habitats, generally
for two to three years, but up to
six or seven years is possible.
They smolt and migrate to sea in
the spring.
DRAFT
281
-------
DRAFT
Modeling: Estimates of Exposure to Carbaryl, Carbofuran, and Methomyl
Exposure estimates for non-crop pesticide applications
EPA's BEs indicate that pesticides containing carbaryl, carbofuran, and methomyl are
registered for over 100 use sites (EPA 2003; EPA 2003; EPA 2004). As previously
indicated, many of the uses identified in the BEs have since been modified, or are
scheduled to be phased out or canceled. The BEs provided relatively few estimates of
exposure given the number and variety of uses authorized (Table 48). All modeled
exposure estimates assumed pesticide use in agricultural crops. No estimates were
provided for other uses of carbaryl, carbofuran, or methomyl.
Table 48. Examples of registered uses of carbaryl, carbofuran, methomyl and the
exposure method used by EPA in BEs.
a.i.
Carbaryl
Carbofuran
Examples of Registered Use
Crops: cranberries, cucumbers, beans, eggplant,
grapefruit, grapes, hay, lemons, lettuce, nectarines,
olives, onions, oranges, parley, peaches, peanuts,
pears, pecans, peppers, pistachios, plums, potatoes,
prunes, pumpkins, rice, sod, spinach, squash,
strawberries, sugar beets, sunflowers, sweet corn,
sweet potatoes, tangelos, tangerines, tomatoes,
walnuts, watermelons, and wheat.
Targeted pests: adult mosquitoes, ticks, fleas, fire
ants, and grasshoppers.
Other use sites: home and commercial lawns, flower
beds, around buildings, recreation areas, golf courses,
sod farms, parks, rights-of-way, hedgerows, Christmas
tree plantations, oyster beds, and rural shelter belts.
Crops: All uses will be canceled. Examples of current
uses include Alfalfa, artichoke, banana, barley, coffee,
field corn, sweet corn, pop corn, cotton, cucumber,
melons, squash, grapes, oats, pepper, plantain,
potato, sorghum, soybean, sugar beet, sugarcane,
sunflower, wheat, cotton, spinach grown for seed, and
tobacco.
Other use sites: All uses will be canceled. Examples
of current uses include agricultural fallow land,
ornamental and/or shade trees, ornamental
herbaceous plants, ornamental non-flowering plants,
pine, ornamental woody shrubs, and vines.
Exposure
Characterization in BE
PRZM-EXAMS
9 crops
No estimates provided
No estimates provided
PRZM-EXAMS
Estimates for 5 crops
No estimates provided
DRAFT
282
-------
a.i.
DRAFT
Examples of Registered Use
Exposure
Characterization in BE
Methomyl
Crops: alfalfa, anise, asparagus, barley, beans
(succulent and dry), beets, Bermuda grass (pasture),
blueberries, broccoli, broccoli raab, Brussels sprouts,
cabbage, carrot, cauliflower, celery, chicory, Chinese
broccoli, Chinese cabbage, collards (fresh market),
corn, cotton, cucumber, eggplant, endive, garlic,
horseradish, leafy green vegetables, lentils, lettuce
(head and leaf), lupine, melons, mint, oats, onions (dry
and green), peas, peppers, potato, pumpkin, radishes,
rye, sorghum, soybeans, spinach, strawberry, sugar
beet, summer squash, sweet potato, tomatillo, tomato,
wheat, and orchards including apple, avocado, grapes,
grapefruit, lemon, nectarines, oranges, peaches,
pomegranates, tangelo, and tangerine.
PRZM-EXAMS
Estimates for 4 crops
Other use sites: sod farms, bakeries, beverage plants,
broiler houses, canneries, commercial dumpsters
which are enclosed, commercial use sites
(unspecified), commissaries, dairies, dumpsters, fast
food establishments, feedlots, food processing
establishments, hog houses, kennel, livestock barns,
meat processing establishments, poultry houses,
poultry processing establishments, restaurants,
supermarkets, stables, and warehouses.
No estimates provided
Exposure estimates for crop applications
The BEs provide estimated environmental concentrations (EECs) for carbaryl,
carbofuran, and methomyl in surface water (Table 49). EECs were generated using the
PRZM-EXAMS model and used to estimate exposure of the three a.i.s to listed salmonids
and their prey (EPA 2004). PRZM-EXAMS generates pesticide concentrations for a
generic "farm pond". The pond is assumed to represent all aquatic habitats including
rivers, streams, off-channel habitats, estuaries, and near shore ocean environments. EPA
indicated that the PRZM-EXAMS scenarios provide "worst-case" estimates of salmonid
exposure and it "believes that the EECs from the farm pond model do represent first
order streams, such as those in headwaters areas" used by listed salmon (EPA 2003; EPA
2003; EPA 2004). However, listed salmonids use aquatic habitats with physical
characteristics that would be expected to yield higher pesticide concentrations than would
be predicted with the "farm pond" based model. Juvenile salmonids rely upon a variety
of non-main channel habitats that are critical to rearing. All listed salmonids use shallow,
low flow habitats at some point in their life cycle (Table 47). Below we discuss the
utility of the EECs for the current consultation. NMFS presents information that
DRAFT
283
-------
DRAFT
indicates the EECs do not represent worst-case environmental concentrations that listed
Pacific salmonids may be exposed to. Finally, NMFS provides additional modeling
estimates to evaluate potential exposure in vulnerable off-channel habitats used by
salmonids.
Table 49. PRZM-EXAMS exposure estimates from EPA's BEs.
Scenario:
crop, state
Application:
rate (Ibs a.i./A)/ method/
number of applications
Acute EEC
(MQ/L)
Chronic EEC
60-d average
(ng/L)1
Carbaryl
Sweet corn, OH
Field corn, OH
Apples, PA
Sugar Beets, MN
Citrus, FL
Peaches, CA
Citrus, CA
Tomatoes, CA
Apples, OR
Blackberries, OR
Snap beans, OR
2/aerial/8; 3.4/aerial/2
2/aerial/4; 1/aerial/2
2/aerial/5; 1.2/spray blast/2
1.5/aerial/2; 1.5/aerial/1
5/aerial/4; 3.4/aerial/2
7/aerial/2; 3.5/air blast/1
5/aerial/4; 3.4/aerial/2
2/aerial/4; 0.66/air blast/1
2/aerial/5; 1 .2/aerial/2
2/aerial/5; 1 .9/air blast/1
1 .5/aerial/4; 0.8/ground/1
53; 46
47; 13
31; 12
23; 7
153; 100
57; 14
20; 7
17; 2
19; 3
12; 8
10; 1.2
19; 13
14; 4
7; 2
6; 2
41; 23
12; 3
11;2
7; 1
6; 1
6; 3
1;0.3
Carbofuran
Alfalfa, CA
Alfalfa, PA
Cotton, MS
Grapes, CA
Potatoes, ME
Artichokes, CA
Cotton, CA
Potatoes, ID
1 /foliar/1
1 /foliar/1
1/in-furrow/1
10/soil surface/1
1/not reported/2
2/g round/1
1/in-furrow/1
2/foliar/1 ; 3/in-furrow/1 ;
6/chemigation/1
6
7.9
11
5.5
26
35
0.8
6.2; 0.2; 10.4
3.0
4.1
5.5
2.7
14
19
0.4
4.0; 0.1; 6.2
Methomyl
Lettuce
Sweet corn
Peaches
Cotton
0.9/aerial/10; 0.225/aerial/15
0.45/aerial/16
1 .8/aerial/3
0.6/aerial/3
88; 30
60
99
55
81; 26
54
85
47
The chronic values reported for methomyl are 56-day average concentrations rather than 60-day
average concentrations.
Utility of EECs for consultation
As described in the Approach to the Assessment section, our exposure analysis begins at
the organism (individual) level of biological organization. We consider the number, age
(or life stage), gender, and life histories of the individuals likely to be exposed. This
scale of assessment is essential as adverse effects to individuals may result in population
level consequences, particularly for populations of extremely low abundance.
Characterization of impacts to individuals provides necessary information to assess
DRAFT
284
-------
DRAFT
potential impacts to populations, and ultimately to the species. To assess risk to
individuals, we must consider the highest concentrations to which any individuals of the
population may be exposed. Several lines of evidence discussed below suggest that
EECs in the BEs may underestimate exposure of some listed organisms and designated
critical habitat.
Although EPA characterized these exposure estimates as "worst case" in the BEs, it has
also acknowledged that measured concentrations in the environment sometimes exceed
PRZM-EXAMs EECs (EPA 2007). EPA has subsequently clarified that rather than
providing worst case estimates, PRZM-EXAMS estimates are protective for the vast
majority of applications and aquatic habitats (EPA 2007). NMFS agrees that the model is
designed to produce estimates of exposure that are protective for a large number of
aquatic habitats.
Recent formal consultation and reviews of EPA informal consultations by the Services,
found that concentrations measured in surface water sometimes exceed peak
concentrations predicted with PRZM/EXAMS modeling (NMFS 2007; NMFS 2008;
USFWS 2008). Concentrations of carbaryl, carbofuran, and methomyl were generally
less than concentrations predicted in the BEs using PRZM-EXAMs although there were
examples where measured concentrations in surface water exceeded these estimates (e.g.,
(Hurlburt 1986; Creekman and Hurlburt 1987; Tufts 1989; Tufts 1990; Beyers, Farmer et
al. 1995). These findings demonstrate that the EECs generated using PRZM- EXAMs;
however, there were instances where measured concentrations in surface water exceeded
these estimates (e.g, (Hurlburt 1986; Creekman and Hurlburt 1987; Tufts 1989; Tufts
1990; Beyers, Farmer et al. 1995). These findings demonstrate that the EECs generated
using PRZM-EXAMS can underestimate peak concentrations that actually occur in some
aquatic habitats, and therefore, peak exposure experienced by some individuals of listed
species may be underestimated.
DRAFT 285
-------
DRAFT
Model assumptions and output suggest listed salmonid exposure to carbaryl, carbofuran,
and methomyl may underestimate or overestimate those concentrations predicted using
PRZM-EXAMS. Two assumptions are discussed below that show salmonids may be
exposed to higher concentrations than predicted with PRZM-EXAMS modeling:
Assumption 1: Model output are 90th percentile time-weighted averages It is
important to recognize that the model predicts concentrations based on site-specific
assumptions (e.g., rainfall) and that environmental concentrations provided for the
estimate do not represent the highest aquatic concentrations predicted given the
assumptions. Rather, the exposure estimates provided in the BEs are time-weighted
average concentrations for one day (i.e., peak), 21 d, and 60 d. Although, EPA refers to
the 1 d-averages as peak concentrations, they do not represent the maximum
concentration predicted but represent the average concentration over a 24-hour (h) period.
Additionally, the concentrations reported represent the upper 10th percentile of the
estimates derived using PRZM-EXAMS (Lin 1998). Although that is a relatively
concervative assumption, it suggests that 10% of time concentrations will be higher under
the given set of assumptions.
Assumption 2: Model inputs used the highest use rates and greatest number of
applications. NMFS compiled information on the maximum use rates permitted (single
and seasonal), number of applications allowed, and minimum application intervals
required through EPA product labeling (Tables 1-3).
Several of the PRZM-EXAMS scenarios did not match up well with current use
restrictions and did not account for maximum permitted application rates. For example,
PRZM-EXAMS scenarios for carbaryl evaluated a maximum single application rate of
3.5 Ib a.i./acre versus higher single application rates authorized by EPA for oyster beds,
home lawn, golf course, and sod farms (8-9.1 Ib a.i./acre), nut trees (5 Ibs a.i./acre), and
citrus (7.5 - 16 Ibs a.i./acre). Some of the simulations included multiple applications that
exceed single application rates. For example, there were simulations for 4 applications of
5 Ibs a.i./acre in citrus and 2 applications of 7 Ibs a.i./acre in peaches. However, these
DRAFT 286
-------
DRAFT
scenarios do not cover the maximum rate of carbaryl allowed per applications (16 Ibs
a.i./acre in California citrus) or annually (24 Ibs a.i./acre/year in apple, pear, crabapple,
and others). The maximum single application rate for methomyl is 0.9 Ibs a.i./acre. One
scenario assumed 3 applications of 1.8 Ibs a.i./acre (California peaches). This scenario
may overestimate aquatic concentrations for single applications at comparable use sites.
However, considering multiple applications, the maximum application rate of methomyl
evaluated in the BE was 9 Ibs a.i./acre (lettuce) versus allowable application rates of 10-
32 Ibs a.i./acre/year for alfalfa, broccoli, cabbage, corn, lettuce, onions, and spinach (EPA
2007).
Few crop scenarios were assessed relative to the number of approved uses. The
BEs provided pesticide exposure estimates from uses in relatively few crops considering
the number of registered uses of carbofuran, carbaryl, and methomyl. For example,
estimates of carbaryl exposure were provided for nine agricultural crops. An evaluation
of currently registered uses of a single carbaryl product label (Sevin brand XLR plus
carbaryl insecticide) revealed the product can be applied to more than 75 agricultural
crops in California alone (CDPR 2009). Similarly, the product Furadan 4-F Insecticide -
Nematicide (carbofuran) can be applied to more than 20 use sites in California while
exposure estimates evaluated only 5 agricultural crops. Finally, DuPont Lannate SP
Insecticides (methomyl) is authorized for use on more than 80 agricultural crops in
California while the BE provided exposure estimates for only 4 crops for all methomyl
products. There are logistic considerations that limit the number of scenarios that can be
evaluated. However, information to suggest that the simulations would be representative
of other registered uses was not included in the BEs.
Crop scenarios are likely not representative of the entire action area. The
regional scale that the modeled scenarios are intended to represent is unclear. Many of
the scenarios were conducted for states outside the distribution of listed salmonids. The
methomyl BE did not provide information on geographic locations simulated (e.g.,
county, state, region, etc.). The assumed rainfall and other site-specific input
assumptions can have large impacts on predicted exposure. For example, the carbofuran
DRAFT 287
-------
DRAFT
BE provides a peak EEC of 5.5 (ig/L in water following an application rate of 10 Ibs
a.i./acre. Yet three simulations conducted at a 10-fold lower application rate produced
greater estimates of exposure due to differences in site-specific assumptions (Table 49).
NMFS also questions whether input assumptions were adequate to represent the range in
variability among sites throughout the action area. Site-specific meteorological and soil
conditions vary greatly throughout the four states where listed salmonids are distributed
and crops are grown. The BEs did not indicate site-specific input assumptions of each
scenario nor did they put these assumptions into perspective with regard to the range of
conditions throughout the four states. This makes it difficult to determine the
representativeness of scenario estimates for the complete range of crop uses.
Crop scenarios do not consider application of more than one pesticide. The
pesticide labels NMFS reviewed had few restrictions regarding the co-application (i.e.,
tank mixture applications) or sequential applications of other pesticide products
containing different a.i.s. Also, there were few restrictions for those pesticides containing
ingredients that share a common mode of action (e.g., cholinesterase-inhibiting
insecticides). For example, we saw no restrictions that would prevent either co-
application or sequential application of products containing carbofuran, carbaryl, and
methomyl. Examples offish kill incidents discussed in the Risk Characterization section
of the Opinion indicate combinations of cholinesterase-inhibiting insecticides are
sometimes applied on the same day or over a short interval, increasing the likelihood of
salmonid exposure to chemical mixtures that may have additive or synergistic effects.
Some labels encourage the use of more than one product. The Sevin Brand XLR Plus
Carbaryl Insecticide (EPA registration No. 264-333) advises that 8 applications of
carbaryl at 3 d intervals "may not provide adequate levels of protection under conditions
of rapid growth or severe pest pressure. The use of an alternative product should be
considered in conjunction with this product." Multiple applications of pesticides increase
the likelihood of cumulative exposure. We considered cumulative exposure based on
generated 60 d time-weighted average concentrations to simulate situations where
pesticide products containing the three a.i.s were applied at separate times during the
growing season (Table 49). To address potential variability between sites, we generated
DRAFT 288
-------
DRAFT
exposure values for a few labeled uses using the GENEEC model, which is intended to
provide screening estimates over large geographic regions (Table 50)1. The input
parameters used were consistent with recent EPA model inputs (EPA 2006; EPA 2007;
EPA 2007).
Table 50. GENEEC estimated concentrations of carbaryl, carbofuran, and methomyl in
surface water adjacent to sweet corn, potatoes, sweet corn, potatoes, and citrus.
Chemical use
Foliar/ ground
application
Rate
Ibs/
acre
No
Interval
days
Buffer
ft
EEC (ug/L)
24-h
avg
4-d
avg
21 -d
avg
60-d
avg
90-d
avg
Sweet Corn
Carbaryl
Carbofuran
Methomyl
2
0.5
0.45
0.225
8
2
14
28
3
7
1
1
0
0
25
25
397
56
307
288
390
56
276
259
348
55
163
153
272
54
72
68
229
53
49
46
Potatoes
Carbaryl
Carbofuran
Methomyl
2
1
0.9
6
2
5
7
5
5
0
0
25
236
112
212
232
112
191
207
111
113
162
108
50
137
106
34
Citrus
Carbaryl
Carbofuran
Methomyl
12
-
0.9
1
-
3
-
-
5
0
-
25
485
-
133
476
-
120
424
-
71
332
-
31
280
-
21
Number of applications
-Not applicable
The EECs for EPA's effect determinations were derived primarily using the PRZM-
EXAMS model. This model predicts runoff to a "farm pond" based on application
specifications (rate and method), properties of the a.i. (solubility, soil adsorption
coefficient, soil metabolisms rate, etc.,), assumed meteorological conditions (amount of
rainfall), and other site-specific assumptions [soil type, slope, etc., (EPA 2004)]. The
farm pond scenario is likely a poor surrogate of certain habitats used by salmonids.
In particular, listed salmonids rely extensively upon a variety of non-main channel
habitats that would be expected to yield higher pesticide concentrations than would be
predicted with the "farm pond" based PRZM/EXAMS model. Examples of off-channel
1 EPA characterizes GENEEC as a tier-1 screening model (EPA 2004c). GENEEC is a meta-model of the
PRZM-EXAMS model that incorporates assumptions that are intended to model exposure estimates on a
site vulnerable to runoff. The size of the treated area and aquatic habitat (farm pond) are the same as
described for PRZM-EXAMS.
DRAFT
289
-------
DRAFT
habitats include alcoves, channel edge sloughs, overflow channels, backwaters, terrace
tributaries, off-channel dredge ponds, off channel ponds, and braids (Swift III 1979;
Anderson 1999). Diverse, abundant communities of invertebrates (many of which are
salmonid prey items) also populate these habitats and, in part, are responsible for juvenile
salmonids reliance on off-channel habitats. Juvenile coho salmon, stream-type Chinook
salmon, and steelhead use off-channel habitats for extended durations (several months).
Although these habitats typically vary in surface area, volume, and flow, they are
frequently shallow, low to no-flow systems protected from a river's or a stream's primary
flow. Thus, rearing and migrating juvenile salmonids use these habitats extensively
(Caffrey 1996; Beechie and Bolton 1999; Montgomery 1999; Roni 2002; Opperman and
Merenlender 2004; Beechie, Liermann et al. 2005; Morley, Garcia et al. 2005; Henning
2006).
Small streams and some off-channel habitats represent examples of habitats used by
salmonids that can have a lower capacity to dilute pesticide inputs than the farm pond.
The PRZM-EXAM estimates assume that a 10-hectare (approximately 25 acres) drainage
area is treated and the aquatic habitat is assumed to be static (no inflow or outflow).
Pesticide treatment areas of 10-hectares and larger occur frequently in agricultural crops,
particularly under pest eradication programs. Additionally, aquatic habitats used by
salmon vary in volume and recharge rates and consequently have different dilution
capacities to spray drift and runoff events. The assumed drainage area to water volume
ratio (100,000 m2:20,000 m3) is easily exceeded for small water bodies. For example, a
one acre pond with an average depth of 1 m would exceed this ratio for treated drainage
areas of approximately five acres in size and larger. The assumed aquatic habitat and size
of the treated area for the PRZM-EXAMS scenarios suggest that exposure is
underestimated for listed salmonids that utilize relatively small aquatic habitats with low
dilution capacities.
DRAFT 290
-------
DRAFT
NMFS estimates of potential exposure in shallow water habitats used bysalmonids
Direct over-spray
To estimate potential exposure of salmon to pesticides in off-channel and other shallow-water
habitats we first determined the initial average concentrations that will result from a direct
overspray of shallow surface water. NMFS is unaware of any circumstances where EPA has
authorized direct application of methomyl to aquatic habitats. However, both carbaryl and
carbofuran can be applied to aquatic habitats. For example, carbaryl is registered for use in rice
crops to control tadpole shrimp and other rice pests. The resulting concentrations in surface
water are a function of the amount applied and the volume of the water body when pesticides are
applied directly to aquatic habitats (Table 51). Carbaryl can be applied twice in rice at a rate of
1.5 Ibs a.i./acre. A single application at that rate would result in an average initial carbaryl
concentration of 1,682 |j,g/L in 10 cm of water. Specimen labels for carbaryl do not place any
restrictions on rice paddy discharges, yet they warn "discharge from rice fields may kill aquatic
and estuarine invertebrates (EPA Reg. No. 264-333 and 264-349)." The BE indicates
carbofuran can also be applied in rice, at a rate of 0.5 Ibs a.i./acre. Assuming this rate in 10 cm
of water results in an average initial average concentration of 561 [ig/L.
Carbaryl is registered for use at a rate of 2 Ibs a.i./acre in cranberries. Five applications can be
made at 7 d intervals. The label for Sevin brand XLR Plus warns that the "use in cranberries
may kill shrimp and crabs. Do not use in areas where these are important resources (EPA Reg.
No. 264-333)." However, there are no restrictions regarding applications to standing water, or
label requirements to ensure bog water does not contaminate other surface waters. A single
application of 2 Ibs a.i./acre to water 10 cm of water results in an initial average concentrations
of 2,242
Carbaryl can also be applied at 8 Ibs a.i./acre in estuarine areas in Washington state to kill ghost
shrimp and mud shrimp in commercial oyster beds. Application of 8 Ibs of carbaryl to a static
water body of 10 cm deep would result in an initial average concentration of approximately 9
mg/L (8,968 |J,g/L). This estimate does not consider tidal influences. The 24(c) label for the use
of carbaryl on oyster beds does not specify when applications are to be made with relation to
DRAFT 291
-------
DRAFT
incoming or outgoing tides. The label does specify applicators must obtain an NPDES permit for
this use. A current NPDES permit specifies that applications to these tidal areas are made when
the beds are uncovered by the outgoing tide. Peak concentrations in the water column are
expected as the treated area is re-flooded by the incoming tide. These concentrations are
expected to decrease as the volume of water over the treated area increases. The area
contaminated by the carbaryl application will increase beyond the treatment site, and water
column concentrations will simultaneously decrease due to transport and dilution from the
incoming tide. This use is discussed in more detail in the Risk Characterization portion of the
Opinion (see Field studies in ESA-listed salmonid habitat).
Table 51. Average initial concentration of any a.i. in surface water resulting from a direct
overspray of aquatic habitat.
Application Rate
(Ibs a.i. /acre)
0.25
0.5
1
3
10
0.25
0.5
1
3
10
0.25
0.5
1
3
10
0.25
0.5
1
3
10
0.25
0.5
1
3
10
Water Depth
(meters)
2
2
2
2
2
1
1
1
1
1
0.5
0.5
0.5
0.5
0.5
0.3
0.3
0.3
0.3
0.3
0.1
0.1
0.1
0.1
0.1
A.i. Concentration in Surface
Water
(MQ/L)
14
28
56
168
560
28
56
112
336
1121
56
112
224
673
2242
93
187
374
1121
3736
280
560
1121
3363
11208
DRAFT
292
-------
DRAFT
Pesticide drift
We also provide estimated pesticide concentrations in shallow off-channel habitats associated
with drift from terrestrial applications of pesticides (Table 52). These estimates were derived
using the AgDrift model and estimate downwind deposition from pesticide drift (Teske 2001).
Additional deposition from runoff was not considered. The drift estimates derived represent
mean projected drift. Although AgDrift adequately predicts drift, its field validations studies and
other research show drift is highly variable and influenced by site-specific conditions and
application equipment (Bird, Perry et al. 2002). No-spray buffer zones (or setbacks) may
significantly reduce pesticide exposure to salmonids by reducing runoff and drift inputs.
The methomyl RED requires label statements for methomyl products that specify not to "apply
by ground equipment within 25 ft, or by air within 100 ft of lakes, reservoirs, rivers, estuaries,
commercial fish ponds and natural, permanent streams, marshes or natural, permanent ponds.
Increase the buffer zone to 450 ft from the above aquatic areas when ultra low volume
application is made (EPA 1998)." This label requirement appears to address many, but not all
aquatic habitats used by listed salmonids. For example, labels specify buffers to natural and
permanent streams, marshes, and ponds but there is no mention of important intermittent aquatic
habitats or manmade watercourses such as floodplain restoration sites and irrigation systems that
either contain listed species or drain to such habitats.
Specific buffer zones that correspond to current label requirements for carbaryl, carbofuran, and
methomyl were assessed (Table 52). No buffers to aquatic habitats are required by EPA for
carbaryl or carbofuran products. However, we did consider interim "voluntary" buffers
recommended by CDPR to protect federally listed endangered species
(http://www.cdpr.ca.gov/docs/endspec/prescint.htm). Our simulations assumed the off-channel
habitat had a downwind width of 10 m. Pesticide concentrations were predicted for habitats that
ranged in depths from 0.1 to 2 m. These dimensions were assumed based on research of
salmonid use of off-channel habitats (Montgomery 1999; Roni 2002; Beechie, Liermann et al.
2005; Morley, Garcia et al. 2005; Henning 2006). Average initial concentration estimates
derived from the simulations ranged from 0.2-447 [ig/L for each Ib of a.i. applied. These
simulations indicate that applications of several Ibs a.i. per acre adjacent to some off-channel
DRAFT 293
-------
DRAFT
habitats could result in aquatic concentrations exceeding 1 mg/L, a value that would result in
substantial toxicity to aquatic life, including deaths of exposed salmonids.
Maximum rates of carbaryl permitted for most vegetable crops range from 1-2 Ibs a.i./acre.
Several tree crops allow much higher application rates (3-8 Ibs a.i./acre), with a maximum single
application rate of 16 Ibs a.i./acre approved for use in California on citrus crops. Carbaryl may
be applied by ground boom, chemigation, spray blast, and aerial spray applications. Considering
application rates, methods, and no requirements for buffers to aquatic habitat, the estimated
initial concentrations of carbaryl in surface waters range from 3 (ig/L to 7 mg/L in the modeled
habitats. Voluntary buffers recommended by CDPR for protection of federally listed fish would
considerably reduce the deposition. For example, drift from aerial applications with the 600 ft
buffer to a 10 m wide stream are predicted to be approximately equivalent to 2-6% of the applied
rate versus drift equivalent to 30-39% of the applied rate with no buffer. Additionally, drift from
ground application with the 120 ft buffer predict drift equivalent to 0.3 - 2% of the applied rate
versus 7-23% of the applied rate predicted with no buffer.
Current carbofuran labels reviewed by NMFS specify application rates of 1 Ib a.i./acre or less for
most crops. Carbofuran can be applied by both ground and aerial methods. Simulations
assuming no aquatic buffers at these rates provide estimates of initial average carbofuran
concentrations from the low (ig/L range to several hundred ng/L. The carbofuran BE indicates
several 24(c) uses have previously allowed application of carbofuran at rates as high as 10 Ibs
a.i./acre. Simulations indicate that application rates in this range, with no buffer restrictions,
would result in initial surface water concentrations of 2 |j,g/L to over 4,000
The maximum single application rate for methomyl is 0.9 Ibs a.i./acre. Considering label-
required buffer zones, simulations at the maximum labeled application rate predict initial average
concentrations of 1-57 [ig/L for ground applications and 4-84 [ig/L for aerial applications.
Incorporating voluntary buffers listed by CDPR predict concentrations ranging from 0.2 -
58
DRAFT 294
-------
DRAFT
Table 52. Average initial pesti
pesticide applied based on Ag
Depth of aquatic habitat
(meters)
cide concentration in 10 m wide off-channel habitat per Ib of
Drift simulations.
Buffer to Aquatic Habitat (ft)
Average Initial
Concentration in Surface
Water (iag/L)
Aerial Applications, EPA default (ASAE fine-medium droplet size distribution)
2
1
0.5
0.1
2
1
0.5
0.1
2
1
0.5
0.1
O1
O1
O1
O1
100"
100"
100"
100"
600J
600J
600J
600J
17
34
67
333
5
9
19
93
1
2
4
18
Aerial Applications, (ASAE very fine - fine droplet distribution)
2
1
0.5
0.1
2
1
0.5
0.1
2
1
0.5
0.1
O1
O1
O1
O1
450"
450"
450"
450"
600J
600J
600J
600J
22
45
89
447
4
8
16
79
3
6
13
64
Air Blast Applications, Dormant Spray
2
1
0.5
0.1
2
1
0.5
0.1
2
1
0.5
0.1
O1
O1
O1
O1
25"
25"
25"
25"
120J
120J
120J
120J
11
21
43
214
3
6
12
62
0.3
1
1
6
Ground Application, Low Boom, ASAE very fine-fine distribution, 50th percentile
2
1
0.5
0.1
2
1
0.5
O1
O1
O1
O1
25"
25"
25"
4
8
15
76
1
1
2
DRAFT
295
-------
DRAFT
Depth of aquatic habitat
(meters)
0.1
2
1
0.5
0.1
Buffer to Aquatic Habitat (ft)
25'
120J
120J
120J
120J
Average Initial
Concentration in Surface
Water (iag/L)
11
0.2
0.4
1
4
Ground Application, High Boom, ASAE very fine-fine distribution, 50m percentile
2
1
0.5
0.1
2
1
0.5
0.1
2
1
0.5
0.1
O1
O1
O1
O1
25'
25'
25'
25'
120J
120J
120J
120J
13
26
52
261
3
6
13
64
1
2
3
17
Carbaryl and carbofuran labels do not require buffers to aquatic habitats.
2Bufferto aquatic habitats specified on current methomyl product labels, as required by the 1998
methomyl RED.
Voluntary "interim bulletin" measures developed by California Department of Pesticide Regulation to
protect threatened and endangered species.
Monitoring Data: Measured Concentrations of Carbaryl Carbofuran, Methomyl
1-Napthol and 3-Hydroxyfuran
Data Described in USEPA's Biological Evaluations
The BE for carbofuran (EPA 2004) summarized national scale surface water monitoring data
available from the USGS NAWQA program and also provided state level summaries based on
NAWQA and CDPR data . It should be noted that some USGS data is also reported in the
CDPR data base. National scale NAWQA data (1991-2001) was presented by land use category,
and included information on reporting limits (0.02 ng/L), frequency of detections, and maximum
concentrations. Maximum concentrations reported for the various land categories were:
agricultural land (7 ng/L), mixed land use (0.678 ng/L), urban land use (0.034 ng/L), and
undeveloped land use (0.034 ng/L). An additional table summarizing NAWQA data (1984-
2004) by state (California, Idaho, Oregon, and Washington) reported a maximum concentration
of 32.2 ng/L. This concentration occurred in surface water sampled from Zollner Creek, near
DRAFT
296
-------
DRAFT
Mt. Angel, Oregon on April 17, 2002, and was also the maximum residue found in the database
on a national scale. CDPR data (1990-2003) were presented by county. Zollner Creek is a
tributary to the Willamette River and within the distribution of threatened Upper Willamette
River Chinook salmon and Upper Willamette River steelhead. The maximum concentration
measured in California was 5.5 |J,g/L, in Imperial County. Date and specific location were not
provided.
The BEs for carbaryl (EPA 2003) and methomyl (EPA 2003) provided less detailed information
about monitoring data on a national basis. The BE for carbaryl provided a national maximum
reported concentration of 5.5 |J,g/L, and noted that carbaryl was the second most commonly
detected insecticide in the NAWQA database. It also noted there were more frequent detections
and higher concentrations in urban streams than in streams draining agricultural or mixed land
use areas. The BE for methomyl reported that few national level monitoring data were available,
and reported data located in EPA's STORET database. The highest reported concentration
nationally was 1 (ig/L (in Texas).
Table 53. National Maximum Concentrations of Carbaryl, Carbofuran, and Methomyl as Reported
in EPA BEs (EPA 2003a, EPA 2003b, and EPA 2004)
Pesticide
Carbaryl
Carbofuran
Methomyl
1-Napthol
Maximum
Concentration (M-g/L)
5.5
32.2
1.0
Data Source
USGS NAWQA
USGS NAWQA
EPA STORET
Years Reported
1991-1998
1984-2004
Not Reported
Not Reported
USGS NAWQA Data for California, Idaho, Oregon, and Washington
We obtained updated data from the USGS NAWQA database to evaluate the occurrence of
carbofuran, carbaryl, methomyl, and 1-napthol (a degradate of carbaryl) in surface waters
monitored in California, Idaho, Oregon, and Washington. The database query resulted in
approximately 5,000 samples in which one or more of the compounds were detected. The
percentage of samples containing the chemicals of interest in comparison to all samples
evaluated by USGS could not be determined from the data query. Approximately 350 unique
sampling locations had detectable concentrations of one or more of the compounds. Available
DRAFT 297
-------
DRAFT
data covered a range of 15 years, from 1992-2007. Land uses associated with the sampling
stations included agriculture, forest, rangeland, urban, and mixed use. Some stations occurred
only once in the data set, others appeared multiple times across a span of years. The frequency
of detection is a combination of the actual occurrence of pesticides in the water and the sampling
intensity. Values reported are for a filtered water sample (0.7 micron glass fiber filter),
representing the dissolved phase. Chemicals transported primarily in the particulate phase would
be underreported in this data set. No sediment or tissue data were available from USGS for these
compounds. Because the USGS monitoring program does not generally coordinate sampling
efforts with specific pesticide applications or runoff events, detected concentrations are likely to
be lower than actual peak concentrations that occur.
Summary information for carbaryl, carbofuran, methomyl, and 1-napthol is reported below
(Table 54). In many instances, the pesticides could be detected, but were below the analytical
limit of quantification (LOQ), and concentrations were reported as less than ("<") the LOQ for
that sample. Ranges of LOQ for each pesticide are reported, as analytical methods varied.
Summary statistics were calculated on samples not designated as ("<")• Many of the
concentrations that could be quantified were designated as "E," meaning the concentrations were
estimated. These data are included in the summary statistics.
Carbaryl was the most commonly occurring pesticide, detected in 99.7% of the samples in the
data set, and quantifiable in 69.7% of them. Carbofuran was detected in nearly as many (91.8%),
but was quantifiable in significantly fewer (7.8%). Methomyl and 1-napthol were detected less
frequently (26.2% and 25.4%, respectively), and concentrations were quantifiable in only a small
number of those instances (2.3% and 8.4%, respectively). Quantifiable concentrations ranged
from 0.0001 - 33.5 ng/L for carbaryl, 0.0015 - 32.2 ng/L for carbofuran, 0.0039 - 0.8222
for methomyl, and 0.0007 - 1.6 (J,g/L for 1-napthol.
DRAFT 298
-------
DRAFT
Table 54. Summary of Occurrences of Carbaryl, Carbofuran, Methomyl, and 1-Napthol in USGS
NAWQA Database (1992-2007) for California, Idaho, Oregon, and Washington
Statistic
Detects in data set
Samples with detects
Detects above LOQ
LOQ range (ng/L)
Minimum concentration (M-g/L)
Maximum concentration (ng/L)
Arithmetic mean concentration (ng/L)
Standard deviation (M-g/L)
Median concentration
Carbaryl
99.7%
4938
69.7%
0.0005-5.2
0.0001
33.5
0.0943
1.11
0.0314
Carbofuran
91 .8%
4545
7.8%
0.002-0.150
0.0015
32.2
0.3846
2.13
0.0318
Methomyl
26.2%
1297
2.3%
0.0044-1 .22
0.0039
0.8222
0.1171
0.19
0.0440
1-Napthol
25.4%
1256
8.4%
0.007-0.09
0.0007
1.6
0.0258
0.15
0.0072
Monitoring Data from California Department of Pesticide Regulation
We evaluated monitoring data available from the CDPR, which maintains a public database of
pesticide monitoring data for surface waters in California (CDPR 2008). Data were available for
carbaryl, Carbofuran, methomyl and 3-hydroxycarborfuran (a degradate of Carbofuran), but not 1-
napthol. Data in the database (http://www.cdpr.ca.gov/docs/emon/surfwtr/surfdata.htm) are from
multiple sources, including monitoring conducted by CDPR, USGS (data from the NAWQA
program as well as other studies); state, city, and county water resource agencies; and some non-
governmental or inter-governmental groups such as Deltakeeper. The CDPR requires a formal
QA/QC protocol for data submitted or does a separate QA/QC review, thus only data subject to
appropriate QA/QC procedures are included in the surface water database. Unlike the USGS
NAWQA data set, the CDPR database may contain whole water samples as well as filtered
samples. If whole water concentrations are reported for compounds that sorb significantly to the
particulate phase, concentrations would appear higher than in a filtered sample, which represents
only the dissolved phase. The majority of the studies, which are described in metadata available
from CDPR, are not targeted at correlating water concentrations with specific application
practices, with the exception of some studies evaluating rice pesticides.
Summary information for carbaryl, Carbofuran, methomyl, and 3-hydroxycarbofuran is reported
below (Table 55). No monitoring for 1-napthol was reported in the database. The database, last
updated in June 2008, consists of approximately 270,000 data records. Each record reports a
specific sampling site, date, and analyte. The number of records associated with a particular
DRAFT
299
-------
DRAFT
compound is indicative of monitoring intensity rather than actual occurrence in surface waters.
In this database, detections below the LOQ are reported as 0 ng/L. Summary statistics were
calculated on samples with values above the LOQ.
Carbaryl (5,265 records) and carbofuran (5,864 records) were the subject of greater monitoring
intensity than either methomyl (1,982 records) or 3-hydroxycarbofuran (1,149 records).
Quantifiable amounts of methomyl appeared in 11.0% of the samples. Carbaryl (3.6%) and
carbofuran (5.9%) were quantifiable less often, and 3-hydroxycarbofuran (0.3%) was only rarely
quantifiable. Quantifiable concentrations ranged from 0.0030 - 8.4 [ig/L for carbaryl, 0.0066 -
5.2 |ig/L for carbofuran, 0.0150 - 5.4 ng/L for methomyl, and 0.0600 - 0.1800 ng/L for 3-
hydroxyfuran.
Table 55. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, and 3-Hydroxycarbofuran
in CDPR Database (1991-2006)
Statistic
Samples in data set
Quantifiable samples
% of quantifications in data set
LOQ range (|ig/L)
Minimum concentration (|ig/L)
Maximum concentration (|ig/L)
Arithmetic mean concentration (|ig/L)
Standard deviation (|ig/L)
Median concentration
Carbaryl
5265
191
3.6%
0.003-1.0
0.0030
8.4
0.3733
0.8999
0.1200
Carbofuran
5864
347
5.9%
0.002
0.0066
5.2
0.3197
0.5521
0.1100
Methomyl
1982
218
11.0%
0.002-1.0
0.0150
5.4
0.2984
0.5223
0.1550
3-Hydroxycarbofuran
1149
3
0.3%
0.0058-0.69
0.0600
0.18
0.1367
0.0666
0.1700
Monitoring Data from Washington State
Data from -30 pesticide monitoring studies conducted in the state of Washington are included in
Department of Ecology Environmental Information Management (ELM) database
(http://www.ecy.wa.gov/eim/). Data in the database are from multiple sources, including state
agencies, and may contain whole water samples as well as filtered samples. If whole water
concentrations are reported for compounds that sorb significantly to the particulate phase,
concentrations would appear higher than in a filtered sample, which represents only the dissolved
phase. The EIM requires a formal QA/QC protocol for data submitted contains or does a
separate QA/QC review, thus only data subject to appropriate QA/QC procedures are included.
Some of the studies contained in this database are targeted with respect to specific pesticide uses,
while others are more generalized water quality surveys. Results from a database query on
DRAFT 300
-------
DRAFT
carbaryl, carbofuran, methomyl, 1-napthol, and 3-hydroxycarbofuran, conducted by NMFS, are
provided below (Table 56). Maximum concentrations for the three a.i.s and the two degradates
occurred in watersheds used by listed Pacific salmonids.
Included in the EIM database are monitoring efforts conducted jointly by the Washington
Departments of Agriculture and of Ecology in some salmon-bearing streams. Final reports for
2003-2007 seasons are publically available on their website
(http://agr.wa.gov/PestFert/natresources/SWM/default.htm). A separate summary of data from
those investigations is provided below. Water samples are not filtered, and thus concentrations
reported include pesticides in both dissolved and particulate phases, although the sampling
protocol specifies an attempt to avoid collection of excessive particulates (Johnson and Cowles
2003). Whole water concentrations for compounds that sorb significantly to the particulate
phase will appear higher than those for a filtered sample, which represents only the dissolved
phase.
The procedure for reporting in the EIM database includes reporting non-detects as the reporting
limit for that particular sample, and adding a "U" data qualifier. The reporting limit was not
specified in the data accessed by NMFS, thus LOQ ranges (Table 56) were estimated based on
"U"-qualified data. Summary statistics were calculated on samples with values above the LOQ
(i.e., not qualified with a "U").
Summary information for carbaryl, carbofuran, methomyl, 1-napthol, and 3-hydroxycarbofuran
based on all data in the database is reported below (Table 56). Sampling intensity was
approximately the same for the three parent compounds considered in this Opinion (-1,400
samples each), with slightly fewer sampling events for the degradates. Based on the method of
reporting in the database, we cannot determine how many samples may have contained detection
of any of the compounds that were below LOQs. Carbaryl appeared most often in the samples
(8.5%), followed by carbofuran (4.0%), and methomyl (3.5%). The degradates were detected
slightly less often when they were analytes, with 1-napthol quantifiable in 1.8% of the samples,
and 3-hydroxycarbofuran quantifiable in 3.3%. Quantifiable concentrations ranged from 0.002 -
DRAFT 301
-------
DRAFT
10.0 ng/L for carbaryl, 0.01 - 2.4 ng/L for carbofuran, 0.0150 - 0.17 ng/L for methomyl, 0.01-
0.64 for 1-napthol, and 0.05 - 0.42 |j,g/L for 3-hydroxycarbofuran.
Table 56. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, 1-Napthol and 3-
Hydroxycarbofuran in Surface Water - Washington EIM Database (1988-2007)
Statistic
Samples in data set
Quantifiable samples
% of quantifications in data set
LOQ range (ng/L)
Minimum concentration (ng/L)
Maximum concentration (|ig/L)
Arithmetic mean concentration
(ng/L)
Standard deviation (ng/L)
Median concentration
Carbaryl
1477
126
8.5%
0.002-
25.0
0.0020
10.0
0.27
0.92
0.05
Carbofuran
1410
57
4.0%
0.01-25.0
0.0100
2.3
0.34
0.38
0.16
Methomyl
1402
49
3.5%
0.01-5.0
0.0150
0.17
0.08
0.04
0.07
1-Napthol
1173
23
2.0%
0.01-5.0
0.0100
0.64
0.08
0.13
0.05
3-Hydroxy
carbofuran
1396
46
3.3%
0.01-10.0
0.0500
0.42
0.10
0.06
0.12
Summary information for carbaryl, carbofuran, methomyl, 1-napthol, and 3-hydroxycarbofuran
based on the recent studies conducted by Washington Departments of Agriculture and Ecology
are presented below (Table 57). All sampling was done in streams that contain listed Pacific
salmonids. These data are a subset of the data listed in Table 56. Sampling intensity was
approximately the same for the three parent compounds considered in this Opinion (-1,200
samples each), with slightly fewer sampling events for the degradates. Based on the method of
reporting in the database, we cannot determine how many samples may have contained detection
of any of the compounds that were below LOQs. Carbaryl appeared most often in the samples
(4.3%), followed by 1-napthol (1.0%), and methomyl (0.7%). Carbofuran and 3-
hydroxycarbofuran appeared less often (0.2% and 0.2% respectively). Quantifiable
concentrations ranged from 0.01 - 10.0 [ig/L for carbaryl, 0.028 - 0.16 [ig/L for carbofuran,
0.015 - 0.17 ng/L for methomyl, 0.01 - 0.641 for 1-napthol, and 0.095 - 0.15 ng/L for 3-
hydroxyfuran.
DRAFT
302
-------
DRAFT
Table 57. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, 1-Napthol and 3-
Hydroxycarbofuran in recent studies by Washington Department of Ecology (2003-2007)
Statistic
Samples in data set
Quantifiable samples
% of quantifications in data set
LOQ range (ng/L)
Minimum concentration (ng/L)
Maximum concentration (|ig/L)
Arithmetic mean concentration
(ng/L)
Standard deviation (ng/L)
Median concentration
Carbaryl
1,223
52
4.3%
0.01-0.25
0.0100
10.0
0.30
1.40
0.03
Carbofuran
1,208
3
0.2%
0.01-0.25
0.0280
0.16
0.09
0.07
0.08
Methomyl
1,207
9
0.7%
0.01-0.25
0.015
0.17
0.043
0.050
0.019
1-Napthol
1,003
10
1.0%
0.03-0.25
0.01
0.641
0.1338
0.189
0.078
3-Hydroxy
carbofuran
102
2
0.2%
0.02-0.25
0.0950
0.15
0.12
0.04
0.12
Three studies in the EIM database reported sediment concentrations. One was conducted to
determine the concentrations of carbaryl and 1-napthol in the sediment of Willapa Bay, and two
were sediment toxicity assessments. The sediment toxicity assessments, which provided the
reported values (Table 58) for carbofuran, methomyl, and 3-hy doxy carbofuran as well as some
data on carbaryl and 1-napthol, had an LOQ of 100 ng/kg. The LOQs for the Willapa Bay
analysis were lower, ranging from 22-37 ng/kg. Neither of the sediment toxicity studies reported
detectable concentration of any of the chemicals. In the Willapa Bay study, both carbaryl (range
29-3,400 ng/kg) and 1-napthol (range 34-280 ng/kg) were detected. These data represent a very
small subset of potentially contaminated sediments, and the negative results should not
necessarily be interpreted to mean that these chemicals are not in the sediment.
Table 58. Summary of Occurrences of Carbofuran, Carbaryl, Methomyl, 1-Napthol and 3-
Hydroxycarbofuran in Sediments in Washington EIM Database (1988-2007)
Statistic
Samples in data set
Quantifiable samples
% of quantifications in data set
LOQ range (jig/kg)
Minimum concentration (ng/kg)
Maximum concentration (jig/kg)
Arithmetic mean concentration
(i-ig/kg)
Standard deviation (ng/kg)
Median concentration (jig/kg)
Carbaryl
26
17
65.32%
22-160
29
3,400
762
1,166
200
Carbofuran
26
0
0%
100
NA
NA
NA
NA
NA
Methomyl
26
0
0%
100
NA
NA
NA
NA
NA
1-Napthol
69
9
13.0%
22-58
34
280
122
75
120
3-Hydroxy
carbofuran
26
0
0%
100
NA
NA
NA
NA
NA
DRAFT
303
-------
DRAFT
Summary of National and State Monitoring Databases
Overall, data from the three sets of monitoring data examined by NMFS is relatively consistent,
with carbaryl generally being the most frequently quantifiable parent compound. No state
monitoring data were available from Oregon or Idaho. Carbaryl and carbofuran were measured
in concentrations ranging from 0.0001-33.5 |J,g/L. Methomyl generally was measured at slightly
lower concentrations, ranging from 0.004-5.4 |J,g/L. Both 1-napthol and 3-hydroxycarbofuran
were detected in slightly lower concentrations, ranging from 0.0007-0.64 |J,g/L, than any of the
parent compounds. Water concentrations based on these monitoring data, as anticipated, were
generally lower than those based on modeling estimates.
Targeted Monitoring Studies
Numerous studies have been conducted in coordination with applications of carbaryl in
Washington State to control burrowing shrimp in commercial oyster beds. Several investigators
have documented water column concentrations exceeding several mg/L, with a peak detection of
27,800 (ig/L (27.8 mg/L) in one estuary. Those monitoring studies are discussed in greater detail
in the Risk Characterization portion of the Opinion.
The BEs for carbofuran (EPA 2004) and methomyl (EPA 2003) report some targeted studies
evaluating pesticide concentrations in field runoff and receiving water bodies. The carbofuran
BE reports concentrations of 16-28 |J,g/L of carbofuran in runoff from treated rice fields 26 d
after flooding (Nicosia, Carr et al. 1991) as cited in (EPA 2004)). Concentrations declined to <5
Hg/L by 37 d after flooding. The discussion does not describe when or how the carbofuran was
applied prior to flooding. Other studies described merely note that carbofuran does move into
surface water, and that concentrations decline over time. Ground water monitoring data for
carbofuran reported in the BE included concentrations of up 4.3 (ig/L in wells on Long Island,
NY 20 years after use was prohibited, and a maximum reported concentration of 176 (ig/L in
Suffolk County, NY.
The methomyl BE reports data from monitoring studies for specific crops that appear to have
been designed to approximate PRZM-EXAMS scenarios. These studies evaluated applications
rates ranging from 0.3 - 1.35 Ib ai/acre on crops such as cantaloupe, sweet corn, apples, lettuce,
DRAFT 304
-------
DRAFT
and tomatoes. All of the studies included multiple applications (from 5 to 29) and short re-
application intervals (1-5 d). Some studies measured concentrations in field runoff (96-1,320
Hg/L), and all measured concentrations in various receiving waterbodies (4.6 - 175 ng/L).
Carbaryl is one of the pesticides used by the U.S. Department of Agriculture's Animal and Plant
Health Inspection Service (APHIS) to control grasshopper infestations. Research was conducted
to evaluate the effects of rangeland aerial applications of a carbaryl formulation known as Sevin-
4-Oil (Beyers, Farmer et al. 1995). Fixed-wing aircraft were used to apply the carbaryl to
rangeland on both sides of the Little Missouri River. Pesticide applicators were instructed to
observe a 152 m no-spray buffer around the Little Missouri River. Carbaryl was applied at a rate
of 0.5 Ibs a.i./acre in 1991 and 0.4 Ibs per acre in 1993. Surface water concentrations were
monitored during a drought and non-drought year. During the study, discharge in the river
ranged from 0.026 to 0.057 m3/s in 1991 and 37.7 to 49.8 m3/s in 1993. Concentrations of
carbaryl in the river were monitored for 3 d following application (Table 59). Maximum values
in surface water were measured on the day of application and peaked at 85.1 and 12.6 ng/L in
1991 and 1993. Concentrations declined but were detectable when the study was terminated 96
h post treatment. Peak values were approximately 7 fold during the drought year when flows
were low, demonstrating the influence of dilution capacity on receiving water concentrations.
These values represent concentrations that might be observed in similar habitats used by listed
salmonids. The 85.1 |j,g/L represents one of the higher values observed in the monitoring data
despite the low use rate (0.5 Ibs a.i./acre) and no-spray buffer of almost 500 ft.
Table 59. Mean concentration (|o.g/L) of carbaryl in the Little Missouri River following rangeland
application of Sevin-4-Oil
Hours after application
1991
1993
1
85.1
12.0
2
-
12.6
4
12.3
3.84
8
3.08
4.01
12
5.30
-
24
10.3
4.51
96
0.100
5.14
-No Data
The CDPR database contained data from several studies conducted by CDPR in the 1990s
evaluating concentrations of rice chemicals in the Colusa Basin Drain, Butte Slough, and the
Sacramento River (referenced in their metadata as studies 17, 30, 34, and 40). Carbofuran, along
with molinate, malathion, methyl parathion, and thiobencarb, was one of the analytes in all years.
DRAFT
305
-------
DRAFT
Some years, analytes also included other pesticides such as 2,4-D, triclopyr, and propanil
(Norbergking, Durhan et al. 1991). Sampling was conducted in March-July. Carbofuran was
quantifiable in 100% of the samples for which it was an analyte. Concentrations across all years
ranged from 0.087 - 2.97 (ig/L (Table 60). Another study located in open literature, conducted
in 1991-1992 (Crepeau and Kuivila 2000), measured peak concentrations in the Colusa River
Basin of 0.6-1.1 ng/L. This study also measured concentrations in the Sacramento River at
Sacramento and the Sacramento River at Rio Vista. Carbofuran concentrations in the river were
typically an order of magnitude lower than the concentrations in the Colusa Basin Drain. The
CDPR studies included acute biotoxicity studies (test organism not specified in metadata) and
generally found no significant toxicity. However, a study on water from the Colusa Basin Drain
using Ceriodaphnia dubia as the test organism in a Toxicity Indicator Evaluation (TIE)
procedure found that an extract of Colusa Basin Drain water caused toxicity in laboratory tests
(Norbergking, Durhan et al. 1991) The procedure indicated that Carbofuran and methyl parathion
accounted for the toxicity of the sample, although other chemicals such as molinate and
thiobencarb were present as well. The concentrations observed in surface water monitoring are,
as expected much lower than those predicted for the flooded rice paddies themselves. It should
be noted that California requires rice growers to obtain discharge permits or get conditional
waivers before discharging water from treated rice fields. Obtaining the permits or waivers
typically requires holding periods following pesticide applications, which helps reduce pesticide
concentrations in the discharge. Those conditions are not a requirement of the federal label.
Table 60. Carbofuran Concentrations in CDPR Studies of Rice Effluents (1995-1998)
Statistic
Samples in data set
Quantifiable samples
% of quantifications in data set
LOQ range (ng/L)
Minimum concentration (|ig/L)
Maximum concentration (ng/L)
Arithmetic mean concentration (ng/L)
Standard deviation (ng/L)
Median concentration
All years
97
97
100%
0.1-0.35
0.0872
2.97
0.5661
0.5098
0.4125
1995
22
22
100%
0.05-0.35
0.1240
0.7000
0.3907
0.1465
0.3700
1996
29
29
100%
0.05-0.35
0.1620
2.97
0.8995
0.7862
0.6100
1997
25
25
100%
0.05-0.35
0.1400
0.6600
0.3998
0.1472
0.4080
1998
21
21
100%
0.05-0.35
0.0872
1.35
0.4800
0.3276
0.3800
Based on these studies, other targeted monitoring studies described in EPA BEs, and open
literature evaluated for the Opinion rendered on chlorpyrifos, diazinon, and malathion (NMFS
DRAFT
306
-------
DRAFT
2008), and the general state of knowledge regarding field runoff from pesticide applications, we
anticipate the following:
• edge-of-field runoff concentrations will be higher than concentrations measured in
waterbodies with substantial diluting volume,
• low-flow or runoff-dominated systems may contain the highest concentrations
(approaching or exceeding modeled concentrations), and
• measured concentrations are likely to be lower than peak runoff concentrations, as
sampling may not coincide with initial application and/or runoff events.
Monitoring data considerations
Surface water monitoring can provide useful information regarding real-time exposure and the
occurrence of environmental mixtures. A primary consideration in evaluating monitoring data is
whether the study design is sufficient to address exposure in a qualitative, quantitative, or
probabilistic manner. The available monitoring studies were conducted under a variety of
protocols and for varying purposes. None of the datasets discussed above were designed to
capture peak concentrations of exposure for salmon, or define exposure distributions.
Of the monitoring programs discussed, the studies conducted by the Washington State DOE were
designed to evaluate exposure to listed Pacific salmonid habitats in several Washington State
watersheds. This sampling program was intended to evaluate pesticide occurrence in some
salmonid-bearing streams during the pesticide application seasons (Johnson and Cowles 2003).
Sample sites for this study are best characterized as integration sites selected based on the
presence of the listed Yakima salmonid population (one of 17 independent populations that
comprise the Middle Columbia River steelhead DPS) and high diversity and intensity of
agriculture. The study design included sampling during the pesticide application season but did
not target specific applications of pesticides nor did it target salmonid habitats that would be
expected to produce the highest concentrations of pesticides (e.g., shallow off-channel habitat in
close proximity to pesticide application sites). Sampling was generally conducted on a weekly
basis, so it is likely peak concentrations associated with drift and runoff events were not
captured. Sampling stations included both agricultural- and urban-dominated watersheds, and
some storm events are captured in the sampling. Sampling favored the detection of multiple
pesticides, rather than peak concentrations in some habitats used by Middle Columbia River
steelhead.
DRAFT 307
-------
DRAFT
Other available monitoring data are also applicable to assessing exposure in listed salmon, but to
varying degrees. Common aspects that limit the utility of the available monitoring data as
accurate depictions of exposure within listed salmonid habitats include: 1) protocols were not
designed to capture peak concentrations or durations of exposure in habitats occupied by listed
species; 2) limited utility as a surrogate for other non-sampled surface waters; 3) lack of
representativeness of current and future pesticide uses and conditions; and 4) lack of information
on actual pesticide use to correlate with observed surface water concentrations.
Protocols not designed to capture peak exposure. The NAWQA monitoring studies
contain the largest data set evaluated. However, these studies were designed to evaluate trends
in water quality and were not designed to characterize exposure of pesticides to listed salmonids
(Hirsch 1988). The NAWQA design does not result in an unbiased representation of surface
waters. For example, some agricultural activities and related pesticide uses that may be very
important in a particular region may not be represented in the locations sampled. Sampling from
the NAWQA studies and other studies reviewed was typically not conducted in coordination
with specific applications of carbaryl, carbofuran, and methomyl. Similarly, sampling was not
designed with consideration to salmon distribution or to target the salmonid habitats most likely
to contain the greatest concentrations of pesticides. Given the relatively rapid dissipation of
these pesticides in flowing water habitats, it is not surprising that pesticide concentrations from
these datasets were generally much lower than predicted by modeling efforts.
Limited applicability to other locations. Pesticide runoff and drift are influenced by a
variety of site-specific variables such as meteorological conditions, soil type, slope, and physical
barriers to runoff and drift. Additionally, surface water variables such as volume, flow, and pH
influence both initial concentrations and persistence of pesticides in aquatic habitats. Finally,
cropping patterns and pesticide use have high spatial variability. Given these and other site-
specific factors, caution should be used when extrapolating monitoring data to other sites.
Representativeness of current and future uses. Pesticide use varies annually depending on
regulatory changes, market forces, cropping patterns, and pest pressure. The use of
DRAFT 308
-------
DRAFT
cholinesterase inhibiting insecticides has declined in California over recent decades. However,
pesticide use patterns change annually and may result in either increases or decreases in use of
pesticide products for specific uses. There is considerable uncertainty regarding the
representativeness of monitoring conditions to forecast future use of products containing the
three a.i.s.
Lack of information on actual use to correlate with observed concentrations. A
common constraint in the monitoring data was lack of information on actual use of pesticides
containing the three a.i.s. For example, the ability to relate surface water monitoring data to the
proposed action was severely hampered because information on application rates,
setbacks/buffers, and applications methods associated with the monitoring were generally not
reported. In most cases, the temporal and spatial aspect of pesticide use relative to sampling was
not reported, further limiting the utility of the information.
Exposure to Other Action Stressors
Stressors of the action also include the metabolites and degradates of the a.i.s, other active and
inert ingredients included in their product formulations, and tank mixtures and adjuvants
authorized on their product labels. Below we summarize information presented in the BEs and
provide additional information to characterize exposure to these Stressors.
Metabolites and degradates ofcarbaryl carbofuran, and methomyl
Available monitoring data for some metabolites and degrades of the three a.i.s in surface water is
presented above. The primary degradate and major metabolite ofcarbaryl is 1-naphthol (EPA
2003). The BE indicates that degradation studies show 1-naphthanol is found at up to 67% of the
applied carbaryl. It is also formed in the environment by degradation of naphthalene and other
polyaromatic hydrocarbon compounds. EPA suggests that salmonid exposure to 1-naphthol is
expected, particularly in alkaline waters, but indicates exposure to aquatic organisms cannot be
calculated due to lack of environmental fate and transport data for this degradate.
The major transformation product of carbofuran in water is 7-phenol (EPA 2004). The BE
indicates other transformation products have the potential to reach the aquatic environment,
DRAFT 309
-------
DRAFT
including 3-hydroxycarbofuran and 3-ketocarbofuran, and that these typically occur in small
amounts and are relatively short lived as compared to the parent. The major degradate in a soil
metabolism study was 3-keto carbofuran, which peaked at 12% of the amount applied after 181
days.
Several transformation products were also identified in the methomyl BE (EPA 2003). CO2 was
identified as the major degradate in most metabolism studies. S-methyl-N-
hydroxythioacetamidate was identified as a highly mobile product of alkaline hydrolysis. In an
aquatic metabolism study, after 7 d, acetonitrile comprised a maximum of 17% and acetamide up
to 14% of the amount of methomyl applied. After 102 d, volatilized acetonitrile totaled up to
27% of the applied methomyl.
The BEs recognized that listed salmonids are likely exposed to several metabolites and
degradates of the a.i.s. However, estimates quantifying exposure to these transformation
products were not provided and remain a considerable source of uncertainty.
Other ingredients in formulated products
Registered pesticide products containing carbaryl, carbofuran, and metholmyl generally include
other ingredients such as carriers and surfactants. NMFS searched NPIRS and located several
pesticide products that contain multiple a.i.s. NMFS located several labels of currently
registered products that contain more than one a.i. (Table 61). Carbaryl is a common component
in several fertilizer products used on turf. It is also commonly formulated with other pesticidal
ingredients including rotenone, a botanical extract used to eradicate fish and as an insecticide.
Carbaryl is formulated with malathion (another cholinesterase-inhibiting insecticide), bifenthrin
(a neurotoxic synthetic pyrethroid) and copper sulfate (registered for use as an insecticide,
algicide, and fungicide). Methomyl is formulated with muscalure, a fly attractant use in fly bait
formulations. NMFS is not aware of any carbofuran products that contain other a.i.s.
DRAFT 310
-------
DRAFT
Table 61. Examples of listed a.i.s on pesticide products containing carbaryl and methomyl.
EPA Product
Registration
Number
4-29
4-59
4-122
4-458
4-333,
239-2514,
8119-5,
71096-15
9198-233,
9198-234,
9198-235
270-255
2724-274
7319-6
53871-3
Active Ingredients
Basic cupric sulfate 7%, carbaryl 1 .25%, rotenone 0.5%, cube
resins other than rotenone 1%
Carbaryl 0.5%, malathion 3%, captan 5.87%
Carbaryl 0.3%, malathion 6%, captan 11.7%
Basic cupric sulfate 7%, carbaryl 2%
Metaldehyde 2%, carbaryl 5%
Carbaryl 2.3%, Bifenthrin 0.058%
Methomyl 1%, component of muscalure 0.025%
Methomyl 1%, component of muscalure 0.049%
Methomyl 1%, component of muscalure 0.026%
Methomyl 1%, component of muscalure 0.04%
Other
Ingredients
90.25%
90. 63%
82%
19%
93%
97.642%
98.975%
98.951%
98.974%
98.96%
Nonylphenol (NP) and nonylphenol polyethoxylates are inert ingredients that may be part of a
pesticide product formulation and are common adjuvant ingredients added during pesticide
applications. NP and nonylphenol polyethoxylates are also ingredients in detergents, cosmetics,
and other industrial products and are a common wastewater contaminant from industrial and
municipal sources. NP has been linked to endocrine disrupting effects in aquatic systems. A
national survey of streams found that NP was among the most ubiquitous organic wastewater
contaminants in the U.S., detected in more than 50% of the samples tested. The median
concentration of NP in streams surveyed was 0.8 (ig/L and the maximum concentration detected
was 40.0 ug/L (Table 62). Related compounds were also detected at a relatively high frequency
(Koplin, Furlong et al. 2002).
Table 62. Detection of nonionic detergent degradates in streams of the U.S. (Koplin, Furlong et al.
2002)
Chemical
4-nonylphenol
4-nonylphenol monoethoxylate
4-nonylphenol diethoxylate
4-octylphenol monoethoxylate
4-octylphenol diethoxylate
Frequency
Detected
50.6
45.9
36.5
43.5
23.5
Maximum (|j.g/L)
40
20
9
2
1
Median
(ug/L)
0.8
1
1
0.2
0.1
DRAFT
311
-------
DRAFT
We are uncertain to what degree NP and NP-ethoxylates may or may not occur in carbaryl,
carbofuran, and methomyl product formulations and/or are added prior to application. Inert
ingredients are often not specified on product labels. Additionally, NP and NP-ethoxylates
represent a very small portion of the more than 4,000 inert ingredients that EPA permits for use
in pesticide formulations (EPA 2008). Many of these inerts are known to be hazardous in their
own right (e.g., xylene is a neurotoxin and coal tar is a known carcinogen). Several permitted
inerts are also registered a.i.s (e.g., copper, zinc, chloropictrin, chlorothalonil). Inerts can be
more than 50% of the mass of pesticide products, and millions of Ibs of these products are
applied to the landscape each year (CDPR 2007). This equates to large contaminant loads of
inerts that may adversely affect salmon or their habitat. Uncertainty regarding exposure to these
ingredients will be qualitatively incorporated into our analysis.
Tank Mixtures
Several pesticide labels authorize the co-application of other pesticide products and other
materials in tank mixes, thereby increasing the likelihood of exposure to multiple chemical
stressors. For example, Sevin XLR Plus (EPA Reg. No. 264-333), which contains carbaryl,
specifies the product is compatible with a wide range of pesticides. The label indicates that
when used as a fruit thinner, Sevin XLR Plus may be mixed with other fruit thinners. Although
mixtures with other pesticides products are not specifically recommended on the Furadan LFR
label (EPA Reg. No. 279-3310, containing carbofuran), they are not prohibited. Furadan LFR
does specifically indicate that the product may be mixed with liquid fertilizer. Lannate SP (EPA
Reg. No. 352-342, containing methomyl) also provides instructions for tank mixing with other
products and identifies some tank mixes that are not compatible. These ingredients and the other
inert ingredients in these products are considered part of the action because they are authorized
by EPA's approval of the FIFRA label. Exposure, and consequently risk associated with
potential ingredients in tank mixtures were not addressed in EPA's BEs and remain a significant
source of uncertainty.
Environmental Mixtures
As described in the Approach to the Assessment, we analyze the status of listed species, in
conjuction with the Environmental Baseline in evaluating the likelihood that action stressors will
DRAFT 312
-------
DRAFT
reduce the viability of populations of listed salmonids. This involves considering interactions
between the stressors of the action and the Environmental Baseline. For example, we consider
that listed salmonids may be exposed to the wide array of chemical stressors that occur in the
various marine, estuarine, and freshwater habitats they occupy throughout their life cycle.
Exposure to multiple pesticide ingredients is most likely in freshwater habitats and nearshore
environments adjacent to areas where pesticides are used. As of 1997, about 900 a.i.s were
registered in the U.S. for use in more than 20,000 different pesticide products (Aspelin and
Grube 1999). Typically 10 to 20 new a.i.s are registered each year (Aspelin and Grube 1999). In
a typical year in the U.S., pesticides are applied at a rate of approximately five billion Ibs of a.i.
per year (Kiely, Donaldson et al. 2004). Pesticide contamination in the nation's freshwater
habitats is ubiquitous and pesticides usually occur in the environment as mixtures (Gilliom,
Barbash et al. 2006). "More than 90% of the time, water from streams with agricultural, urban,
or mixed-land-use watersheds had detections of two or more pesticides or degradates, and about
20% of the time they had detections of 10 or more (Gilliom, Barbash et al. 2006)." The
likelihood of exposure to multiple pesticides throughout a listed salmonids' lifetime is great
considering their migration routes and habitats occupied for spawning and rearing. In a three-
year monitoring study conducted by the Washington DOE, pesticide mixtures were found to be
common in both urban and agricultural watersheds (Burke, Anderson et al. 2006). An average of
three pesticides was found in each sample collected on urban sampling sites with as many as nine
pesticides found in a single sample. Agricultural sites averaged three to five pesticides per
sample with as many as 14 pesticides being detected in a single sample (Burke, Anderson et al.
2006). Mixtures of chemical that share a common mode or mechanism of action are of particular
concern. Six to 11 million Ibs of cholinesterase-inhibiting insecticides are used annually in
California (CDPR 2007). One a.i., thiodicarb, degrades into methomyl. Potential effects of
thiodicarb and other pesticide mixtures containing carbaryl, carbofuran, and methomyl were not
addressed in the BEs.
Cholinesterase inhibiting insecticides, including carbaryl, diazinon, chlorpyrifos, and malathion
are the most frequently detected mixtures in urban streams across the U.S. (Gilliom, Barbash et
al. 2006). Additionally, the high frequency for which carbaryl (99.7%), carbofuran (91.8%),
methomyl (26.2%), 1-napthol (25.4%), and other compounds are detected in surface waters of
DRAFT 313
-------
DRAFT
California, Idaho, Oregon, and Washing suggests that listed salmonids are commonly exposed to
combinations of these, and other mixtures of cholinesterase-inhibiting insecticides (Table 54).
Gilliom and others (2006) suggested that assessment of pesticide mixture toxicity to aquatic life
is needed given the widespread and common occurrence of pesticide mixtures, particularly in
streams, because the total combined toxicity of pesticides in water is often greater than that of
any single pesticide compound. Exposure to multiple pesticide ingredients can result in additive
and synergistic responses as described below in the Risk Characterization section. It is
reasonable to conclude that compounds sharing a common mode of action cause additive effects
and in some cases synergistic effects. CDPR's most recent pesticide use report indicates
6,857,530 Ibs of cholinesterase-inhibiting insecticides were applied in California during 2006.
Over 60 cholinesterase-inhibiting a.i.s are currently registered in California (CDPR 2007).
Exposure to these compounds and other baseline stressors (e.g., thermal stress) was not a
consideration in the BEs. Therefore, risk to listed species may be underestimated.
Exposure Conclusions
Pacific salmon and steelhead use a wide range of freshwater, estuarine, and marine habitats and
many migrate hundreds of miles to complete their life cycle. Carbaryl, carbofuran, and
methomyl are commonly detected in freshwater habitats within the four western states where
listed Pacific salmonids are distributed. Because the proposed action of registration of the three
a.i.s for the next 15 years authorizes many of the same uses, these three a.i.s will continue to be
present in the action area. Therefore, we expect some individuals within all the listed Pacific
salmon and steelhead ESUs/DPSs will be exposed to these chemicals and other stressors of the
action. Carbaryl can exceed several mg/L in coastal estuaries based on measured environmental
concentrations. Carbaryl and carbofuran concentrations in off-channel habitat can also exceed
several mg/L where buffers are lacking. Peak concentrations of methomyl are expected to be
significantly lower given label restrictions that require buffers to aquatic habitats and a relatively
low maximum application rate. However, model estimates indicate methomyl can reach
concentrations of several hundred |j,g/Lin surface waters given some uses that allow repeated
applications at short re-application intervals. Given variable use of these pesticides across the
landscape, and variable temporal and spatial distributions of listed salmonids, we expect
DRAFT 314
-------
DRAFT
exposure is also highly variable among individuals and populations of listed salmon. However,
defining exposure and distributions of exposure among differing life stages of each independent
population is complicated by several factors. Paramount among these is the uncertainty
associated with the use of pesticide products containing these a.i.s. More specifically:
• Although the BEs and RED documents provide information on EPA regulatory decisions,
they lack a full characterization of label-specific information needed to assess exposure
(e.g., application restrictions including application methods, rates, and intervals are
lacking for many non-agricultural uses);
• EPA-authorized labels contain language that frequently does not provide clear boundaries
on product use (e.g., the maximum number of applications is commonly not specified and
labels often instruct applicators to repeat applications "as necessary");
• Product labels authorize the application of chemical mixtures that are not specified or not
clearly defined (e.g., the ingredients of pesticide formulations are not fully disclosed,
labels recommend tank mixture applications with other pesticides and adjuvants and tank
mixtures with other pesticides are permitted unless specifically stated otherwise);
• Defining use of these products is highly uncertain because products are not likely to be
used to the full extent permitted on the labels and historical use information is limited and
may not reflect future use.
A major limitation of these assessments is that the majority of monitoring data used were not
designed to determine exposure to listed salmonids, with the exception of specific studies
conducted in Washington. Therefore, caution should be exercised in using these data for that
purpose. Additionally, the assessments lack uncertainty analyses of the monitoring and toxicity
data used, which limit the confidence in the given estimates (Warren-Hicks and Moore 1998).
Given the complexity and scale of this action we are unable to accurately define exposure
distributions for the chemical stressors. We assume the highest probability of exposure occurs in
freshwater, and nearshore estuarine/marine environments with close proximity to areas where
pesticide products containing carbaryl, carbofuran, and methomyl are applied. We considered
several sources of information to define the range of potential exposure to action stressors. EPA
provided a number of exposure estimates with maximum concentrations of 153, 35, and 99 (ig/L
predicted for registered uses of carbaryl, carbofuran, and methomyl, respectively. We generated
DRAFT 315
-------
DRAFT
additional exposure estimates for shallow off-channel habitats with predicted concentrations
exceeding 1,000 (ig/L for carbaryl and carbofuran, and a peak concentration of 58 (ig/L for
methomyl. Additionally, we considered monitoring data presented by EPA and from other
sources which indicate comparable concentrations of carbaryl, carbofuran, and methomyl have
been detected in surface waters within the four states where the listed salmon and steelhead are
distributed (85, 32, and 175 ng/L, respectively).
We assume that the exposure estimates provided by EPA in the BEs and additional modeling and
monitoring information provided above represent realistic exposure levels for some individuals
of the listed species. Further, we assume the distribution within the range of exposures is a
function of pesticide use and the duration of time listed salmonids spend in these habitats. All
listed Pacific salmon and steelhead occupy habitats that could contain high concentrations of
these pesticides at one or more life stages. However, the time spent in these habitats varies
among species. Adult salmon and steelhead spend weeks to several months in freshwater
habitats during their migration and spawning activities. Immediately after emerging from the
gravel substrate and transitioning from alevins to fry, salmonids move to habitats where they can
swim freely and forage. At this point in their development most salmon occupy freshwater
habitats. Chum salmon are an exception. They immediately migrate downstream following
emergence to nearshore environments in estuaries near the mouth of their parent stream. Upon
arrival in the estuary the chum salmon fry inhabit nearshore areas at a preferred depth of 1.5-5 m.
In Puget Sound surveys indicate chum salmon fry are distributed extremely close to the shoreline
and concentrated in the top 6 inches of water. Chum salmon fry are less likely to be exposed to
high concentrations of pesticides than other salmonids given the habitat they occupy and the
duration of time spent in the shallow water habitats. They may reside immediately next to the
shore in estuaries for as little as one or two weeks before moving offshore or into deeper-water
habitats within the nearshore environment. Sockeye salmon fry most frequently distribute to
shallow beach areas in the littoral zones of lakes. They initially occupy shoreline habitats of
only a few centimeters in depth before moving further off-shore and taking on a more pelagic
existence. Coho salmon, Chinook salmon, and steelhead fry typically select off-channel habitats
associated with their natal rivers and streams. These species are most likely to experience higher
pesticide exposures given their utilization of shallow freshwater habitats as juveniles for rearing.
DRAFT 316
-------
DRAFT
Coho salmon and steelhead have a greater preference for the shallow habitats and rear in
freshwater for more than a year.
Substantial data gaps in EPA's exposure characterization include exposure estimates associated
with product uses on many crops and particularly, on non-crop uses. The highest concentrations
detected in surface waters were those associated with applications to aquatic habitats. Those
types of applications although mentioned, were not evaluated in EPA's BEs. Additionally,
exposure estimates for other chemical stressors including other ingredients in pesticide
formulations, other pesticide products authorized for co-application, adjuvants, degradates, and
metabolites are not available or are non-existent. Although NMFS is unable to comprehensively
quantify exposure to these chemical stressors, we are aware that exposure to these stressors is
likely. We assume these chemical stressors may pose additional risk to listed Pacific salmonids.
However, in order to ensure that EPA's action is not likely to jeopardize listed species or destroy
or adversely modify critical habitat, NMFS analyzes exposure based on all stressors that could
result from all uses authorized by EPA's action.
Response Analysis
In this section, we identify and evaluate toxicity information from the stressors of the action and
organize the information under assessment endpoints. The endpoints target potential effects
from the stressors of the action to individual salmonids and their supporting habitats. The
assessment endpoints represent biological attributes that, when adversely affected, lead to
reduced fitness of individual salmonids or degrade PCEs (e.g., prey abundance and water
quality). We constructed a visual conceptual model to guide development of risk hypotheses and
assessment endpoints to highlight potential uncertainties uncovered by our analysis of the
available information. We begin the response analysis by describing the toxic mode and
mechanism of action of carabaryl, carbofuran, and methomyl. Next we summarize the toxicity
data presented in the three BEs and organize the information to applicable assessment endpoints
(e.g., survival, growth, etc.). The information provided by EPA addressed aspects of survival,
growth and reproduction of aquatic species (freshwater and saltwater), as well as providing some
discussion on other information found in the open literature, such as results from some field
experiments and experiments that evaluated sublethal effects. NMFS is charged under the ESA
DRAFT 317
-------
DRAFT
to evaluate all direct and indirect effects of an action. We therefore evaluate all aspects of an
action that may reduce fitness of individuals or reduce primary constituent elements of
designated critical habitat. The evaluation includes information that EPA provided on survival,
growth, or reproduction, but also encompasses a broader range of endpoints including behaviors,
endocrine disruption, and other physiological alterations. The information we assessed is
derived from published, scientific journals and information from government agency reports,
theses, books, information and data provided by the registrants identified as applicants, and
independent reports. Typically, the most relevant study results are those that directly measure
effects to an identified assessment endpoint derived from experiments with salmonids, preferably
listed Pacific salmonids or hatchery surrogates, exposed to the stressors of the action.
Effects of pesticide products on
ESA-listed species and their habitat
Response Profile
Figure 37. Response analysis
Mode and Mechanism of Action
Carbaryl, carbofuran, and methomyl share a similar mode and mechanism of toxic action and are
a part of a group known as TV-methyl carbamates. All three have similar chemical structures and
act as neurotoxicants by impairing nerve cell transmission in vertebrates and invertebrates. They
inhibit the enzyme AChE, which is present in cholinergic synapses. The normal function of
AChE is to break down (hydrolyze) the neurotransmitter, acetylcholine, thereby serving as an
"off-switch" for the electrochemical signal transmissions along nerve cells and neuromuscular
junctions. AChE is prevalent in a variety of cell and organ types throughout the body of
vertebrates and invertebrates (Walker and Thompson 1991). Interference of normal nerve
transmission by TV-methyl carbamates may affect a wide array of physiological systems in fish
(Figure 4). Organophosphates (OPs) share this mode of action and physiological responses are
similar.
DRAFT
318
-------
DRAFT
The mechanism of action of TV-methyl carbamates, inhibition of AChE, involves a series of
enzyme-mediated reactions. Briefly, in a reversible reaction carbamates bind to AChE, thereby
inhibiting AChE's normal activity to hydrolyze the neurotransmitter acetylcholine at nerve
synapses. This reaction is similar to organophosphorus insecticides with the main exception
being a carbamylation of AChE instead of a phosphorylation. Carbamate inhibition of AChE is
"reversible" in cases of sublethal exposure and recovery of TV-methyl carbamate-inhibited AChE
is typically rapid compared to OP-inhibited AChE. The key result of AChE inhibition by
carbamate and OP insecticides is accumulation of acetylcholine in a cholinergic synapse. The
buildup of acetylcholine causes continuous nerve firing and eventual failure of nerve impulse
propagation. A variety of adverse effects to organisms can result ranging from sublethal
behavioral effects to death (Mineau 1991).
Incidences of acute poisoning from AChE inhibitors are prevalent for wildlife, particularly for
birds and fish (Mineau 1991). The following passage describes the classic signs of AChE-
inhibiting insecticide poisonings offish:
"Fish initially change normal swimming behavior to rapid darting about with loss of
balance. This hyper excitability is accompanied by sharp tremors which shake the
entire fish. The pectoral fins are extended stiffly at right angles from the body
instead of showing the usual slow back and forth motion normally used to maintain
balance. The gill covers open wide, and opercular movements become more rapid.
With death the mouth is open and the gill covers are extended. Hemorrhaging
appears around the pectoral girdle and base of the fins (Weiss and Botts 1957)."
Numerous reports, peer-reviewed journal articles (Williams and Sova 1966; Holland, Coppage et
al. 1967; Coppage and Matthews 1974; Rabeni and Stanley 1975; Haines 1981; Antwi 1985) as
well as multiple reviews, text books (Mineau 1991; Smith 1993; Geisy, Solomon et al. 1999),
and wildlife poisoning cases document inhibition of AChE activity in exposed invertebrates
(Detra and Collins 1986; Detra and Collins 1991) and vertebrates including salmonids following
exposures to carbamates and OPs (Zinkl, Shea et al. 1987; Hoy, Horsberg et al. 1991; Beyers
and Sikoski 1994; Li and Fan 1996; Grange 2002; Sandahl, Baldwin et al. 2004; St. Aubin 2004;
Sandahl, Baldwin et al. 2005; Scholz, Truelove et al. 2006; Yi, Liu et al. 2006; Eder, Kohler et
DRAFT 319
-------
DRAFT
al. 2007; Liu, Yi et al. 2007; Tierney, Casselman et al. 2007; Li, Jiang et al. 2008; Laetz,
Baldwin et al. In Press).
One highly relevant study measured inhibition of brain AChE, duration of recovery, survival at
24 h, and tissue concentrations in juvenile rainbow trout (O. mykiss) following exposure to
carbaryl at 0, 250, 500, 1,000, 2,000, and 4,000 ng/L (Zinkl, Shea et al. 1987). Rainbow trout
showed dose-dependent AChE inhibition from 61 to 91 % when exposed to 250 - 4,000 [ig/L for
24 h. Most trout that died had 85% or greater inhibition. Trout recovered AChE activity
following 24 h in uncontaminated water, indicating that fish recover if given the opportunity
following carbamate exposures (Zinkl, Shea et al. 1987). This study showed that carbaryl is
acutely toxic to rainbow trout in a matter of hours (incidences of death at 1.5 - 4 h) at
concentrations at or above 1,000 ug/L.
pH and toxicity
We located several data sets on toxicity to freshwater fish and aquatic invertebrates exposed to
carbaryl and carbofuran at different pHs (Mayer and Ellersieck 1986). Acute toxicity of carbaryl
and carbofuran increases as pH increased, based on the available freshwater fish assays (Mayer
and Ellersieck 1986). This largely is considered a result of the influence pH has on the creation
of hydrolysis products (Mayer and Ellersieck 1986). However, the persistence of carbaryl and
carbofuran is reduced as pH increases. As noted in the exposure section, degradation half-lives
can vary from hours (in alkaline waters) to weeks (in acidic waters) depending on pH. Within
the Pacifc Northwest and California pH varies seasonally and typically may range from 6-9.
For methomyl, pH seems to have less of an influence on hydrolysis rates. Due to the influence
of pH on persistence of carbaryl and carbofuran, we evaluated reported toxic effects within the
context of pH if provided and discuss pH in relationship to salmonid habitat utilization.
Temperature and toxicity
We found no consistent correlation with temperature and toxicity of the three TV-methyl
carbamates.
DRAFT 320
-------
DRAFT
Studies with mixtures of AChE inhibiting insecticides
Because the three carbamates share a common mechanism of action, are registered for use and
applied in the same watersheds, and have demonstrated additive and synergistic effects in aquatic
organisms, we evaluate the response of salmonids and their habitat not just from exposure to
single carbamates, but also to common mixtures of TV-methyl carbamates. We therefore include
an analysis of combinations of carbaryl, carbofuran, and methomyl based on additive toxicity
observed in recent publications with Chinook and coho salmon (discussed in the Risk
Characterization section) (Scholz, Truelove et al. 2006; Laetz, Baldwin et al. In Press). Because
NMFS identified mixture effects to listed salmonids as a critical data gap in understanding the
occurrences of multiple insecticides in salmonid habitats, a number of the experiments described
below were conducted by researchers at or associated with NOAA' s Northwest Fisheries Science
Center.
One of the earliest mixture studies available evaluated bluegill survival following a range of
exposure durations (24, 48, 72, or 96 h) to binary combinations of 19 insecticide mixtures
(Macek 1975). Carbaryl and several of the OP pesticides were tested. Macek (1975) used the
equation AB/(A+B) = X to calculate mixture toxicity; where AB was the number of dead fish
from a mixture of pesticides A and B, and A + B was the sum of dead fish from A and B alone.
The resulting ratios, X, were designated by the author as less than additive for a ratio of less than
0.5, additive when the ratio fell between 0.5 and 1.5, and synergistic for a ratio of more than 1.5.
Carbaryl containing mixtures resulted in additive toxicity for some compounds (DDT,
methoxychlor, parathion, methyl parathion) and synergistic toxicity for other compounds
(malathion, copper sulfate). Antagonism is when the cumulative toxicity of a mixture is less
than additive. In this study, mixtures containing carbaryl were not atagonistic. Differences in
classification between additive and synergistic combinations should be interpreted cautiously, as
the threshold for synergism was arbitrarily set at 1.5 by the authors. Mixture results with DDT
and toxaphene were 1.31 and 1.14, respectively. The binary combination of diazinon and
parathion were classified as synergistic toxicants, (i.e., more fish died than predicted based on an
additive response). Validation of chemical concentrations with analytical chemistry was not
conducted. Although the lack of raw data makes it difficult to determine exact concentrations
tested and the lack of analytical confirmation precludes determination of precise concentrations,
DRAFT 321
-------
DRAFT
the study shows that binary combinations containing carbaryl exhibit greater toxicity than
exposure to carbaryl alone. Additionally, the majority of pesticide-containing mixtures tested
resulted in either additive or syngeristic responses including mixtures containing carbamate and
OP insecticides.
Additive toxicity of binary combinations of carbamates and OPs at a cellular level was
demonstrated from in vitro experiments with Chinook salmon (Scholz, Truelove et al. 2006).
Carbaryl and carbofuran, in addition to the oxons of diazinon, chlorpyrifos, and malathion
caused additive toxicity as measured by AChE inhibition in salmonid brain tissue (Scholz,
Truelove et al. 2006). Further, the joint toxicity of the mixtures could be accurately predicted
from each insecticide's toxic potency, simply by adding the two potencies together at a given
concentration. Since the experiments were conducted using in vitro exposures with the oxon
degradates and not with the parent compounds, the authors conducted subsequent sets of
experiments to investigate whether additive toxicity as measured by AChE inhibition also
occurred when live, juvenile coho salmon were exposed for 96 h to the parent compounds, i.e., in
vivo exposures.
The results of the second set of experiments were unexpected by the authors (Laetz, Baldwin et
al. In Press). Measured AChE inhibition from some of the binary combinations was
significantly greater than the expected additive toxicity, i.e., synergistic toxic responses were
found (Laetz, Baldwin et al. In Press). As with the in vitro study, brain AChE inhibition in
juvenile coho salmon (O. kisutch) exposed to sublethal concentrations of the carbamates carbaryl
and carbofuran, as well as the OPs chlorpyrifos, diazinon, and malathion, was measured (Laetz,
Baldwin et al. In Press). Dose-response data for individual chemicals were normalized to their
respective ECso concentrations (AChE activity compared to control) and collectively fit to a non-
linear regression. The regression line was used to determine whether toxicological responses to
binary mixtures were antagonistic, additive, or synergistic. No binary mixtures resulted in
antagonism. Additivity and synergism were both observed, with a greater degree of synergism at
higher exposure concentrations. Moreover, certain combinations of OPs were lethal at
concentrations that were sublethal in single chemical trials. Concentrations of each insecticide
are listed in Table 20. Based on a default assumption of dose-addition, the five pesticides were
DRAFT 322
-------
DRAFT
combined in all possible pairings to yield target levels of AChE inhibitions in the brains of
exposed coho salmon.
Table 63. Concentrations (ng/L) of insecticides used in mixture exposures. ECSOs were
calculated from dose-response data of AChE activity using non-linear regression. Coho salmon
exposed to 1.0, 0.4, or 0.1 EC50 treatments had an equipotent amount of each carbamate or OP
within the treatment e.g., to attain the 1.0 EC50 treatment for diazinon and chlorpyrifos, 1.0 jig/L of
chlorpyrifos (0.5 the EC50) was combined with 72.5 iig/L (0.5 of the EC50).
Insecticide
Carbaryl
Carbofuran
Chlorpyrifos
Diazinon
Malathion
Measured EC50
145.8
58.4
2.0
145.0
74.5
Concentration of each ingredient in binary
combination to achieve treatment level
1.0EC50 units
72.9
29.2
1.0
72.5
37.3
0.40 EC50
units
29.2
11.7
0.4
29.0
14.9
0.10EC50
units
7.3
2.9
0.1
7.3
3.7
As determined by the regression, these levels of enzyme inhibition would result from exposure to
0.1, 0.4, and 1.0 EC50 units, respectively. Two thirds (20/30) of pesticide pairs yielded AChE
levels that were significantly lower, i.e., indicative of synergism, than would be expected based
on additivity i.e., dose-addition (t-test with Bonferroni correction, p < 0.005). The number of
combinations that were statistically synergistic increased with increasing exposure
concentrations. All pairings analyzed at 1.0 EC50 showed synergism in AChE inhibition.
The combination of carbaryl at 79.9 [ig/L and carbofuran at 29.2 [ig/L to achieve 1.0 EC50
produced synergistic toxicity, however no incidences of coho mortality were observed. At 0.1
and 0.4 EC50 levels with the two carbamates, synergistic toxicity was observed, although the
deviation from the predicted additive response was not statistically significant. In binary
mixtures containing an OP and a carbamate synergistic toxicity occurred with some of the
combinations at 0.1 and 0.4 EC50 treatments. The mechanism for synergistic toxicity in
salmonids is unknown.
Additionally, pairings of two OPs produced a greater degree of synergism than mixtures
containing one or two carbamates. This was particularly true for mixtures containing malathion
coupled with either diazinon or chlorpyrifos. At the highest exposure treatment, 1.0 EC50
DRAFT
323
-------
DRAFT
(malathion at 37.3, chlorpyrifos at 2, diazinon at 72.5 (ig/L ), binary combinations produced
synergistic toxicity. Many fish species die following high rates of acute brain AChE inhibition,
between 70-90% (Fulton and Key 2001).
Coho salmon exposed to combinations of diazinon and malathion (1.0 and 0.4 EC50) as well as
chlorpyrifos and malathion (1.0 ECso) all died (Laetz et al. In Press). This result was in contrast
to the predicted AChE inhibition from in vitro tests with Chinook salmon. Coho exposed to
these OP mixtures showed toxic symptoms of inhibition of AChE, including loss of equilibrium,
rapid gilling, altered startle response, and increased mucus production before dying. OP
combinations were also synergistic at the lowest concentrations tested. Diazinon and
chlorpyrifos were synergistic when combined at 7.3 |j,g/L and 0.1 ng/L, respectively. The
pairing of diazinon (7.3 ng/L) with malathion (3.7 ng/L) produced severe (> 90%) AChE
inhibition, including classical signs of poisoning as well as death with some combinations. We
expect that juvenile salmonids exposed to these effect concentrations in the environment will
respond similarly.
Multiple studies indicate compounds that share a common mode of action frequently result in
additive and at times synergistic responses in aquatic organisms (Macek 1975; Bocquene,
Bellanger et al. 1995; Monserrat, Yunes et al. 2001; Anderson and Lydy 2002; Jin-Clark, Lydy
et al. 2002; Lydy and Austin 2005; Scholz, Truelove et al. 2006; Belden, Gilliom et al. 2007;
Laetz, Baldwin et al. In Press). Unfortunately, we are unable to create a predictive model of
synergistic toxicity as dose-response relationships with multiple ratios of pesticides are not
available at the present time and the mechanism of synergism remains to be determined. That
said, we conducted a mixture analysis with carbaryl, carbofuran, and methomyl based on
additive toxicity with the caveat that synergism is likely where circumstances mirror the
experimental conditions of this study, i.e., similar exposure durations and pesticide
concentrations. This is a reasonable approach based on the current state of the science. We used
the mixture work of Laetz et al. In Press and Scholz et al 2006 to construct mixture dose-
response relationships predicated on additivity (see mixture analyses in the Risk
Characterization section).
DRAFT 324
-------
DRAFT
Summary ofToxicity Information Presented in the Biological Evaluations
Each BE primarily summarized acute and chronic toxicity data from "standardized toxicity tests"
submitted by pesticide registrants during the registration process, tests from government
laboratories available in EPA databases, or from published, peer-reviewed scientific publications
(books and journals). The assessment endpoints from these tests for an individual organism
generally included aspects of survival (death), reproduction, and growth measured in laboratory
dose-response experiments (EPA 2004). Survival is measured in both acute and chronic tests.
Reproduction and growth are generally measured and reported in the chronic tests. Population-
level endpoints and analyses were generally absent in the BEs, other than a few measurements of
fish and aquatic invertebrate reproduction. Adverse effects to organisms were not translated into
consequences to populations. The BEs also presented some information on multispecies
microcosm and mesocosm studies. For this Opinion, NMFS translates effects to individual
salmonids into potential population level consequences as explained in the Risk Characterization
portion of the Effects of the Proposed Action section, and ultimately draws a conclusion on the
likely risk to listed salmonids based on exposure and anticipated individual and population level
effects.
Survival of individual fish is typically measured by incidences of death following 96 h exposures
(acute test) and incidences of death following 21 d, 30 d, 32 d, and "full life cycle" exposures
(chronic tests) to a subset of freshwater and marine fish species reared in laboratories under
controlled conditions (temperature, pH, light, salinity, dissolved oxygen, etc.,) (EPA 2004).
Lethality of the pesticide is usually reported as the median lethal concentration (LC50), the
statistically-derived concentration sufficient to kill 50% of the test population. For aquatic
invertebrates it may be reported as an EC50, because death of these organisms may be difficult
to detect and immobilization is considered a terminal endpoint. An LC50 is derived from the
number of surviving individuals at each concentration tested following a 96 h exposure and is
typically estimated by probit or logit analysis and recently by statistical curve fitting techniques.
In FIFRA guideline tests, LCSOs are typically calculated by probit analysis. If the data are not
normally distributed for a probit analysis, than either a moving average or binomial is used,
resulting in no slope being reported. Ideally, to maximize the utility of a given LC50 study, a
slope, variability around the LC50, and a description of the experimental design- such as
DRAFT 325
-------
DRAFT
experimental concentrations tested, number of treatments and replicates used, solvent controls,
etc.- are needed. The slope of the observed dose-response relationship is particularly useful in
interpolating incidences of death at concentrations below or above an estimated LC50. The
variability of an LC50 is usually depicted by a confidence interval (95% CI) or standard
deviation/error and is illustrative of the degree of confidence associated with a given LC50
estimate i.e., the smaller the range of uncertainty the higher the confidence in the estimate.
Without an estimate of variability, it is difficult to infer the precision of the estimate.
Furthermore, survival experiments are of most utility when conducted with the most sensitive
life stage of the listed species or a representative surrogate. In the case of ESA-listed Pacific
salmonids, there are several surrogates that are available for toxicity testing including hatchery
reared coho salmon, Chinook salmon, steelhead, and chum salmon, as well as rainbow trout2.
The available toxicity data include a varitety of salmonids. Unfortunately, slopes, estimates of
variability for an LC50, and experimental concentrations frequently are not reported. In our
review of the BEs, we did not locate any reported slopes of dose-response curves, although some
of this information was presented in some of the corresponding Science Chapters. Consequently,
we must err on the side of the species in the face of these uncertainties and select LCSOs from the
lower range of available salmonid studies. We selected LCSOs and associated slopes as input in
the population modeling exercises discussed later. We evaluate the likelihood of concentrations
that are expected to kill fish and apply qualitative and quantitative methods to infer population-
level responses of ESA-listed salmonids within the Risk Characterization section (Figure 2).
Growth of individual organisms is an assessment endpoint derived from standard chronic fish
and invertebrate toxicity tests summarized in the BEs. It is difficult to translate the significance
of impacted growth derived from a guideline study to fish growth in aquatic ecosystems. The
health of the fish, availability and abundance of prey items, and the ability of the fish to
adequately feed are not assessed in standard chronic fish tests. These are important factors
affecting the survival of wild fish. What is generally assessed is size or weight offish measured
2 Rainbow trout and steelhead are the same genus species (Oncorhynchus mykiss), with the key differentiation that
steelhead migrate to the ocean while rainbow trout remain in freshwaters. Rainbow trout are therefore good
lexicological surrogates for freshwater life stages of steelhead, but are less useful as surrogates for life stages that
use estuarine and ocean environments.
DRAFT 326
-------
DRAFT
at several times during an experiment. The test fish are usually fed twice daily, ad libitum, i.e.,
an over abundance of food is available to the fish. Therefore, any reductions in size are a result
offish being affected to such an extent that they are not feeding even when presented with an
abundance of food. Subtle changes in feeding behaviors or availability of food would not be
detected from these types of experiments. If growth is affected in these experiments, it is highly
probable that growth offish in natural aquatic systems would be severely affected. If effects to
growth are likely, we assess salmonid population-level consequences based on reductions in
juvenile growth and subsequent reduction in size prior to ocean entry.
Reproduction, at the scale of an individual, can be measured by the number of offspring per
female (fecundity), and at the scale of a population by measuring the number of offspring per
females in a population over multiple generations. The BEs summarized reproductive endpoints
at the individual scale from chronic freshwater fish experiments where hatchability and juvenile
and larval survival are measured. NMFS considers many other assessment measures of
reproduction, including egg size, spawning success, sperm and egg viability, gonadal
development, reproductive behaviors, and hormone levels. These endpoints are not generally
measured in standardized toxicity assays used in pesticide registration.
Sometimes qualitative observations of sublethal effects are summarized from 96 h lethality dose-
response bioassays in EPA's risk assessments. These observations generally were limited in the
BEs for carabaryl, carbofuran, and methomyl, and when noted, pertained to unusual swimming
behaviors. None of these behaviors were rigorously measured and therefore are of limited value
in assessing the effects of the three insecticides on Pacific salmonids. We do, however note a
few of the observations when they pertained to a relevant assessment endpoint, such as impaired
swimming. Some BEs presented toxicity information on degradates, metabolites, and
formulations. However, toxicity information on other or "inert" ingredients found in pesticide
formulations was usually not presented.
Results from multiple species tests, called microcosm and mesocosm studies, were also
discussed in the BEs to a varying degree. These types of experiments are likely closer
approximations of potential ecosystem-level responses such as interactions among species
(predator-prey dynamics), recovery of species, and indirect effects to fish. However, the
DRAFT 327
-------
DRAFT
interpretation of results is complicated by how well the results represent natural aquatic
ecosystems and how well the studies apply to salmonid-specific assessment endpoints and risk
hypotheses. These studies typically measured individual responses of aquatic organisms to
contaminants in the presence of other species. Some are applicable to questions of trophic
effects and invertebrate recovery, as well as providing pesticide fate information. The most
useful mesocosm study results for this Opinion are those that directly pertain to identified
assessment endpoints and risk hypotheses. We discuss study results in the context of salmonid
prey responses, emphasizing survival and recovery of prey taxa as well as shifts from preferred
taxa to other taxa if measured. One of the notable limitations of these study types is they do not
represent real world aquatic ecosystems that are degraded from various stressors including
contaminants.
Results from aquatic field studies were generally not discussed in great detail within the BEs.
We discuss field studies that evaluated identified assessment endpoints, particularly those which
address salmonid prey responses in systems with ESA-listed salmonids.
Ranges in toxicity values presented in the BEs for each a.i. are summarized in Table 21. Ranges
in toxicity values (ng/L) are organized by assessment endpoint and associated assessment
measures. The BEs provided toxicity information from EPA's EFED Pesticide Toxicity
Database and from the ACQUIRE database.
DRAFT 328
-------
DRAFT
Table 64 .Assessment endpoint toxicity values (u.g/L) presented in BEs and REDs for carbaryl, carbofuran, and methomyl
Assessment
Endpoint
Survival
Survival
Reproduction
or larval
survival
Fish growth
Habitat-
salmonid prey
Assessment
measure
Freshwater fish LCso
Salmonid LCso
NOEC/LOEC
NOEC/LOEC
invertebrate survival
invertebrate fecundity/
emergence/
developmental rate
invertebrate
reproduction
carbaryl (|ig/L)
> 95% a.i.
250-
20,000
n=17
250-
3,000 n=7
fathead
minnows
210/680
n=1
—
1.7-26
n=5
500/
1,000
1.5/3.3
n=1
< 95% a.i.
14,000-
290,000
n=8
1,400-
4,500 n=4
—
—
4.3-13.0
n=5
—
degradate
(s)
1,400-1,800
n=6
1,400, 1,600
n=2
—
—
200-2,100
n=5
—
carbofuran (p.g/L)
> 95% a.i.
88-1,990
n=20
164-600
n=10
rainbow
trout
24.8/56.7
n=1
(larval
survival1)
rainbow
trout2
24.8/56.7
n=1
2.2-2,700
n=3
—
9.8/27
n=1
< 95% a.i.
240-
3,100n=6
610
n=1
—
—
41
n=1
—
degradate
(s)
—
—
—
—
—
—
methomyl (|ig/L)
> 95% a.i.
480-
6,800
n=11
560-
6,800
n=6
fathead
minnow
57/117
n=1
(larval
survival1)
fathead
minnow
76.117/
142,243
n=2
8.8-920
n=8
0.4/0.8
n=1
1.6/3.1
n=1
< 95% a.i.
300-7,700
n=16
1,200-
3,200
n=7
—
—
7.6 - 720
n=6
—
degradate
(s)
462,000
n=1
—
—
—
—
—
1. Larval survival derived from 28 day fish assays
2. Juvenile growth measured at 60, 75, 90 d exposure
DRAFT
329
-------
DRAFT
Assessment endpoint: Fish survival
Assessment measure: 96 h survival from laboratory bioassays reported as an LC50.
Carbofuran is the most toxic of three insecticides based on fish survival values (LCSOs) followed
by carbaryl and methomyl, respectively (Table 21). All three carbamates have a range of acute
freshwater fish LCSOs spanning 1 to 2 orders of magnitude. EPA reported the following ranges
of LCSOs: carbofuran LCSOs ranged from 88-3,100 |J,g/L; carbaryl LCSOs ranged from 250-
290,000 |ig/L; and methomyl LCSOs ranged from 300-7,700 [ig/L. Based on these LC50 ranges,
EPA classified these insecticides as "highly toxic" to "moderately toxic". Salmonids were well
represented in the data set, with 11 results for carbaryl, 11 for carbofuran, and 13 for methomyl.
A cumulative frequency distribution of carbaryl LCSOs for freshwater fish indicated that Atlantic
salmon were the most sensitive of the species tested, and salmonids as a group were much more
sensitive than fathead minnow and bluegill sunfish (EPA 2003).
EPA classified the three carbamates as very highly toxic to moderately toxic to estuarine and salt
water species depending on the chemical and the fish species tested. Carbofuran LCSOs for
marine and estuarine fish ranged from 33 [ig/L to more than 100 [ig/L (n=5), indicating that
saltwater/estuarine species were more sensitive to carbofuran than salmonids tested in
freshwater. Two LCSOs (2,200 and 2,600 |J,g/L, both sheepshead minnow) were reported for
carbaryl and one LC50 (1,160 |J,g/L, sheepshead minnow) was reported for methomyl. Based on
available data presented in EPA documents, it is uncertain whether marine and estuarine species
are more sensitive, less sensitive, or equally sensitive to carbaryl and methomyl compared to
freshwater fish. No data were presented on salmonids exposed in saline environments.
Assessment endpoint: Reproduction
Assessment measure: Number of offspring, hatchability, number offish that attained sexual
maturity by 136 d, and number of spawns per spawning pair
For carbaryl, one chronic study was listed in the BE (EPA 2003), and briefly discussed in the
Science Chapter (EPA 2003c), that evaluated a variety of assessment endpoints of fathead
minnows including reproductive endpoints (referenced in EPA 2003c as TOUCAR05 Carlson
1972, although this is an erroneous citation of (Carlson 1971)). Reproductive endpoints
measured included number of mature males, number of mature females, number of immature
DRAFT 330
-------
DRAFT
fish, number of eggs per mature female, number of eggs spawned, and, hatchability of eggs
(Carlson 1971). Fathead minnows were exposed to five treatments (8, 17, 62, 210, and 680 (ig/L
[analytically verified]) of carbaryl in a flow through system for nine months; capturing the life
cycle of the fathead minnow. Fathead minnows showed reduced number of eggs per female and
reduced number of eggs spawned when exposed to 680 |j,g/L and of the eggs spawned, none
hatched (Carlson 1971).
The carbofuran BE (EPA 2004) stated that no full life cycle fish tests were available, and as it
did not report any early-life stage test results, presumably none of those were available either, as
the only data reported were for a partial life cycle test for rainbow trout (O. mykiss). Reported
NOAECs were based on growth effects, and no reproductive endpoints were discussed. A
definitive reference to the study was not provided in the BE. The carbofuran Science Chapter
(EPA 2005) reports on an early life stage test for rainbow trout, referring to it as Acc.#
GEOCAR08. We were unable to locate a specific citation for this study in the references but the
reported NOAEC (24.8 ng/L) and LOAEC (56.7 ng/L) were the same as the unreferenced study
discussed in the BE. However, the Science Chapter stated that larval survival was the most
sensitive endpoint of those evaluated, and that scoliosis was observed in the larval fish at
carbofuran concentrations >56.7 [ig/L. The Science Chapter also references a sheepshead
minnow (Cyprinodon varigatus) early life stage study (MRID 432505-01) that resulted in a
NOAEC of 2.6 ng/L and LOAEC of 6.0 [ig/L based on reduced embryo hatching. Specific
extent of reduced hatching was not reported. No sublethal effects, such as the scoliosis in the
rainbow trout, were mentioned in this study or two other studies on sheepshead minnow (MRIDs
408184-01 and 426974-01) that were submitted but did not produce definitive NOAECs or
LOAECs.
The Science Chapter for methomyl (EPA 1998) reports a NOAEC (57 [ig/L) and LOAEC
(117 |J,g/L) for fathead minnow (Pimephalespromelus) based on larval survival. Specific
reductions in larval survival were not reported. In the table, the study is referred to as MRID
00131255 and Driscoll 1982. The paragraph below the table appears to reference the same study
as Ace. 251424. The same data are reported in the BE (EPA 2003b), but without any citation.
DRAFT 331
-------
DRAFT
We located the data report and confirmed that embryo hatchability was not affected by the dose
regime tested: 0, 57, 113, 254, 395, and 972 (ig/L (measured concentrations).
Assessment endpoint: Fish growth
Assessment Measure: Growth rate, weight, length, or biomass of second generation as
measured in chronic toxicity tests
The carbaryl BE (EPA 2003a) did not report any effects on growth for either freshwater or
estuarine/marine species, although reduced growth is listed as the affected endpoint in a fathead
minnow (Pimephalespromelas) study referenced in the Science Chapter (EPA 2003c). For this
study, the NOAEC is given as 210 (ig/L and the LOAEC as 680 |j,g/L However, the fathead
minnow study discussed within the reproductive endpoints summary measured growth of fathead
minnow larvae (length) at 30 and 60 days (Carlson 1971). Although statistical tests were not
used, no differences in growth were apparent between exposed and control fish at the
concentrations tested: 8, 17, 62, 210, and 680 ug/L (Carlson 1971).
The carbofuran BE (EPA 2004) reports larval rainbow trout (O. mykiss) exposed to 56.7 and 88.7
Hg/L carbofuran had significantly reduced lengths and reduced survival at 60, 75, and 90 d
compared to unexposed trout in a partial life cycle test. Additionally, trout weight was
significantly reduced after 75 d of exposure to 56.7 and 88.7 ug/L. Although, no definitive
reference was provided for these data, NMFS obtained the study report to verify study results.
Upon review, several additional adverse effects to fry were noted throughout the experiment. At
day 20, trout exposed to 57.8 and 88.7 (ig/L respired rapidly compared to control fish and were
hyperactive at 88.7 ug/L. These effects continued for the remainder of the experiment (~ 70
days). Following 75 d of exposure to carbofuran at 56.7 and 88.7 ug/L, 23 and 35% of the fish,
respectively, had curved spines. The Science Chapter (EPA 2005) reports growth was the most
sensitive endpoint in two sheepshed minnow studies (Cyprinodon varigatus, MRIDs 40818401
and 42697401), but notes that neither of these tests produced a definitive NOAEC or LOAEC.
The methomyl BE (EPA 2003) lists a NOAEC of 76 ug/L and a LOAEC of 142 ug/L based on
growth endpoints for fathead minnow (Pimephalespromelas), a freshwater species. It also lists a
NOAEC of 260 (J,g/L and a LOAEC of 490 (J,g/L for sheepshead minnow (Cyprinodon
DRAFT 332
-------
DRAFT
varigatus), an estuarine species, but the endpoint affected is described as reproduction and/or
growth, so whether or not growth was affected is uncertain. The Science Chapter (EPA 1998)
does not describe any data for estuarine/marine species, as no guideline test had been submitted
at that time.
Assessment endpoint: Habitat- salmonid prey
Assessment measure: Aquatic invertebrate survival, growth, reproduction from acute and
chronic laboratory toxicity tests
The carbaryl BE (EPA 2003a) lists a range of acute EC50 values for freshwater aquatic
invertebrate survival. Six EC50 values are given for Daphnia magna, ranging from 4.3 -13.0
Hg/L (mean 7.3 |J,g/L). Although not specified as such, given the test organism, the fact that they
are noted as 48 h tests, and the range of percent a.i., we assume that these are guideline tests
conducted by the registrant or an acceptable government lab. Most of the EC50 values are
consistent, with the exception of one test which lists an EC50 of 13.0 [ig/L for a test on a 43.9%
formulation. Discounting this value, mean EC50 is 6.2 [ig/L. ECSOs were also given for three
species of stoneflies, ranging from 1.7-5.6 [ig/L (mean 3.6 ng/L). An EC50 for an amphipod
(Gammarus fasciatus) of 26 [ig/L was also given. There were also EC50 values for a number of
estuarine species. There were five EC50 values given for mysid shrimp (Mysidopsis bahia), a
common guideline test organism, ranging from 5.7-20.2 [ig/L (mean 10.3 ng/L). Discarding the
EC50 value of 20.2 ng/L for a 43.7% formulation, which is also a statistical outlier, the mean
EC50 value for mysid would be 7.8 [ig/L. ECSOs for various other shrimp species were given,
ranging from 1.5-170 [ig/L. All of these tests appear to have been conducted on the a.i.. An
LC50 for blue crab of 320 [ig/L was given. The BE also reported a number of EC50/LC50
values for aquatic insects and other invertebrates derived from the USEPA ECOTOX database.
The carbaryl Science Chapter (EPA 2003c) used an acute value of 5.1 ng/L (stonefly, Chlorperla
grammaticd) as the survival assessment endpoint for freshwater aquatic invertebrates, and
5.7 ng/L (mysid shrimp) as the survival assessment endpoint for estuarine/marine aquatic
invertebrates. The chapter also notes:
"Studies have indicated that acute exposure to carbaryl impacts
predator avoidance mechanisms in invertebrates, reduces overall
zooplankton abundance (Hanazato and Yasuno 1989; Havens
DRAFT 333
-------
DRAFT
1995), and may actually promote phytoplankton growth through
reduced predation by zooplankton."
Acute toxicity data for 1-napthol presented in the BE (EPA 2003a) listed 48 h ECSOs of 700-730
for D. magna, and ECSOs of 200-210 |J,g/L forM bahia. No additional information
regarding acute toxicity of 1-napthol to aquatic invertebrates was provided in the Science
Chapter (EPA 2003 c).
Overall, results presented show that carbaryl and formulations of carbaryl are acutely toxic to a
wide array of aquatic invertebrates in the low (ig/L range, frequently with EC50s/LC50s of less
than 10 ng/L. The degradate, 1-napthol, appears less toxic with respect to comparable
invertebrates; acute survival ECSOs ranged from 200-730 ng/L. However, no data for the genera
more sensitive to the parent carbaryl, such as the stoneflies, caddisflies, or mayflies, are
available. Thus the lower end of the toxicity range is not well established.
The carbaryl BE (EPA 2003a) lists a NO AEC of 1.5 |ig/L and aLOAEC of 3.3 ng/LforZ).
magna based on reproduction and a NO AEC of 500 |J,g/L and a LOAEC of 1,000 |J,g/L for the
midge fly (Chironomous riparius) based on emergence/developmental rate. Specific percentages
of inhibition are not noted. The Science Chapter (EPA 2003c) does not add any additional detail,
but does note:
"midge larvae are benthic macroinvertebrates and exposure may
have been better characterized had it been based on sediment pore
water concentrations as opposed to carbaryl concentrations in
overlying water"
as a potential explanation for the difference in sensitivity between D. magna and C. riparius. No
data regarding reproductive endpoints were presented for 1-napthol.
The carbofuran BE (EPA 2004) included acute ECSOs for the freshwater aquatic invertebrate D.
magna (29-38.6 |J,g/L) and a estuarine/marine invertebrate, pink shrimp (Penaeus duorarum, 4.6-
7.3 |J,g/L). All tests appear to have been conducted with the a.i.. ECSOs of 2.2-2.6 ng/L were
reported in the Science Chapter for the freshwater water flea Ceriodaphnia dubia. The Science
Chapter also provided toxicity values for the freshwater red crayfish (Procambarus clarkii, LC50
2,700 |J,g/L) and the eastern oyster (Crassostrea virginica, EC50 >1,000->5,000 ng/L), Probit
DRAFT 334
-------
DRAFT
slopes were included in Appendix H to Appendix 1 (EPA 2005) for studies where they were
reported or data were available to calculate them. Probit slopes were not available for/), magna
or C. dubia, but were available for the red crayfish (slope 2.91) and the pink shrimp (slope 2.25).
Data provided from chronic toxicity tests included a NOAEC of 9.8 (ig/L and a LOAEC of 27
(ig/L for the freshwater aquatic invertebrate D. magna, and NOAEC of 0.4 (ig/L and a LOAEC
of 0.98 ng/L for the estuarine invertebrate M bahia. In both cases, the most sensitive endpoints
appeared to be survival of the adults and/or growth rather than decreased production of offspring.
The BE (EPA 2004) and Science Chapter (EPA 2005) reference several field studies from the
open literature that examined effects on aquatic invertebrates following application of carbofuran
either to fields or directly to water. One study (Matthiessen, Sheahan et al. 1995) noted complete
mortality of caged amphipods (Gammaruspulex) in a stream draining a field treated with
granular carbofuran at 2.7 Ibs a.i./acre. EPA documents reported a pond enclosure study
(Wayland 1991) noting decreases in amphipod (Hyallela azteca) abundance and biomass and
chironomid biomass at concentrations of 25 (ig/L in a pond enclosure study. EPA also reported
several studies in which carbofuran was applied directly to water caused mortality in aquatic
invertebrates at concentrations ranging from 5-25 (ig/L (Flickinger, Mitchell et al. 1986;
Wayland and Boag 1990; Mullie, Verwey et al. 1991).
The methomyl BE (EPA 2003b) reported acute survival ECSOs for the freshwater invertebrates
water flea (D. magna), amphipod (Gammarus pseudolimnaeus), midge (Chironomousplumosus)
and three genera of stone flies (Skwala sp., Pteronarcella badia, and Isogenus sp.). Most values
appeared to be derived from 48-or 96 h standard laboratory tests. Many of the tests were with
a.i. (95-99% a.i.), but others were conducted with a 24% formulation (EPA 1998). The product
name was not specified. Information in the BE is noted as having come from the EFED
Pesticide Ecotoxicity Database, not the RED Science Chapter (EPA 1998). There are some
inconsistencies between these two documents regarding which EC50 is associated with the
formulation versus the technical a.i. Based on information in the BE, which lists data for a
formulation test on all species, the 24% formulation appears to be more toxic than the technical.
ECSOs for the technical range from 8.8-31.8 |J,g/L for D. magna, and are reported as 920 |J,g/L for
G. psuedolimnaeus, 88 |J,g/L for C .plumosus, 34 |J,g/L for Skwala sp, 69 [ig/L for P. badia, and
DRAFT 335
-------
DRAFT
343 for Isogenus sp. ECSOs for the 24% formulation are reported as 7.6 |J,g/L for D. magna, and
are reported as 720 |J,g/L for G. psuedolimnaeus, 32 |J,g/L for C. plumosus 29 |J,g/L for Skwala sp,
60 ng/L for/1, badia, and 29 for Isogenus sp.
Acute ECSOs are also reported for several species of estuarine/marine shrimp in both the BE
(EPA 2003b) and the RED Science Chapter (EPA 1998). Again, there are inconsistencies
between the documents, with the same ECSOs from formulation tests reported in the BE as 30%
a.i. and in the Science Chapter as 24% a.i.. Some ECSOs were also reported for the a.i. Only one
species, grass shrimp (Palmonetes vulgaris) appears to have been tested with both the a.i. and the
formulation (EPA 2003b). In this case, the formulation (EC50 130 |J,g/L, reported as a 30% a.i.),
appears less toxic than the technical (EC50 49 ng/L). Other ECSOs listed in the BE were for
pink shrimp (P. duorarum, 19 |J,g/L) and mysid shrimp (M bahia, 230 ng/L), both of which
appear to be technical a.i. (90-98.4%).
The methomyl BE (EPA 2003b) reports NOAECs and LOAECs for reproductive endpoints for
two studies conducted on D. magna, both with the technical a.i.. In one study, with a NOAEC of
0.4 ng/L and a LOAEC of 0.8 |J,g/L, the number of young per female was reduced. A second
resulted in a NOAEC of 1.6 ng/L and a LOAEC of 3.1 |J,g/L, based on an unspecified
reproductive endpoint. The BE also reports a NOAEC of 29 |J,g/L and a LOAEC of 59 |J,g/L for
the estuarine mysid shrimp, based on "reproduction and/or growth."
The BE and Science Chapter both report on an outdoor microcosom study conducted with
methomyl (MRID 437444-02). The Science Chapter, Appendix C (EPA 1998), describes the
study design as: "Methomyl was applied to seven treatment groups, at two application rates, at
three different application intervals, over a period of 22 days (pg 51)." Specific application rates
and intervals were not provided, nor was use of a control specifically mentioned. Zooplankton
(Cladocera, Copepodia and Rotifera) abundance and community composition were altered in at
least some treatments, and Ephemeroptera abundance decreased in the two highest treatments.
DRAFT 336
-------
DRAFT
Toxicity of Carbaryl Carbofuran, and Methomyl Degradates
The BEs briefly addressed the issues of degradates. The carbaryl BE (EPA 2003a) provides
acute fish and invertebrate toxicity data for 1-napthol, the principal hydrolysis degradate of
carbaryl. The LCSOs for three species offish tested (freshwater and marine estuarine) range
from 750 - 1,800 |J,g/L. LCSOs presented for aquatic invertebrates include 700-730 |J,g/L for D.
magna (freshwater organism), 200-210 |J,g/L forM bahia (estuarine organism) and 2,100 |J,g/L
for C. virginica (estuarine organism).
During our open literature review, we located a study that compared acute lethalities between
carbaryl and 1-naphthol in two species offish (Shea and Berry 1983). In goldfish (Carassius
auratus) and killifish (Fundulus heteroclitus), 1-naphtol was significantly more toxic than
carbaryl based on 10 d acute lethality tests (Shea and Berry 1983). The degradate 1-naphthol
was approximately five times more toxic than carbaryl in goldfish, and in killifish twice as toxic
as carbaryl (Shea and Berry 1983). Additionally, fish exposed to 1-naphthol showed
neurological trauma including pronounced erratic swimming behaviors and increased opercula
beats following exposure to 5 and 10 mg/L. None of these symptoms were observed in the
carbaryl treatments (Shea and Berry 1983).
Other degradates identified in fate studies submitted to EPA included 5-hydroxy-l-
napthylmethylcarbamate (aerobic soil metabolism and anaerobic aquatic); 1-naphthyl
(hydroxymethyl) carbamate (aerobic soil metabolism and anaerobic aquatic); 1,4-
napththoquinones (degradates of 1-napthol); 4-hydroxy-l-napthyl methylcarbamate (anaerobic
aquatic); 1,5-naphthalenediol (anaerobic aquatic); and 1,4-naphthalenediol (anaerobic aquatic).
No toxicity information was presented for these degradates.
The carbofuran BE (EPA 2004) describes three degradates: 3-hydroxycarbofuran, 3-
ketocarbofuran, and carbofuran 7-phenol in the toxicity section, and refers to two additional
degradates: 3-hydroxy-7-phenol, and 3-keto-7-phenol in the section on environmental fate and
transport. No toxicity data are included, and the document noted that "inclusion of
environmental transformation products in the risk analysis of carbofuran would not be expected
DRAFT 337
-------
DRAFT
to result in substantive changes to conclusions drawn using the parent alone (pg 30)" (EPA
2004).
Based on the more detailed fate information in the Science Chapter for the Carbofuran RED
(EPA 2005), carbofuran 7-phenol is a hydrolysis product (up to 75% of applied). Carbofuran 7-
phenol is expected to be less toxic than parent carbofuran, as the toxic moiety (the carbamylating
radical) has already dissociated (EPA 2004). No toxicity data for this compound are provided in
either the BE or the Science Chapter. The degradates 3-hydroxycarbofuran and 3-
ketocarbofuran are detected in soil photolysis and aerobic soil metabolism studies. These
compounds were each approximately 3-5% of applied (EPA 2005), and are structurally more
similar to the parent carbofuran. The Science Chapter cites an open literature study using
Microtox (Kross, et. al., as cited in EPA 2005) noting "3-ketocarbofuran appears as toxic or
slightly more toxic than the parent, but 3-hydroxycarbofuran is much less toxic." A second
study evaluated (Gupta 1994, as cited in EPA 2005) indicates 3-hydroxycarbofuran is "equally as
toxic as the parent." No specific toxicity values were presented. Other degradates mentioned in
the fate summary are 3-hydroxycarbofuran phenol (3-hydrocy-7-pheonl) and 3-ketocarbofuran
phenol (3-keto-7-phenol), which are less than 5% of applied (EPA 2005).
The methomyl BE notes "a degradate (thiolacetohyroxamic acid 5-methyl ester)... was tested
and found to be practically nontoxic to bluegill" (EPA 2003b). This study is also mentioned in
the Environmental Risk Assessment for the RED (EPA 1998). The bluegill LC50 for this
degradate is 462,000 ng/L. No other degradate toxicity data were presented. The fate portion of
the BE (EPA 2003b) and the Environmental Risk Assessment (EPA 1998) both reference S-
methyl-N-hydroxythioacetamidate as a product of hydrolysis in both water and soil. This may
be the same compound expressed under different naming conventions, but no structures were
provided to confirm this identification. In laboratory tests, methomyl photolyses rapidly in water
(ti/2=l d), but hydolyses more slowly (stable at pH 7, ti/2=30 days at pH 9) (EPA 1998). In a
water-sediment system study, methomyl estimated system half-life was 4-5 d (EPA 1998).
Major degradates in soil and aquatic systems include acetonitrile (17-66%), acetonitrile (-14%,
water-sediment study), and CC>2 (EPA 1998). No toxicity data were presented for acetonitrile or
acetamide.
DRAFT 338
-------
DRAFT
Formulations and other (inert) ingredients found in carbaryl's, carbofuran's, and methomyl's
formulations
Assessment endpoint: Fish survival, aquatic invertebrate survival,and primary production
Assessment measure: Aquatic invertebrate survival, growth, and reproduction from acute
and chronic laboratory toxicity tests
The carabaryl BE (EPA 2003a) and the carbaryl Science Chapter (EPA 2003c) provide some
toxicity data on formulations containing 5-81.5% carbaryl for both aquatic invertebrates and fish.
We did not receive information on whether the formulations tested are currently registered. Data
for formulations (referred to in the Science Chapter (EPA 2003c) as technical end-product or
TEP) are contained in Appendix D-l. The technical a.i. EC50 for water flea (D. magna), based
on a single test, is 5.6 ng/L. No 95% confidence interval is given for the test. Data for several
formulations with 43.7-81.5% carbaryl are presented. ECSOs for the formulations range from
4.3-13.0 ng/L. Data are also available to compare toxicity of the technical grade to formulations
(43.7-81.5%) for the estuarine mysid shrimp, M. bahia. No 95% confidence intervals are given.
Two technical a.i. LCSOs were provided in the data, 5.7 ng/L and 6.7 [ig/L. Tests with
formulations (n=3) resulted in LCSOs of 9.3-20.2 [ig/L. No chronic tests, which evaluate
reproduction and growth, were described for any formulations.
For carbaryl, acute LC50 formulation data are also given for two species offish, rainbow trout
(O. mykiss, 44-81.5 % carbaryl) and bluegill sunfish (Lepomis macrochirus, 5-50% carbaryl).
No 95% confidence intervals are given. Although other fish data for the technical carbofuran are
provided, a same-species comparison of toxicity data is better than evaluating against a range of
species, as the various species will exhibit a range of sensitivity. A single a.i. LC50 (1,200 ng/L)
is given for rainbow trout. A total of four LCSOs are given for formulations, ranging from 1,400-
4,500ng/L. Three a.i. LCSOs for bluegill are given, ranging from 5,000-14,000 [ig/L. The
bluegill LCSOs (n=4) for formulations range from 9,800-290,000 (ig/L. Based on these data, it
appears the formulations tested are relatively similar in toxicity to the a.i. on an acute lethality
basis (survival endpoint). No chronic tests, which evaluate reproduction and growth, were
described for any formulations.
DRAFT 339
-------
DRAFT
The carbofuran BE (EPA 2004) lists some formulation toxicity data for aquatic invertebrates,
referenced to the EFED Science Chapter and other data referenced to the EPA Acquire Database.
In some cases, the formulation data referenced to the Science Chapter provides percent a.i. and in
others it contains designations such as "5G" or "50 WP". In cases where the percent a.i. is
reported as a number, we presume the toxicity value has been corrected to percent a.i. It is
unclear whether the others have been corrected, whether the data from the Acquire database has
been corrected, or the specific products with which the toxicity data are associated.
Toxicity data were presented in the carbofuran Science Chapter (EPA 2005) were found in
Appendix H of Appendix 1 to the main document, titled Animal Toxicity Tests with DERs.
Study data on formulations were presented for bluegill sunfish (L. macrochirus) and rainbow
trout (O. mykiss). A single test on rainbow trout was conducted using the formulation Furadan
75WP with a reported LC50 of 458 |J,g/L. LCSOs for rainbow trout tested with technical
carbofuran ranged from 362-600 |j,g /L. A number of different formulations (Furadan 4F,
Furadan 10G, Furadan 50WP and Furadan 75WP) were tested with bluegill sunfish. LCSOs
ranged from 240-488 |j,g carbofuran/L. LCSOs for bluegill sunfish tested with technical
carbofuran ranged from 88-126 |j,g a.i/L. Only one formulation test for freshwater invertebrates
was reported. D. magna exposed to "5G" had a survival EC50 of 41 |j,g /L. Tests with D. magna
on technical carbofuran produced survival ECSOs of 29-38.6 |j,g /L. The estuarine fish Atlantic
silverside (Mendia menidid) was tested in two formulations, Furandan 4F and Furandan 15G.
LCSOs from these tests were 36 |j,g a.i./L. and 64 |j,g a.i./L respectively. A test with technical
grade on the same species produced an LC50 of 33 |j,g a.i./L. (95% CI 27-41 |j,g a.i./L). The
estuarine/marine invertebrate pink shrimp was tested both in a formulation (Furandan 15G, EC50
13.3 |j,g a.i./L and with technical (ECSOs 4.6-7.3 |j,g a.i./L, n=2). Based on these data, there is
no indication that the carbofuran formulations tested are more toxic than the a.i. on survival as an
endpoint). No long term studies, such as life cycle studies, with formulations were reported for
either fish or invertebrates.
The methomyl BE (EPA 2003b) and Science Chapter for the RED (EPA 1998) both reported
toxicity data for a 24% formulation and a 29% formulation for both fish and aquatic
invertebrates. In addition, the BE reported some toxicity data for a 30% formulation. While it is
DRAFT 340
-------
DRAFT
difficult to make a definitive comparison given the structure of the data, the 24% formulation
appears more toxic to both fish and invertebrates. Neither the toxicity section of the BE nor the
use characterization in the Science Chapter list specific products. Thus, we cannot cross-
reference the data to a product or determine if it is a currently registered product.
Identified data gaps and uncertainties of carbofuran's, carbaryl's and methomyl's'
toxicity information presented in BEs and REDs:
Overall, data provided in the BEs and their Science Chapters were insufficient to allow a
thorough evaluation of all identified assessment endpoints and measures considered by NMFS.
The Data Evaluation Reviews (DERs) or the original studies may have helped reduce some of
the uncertainties related to experimental design, dose-response slopes, and confidence estimates
of the data, but we did not receive summaries of this information from EPA when preparing this
Opinion. However, even if the DER data were submitted, other aspects of EPA's assessment
endpoints (survival, reproduction, and growth) were not presented. When this missing
information is combined with the absence of information presented on other non-assessed
endpoints such as AChE inhibition in fish and invertebrates, swimming behaviors, and olfactory-
mediated impairments, an incomplete picture emerges on the potential effects of EPA's action to
listed salmonid and habitat endpoints.
Toxicity information presented in the BEs (EPA 2003a, EPA 2003b, and EPA 2004) lacks details
NMFS requires for analyses in support of this consultation. More complete information is
presented in the Science Chapters for carbaryl (EPA 2003c) and carbofuran (EPA 2005), yet the
information was not applied by EPA in its characterization of risk to listed salmonids or
salmonid habitats. Specific data gaps identified include the following, although not all gaps
apply to all three carbamates.
• Reported LCSOs not accompanied by slopes, experimental design (number of treatments
and replicates, life stage of organism, concentrations tested), measures of variability such
as confidence intervals or standard deviations/errors;
• No analysis of the degree or magnitude of inhibition of acetylcholinesterase by the three
carbamates and expected response by listed salmonids or aquatic, salmonid prey
communities;
• Summary and discussion offish sublethal data absent from BEs including effects to
swimming and chemoreception;
DRAFT 341
-------
DRAFT
• Limited or no toxicity data on current formulations,
• Limited or no toxicity data presented for identified surface water degradates of the three
carbamates;
• Sensitivity of surrogate lab strains compared to wild, listed fish, particularly comparisons
between warm and cold water fish species used in chronic guideline tests;
• No data summarized for mixture toxicity including tank mixtures and environmental
mixtures to assessment endpoints;
• No toxicity data presented on "inert" or other ingredients present in formulations
containing each of the a.i.s;
• No analysis of environmental factors (pH, temperature) influence on exposure and
toxicity.
Summary of Toxicity Information from Open Literature
To organize the available toxicity information on listed salmonids and habitat, we developed risk
hypotheses with associated assessment endpoints as described in the Approach to the Assessment
section. Recall that assessment endpoints are biological attributes of salmonids and their habitat
potentially susceptible to the stressors of the action. In addition to toxicity data presented in the
BEs, we also considered information from other sources to evaluate both individual and
population level endpoints. The results of those studies are summarized below under
corresponding assessment endpoints. We qualitatively assigned the most significance to study
results that were: 1) derived from experiments using salmonids (preferably listed Pacific
salmonids or hatchery surrogates); 2) measured an assessment endpoint of concern e.g.,
survival, growth, behavior, reproduction, abundance etc., identified in a risk hypothesis; 3)
resulted from exposure to stressors of the action or relevant chemical surrogates (i.e., other
AChE inhibitors); and 4) had no substantial flaws in the experimental design. When a study did
not meet these criteria, we highlighted the issue(s) and discussed how the information was used
or why the information could not be used.
Assessment endpoint: Swimming
Assessment measures: Burst swimming speed, distance swam, rate of turning, baseline speed,
tortuosity of path, acceleration, swimming stamina, and spontaneous swimming activity,
Swimming is a critical function for anadromous salmonids that is necessary to complete their life
cycle. Impairment of swimming may affect feeding, migrating, predator avoidance, and
DRAFT 342
-------
DRAFT
spawning (Little and Finger 1990). It is the most frequently assessed behavioral response of
toxicity investigations with fish (Little and Finger 1990). Swimming activity and swimming
capacity of salmonids have been measured following exposures to a variety of AChE-inhibiting
insecticides including the carbamates carbaryl and carbofuran and a variety of OP insecticides.
Swimming capacity is a measure of orientation to flow as well as the physical capacity to swim
against it (Howard 1975; Dodson and Mayfield 1979). Swimming activity includes
measurements of frequency and duration of movements, speed and distance traveled, frequency
and angle of turns, position in the water column, and form and pattern of swimming. A review
paper published in 1990 summarized many of the experimental swimming behavioral studies and
concluded that effects to swimming activity generally occur at lower concentrations than effects
to swimming capacity (Little and Finger 1990). Therefore, measurements of swimming activity
are usually more sensitive than measurements of swimming capacity. A likely reason is that
fishes having impaired swimming to the degree that they cannot orient to flow or maintain
position in the water column are moribund (i.e., death is imminent). The authors of the review
also concluded that swimming-mediated behaviors are frequently adversely affected at 0.3 -
5.0% of reported fish LCSOs3, and that 75% of reported adverse effects to swimming occurred at
concentrations lower than reported LCSOs (Little and Finger 1990). Both swimming activity and
swimming capacity are adversely affected by AChE-inhibiting insecticides.
We located studies that measured impacts to salmonid swimming behaviors from exposure to
carbaryl. Several studies also measured AChE inhibition from OPs and provided correlations
between AChE activity and swimming behaviors. We did not locate any studies that tested
mixtures of AChE-inhibiting insecticides on swimming behaviors of any aquatic species.
Carbaryl
Experiments with carbaryl have shown that cutthroat trout's swimming abilities are
compromised by 6 h exposures to 750 and 1,000 |J,g/L, resulting in increased susceptibility to
predation (Labenia, Baldwin et al. 2007). Cutthroat trout swimming capacity was not
The current hazard quotient-derived threshold for effects to threatened and endangered species used by EPA is 5 % (1/20"1) of the lowest fish
LC50 reported. If the exposure concentration is less than 5 % of the LC50 a no effect determination is made which likely underestimates risk to
listed salmonids based on swimming behaviors.
DRAFT 343
-------
DRAFT
statistically significantly affected by carbaryl at 500 ug/L, although muscle AChE activity was
29% relative to unexposed cutthroat. More sensitive swimming activity measurements were not
evaluated. At 750 and 1,000 ug/L, AChE activity was 24 and 23% relative to unexposed fish,
respectively. A known predator of juvenile cutthroat, lingcod, consumed on average more
cutthroat that were exposed to carbaryl compared to those that were not exposed to carbaryl.
The experimental design ensured that the predator was exposed to minimal concentrations when
carbaryl-exposed fish were transferred in a large experimental chamber along with unexposed
cuthroat. In the predation experiment with lingcod, cutthroat were exposed for 2 h to 200, 500,
and 1,000 [ig/L carbaryl. Results indicated a dose-dependent decrease in the ability of carbaryl-
exposed trout to avoid being eaten by the lingcod predator. At 200 |J,g/L, an increase in
predation was evident, although not statistically significant. At 500 and 1,000 ug/L carbaryl,
cutthroat trout were consumed at statistically significantly higher rates that unexposed fish.
Cutthroat trout's AChE activity was also reduced in a dose-dependent fashion showing greater
reductions with increasing carbaryl concentrations. Six hour exposures to cutthroat trout
significantly reduced AChE activity in brain and muscle where 50% reductions (IC50) in brain
AChE occurred at 213 |J,g/L carbaryl. The onset of inhibition occurred within 2 hours and was
near maximal. Benchmark concentrations corresponding to 20% inhibition were 32 |J,g/L in
brain and 23 |j,g/L in muscle tissues. Recovery of AChE activity took 42 hours at 500 (ig/L
carbaryl to return to pre-exposure levels. We ranked these sets of experiments as highly relevant
to understanding the effects of carbaryl on inhibition of AChE and subsequent salmonid
swimming behaviors. The test also provided information on lack of predator avoidance
behaviors by salmonids.
Catfish (Mystus vittatus) showed increased swimming activity following 72 h of exposure to
12,500 |J,g/L and no mortalities were noted (Arunachalam, Jeyalakshmi et al. 1980). The 72 h
LC50 for catfish was 17,000 |J,g/L, indicating this species of catfish is much less sensitive to
acute concentrations of carbaryl than salmonids. Other sublethal endpoints were also measured
and were affected by carbaryl, including food intake, growth, metabolism, and rate of opercular
beats. Catfish increased their rates of opercular beats from 74 per minute exposed to freshwater
alone to 124 per minute when exposed to 12,500 |j,g/L carbaryl, which the authors attributed to
cute stress from carbaryl. The study showed that swimming activity is affected prior to the onset
DRAFT 344
-------
DRAFT
of death and that it is a more sensitive endpoiont than lethality. We ranked this study as relevant
because swimming was measured, but the high concentrations used, lack of chemical
verification, and lack of compatibility to listed salmonids introduces uncertainty.
Multiple swimming related behavioral responses in rainbow trout fry (0.5-1 g) were assessed
following 96 h exposures to carbaryl at 10, 100, and 1,000 |J,g/L; nominal concentrations (Little,
Archeski et al. 1990). These included swimming capacity (cm/sec), swimming activity (sec),
prey strike frequency, daphnids consumed, percent consuming daphnids, and percent survival
from predation. All endpoints were significantly affected at 1,000 [ig/L carbaryl relative to
unexposed fry. At 10, 100, and 1,000 [ig/L carbaryl, significantly more rainbow trout were
consumed relative to unexposed fish. By using very young fry, the studies also provide
information on a sensitive early lifestage where swimming behaviors are critical to survival i.e.,
feeding and predator avoidance. We ranked these studies as highly relevant to a variety of
essential swimming-associated behaviors.
One study tested whether schooling offish in estuarine waters was affected by a single short-
term exposure of carbaryl at 100 |j,g/L (Weis and Weis 1974). Carbaryl impaired schooling of
Atlantic silversides (Menidia medidid) by increasing the area used by up to twice that relative to
control groups. Recovery took three days for fish to attain similar schooling behaviors as the
control treatment (Weis and Weis 1974). The authors suggested that increased areas of
schooling fish would increase energy expenditures of individual fish and also increases rates of
predation. The study lacked analytical verification of exposure concentrations; only one
concentration was tested; and the study was conducted with a non-salmonid fish. The study is
relevant because it addressed an important swimming behavior i.e., schooling, which juvenile
salmonids use at times, and presented data on time to recovery following an adverse behavioral
affect. However, it remains uncertain at which concentrations juvenile salmonids that school
would be affected.
Neurological effects on the startle response and ability to avoid predation by juvenile medaka
were evaluated following exposures of 2.5, 5.1, 7.0, and 9.4 mg/L carbaryl (Carlson, Bradbury et
al. 1998). At 5.1 mg/L and higher, carbaryl increased the time between motor neuron peak and
DRAFT 345
-------
DRAFT
initiation of muscle activity (i.e., swimming response) and at 7.0 mg/L and higher response to
stimuli ratios were also higher. Predation trials on exposed medaka showed no differences
between fish exposed to carbaryl and unexposed fish. While these concentrations are extremely
high compared to concentrations affecting salmonids, the results indicate that neurological-
associated swimming behaviors are affected by carbaryl. Given that the 48 h LC50 of juvenile
medaka is estimated at 9.4 mg/L and that detectable sublethal neurological effects occurred at 5.1
mg/L, roughly half the LC50, it is uncertain at what concentrations juvenile salmonids
neurological endpoints would be affected. We ranked this experiment as relevant as it provided
evidence that neurological effects manifest at lower concentrations than 48 h LCSOs.
Carbofuran
Carbofuran adversely affected swimming behaviors in goldfish (Carassius auratus) following 24
and 48 h exposures to the lowest concentration tested, 5 (ig/L (Bretaud, Saglio et al. 2002).
Swimming activity (fish swimming from one zone to another), the least sensitive endpoint, was
significantly affected at 500 |j,g/L carbofuran, while burst swimming, the most sensitive
endpoint, was significantly affected at 5 |j,g/L following 24 h exposure (Bretaud, Saglio et al.
2002). Burst swimming behavior, sheltering, and nipping in goldfish were significantly
increased by a 4 h exposure to 1 (ig/L carbofuran (Saglio, Trijasse et al. 1996). At 12 h
exposures, significant effects were observed at 100 (ig/L for sheltering, nipping, and burst
swimming. Grouping of goldfish showed non dose-dependent responses as significant effects
(p<0.05) appeared at 10 ng/L, but were absent at 1 and 100 (ig/L (Saglio, Trijasse et al. 1996).
We ranked both of these studies as relevant, but due to difficulties inherent in translating the
observed behavioral effects in goldfish to salmonids combined with the lack of analytical
verification of concentrations, the study did not receive a highly relevant ranking. The results
support that ecologically relevant swimming-related behaviors are impacted at lower
concentrations than more coarse measures of swimming such as swimming stamina or swimming
capacity.
Methomyl
We were unable to locate any studies that evaluated effects to swimming related behaviors in
fish. Although this is a data gap, we assume methomyl does inhibit swimming in the same
DRAFT 346
-------
DRAFT
fashion as the other carbamates because swimming behaviors are typically affected when AChE
is sufficiently inhibited e.g. by 30% or more. A review also concluded that swimming-mediated
behaviors are frequently adversely affected at 0.3 - 5.0% of reported fish LCSOs, and that 75%
of reported adverse effects to swimming occurred at concentrations lower than reported LCSOs
(Little and Finger 1990).
Other AChE inhibiting insecticides effects on swimming and related behaviors-
We also reviewed study results conducted with other carbamate and OP insecticides because
both classes of compounds share the same mode of toxic action, inhibition of AChE. Recovery
of swimming to pre-exposure levels is much more rapid in carbamate-affected fish compared to
OP-affected fish due to the reversible binding of carbamates with AChE (Mineau 1991). In one
study with carbaryl and rainbow trout, AChE activity returned to levels measured in control
animals after 24 h in clean water (Zinkl, Shea et al. 1987) and in another study within 42 h in
cutthroat trout (Labenia, Baldwin et al. 2007). In the latter study, cutthroat trout recovered
approximately half of their pre-exposure AChE activity within 6 h, following peak AChE
inhibition within 2 h. Both of these studies indicate that recovery of AChE activity following
exposure to carbamates occurs quickly and but is highly dependent on environmental exposure
conditions found in the aquatic habitats.
We did not locate additional studies with other carbamates, but located multiple studies with OP
insecticides. The OPs chlorpyrifos, diazinon, and malathion showed significant and persistent
effects to a suite of swimming related behaviors in salmonids at concentrations expected in
salmonid habitats; as reviewed in NMFS' November 18, 2008 Opinion on these three OP
insecticides (NMFS 2008). Robust evidence of the three OPs showed reductions in swimming
speed (Brewer, Little et al. 2001), distance swam (Brewer, Little et al. 2001), acceleration
(Tierney, Casselman et al. 2007), food strikes (Sandahl, Baldwin et al. 2005) and significant
correlations with AChE activity (Brewer, Little et al. 2001; Sandahl, Baldwin et al. 2005).
Additionally, other OPs including fenitrothion, parathion, and methyl parathion, adversely
affected a suite of swimming behaviors reviewed in (Little and Finger 1990). One noteworthy
study investigated the effects of six pesticides including methyl parathion (OP) and DEF (OP) on
rainbow trout swimming behavior (Little, Archeski et al. 1990). All insecticides adversely
affected spontaneous swimming activity, while DEF also reduced swimming capacity in juvenile
DRAFT 347
-------
DRAFT
rainbow trout (Little, Archeski et al. 1990). In bluegill methyl parathion adversely affected burst
swimming behavior at 300 (ig/L (Henry and Atchison 1984). Respiratory disruptions, comfort
movements, and aggression behaviors in bluegill were all adversely affected by 24 h exposures
to methyl parathion at 3.5 ng/L. This suggests that these social behaviors are very sensitive to
AChE inhibition (Henry and Atchison 1984).
Two month old juvenile rainbow trout, brook trout, and coho were exposed to malathion
(Phillaps Malathion 55%) for 7-10 days depending on species (Post and Leasure 1974).
Swimming performance, brain AChE activity, and recovery time were measured following
exposure to malathion concentrations of 0, 40, 90, 120 [ig/L in brook trout; 0, 55, 112, 175 [ig/L
in rainbow trout; and 0, 100, 200, 300 [ig/L in coho. Additionally, once fish recovered AChE
activity, they were subjected to a second exposure to determine if prior exposure altered
susceptibility to malathion. Swimming performance and AChE activity did not differ from
values of the initial exposure i.e., a second exposure resulted in no evidence of increased
susceptibility. Brook trout were the most sensitive based on AChE inhibition followed by
rainbow trout and coho salmon, respectively. AChE inhibition of 25% relative to control fish
occurred at 40 |j,g/L (brook trout), 55 (ig/L (rainbow trout), and 100 (ig/L (coho). Coho required
at least twice the concentration of malathion compared to brook and rainbow trout to inhibit
AChE activity. Swimming performance was affected at the lowest concentrations tested in each
salmonid species and showed a dose-dependent decrease in swimming performance as malathion
concentration increased. The data indicated that AChE inhibition of approximately 20-30%
resulted in a 5% or less reduction in swimming performance and as inhibition increased,
swimming performance decreased. Note, however, that the swimming test conducted in the
study is a coarse measure of swimming capacity. Thus, other non-measured swimming activity
endpoints would likely be affected at lower concentrations (Little and Finger 1990; Little,
Archeski et al. 1990). Recovery of AChE in exposed salmonids took 25 d for brook trout, 35 d
for rainbow trout, and 42 d for coho. There was no difference in recovery time based on
concentrations tested within species. Post and Leasure (1974) concluded, "these figures are
significant in that they point out the need for spacing malathion insecticide usage in ecosystems
where this insecticide is used at intervals during a growing season." Additionally, Post and
Leasure (1974) emphasized that where OP insecticides are used, "their effect must also be taken
DRAFT 348
-------
DRAFT
into consideration". We ranked this experiment as relevant as several salmonid species were
tested using a rigorous experimental design, although validation of malathion concentrations was
not performed.
In summary, the information presented on swimming behaviors from AChE-inhibiting
insecticides provide a weight of evidence that carbamates (and OPs) adversely affect swimming
behaviors at sublethal concentrations which can reduce the fitness of affected salmonids.
Assessment endpoints: Olfaction and olfactory-mediated behaviors: Predator avoidance, prey
detection and subsequent growth, imprinting of juvenile fish to natal waters, homing of adults
returning from the ocean, and spawning/reproduction
Assessment measures: Olfactory recordings (electro-olfactogram), behavioral measurements
such as detection of predator cues and alarm response, adult homing success, AChE activity in
olfactory rosettes and bulbs, and avoidance/preference
The olfactory sensory system in salmonids is particularly sensitive to toxic effects of metals and
organic contaminants. This is likely a result of the direct contact of olfactory neurons and
dissolved contaminants in surface waters. Olfactory-mediated behaviors play an essential role in
the successful completion of anadromous salmonid life cycles, and include detecting and
avoiding predators, recognizing kin, imprinting and homing in natal waters, and reproducing. It
is well established that Pacific salmon lose navigation skills when olfactory function is lost and
consequently are unable to return to natal streams (Wisby and Hasler 1954).
We located studies that measured olfactory responses offish to carbofuran and carbaryl. We
found no studies with methomyl. Several studies with other carbamates were found, but most
were with thiocarbamates which have a different mode of action than TV-methyl carbamates, i.e.,
they do inhibit acetylcholineterase. We do not discuss or use studies with OP-induced olfaction
in salmonids because AChE inhibition does not appear to be the putative mode of action
affecting olfaction, although more empirical data are needed to confirm this. Below we discuss
the available literature on carbamate effects to fish olfaction.
DRAFT 349
-------
DRAFT
The olfactory activity of juvenile cutthroat appeared unresponsive to concentrations of carbaryl
following 10 second pulses across the olfactory epithelia and juveniles showed no
preference/avoidance to carbaryl (Labenia, Baldwin et al. 2007). No statistically significant
departures relative to unexposed fish were observed at 5, 50, or 500 |j,g/L carbaryl from
neurophysiological recordings i.e., electro-olfactograms. Additionally, in a behavioral avoidance
assay, cutthroat trout did not avoid seawater containing carbaryl at 500 ng/L. These results
suggest that cutthroat trout do not actively avoid carbaryl and that short term exposures do not
affect olfaction at the concentrations tested. We found no other studies that evaluated other
salmonids in estuarine conditions or in freshwaters. We ranked these study results as highly
relevant to the effects of carbaryl on salmonid olfaction and an olfactory-mediated behavior,
avoidance.
In one set of experiments, coho salmon exposed for 30 minutes to three carbamates (carbofuran,
antisapstain IPBC, mancozeb), separately, had reduced olfactory ability as well as disruption of
normal AChE activity (Jarrard, Delaney et al. 2004). Carbofuran reduced olfaction by 50%
(EC50) at 10.4 ng/L; IPBC reduced olfaction at 1.28 ng/L (EC50); and mancozeb reduced
olfaction at 2.05 mg/L (EC50). All three carbamates also affected AChE activity with highly
variable results. Carbofuran reduced AChE activity in the olfactory receptor at 10 [ig/L (45%)
and statistically significantly at 200 [ig/L (68%), but none of the treatments (0.1, 10, or
200 ng/L) reduced brain or olfactory bulb AChE activity. This could be a result of the limited
exposure duration. Coho salmon brain AChE activities increased following 30 min exposures to
mancozoceb and IPBC, while olfactory measurements showed reductions (i.e., inhibition of
olfaction). Mancozeb is a manganese zinc ethylene-^/5 -dithiocarbamate that is used as a
fungicide and mildewcide. IPBC is a dithiocarbamate used as a wood preservative and currently
its mechanism of action remains unknown. The authors concluded that too little information
exists to develop a causal relationship between AChE inhibition and olfaction (Jarrard, Delaney
et al. 2004). The data do show that carbofuran inhibited both AChE activity and olfaction at 10
and 200 ng/L, respectively. We ranked these study results as highly relevant to carbofuran's
effect on salmonid olfaction.
DRAFT 350
-------
DRAFT
Goldfish showed no avoidance of carbofuran-contaminated water up to a concentration of 10
mg/L, but at the onset of the contaminated flow goldfish showed increased and immediate burst
swimming responses relative to unexposed fish (Saglio, Trijasse et al. 1996). In the same study,
olfactometric tests were performed to assess the influence of 1, 10, and 100 |j,g/L carbofuran on a
suite of behavioral responses to a food extract (a chironomid-containing solution). Sheltering,
grouping, nipping, burst swimming, and attraction behaviors were assessed for ten minutes at 4,
8, and 12 h time points. Significant effects to goldfish behavior in the presence of food stimuli
were observed as low as 1 ng/L carbofuran. Sheltering activity increased, while attraction
decreased in a dose-dependent manner at 4 h across treatment range (1-100 ug/L) relative to
unexposed fish. At 10 and 100 ug/L, goldfish nipped more, relative to controls; representative of
increased aggression and stress. The authors concluded that olfactory-mediated behaviors in
goldfish were adversely affected by sublethal concentrations of carbofuran. We ranked this
study as relevant to the assessment endpoint of olfaction as ecologically relevant olfactory-
mediated behaviors were tested with sublethal concentrations of carbofuran. We note that
concentrations were not analytically verified and direct correlation of olfactory-affected goldfish
behaviors to salmonid behaviors remains uncertain.
The olfactory ability of male Atlantic salmon parr to detect a female priming hormone was
measured following 5 d exposures to carbofuran (1.1, 2.7, 6.5, 13.9, 22.7 [ig/L ). The response
to prostaglandin F2a by male olfactory epithelium was statistically significantly reduced at
carbofuran concentrations as low as 1.0 ug/L, while the threshold for detection was reduced 10-
fold. Furthermore at 2.7 [ig/L carbofuran male fish completely lost priming ability induced by
the female pheromone resulting in no increase in milt or plasma steroids at 2.7 [ig/L. These
results suggest that Atlantic salmon exposed to these effect concentrations would have difficulty
preparing for spawning due to their impaired priming. Reductions in productivity are possible if
male fish miss spawning opportunities. We ranked this study as highly relevant to the effects of
carbofuran on olfactory-mediated behaviors such as spawning synchronization.
Mixtures containing carbaryl, carbofuran, and methomyl
We located no experiments that tested mixtures of the three a.i.s, nor did we find mixture studies
with other TV-methyl carbamates. Therefore, the potential for mixture toxicity to olfactory
DRAFT 351
-------
DRAFT
endpoints of salmonids remains a recognized data gap. Given the differences in olfactory
toxicity in salmonids for carbaryl (no observed toxicity) and carbofuran (inhibited olfaction at 10
Hg/L) combined with no available information on methomyrs affect on olfaction, no apparent
dose-response pattern of toxicity emerges for the three a.i.s.
Assessment endpoints: Toxic effects in salmonids from consuming contaminated prey
Assessment measures: Survival, swimming performance
A current uncertainty is the degree to which secondary poisoning of juvenile salmonids may
occur from feeding on dead and dying drifting insects. Secondary poisoning is a frequent
occurrence with OPs and carbamates in bird deaths (Mineau 1991), yet is much less studied in
fish. Uptake, metabolism, and accumulation of carbaryl by a salmonid prey item, Chironomus
riparius (midge), exposed for 24 h indicated significant uptake over the first 8 h, significant
metabolism (more than 85-99%) of parent carbaryl to metabolites and low bioconcentration
factors (5-10) (Lohner and Fisher 1990). These results suggest that contaminated prey items,
such as aquatic invertebrates, do not accumulate significant carbaryl, and what they do
accumulate is likely rapidly metabolized. That said, juvenile salmonids could still get a dose of
carbaryl from feeding on drifting, contaminated insects that have not had time to metabolize
carbaryl. Juvenile brook trout gorged on drifting insects following applications of carbaryl and
AChE activity was reduced (15-34%) in the trout (Haines 1981). However, it is not possible to
differentiate the contribution to AChE inhibition from the aqueous and dietary routes because
concentrations were not measured in the water, prey, or fish. In another study, resident trout
feeding on dying and dead drifting invertebrates (from the pyrethroid cypermethrin) caused a
range of physiological symptoms in brook trout: loss of self-righting ability and startle response;
lethargy; hardening and haemolysis of muscular tissue similar to muscle tetany; and anemic
appearance of blood and gills (Davies and Cook 1993). The possibility that the adverse effects
in the trout manifested from exposure to the water column instead of from feeding on
contaminated prey was ruled out by the authors as measured field concentrations of pesticides
did not produce known toxic responses. In a laboratory feeding study with the OP fenitrothion,
brook trout (S. fontmalis) were fed contaminated pellets (1 or 10 mg/g fenitrothion for four wks)
(Wildish and Lister 1973). Growth was reduced in both treatments. AChE inhibition was
measured at 2, 12, and 27 d following termination of contaminated diet treatments. Trout had
lower AChE activity than unexposed fish at both treatments, and by 27 d following termination,
DRAFT 352
-------
DRAFT
contaminated diet-induced AChE levels regained some of their activity. The treatment
concentrations used in this study are very high and indicate that brook trout are not sensitive to
diet-induced toxicity of fenitrothion. The experiment did show that AChE inhibition from the
diet is possible, yet it is difficult to determine the relative toxicity of carbaryl, carbofuran, and
methomyl found in contaminated insects consumed by Pacific salmonids.
Habitat assessment endpoints:
Prey survival, prey drift, nutritional quality of prey, abundance of prey, health of aquatic prey
community, and recovery of aquatic communities following N-methyl carbamate exposures
Assessment measures: 24, 48, and 96 h survival of prey items from laboratory bioassays
reported as EC/LC50s; sublethal effects to prey items; field studies on community abundance;
indices of biological integrity (IBI); community richness; and community diversity.
Death of aquatic invertebrates in laboratory toxicity tests were reported in each of the BEs.
Salmonid aquatic and terrestrial prey are highly sensitive to the three carbamate insecticides.
Death of individuals and reductions in individual taxa and prey communities have been
documented and are expected following applications of TV-methyl carbamates that achieve effect
concentrations-several of which are at the low ng/L levels (Schulz 2004). Complete or partial
elimination of aquatic invertebrates from streams contaminated by insecticides has been
documented for carbaryl (Muirhead-Thomson 1987) as well as many other insecticides. A
review of more than 60 field studies (published from 1982-2003) on insecticide contamination
concluded that "about 15 of the 42 studies revealed a clear relationship between quantified, non-
experimental exposure and observed effects in situ, on abundance [aquatic invertebrate], drift,
community structure, or dynamics" (Schulz 2004). Although the top three insecticides most
frequently detected at levels expected to result in toxicity were chlorpyrifos (OP), azinphos-
methyl (OP), and endosulfan, the TV-methyl carbamates carbaryl, carbofuran, oxymyl, and
fenobucarb all showed clear or assumed relationships between exposure and effect (Schulz
2004).
Drift, feeding behavior, swimming activity, and growth are sublethal endpoints of aquatic prey
negatively affected by exposure to AChE inhibitors (Courtemanch and Gibbs 1980; Haines 1981;
Hatakeyama, Shiraishi et al. 1990; Davies and Cook 1993; Beyers, Farmer et al. 1995; Schulz
2004). Drift of aquatic invertebrates is an evolutionary response to aquatic stressors. However,
DRAFT 353
-------
DRAFT
insecticides, particularly carbamates and OPs, can trigger catastrophic drift of salmonid prey
items (Courtemanch and Gibbs 1980; Haines 1981; Hatakeyama, Shiraishi et al. 1990; Davies
and Cook 1993; Schulz 2004). Some invertebrates may drift actively to avoid pesticides and
settle further downstream, which can provide temporary spikes in available food items for
feeding salmonids. Catastrophic drift can also deplete benthic populations resulting in long-term
prey reduction that may affect salmonid growth at critical time periods. We located no studies
that address this line of reasoning directly with Pacific salmonids. Davies and Cook (1993) did
show aquatic invertebrate community changes, mortality of invertebrates, drift of dying and dead
invertebrates, and affected trout following spraying of a pyrethroid pesticide, cypermethrin, an
invertebrate and fish neurotoxicant (Davies and Cook 1993). Effect concentrations were
estimated at 0.1-0.5 (ig/L cypermethrin. It is difficult to compare these effect concentrations to
carbamate insecticides. However, it is illustrative of how insecticides can damage multiple
endpoints of an aquatic community including reducing abundance of prey (Davies and Cook
1993).
Several scientific peer-reviewed publications and EPA documents have reviewed aspects of the
available information on multi-organism microcosm, mesocosm, and field test results for the
AChE insecticides (Leeuwangh 1994; Barren and Woodburn 1995; Schulz 2004; Van
Wijngaarden, Brock et al. 2005). Van Wijngaarden et al. (2005) conducted a literature review
that listed ecological threshold values (e.g., NOECeco and LOECeco) for carbaryl, carbofuran, and
bendiocarb (another TV-methyl carbamate) from model ecosystems or "adequate" field studies. A
NOECeco represented "the highest tested concentration at which no, or hardly any, effects on the
structure and functioning of the studied model ecosystem were observed. The LOECeco is the
lowest tested concentration at which significant treatment-related effects occurred" (Van
Wijngaarden, Brock et al. 2005). Below we discuss some of this information in relation to
effects on salmonid prey. The majority of studies were conducted in littoral systems, i.e., ponds,
and other static systems, but one study with carbaryl was conducted in a running water (lotic)
system (Courtemanch and Gibbs 1980). Population densities of salmonid prey items (i.e.,
Ephemeroptera, Diptera, Amphipoda, Cladocera, Copepoda, Isopoda, Ostracada, Trichoptera)
decline following exposures to AChE-inhibiting insecticide concentrations, including carbaryl
and carbofuran (Van Wijngaarden, Brock et al. 2005). Adverse effects to these groups occurred
DRAFT 354
-------
DRAFT
at or below "1 toxic unif'-where a toxic unit equals field concentrations normalized by dividing
them by the 48 h EC50 ofDaphnia magna for a given AChE inhibitor.
We did not locate any microcosm, mesocosm, or field experiments that measured responses of
aquatic communities that contained salmonids and salmonid prey items simultaneously; a
recognized data gap. Several studies evaluated aquatic invertebrate responses to TV-methyl
carbamates (carbaryl, carbofuran, and bendiocarb) in static and running water systems. We
found no field studies with methomyl. We summarize open literature studies with aquatic
invertebrates organized by insecticide in Table 22. We found no studies that addressed effects of
methomyl on aquatic invertebrates in the field or from multispecies microcosms or mesocosms.
The available literature from field experiments indicates that populations of aquatic insects and
crustaceans are likely the first aquatic organisms damaged by exposures to carbaryl, carbofuran,
and methomyl contamination. Benthic community shifts from sensitive mayfly, stonefly and
caddisfly taxa, the preferred prey of salmonids, to worms and midges occur in areas with
degraded water quality including from contaminants such as pesticides (Cuffney, Meador et al.
1997; Hall, Killen et al. 2006). Reduced salmonid prey availability correlated to OP use in
salmonid bearing watersheds (Hall, Killen et al. 2006). We found no studies that evaluated the
effects on carbamates and that made correlations to salmonid bearing watersheds. Subsequent
effects to salmonid's growth from reduced prey availability and quality remain untested and are a
current data gap.
We located one highly relevant study that focused on fish growth following a single exposure to
chlorpyrifos. The study indicated that native fathead minnows exposed to chlorpyrifos had
reduced growth due to reductions in prey item abundance in littoral enclosures (pond
compartments) (Brazner and Kline 1990). The experiment tested the hypothesis that, "addition
of chlorpyrifos would reduce the abundance of invertebrates and cause diet changes that would
result in reduced growth rates." Nominal, chlorpyrifos treatment concentrations of 0.5, 5.0, and
20 |J,g/L (chemical analysis of water concentrations provided at 0, 12, 24, 96, 384, 768 h) all
resulted in statistically significant reductions in growth at 31 days. A single pulse of chlorpyrifos
was introduced into each enclosure at day 0. Invertebrate abundance was determined in each
DRAFT 355
-------
DRAFT
replicate on days -3, 4, 16, and 32. Fathead minnows were sampled from enclosures on days -2,
7, 15, and 31 where fish were weighed, measured, and dissected to determine gut content
(dietary items identified). By day 7, significant differences in mean numbers of rotifers,
cladocerans, protozoans, chironomids, mean total number of prey being eaten per fish, and mean
species richness were greater in fish from the control enclosures than in some of the treatments.
By day 15, control minnows were significantly larger than fish from treated levels. These
experimental results support the conclusion that reductions in abundance of prey to juvenile fish
can result in significant growth effects. It is reasonable to assume that reductions in prey from
carbamate insecticides can also result in reduced juvenile salmonid growth and ultimately
reduced survival and productivity. The precise levels of prey reduction necessary to cause
subsequent reductions in salmonid growth remain a recognized data gap.
Although the cause is unknown, recent declines in aquatic species in the Sacramento-San
Joaquin River Delta in California have been attributed to toxic pollutants, including pesticides
(Werner, Deanovic et al. 2000). Significant mortality or reproductive toxicity in C. dubia was
detected in water samples collected at 24 sites in the Sacramento-San Joaquin River Delta in
California. Ecologically important back sloughs had the largest percentage of toxic samples (14
- 19%). The TIEs identified carbofuran and carbaryl as well as the OPs chlorpyrifos, diazinon,
and malathion as the primary toxicants in these samples responsible for the adverse effects
(Werner, Deanovic et al. 2000).
Recovery of salmonid prey communities following acute and chronic exposures from carbaryl,
carbofuran, and methomyl depends on the organism's sensitivity, life stage, length of life cycle,
among other characteristics. Univoltine species will take longer than multivoltine species to
recover (Liess and Schulz 1999). Recovery of salmonid prey items such as caddisflies,
stoneflies, and mayflies will be slow, considering their long life cycles and infrequent
reproduction. Additionally, these species also require clean, cool waters to both recover and
maintain self-sustaining populations. In several salmonid-supporting systems these habitats are
continually exposed to anthropogenic disturbances, including pesticide contamination, which
limits their recovery and can also limit recovery of multivoltine species. For example, urban
environments are seasonally affected by stormwater runoff that introduces toxic levels of
DRAFT 356
-------
DRAFT
contaminants and scours stream bottoms with high flows. Consequently, urban environments do
not typically support diverse communities of aquatic invertebrates (Paul and Meyer 2001;
Morley and Karr 2002).
Similarly, yet due to a different set of circumstances, watersheds with intensive agriculture land
uses show compromised invertebrate communities (Cuffney, Meador et al. 1997). Indices of
biological integrity (IBI) and other invertebrate community metrics are useful measures of the
health of an aquatic community because cumulative impacts of aquatic stressors are integrated
over time. The IBI is also valuable because it converts relative abundance data of a species
assemblage into a single index of biological integrity (Allan 1995). Salmonid-inhabited
watersheds have been assessed using IBIs and other metrics of aquatic community health.
A study on the condition of Yakima River Basin's aquatic benthic community found that
invertebrate taxa richness was directly related to the intensity of agriculture i.e., at higher
agriculture intensities taxa richness declined significantly both for invertebrates as well as for
fish (Cuffney, Meador et al. 1997). Locations with high levels of impairment were associated
with high levels of pesticides and other agricultural activities, which together with habitat
degradation were likely responsible for poor aquatic conditions (Cuffney, Meador et al. 1997).
Salmonid ESUs and DPSs occur in the Yakima River Basin as well as other watersheds where
invertebrate community measurements indicate severely compromised aquatic invertebrate
communities such as the Willamette River Basin, Puget Sound Basin, and the Sacramento-San
Joaquin River Basin.
DRAFT 357
-------
DRAFT
Table 65. Study designs and results with freshwater aquatic invertebrates
Chemical
Carbaryl
Carbaryl
Carbaryl
Carbaryl
Taxa/species
multiple n= 25;
including predators, herbivores,
zooplankton
1 . C. californica (stonefly)
2. C. sp. (mayfly)
3. A. sp (mayfly)
4. B. americanus (caddisfly)
5. P. sp. (caddisfly) early instar
6. P. sp. (caddisfly) late instar
7. L unicolor (caddisfly)
Coleopteran (beetles), dipteral
(flies), trichoptera (caddisflies),
terrestrial, emphemeroptera
(mayflies)
Chironomus riparius (midge)
Assessment
measures
species richness,
biomass,
abundance, survival
Survival; LC1.LC50
salmonid stomach
contents
Survival (EC50);
Uptake rate;
Bioconcentration
factors (BCFs)
Concentrations tested
510 |ig/L; single
application; simulating
direct overspray
1-30 |ig/L
4-100,000 |ig/L
10-28 |ig/L
5-55 |ig/L
5-80 |ig/L
5-80 |ig/L
5-80 |ig/L
0.5 Ibs/acre, 2
applications, 7 days
interval
Multiple concentration
ranges at pH 4,6,8;
temps 12, 20, 30 °C
Exposure
duration
13d
96 h
na
24 h
Effects
Reduced richness of
community (by 15%),
predators, zoo plankton;
reduced biomass of predators;
increased abundance of large
herbivores;
No effect on zooplankton
abundance;
elimination of diving beetle
larvae, Daphnia pulex, D.
ambigua
9.0 (LC1); 17.3(LC50)ug/L
3.0; 11.1
7.5; 20.4
28.8; 41.2
14.8; 30.3
33.8; 61.0
9.5; 29.0
Increased numbers of diptera
in stomach after first spray;
increased numbers of
emphemeroptera, trichoptera,
diptera, and terrestrial insects
indicating drift; condition factor
offish increased following
applications due to increased
feeding; fish AChE inhibited
15-34%
EC50s(61-133|ig/L);
Lowest EC50 at 30 °C, pH 4
and 6; Highest ECSOs at 10°C,
pH 4 and 6.
Uptake rates increased with
temp.
BCFs = 5.36-10.12
Data source
(Relyea
2005)
(Peterson,
Jepson et al.
2001)
(Haines
1981)
(Lohnerand
Fisher 1990)
DRAFT
358
-------
DRAFT
Chemical
Carbaryl
carbaryl
carbaryl
carbarvl
Taxa/species
C. californica (stonefly)
Cinygma sp. (mayfly)
Stream invertebrate aquatic
community
Xathocnemis zealandica
(damselfly)
Daphnia magna (daphnids)
Assessment
measures
Survival (EC50)
Aquatic invertebrate
drift, survival;
presence and
abundance of riffle
invertebrates;
recovery of benthic
invertebrates
Emergence of
damselflies, %
emerged,
Population
abundance at
different phases;
Life stage sensitivity
(survival, 48 h LC50)
Concentrations tested
17.3, 173, 1730 ng/L
10.2, 102,204,408,
1020 ng/L
0.75 Ibs/ acre;
1 Ib/acre
1, 10, 100 |ig/L
(concentrations
replenished every 12
days at 1/4the nominal
treatment level)
15 |ig/L applied as a
single pulse at different
population phases
(treatments): growing,
density peak, stable
Exposure
duration
15 min
30
60
15 min
30
60
96 h recovery
period
2, 30, 60, 365
day drift
sampling post
spray; 1,2, 3,
30, 60 day
benthic
samples
67 days
1 pulse at day
8 to growing
treatment (3
replicates),
day 1 1 to
density peak
treatment ,
day 25 stable
treatment;
48 h
Effects
EC50= na
EC50= na
EC50 = 1139 95%CI(370-
15400) |ig/L
EC50 = 848 no Cl
EC50 = 220 no Cl
EC50 = 165 95% Cl (124-232)
170fold increase in drift at 2 d.
All invertebrates in drift
samples dead at 2 d
(plecoptera, emphemeroptera
diptera were common), at
SOand 60 d drift below prey-
spray levels.
Within hrs larger stream
inverts found dead (plecoptera,
trichoptera, emphemerptera),
tricoptera abandoned their
cases; at 30 and 60 d
significant population declines
of salmonid-prey invertebrates
compared to pre-spray levels.
Total number of organisms not
significantly reduced.
90% reduction in emergence
at 100 |ig/L no significant
differences at 1 and 10 ug/L.
70% decline in abundance
during growth phase, 92.6%
decline in peak phase, 34.8 %
decline in stable phase ,
compare to a 61% decline in
control population;
Survival (48h LCSOs)
Data source
(Peterson,
Jepson et al.
2001)
(Courtemanc
h and Gibbs
1980)
(Hardersen
and Wratten
1998)
(Takahashi
and
Hanazato
2007)
DRAFT
359
-------
DRAFT
Chemical
carbaryl
carbaryl
formulation
(43.3% a.i.)
carbaryl
formulation
(Sevin-4-Oil)
Taxa/species
Zooplankton pond community
Bluegill sunfish, multiple aquatic
species including
phytoplankton, macrophytes,
zooplankton, and
macroinvertebrates
Aquatic macroinvertebrate
community;
Flathead chub (Platygobio
gracilis)
Assessment
measures
Abundance of
zooplankton species
Abundance and
richness of species
groups;
Fish survival and
growth
Invertebrate drift;
AChE inhibition in
resident fish
Concentrations tested
500|ig/L; 1-3
applications to pond
mesocosms;
0.1 mg/L
2, 6, 20, 60, 200 |ig/L;
applications at weekly
intervals for 6 weeks
Aerial application in
1991 and 1993 applied
atO.SIbs
carabaryl/acre and 0.4
Ibs/acre, respectively
Exposure
duration
0-3 months
0- 12 weeks,
with 6 weekly
pulses
beginning at
week 10
River water
sampled at 1,
2,4, 8, 12,
24, 48, 96 h;
AChE
inhibition
measured at
24 h;
Drift sampled
4 times daily
for 4 days
Effects
At 500 ug/L sustained
reductions in daphnid and
copepod populations. No
observable reductions in rotifer
populations. At 100 |ig/L,
reductions in daphnid
population observed, but not
with copepods or rotifers
no long term reductions in
major invertebrate groups
were observed; no detectable
differences in bluegill survival
or growth between control and
pesticide treatments,
carbaryl half life was 30 min,
mean pH of treatment groups
in application events ranged
from 9.2-10.3; mean
temperatures ranged from 17 -
27 °C, with the majority
between 20-24 °C.
1 h average 85 |ig/L and
declined to 0.100 |ig/L by 96 h
in 1991;
2 h average 12 |ig/L L and
declined to 5.14 |ig/L by 96 h
in 1993;
No detectable differences in
AChE inhibition at 24 h, low
statistical power (0.82 and 1);
Note: AChE is expected to
recover by 24 h;
Increase in coefficient of
variation of drift measure from
day land 2 post application
compared to reference site
drift in 1991; Similar increase
on day 2 in 1993; drift
composed primarily of live
mayflies (emphemeroptera)
Data source
(Hanazato
and Yasuno
1990)
Registrant-
submitted
study (1993)
(Beyers,
Farmer et al.
1995)
DRAFT
360
-------
DRAFT
Chemical
carbofuran
carbofuran
Taxa/species
Gammarus pulex (fw amphipod)
8 aquatic invertebrate taxa:
Hyalella azteca (amphipod)
Chirinominae larvae
Tanypodinae larvae
2 snails
1 Leech
Damsel fly nymphs
Mayfly nymphs
Assessment
measures
in situ survival,
feeding rates;
laboratory bioassays
measured 24 h
survival (LC50) and
feeding rates
Taxa abundance
and biomass
Concentrations tested
Field study: measured
stream concentrations
in stormwater runoff
following an
application of 3 kg a.i./
ha; 27 |ig/L stream,
256 |ig/L in agricultural
field ditch
Lab study:
Feeding rate-
0.75,1.25,2.25,4.0,7.0,
12.0 ng/L
5 ng/L
25 ng/L
Note: pH of 9 in
experimental pond
reduced persistence of
carbofuran by
increasing the rate of
hydrolysis (Chapman
and Cole 1982)
Exposure
duration
Field study:
5 day; 2 rain
events
captured
Lab study:
Survival-
24-, 48- 96-hr
Feeding rate-
7 day
exposures
Taxa
assessed at
5, 11, 32, 55
days post
application
Effects
Field study:
All amphipod chambers
showed reduced feeding rates
at time during rain events and
all amphipods died in 10
experimental chambers
Lab study:
Survival- |ig/L (95% Cl)
24hLC50 = 21 (14.7-30)
48 hLC50 = 12.5 (5.7-27.5)
96hLC50 = 9(5.8-13.9)
Feeding rate-
>3 |ig/L feeding rates reduced
toO
No detectable reductions to
abundance or biomass of taxa
from 5 |ig/L;
H azteca: 10% reduction in
abundance; 6% reduction in
biomass
Chironominae: 77% dead in
48-72 h dip net samples
compared to 0% dead in
control and 5 |ig/L treatment;
biomass reduced by 17% at 55
d; evidence that younger
larvae were less sensitive than
older larvae;
No detectable reductions in
remaining taxa
Data source
(Matthiessen
, Shearan et
al. 1995)
(Wayland
1991)
DRAFT
361
-------
DRAFT
Chemical
carbofuran
carbofuran
Taxa/species
Aquatic invertebrate community
in rice fields:
ostracada, copepod, cladocera,
mosquito larvae, chironomid
larvae
Cerodaphnia dubia (daphnids)
Assessment
measures
Abundance of
ostracada, copepod,
cladocera, mosquito
larvae, chironomid
larvae
Survival (24 and 48
h LCSOs); survival
from mixture of
carbofuran and
methylpararthion (48
h LC50);
Mean young per
female and percent
survival from 7 day
chronic assays
Concentrations tested
0.1 kg carbofuran/ha-
applied once
0.3 kg carbofuran/ha
applied in two regimes:
3, 52, 85, 97 days; 16,
57, 69 days
Irrigation filed-collected
water samples (toxicity
identification
evaluations employed)
Mixture toxicity (1:1,
3:1, 1:3 ratios tested)
0.16, 0.33, 0.65, 1.3,
2.6 ug/L
Exposure
duration
variable
application
intervals and
frequencies
24, 48 hours
7 days
Effects
Ostracada: timing and
magnitude of peak
abundances not affected by
carbofuran, multiple
application regimes of 0.3 k/ha
significantly reduced
abundance compared to one
application of 0.1 kg/ha
Copepods: significant
reductions at 45 and 57 d at
0.3 kg/ha.
Cladoceran: rate of population
growth significantly affected
from 52 and 85 days at 0.3
kg/ha. Abundance lower at 0.1
kg/ha compared to 0.3 kg/ha.
Chironomid larvae: isolated
significant differences in
abundances were found
(P<0.05), but showed no dose-
response relationship
Mosquito larvae:
No change in abundances
Carbofuran and methyl-
parathion putative agents of
toxicity in field-collected
samples from CA rice drain.
Carbofuran acute toxicity :24 h
LC50 3.4 |ig/L (95% Cl 2.8-
4.3; 48 h LC50 2.6 |ig/L (95%
Cl NA); chronic study- NOEC
1.3|ig/L; LOECat2.6 |ig/L
Strict additive toxicity of
combinations tested of
carbofuran and methyl-
parathion;
Data source
(Simpson,
Roger et al.
1994)
(Norberg-
King,
Durhan et al.
1991)
DRAFT
362
-------
DRAFT
Chemical
carbofuran
carbofuran
carbofuran
carbofuran
Taxa/species
Chironomus tentans
(chironomid)
Chorophium vo/ufafor(estuarine
amphipod)
Daphnia magna (daphnid)
Hydra attenuate (fw jellyfish)
Macrobrachium rosenbergii (fw
shrimp)
Aquatic macroinvertebrate
community
Ceriodaphnia dubia (daphnid)
Neomysis mercedi)( mysid)
Pimephales promelas (fathead
minnow)
Assessment
measures
Survival in spiked-
sediments, mixed
with atrazine, and
field-collected runoff
Survival and
avoidance behavior
in contaminated
spiked sediments,
Survival (acute
LC50);
macroinvertebrate
community structure
Survival
(% mortality)
Concentrations tested
Spiked sediment
concentrations:
25,45, 83, 150|ig/kg;
Mixture
concentrations:
25, 45, 80 ng/kg
carbofuran + 5, 10, 20
mg/kg atrazine;
Field collected
sediments in runoff:
collected sediments in
runoff:
186 |ig/kg sediment-
associated
73 |ig/L pore water
2.5, 25, 250 ng/g (dry
weight sediment)
Average
concentrations
detected 5 d post
application 0.7 and
0.16 |ig/L maximum
carbofuran detection
2.1 ng/L
Not reported;
conducted toxicity
identification
evaluations (TIE)
Exposure
duration
10 days
48 and 96 h
exposures
24 , 48, 72,
96 h
exposures to
field collected
water
samples
96 h
exposures to
field collected
water
samples
Effects
Spiked sediment results:
Nominal [carbofuran] LC50
47.9 ng/kg (95% Cl 43.9-52.1);
Sediment bound LC50 20.9
|ig/kg(95%CI 18.4-33.2);
Interstitial water LC50 11.8
|ig/L(95%CI 10.9-17.1)
Mixture results:
Additive toxicity present with
atrazine + carbofuran, no
synergistic or antagonistic
toxicity responses. Atrazine
affected surnvial in the low
mg/kg range.
Field collected sediments in
runoff:
No chironomids survived
exposure, note control survival
was low (57%)
48 h LC 50 = 73 ng/g (95% Cl
0-128);48hLC20 = 41 (95%
Cl 2-82) ng/g; no effect on
avoidace behavior of sediment
No acute toxicity reported to
tested species.
Significant reduction of % EPT
taxa and increase in
community loss indices at
treated sites. Mayflies were
primary affected species from
carbofuran applications
3/47 irrigation samples had
100% toxicity to C. dubia
attributed to carbofuran
Data source
(Douglas,
Mclntosh et
al. 1993)
(Hellou,
Leonard et
al. 2009)
(Castillo,
Martinez et
al. 2006)
(de Vlaming,
DiGiorgio et
al. 2004)
DRAFT
363
-------
DRAFT
Chemical
carbofuran
Taxa/species
Wetland invertebrate aquatic
community:
Daphnia magna,
Chironomus riparius
Assessment
measures
Survival (EC50)
Primary productivity
Concentrations tested
10, 100, 1,000|ig/L
water only 48 h test;
10, 100, 1,000|ig/L
mixed with soil and
added to microcosms
Exposure
duration
48- hr;
30 day
microcosm
test
Effects
D. magna EC50 = 48 |ig/L
C. riparius EC50 = 56 |ig/L
Populations of D. magna were
viable at day 1 in 10 and
100ug/L, it took 4 d for a viable
population (i.e., 5 or more live
diohnids) to establish at 1000
|ig/L . No dead daphnids or
chironomids observed at 14 d
and 30 d. No effects to
microbial community enzyme
activity were found.
Data source
(Johnson
1986)
DRAFT
364
-------
DRAFT
Studies with other AChE inhibitors on salmonid prey items:
Robust evidence shows that salmonid prey taxa and communities can be substantially reduced
following exposures to the OP insecticides chlorpyrifos, diazinon, and malathion. NMFS
reviewed these data and presented its findings in a biological opinion (NMFS 2007d). We use
these findings to show the types of aquatic community responses following exposures to AChE
inhibiting insecticides. The toxic potency of a pesticide is a function of concentration and
duration of exposure, which in turn is a function of a pesticide's physical properties and
interactions of the pesticide with environmental variables such as temperature, pH, sunlight, soil
micro-organisms, etc. With this in mind, if carbaryl, carbofuran, and methomyl are at
concentrations individually (or together in mixtures) expected to reduce salmonid prey
communities, we infer a similar magnitude of response (reductions in abundance from death and
catastrophic drift) and similar recovery period for the as those observed in affected aquatic
communities treated with OP insecticides.
We note that individual aquatic invertebrates exhibiting adverse sublethal responses from
carbamates will likely recover much more quickly compared to than those exposed to OPs
(Kallander, Fisher et al. 1997). The midge, Chironomus tentans, showed complete recovery of
AChE activity in 24 h following two 1 h pulses of carbaryl separated by 24 h. Additionally, two-
1 h pulses of carbamates (carbaryl, carbofuran, aldicarb, propoxur) caused significantly fewer
symptoms of intoxication than 2 h of continuous exposure when chironomids were given 2-6 h
of recovery in clean water between doses (Kallander, Fisher et al. 1997). In contrast, sublethal
exposures to OPs were equally toxic when exposures are either pulsed or continuous.
Reviews of field, mesocosm, and microcosm studies with the three OPs document reductions in
aquatic invertebrate populations and lengthy recovery times for populations of some taxa (Giesy,
Solomon et al. 1998; Van Wijngaarden, Brock et al. 2005). A recent study found significant
changes to macroinvertebrate assemblages of artificial stream systems following a 6 h exposure
to chlorpyrifos at 1.2 |J,g/L; the lowest concentration tested (Colville, Jones et al. 2008). The
addition of chlorpyrifos to the artificial streams resulted in a rapid (6 h) change in the
macroinvertebrate assemblages of the streams, which persisted for at least 124 d after dosing
(Colville, Jones et al. 2008). The chlorpyrifos dissipated from the system within 48 h (Pablo,
DRAFT 365
-------
DRAFT
Krassoi et al. 2008), however the macroinvertebrate community did not recover rapidly. Several
species similar to salmonid prey items were significantly affected.
Zooplankton and insect taxa appeared the most sensitive in studies with diazinon and in
particular, the salmonid prey taxa trichoptera, diptera, and cladocera were highly sensitive
(Giddings, Hall et al. 2000). Field studies in salmonid habitat also show reductions in salmonid
prey abundances. For example, in listed steelhead habitat in the Salinas River, California,
abundances of the salmonid prey items including mayfly taxa, daphnids, and an amphipod
(Hyalella aztecd) were significantly reduced downstream of an irrigation return drain compared
to upstream (Anderson, Hunt et al. 2003; Anderson, Hunt et al. 2003; Anderson, Phillips et al.
2006). In this study, diazinon and chlorpyrifos were detected above acute toxicity thresholds in
surface waters and sediments. Combined toxicity of the two OPs using a toxic unit approach
correlated strongly with mortality of daphnids. For H. azteca, acute toxicity was attributed to
sediment pore-water concentrations of chlorpyrifos (Anderson, Hunt et al. 2003). Other
pesticides were likely present and responsible for some of the toxicity in the Salinas River,
including carbaryl. In a subsequent study on the Salinas River, TIE demonstrated that
chlorpyrifos and diazinon were responsible for the observed death of the daphnid C. dubia (Hunt,
Anderson et al. 2003). These data support the line of evidence that field concentrations of OPs
can and do adversely affect aquatic invertebrates in salmonid habitats. We believe the same
situation occurs with carbamates, given the toxicity of the compounds and the similar mode of
action.
Adjuvant toxicity
Assessment endpoints: Survival offish and aquatic prey items, endocrine disruption in fish
Assessment measures: 24, 48, 96 h LCSOs, and vitellogenin levels in fish plasma
Although no data were provided in the BEs related to adjuvant toxicity, an abundance of toxicity
information is available on the effects of the alkylphenol polyethoxylates, a family of non-ionic
surfactants used extensively in combination with pesticides as dispersing agents, detergents,
emulsifiers, adjuvants, and solubilizers (Xie, Thrippleton et al. 2005). Two types of alkylphenol
polyethoxylates, NP ethoxylates and octylphenol ethoxylates, degrade in aquatic environments to
the more persistent, toxic, and bioaccumulative degradates, octylphenol and NP, respectively.
We note that the technical registrant of methomyl stated that no nonylphenol ethoxylates are
DRAFT 366
-------
DRAFT
used within methomyl formulations. We did not receive information on the presence or absence
of alkylphenolpolyethoxylates in carbaryl- or carbofuran- containing formulations. Adjuvants
are frequently mixed with formulations prior to applications, so although they may not be present
in the formulations they could still be co-applied. Below we discuss NP's toxicity as an example
of potential adjuvant toxicity, as we received no information on adjuvant use or toxicity within
the BEs.
We queried EPA's ECOTOX online database and retrieved 707 records of NP's acute toxicity to
freshwater and saltwater species. The lowest reported LC50 for a salmonid was 130 (ig/L for
Atlantic salmon. Aquatic invertebrates, particularly crustaceans, were killed at low
concentrations of NP, lowest reported LC50 = 1 |j,g/L for H. azteca. These data indicate that a
wide array of aquatic species are killed by NP at |j,g/L concentrations. We also queried EPA's
ECOTOX database for sublethal toxicity and retrieved 689 records of freshwater and saltwater
species tested in chronic experiments. The lowest fish LOEC reported was 0.15 (ig/L for fathead
minnow reproduction. Numerous fish studies reported LOECs at or below 10 ng/L.
Additionally, salmonid prey species are sensitive to sublethal effects of NP. The amphipod,
Corophium volutator, grew less and had disrupted sexual differentiation (Brown, Conradi et al.
1999). Multiple studies with fish indicated that NP disrupts fish endocrine systems by
mimicking the female hormone 17|3-estradiol (Brown and Fairchild 2003; Arsenault, Fairchild et
al. 2004; Madsen, Skovbolling et al. 2004; Jardine, MacLatchy et al. 2005; Luo, Ban et al. 2005;
McCormick, O'Dea et al. 2005; Segner 2005; Hutchinson, Ankley et al. 2006; Lerner, Bjornsson
et al. 2007; Lerner, Bjornsson et al. 2007). NP induced the production of vitellogenin in fish at
concentrations ranging from 5-100 |J,g/L (Hemmer, Bowman et al. 2002; Ishibashi, Hirano et al.
2006; Arukwe and Roe 2008; Schoenfuss, Bartell et al. 2008). Vitellogenin is an egg yolk
protein produced by mature females in response to 17-P estradiol, however immature male fish
have the capacity to produce vitellogenin if exposed to estrogenic compounds. As such,
vitellogenin is a robust biomarker of exposure. A retrospective analysis of an Atlantic salmon
population crash suggested the crash was due to NP applied as an adjuvant in a series of
pesticide applications in Canada (Fairchild, Swansburg et al. 1999; Brown and Fairchild 2003).
Additionally, processes involved in sea water adaptation of salmonid smolts are impaired by NP
DRAFT 367
-------
DRAFT
(Madsen, Skovbolling et al. 2004; Jardine, MacLatchy et al. 2005; Luo, Ban et al. 2005;
McCormick, O'Dea et al. 2005; Lerner, Bjornsson et al. 2007; Lerner, Bjornsson et al. 2007).
These results demonstrate NP is of concern to aquatic life, particularly salmonid endocrine
systems involved in reproduction and smoltification. This summary is for one of the more than
4,000 inerts/other ingredients and adjuvants currently registered for use in pesticide formulations
and there are likely others with equally deleterious effects. Unfortunately we received minimal
information on the constituents found in carbaryl-, carbofuran-, and methomyl-containing
formulations.
Consequently, the effects that these other ingredients may have on listed salmonids and
designated critical habitat remain an uncertainty and are a recognized data gap in EPA's action
under this consultation.
Summary of Response Analysis:
We summarize the available toxicity information by assessment endpoint in Table 23. Data and
information reviewed for each assessment endpoint was assigned a general qualitative ranking of
either "low", "medium", or "high." To achieve a high confidence ranking, the information
stemmed from direct measurements of an assessment endpoint, conducted with a listed species or
appropriate surrogate, and was from a well-conducted experiment. A medium ranking was
assigned if one of these three general criteria was absent and low ranking was assigned if two
criteria were absent. Evidence of adverse effects to assessment endpoints for salmonids and their
habitat from the three a.i.s was prevalent. However, much less information was available for
other ingredients, in part, due to the lack of formulation information provided in the BEs as well
as the statutory mandate under FIFRA for toxicity data on the a.i.s to support registration. We
did locate a substantial amount of data on one group of adjuvants/surfactants, the NP ethoxylates.
However, we received and located minimal information for the majority of tank mixes and other
ingredients within formulations.
DRAFT 368
-------
DRAFT
Table 66. Summary of assessment endpoints and effect concentrations
Assessment Endpoint
Carbaryl
Fish:
-survival (LC50)
-growth
-reproduction
-swimming
-olfactory-mediated behaviors
Habitat: -prey survival (LC50)
Carbofuran
Fish:
-survival (LC50)
-growth
-reproduction
-swimming
-olfactory-mediated behaviors
Habitat: -prey survival (LC50)
Methomyl
Fish:
-survival (LC50)
-growth
-reproduction
-swimming
-olfactory-mediated behaviors
Habitat: -prey survival (LC50)
Other ingredient: Nonylphenol
Fish:
-survival (LC50)
-reproduction
-smoltification
-endocrine disruption
Habitat: -prey survival (LC50)
Additive toxicity of
/V-methyl carbamates
Synergistic toxicity of
/V-methyl carbamates
Evidence of
arivprQp
CIU Vd 9C
responses
(yes/no)
yes
no
yes
yes
no
yes
yes
yes
yes
yes
yes
yes
yes
yes
no
-
-
yes
yes
yes
yes
yes
yes
yes
yes
Concentration
range of observed
effect or
concentrations
tested showing
absence of effect
(ng/L)
250-4,500
8,17,62,210,680
680
10-12,500
5, 50, 500
1.5-351
88-3,100
56.7
6
5, 100
1-200
1.62-2,700
300-7,700
142,243
31-500
-
-
8.8-920
130->1,000
0.15-10
5-100
5.0-100
1->1,000
multiple
multiple
Degree of confidence
in effects
(low, medium, high)
high
medium
medium
high
high
high
high
medium
low
medium
high
high
high
medium
low
-
-
high
high
high
medium
high
high
high
high
reported 48 h EC50s of D. magna are low and high
' 48 h survival EC50 of Chironomus tentans (Karnak
values of the range (Takahashi and Hanazato 2007)
and Collins 1974)
DRAFT
369
-------
DRAFT
Risk Characterization
In this section we integrate our exposure and response analyses to evaluate the likelihood of
adverse effects to individuals, populations, species, and designated critical habitat. We
combined the exposure analysis with the response analysis to: 1) determine the likelihood of
salmonid and habitat effects occurring from the stressors of the action; 2) evaluate the evidence
presented in the exposure and response analyses to support or refute risk hypotheses; 3) translate
fitness level consequences of individual salmonids to population-level effects; and 4) translate
habitat-associated effects to potential impacts on PCEs of critical habitat. The risk
characterization section concludes with a general summary of species responses from population
level effects. We then evaluate the effects to specific ESUs and designated critical habitat in the
Integration and Synthesis section.
Exposure Profile
Effect on individuals
Effects on populations
Effects on species
(ESU or DPS)
I
Can EPA ensure that its
action is not likely to
jeopardize the continued
existence of the species?
Risk
Characterization
Analyzed within the context
of the Environmental
Baseline (including
multiple stressors such as
temperature and
environmental mixtures of
pesticides); the Status of
Listed Resources and
Cumulative Effects
Addressed in the
Integration and
Synthesis Section
Response Profile
Effects on habitat
Effects on primary
constituent elements
Effects on conservation
value of designated
habitat
I
Can EPA ensure that its
action is not likely to
adversely modify or destroy
designated critical habitat?
Figure 38. Schematic of the Risk Characterization Phase
DRAFT
370
-------
DRAFT
Exposure and Response Integration
In Figures 34-36, we show the overlap between exposure estimates for the three carbamates and
concentrations that affect assessment endpoints. The figures show the exposure concentration
ranges (minimum - maximum values) gleaned from the three predominant sources of exposure
data we analyzed: monitoring data; EPA's estimates presented in the BEs that represent crop
uses; and NMFS' modeling estimates for off-channel habitats. None of the modeled exposure
estimates were derived for non-crop use. This is a major data gap as carbaryl is used extensively
in urban and residential areas. The effect concentrations are values taken from the toxicity data
reviewed in the Response Analysis Section. With respect to the assessment endpoint survival,
recall that the effect concentrations are LCSOs, thus death of sensitive individuals is not
represented by this metric and can occur at concentrations well below LCSOs. Additionally, we
cannot accurately predict at what concentrations death first occurs because dose-response slope
information was generally not provided for most of the acute lethality studies. Although we are
unable to determine at what concentration an individual organism might die, we do incorporate
survival endpoints from acute 96 h studies using a default slope in a population modeling
exercise discussed below. This slope is recommended by EPA when more relevant information
is unavailable (EPA 2004). Where overlap occurs between exposure concentrations and effect
concentrations NMFS explores the likelihood of adverse effects. If data suggest exposure
exceeds adverse effects thresholds, we discuss the likelihood and expected frequency of effects
based on species information and results of the exposure and response analyses.
This is a coarse analysis because it does not present temporal aspects of exposure nor does it
show the distribution of toxicity values. However, it does allow us to systematically address
which assessment endpoints are affected from carbaryl, carbofuran, and methomyl exposure.
Where significant uncertainty arises, NMFS highlights the information and discusses its
influence on our inferences and conclusions. We discuss the uncertainties related to these
endpoints under associated risk hypotheses later in this section.
Carbaryl
Concentration ranges overlap with most of the assessment endpoint ranges indicating that
adverse effects are expected in salmonids if exposed for sufficient durations (Figure 34). Prey
DRAFT 371
-------
DRAFT
survival appears to be the most sensitive endpoint as all three ranges encompass this endpoint
and the maximum concentration values far exceed the prey survival range. Swimming is likely
impaired at the higher end of the concentration range from monitoring data, at the middle and
higher range from EPA estimates, and throughout the range predicted in the off-channel habitat
estimates. Concentrations occurring in the off-channel habitats that NMFS modeled will kill
juvenile salmonids. Furthermore, given the LC50 values for salmonids following 96 h
exposures, we expect that fewer deaths of juveniles will occur in many of the freshwater habitats
exposed to carbaryl. Fish reproduction (based on a single fathead minnow study) would be
affected by concentrations in the off-channel habitat. We also note minimal data exist on effects
to fish growth: only a single study with fathead minnows. We expect that carbaryl will impair
swimming of salmonids, kill salmonid prey, and in certain circumstances kill salmonids when
exposed for sufficient durations. The effect concentrations shown in the figure do not account
for the potential enhanced toxicity of carbaryl to salmonids or their prey items in aquatic habitats
where other AChE inhibitors are present. We also note that pH is a major factor in carbaryl's
persistence in aquatic habitats. At pHs above 8, carbaryl breaks down fairly rapidly (half-life of
24 h) while at pHs less than 8 carbaryl is much more resistant to hydrolysis (half-life of 1- 30 d
for pH of 7.9 - 5.7). The pH of natural surface waters commonly ranges from 7 to 9, thus pH is
an important consideration when evaluating toxicity of carbaryl.
DRAFT 372
-------
DRAFT
Carbaryl Exposure Concentrations
Monitoring
data
EPA estimates
Off-Chnmiel habitat
estimates
Effect Concentrations for Salmonid Assessment Endpoints
Swumnm? <•—-"- - - - ^ — •+
Salmonid survival
Fish reproduction
0 00001 0 001 0 1 10 1
Carbarvl concentration (112: L)
Figure 39. Carbaryl exposure concentrations and salmonid assessment endpoints' effect
concentrations in ug/L
Carbofuran
Concentration ranges overlap with most of the assessment endpoint ranges, indicating that
adverse effects are expected in salmonids if exposed for a sufficient duration (Figure 40).
Salmonid prey items appear just as sensitive to carbofuran as to carbaryl. Significant variation
among prey species were observed in reported acute survival EC/LC 50s. Fish reproduction,
swimming and salmonid olfactory-mediated behaviors are encompassed or exceeded by all three
exposure ranges. The fish growth value (56 ug/L LOEC) was exceeded by the off-channel
habitat concentrations. Salmonid survival was the least sensitive response reviewed. As with
carbaryl, concentrations of carbofuran occurring in the off-channel habitats that NMFS modeled
will kill juvenile salmonids. Furthermore, given the LC50 values for salmonids following 96 h
exposures, we expect that fewer deaths of juveniles will occur in many of the freshwater habitats
exposed to carbofuran. As with carbaryl, we note that pH is a major factor in the degree of
DRAFT 373
-------
DRAFT
toxicity of carbofuran. At pHs above 9, carbofuran breaks down fairly rapidly (half-life of 0.8-
15 h) while at pH 7 reported half-lives range from 2- 28 d. Therefore, pH is an important
consideration when evaluating toxicity of carbofuran in the aquatic environment. We discuss
these effects and other considerations in more detail under the risk hypotheses.
Carbofuran Exposure Concentrations
Monitoring ^^^_
data
EPA estimates
Off-Channel habitat
estimates
Effect Concentrations for Salmonul Assessment Endpoints
Swuumins
Olfactory-mediated
behaviors
Prey survival
Fish reproduction
Salmonid survival
Fish growth
000001 0001 0 1 10 1000
Carbofuran concentration (ng L)
Figure 40. Carbofuran exposure concentrations and salmonid assessment endpoints' effect
concentrations in ug/L
Methomyl
We do expect a portion offish to be exposed to upper end concentrations which would overlap
with most of the effect thresholds. (Figure 41). As with carbaryl and carbofuran, prey survival
was the most sensitive endpoint assessed. More sensitive prey, with ECSOs at the lower end of
the distribution, are expected to die based on all three exposure estimate ranges, with the highest
DRAFT 374
-------
DRAFT
rates of mortality based on EPA estimates and NMFS' off-channel habitat estimates. The
magnitude of reduction in prey abundance will depend on which taxa are present in the exposed
water body and the actual concentrations and exposure durations. If reductions in prey
abundance, especially of the smaller organisms coincide with fry feeding for the first time
following yolk sac absorption, starvation is likely. We discuss this in greater detail in the risk
hypotheses below. The lower value of the fish reproduction endpoint, 31 |ig/L, overlapped with
both EPA's and NMFS' off-channel habitat estimates, and the highest value in the monitoring
data was greater. A notable data gap is the absence of information on methomyrs toxicity to
olfactory-mediated behaviors. Although we found no studies that evaluated swimming, we
expect swimming impairment at concentrations less than LCSOs because swimming is impaired
with other AChE inhibitors and AChE inhibition correlates to impaired swimming (Little and
Finger 1990; Little, Archeski et al. 1990).
DRAFT 375
-------
DRAFT
Methomyl Exposure Concentrations
Monitoring
data ^
EPA estimates
Off-Channel habitat
estimates
Effect Concentralions for Siilmonid Assessment Endpoints
Fish reproduction '*• •*'
Snlmomcl survival ***"
Fish growth
0.00001 0001 0.1 10 1000
Methomyl concentration (ng/'L)
Figure 41. Methomyl exposure concentrations and salmonid assessment endpoints' effect
concentrations in ug/L
Relationship of pesticide use to effects in the field
Schulz reviewed 45 field and in situ studies published in peer-reviewed journals (from 1982-
2003) that evaluated relationships between insecticide contamination and biological effects in
freshwater aquatic ecosystems and included invertebrates and fishes (Schulz 2004). For each
study, the authors classified the relationship of exposure to effect in one of four categories: no
relation, assumed relation, likely relation, and clear relation based on the cited authors' judgment
of their own results. A relationship was classified as clear only if the exposure was quantified
and the effects were linked to exposure temporally and spatially. The review concluded that
"about 15 of the 42 studies revealed a clear relationship between quantified, non-experimental
exposure and observed effects in situ, on abundance [aquatic invertebrate], drift, community
structure, or dynamics" (Schulz 2004). Although the top three insecticides most frequently
DRAFT 376
-------
DRAFT
detected at levels expected to result in toxicity were chlorpyrifos (OP), azinphos-methyl (OP),
and endosulfan, the TV-methyl carbamates carbaryl, carbofuran, oxymyl, and fenobucarb all
showed clear, likely or assumed relationships between exposure and effect (Schulz 2004). No
studies involving methomyl were evaluated. It should be noted that these studies were not
designed to establish effect thresholds and in our review, are not sufficient to define thresholds.
However, the data do provide information on concentrations of insecticides known to cause
biological and ecological effects under field conditions.
Schulz noted that for all of the studies "that seem to establish a clear link between exposure and
effect, the pesticide concentrations measured in the field were not high enough to support an
explanation of the observed effects simply based on [laboratory bioassays] acute toxicity." Some
authors have suggested differences in field measured exposure and actual organismal exposure
(aquatic, sediment, dietary; environmental variables that affect exposure) as a reason for higher
mortalities in situ than predicted by laboratory toxicity data. Schulz concluded that on the basis
of present knowledge, it cannot be determined whether the measured concentration in the field
regularly underestimates the actual exposure or if a general difference between the field and
laboratory reactions of aquatic invertebrates is responsible; both are reasonable assertions. The
review by Schulz shows a body of evidence that natural aquatic ecosystems can be adversely
affected by AChE inhibitors including carbaryl and carbofuran. We expect a similar relationship
for methomyl because it shares the same mechanism of action as carbaryl and carbofuran (AChE
inhibition) and concentrations that lead to death of aquatic invertbrates are similar to carbaryl
and carbofuran.
We reviewed multiple studies that showed direct, adverse effects to salmonid prey species
following exposures to TV-methyl carbamates (Table 67). These included reduction in individual
aquatic species, changes in community abundance, richness, and diversity with salmonid prey
items.
DRAFT 377
-------
DRAFT
Table 67. Published field studies designed to establish a relationship between W-methyl
carbamate contamination of aquatic habitats due to agricultural practices (adapted from Table 2 in
Schulz 2004).
Source
Concentration
^g/L
Duration
Endpoint
Species
Relationship
of exposure
and effect
Carbaryl
Aerial
application
0.1-85.1
24
aquatic
invertebrate
Drift
Various
invertebrates
Likely
Carbofuran
0.05-26.8
Few
hours
mortality
Amphipod
(Gammarus
pulex)
Clear
Fie Id studies in ESA-listed salmonid habitats with other AChE inhibitors
A group of field studies evaluated macroinvertebrate community responses in the orchard-
dominated Hood River Basin, Oregon and correlated results with chlorpyrifos and azinphos-
methyl use and detections (St. Aubin 2004)Van der Linde 2005; (Grange 2002). Hood River
Basin contains several listed anadromous salmonids, including lower Columbia River steelhead.
The goals of the studies were to determine whether in-stream OPs affected steelhead AChE
activity and changed the aquatic macroinvertebrate community. An additional second objective
addressed how changes in macroinvertebrate community might affect salmonid growth. A suite
of reference and orchard-dominated sampling sites within the Hood River Basin were sampled
pre and post the two primary application seasons, spring (chlorpyrifos) and summer (azinphos-
methyl). Significant differences in macroinvertebrate community assemblages were found
between upstream reference sites and downstream agricultural sites (St. Aubin 2004), similar to
the results described in a California stream (Hall, Killen et al. 2006). However, no significant
differences were found at each individual site, before and after summer spraying (St. Aubin
2004). Therefore, the second Hood River study investigated the spring spray events as well as
the summer spray events to determine seasonal effects (Van der Linde 2005). Sharp declines in
species abundance between reference sites and downstream sites during the spring-spray period
correlated to chlorpyrifos applications and subsequent aquatic detections (one site over an 8 d
period showed chlorpyrifos ranging from 0.032 -0.183 |ig/L). There were more pollutant
tolerant taxa and less intolerant taxa at the agricultural sites (Van der Linde 2005). Collector-
DRAFT
378
-------
DRAFT
gatherer species, many of which are salmonid prey items, declined rapidly at agricultural sites
compared to abundances at the reference sites following applications. Interestingly, reductions
in biodiversity in 2001 agricultural sites compared to reference sites was not seen in 2002 (Van
der Linde 2005). The authors commented that diversity metrics do not always behave
consistently or predictably in response to environmental stress. More than two years of data are
likely needed to more sufficiently address community variability at this site.
Two sets of field experiments directly investigated juvenile steelhead (hatchery- reared) AChE
activity from caged-fish studies in an agricultural basin in Hood River Basin, OR (Grange 2002;
St. Aubin 2004). The studies analyzed water samples for chlorpyrifos, azinphos-methyl, and
malathion before, during, and after orchard spray periods. One of the studies also monitored the
aquatic invertebrate community's response (discussed later under prey effects) in conjunction
with the AChE inhibition (St. Aubin 2004). Steelhead from reference sites had statistically
significantly greater AChE activity than steelhead from orchard-dominated areas. The
reductions in AChE activity corresponded to the application seasons and detections of
chlorpyrifos and azinphos-methyl insecticides.
The data from one study indicated that OP-insecticides inhibited AChE activity in steelhead held
in cages in the Hood River Basin and this inhibition correlated to chlorpyrifos and
azinphosmethyl detections and to a lesser degree with malathion detections (Grange 2002).
None of the pesticides were detected at reference sites and both chlorpyrifos (range in maxima of
(0.077- 0.196 |ig/L) and azinphos-methyl were frequently detected at orchard stream and river
sites. AChE activity was inhibited up to 21% in smolts, and 33% in juveniles relative to
reference locations. Temperature was a confounding factor as lower temperatures showed lower
AChE activity while higher temperatures showed higher AChE activity at reference sites. The
authors normalized data to temperature and found a greater number of statistically significant
reductions in AChE in steelhead. Study results show that, steelhead in these systems exposed to
OP insecticides lose AChE activity (up to 33%) and, depending on the percentage of inhibition,
can manifest into fitness level consequences (Grange 2002; St. Aubin 2004).
DRAFT 379
-------
DRAFT
The field studies conducted in Hood River Basin, Oregon show that salmonids' AChE activity
was reduced in orchard-dominated streams during chlorpyrifos and azinphos-methyl
applications. Additionally, the macroinvertebrate communities in these systems were
compromised to such an extent that there was a reduction in salmonid prey abundance. These
findings also highlight the importance of characterizing the presence of not only TV-methyl
carbamate insecticides, but also concentrations of OP insecticides in evaluating toxicity of
carbaryl, carbofuran, and methomyl to Pacific salmonids.
Field studies in ESA-listed salmonid habitats: Willapa Bay and Gray's Harbor Washington
History of Carbaryl use in Washington estuaries
Willapa Bay and Gray's Harbor support large commercial oyster-producing areas (approximately
600 acres in Willapa and 200 acres in Gray's Harbor) and are located north of the Columbia
River along the southwestern coast of Washington. The Pacific oyster (Crassostrea gigas),
introduced from Japan in the 1920s, is the principal species cultivated in those estuaries
(Feldman, Armstrong et al. 2000). Two species of endemic shrimp, ghost shrimp (Neotrypaea
californiensis) and mud shrimp (Upogebiapugettensis\ are abundant in the oyster growing areas
of these estuaries and create burrows in the sediment that are unfavorable for optimal oyster
cultivation. The Washington State oyster industry has used aerial applications of carbaryl to kill
the burrowing shrimp in oyster beds since the early 1960s. EPA authorizes the use of carbaryl
on oyster beds in Washington through a SLN label, or Section 24(C) (EPA Reg. No. 264-316).
Carbaryl is applied directly to intertidal areas when they are exposed at low tide, primarily by
helicopter (Dumbauld, Brooks et al. 2001). This use has been controversial and there have been
numerous studies dating back several decades that evaluated the ecological impact of carbaryl
use in these two estuaries (Feldman, Armstrong et al. 2000). Although the EPA label does not
limit applications to Willapa Bay and Gray's Harbor, historically these have been the only areas
treated with carbaryl. Additionally, Washington State requires applicators to obtain NPDES
permits for applications to aquatic habitats.
ESA-Listed Pacific salmonids that occur in Willapa Bay and Gray's Harbor
Juvenile and adult Chinook salmon from ESA-listed populations are expected to be present in
Willapa Bay and Gray's Harbor at the time of carbaryl applications. It is expected that juvenile
DRAFT 380
-------
DRAFT
Chinook salmon from the Lower Columbia River use these estuaries during the period when
carbaryl is applied based on their behaviors (Personal communication Eduardo Casillas NMFS
2/18/2009, Anne Shaffer WDFW 3/04/2009, and Thorn Hooper NMFS, 2/12/2009).
Specifically, their use of dendritic channels to access inundated mud flats suggests juvenile
Chinook salmon will be exposed to peak carbaryl concentrations in the water column and also
from feeding on live and dead contaminated prey. Also, recent information (six years of
coastline surveys) shows that juvenile Columbia River Chinook salmon salmon "bounce" north
along the Washington coastline and south along the Oregon coastline in the nearshore surf zones
(Personal communication Eduardo Casillas NMFS 2/18/2009 ) and have been found occupying
nearshore habitats as far North as the Strait of Juan de Fuca (Shaffer, Grain et al. submitted). In
one study, juvenile Chinook salmon salmon captured along the coast Washington State's were
genetically analyzed to determine natal river origins. Of the fish collected for genetic analysis,
45% of juvenile Chinook salmon collected in Crescent Bay, 75% of those in Freshwater Bay,
and 60% of those in Pysht Bay were from ESA-listed stocks of the Columbia River (Shaffer,
Grain et al. submitted).
DNA analysis revealed that adult salmonids harvested in Willapa Bay in July and August, when
carbaryl is typically applied include several ESA-listed species: Lower Columbia River Chinook
salmon, California Coastal Chinook salmon, Puget Sound Chinook salmon, and Snake River
Fall-run Chinook salmon (Kassler and Marshall 2004). Given this information we also expect
these same species of Chinook salmon will be present in Gray's Harbor as Grays Harbor is
located just North of Willapa Bay. Further, we expect adults of these stocks may be exposed to
carbaryl given that they occupy the estuary during the application periods and oyster plots occur
throughout the Bay. It is possible that chum salmon from the Columbia River also occur in the
Willipa Bay and Gray's Harbor estuaries. However, we are unaware of documentation of their
occurrence at those locations, and unlike juvenile Chinook salmon from the Columbia River,
juvenile chum have not been found in recent sampling efforts of nearshore surf zones. We
therefore conclude that Columbia River ESA-listed chum are unlikely to be present in Willapa
Bay or Gray's Harbor.
DRAFT 381
-------
DRAFT
Measured Concentrations in Willapa Bay and Gray's Harbor following carbaryl applications
Numerous monitoring studies have been conducted in coordination with applications of carbaryl
to control burrowing shrimp in commercial oyster beds in Washington State. The results of the
studies are quite variable, as are the study designs and objectives. Several investigations have
documented water column concentrations exceeding several mg/L (which is substantially higher
than the |j,g/L concentrations observed in freshwater aquatic habitats) following the first flood
tide post-application (Hurlburt 1986; Creekman and Hurlburt 1987; Tufts 1989; Tufts 1990).
Water column concentrations measured at Willapa Bay in 1984 detected a mean concentration of
10.6 mg/L upon initial flooding of the treated area (Hurlburt 1986). Concentrations were
monitored at varying depths for several hours following flooding. The concentrations decreased
rapidly but were detectable in the low ng/L range throughout sampling. In 1985 samples were
collected along transects from treated areas following the direction of the tide to monitor the
dispersal of carbaryl (Creekman and Hurlburt 1987). The data indicate that carbaryl is
transported off the treated area by the tide. Average concentrations measured above the treated
area ranged from 2.4 - 5.5 mg/L. The highest average concentration (7.9 mg/L) was collected
510 ft from the treated area. Average concentrations at the most distant sampling point (650 ft
from initial treated area) were 2.5 mg/L. In 1996, further study was taken to evaluate carbaryl
movement off the treated spray area (Tufts 1989). Samples were collected along a transect
corresponding to the direction of the incoming tide. The peak concentration measured was 27.8
mg/L. Over one treated area, the concentration of carbaryl decreased from 13.2 mg/L to 9.3
mg/L as the water depth increased from 1.5 to 10 inches. The concentration further decreased to
600 |ig/L when the water rose to 16 inches. The monitoring indicated concentrations of carbaryl
decreased with increasing distances along the transect, but concentrations greater than 1 mg/L
were detected in several instances at distances several hundred feet from treated plots. Carbaryl
was found in the water column at the detection limit (0.1 mg/L) as far as 1,725 ft from the treated
area.
In 1987, surface water monitoring was conducted in Willapa Bay to determine "the dilution
pattern of carbaryl washed from sprayed oyster beds, and to determine carbaryl concentrations in
shallow pools and streams where marine fish might be found" (Tufts 1990). Carbaryl
concentrations in the mg/L range were frequently detected with the incoming tide. Peak
DRAFT 382
-------
DRAFT
concentration for one sample site was 18.8 mg/L upon initial flooding. The concentrations
decreased to 0.2 mg/L when covered by 18 inches of water. Concentrations of carbaryl were
variable among sites. For example, at one of the sampling sites carbaryl was not detected until
the water was 7 inches deep. A peak concentration of 8 mg/L was detected at this site when the
water depth was 11 inches. This sample station was located 300 ft from a treated area.
Maximum carbaryl concentrations detected at other sites were 17.4, 7.8, and 4 mg/L with
samples collected at depths of 6, 4, and 13 inches from the bottom. Carbaryl's primary toxic
degradate, 1-naphthol, was detected at concentrations as high as 1.4 mg/L. Average
concentrations of carbaryl detected in tide pools and small streams ranged from 3.6 to 11.2 mg/L.
The 11.2 mg/L detection was associated with application of 5 Ibs carbaryI/acre. An average
concentration of 7.8 mg/L in tide pools and streams was associated with an application rate of 4
Ibs carbary I/acre, approximately half the maximum rate allowed. Similarly, a more recent study
found a peak of 820 ng/L in the water column 50 ft from the application site following a typical
treatment application at the maximum rate of 8 Ibs/acre (n=3) (Weisskopf and Felsot 1998).
Several studies demonstrate that carbaryl dissipates fairly rapidly from over the treatment site
due to degradation, metabolism, dilution, and off-site transport (Hurlburt 1986; Creekman and
Hurlburt 1987; Tufts 1989; Tufts 1990; Weisskopf and Felsot 1998). In 2006 and 2007, samples
collected at the mouth of channels adjacent to treated oyster beds in Willapa Bay had maximum
concentrations of 29.1 |ig/L after 6 h (high tide), 38 |ig/L after 12 h (low tide), and21.1 |ig/L
after 24 h (low tide) (Major, Grue et al. 2005).
The NPDES permit for Willapa Bay and Gray's Harbor requires annual monitoring of water
column concentrations in treated areas. It specifies an acute effluent limit of 3 jig/L and a
chronic limit of 0.06 |ig/L. However, those data are of highly questionable value because the
monitoring plan specifies that monitoring is suspended for the first 48 h following application to
assess the acute effluent, and further suspends monitoring for 30 days from the last application to
assess the chronic limit. We expect that most of the carbaryl will be degraded and transported to
other locations by the time carbaryl monitoring is initiated.
DRAFT 383
-------
DRAFT
Carbaryl sprayed on mud flats can be transported substantial distances at concentrations that may
have ecological impacts. Researchers found that close to 100% of Dungeness crabs were killed
up to 100 m off the carbaryl application area (Doty, Armstrong et al. 1990). Levels decrease to
below 1 mg/L when transported more than 200m or more. Washington DOE reports that
carbaryl concentrations in the "potential effects threshold range" of 0.1-0.7 [ig/L have been
detected at locations several miles from oyster beds soon after large areas were treated (Johnson
2001).
Measured Concentrations in Invertebrates
Sample of crustaceans from carbaryl treated areas in 1984 showed high tissue levels. A single
ghost shrimp and Dungeness crab were analyzed and contained concentrations of 24.9 and 41.9
mg/kg (ppm), respectively (Hurlburt 1986). Analysis of dead shrimp in 1985 following
treatment at 7.5 Ibs carbaryl/acre revealed average concentrations of approximately 4.5 mg/kg.
When left on the treated oyster beds the concentrations declined to approximately 10% of the
initial concentration 24 h post-treatment, then remained relatively stable at the 48, 72, and 96 h
sampling events. Concentrations in shrimp and annelid worms were investigated following the
1985 applications and associated with different application rates (Tufts 1989). Average
concentrations in shrimp ranged from 5.3 to 13.8 mg/kg. The concentration in worms ranged
from 57-58.6 mg/kg. The 10 Ib application rate is less relevant as the current label restricts
applications to 8 Ibs a.i./acre. Concentrations in other, more relevant salmonid prey items were
apparently not assessed although we assume they may be comparable for other organisms that
occupy similar habitat.
Table 68. Carbaryl concentrations in aquatic organisms on treated sites (mg/kg).
Year
1984
1985
1986
Dungeness Crab
41.9
-
-
Burrowing Shrimp
24.9 (rate not reported)
4.5 (7.5 Ib a.i./acre)
1 3.8 (10lb a.i./acre)
8.7 (7.5 Ib a.i./acre)
5.3 (5 Ib a.i./acre)
Annelid Worms
-
-
75.5 (5 Ib a.i./acre)
57 (7.5 Ib a.i./acre)
58.6 (5 Ib a.i./acre)
Ecological Effects
There have been concerns that carbaryl applications in Willapa Bay and Gray's Harbor may have
adverse effects to the commercial Dungeness crab fishery (Feldman, Armstrong et al. 2000).
DRAFT
384
-------
DRAFT
Consequently, Washington Department of Fisheries conducted several surveys to estimate the
number of crabs killed (Hurlburt 1986; Creekman and Hurlburt 1987; Tufts 1989; Tufts 1990).
Transect surveys conducted on treatment beds the day following applications indicate large
numbers of crabs are killed by the applications (Table 69). Other acute mortalities to non-target
organisms were generally not reported. However, transect surveys were conducted during 1986
and 1987 due to concerns over potential impacts to fish given staff observations from the
Washington Department of Fisheries reported that fish mortality was "routinely" noted, but not
quantitatively assessed following applications of carbaryl (Tufts 1989). Mortalities were
characterized as small fish apparently trapped in shallow pools during low tide and directly
exposed during carbaryl treatments. The surveys indicate several thousand fish were killed each
year from carbaryl applications to approximately 400 acres at rates of < 7.5 Ibs a.i./acre.
Currently, the NPDES permit allow for treatment of up to 600 acres. The current 24 (C) label
does not specify acreage restrictions and allows for applications of carbaryl up to 8 Ibs a.i./acre.
Table 69. Estimated dungeness crab mortalities resulting from carbaryl applications in Willapa
Bay and Gray's Harbor, Washington
Year
1984
1985
1986
1987
Maximum Application Rates*
< 10 Ibs a.i. /acre
< 7.5 Ibs a.i./acre
< 7.5 Ibs a.i./acre
< 7.5 Ibs a.i./acre
Total Acres Treated**
490
391
398
434
Dungeness Crab Killed
38,410
59,933
16,286
44,053
* Current 24(c) label allows for application of up to 8 Ibs a.i./acre
** Current NPDES permit allows for a total of 600 hundred acres to be treated in Willapa Bay and Gray's
Harbor, WA. The Label places no restrictions on acres or geographic restrictions for use of carbaryl on
oyster beds in Washington.
Given that the fish that reside in standing water on mud flats are likely to be the most vulnerable
to carbaryl exposure, staff from the Washington Department of Fisheries also estimated the
available marine fish habitat that existed on exposed mudflats of treated areas. They
characterized water that was at least 2 inches in depth, a depth that is adequate for salmon fry as
marine habitat. This habitat comprised substantial portions of the treated areas. In 1986, surveys
indicate 135 acres of the 398 of the treated area (approximately 34%) were marine fish habitat
(Tufts 1989). In 1987, 67 of the 434 (15%) acres were characterized by Washington Department
of Fisheries as marine fish habitat. The current NPDES permit specifies that there be a 200 ft
buffer zone for sloughs and channels when carbaryl is applied by helicopter. A 50 ft buffer is
required for those aquatic habitats when carbaryl is applied by hand sprayer. AgDrift estimates
DRAFT 385
-------
DRAFT
for aerial application at an application rate of 8 Ibs a.i./acre with a 200 ft buffer indicate the
average initial concentration from drift to a body of water that is 2 inches deep (~ 5 cm) and 10
m wide would be 771 u.g/L. It seems highly unlikely that helicopter applications would be able
to successfully avoid direct overspray of some of the marine fish habitats on the exposed
mudflats. Additionally, the 24(C) labels do not contain buffer zone restrictions. A direct
overspray of 2 inches of water would result in an average initial concentration of approximately
18 mg/L, a concentration comparable to measured concentrations associated with initial tidal
inundation of treated mud flats (see surface water detections above). Acute exposure to these
concentrations are expected to kill a portion of exposed salmonids in a matter of hours i.e., 24 h
LCSOs for carbaryl range from 948-8,000 [ig/L, n= 60 (Mayer and Ellersieck 1986), and
significantly reduce AChE activity that will lead to myriad sublethal effects such as impaired
swimming.
Table 70. Estimated fish mortalities resulting from carbaryl applications in Willapa Bay and Gray's
Harbor, Washington
Year
1986
1987
Maximum Application Rates*
< 7.5 Ibs a.i./acre
< 7.5 Ibs a.i./acre
Total Acres Treated**
398
434
Fish Killed
14,954
8,041
* Current 24(c) label allows for application of up to 8 Ibs a.i./acre
** Current NPDES permit allows for a total of 600 hundred acres to be treated in Willapa Bay and Grays
Harbor, WA. The Label places no restrictions on acres or geographic restrictions for use of carbaryl on
Oyster beds in Washington.
Salmonids were not reported in the sampling data which occurred over two seasons. The authors
report that the fish kills were extremely variable and unpredictable. The fish mortality reported
for transect surveys included only four species in 1986: saddleback gunnel, threespine
stickleback, staghorn sculpin, and arrow goby. In 1987, mortalities included English sole, sand
sole, kelp greenlings, shiner perch, saddleback gunnels, staghorn sculpin, and arrow goby.
Response ofbenthic community
Samples collected in 1985 to assess impacts of carbaryl on the intertidal invertebrate community
(Tufts 1990). Treated area showed decreases in numbers of infaunal benthic crustaceans at 15
and 30 d post-treatment while a concurrent increase in crustacean abundance was observed on
control plots (Table 71). Abundance ofbenthic crustaceans was low on both control and treated
DRAFT 386
-------
DRAFT
plots 5 months after application in December. The authors suggest the data may represent a
seasonal decline in benthic crustaceans.
Table 71. Total number of benthic crustaceans (tanaids, cumaceans, amphipods, copepods, and
ostracods; adapted from Tufts 1990).
Date
June 30
July 16
August 01
December 10
Sampling
Pretreatment
15 days post-treatment
30 days post-treatment
161 days post-treatment
Control Plot
(change from pretreatment)
75
96 (+28%)
136(+181%)
2
Treated Plot
(change from pretreatment)
33
2 (-94%)
4 (-82%)
6
Field incidents reported in EPA incident database
NMFS reviewed reported incidents offish deaths from field observations throughout the U.S.
because the information reflects real world scenarios of pesticide applications and corresponding
death of freshwater fish. We recognize that much of the information is not described in
sufficient detail to attribute an incident to a label-permitted use leading to the death offish, or to
make conclusions regarding the frequency offish kills that may be associated with the use of
pesticides. NMFS uses the information as a component to evaluate a line of evidence- whether
or not fish kills have been observed from labeled uses of the three pesticide products. EPA
categorizes incidents in the database into one of five levels of certainty: highly probable,
probable, possible, unlikely, or unrelated. The certainty level indicates the likelihood that a
particular pesticide caused the observed effects. EPA uses the following definitions to classify
fish kill incidents:
• Highly probable (4): Pesticide was confirmed as the cause through residue analysis or
other reliable evidence, or the circumstances of the incident along with knowledge of the
pesticides toxicity or history of previous incidents give strong support that this pesticide
was the cause.
• Probable (3): Circumstances of the incident and properties of the pesticide indicate that
this pesticide was the cause, but confirming evidence is lacking.
• Possible (2): The pesticide possibly could have caused the incident, but there are
possible explanations that are at least as plausible. Often used when organisms were
exposed to more than one pesticide.
DRAFT
387
-------
DRAFT
• Unlikely (1): Evidence exists that a stressor other than exposure to this pesticide caused
the incident, but that evidence is not conclusive.
• Unrelated (0): Conclusive evidence exists that a stressor other than exposure to the given
pesticide caused the incident.
NMFS reviewed several incident reports provided by EPA from OPP's incident database. This
database is populated with reports received by EPA from registrants that are defined as
reportable under FIFRA 6(a)(2) and includes other information received from registrants and
other sources.
Relatively few incidents involving lethality to fish were documented in the database, considering
the length of time these products have been registered and their toxicity profiles. EPA provided
the following table summarizing known mortality incidents (Table 72). NMFS is uncertain as to
how representative this database is of known fish kills and the level of coordination that has
occurred with various state and federal agencies that investigate these incidents. The mortality
events discussed above for carbaryl applications in Willapa Bay and Gray's Harbor were not
recorded in the database. Several of the incidents provided by EPA are discussed in more detail
below.
Table 72. EPA summary of field incident data with carbaryl, carbofuran, and methomyl;
highlighted incidents (yellow) are discussed.
Carbaryl
Incident ID
BOOOO-246-01
BOOOO-501-92
I000799-003
1000910-001
I007720-020
1013436-001
1018436-001
1015419-644
Certainty
Probable
Probable
Possible
Unlikely
Probable
Possible
Possible
Carbofuran
Incident ID
1000165-052
I000599-008
1001605-003
1005416-001
1012265-001
1012265-003
1012265-004
1013436-001
Certainty
Possible
Probable
Probable
Highly
Probable
Possible
Probable
Possible
Possible
Methomyl
Incident ID
BOOOO-216-19
BOOOO-501-36
1000108-001
1013436-001
Certainty
Unlikely
Unlikely
Probable
Possible
DRAFT
388
-------
DRAFT
Incident 1013436-001: A report of a large fish kill on October 16, 2001, was investigated by the
California Department of Fish and Game. The kill involved several thousand fish. Although
salmonids were not identified, it occurred in the San Joaquin River, which is used by listed
salmonids. EPA characterized this event as "possible" regarding its association with carbaryl,
carbofuran, and methomyl. Methomyl was detected in fish gills and several other pesticides
were detected in the water including azinphos-methyl, another cholinesterase inhibitor, and
several chlorinated hydrocarbons (chlordane, DDT, dieldrin, methoxychlor, mirex, and others).
It is likely that the kill was in part due to mixture toxicity, particularly with the combinations of
carbamates and OP insecticides.
Incident 1000599-008: A report by the California Department of Fish and Game indicates
approximately 3,000 fish, 4,000 crayfish and frogs, 200 birds, and at least 5,000 invertebrates
were killed on North Temple Creek in San Joaquin County, California. The adjacent vineyard
had been treated with Furadan 4F (carbofuran). Chemical analysis confirmed carbofuran in the
gizzards of starling up to 9.8 mg/kg, crayfish collected on site had 0.6 mg/kg, the stomach
contents of great egrets, which were primarily crayfish, contained 0.3 mg/kg, and surface water
had detections up to 7,800 |ig/L. The California Department of Fish and Game concluded that
the preponderance of evidence indicated carbofuran was the cause of death. EPA characterized
the cause of death as probable.
Incident BOOOO-246-01: According to a 1975 report by EPA, 22,000 catfish were killed in 7
miles of stream at an agricultural site in Oklahoma. Minimal information on the incident was
reported. Chemical analysis was not conducted but EPA classified this association with carbaryl
as probable. Other a.i.s were identified as possibly contributing to the kill were toxaphene,
methyl parathion, and endrin.
Incident BOOOO-501-92: EPA concluded that a fish kill in a New Jersey pond was caused by
drift of carbaryl following the application of Sevin in 2000. This incident was characterized as
probable. No chemical analysis was conducted.
DRAFT 389
-------
DRAFT
FMC reported several incidents called in to their emergency telephone number in the second
quarter of 1992. One incident reported a small number of dead bluegill observed in a drain pond
adjacent to corn and tobacco fields. The caller indicated Furadan 15G (carbofuran), Counter
(chlorpyrifos), and Temik (aldicarb) had been applied to the crops. Chlorpyrifos is an AChE
inhibiting organophosphate insecticide. Aldicarb and carbaryl are both cholinesterase inhibiting
TV-methyl carbamates. Dead fish were observed following heavy rains. No analysis was done on
the fish to determine the cause of death. EPA characterized carbofuran as a "possible" cause of
the incident. It seems likely that the three cholinesterase inhibiting insecticides may have
resulted in a cumulative exposure that was sufficient to cause the fish kill due to mixture toxicity.
Incident 1000108-001: DuPont reported a fish kill in 1992 in southern Georgia that occurred
following an application of Lannate LV (methomyl). One hundred and twenty-five fish were
found dead in a pond located 50 to 75 yards from a large field of sweet corn. The 200 acre field
was treated with Lannate LV (methomyl) five times between June 1 and June 16. Additionally,
Lorsban insecticide (chlorpyrifos) was applied four times during the same interval including co-
applications of four fertilizer treatments. On the day of the fish kill (June 16), applications of
both Lannate and Lorsban were made. Lannate LV was applied at rates of 1.5 pints/acre and
Lorsban was applied at a rate of 1 pint/acre. During the first 16 days of June a total of 10.52
inches of rain were received. The State of Georgia took water samples on June 17. Analysis
indicated 136 |ig/L of methomyl in the pond. Chlorpyrifos concentrations were not reported and
it is not known if they were analytes. Water temperatures were relatively high (reported to be
about 90°F) but dissolved oxygen was reported to be "normal (7.2-10)". DuPont suggested the
high temperatures, fertilizers, and suspended solids could be stressful to fish and low oxygen
content at night might have played a part in the deaths. The species involved in the incident
(carp and bluegill) were warm water fishes. EPA characterized methomyl as a "probable" cause
of the fish kill. Given the reported rapid dissipation of methomyl it is likely that the
concentrations measured do not represent peak concentrations that occurred in the pond. It
seems likely that chlorpyrifos may have also been a contributing factor in this incident.
DRAFT 390
-------
DRAFT
Mixture Analysis ofCarbaryl, Carbofuran, andMethomyl
As noted earlier, pesticides most often occur in the aquatic environment as mixtures. In our
review and synthesis of the available exposure and response information, we find the three TV-
methyl carbamates carbaryl, carbofuran, and methomyl, share the same mechanism of toxic
action (AChE inhibition) and are expected to co-occur in salmonid habitats. Therefore, we
employ a simple mixture analysis derived from empirical data with Pacific salmonids to predict
potential effects to individual salmonid's AChE activity and their survival from short-term
exposures. The analysis is predicated on the toxic potencies of the three insecticides added
together to predict the resulting cumulative effect to AChE activity and mortality.
Mixture toxicity is typically described by three general responses: antagonistic, additive, or
synergistic. Antagonism and synergism are where the toxic response is not predicted by the
individual potencies of the pesticides found in the mixture. Antagonistic effects of a mixture
lead to less than expected toxicity on the organismal endpoint. Mechanistically, the pesticides
are likely interacting with one another to reduce the toxic potency of individual pesticides.
Synergistic effects of a mixture lead to a greater than expected effect on the organismal endpoint
and the pesticides within the mixture enhance the toxicity of one another. The third general type
of mixture toxicity and the one most frequently reported is additivity (known also as dose-
addition or concentration-addition). This type of response is defined by adding the individual
potencies of pesticides together to predict the effect on the biological endpoint. Additivity has
been demonstrated for many pesticide classes as well as other organic compounds such as PAHs,
PCBs, and dioxins.
Additive toxicity of chemicals that share a mode or mechanism of toxic action is well established
in the scientific literature, and as a result has been informing regulatory decisions for more than a
decade. In 1996, the National Academy of Sciences recommended a dose-additive approach to
assessing risks to human infants and children from pesticide exposure. EPA currently assesses
human risk of pesticide mixtures for pesticides that share a common mechanism of toxic action
e.g., TV-methyl carbamates [such as carbaryl, carbofuran, methomyl], organophosphorus
insecticides, chloroacetanilide and triazine herbicides, as mandated by FQPA. The analysis EPA
conducts is predicated on additive toxicity and applies dose-addition to set tolerance limits of
DRAFT 391
-------
DRAFT
pesticide residues on food. For example, the toxic potencies of the TV-methyl carbamates are
added together to determine pesticide tolerance limits for edible crops. Although additive
toxicity is evaluated when determining risk to humans, EPA OPP has yet to apply a similar
approach to address cumulative toxicity of pesticides that share a common mode or mechanism
of action in the evaluation of terrestrial and aquatic species. That said, the use of dose-addition
for mixtures containing acetylcholinesterase-inhibiting pesticides is well established and has
been extended to protection of aquatic life (Belden, Gilliom et al. 2007).
Dose-addition assumes the cumulative toxicity of the mixture can be predicted from the sum of
the individual toxic potencies of each component of the mixture. Within the past decade,
government regulatory bodies started to use dose-addition models to predict toxicity for those
chemicals that share a common mode of action. In California, the CVRWQB uses dose-addition
models (based on the toxic-unit approach) to develop TMDLs for the OP insecticides, diazinon
and malathion. NMFS Biological Opinions have also recognized the environmental reality of
co-occurring pesticides in species' aquatic habitats and applied additive toxicity models to
predict potential responses of salmonids NMFS 2004, NMFS 2005a, NMFS 2005b, NMFS
2005c(NMFS 2008).
In salmon, dose-additive inhibition of brain AChE activity by mixtures of OPs and carbamates
was demonstrated in vitro (Scholz, Truelove et al. 2006). More recently, it has been found that
salmonid responses to OP and carbamate mixtures vary in vivo; some interactions were
synergistic, rather than just additive (Laetz, Baldwin et al. In Press). We used the dose-addition
method to predict responses by applying the modeling exposure estimates and
measured concentrations of carbaryl, carbofuran, and methomyl presented in the Exposure
Analysis. Effects of the three carbamates individually and in combination on AChE inhibition
Figure 42A) and survival (B) are shown below. Based on additivity, the mixture is expected to
be more toxic than the individual carbamates for both endpoints. Due to the steep slopes of the
two dose-response curves, and especially the mortality slope, small changes in concentrations
elicit large changes in observed toxicity. The exposure values represent concentrations from
EPA PRZM-EXAMs 60 d average modeling estimates for surface waters (carbaryl: 4 aerial
applications at 5 Ibs a.i./acre, citrus in FL; carbofuran: 1 ground application at 2 Ibs a.i./acre,
DRAFT 392
-------
DRAFT
artichokes in CA; methomyl: 3 aerial applications at 1.8 Ibs a.i./acre, peaches). We recognize
that this approach is likely to under-predict toxicity for mixtures that produce synergistic rather
than additive responses (Laetz, Baldwin et al. In Press).
0.01 0.1 1 10 100
Concentration relative to EC50
% 40 -
Cl 41 ug/L
Cnl9 ug/L
M 85 ug/L
0.01 0.1 1 10 100
Concentration relative to LC50
Figure 42. Percent AChE inhibition (A.) and percent mortality (B.) expected from exposure to
carbaryl (Cl), carbofuran (Cn), and methomyl (M) as separate constituents and as mixtures (Cl 41
ug/L, Cn 19 ug/L, and M 85 ug/L)4.
We used a variety of exposure estimates and monitoring data to evaluate responses to different
mixtures of carbaryl, carbofuran, and methomyl (Table 73). The predicted additive responses
from these mixtures ranged from 16-74% inhibition of AChE and 0.01-74% mortality. The
predicted additive response to AChE inhibition is likely to result in increased behavioral
consequences to salmonids. What is not captured in these responses is the likelihood of exposure
to the various mixture concentrations. The PRZM-EXAMS values were estimates selected from
EPA simulations of western crops. The scenarios were representative of use rates and numbers
of applications on current product labels. Additionally, we used 60 d, time-weighted averages
estimates of exposure rather than predicted peak concentrations as exposure to multiple
4 EPA's default pesticide slope was used for acute mortality (3.63 or probit slope of 4.5) [EPA 2004]. The slope
used for AChE inhibition was based on pooling data from five cholinesterase-inhibiting insecticides; including
carbofuran, carbaryl, chlorpyrifos, diazinon, and malathion Laetz, C. A., D. H. Baldwin, et al. (In Press). "The
synergistic toxicity of pesticide mixtures: Implications for ecological risk assessment and the conservation of
threatened Pacific salmon." Environmental Health Perspectives ehponline.org: doi:10.1289/ehp.0800096.
DRAFT
393
-------
DRAFT
pesticides would be expected to occur more frequently over chronic durations. This may
underestimate effects as responses assumed 96 h exposure. Site specific considerations will also
have an influence on the frequency of exposure.
Table 73. Predicted AChE inhibition and mortality from estimated and measured exposure to
carbaryl, carbofuran, and methomyl.
Concentration (ug/L)
% AChE Inhibition
% Mortality
Modeling: PRZM-EXAMS 60-day averages1 (from Table 49)
Carbaryl
Carbofuran
Methomyl
Additive response
12
19
81
8.34
25.39
28.33
44.30
0.00
0.04
0.09
1.38
Modeling: GENEEC 90-day averages (from Table 50)
Sweet Corn
Carbaryl
Carbofuran
Methomyl
Additive response
229
53
49
60.67
47.67
19.61
72.24
42.10
1.63
0.01
73.62
Potatoes
Carbaryl
Carbofuran
Methomyl
Additive response
137
106
34
48.51
63.93
14.66
73.63
10.13
17.02
0.00
69.52
Citrus
Carbaryl
Methomyl
Additive response
280
21
65.25
9.76
66.25
60.14
0.00
62.97
Monitoring: NAWQA maxima in 4 states (from Table 54)
Carbaryl
Carbofuran
Methomyl
Additive response
33.5
32.2
0.82
19.59
36.02
0.48
44.22
0.07
0.27
0.00
1.79
Monitoring: California CDPR database maxima (from Table 55)
Carbaryl
Carbofuran
Methomyl
Additive response
8.4
5.2
5.4
6.07
8.93
2.85
15.58
0.00
0.00
0.00
0.01
Monitoring: Washington EIM database maxima (from Table 56)
Carbaryl
Carbofuran
Methomyl
Additive response
10
2.3
0.17
7.09
4.29
0.11
10.60
0.00
0.00
0.00
0.00
PRZM-EXAMS estimates for carbaryl in California peaches (2 applications at 7 Ib a.i./acre), carbofuran in
California artichokes (1 application at 2 Ib a.i./acre), and methomyl in lettuce (10 applications at 0.9 Ib
a.i./acre). Although current labeling may not be consistent for these uses (peaches, artichokes, and
lettuce), the use rate and number of applications are consistent with other labeled uses within the action
area.
DRAFT
394
-------
DRAFT
The GENEEC estimates are 90 d, time-weighted averages that were based on labeled uses of the
three compounds in sweet corn, potatoes, and citrus. We found no restrictions that would
prevent co-application or sequential applications of carbaryl, carbofuran, and methomyl. The
application rates assumed were consistent with current labels and generally representative of use
rates authorized for many crop and non-crop uses. An exception was the assumed carbaryl use
rate in citrus (12 Ib a.i./acre), which is substantially higher than the maximum use rate approved
for most crops.
The NAWQA, CDPR, and EIM monitoring values represent the maximum concentrations found
in the respective databases. The values cited were all measured within the four states and the
vast majority is from waters that contain or drain to listed salmonid habitats. Most of the
detections in these, and other monitoring studies that did not target specific applications of the
three chemicals occurred at or below the |ig/L level. We expect that exposure at these levels will
be common in drainages where the three products are used extensively.
Evaluation of Risk Hypotheses: Individual Salmonids
In this phase of our analysis we examine the weight of evidence from the scientific and
commercial data to determine whether it supports or refutes a given risk hypothesis. We also
highlight general uncertainties and data gaps associated with the data. In some instances there
may be no information related to a given hypothesis. If the evidence supports the hypothesis we
determine whether it warrants an assessment either at the population level, or affects PCEs to
such a degree, to warrant an analysis on the potential to reduce the conservation value of
designated critical habitat.
1. Exposure to carbaryl carbofuran, and methomyl is sufficient to:
A. Kill salmonids from direct, acute exposure.
A large body of laboratory toxicity data indicates that anadromous salmonids die following
short-term (< 96 h) exposure to the three insecticides at levels above 200 |ig/L. We expect
concentrations of carbaryl, carbofuran, and methomyl in salmonid off-channel habitats will reach
lethal levels based on exposure concentrations derived from monitoring data, EPA's modeling
DRAFT 395
-------
DRAFT
estimates, and NMFS modeling estimates. The youngest swimming salmonids appear to be the
most likely to die from short-term, acutely toxic exposures in these habitats. It is less likely that
adults would be killed by acute concentrations in most freshwater aquatic habitats compared to
juveniles. However, if adults are present in smaller off-channel habitats during aerial
applications or severe runoff events death is possible, particularly from carbaryl and carbofuran.
The available monitoring data, if representative of salmonid habitats, indicated that
concentrations rarely achieve LC50 values for the three compounds in freshwaters. However, it
is unlikely that peak concentrations are reflected in the monitoring data, and given the acutely
toxic nature of carbamates, a brief exposure would be sufficient to cause effects. As described in
the exposure section, monitoring data are limited when compared to the range of habitats used by
salmonids. Few data were found that targeted applications and subsequent concentrations in
edge of field habitats which typically show much higher concentrations than weekly, monthly, or
seasonal monitoring efforts. Although we found no information on egg survival following acute
exposures, we do not expect death of eggs from these insecticides as entry into the eggs via the
water column is unlikely. Further support for acute lethality to fish is found in field incidences
of death attributed to carbaryl, carbofuran, and methomyl that EPA ranked as "probable". We
located several incidents showing death of freshwater fish following exposures to TV-methyl
carbamates. We expect juveniles of listed salmonids to be at the highest risk of death when in
small freshwater off-channel and edge habitats, and secondarily in estuarine habitats where
carbaryl is applied directly to mudflats. In conclusion, the available information on measured
and expected concentrations of the three insecticides supports this hypothesis. We translate the
fitness level consequences of reduced survival from mortality of juvenile salmonids to potential
population level consequences using population models (see population modeling section
below).
B. Reduce salmonid survival through impacts to growth.
Fish growth is reduced following long-term exposures to carbofuran and methomyl in fathead
minnows and rainbow trout, respectively. EPA reported on a single test that measured growth
effects to fathead minnows following 30 and 60 d exposure to carbaryl at 0, 8, 17, 62, 210, and
680 |ig/L (Carlson 1971). No statistically significant effects on growth were reported in this
study. It is difficult to extrapolate from this one study with a warm water species to potential
DRAFT 396
-------
DRAFT
growth effects to ESA-listed salmonids especially given that salmonids appear substantially
more sensitive to carbaryl's acute toxicity than fathead minnows (96 h LCSOs for salmonids
range from 250-4,50Q|ig/L carbaryl and for fathead minnows 7,700-14,600 |ig/L). Additionally,
20% and 50% inhibition of AChE in salmonids occurs at concentrations as low as 23 and 185
|ig/L, respectively (Labenia, Baldwin et al. 2007) and this inhibition was found to affect
swimming behavior (Labenia, Baldwin et al. 2007). Reduced growth occurred at 56 |ig/L for
carbofuran and 142 |ig/L for methomyl. Only one test result was reported for each pesticide.
We did not identify any studies that provided a quantitative relationship between growth and fish
survival in the field or lab. However, there is abundant literature that shows salmonids that are
smaller in size have reduced first year survival (Appendix 1). Additionally, exposure to
sublethal concentrations of other AChE inhibitors (chlorpyrifos and diazinon) for acute durations
does cause reduced feeding success, which likely results in impacts to growth (Scholz, Truelove
et al. 2000; Sandahl, Baldwin et al. 2005). We expect that juvenile fish exposed to carbaryl,
carbofuran, and methomyl during their freshwater residency will feed less successfully, resulting
in lower growth rates and reduced sizes. Exposure concentrations will likely vary temporally
and spatially for salmonids depending on life history, pesticide use, and environmental
conditions. The available information support that growth is likely reduced where salmonids are
exposed to more than 56 |ig/L carbofuran and to concentrations below LC50 for carbaryl and
methomyl, although the exact concentrations that may cause growth reduction are currently
unknown. The weight of evidence supports the conclusion that fitness level consequences from
reduced size are likely to occur in individual salmonids exposed to the three TV-methyl
carbamates. Therefore, we address the potential for population level repercussions due to
reduced growth using species-specific population models.
C. Reduce salmonidgrowth through impacts on the availability and quantity ofsalmonid prey
We address several lines of evidence to determine the likelihood of reduced salmonid growth
from impacts to aquatic invertebrate prey. The first line of evidence we evaluated is whether
salmonid prey items are sensitive to acute and chronic exposures from expected concentrations
of the three carbamates. These primarily involved evaluating laboratory experimental results
that reported on incidences of death or sublethal effects. We located a total of 10, 4, and 14
survival estimates (24, 48, 72, and 96 h EC/LC50s) for carbaryl, carbofuran, and methomyl,
DRAFT 397
-------
DRAFT
respectively. Based on an evaluation of the assessment endpoints, we found a robust body of
exposure and toxicity data that indicated salmonid aquatic prey are highly sensitive and affected
by expected exposures to each of the insecticides as well as from mixtures containing the three
insecticides. We expect death and a variety of sublethal effects to salmonid prey items.
The second line of evidence is whether field level reductions in aquatic invertebrates correlate to
TV-methyl carbamate insecticide use and/or concentrations in salmonid habitats. We found
several examples supporting this line of evidence. The available laboratory and field data show
reductions in aquatic invertebrate taxa and reductions in invertebrate abundances following
applications of carbofuran and carbaryl (Table 67). Furthermore field studies explicitly
investigated whether real world applications and subsequent pesticide drift and/or runoff into
aquatic environments affected receiving water aquatic communities. One compelling study
measured pre- and post aquatic community responses from field applications of carbaryl at 0.75
and 1 Ib/acre (these rates are at the lower end of approved maximum application rates, up to 12
Ibs/acre) (Courtemanch and Gibbs 1980). Applications correlated to substantial drift of salmonid
type aquatic invertebrates, of which most were dead or dying. More striking was that population
reductions of plecoptera (stoneflies) persisted for sixty days (Courtemanch and Gibbs 1980).
Runoff from fields treated with carbofuran at 2.68 Ibs a.i./ acre contained concentrations as high
as 264 (ig/L in the field ditches that drained to a stream where carbofuran was measured at 27
Hg/L (Matthiessen, Shearan et al. 1995). Caged amphipods present in the stream all died during
peak runoff following a rain event one month after the carbofuran application. This indicated
that carbofuran can persist in soil and mobilize by rain into subsequent runoff and ultimately
elicit toxic responses from stream invertebrates (Matthiessen, Shearan et al. 1995). Although we
found no field studies with methomyl, we expect that it would similarly reduce aquatic
invertebrate populations at higher concentrations than carbaryl and carbofuran based on its
toxicity profile.
The third line of evidence we evaluated was whether salmonids showed reduced growth in areas
of low prey availability, particularly those areas coinciding with use of carbaryl, carbofuran, and
methomyl. An evaluation of this line is complicated by multiple factors affecting habitat quality
i.e., water quantity, quality, riparian zone condition, etc., which in turn affects prey items and
DRAFT 398
-------
DRAFT
salmonids. We were unable to locate information attributing reduced growth in salmonids to
specific insecticide exposures that reduced prey, as most studies focused on measuring direct
effects on salmonids or direct effects on invertebrates (see review by Schulz 2004). However,
there are multiple field experiments and studies that demonstrate reduced fish growth resulting
from reduced prey availability (Brazner and Kline 1990; Metcalfe, Fraser et al. 1999; Baxter,
Fresh et al. 2007) or document fish growth rates below maximal potential growth rates when
prey are limited (Dineen, Harrison et al. 2007).
One study in particular, tested the hypothesis that single applications of the OP insecticide
chlorpyrifos (0.5, 5, 20 ng/L) to outdoor ponds (littoral enclosures) would reduce the abundance
of invertebrates and cause diet changes that would result in reduced growth rates of juvenile fish
(Brazner and Kline 1990). The results are direct, empirical evidence that support this hypothesis.
Growth rates of fathead minnow larvae were reduced significantly in all chlorpyrifos-containing
treatments due to reduction in prey abundance. At 15 d post-treatment, the reductions in growth
rate compared to control fish were the most pronounced and coincided with the greatest
reductions in invertebrates. Stomach contents of minnows were identified throughout the
experiment. By day 7 mean numbers of protozoans, chironomids, rotifers, cladocerans, mean
total number of prey being eaten per fish, and mean species richness were greater in unexposed
treatments compared to some of the other treatments. On day 15, most of the differences were
more pronounced. The results strongly support the conclusion that foraging opportunities were
better in untreated enclosures and unexposed larvae grew significantly more compared to
chlorpyrifos-treated enclosures. Furthermore, the reductions in prey items in diets mirrored the
reduction in prey items in the enclosures. We did not find any study results with TV-methyl
carbamates, but we make the inference that concentrations of the three TV-methyl carbamates that
are sufficient to reduce aquatic prey would also lead to reduced fish growth. This further
supports the hypothesis that reduction in prey abundances translates to reductions in subsequent
ration as well as individual growth. The authors concluded that "low levels of contaminants that
induce slower growth in young-of-the-year fish through food chain effects or other means may
eventually reduce the survival and recruitment of these fish."
DRAFT 399
-------
DRAFT
Collectively, the lines of evidence strongly support the overall hypothesis. Thus, we carry
reduced prey impacts to the next level of analysis (i.e., the population-level). We conducted
population modeling exercises based on reduced abundances of salmonid prey, presented in the
next section.
D. Impair swimming which leads to reduced growth (via reductions in feeding), delayed and
interrupted migration patterns, survival (via reduced predator avoidance), and reproduction
(reduced spawning success).
Swimming is a critical function for anadromous salmonids. The primary line of evidence for this
hypothesis is impaired swimming behaviors following exposure to carbaryl, carbofuran, and
methomyl. A secondary line of evidence for this hypothesis are studies showing swimming
behavior modification following exposure to other AChE inhibiting chemicals such as the OPs,
as we anticipate the results are similar in nature. Several studies regarding the effects of carbaryl
on the swimming related behaviors were reviewed, with effects on predator avoidance,
schooling, and feeding behaviors occurring at carbaryl concentrations of -100-1,000 |ig/L in
laboratory studies (Labenia, Baldwin et al. 2007) (Little, Archeski et al. 1990) (Arunachalam,
Jeyalakshmi et al. 1980) (Weis and Weis 1974) (Carlson, Bradbury et al. 1998). Only one study
was available for carbofuran showing effects to social behaviors at 5 jig/L (Saglio, Trijasse et al.
1996), and no studies were located for methomyl. Concentrations at which effects were noted
for carbaryl and carbofuran are within the ranges of concentrations estimated by the various
modeling methods, especially for the off-channel habitat locations. We anticipate similar effects
will occur with methomyl because it has the same mechanism of action as carbaryl and
carbofuran, inhihition of AChE (which is correlated to swimming impacts). However, we expect
it will take greater concentrations of methomyl compared to carbaryl and carbofuran to affect
swimming. We expect that these levels are likely from concentrations resulting from aerial drift
into off-channel habitats. This is becauseconcentrations of methomyl necessary forAChE
inhibition and acute lethality (LC50) are greater than carbaryl and carbofuran concentrations
necessary to inhibit AChE and kill fish . We also discussed compelling evidence that OPs impair
salmonid swimming behaviors and also show associated reductions in AChE activity.
DRAFT 400
-------
DRAFT
The most sensitive swimming endpoints are those associated with swimming activity compared
to those that measure swimming capacity (Little and Finger 1990; Little, Archeski et al. 1990).
The ecological consequences to salmonids from aberrant swimming behaviors are implied
primarily through the impairment of feeding, translating to reduced growth; aberrant migratory
patterns that ultimately reduce survival and reproduction. Impaired swimming behavior
correlated with both AChE inhibition and increased depredation rates (Labenia, Baldwin et al.
2007). Although NMFS was unable to locate results from field or laboratory experiments for the
other remaining endpoints of this hypothesis, we conclude that swimming behaviors are affected
by the three insecticides. Adverse effects to swimming-associated behaviors are directly
attributed to AChE inhibition, leading to potential reductions in an individual's fitness (i.e.,
growth, migration, survival, and reproduction). We therefore translate impaired swimming to
potential impacts on salmonid populations. Based on concentrations generated in modeling,
NMFS believes concentrations of carbaryl and carbofuran, applied in accordance with current
labels, are sufficient to impair swimming behavior of salmonids in some environments. As we
located no studies that evaluated the effects of methomyl on swimming behaviors in fish, we do
not know the exact levels sufficient to impair swimming except to say that levels well below
reported LCSOs can impair swimming. Swimming-mediated behaviors are frequently impaired
at 0.3 - 5.0% of reported fish LCSOs, and that 75% of reported adverse effects to swimming
occurred at concentrations lower than reported LCSOs (Little and Finger 1990). Taken together
this information supports methomyl affecting swimming behaviors below reported LCSOs. We
expect methomyl at hundreds of |j,g/L (expected from currently approved applications) to impair
swimming.
E. Reduce olfactory-mediated behaviors resulting in consequences to survival, migration, and
reproduction.
In the Opinion regarding effects of chlorpyrifos, diazinon, and malathion on listed salmonids
(NMFS 2008), sufficient data were available to conclude that olfactory-mediated behaviors were
affected by those pesticides. Less data were available to assess the effects of carbaryl,
carbofuran, and methomyl on these endpoints. Evidence is unclear as to mode of action for
insecticide-mediated olfactory impairment. Thus, conclusions drawn based on data for OPs may
not necessarily be applicable to TV-methyl carbamates. Data currently available are conflicting,
DRAFT 401
-------
DRAFT
with the one study available for carbaryl showing no apparent effect on olfaction on juvenile
cutthroat trout at concentrations of up to 500 |ig/L. Three studies regarding olfactory effects of
carbofuran were located and reviewed, and these indicated olfactory effects in several species of
fish (including two salmonids) at concentrations ranging from 1-10 |ig/L. These concentrations
are within both the ranges estimated by modeling, and the ranges measured in all monitoring data
sets. No data were available for methomyl. While data are not conclusive, based on what is
known, and giving the benefit of the doubt to the species, NMFS believes it is reasonable to
assume that these types of effects will occur from carbofuran exposures. Therefore, we discuss
the potential implications at the population level.
F. Exposure to mixtures of carbaryl carbofuran, and methomyl can act in combination to
increase adverse effects to salmonids and salmonid habitat.
The exposure and toxicity information we compiled, reviewed, and analyzed support the risk
hypothesis, although less mixture data were available for the carbamates addressed in this
Opinion than the OPs addressed in the previous Opinion (NMFS 2008). Evidence of additive
and synergistic effects on survival and AChE inhibition in salmonids were identified. Multiple,
independent study results supported additive toxicity from measured AChE inhibition. We
therefore conducted an analysis of potential mixtures on the levels of AChE inhibition and the
potential for an increased, reduced survival predicated on simple additivity. The analysis
showed that both survival and AChE inhibition of individuals is likely affected to a greater
degree than from exposure to a single chemical alone. We also expect that assessment endpoints
influenced by AChE inhibition are likely affected to a greater degree when in the presence of
more than one of the three insecticides. Considerable uncertainty arises as to the level of
impairment caused by mixtures for some endpoints, as dose responses have not been
characterized for some pesticide combinations. We conclude that this hypothesis is supported by
the available information and we assess the potential for population level consequences below.
DRAFT 402
-------
DRAFT
G. Exposure to other stressors of the action including degradates, adjuvants, tank mixtures,
and other active and other ingredients in pesticide products containing carbaryl, carbofuran,
and methomyl cause adverse effects to salmonids and their habitat.
Fifteen of the carbamate formulations contain other a.i.s, including malathion (OP insecticide),
bifenthrin (pyrethroid insecticide), rotenone (insecticide/piscicide), metaldehyde (mulloscicide),
captan (phthalidimide fungicide), and cupric sulfate (toxic to algae, aquatic invertebrates, and
fish). No data regarding the joint toxicity of these multi-a.i. products were presented in the BEs
or located in open literature. However, given that these pesticides are also toxic to aquatic life it
is reasonable to assume the salmonids and/or their prey items will exhibit a greater toxic
response when exposed to multiple a.i.s than to a single one. We are unable to estimate the
magnitude of such a response, and other than for malathion, are not currently able to evaluate
whether the response would fall into an additive, synergistic, or antagonistic category. Based on
mixture data discussed elsewhere in this Opinion, we assume additivity for carbaryl, carbofuran,
and methomyl.
Some toxicity data were available in the BEs for formulations not containing additional a.i.s.
Based on that data, no specific conclusions can be drawn as to whether formulations are, in
general, more or less toxic than the technical grade a.i. However, it should be noted that
adjuvants which increase either uptake (such as penetrants or surfactants that make physiological
membranes more permeable) or length of exposure (attractants, or emulsions that increase time
in the water column) are likely to increase toxicity of the a.i.
Only a limited amount of data were available to evaluate toxicity of the degradates of carbaryl,
carbofuran, and methomyl. Monitoring data indicate that 1-napthol, a degradate of carbaryl, and
3-hy doxy carbofuran, a degradate of carbofuran, occur in measurable quantities in the
environment. Based on the monitoring data, 1-napthol co-occurs with carbaryl in both water
column and sediment. Toxicity information presented in the BE indicates that 1-napthol is
approximately an order of magnitude less toxic than parent carbaryl for aquatic invertebrates.
No data were given in the BE regarding the toxicity of 1-napthol to fish. We did locate one open
literature study evaluating the effects of 1-napthol on fish, which indicated it was 2-5 times more
toxic than the parent carbaryl (Shea and Berry 1983). No toxicity data were located for 3-
hydroxycarbofuran or 3-ketocarbofuran, but based on structural similarities, the toxicity is likely
DRAFT 403
-------
DRAFT
similar to the parent carbofuran. Based on fate data presented in the BEs, 3-hydroxycarbofuran
and 3-ketocarbofuran appear to occur in soil rather than in the water column. Both are shown as
being non-detectable to representing 2-5% of applied parent material. Together, they could
represent 6-7% of applied, assuming they were measured in the same test. Neither the BE nor
the Science Chapter specify if that is the case.
Overall, while NMFS cannot quantify the increased toxicity of exposure to other stressors of the
action such as additional a.i.s, degradates, and adjuvants, the existing body of information
indicates these compounds are likely to cause and/or exacerbate adverse effects on salmonids
and their habitat caused by carbaryl, carbofuran, and methomyl.
We did not receive a complete list of the currently registered formulations containing carbaryl,
carbofuran, and methomyl from EPA. Thus, we cannot make any definitive conclusions for
every stressor of the action. However, we did evaluate the exposure and response to a commonly
used surfactant/adjuvant mixed with, or found in pesticide formulations. We reasoned if the data
support adverse effects from this one of more than 4,000 substances, then other unidentified inert
ingredients could also be toxic and may pose a significant risk to salmonids and their habitat.
We selected NP polyethoxylates and NP because of their widespread use in pesticide
formulations and abundance of information regarding environmental concentrations and adverse
effects to salmonids and their prey. The data indicated that these surfactants can kill outright,
disrupt endocrine systems, particularly reproductive physiology, and bioaccumulate in benthic
invertebrates from expected concentrations in the environment (Arsenault, Fairchild et al. 2004;
Madsen, Skovbolling et al. 2004; Jardine, MacLatchy et al. 2005; Luo, Ban et al. 2005;
McCormick, O'Dea et al. 2005; Segner 2005; Hutchinson, Ankley et al. 2006; Lerner, Bjornsson
et al. 2007; Lerner, Bjornsson et al. 2007). Importantly, we found studies that linked Atlantic
salmon population crashes in Canada to use of NP in insecticide formulations. However, the
BEs did not provide any information as to the prevalence of this material in formulations of the
three OP insecticides that pertain to this consultation. We did receive confirmation from the
technical registrant of methomyl that no methomyl containing formulation they make contains
NP. Significant uncertainty surrounds the number and type of compounds, as well as the toxicity
of these other materials used in pesticide formulations. As a result, we must caveat our
DRAFT 404
-------
DRAFT
conclusions regarding population-level responses with the uncertainty that the actual risk posed
to listed salmonids and their habitat is likely greater when all ingredients are taken into account.
H. Exposure to other pesticides present in the action area can act in combination with
carbaryl carbofuran, and methomyl to increase effects to salmonids and their habitat.
The available toxicity and exposure data support the hypothesis. Other carbamates and OPs
found in the action area likely result in additive or synergistic effects to exposed salmonids and
aquatic invertebrates. The magnitude of effects will depend on the duration and concentrations
of exposure.
/. Exposure to elevated temperatures can enhance the toxicity of the stressors of the action.
We found no consistent correlation with elevated temperature and toxicity of the three TV-methyl
carbamates. However, salmonids are coldwater species, and exposure to elevated temperatures
increases physiological stress, thus making them more susceptible to other stressors.
Additionally, other AChE inhibitors such as OPs are more toxic to salmonids and aquatic
invertebrates exposed to elevated temperatures, so where elevated temperatures co-occur with
OPs and carbamates the effect to aquatic life is likely greater than at lower temperatures. Many
salmonid populations reside in watersheds which have been listed by the four western states as
impaired due to temperature exceedances. We therefore discuss qualitatively temperature
impacts on salmonids population responses to the stressors of the action.
/. Exposure to specific pH ranges can affect the toxicity of the stressors of the action.
Some data indicate acute toxicity of carbaryl and carbofuran increases as pH increased based on
the available freshwater fish assays (Mayer and Ellersieck 1986). Other data indicated that
toxicity was reduced in pH greater than 9 due to rapid hydrolysis i.e., a half-life of 30 minutes in
pond mesocosms. For methomyl, pH seems to have less of an influence on hydrolysis rates.
Within the Pacific Northwest and California pH varies seasonally and typically may range from
6 - >9. We expect that salmonids exposed to both pH ranges at or near physiological tolerance
limits and the three insecticides concurrently may die at relatively lower concentrations
compared to salmonids exposed in laboratory assays. We therefore discuss qualitatively pH
impacts on salmonids population responses to the stressors of the action.
DRAFT 405
-------
DRAFT
Effects to Salmonid Populations from the Proposed Action
Here we translate individual fitness consequences to potential population-level effects using both
quantitative and qualitative methods. We quantitatively translate reduced survival of juveniles
based on four-day exposures to four populations of salmonids including ocean-type Chinook,
stream-type Chinook, coho, and sockeye salmon. We employ a life history population model
that incorporates changes in first-year juvenile survival rates and then translates them into
predicted changes in the modeled population's intrinsic rate of growth, i.e., lambda (Appendix
1). We discuss the percent change in lambda in the context of expected concentrations of the
three insecticides in salmonid habitats. We focus on the concentrations at which a significant
departure occurs from the unexposed population and compare them to expected environmental
concentrations described in the Exposure Analysis. We also discuss in general terms the
likelihood of exposure to the range of pesticide concentrations that occur in salmonid habitats.
In a second modeling exercise, we translate reductions in growth of juvenile salmon from AChE
inhibition and from reduced prey abundances to potential population impacts using individual-
based growth and life history population models (Appendix 1). These two endpoints (AChE
inhibition and reduced prey abundance) are combined in the model to evaluate population-level
effects due to reductions in first year survival of juveniles (Appendix 1). Similar to the survival
models, percent change in lambda is the output. We discuss the significance of population
changes in the context of departures from normal variability of the unexposed population and
expected environmental concentrations. We conclude this section on population-level effects
with a discussion of population-level responses to other affected salmonid endpoints that were
not modeled. These include effects from other stressors of the proposed action, mixture effects,
and effects to behaviors from impaired olfaction and AChE inhibition such as swimming
behaviors.
Salmonid Population Models
We selected four generalized life history strategies to model (Appendix 1). We ran general life
history matrix models for coho salmon (Oncorhynchus kisutch), sockeye salmon (O. nerka) and
ocean-type and stream-type Chinook salmon (O. tshawytscha). We did not construct a steelhead
(O. mykiss) life history model due to the lack of demographic information. Chum salmon (O.
DRAFT 406
-------
DRAFT
ketd) were omitted from the growth model exercise because they migrate to marine systems soon
after emerging from the gravel and the model assesses early life stage growth effects over a
minimum of 140 d in freshwater systems. The basic salmonid life history we modeled consisted
of hatching and rearing in freshwater, smoltification in estuaries, migration to the ocean,
maturation at sea, and returning to the natal freshwater stream for spawning followed shortly by
death. For specific information on how we constructed the models see Appendix 1.
Effects to salmonid populations from death of sub-yearling juveniles
An acute toxicity model was constructed that estimated the population-level impacts of
subyearling juvenile (referred to as juveniles within this section) mortality resulting from
exposure to concentrations of carbaryl, carbofuran, and methomyl. These models excluded
sublethal and indirect effects of the pesticide exposures and focused on the population-level
outcomes resulting from an annual 4 d exposure of all juveniles in the population to carbaryl,
carbofuran, and methomyl from single exposures. Death of juveniles was implemented as a
change in first-year survival rate for each of the salmon lifehistory strategies modeled. We also
addressed mixture toxicity of the three insecticides in the model using exposure concentrations
derived from EPA modeling estimates and from NMFS' modeling estimates of off-channel
habitats. Model output is displayed in Tables 31-34.
The percent changes in lambdas increased as concentrations of the three carbamates increased.
Increases in direct mortality during the first year of life produced large impacts on the
population growth rates for all the life history strategies. Model results for stream-type Chinook
salmon showed significant impacts at lower concentrations than the other modeled populations.
This result is primarily due to the size of the standard deviation of the unexposed population.
Percent changes in lambda were deemed significant if they were outside of one standard
deviation from the unexposed population. The relative sensitivity of the life history models
producing the greatest to the least changes in population growth rate for equivalent impact on
survival rates was coho salmon, ocean-type Chinook salmon, stream-type Chinook salmon, and
sockeye salmon. We note that the choice of LC50 is an important driver for these results.
Therefore, an LC50 above or below the ones used here will result in a different dose-response.
We selected the lowest reported salmonid LC50 from the available information to ensure that
risk is not underestimated. However, if the actual environmental 96 h LC50 is lower, then the
DRAFT 407
-------
DRAFT
model will under-predict mortality. If the actual environmental acute LC50 is higher, then the
model will over-predict mortality.
These results indicate that exposure of salmonid populations to carbaryl, carbofuran, and
methomyl for four days at the reported LCSOs would have severe consequences to the
population's growth rate. If exposure occurred every year for each new cohort, population
abundance would decline and recovery efforts would be slowed. For those natural populations
with current lambdas of less than one, risk of extinction would increase substantially, especially
if several successive generations were exposed. For each of the combinations of species and
insecticide, we denoted the relative concentration at which the percent change in lambda is
deemed significantly different from the unexposed populations e.g., a 9.1% change in lambda is
estimated at 190 |j,g/L carbaryl for ocean-type Chinook salmon. We consider these
concentrations at four day exposure durations as population effect thresholds when above the
threshold a more pronounced reduction in lambda is predicted and below the threshold if is
uncertain whether the change in lambda is from the insecticide. The population effect thresholds
calculated here assume all the juveniles in the population are exposed to the insecticide.
Carbaryl's thresholds for the four species ranged from 129-190 |J,g/L, carbofuran's from 89-126
Hg/L, and methomyl's from 287-431 [ig/L. These results can be compared to expected
concentrations shown in Figure 39, Figure 40, and Figure 41. All of the thresholds overlap with
expected concentrations for the three insecticides. The overlap primarily occurs with NMFS
modeled estimates for off-channel habitats.
When we compare the population threshold concentrations to expected levels in salmonid
habitats described in the exposure section, it is likely that some individuals within a population
will be exposed during their freshwater juvenile life stage, particularly those juveniles exposed
while using off-channel habitats. Additionally, four day averages from GENEEC runs, but not
from PRZM-EXAMS, exceed these threshold concentrations. It is uncertain how appropriate
EPA's model estimates are for salmonid habitats, but taken at face value juveniles exposed to
GENEEC-derived concentrations would die. Population-level effects are expected from these
exposures if the majority of the individuals that comprise the populations are exposed during
their freshwater residency. The likelihood of population effects from death of juveniles increases
DRAFT 408
-------
DRAFT
for those populations that spend longer periods in freshwaters such as steelhead, stream-type
Chinook, and coho salmon. We also expect additional acute mortalities from juveniles that are
exposed to 24(c) carbaryl applications in Washington estuaries, although we do not know how
many individuals are exposed each year. For those populations with lambdas greater than one,
reductions in lambda from death of juveniles can also lead to consequences to abundance and
productivity. Attainment of recovery goals would likely not be met for populations with reduced
lambdas. Many of the populations that are categorized as core populations or are important to
individual strata, have lambdas just above one and are essential to survival and recovery goals.
Slight changes in lambda, even as small as 3-4%, would result in reduced abundances and
increased time to meet population recovery goals.
DRAFT 409
-------
DRAFT
Table 74. Modeled output for Ocean-type Chinook salmon exposed to 4 d exposures of carbaryl,
carbofuran, and methomyl reporting the impacted factors of survival as percent dead, lambda and
standard deviation, and percent change in lambda compared to an unexposed population.
Carbaryl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Carbofuran
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Methomyl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
0
^g/L
0
1.09
(0.1)
NA
50
^g/L
0
1.08
(0.1)
NS
100
^g/L
3
1.08
(0.1)
NS (-1)
200
^g/L
31
0.98
(0.09)
-10
250
^g/L
50
0.89
(0.08)
-18
350
^g/L
77
0.71
(0.06)
-34
500
^g/L
93
.53
(0.05)
-52
750
^g/L
98
0.36
(0.03)
-67
-9.1 %~190|ag/L
0
^g/L
0
1.09
(0.10)
NA
50
^g/L
1
1.09
(0.10)
NS
100
^g/L
14
1.04
(0.10)
NS (-4)
150
^g/L
42
0.93
(0.08)
-15
164
^g/L
50
0.89
(0.08)
-18
200
^g/L
67
0.78
(0.07)
-27
250
^g/L
82
0.67
(0.06)
-39
350
^g/L
94
0.5
(0.04)
-54
-9.1 %~126|ag/L
0
^g/L
0
1.09
(0.10)
NA
250
^g/L
5
1.07
(0.10)
NS(-1)
400
^g/L
23
1.08
(0.9)
NS(-7)
500
^g/L
40
0.94
(0.08)
-14
560
^g/L
50
0.89
(0.08)
-18
700
^g/L
69
0.78
(0.07)
-29
900
^g/L
85
0.64
(0.06)
-41
1,100
^g/L
92
0.54
(0.05)
-51
-9.1 % ~ 431 (ig/L
NA denotes non applicable; NS denotes values less than one standard deviation of lambda expressed as
the percent of lambda. (Calculated value, omitted when less than or equal to one)
DRAFT
410
-------
DRAFT
Table 75. Modeled output for Stream-type Chinook salmon exposed to 4 d exposures of carbaryl,
carbofuran, and methomyl reporting the impacted factors of survival as percent dead, lambda and
standard deviation, and percent change in lambda compared to an unexposed population
Carbaryl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Carbofuran
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Methomyl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
0
^g/L
0
1.00
(0.03)
NA
50
^g/L
0
1.00
(0.03)
NS
100
^g/L
3
0.99
(0.03)
NS (-1)
200
^g/L
31
0.91
(0.03)
-9
250
^g/L
50
0.84
(0.03)
-16
350
^g/L
77
0.69
(0.02)
-31
500
^g/L
93
0.53
(0.02)
-47
750
^g/L
98
0.37
(0.01)
-63
-3.1 %~129|ag/L
0
^g/L
0
1.0
(0.03)
NA
50
^g/L
1
1.0
(0.03)
NS
100
^g/L
14
0.96
(0.03)
-4
150
^g/L
42
0.87
(0.03)
-13
164
^g/L
50
0.84
(0.03)
-16
200
^g/L
67
0.76
(0.02)
-24
250
^g/L
82
0.65
(0.02)
-35
350
^g/L
94
0.5
(0.01)
-50
-3.1 %~89|ag/L
0
^g/L
0
1.0
(0.03)
NA
250
^g/L
5
0.99
(0.03)
NS(-1)
400
^g/L
23
0.94
(0.03)
-6
500
^g/L
40
0.88
(0.03)
-12
560
^g/L
50
0.84
(0.03)
-16
700
^g/L
69
0.75
(0.02)
-25
900
^g/L
85
0.63
(0.02)
-37
1,100
^g/L
92
0.54
(0.02)
-46
-3.1 % ~ 287 |ag/L
NA denotes non applicable; NS denotes values less than one standard deviation of lambda expressed as
the percent of lambda. (Calculated value, omitted when less than or equal to one)
DRAFT
411
-------
DRAFT
Table 76. Modeled output for Coho salmon exposed to 4 d exposures of carbaryl, carbofuran, and
methomyl reporting the impacted factors of survival as percent dead, lambda and standard
deviation, and percent change in lambda compared to an unexposed population.
Carbaryl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Carbofuran
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Methomyl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
0
^g/L
0
1.03
(0.05)
NA
50
^g/L
0
1.03
(0.05)
NS
100
^g/L
3
1.02
(0.05)
NS(-1)
200
^g/L
31
0.91
(0.05)
-12
250
^g/L
50
0.82
(0.04)
-21
350
^g/L
77
0.69
(0.03)
-39
500
^g/L
93
0.42
(0.02)
-58
750
^g/L
98
0.27
(0.01)
-74
-5.3% -144 (og/L
0
^g/L
0
1.03
(0.05)
NA
50
^g/L
1
1.02
(0.05)
NS(-1)
100
^g/L
14
0.98
(0.05)
NS)-5
150
^g/L
42
0.86
(0.05)
-17
164
^g/L
50
0.82
(0.04)
-21
200
^g/L
67
0.71
(0.04)
-31
250
^g/L
82
0.58
(0.03)
-44
350
^g/L
94
0.40
(0.02)
-61
-5.3% -98 (4,g/L
0
^g/L
0
1.03
(0.05)
NA
250
^g/L
5
1.01
(0.05)
NS(-2)
400
^g/L
23
0.94
(0.05)
-8
500
^g/L
40
0.87
(0.05)
-16
560
^g/L
50
0.82
(0.04)
-21
700
^g/L
69
0.69
(0.04)
-32
900
^g/L
85
0.55
(0.03)
-47
1,100
^g/L
92
0.44
(0.02)
-57
-5.3% -338 ng/L
NA denotes non applicable; NS denotes values less than one standard deviation of lambda expressed as
the percent of lambda. (Calculated value, omitted when less than or equal to one)
DRAFT
412
-------
DRAFT
Table 77. Modeled output for Sockeye salmon exposed to 4 d exposures of carbaryl, carbofuran,
and methomyl reporting the impacted factors of survival as percent dead, lambda and standard
deviation, and percent change in lambda compared to an unexposed population
Carbaryl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Carbofuran
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
Methomyl
% dead
Lambda
(STD)
% change in lambda
Threshold for
significant change
in lambda
0
^g/L
0
1.01
(0.06)
NA
50
^g/L
0
1.01
(0.06)
NS
100
^g/L
3
1.00
(0.06)
NS(-1)
200
^g/L
31
0.92
(0.05)
-8
250
^g/L
50
0.86
(0.05)
-15
350
^g/L
77
0.71
(0.04)
-30
500
^g/L
93
0.55
(0.03)
-46
750
^g/L
98
0.40
(0.02)
-61
-5.6% -169 (og/L
0
^g/L
0
1.01
(0.06)
NA
50
^g/L
1
1.01
(0.06)
NS
100
^g/L
14
0.97
(0.05)
NS(-4)
150
^g/L
42
0.89
(0.05)
-12
164
^g/L
50
0.86
(0.05)
-15
200
^g/L
67
0.78
(0.04)
-23
250
^g/L
82
0.67
(0.04)
-34
350
^g/L
94
0.52
(0.03)
-48
-5.6% -114 |ag/L
0
^g/L
0
1.01
(0.06)
NA
250
^g/L
5
1.00
(0.06)
NS(-1)
400
^g/L
23
0.95
(0.05)
-6
500
^g/L
40
0.89
(0.05)
-11
560
^g/L
50
0.86
(0.05)
-15
700
^g/L
69
0.76
(0.04)
-24
900
^g/L
85
0.65
(0.04)
-36
1,100
^g/L
92
0.56
(0.03)
-45
-5.3% -391 |ag/L
NA denotes non applicable; NS denotes values less than one standard deviation of lambda expressed as
the percent of lambda. (Calculated value, omitted when less than or equal to one)
The population exercises discussed thus far focused on the effects from exposures to a single
annual application of an insecticide; however, we know that these insecticides likely occur
together in environmental mixtures and are frequently applied multiple times. To address the
potential population-level effects to environmental mixtures of the three insecticides, we ran the
same models with a calculated percent mortality predicted using dose-addition. We applied the
DRAFT 413
-------
DRAFT
same additivity model described in the previous Mixtures section to predict the cumulative
percentage of death in an exposed population. Table 78 shows the model's prediction from three
scenarios. We did not use the data from ambient monitoring programs to devise a scenario
because the programs were not designed to capture peak concentrations from drift or runoff into
juvenile salmonid rearing areas. We did run a few modeling runs with median values of the
three insecticides taken from ambient monitoring data. The results showed no statistically
significant reductions in lambda for the four populations. In scenario 1, 5% of exposed
individuals are expected to die following 4 d exposures to the estimated concentrations from
PRZM-EXAMS 24 h averages from EPA's BEs. For carbaryl, we selected a 2 Ibs/acre applied
aerially with four applications to apples in Oregon which resulted in a concentration of 19 ng/L.
For carbofuran, we selected a 2 Ib/acre ground application to artichokes in California which
resulted in a concentration of 35 ng/L. For methomyl, we selected 0.9 Ibs/acre applied ten times
to lettuce in California which resulted in a concentration of 88 ng/L. If these events were to
occur in a watershed, 5% of the individuals are expected to die which would not lead to a
population-level effect i.e., reduction in lambda, based on juvenile mortality.
In contrast, the other two scenarios showed substantial and severe percent reductions in lambda
for each of the modeled populations; reductions in lambda ranged from 27-38%. In scenario 2,
we selected 90 d GENEEC-modeled scenarios for corn from Table 50. Concentrations of the
three insecticides were as follows: carbaryl = 229 |J,g/L, carbofuran = 53 |J,g/L, and methomyl =
49 |J,g/L. The toxicity is largely a result of the carbaryl concentration, as it was close to the 96 h
LC50 of 250 ng/L. In scenario 3, concentrations from drift into off-channel habitats (0.5 m
deep) were calculated using the AgDrift model and included a 100 ft no-spray buffer for
methomyl. For carbaryl, 5 Ib/acre applied once aerially with a fine-medium spray droplet size
resulted in a concentration of 335|j,g/L. For carbofuran, 1 Ib/acre applied once also with a fine-
medium spray droplet size resulted in a concentration of 67 [ig/L. For methomyl, 0.9 Ibs/acre
applied once resulted in a concentration of 17. l|j,g/L. Cumulatively, these three concentrations
would result in 89% mortality of exposed juveniles which reduce lambdas of the four salmonid
populations by 41-52%, a substantial effect.
DRAFT 414
-------
DRAFT
The likelihood of scenarios 2 and 3 occurring is difficult to accurately predict due to the lack of
detailed information on watershed characteristics, salmonid presence, the numbers of salmonids
exposed, the duration of exposure, and the climatic variables leading to drift. Scenarios 2 and 3
may represent infrequent events, but if a substantial part of a population of listed salmonids is
exposed to these mixtures, a severe reduction in a population's abundance is expected.
Table 78. Modeled output for Ocean-type Chinook, Stream-type Chinook, Sockeye, and Coho salmon
exposed to 4 d exposures of carbaryl-, carbofuran-, and methomyl-containing mixtures. The table
denotes the impacted factors of survival as percent dead, lambda and standard deviation, and percent
change in lambda compared to an unexposed population
Scenario 1 :
PRZM-EXAMS
24- h averages
% dead
Lambda
(STD)
% change in lambda
Scenario 2:
GENEEC
90-d averages
% dead
Lambda
(STD)
% change in lambda
Scenario 3:
Offchannel habitats
0.5 m deep
% dead
Lambda
(STD)
% change in lambda
Ocean-type Chinook
5
1.07
(0.10)
NS(-1)
Ocean-type Chinook
74
0.74
(0.07)
-32
Ocean-type Chinook
89
0.59
(0.05)
-46
Stream-type Chinook
5
0.99
(0.03)
NS(-1)
Stream-type Chinook
74
0.72
(0.02)
-28
Stream-type Chinook
89
0.58
(0.02)
-42
Sockeye
5
1.00
(0.06)
NS(-1)
Sockeye
74
0.74
(0.04)
-27
Sockeye
89
0.60
(0.03)
-41
Coho
5
1.01
(0.05)
NS(-2)
Coho
74
0.66
(0.04)
-36
Coho
89
0.49
(0.03)
-52
NA denotes non applicable; NS denotes values less than one standard deviation of lambda expressed as
the percent of lambda. (Calculated value, omitted when less than or equal to one)
DRAFT
415
-------
DRAFT
Effects to salmonid populations from reduced size of juveniles due to impaired feeding and
reduced abundance of aquatic prey
To evaluate the potential for adverse effects to juvenile growth resulting from carbaryl,
carbofuran, and methomyl on Pacific salmonid populations, we developed a model (Appendix
1). The model links AChE inhibition, feeding behavior, prey availability, and somatic growth of
individual salmon to the productivity of salmon populations expressed as a percent change in
lambda (a population's intrinsic rate of growth). The model scenarios (single insecticide or
multiple pulses of single insecticides) assume annual exposure of all the subyearling juveniles in
the population and their prey to the insecticide. Similar to the survival model, we developed the
growth model for four populations of salmonids: ocean-type Chinook, stream-type Chinook,
sockeye, and coho salmon.
We integrated two avenues of effect to juvenile salmonids' growth from exposure to the three TV-
methyl carbamates. The first avenue is a result of AChE inhibition on the feeding success and
subsequent effects to growth of juvenile salmonids. Study results with juvenile salmonids show
that feeding success is reduced following exposures to AChE inhibitors (Sandahl, Baldwin et al.
2005). The second avenue the model addresses is the potential for reductions in juvenile growth
due to reduction in available prey. Salmon are often found to be food limited in freshwater
aquatic habitats, suggesting that a reduction in prey due to insecticide exposure may further
stress salmon and lead to reduced growth rates. Field mesocosm data support this assertion,
showing reduced growth of juvenile fish following exposure to the AChE inhibitor, chlorpyrifos
(Brazner and Kline 1990). Furthermore, based on our review of the sensitivities of aquatic
invertebrates to the three insecticides, we expect reductions in densities and altered composition
of the salmonid prey communities.
Reductions in aquatic prey are included in the model because of the high relative toxicity of
pesticides to salmonid prey and the extended duration of effects on prey communities. Juvenile
salmonids are largely opportunistic, feeding on a diverse community of aquatic and terrestrial
invertebrate taxa that are entrained in the water column or on the surface (Higgs, Macdonald et
al. 1995). As a group, these invertebrates are among the more sensitive taxa for which there is
toxicity data, but within this group, there is a wide range of sensitivities (Table 64, Table 65).
DRAFT 416
-------
DRAFT
The three insecticides are highly toxic to aquatic macroinvertebrates; and concentrations that are
not expected to kill salmonids are often lethal for their invertebrate prey (e.g., for carbaryl, range
of mean LCSOs for salmonids = 250-3,000 |J,g/L, vs. range of geometric mean ECSOs for water
fleas = 3-120 ng/L). In particular, prey items that are preferred by small juvenile salmonids
(including midge larvae, water fleas, mayflies, caddisflies, and stoneflies) are among the most
sensitive aquatic macroinvertebrates. In addition, effects on the prey community can persist for
extended periods of time (weeks, months, years), resulting in effects on fish feeding and growth
long after an exposure has ended (Ward, Arthington et al. 1995; Van den Brink, van
Wijngaarden et al. 1996; Liess and Schulz 1999; Colville, Jones et al. 2008).
Selection of aquatic invertebrate toxicity values to represent salmonid prey items
The model requires for each insecticide an EC50 (defined as a 50% reduction in the biomass of
salmonid prey items) and a corresponding slope (Appendix 1). The term "EC50" will be used in
this section to describe short-term survival (death and immobility) data for aquatic invertebrates.
To determine what levels of the three pesticides reduce aquatic invertebrate numbers, we
reviewed the available field and laboratory studies. We found robust data for carbaryl,
carbofuran, and methomyl with respect to laboratory acute toxicity tests that measured survival
at 24-, 48-, 72-, and 96 h with an array of aquatic invertebrates. We did not locate a field study
that measured aquatic community response to a range of concentrations of the three insecticides.
Therefore, we did not select concentration data from field experiments as we did in NMFS' 2008
Opinion on the registration of chlorpyrifos, diazinon, and malathion (NMFS 2008).
To determine a single effect concentration to use in the model analyses, a search was completed
using the EPA's ECOTOX database for each pesticide (http://cfpub.epa.gov/ecotox/). Several
criteria were used to determine which reported effect concentrations were included in the final
analysis. The data included were from studies on taxa that are known to be salmonid prey (or are
functionally similar to salmonid prey); these include a diverse group aquatic insects and worms
and fresh and saltwater crustaceans. Studies with exposures of at least 24 h and not more than 96
h were included. Studies examining shorter and longer exposure times are known to affect
invertebrates (Peterson, Jepson et al. 2001), but these were excluded so that estimated ECSOs
would be comparable. Studies reporting survival ECSOs in which mortality or immobilization
was the recorded endpoint were included. Data derived for sublethal endpoints (e.g., growth or
DRAFT 417
-------
DRAFT
reproduction) were not included. If specific data were represented more than once in the
ECOTOX output, duplicates were eliminated. Data from recent peer-reviewed studies that report
survival ECSOs were also included (Norberg-King, Durhan et al. 1991; Takahashi and Hanazato
2007). Next, we calculated the geometric mean when more than one survival EC50 was reported
for a species.
Probability distributions of aquatic prey survival toxicity values (ECSOs)
We plotted the survival EC50 data for each of the three insecticides using cumulative probability
distributions. We also plotted the data based on all test results for the species without taking the
geometric means. From the distributions of the data, a single effect concentration and slope were
derived to best represent the diverse community of prey available in juvenile salmonid
freshwater and estuarine habitats. The distributions of individual ECSOs and the geometric
means of ECSOs by taxa were analyzed to estimate the 50th, 10th, and 5th percentiles. Figure 43
shows the distributions of geometric means of ECSOs by taxa. Specifically, for each pesticide, a
probability plot was used to graph the EC50 concentrations normalized to a normal probability
distribution. For each plot, the X axis is scaled in probability (between zero and 100%) and
shows the percentage of entire data whose value is less than the data point. The Y axis displays
the range of the data on a log scale. The results of a linear regression of the log-transformed
concentrations are shown and highlight the lognormal distribution of the data Figure 43. In the
regression equation, the normsinv function returns the inverse of the standard normal cumulative
distribution. The standard normal distribution has a mean of zero and a standard deviation of
one. For example, given a percentile value of 50 (i.e., a probability of 0.5), normsinv(50) returns
a value of zero. The plots and regressions were performed using KaleidaGraph 4.03 (Synergy
Software).
DRAFT 418
-------
DRAFT
B.
ios-3
carbofuran
(conce
21
r2 = 0.98
log(concentration) = 1.92 + 1.43 * normsinv(percentile)
= 21
1000^
0.001
.1 1 5 10 20 30 50 70 80 90 95 99 99.9
percentile
10'-,
1000-
10-
c.
methomyl
log(concentration) = 2.18 + 0.67 * normsinv(percentile)
n = 15
i2 = 0.95
.1 1 510 20 30 50 70 80 90 95 99 99.9
percentile
10-
.1 1 510 20 30 50 70 80 90 95 99 99.9
percentile
Figure 43. Probability plot for each pesticide showing the distribution of the geometric means of
the ECSOs for each aquatic invertebrate species. The straight line shows the result of a linear
regression. In the regression equation, the normsinv() function returns the inverse of the standard
normal cumulative distribution. See text for more details. A) Plot of carbaryl ECSOs. B) Plot of
carbofuran ECSOs. C) Plot of methomyl ECSOs.
In Table 79, concentrations are reported for each insecticide associated with either the 50th, 10th,
or 5th percentiles derived from probability distribution plots with all study results (plots not
shown) or with species geometric means. Percentiles from the species geometric means show
DRAFT
419
-------
DRAFT
higher concentrations compared to the "all studies" plots. This is likely a reflection that when
individual study results are considered separately, the species with the greater number of EC50
study results is found at the lower end of the distribution.
Table 79. Carbaryl, carbofuran, and methomyl survival EC50 concentrations at 50th, 10th, and 5th
percentiles from probability distribution plots.
Concentration of EC50 at each percentile (|j.g/L)
50% 10% 5%
All studies probability distribution plot
Carbaryl
Carbofuran
Methomyl
45.23
58.95
128.9
2.29
0.94
12.93
0.98
0.29
6.74
Species geometric means probability distribution plot
Carbaryl
Carbofuran
Methomyl
69.53
89.95
150.76
4.33
1.22
20.74
1.97
0.37
11.82
We selected the 10th percentile from each of the pesticide plots to represent the survival EC50 for
salmonid prey. The associated 10th percentile for carbaryl (4.33 ng/L), carbofuran (1.22 ng/L),
and methomyl ( 20.74 ng/L) was used as input for the population growth model exercises. The
10th percentile is a reasonable selection because the data included in the meta-analysis were
limited to concentrations that caused mortality or immobilization within a short period of time
(1-4 days). A growing number of studies on a variety of insecticides have reported that
concentrations well below LCSOs can cause delayed mortality or sublethal effects that may scale
up to affect aquatic invertebrate populations, especially in scenarios with multiple exposures
and/or other stressors. Evidence for ecologically significant sublethal or delayed effects to
aquatic invertebrates includes reduced growth rates (Schulz and Liess 2001; Forbes and Cold
2005), altered behavior (Johnson, Jepson et al. 2008), reduced emergence (Schulz and Liess
2001; Johnson, Jepson et al. 2008), reduced reproduction (Cold and Forbes 2004; Forbes and
Cold 2005), and reduced predator defenses (Sakamoto, Chang et al. 2006; Johnson, Jepson et al.
2008).
Additionally, the available toxicity data - and therefore the data included for these analyses- are
from studies using taxa hearty enough to survive laboratory conditions. Studies specifically
examining salmonid prey that are more difficult to rear in the laboratory have documented
DRAFT
420
-------
DRAFT
relatively low survival EC50 values when exposed to current use insecticides (Johnson, Jepson et
al. 2008), including carbaryl (Peterson, Jepson et al. 2001; Johnson, Jepson et al. 2008)
The selection of the 10th percentile is a reasonable choice in keeping with general risk
management practices of protecting the aquatic community. The standard procedure the U.S.
EPA Office of Water uses in establishing Aquatic Life Criteria is to protect to the 5th percentile
of a species sensitivity distribution that includes all genera of aquatic species for which valid
data are available (EPA 1995). In the probabilistic risk assessment for carbofuran (EPA 2005),
OPP evaluated effects on the 5th, 50th, and 95th percentiles of separate species sensitivity
distributions for fish and invertebrates, although it was not clear in the document which was the
preferred percentile for risk management.
Modeling availability of unaffected prey
Reductions in benthic invertebrate densities can lead to long-term reductions in prey availability
and reductions in fish growth (Davies and Cook 1993). That said, prey densities are not usually
reduced to zero (Wallace, Lugthart et al. 1989). Therefore, it is assumed that regardless of the
exposure scenario, prey abundance would not drop below a specific "floor" of prey availability.
This floor is included in the model to reflect an assumption that a minimal yet constant terrestrial
subsidy of prey and/or an aquatic community with tolerant individuals would be available as
prey, regardless of pesticide exposure and in addition to the constant recovery rate (see below).
Therefore, even in extreme exposure scenarios, some prey will be available, as determined by the
value assigned to the floor; in some highly degraded systems this may or may not be the case.
No studies have quantified this floor for the purpose of estimating prey availability, but several
studies have documented reductions in overall benthic insect densities of 75-98% (Wallace et al.
1989, Anderson et al. 2003, Anderson et al. 2006). Because benthic densities are typically
correlated with drift densities (Hildebrand 1974, Waters and Hokenstrom 1980), these reductions
likely result in similar reductions of prey. Therefore, assuming there is also some constant rate
of terrestrial invertebrate subsidy in addition to a residual aquatic community, a floor of 0.20, or
20% offish ration, is reasonable. The model does not include any additional impacts to fish via
DRAFT 421
-------
DRAFT
dietary exposure from contaminated prey, or any potential synergistic or additive effects to the
aquatic invertebrates that may be result from multiple stressors (Schulz and Liess 2001).
Modeling spikes in invertebrate drift following insecticide exposure
"Catastrophic drift" of invertebrates, due to acute mortality and/or emigration of benthic prey
into the water column is frequently observed following exposure to insecticides (Davies and
Cook 1993; Schulz and Liess 2001; Schulz 2004). Drift rates within hours of exposure can be
more than 10,000 times the natural background drift (Cuffney 1984), and fish have been found to
exploit this by feeding beyond satiation (Haines 1981; Davies and Cook 1993). The duration and
magnitude of the spike in drift of prey is dependent in part on the physical properties and dose of
the pesticide; however, the spike is generally ephemeral and returns to natural, background levels
within hours to days (Haines 1981; Kreutzweiser and Sibley 1991). Likewise, the magnitude of
the spike is dependent in part on the benthic density of prey; the spike in drift from communities
that have been reduced by previous exposures is smaller than the spike from previously
undisturbed communities (Cuffney 1984; Wallace, Huryn et al. 1991). To reflect this temporary
increase in prey availability, the model includes a one-day prey spike for the day following an
exposure (Appendix 1). The model also accounts for this short-term increase in prey availability
by allowing fish to feed at a maximum rate of 1.5 times their normal, optimal ration.
Modeling recovery ofsalmonid prey
We selected a 1% recovery in prey biomass per day. Reports of recovery of invertebrate prey
populations, once pesticide exposure has ended, range from within days to more than a year
(Cuffney 1984; Kreutzweiser and Sibley 1991; Pusey, Arthington et al. 1994; Ward, Arthington
et al. 1995; Van den Brink, van Wijngaarden et al. 1996; Liess and Schulz 1999; Colville, Jones
et al. 2008). The dynamics of recovery are complicated by several factors, including the details
of the pesticide exposure(s) as well as habitat and landscape conditions (Liess and Schulz 1999)
(Van den Brink, Baveco et al. 2007). In watersheds with undisturbed upstream habitats,
recovery can be rapid due to a healthy source of invertebrates that can immigrate via drift and/or
aerial colonization (for adult insects) (Heckmann and Friberg 2005). However, in watersheds
dominated by agricultural or urban land uses, healthy upstream or nearby habitats may be limited
DRAFT 422
-------
DRAFT
and consequently, recolonization by salmonid prey is likely reduced (Liess and Von der Ohe
2005; Schriever, Ball et al. 2007). Additionally, many large, high-quality prey take a year or
more to develop (Merritt and Cummins 1995) indicating that recovery of biomass (as compared
to prey density) is likely a limiting factor (Cuffney 1984). Recovery to pre-disturbance levels is
unlikely in aquatic habitats where invertebrate abundances are repeatedly reduced by stressors.
We consider a 1% (control prey abundance per day) recovery rate as ecologically realistic to
represent recolonization by invertebrates in salmonid habitats (Ward, Arthington et al. 1995; Van
den Brink, van Wijngaarden et al. 1996; Colville, Jones et al. 2008).
Growth model results
Exposure to single insecticides for 4-, 21-, and 60 day exposure durations
Population model outputs for the four salmon populations are summarized as dose-response
curves in Figures 39-42. As expected, greater reductions in population growth resulted from
longer exposures to the insecticides. The primary factor driving the magnitude of change in
lambda was the Prey Abundance parameter for each insecticide i.e., the 10th percentile survival
EC50 for salmonid prey. The AChE parameter for each of the insecticides was a secondary
factor compared to Prey Abundance. This is largely because the salmonid ECSOs for AChE were
much higher, typically by an order of magnitude, than the prey survival ECSOs.
Similar trends in effects were seen for each pesticide across all four life history strategies
modeled. This is apparent by the similar shape of the dose-response curves across species. The
curves plateau when there is no more reduction possible in the aquatic community i.e., the 20%
biomass of the aquatic invertebrate community is reached. Once that plateau is achieved, further
reductions in lambda are minimal with increasing concentrations. Clearly the most toxic of the
pesticides affected salmon populations at lower concentrations as is observed with carbaryl and
carbofuran where concentrations in the low |j,g/L range are sufficient to reduce populations'
growth rates compared to methomyl where 20 |j,g/L or greater are needed to reduce lambdas.
One factor that contributed to the similar responses observed was the use of the same surrogate
toxicity values for all four life history strategies. The stream-type Chinook salmon (Figure 45)
and sockeye salmon (Figure 47) models produced very similar results as measured as the final
output of percent change in population growth rate. The ocean-type Chinook salmon model
DRAFT 423
-------
DRAFT
output produced the next most extreme response; coho salmon output (Figure 46) showed the
greatest changes in lambda resulting from the pesticide exposures. When looking for similarities
in parameters to explain the ranking, no single life history parameter or characteristic, such as
lifespan, reproductive ages, age distribution, lambda and standard deviation, or first-year survival
show a pattern that matches this consistent output (Appendix 1). Combining these factors into
the transition matrix for each life history and conducting the sensitivity and elasticity analyses
revealed that changes in first-year survival produced the greatest changes in lambda. While
some life history characteristics may lead a population to be more vulnerable to an impact, the
culmination of age structure, survival and reproductive rates as a whole strongly influences the
population-level response.
ocean-type chinook
o -
8 -5
°
-10 -
.E -15 -
ra
c
eo
-20 -
-25 J
duration
o»4 d
d
o»60d
10 100
[carbaryl], ug/L
10
[carbofuran], ug/L
10
100
[methomyl], ug/L
Figure 44. Percent change in lambda for Ocean-type Chinook salmon following 4 d, 21 d, and 60 d
exposures to carbaryl, carbofuran, and methomyl. Open symbols denote a percent change in
lambda of less than one standard deviation from control population. Closed symbols represent a
percent change in lambda of more than one standard deviation from control population.
DRAFT
424
-------
DRAFT
stream-type Chinook
o -
o -5-
£ -10 -
.E -15 -
-20 -
-25 J
duration
o»4 d
n«21 d
»*60d
10
[carbaryl], ug/L
100
10
[carbofuran], |jg/L
10
100
[methomyl], ug/L
Figure 45. Percent change in lambda for Stream-type Chinook salmon following 4 d, 21 d, and 60
d exposures to carbaryl, carbofuran, and methomyl. Open symbols denote a percent change in
lambda of less than one standard deviation from control population. Closed symbols represent a
percent change in lambda of more than one standard deviation from control population.
coho
0-
oJ-10-
c -15-
-20-
-25 J
duration
o«4 d
n«21 d
o« 60 d
10
[carbaryl], ug/L
100
10
[carbofuran], ug/L
10
100
[methomyl], ug/L
Figure 46. Percent change in lambda for Coho salmon following 4 d, 21 d, and 60 d exposures to
carbaryl, carbofuran, and methomyl. Open symbols denote a percent change in lambda of less
than one standard deviation from control population. Closed symbols represent a percent change
in lambda of more than one standard deviation from control population.
DRAFT
425
-------
DRAFT
sockeye
o-
-5H
ra-10 -
.5 -15
-------
DRAFT
Populations exposed to carbaryl drift within off-channel habitats following four applications
would experience severe declines in population growth rate ranging from 15-19%. Interestingly,
a single application would result in notable reductions in the four modeled population's lambdas
as well, ranging from 8-11%. Similar to the carbaryl modeled scenario, methomyl applied to
sweet corn ten times (the label allows for 14 applications at 1 d intervals) resulted in significant
reductions in lambdas ranging from 6-8%. A single application did not result in a significant
reduction in lambda for any of the four populations.
Table 80. Multiple application scenarios for carbaryl and methomyl and predicted percent change
in lambdas for salmon populations
Crop examples
Application rate
Number of
applications/yr
Application
interval
Method of
application
No-application
Buffer
Off-channel
habitat
characteristics
Carbaryl
Almonds, chestnuts, pecans,
filberts, walnuts, pistachios
5 Ibs a.i./acre
4
14 days
Aerial (fine-medium droplet
distribution)
none
water depth = 0.5 m
Initial average concentration
335 |j,g/L; 24 h exposure
Methomyl
Sweet corn
0.45 Ibs a.i./acre
10
3 days
aerial (fine-medium droplet distribution)
100ft
water depth = 0.5 m
Initial average concentration 8.55 |j,g/L;
96 h exposure
% change in Lambda
Ocean-type
Chinook
Stream-type
Chinook
Sockeye
Coho
-1 9%
-1 5%
-16%
-1 8%
-8%
-6%
-7%
-8%
Population-level consequences from other affected salmonid assessment endpoints and other
stressors of the action
In this section we present the population-level consequences from individual effects that are not
amenable to population modeling. In most cases we lack the empirical data to conduct
population modeling for these endpoints. Thus, we use qualitative methods to infer population-
level responses. We focus on the population metrics of abundance and productivity. Both are
metrics used by NMFS to assess a population's viability and both can be compromised by the
chemicals assessed in this Opinion. Individual fitness consequences that reduce survival,
DRAFT
427
-------
DRAFT
growth, reproduction, or migration can lead to reduced salmonid population viability if sufficient
numbers of individuals comprising a population are affected, and are more pronounced when
affected over multiple generations. If the reductions in fitness result in reducing a population's
survival or recovery potential, then we consider whether the ESU or DPS is impacted (See
Integration and Synthesis section).
With the proposed action it is difficult to place an exact number on the percentage of a
population that is affected or how frequently a population is affected because of the lack of
information on the spatial and temporal uses of the registered formulations containing carbaryl,
carbofuran, and methomyl, compounded by the imperfect data on where salmonids are at any
given time. However, NMFS has sufficient information to make reasonable inferences from the
available use, exposure, and response data on the likelihood of population level consequences.
Below we address whether the remaining fitness level consequences identified from the risk
hypotheses affect the viability of salmonid populations. As mentioned earlier, we focus on the
potential for reduced population abundance and productivity.
Impaired swimming and olfactory-mediated behaviors
All life stages of salmonids rely on their inherent ability to smell, to swim, and to navigate
through a variety of habitats over their life span and to ultimately spawn successfully in natal
waters and complete their life cycle. We have shown that exposure to carbaryl and carbofuran,
compared to effects concentrations necessary to impair swimming behaviors, appears sufficient
to do so in some environments, especially in rearing locations for the juveniles. Although, no
data were available to evaluate the effect of methomyl on swimming behavior, we find it
reasonable to apply the conclusions drawn for carbaryl and carbofuran due to the chemicals
sharing a similar mechanism of toxic action. Specifically, we expect that salmonids with
impaired swimming behaviors from AChE inhibition will show reduced feeding, delayed or
interrupted migration, reduced survival, and reduced reproductive success. We conclude that
exposed populations are likely to have reduced abundance and productivity as a result of
impaired swimming.
Based on the information we reviewed for carbaryl, carbofuran, and methomyl on salmonid
olfaction, we find differences in expected responses. For carbaryl, we conclude that it is unlikely
DRAFT 428
-------
DRAFT
to affect salmonid olfaction at estimated concentrations. For methomyl, we located no
information on its effects to fish olfaction, and given the variation olfactory responses measured
in from other AChE inhibitors (OPs and carbamates) it is uncertain whether methomyl will affect
olfaction. For carbofuran, definitive evidence shows that olfaction in fish is affected at low |j,g/L
concentrations.
Because olfaction plays an important role in a suite of ecologically relevant behaviors that are
affected when an individual salmonid's olfaction is impaired, we include this endpoint in our
analysis. Lack of predator avoidance behaviors by juvenile and adult salmonids reduces the
probability of surviving predation events. Juvenile salmonids with impaired olfaction may fail to
properly imprint on their natal waters, which later in life leads to adults straying i.e., migrating
into and spawning in streams other than their natal stream. Adults that do not return to natal
waters are a functional loss to recruitment of a population. Adult male salmonids that do find
their way back to natal stream or river reaches and are subsequently exposed to the three
insecticides may lose some or all of their olfactory capacity, even from a short-term exposure.
Female salmonids release odorants to trigger male priming hormones and to alert males of a
female's spawning condition. However, male fish with reduced olfactory capacity may not
detect these cues, as demonstrated in a study on carbofuran (Bretaud, Saglio et al. 2002) thus
spawning synchronization could be compromised and recently laid eggs may go unfertilized.
Unfertilized eggs may result in reduced productivity and abundance for a population if sufficient
numbers of spawning events are missed. Again, we find it difficult to accurately predict when
these impairments and missed spawning opportunities occur, primarily as a result of incomplete
pesticide use information, difficulty in conducting field experiments with adult salmonids, and
uncertainties surrounding extent of effects and concentrations which may trigger them. Because
imprinting, avoiding predators, homing, and spawning are likely affected when exposed to
carbofuran, we conclude these additional effects cannot be dismissed. Therefore, we expect
populations exposed to carbofuran will show reduced reproductive rates, reduced return rates,
and reduced intrinsic rates of growth when sufficient numbers of individuals are affected.
Starvation during a critical life stage transition
Salmonids emerge from redds (nests) with a yolk-sac as their initial food source (yolk-sac fry).
Once the yolk-sac has been completely absorbed, they must begin exogenous feeding. Fry have
DRAFT 429
-------
DRAFT
limited energy reserves, and if they are unable to properly swim or detect and capture prey the
onset of starvation occurs rapidly. Because prey items are limited by gape width, prey at this
stage are limited to very small aquatic invertebrates. The stressors of the action likely affect this
critical life stage transition in several ways, leading to increased early life stage mortality.
Impaired swimming and olfaction affects the fry's ability to detect and capture prey. Prey may
be killed outright by the stressors of the action leading to reduced prey availability or the
complete absence of prey. These same areas also have off-channel habitats where fry seek
shelter and food and those areas are highly susceptible to the highest concentrations of the three
insecticides. Therefore, we expect reductions in a population's abundance where transit!oning
yolk-sac fry are exposed to the stressors of the action. All salmonid life histories share this
common life stage transition and therefore are at risk.
Death of returning adults
We discussed and analyzed with models the importance of juveniles to population viability.
However, we did not address possible implications of returning adults dying from exposure to
the stressors of the action. An adult returning from the ocean to natal freshwaters is important to
a population's survival and recovery for many reasons. Notably, less than one percent of adults
generally complete their life cycle. For populations with lambdas well below 1, every adult is
crucial to a population's viability. We expect that some sensitive adults will die from short-term
exposures before they spawn within some of the populations, particularly those that spawn in
intensive agricultural watersheds and urban/suburban environments where elevated temperatures,
other AChE-inhibiting insecticides, and low pHs are present. The lower end concentrations that
kill 50% of exposed salmonids fall in the hundreds of ng/L. For methomyl, we expect that fewer
adults would die compared to carbofuran and carbaryl based on differences in toxic potentcies.
We still expect that sensitive individuals exposed concentrations below the LC50 will die. These
concentrations are expected in habitats receiving drift from aerial applications. The persistence
of these concentrations will vary with habitat, but for those habitats with lower pHs and minimal
flow sensitive adult salmonids are expected to be killed. We cannot quantify the number of
adults lost to a given population in a given year. For those populations where each adult
salmonid is important to viability, we expect reductions in both productivity and abundance.
DRAFT 430
-------
DRAFT
Synergistic toxicity
With certain combinations and specific concentrations of carbaryl and carbofuran, synergism
occurs, translating into increased inhibition of AChE and in some cases increased rates of
mortality among exposed salmonids. We have no data either supporting or refuting synergism
for methomyl. We have no predictive models for synergistic toxicity. However, where we
expect co-occurrence of the three insecticides, we expect synergism may occur if high enough
concentrations exist. Generally, these concentrations must exceed individual LCSOs for the three
compounds which is most likely to occur in areas with extensive crop uses where applications
overlap in space and time. In these areas, even more fish would die from synergistic effects than
predicted based on the additive toxicity for carbaryl, carbofuran, and methomyl. Juveniles and
returning adults may experience synergistic toxicity. Whether or not death occurs is dependent
on exposure duration and concentrations of the three insecticides. Typically, adults are less
sensitive than early lifestages, however it is very difficult to conduct toxicity assays with pre-
spawn adult salmoninds. Prespawn adults have used up most of their accumulated fat,
converting it into gamete production, and will die soon after spawning. We are unsure how
sensitive these adults are to toxic pesticides, but expect in their physiological state that they will
be as or more sensitive than juveniles. We expect that some adults that occupy or spawn in
shallow, low flow systems will be impacted from synergistic toxicity. Therefore, population-
level effects could be more pronounced, depending upon the number of individuals and the
importance of those individuals to the survival and recovery of the population. We conclude that
based on the expected environmental concentrations of the three insecticides, synergism is likely
in many off-channel habitats resulting in increased rates of death to juveniles and to adults.
Toxicity from other stressors of the action
We identified inert ingredients, adjuvants (NP), tank mixtures (recommended on pesticide
product labels), degradates (1-napthol, 3-hydroxycarbofuran), and other pesticide a.i.s
(malathion, bifenthrin, rotenone, metaldehyde, captan, and cupric sulfate) as toxic to salmonids
and their prey. There remain substantial data gaps on the expected concentrations of these
chemicals in salmonid habitats. However, some chemicals are detected at concentrations that
pose substantial risk to listed salmonids and their prey e.g., malathion, NP. The risk posed by
these other stressors to salmonid populations is complicated by the same factors we discussed for
carbaryl, carbofuran, and methomyl (i.e., the numbers of individuals exposed, the uncertainty
DRAFT 431
-------
DRAFT
surrounding the temporal and spatial uses of these chemicals, etc.). Severe population reductions
of Atlantic salmon in Canada were attributed specifically to the use of NP within a pesticide
formulation (Fairchild, Swansburg et al. 1999; Brown and Fairchild 2003). We conclude that
given the use and co-application of these chemicals with carbaryl, carbofuran, and methomyl,
exposed individuals are at increased risk of the suite of toxic effects expected from these
substances. Substantial uncertainty also exists regarding the identity of other ingredients in
formulations further complicating our ability to predict overall toxicity to salmonids and their
prey as a result of pesticide use. Exposed populations are at increased risk of reduced abundance
and productivity from these chemicals. However, NMFS is unable to accurately describe the
level of risk.
Conclusion on population-level effects
We conclude that many of the populations of threatened and endangered salmonids covered by
this consultation will likely show reductions in viability, particularly those that are comprised of
juvenile life histories that rear for weeks to years in freshwater habitats found in intensive
agricultural and residential/urban areas. Juvenile coho, steelhead, and ocean- and stream-type
Chinook salmon use these type of rearing areas for extended periods which overlaps with
pesticide applications. Of greatest concern are for those independent populations for each ESU
or DPS overlapping with uses of carbaryl, carbofuran, and methomyl.
Based on these facts, we expect that where the geographic range of listed populations overlap
with intensive cropping patterns and urban/residential areas severe effects to abundance and
productivity are anticipated from exposure to the three pesticides. However, due to the
impending phase out of carbofuran, we expect the likelihood of exposure and therefore toxicity
to decline rapidly as existing products stocks are depleted. Population effects are largely a result
of reduced salmonid prey abundances and subsequent reduced growth of juveniles. Although
less of a factor, individual fitness consequences are also likely in some areas due to impaired
swimming.
DRAFT 432
-------
DRAFT
Effects to Designated Critical Habitat: Evaluation of Risk Hypotheses
Presently, critical habitat has been designated for 26 of the 28 listed salmonids, all of which is
located within the action area. Designated critical habitat within the action area includes
spawning and rearing areas, freshwater migratory corridors, and nearshore and estuarine areas,
and includes essential physical and biological features. The effects of the proposed action on
prey and water quality PCEs are addressed below by the following risk hypotheses. If the PCEs
are negatively impacted, we address the potential for reductions to the associated conservation
value of the designated critical habitat in the Conclusion Section.
Risk Hypotheses:
1. Exposure to the stressors of the action is sufficient to reduce abundances of aquatic
prey items of salmonids.
We evaluated two lines of evidence to determine whether this hypothesis is supported by the
available information. The first is whether data support the occurrence of adverse effects to
salmonid prey items from the stressors of the action. The second is whether reductions in
abundances of salmonid prey items occur in areas of documented exposure to the stressors of the
action. We found overwhelming evidence in support of the first line of evidence. The stressors
of the action are expected to kill large numbers and types of aquatic species that serve as prey to
salmonids, especially when carbaryl, carbofuran, and methomyl are present together and/or co-
occur with other insecticides. The concentrations we summarized indicate that alone each of the
insecticides can also kill prey at expected environmental concentrations.
The IBI and other metrics of aquatic community health were reviewed to evaluate the second
line of evidence. In areas of intensive agriculture, where we expect use of the stressors of the
action, biological integrity is often significantly reduced (Cuffney et al 1997). Many of the
preferred salmonid prey items are present only in low numbers or absent altogether in these
areas. We see similar depauperate communities in urban areas. We recognize many other
limiting factors contribute to poor condition of these aquatic communities. However, these
insecticides and their formulations may be responsible for a substantial portion. In fact, several
studies have shown toxicity to salmonid prey items from field collected waters and sediment due
DRAFT 433
-------
DRAFT
to pesticide residues (Cuffney, Meador et al. 1997; Hall, Killen et al. 2006). In summary, the
available information shows prey items of ESA-listed salmonids are affected by the stressors of
the action to an extent warranting an analysis of whether the conservation value of designated
critical habitat is reduced.
2. Exposure to the stressors of the action is sufficient to degrade water quality in
designated critical habitat.
We evaluated this hypothesis by applying exposure concentrations evaluated in the Exposure
Analysis and toxicity data from the Response Analysis. We did not compare expected
concentrations of these pesticides to U.S. Water Quality Criteria as none are published.
We evaluated whether any of the state waters within designated critical habitat are listed as
impaired by carbaryl, carbofuran, and/or methomyl by searching 303(d) lists.
• The expected concentrations from the proposed action trigger adverse effect levels for
salmonids and their prey (see Exposure Analysis and Response Analysis). We expect
these concentrations to be present in designated critical habitat and therefore to degrade
water quality.
• Several waterbodies in the action area are listed under section 303(d) specifically for the
pesticides considered in this Opinion, including: Willapa Bay, WA (carbaryl, area not
specified, 18 separate listings), Gray's Harbor County Drainage Ditch #1, WA (carbaryl,
area not specified), Pacific County Drainage Ditch #1, WA (carbaryl, area not specified),
North River, WA (carbaryl, area not specified), Palix River (carbaryl, area not specified),
Johnson Creek, OR (carbaryl & carbofuran, 23.7 river miles), Beaverton Creek, OR
(carbaryl, 9.8 river miles), Tualatin River, OR (carbaryl, 44.7 river miles), Mill Creek,
OR (carbofuran, 25.7 river miles), and Colusa Basin Drain, CA (carbofuran, 49 river
miles). We located no listings specifically for methomyl, and we located no listings
specifically for these pesticides in the state of Idaho.
In many of the watersheds containing designated critical habitats water quality is identified as a
major limiting factor to salmonid production. The proposed action is likely to further degrade
water quality. Collectively, this information supports the conclusion that designated critical
habitats are likely degraded throughout the four states and further analysis is warranted to
determine the potential to reduce the conservation value of designated critical habitats.
DRAFT 434
-------
DRAFT
Areas of Uncertainty:
In this section we list the predominant uncertainties and data gaps uncovered by our analysis of
the effects of the proposed action. We do not discuss the entire suite of uncertainties, but
highlight those likely have the most influence on the present analysis.
• Description of the action. We lacked a complete description of EPA-authorized uses of
pesticides containing carbaryl, carbofuran, and methomyl as described in labeling of all
pesticide products containing these a.i.s.
• Exposure resulting from non-agricultural uses. We lacked exposure estimates of stressors
of the action associated with non-agricultural uses of these pesticides.
• Exposure to and toxicity of pesticide formulations and adjuvants. Minimal information
was found on formulations, adjuvants, and on other/inert ingredients within registered
formulations.
• Exposure to Mixtures. We lacked information on permitted tank mixtures. Additionally,
given that relatively few tank mix combinations are prohibited, it was not feasible to
evaluate all potential combinations of tank mixtures. Pesticide mixtures are found in
freshwater throughout the listed-salmonid habitat areas. However, mixture constituents
and concentrations are highly variable.
• Toxicity of mixtures. The toxicity of most environmental mixtures is unknown.
• Synergistic responses. Exposure to combinations of carbaryl, carbofuran, methomyl,
and/or other combinations of OP and carbamate insecticides can result in synergistic
responses. However, we are not aware of a method to predict synergistic responses.
DRAFT 435
-------
DRAFT
1 ' i
Cumulative effects include the effects of future state, tribal, local, or private actions that are
reasonably certain to occur in the action area considered by this Opinion. Future federal actions
that are unrelated to the proposed action are not considered in this section because they require
separate consultation pursuant to section 7 of the ESA.
During this consultation, NMFS searched for information on future state, tribal, local, or private
actions that were reasonably certain to occur in the action area. NMFS conducted electronic
searches of business journals, trade journals, and newspapers using Google and other electronic
search engines. Those searches produced reports on projected population growth, commercial
and industrial growth, and global warming. Trends described below highlight the effects of
population growth on existing populations and habitats for all 28 ESUs/DPSs. Changes in the
near-term (five-years; 2014) are more likely to occur than longer-term projects (10-years; 2019).
Projections are based upon recognized organizations producing best available information and
reasonable rough-trend estimates of change stemming from these data. NMFS analysis provides
a snapshot of the effects from these future trends on listed ESUs.
States along the Pacific west coast, which also contribute water to major river systems, are
projected to have the most rapid growth of any area in the U.S. within the next few decades.
This is particularly true for coastal states. California, Idaho, Oregon, and Washington are
forecasted to have double digit increases in population for each decade from 2000 to 2030
(USCB 2005). Overall, the west coast region [which also includes four additional states
(Arizona, Utah, Nevada, and Alaska) beyond the action area] had a projected population of 65.6
million people in 2005. This figure will eventually grow to 70.0 million in 2010 and 74.4
million in 2015. At this rate, such growth will make the Pacific coast states the most populous
region in the nation.
Although general population growth stems from development of metropolitan areas, growth in
the western states is projected from the enlargement of smaller cities rather than from major
metropolitan areas. Of the 42 metropolitan areas that experienced a 10% growth or greater
DRAFT 436
-------
DRAFT
between 2000 and 2007, only seven have populations greater than one million people. Of these
major cities, one and two cities are from Oregon and California, respectively. They include
Portland-Vancouver-Beaverton, OR (1.83%/year), Riverside-San Bernadino-Ontario, CA
(3.63%/year), and Sacramento-Arden-Arcade-Roseville, CA (2.34%/year).
Urban Growth
As these cities border coastal or riverine systems, diffuse and extensive growth will increase
overall volume of contaminant loading from wastewater treatment plants and sediments from
sprawling urban and suburban development into riverine, estuarine, and marine habitats. Urban
runoff from impervious surfaces and roadways may also contain oil, copper, PAHs, and other
chemical pollutants and flow into state surface waters. Inputs of these point and non-point
pollution sources into numerous rivers and their tributaries will affect water quality in available
spawning and rearing habitat for salmon. Based on the increase in human population growth, we
expect an associated increase in the number of NPDES permits issued and the potential listing of
more 303(d) waters with high pollutant concentrations in state surface waters.
Mining
Mining has historically been a major component of western state economies. With national
output for metals increasing at 4.3% annually (little oil, but some gas is drawn from western
states), output of western mines should increase markedly (Woods and Figueroa 2007).
Increases in mining activity will add to existing significant levels of mining contaminants
entering river basins. Given this trend, we expect existing water degradation in many western
streams that feed into or provide spawning habitat for threatened and endangered salmonid
populations will be exacerbated.
Agriculture
As the western states have large tracts of irrigated agriculture, a rise in agricultural output is
anticipated. Impacts from heightened agricultural production will likely result in two negative
impacts on listed Pacific salmonids (Woods and Figueroa 2007). The first impact is the greater
use and application of pesticide, fertilizers, and herbicides and their increased concentrations and
DRAFT 437
-------
DRAFT
entry into freshwater systems. Carbaryl, carbofuran, and methomyl, and other pollutants from
agricultural runoff may further degrade existing salmonid habitats. Second, increased output and
water diversions for agriculture may also place greater demands upon limited water resources.
Water diversions will reduce flow rates and alter habitat throughout freshwater systems. As
water is drawn off, contaminants will become more concentrated in these systems, exacerbating
contamination issues in habitats for protected species.
Recreation
The western states are widely known for scenic and natural beauty. Increasing resident and
tourist use will place additional strain on the natural state of park and nature areas that are also
utilized by protected species. Hiking, camping, and recreational fishing in these natural areas is
unlikely to have any extensive effects on water quality.
The above non-federal actions are likely to pose continuous unquantifiable negative effects on
listed salmonids addressed in this Opinion. Each activity has undesirable negative effects on
water quality. They include increases in sedimentation, loss of riparian shade (increasing
temperatures), increased point and non-point pollution discharges, decreased infiltration of
rainwater (leading to decreases in shallow groundwater recharge, decreases in hyporheic flow,
and decreases in summer low flows).
Nevertheless, there are also non-federal actions likely to occur in or near surface waters in the
action area that may have beneficial effects on the 28 ESUs. They include implementation of
riparian improvement measures, fish habitat restoration projects, and BMPs (e.g., associated with
timber harvest, grazing, agricultural activities, urban development, road building, recreational
activities, and other non-point source pollution controls).
NMFS expects many of the current anthropogenic effects described in the Environmental
Baseline will continue. Listed Pacific salmonids are exposed to harvest, hatchery, hydropower,
and habitat degradation activities. With regard to water quality, fish are continually exposed to
pesticides, contaminants, and other pollutants during their early life history phase and during
adult migratory returns to their natal streams for spawning.
DRAFT 438
-------
DRAFT
NMFS also expects the natural phenomena in the action area (e.g., oceanographic features,
ongoing and future climate change, storms, natural mortality) will continue to influence listed
Pacific salmonids as described in the Environmental Baseline. Climate change effects are
expected to be evident as alterations to water yield, peak flows, and stream temperature. Other
effects, such as increased vulnerability to catastrophic wildfires, may occur as climate change
alters the structure and distribution of forest and aquatic systems.
Coupled with EPA's registration of carbaryl, carbofuran, and methomyl, climate change, and the
effects from anthropogenic growth on the natural environment will continue to affect and
influence the overall distribution, survival, and recovery of Pacific salmonids in California,
Idaho, Oregon, and Washington.
-;,',- -.,• ••- •". .,
The Integration and Synthesis section is the final step of NMFS' assessment of the risk posed to
listed Pacific salmonids and critical habitat as a result of EPA's registration of carbaryl,
carbofuran, and methomyl. In this section, we perform two evaluations: whether it is reasonable
to expect the proposed action is likely to (1) reduce both survival and recovery of the species in
the wild and (2) result in the destruction or adverse modification of designated critical habitat.
To address species survival and recovery, we discuss the likelihood of the proposed action to
reduce the viability of the salmonid species to such an extent that increased extinction rates are
likely. Specifically, we address whether we expect reductions in a population's abundance and
productivity from the stressors of the action to affect a species (ESU or DPS). We also address
whether the critical habitat will remain functional to serve the intended conservation role for 26
listed Pacific salmonid ESUs/DPSs or retain its current ability to establish those features and
functions essential to the conservation of the species. We address whether we expect reductions
in the PCEs for prey availability and degradation of water quality from the stressors of the action
to reduce the overall conservation value of designated critical habitat. Conclusions for each
ESU/DPS and associated designated critical habitat are found in the Conclusion section.
DRAFT 439
-------
DRAFT
Effects of the proposed action at the species level
In our Effects of the Proposed Action section, we assessed the effects of the federal action to
listed Pacific salmonids. We discussed the exposure, response, and risks to individuals when
they co-occur with the allowable uses of the three a.i.s., their metabolites and degradates, other
active and inert ingredients included in their product formulations, tank mixtures, and adjuvants
authorized on their product labels within the action area. Our analysis also considered
combinations of cholinesterase-inhibiting pesticides that may be applied on the same day or over
a short interval or occur within freshwater habitats, and the cumulative exposure of multiple
pesticide applications containing the three a.is. that may have additive or synergistic effects to
salmonids. We coupled these stressors with the environmental mixtures, degraded habitat
conditions, and common detections of carbaryl, carbofuran, and methomyl present in the
baseline within the four western states where listed salmonids are distributed.
As listed salmonids use a wide range of freshwater, estuarine, and marine habitats, many migrate
hundreds and thousands of miles to complete their life cycle. Adult salmon and steelhead spend
weeks to several months in freshwater habitats during their migration and spawning activities.
Immediately after emerging from the gravel substrate and transitioning from alevins to fry,
salmonids move to habitat where they can swim freely and forage. At this point in their
development most salmon occupy freshwater habitats. Chum salmon are an exception. They
immediately migrate downstream following emergence to nearshore environments in estuaries
near the mouth of their parent stream. Chum fry may reside immediately next to the shore in
estuaries for as little as one or two weeks before moving offshore or into deeper water habitats
within the nearshore environment. Sockeye salmon fry most frequently distribute to shallow
beach areas in the littoral zones of lakes before moving offshore and taking on a more pelagic
existence. Coho salmon, Chinook salmon, and steelhead fry typically select off-channel habitats
associated with their natal rivers and streams. Coho salmon and steelhead actively seek shallow
habitats and rear in freshwater for more than a year.
Salmonids use off-channel habitats as protection from a river's or a stream's primary flow and to
take advantage of feeding opportunities. Diverse, abundant communities of invertebrates (many
of which are salmonid prey items) populate these habitats and, in part, are responsible for
DRAFT 440
-------
DRAFT
juvenile salmonids' reliance on off-channel habitats. Thus, rearing and migrating juvenile
salmonids use these habitats extensively. We expect that applications of several Ibs of a.i. per
acre adjacent to off-channel habitats result in substantial aquatic toxicity to aquatic life, including
death of exposed salmonids and substantial reduction in salmonid prey taxa. Based on general
state of knowledge for field runoff and drift from pesticide applications, we expect higher
pesticide concentrations in edge-of-field, low-flow, and shallow aquatic systems.
Based on the above, NMFS expects high exposure for listed salmonids to multiple pesticides and
stressors of the action throughout their lifetime. We further expect exposure to multiple
pesticide ingredients in freshwater habitats and nearshore environments adjacent to areas where
pesticides are used. As the duration of the proposed action is 15 years, authorized uses of
carbaryl, carbofuran, and methomyl, listed Pacific salmonids are continually exposed to these
inputs in the action area.
We expect the a.i.s found in surface water runoff and pesticide drift to cause acute lethality and
sublethal effects to juvenile salmonids and consequently depress the lambda of a population.
Additionally, we expect indirect effects through reduced salmonid prey abundance and
subsequent reduced juvenile growth to lead to increased juvenile death. Impaired swimming and
olfactory-mediated behaviors of individuals can also result in adverse population-level effects.
Swimming is a critical function for salmonids to complete their complex life cycle. Salmonids
with impaired swimming behaviors from AChE inhibition will show reduced feeding, delayed or
interrupted migration, reduced survival, and reduced reproductive success. We expect that the
stressors of the action will impair swimming. Olfactory-mediated behaviors are also important
for the successful completion of salmonids' life cycles. They include detecting and avoiding
predators, recognizing kin, locating natal waters for spawning, and reproduction. Juvenile
salmonids with impaired olfactory-mediated behaviors can lose their ability to imprint on natal
streams, and avoid predators. Adults may lose their ability to recognize spawning events and
resulting in reduced reproduction.
DRAFT 441
-------
DRAFT
If exposed to carbofuran at low ug/L concentrations, impairment of olfactory-mediated behaviors
such as feeding, imprinting, homing, spawning, and predator avoidance will likely result.
Consequently, the above conditions lead to increased predation and ultimately reduced juvenile
and adult survivorship. As exposure increases we expect to see more pronounced effects due to
additive and synergistic effects.
Based on exposure of salmonid populations to carbaryl, carbofuran, and methomyl for four days
at the reported LCSOs, severe consequences to a salmonid's population growth rate is expected.
The most pronounced effects are within off-channel habitats based on NMFS exposure estimates.
The likelihood of population effects from death of salmonid juveniles increases for those
populations that spend longer periods in freshwaters especially during pesticide applications
adjacent to off-channel habitats and shallow water systems. They include juvenile steelhead,
coho, stream-type Chinook, and coho salmon. Methomyl concentrations required to result in
population level impacts from acute mortality of juveniles are not expected in the majority of
aquatic habitats used by juveniles. However, we expect that carbaryl and carbofuran
applications could cause population level effects through acute mortality of juveniles particularly
in areas of intensive agriculture and large urban/residential areas. We also expect additional
acute mortalities of juveniles from applications of carbaryl to forests and from exposure to 24(c)
carbaryl applications in estuaries in Washington State. However, we do not know how many
individuals are exposed each year following these applications.
We also expect indirect effects from the a.i.s through reduced salmonid prey abundance and
subsequent reduced juvenile growth and probability of survival. Juvenile feeding during the
period immediately following emergence is vital for the survival of salmonid fry and shared by
all ESA-listed Pacific salmonids. Once juveniles have completely absorbed the yolk-sac, they
are dependent on the local invertebrate population. Fry may starve if sufficient food is not
available. As salmonid prey are sensitive to the expected exposures to each of the insecticides as
well as from mixtures containing the three a.i.s, we expect death and a variety of sublethal
effects to salmonid prey items. If the salmonid prey base is limited while fry are in this life
stage, juvenile growth will be significantly hindered. Thus, juvenile survivorship will be low
and may consequently lead to a decrease in lambda for the affected population.
DRAFT 442
-------
DRAFT
Based on the NMFS growth models, a four-day exposure to expected concentrations of carbaryl,
carbofuran, and methomyl will substantially reduce a population's growth rate due to prey base
reduction. All four life history types modeled demonstrated this effect. As exposure duration
increases, we expect more pronounced effects on salmonid prey abundance and a longer prey
recovery period leading to further reduced salmonid growth and decreased survival.
We expect that all effects will be more pronounced in drainage basins that include high use
areas, i.e., major agricultural, urban, or residential centers. To determine which ESUs/DPSs will
be affected by the action, we compare the range of each species to the locations of high use land
classes. Species that occupy major agricultural basins, such as the Willamette basin or California
Central Valley, have a much higher risk of exposure than species found in uninhabited, non-
agricultural areas.
California Coastal Chinook Salmon
The CC Chinook salmon ESU consists of 10 historically functionally independent populations in
California from Humboldt County in the north to Sonoma County in the south. Historic annual
escapement was around 73,000 fish with most of these being produced in the Eel River. Today,
annual returns to the Eel River system is estimated at 150 to 2,800 fish. Runs in the Russian
River may be viable, though the short-term trend is negative. Lack of data precludes
development of population growth rates or trends for all other populations.
The Status of Listed Resources and Environmental Baseline sections indicate that fisheries,
timber harvest, vineyards and other agriculture, introduced fish species, and migration barriers
negatively affect this ESU. Adverse effects on Chinook salmon habitat include a high
percentage of fines in the streams' bottom substrate, lack of large instream woody debris,
reduced riparian vegetation, elevated water temperatures, increased predation, and barriers that
limit access to tributaries. The Mediterranean climate in California may result in high
concentration of contaminants in runoff during the onset of the rainy season. The early spawning
runs expose adults to the first flush runoff of contaminants. The cumulative impacts from these
multiple threats continue to affect the CC Chinook salmon. Accumulation of fine sediment in
DRAFT 443
-------
DRAFT
the bottom substrate, contamination of waters, lack of riparian vegetation, and introduced fish
species are expected to impact invertebrate abundance and composition in streams within the
ESU.
Most of the agriculture and urban developments are concentrated in the Russian River valley and
on alluvial plains of coastal rivers. Vineyards and smaller centers are dispersed along coastal
watersheds. Based on the crop types, we expect carbaryl and methomyl are commonly applied
throughout the growing season in the Russian River basin. We also expect use of the a.i.s on the
alluvial plain of the lower Eel River and its estuary, and to a lesser extent along the lower
portions and estuaries of other coastal streams. Carbaryl is expected to be used the entire year
within urban/residential areas. Existing stocks of carbofuran are also expected to be applied
throughout the growing season. In California, there are 61 pesticides that inhibit AChE approved
for use and we expect that application of other AChE inhibiting pesticides will co-occur with the
a.i.s, exacerbating adverse effects from AChE inhibition. However, monitoring data for
pesticides in streams within the CC Chinook salmon ESU is lacking. Adults within Willapa Bay
during 24(c) carbaryl application to oyster beds are likely exposed to this compound and may
experience additional mortalities.
CC Chinook spawning occurs on gravel beds in the mainstem of rivers and larger tributaries.
Fry use the floodplain, stream margins, and side channels to rear after emergence from the redds
in December through mid-April. Downstream migration of juveniles starts as early as February
and most enter marine waters before mid-summer.
We expect the proposed uses of carbaryl and carbofuran pesticide products that contaminate
aquatic habitats will lead to individual fitness level consequences and subsequent population
level consequences. Land use and crop type data indicate that, while not a major agricultural
center, the CC Chinook will be exposed to all three a.i.s. We expect that exposure to methomyl
may lead to individual fitness consequences, but not to an extent that would affect population
growth rates. The risk to this species' survival and recovery from the stressors of the action is
high for carbaryl and carbofuran, but low for methomyl.
DRAFT 444
-------
DRAFT
Central Valley Spring-run Chinook Salmon
The CV Spring-run Chinook salmon ESU includes four populations of Spring-run populations in
the upper Sacramento River and its tributaries. The distribution within the Sacramento River
basin is now mostly restricted to accessible areas below dams in the mainstem river and in three
of its tributaries: Deer, Mill, and Butte Creeks. Abundance remains far below the estimated
700,000 once entering the Sacramento-San Joaquin Rivers system. The number of Spring-run
Chinook salmon spawning in the Sacramento River has averaged about 9,800 annually since
2000. While most populations within the ESU are at or above replacement, the Sacramento
River population has been steadily decreasing.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include impaired or loss of habitat, predation, contamination, and water
management. Reservoir dams in the Sacramento River have prevented the ESU from using its
historic spawning locations. Physical channel habitat has been altered through sediment input
from mining, levee construction, and removal of riparian vegetation for levee maintenance.
Detected pesticides in the Sacramento River include thiobencarb, carbofuran, molinate, simazine,
metolachlor, and dacthal, chlorpyrifos, carbaryl, and diazinon. State and federal water diversions
in the south Sacramento-San Joaquin Delta (Delta) have resulted in increased mortality through
prolonged migration and entrainment at the water diversion facilities. We also expect that
application of other AChE inhibiting pesticides will co-occur with carbaryl, carbofuran, and
methomyl in the waters within the ESU and exacerbate adverse effects from AChE inhibition.
With the high density of agriculture in the Sacramento River valley and the Sacramento-San
Joaquin Delta (Delta), application of all three a.i.s is expected. Large urban centers occur along
the Sacramento River and San Francisco Bay. Young and adult migrating Chinook salmon are
also exposed to poor water quality from agricultural runoff that enters the Delta from the San
Joaquin River. Carbaryl and methomyl can be applied to several crops throughout the growing
season in the Central Valley. Existing stocks of carbofuran are also expected to be applied
throughout the growing season. Carbaryl is approved for use throughout the year within
urban/residential areas
DRAFT 445
-------
DRAFT
The CV Spring-run Chinook salmon is categorized as an ocean-type fish. The young salmon
emerge from the gravel from November through May and outmigration starts within four
months. While the majority emigrates soon after emergence, some juveniles emigrate as
yearlings with the onset of fall storms. Juveniles' seaward migration follows the mainstem
Sacramento River, through the Delta, and across San Francisco Bay. Off-channel habitats within
the lower Sacramento River floodplains, especially the Yollo Bypass, are particularity important
for CV Chinook salmon juveniles during rearing and migration (Sommer, Nobriga et al. 2001;
Sommer, Harrell et al. 2005).
The widespread uses of these materials have substantial overlap with rearing and migratory
habitat of the CV Spring-run Chinook salmon ESU. Given the CV Spring-run Chinook salmon
high variation in abundance, restricted distribution, and the exposure of juveniles from all
populations during rearing and downstream migration, we expect a high risk to this species'
survival and recovery from the stressors of the proposed action.
Lower Columbia River Chinook salmon
The LCR Chinook salmon ESU includes 32 historical populations in tributaries from the ocean
to the Big White Salmon River, Washington and Hood River, Oregon. The ESU also includes
17 artificial propagation programs. LCR Chinook salmon numbers began to decline by the early
1900s from habitat degradation and excessive harvest rates. Many of these populations have low
abundance. The annual population growth rates for 14 independent populations range from 0.93
to 1.037.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include hydromorphological changes from hydropower development, loss of
tidal marsh and swamp habitat, and reduced or eliminated access to subbasin headwaters from by
the construction of non-federal dams. LCR Chinook salmon spawning and rearing habitats in
tributary mainstems have been adversely affected by sedimentation, elevated water temperature,
and reduced habitat diversity. The survival of yearlings in the ocean is also affected by habitat
conditions in the estuary, such as changes in food availability and the presence of contaminants.
DRAFT 446
-------
DRAFT
NAWQA sampling in surface waters within the ESU range detected more than 50 pesticides in
streams. Concentrations often pesticides, including carbaryl and carbofuran, also exceeded
EPA's chronic toxicity aquatic life criteria (Wentz, Bonn et al. 1998). The cumulative impacts
from these multiple threats continue to affect this ESU.
Most of the highly developed land and agricultural areas in this ESU's range are adjacent to
salmonid habitat. Registered uses of carbaryl, carbofuran, and methomyl include applications to
crop agricultural sites, residential sites, and urban sites. Registered 24(c) uses in Oregon include
carbofuran application to potatoes, nursery stock, sugar beets, and watermelons. Based on land
use patterns, we expect the highest exposure to these chemicals to occur in the Lewis River
basin, Clackamas River basin, and Hood River basin.
Mature LCR Fall-run Chinook salmon enter freshwater in August through October to spawn in
large river mainstems. After emergence, fry typically select off-channel habitats associated with
their natal rivers and streams. Juveniles eventually emigrate from freshwater, usually within six
months of hatching. LCR Spring-run Chinook salmon enter freshwater in March through June to
spawn in upstream tributaries. These fish generally emigrate from freshwater as yearlings. As
juveniles overwinter in shallow, freshwater habitats, they are likely to experience higher
exposure to pesticides, other contaminants, and elevated temperature. In northern rivers,
juveniles may rear in freshwater for two years or more. Given their long residence time in
shallow freshwater habitats, LCR Chinook salmon are vulnerable to high pesticide exposures.
Also, if adults and juveniles are within Willapa Bay during 24(c) carbaryl application to oyster
beds they are likely exposed to this compound and may likely experience additional mortalities.
Given the life history of LCR Chinook salmon, we expect the proposed uses of carbaryl,
carbofuran, and methomyl pesticide products may contaminate aquatic habitats and lead to
individual fitness and subsequent population level consequences. Therefore, the risk to this
species' survival and recovery from the stressors of the action is high.
DRAFT 447
-------
DRAFT
Upper Columbia River Spring-run Chinook salmon
The UCR Spring-run Chinook salmon ESU includes 11 populations and 7 artificial propagation
programs in the state of Washington. The four known annual population growth rates range
from 0.99 to 1.1. Based on 1980-2004 returns, the average annual population growth rate for this
ESU is estimated at 0.93. One historical population is considered extinct.
The major threats to UCR Chinook salmon identified in the Status of Listed Resources and
Environmental Baseline sections include reduced tributary stream flow and impaired fish
passage from hydroelectric dams. Additionally, degradation of the tributary habitat and impaired
water quality from development negatively affect this ESU. Pesticide detections in UCR
Chinook salmon freshwater habitats are well documented. NAWQA sampling from 1992-1995
in the Central Columbia Plateau detected numerous pesticides in surface water (Williamson et al.
1998). Carbaryl was detected in 5% of samples, and carbofuran was detected in 6%. Methomyl
was not detected. While detections of these three chemicals did not exceed water quality
standards, concentrations of six other pesticides exceeded EPA criteria for the protection of
aquatic life. Listed salmonids are commonly exposed to combinations of carbaryl, carbofuran,
and methomyl, other compounds, and other mixtures of cholinesterase-inhibiting insecticides.
Registered uses of carbaryl, carbofuran, and methomyl include applications to crop agricultural
sites, residential sites, and urban sites. Registered 24(c) uses in Washington State include
carbofuran application to potatoes and spinach grown for seed. About 3% of the land has been
developed and 3.5% has been cultivated for crops. Areas with the potential for high exposure in
the Upper Columbia River basin are the Methow, Entiat, and Wenatchee River basins.
UCR Spring-run Chinook salmon begin returning from the ocean in the early spring. After
migration, they hold in freshwater tributaries until spawning in mid- to late August. Fish spawn
and rear in the major tributaries leading to the Columbia River between Rock Island and Chief
Joseph dams. UCR Spring-run Chinook salmon fry typically select off-channel habitats
associated with their natal rivers and streams to rear. Juveniles spend a year in freshwater before
migrating to the ocean in the spring of their second year of life. The duration of juvenile rearing
DRAFT 448
-------
DRAFT
in shallow freshwater habitats increases their susceptibility to higher exposures of pesticides,
contaminants, and elevated temperature.
Given the life history of UCR Spring-run Chinook salmon, we expect the proposed uses of
carbaryl, carbofuran, and methomyl pesticide products that contaminate aquatic habitats will lead
to individual fitness and likely lead to subsequent population level consequences, i.e., reductions
in population viability. Therefore, the risk to this species' survival and recovery from the
stressors of the action is high.
Puget Sound Chinook Salmon
The Puget Sound ESU includes all runs of Chinook salmon in the Puget Sound region from the
North Fork Nooksack River to the Elwha River on the Olympic Peninsula. This includes 31
historic quasi-independent populations and 26 artificial propagations programs. Of the historic
populations, only 22 are considered extant. The estimated total run size for this ESU in the early
1990s was 240,000 fish. During a recent five-year period, the geometric mean of natural
spawners in populations of this ESU ranged from 222 to just over 9,489 fish (Good, Waples et al.
2005). Recent five-year and long-term productivity trends remain below replacement for the
majority of the 22 extant populations of Puget Sound Chinook salmon. The annual population
growth rate known for these populations ranged from 0.75 to 1.17.
The major threats to the Puget Sound Chinook salmon identified in the Status of Listed
Resources and Environmental Baseline sections include degraded freshwater and marine habitat
from agricultural activities and urbanization. Poor forestry practices have also reduced water
quality in the upper river tributaries for this ESU. Elevated temperature, water diversions, and
poor water quality across land use categories pose significant threats to the status of Puget Sound
Chinook salmon. Furthermore, there has been extensive urbanization in this region. Well over
two million people inhabit the area, with most development occurring along rivers and coastline.
Pesticide use and detections in the ESU's watershed are well documented. NAWQA sampling
conducted in 2006 in the Puget Sound basin detected numerous pesticides and other synthetic
organic chemicals in streams and rivers. However, mixtures of chemicals found in agricultural
DRAFT 449
-------
DRAFT
and urban settings differ. Urban streams sampled in Puget Sound showed the highest detections
for carbaryl, diazinon, and malathion. Carbaryl was detected at 60% of urban sampling sites
(Ebbert, Embrey et al. 2000). Diazinon was detected at all urban sites, frequently at
concentrations that exceeded EPA guidelines for protecting aquatic life (Bortleson and Ebbert
2000).
Registered uses of carbaryl, carbofuran, and methomyl include applications to crop agricultural
sites, residential sites, and urban sites. In addition, Washington has 24(c) registrations for
carbofuran use on spinach grown for seed and potatoes. Land classes indicate high-use of the
three a.i.s in the following drainage basins: Nooksack, Skagit River, Stillaguamish, Skykomish,
Snoqualimie, Snohomish, Lake Washington, Duwamish, and Puyallup. Roughly 7% of the
lowland areas (below 1,000 ft elevation) in the Puget Sound region are covered by impervious
surfaces, which increase urban runoff containing pollutants and contaminants into streams.
Pollutants carried into streams from urban runoff include pesticides, heavy metals, PCBs,
PBDEs, PAHs, pharmaceuticals, nutrients, and sediments. As such, we expect salmonid
populations within the ESU to be exposed to carbaryl, carbofuran, and methomyl. Further,
monitoring data shows that we can expect concurrent exposure with other AChE inhibiting
chemicals, which will exacerbate the severity of effects. Adults within Willapa Bay during 24(c)
carbaryl application to oyster beds are likely exposed to this compound and may experience
additional mortalities.
Puget Sound stream-type Chinook salmon fry typically rear in shallow off-channel habitats
associated with their natal rivers and streams. Juveniles generally have long freshwater
residences of one or more years before migrating to the ocean.
Given the life history of the Puget Sound Chinook salmon, we expect that the proposed uses of
carbaryl, carbofuran, and methomyl pesticide products may lead to individual fitness
consequences and subsequent population level consequences, i.e., reductions in population
viability. Therefore, the risk to this species' survival and recovery from the stressors of the
action is high.
DRAFT 450
-------
DRAFT
Sacramento River Winter-run Chinook salmon
The Sacramento River Winter-run Chinook salmon ESU includes only one population in
Sacramento River, California. The current spawning distribution of Sacramento River Winter-
run salmon is restricted to a short portion of the mainstem Sacramento River below Keswick
Dam. Historic run estimates for the Sacramento River are as large as 200,000 fish (Brown et al.
1994). Estimated natural production has fluctuated greatly over the past two decades. In 2007,
estimated natural production was 4,461 fish. The population's annual growth rate ranged from
0.870 to 1.090.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections indicate impaired or loss of habitat, predation, contamination, and water
management negatively affect this ESU. Reservoir dams in the Sacramento River have
eliminated the ESU from its historic spawning locations. Today, the ESU depends on the ability
of the BOR to manage cold water through reservoir storage and releases to support adult holding,
spawning, incubation, and rearing. The physical channel habitat has been altered through
sediment input from mining, levee construction, and removal of riparian vegetation for levee
maintenance. Pesticides are frequently detected in the Sacramento River including thiobencarb,
carbofuran, molinate, simazine, metolachlor, dacthal, chlorpyrifos, carbaryl, and diazinon.
Carbofuran is commonly detected in agricultural areas but less so in urban sites while carbaryl is
commonly detected in all urban sites. Modification of hydrology has resulted in increased
mortality through stranding, increased predation, prolonged migration, and entrainment at water
diversion facilities.
About 10% of the land within this ESU is developed; large areas of urban centers occur along the
Sacramento River and San Francisco Bay. As about 21% of land within the ESU is cultivated,
all three a.i.s are expected to be applied to several crops within the ESU. Agriculture activity is
prominent in the lower Sacramento River and within the Delta. Chinook salmon are also
exposed to poor water quality from agricultural runoff that enters the Delta from the San Joaquin
River. The Mediterranean climate in California, with dry summers and fall storms, may result in
high concentration of these contaminants in run-off during the onset of the rainy season. These
DRAFT 451
-------
DRAFT
high concentrations can overlap with juvenile presence in the river system and movement into
floodplains for rearing.
Winter-run adults enter the Sacramento River in early spring with spawning peaking in May and
June. Spawning occurs in the Sacramento River downstream of the Keswick Dam. Fry rear in
the Sacramento River for a few weeks to months before starting outmigration in late July,
peaking in November and December. During outmigration, the young salmon migrate down the
Sacramento River, through the Delta and San Francisco Bay. Juvenile winter-run Chinook
salmon appear in the Delta from October to early May where they may rear in the fresher
upstream portions for up to two months.
We expect that the proposed uses of carbaryl, carbofuran, and methomyl pesticide products will
lead to both individual fitness level consequences and subsequent population level consequences.
Registered uses of these a.i.s indicate overlap with the spawning, rearing, and migratory habitat
of the one extant population in this ESU. Thus, the risk to this species' survival and recovery
from the stressors of the proposed action is high.
Snake River Fall-run Chinook salmon
The SR Fall-run Chinook salmon ESU is comprised of a single population that spawns and rears
in the mainstem Snake River and its tributaries below Hells Canyon Dam. The range for this
ESU includes the Snake River basin in the states of Idaho, Oregon, and Washington. Only 10 to
15% of the historical range of this ESU remains. Today, the vast majority of spawning occurs
upstream from the Lower Granite Dam. Estimated historical returns from 1938 to 1949 were
72,000 fish annually (Bjornn and Horner 1980). The average abundance (1,273) of SR Fall-run
Chinook salmon over the most recent 10-year period is below the 3,000 natural spawner average
abundance thresholds identified as a minimum for recovery. The annual population growth rate
for this single population is 1.02. Two historical populations are considered extirpated.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include impaired stream flows and barriers to fish passage in tributaries from
hydroelectric dams. During the 1960s and 1970s, approximately 80% of the ESU's historic
DRAFT 452
-------
DRAFT
habitat was eliminated or severely degraded by the construction of the Hells Canyon complex
and the lower Snake River dams. Additionally, degraded freshwater habitats in the estuary,
mainstem, and tributaries from development and land use activities negatively affect this ESU.
Agricultural activities, urban communities, and industries are concentrated along the Snake River
and near the mouths of major tributary valleys. Thus, stream water quality and biological
communities in the downstream portion of the upper Snake River basin are degraded. The Snake
River is directly adjacent to the Lewiston area. The cumulative impacts from these multiple
threats continue to affect SR Fall-run Chinook salmon.
Registered uses of carbaryl, carbofuran, and methomyl include applications to crop agricultural
sites, residential sites, and urban sites. Registered 24(c) uses within this ESU include carbofuran
application to potatoes and spinach grown for seed in Washington and potatoes and sugar beets
in Idaho. The remaining spawning and rearing areas of the SR Fall-run Chinook salmon are
potentially high-use areas for these chemicals. About 14% of the land has been developed.
Agricultural activities, urban communities, and industries are concentrated along the Snake River
and near the mouths of major tributary valleys. Dryland agriculture occurs in lower Clearwater
Basin, with crops including wheat, peas, and lentils. Alfalfa, hay, and grasses are also grown in
the lower Clearwater as well as the lower Salmon River basin. Thus, ESU exposure to these
pesticides is likely. Adults within Willapa Bay during 24(c) carbaryl application to oyster beds
are likely exposed to this compound and may experience additional mortalities.
SR Fall-run Chinook salmon generally spawn and rear in larger, mainstem rivers, such as the
Salmon, Snake, and Clearwater Rivers. Prior to alteration of the Snake River basin by dams, SR
Fall-run Chinook salmon exhibited a largely ocean-type life history. These fish migrate
downstream and rear in the mainstem Snake River during their first year. Today, some SR Fall-
run Chinook salmon in the Snake River basin also exhibit reservoir- type life history. Fish with a
reservoir-type life history overwinter in low velocity pools created by the hydroelectric dams.
This condition prevents juveniles from reaching a suitable size before they migrate out of the
Snake River, increasing their susceptibility to toxins.
DRAFT 453
-------
DRAFT
Adult SR Fall-run Chinook salmon enter the Columbia River in July and August. Spawning
occurs above Lower Granite Dam in the mainstem Snake River and in the lower reaches of the
larger tributaries. Spawning occurs from October through November and fry emerge from redds
beginning in March or April of the following year. They rear for two months or more in the
sandy littoral zone along the river margin. Parr and presmolts move downstream from natal
spawning and early rearing areas from June through early fall. Juveniles migrate along the edges
of rivers, where they are at risk of exposure to higher concentrations of pesticides from drift and
runoff. Their duration in shallow freshwater habitats increases their chances of higher exposure
to pesticides, contaminants, and elevated temperature.
Given the life history of SR Fall-run Chinook salmon, we expect the proposed uses of carbaryl
and carbofuran pesticide products may lead to individual fitness level consequences and
subsequent population level consequences. We expect that exposure to methomyl may lead to
individual fitness consequences, but not to an extent that would affect population growth rates.
Land class data indicate a high probability of exposure to SR Fall-run Chinook salmon. The risk
to this species' survival and recovery from the stressors of the action is high for carbaryl and
carbofuran, but low for methomyl.
Snake River Spring/Summer-run Chinook salmon
This ESU includes 32 historical populations in the Snake River basin which drains portions of
southeastern Washington, northeastern Oregon, and north/central Idaho. Historically, the
Salmon River system may have supported more than 40% of the total return of Spring/Summer-
run Chinook salmon to the Columbia system (Fulton 1968). The long-term trends in
productivity indicate a shrinking population. However, recent trends in productivity, buoyed by
the last five years, are approaching replacement levels. The annual population growth known for
18 populations ranged from 0.97 to 1.1. Historical populations above Hells Canyon are
considered extinct.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include degraded water quality in the freshwater estuary, tributaries, and
coastal habitats from land use activities and hydroelectric dams. Significant threats to SR
DRAFT 454
-------
DRAFT
Spring/Summer-run Chinook salmon include elevated temperature, water diversions, and poor
water quality.
Some spawning and rearing areas of the SR Spring/Summer Run Chinook are in forested areas,
where carbaryl may be applied. The likelihood of exposure in forested areas is unknown.
Registered 24(c) uses within this ESU include carbofuran application to potatoes and spinach
grown for seed in Washington and potatoes and sugar beets in Idaho. Dryland agriculture occurs
in lower Clearwater Basin, with crops including wheat, peas, and lentils. Alfalfa, hay, and
grasses are also grown in the lower Clearwater as well as the lower Salmon River basin. All
three a.i.s are registered for use on alfalfa. Agricultural activities, urban communities, and
industries are concentrated along the Snake River and near the mouths of major tributary valleys,
thus stream water quality and biological communities in the downstream portion of the upper
Snake River basin are degraded.
SR Spring/Summer-run Chinook salmon spawn at high elevations in the headwater tributaries of
the Clearwater, Grande, Ronde, Salmon, and Imnaha Rivers. Spawning is complete by the
second week of September. Eggs incubate and hatch in late winter and early spring of the
following year. The fry typically overwinter in shallow off-channel habitats associated with their
natal rivers and streams. Juveniles become active seaward migrants during the following spring
as yearlings (Connor, Sneva et al. 2005).
Given the life history of SR Spring/Summer-run Chinook salmon, we expect the proposed uses
of carbaryl and carbofuran pesticide products that contaminate aquatic habitats will lead to
individual fitness level consequences and subsequent population level consequences. The uses
of these materials in some natal areas and along the migratory route of SR Spring/Summer-run
Chinook salmon, may cause acute lethality or temporary AChE inhibition. We expect that
exposure to methomyl may lead to individual fitness consequences, but not to an extent that
would affect population growth rates. The risk to this species' survival and recovery from the
stressors of the action is high for carbaryl and carbofuran, but low for methomyl.
DRAFT 455
-------
DRAFT
Upper Willamette River Chinook Salmon
The UWR Chinook salmon ESU includes all eight naturally spawned populations residing in the
Clackamas River and the Upper Willamette River above Willamette Falls. It also includes seven
artificial propagation programs. The population in the McKenzie River is the only population
that is naturally producing, and current estimates indicate a negative growth rate. Historically,
the Upper Willamette River supported large numbers (exceeding 275,000 fish) of UWR Chinook
salmon. Current abundance of natural-origin fish is estimated at less than 10,000.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include habitat loss due to blockages from hydroelectric dams and irrigation
diversions, and degraded water quality within the Willamette mainstem and the lower reaches of
its tributaries. Elevated water temperature also poses a significant threat to the status of UWR
Chinook salmon. Fifty pesticides were detected in streams that drain both agricultural and urban
areas. Forty-nine pesticides were detected in streams draining agricultural land, while 25 were
detected in streams draining urban areas. Ten of these pesticides, including carbaryl and
carbofuran, exceeded EPA criteria for the protection of freshwater aquatic life from chronic
toxicity (Wentz, Bonn et al. 1998). The cumulative impacts from these threats continue to affect
UWR Chinook salmon.
Based on the crop types, we expect that carbaryl and methomyl are commonly applied in the
Willamette Valley throughout the growing season. Existing stocks of carbofuran are also
expected to be applied throughout the growing season. Registered 24(c) uses in Oregon include
carbofuran application to potatoes, nursery stock, sugar beets, and watermelons. The Willamette
Basin is the largest agricultural area in Oregon. In 1992 the Willamette Basin accounted for 51%
of Oregon's total gross farm sales and 58% of Oregon's crop sales. About one-third of the
agricultural land is irrigated and most of it is adjacent to the mainstem Willamette River. Urban
developments are also located primarily in the valley along the mainstem Willamette River. We
expect carbaryl to be used throughout the year in urban and residential areas. Given that major
urban and agricultural areas are located adjacent to the mainstem Willamette, ESU exposure to
these pesticides is likely. We also expect that other AChE inhibiting pesticides will co-occur
DRAFT 456
-------
DRAFT
with carbaryl, carbofuran, and methomyl in the waters of the Willamette Valley and exacerbate
adverse effects from AChE inhibition.
Chinook salmon fry typically select shallow off-channel habitats associated with their natal
rivers and streams. Juveniles generally rear in freshwater for several months to more than one
year before migrating to the ocean. Their duration in shallow freshwater habitats increases their
susceptibility to higher exposures of pesticides, contaminants, and elevated temperature. UWR
Chinook salmon exhibit an earlier time of entry into the Columbia River and estuary than other
spring Chinook salmon ESUs (Meyers, Kope et al. 1998). Although most juveniles from interior
spring Chinook salmon populations reach the mainstem migration corridor as yearlings, some
juvenile Chinook salmon in the lower Willamette River are sub-yearlings (Friesen, Vile et al.
2004). Off-channel habitats within the Willamette Valley floodplain are particularity important
for rearing fry and are actively being identified, reconnected and restored. We expect fry to be
exposed to the three insecticides when applications overlap with fry occurrence and will further
depress abundances of the available prey.
In conclusion we expect that the UWR Chinook salmon ESU will be compromised by reduced
lambdas of affected populations. The risk to this species' survival and recovery from the
stressors of the action is high.
Columbia River Chum Salmon
This ESU includes two remaining populations of 16 historical populations in the lower reaches
(the Lower Gorge tributaries and Gray's River) of the Columbia River. Thus, about 88% of the
historic populations are extirpated or nearly so. In the early 1900s, the run numbered in the
hundreds of thousands to a million returning adults. The size of the Lower Gorge population is
estimated at 400-500 individuals, down from a historical level of greater than 8,900 (Good,
Waples et al. 2005). Previous estimates of the Gray's River population range from 331 to 812
individuals. However, the population increased in 2002 to as many as 10,000 individuals (Good,
Waples et al. 2005). Overall, the lambda values indicate a long-term downward trend at 0.954
and 0.984, respectively.
DRAFT 457
-------
DRAFT
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections are hydromodification and habitat loss. Of the salmonids, chum salmon are
most averse to negotiating obstacles in their migratory pathway. Thus, they are more highly
impacted by the Columbia River hydropower system - specifically the Bonneville Dam
(Johnson, Grant et al. 1997). The water quality in the lower Columbia River is poor. Recent
USGS studies have demonstrated the presence of 25 pesticide compounds in surface waters,
including carbaryl (Ebbert and Embry 2001). Although the habitat restoration project for the
Gray's River will likely provide some benefit to the population, we are unable to quantify the
overall net effect for salmonids at this time.
Land use data indicate that the Columbia River chum salmon may be at risk of pesticide
exposure. In addition to general uses, registered 24(c) uses in Oregon include carbofuran
application to potatoes, nursery stock, sugar beets, and watermelons. The locations of high-
pesticide use areas and the preferential use of river-edge habitat by chum salmon indicate that the
species is at risk of pesticide exposure. The developed area surrounding the cities of Portland
and Vancouver occurs along the migratory route of the Lower Gorge chum.
Columbia River chum salmon fry emerge between March and May and emigrate shortly
thereafter to nearshore estuarine environments (Salo 1991). This is in sharp contrast to other
salmonid behavior and indicates that chum salmon are less dependent on freshwater conditions
for survival. After emergence, juvenile Columbia River chum salmon spend around 24 days
feeding in the estuary. Adults return to spawn in the lower reaches of the Columbia River
between the ages of two and five from mid-October through December. An average often days
is spent in the freshwater by the spawning adults.
Given the life history of the Columbia River chum salmon, we expect that the proposed uses of
carbaryl, carbofuran, and methomyl pesticide products may lead to individual fitness level
consequences. We expect that exposure will occur, resulting in both acute lethality and sublethal
olfactory-mediated effects. However, we do not expect these effects to occur at a scale that
would have population level effects. As chum fry are more precocious and quickly leave natal
streams, they are less reliant on the local invertebrate population. Given the life history of
DRAFT 458
-------
DRAFT
Columbia River chum salmon, the risk to this species' survival and recovery from the proposed
stressors of the action is low.
Hood Canal Summer-run Chum Salmon
This ESU includes 16 historical, naturally spawned populations of summer-run chum salmon in
Olympic Peninsula Rivers between Hood Canal and Dungeness Bay, Washington, as well as
eight artificial propagation programs. Of the historically existing populations, seven are believed
to be extirpated. Most of the extirpated populations occurred on the eastern side of the canal.
Only two of the remaining populations have long-term trends above replacement; long-term
lambda values of the nine existing populations range from 0.85 to 1.39 (Good, Waples et al.
2005). The Hood Canal Summer-run chum salmon populations are the subject of an intense
hatchery program intended to bolster numbers. As much as 60% of the spawning populations are
hatchery-raised fish.
The major threat to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections is habitat degradation. The combined effects of degraded floodplains,
estuarine, and riparian habitats, along with reduced stream flow and sedimentation, have had a
profound negative impact on this ESU.
The land use and environmental data indicate that the Hood Canal Summer-run chum may be
exposed to carbaryl, carbofuran, and methomyl. There is no cultivated crop land and less than
6% of the ESU is developed. Washington's 24(c) registrations for carbofuran use are not
expected to be employed in this region. We do expect some exposure due to residential uses.
The Hood Canal Summer-run chum spawn from mid-September through mid-October (Tynan
1997). Emergence generally occurs from early February through mid April. Upon emerging, fry
immediately commence downstream migration to estuaries (Tynan 1997). Upon arrival in the
estuary, salmon fry inhabit nearshore areas in shallow water. In Puget Sound, they have been
observed to reside in the top 6 inches of surface water and extremely close to the shoreline
(Tynan 1997). This behavior increases the likelihood of acute exposure to drift and runoff events.
DRAFT 459
-------
DRAFT
Given the life history of the Hood Canal Summer-run chum, we expect the proposed uses of
carbofuran, carbaryl, and methomyl pesticide products may contaminate aquatic habitats and
lead to individual fitness level consequences. We expect that exposure will occur, resulting in
lethality and sublethal olfactory-mediated effects. We do not, however, expect that these effects
will have population level consequences. As chum fry are more precocious and quickly leave
natal streams, they are less reliant on the local invertebrate population. Therefore, risk to this
species' survival and recovery from the stressors of the action is low.
Central California Coast Coho Salmon
The CCC coho salmon ESU includes 11 historical independent populations within counties from
Mendocino to Santa Cruz in California. Coho populations in three larger watersheds, as well as
some in smaller watersheds, have been extirpated or are nearly so. Historical escapement has
been estimated between 200,000 and 500,000 fish. Current escapements are not known, though
a minimum of 6,570 adult coho salmon are estimated to return to coastal streams within the ESU.
Long-term population trends do not exist for any of the populations in this ESU. More fish enter
northern streams but variation in abundance between cohorts can be large with one cohort often
dominating. Southern streams produce few naturally spawned fish of all cohorts.
CCC coho salmon populations have been adversely affected by loss of riparian cover, elevated
water temperatures, alteration of channel morphology, loss of winter habitat, and siltation. High
water temperatures prevent coho salmon from inhabiting several streams within the ESU.
Pesticides are expected to enter rivers as drift during application in agricultural areas. Highly
contaminated runoff into the Russian River, San Francisco Bay, and into rivers south of the
Golden Gate Bridge is expected during the first fall storms. We expect that application of other
AChE inhibiting pesticides will co-occur with carbaryl, carbofuran, and methomyl in the waters
within the ESU and exacerbate adverse effects from AChE inhibition.
The majority of agricultural land use is concentrated in the Russian River watershed and
watersheds south of the Golden Gate Bridge. High density urban development and urban centers
occur in the San Francisco Bay and in the Russian River basin. All three a.i.s are expected to be
applied to several crops within the Russian River basin and the Santa Cruz stratum during the
DRAFT 460
-------
DRAFT
growing season. Carbaryl is expected to be used within urban/residential areas throughout the
entire year.
The movements of mature adults are influenced by stream flow; entry often occurs during the
first large winter storms. Similarly, CCC coho salmon adults must wait in the lower river for
sufficient flow to be able to reach spawning grounds. Fry emerge in spring and remain in the
stream for up to 18 months. Newly emerged fry use backwater, side channels, and shallow
channel edges. During winter, juveniles inhabit side channels, sloughs, backwater, and other
protected channel features. The off-channel floodplain habitats provide feeding and growth
opportunities that are important before smoltification and seaward migration in the spring. The
presence of discrete brood years (BYs) makes coho salmon more vulnerable to environmental
perturbations than other salmonids because a failed BY is unlikely to be replaced by other BYs.
Land use data indicates substantial overlap between high-use areas and fish runs in the Russian
River, San Francisco Bay, and the Santa Cruz area. In several streams, one or more BYs are at
the verge of extinction, and heavy pesticide exposure may result in the loss of a BY. We expect
the proposed use of carbaryl, carbofuran, and methomyl pesticide products will lead to both
individual fitness level consequences and subsequent population level consequences. Therefore,
the risk to this species' survival and recovery from the proposed action is high.
Lower Columbia River Coho Salmon
The LCR coho salmon ESU includes all naturally spawned coho salmon populations in streams
and tributaries to the Columbia River in Washington and Oregon from the mouth of the
Columbia up to and including the White Salmon and Hood rivers, and along the Willamette to
Willamette Falls, Oregon. The ESU includes 26 historical populations and 25 artificial
propagation programs. Over 90% of the historic populations of LCR coho salmon are
considered extirpated. Most populations have very low numbers and have been replaced by
hatchery production. Only two populations have a degree of natural spawning - the Sandy River
and the Clackamas River. The annual population growth rate known for the Sandy River and
Clackamas River are 1.102 and 1.028, respectively.
DRAFT 461
-------
DRAFT
The major threats to LCR coho salmon identified in the Status of Listed Resources and
Environmental Baseline sections include reduced water flow in the mainstem and estuary from
irrigation diversions and hydroelectric dams. Additionally, degraded water quality in freshwater
and tributary habitats negatively affects this ESU. Among the various types of habitat threats,
elevated temperature, water diversions, and poor water quality have significant influence on the
status of LCR coho salmon.
Pesticide use and detections in LCR coho salmon freshwater habitats are well documented.
NAWQA sampling in surface waters within the ESU range detected more than 50 pesticides in
streams within this ESU. Ten pesticides exceeded EPA's criteria for the protection of aquatic
life from chronic toxicity, including carbaryl. The cumulative impacts from these multiple
threats continue to affect this ESU.
Registered uses of carbaryl, carbofuran, and methomyl include applications to crop agricultural
sites, residential sites, and urban sites. Registered 24(c) uses in Oregon include carbofuran
application to potatoes, nursery stock, sugar beets, and watermelons. Agricultural and urban
development occurs along the Columbia River and the basins of its major tributaries. Both the
Sandy and Clackamas Rivers drain parts of the Willamette Basin - a major agricultural area.
Thus, ESU exposure to these pesticides is likely.
LCR coho salmon enter freshwater from August through December. Coho salmon spawn in
November and December: emergence from the redds occurs between March and July. Fry
typically select off-channel habitats associated with their natal rivers and streams to rear. The
juvenile coho salmon reside in shallow freshwater habitats for more than one year. The long
residence in these habitats increases their likelihood of experiencing significant exposure to
pesticides and other contaminants.
Given the life history of LCR coho salmon, we expect the proposed uses of carbaryl, carbofuran,
and methomyl pesticide products may lead to both individual fitness level and subsequent
population level consequences. The risk to this species' survival and recovery from the stressors
of the action is high.
DRAFT 462
-------
DRAFT
Southern Oregon/Northern California Coast coho salmon
The SONCC coho salmon ESU includes coho salmon in streams between Cape Blanco, Oregon,
and Punta Gorda, California, and three artificial propagation programs in the Klamath, Trinity,
and Rogue Rivers. Little information on abundance trends exists for the streams within the ESU.
However, available information indicates that Eel River and Southern populations are at high
risk from critically low abundances. Northern populations may have larger runs. Recent
estimated escapement in the Klamath River is about 2,000 fish. The Rogue River spawner run
ranged from 7,800 to 12,213 coho salmon from 1999 to 2001, though many are of hatchery
origin.
The threats to this ESU include road crossings and other migration barriers, timber harvest and
agricultural activities, and water management. Adverse effects on the SONCC coho salmon
consist of barriers that limit access to tributaries, lack of large instream woody debris, reduced
riparian vegetation, and elevated water temperature. The Klamath River, one of the largest
rivers, is 303(d) listed because aquatic habitat has been degraded due to excessively high water
temperatures and algae blooms associated with high nutrient loads, water impoundments, and
agricultural water diversions (USEPA 1993). Pesticides used in land management activities are
expected to enter stream mainstems and tributaries during application and as runoff.
A large portion of the agricultural and municipal land uses within this ESU is concentrated in the
upper Klamath and Rogue Rivers. A diverse array of crops is produced in these areas each year.
Some agriculture and smaller urban centers are dispersed on the lower alluvial coastal plains and
in rivers valleys of coastal rivers. Active forest management occurs throughout the watersheds
within this ESU, and application of pesticide products is anticipated.
Spawning occurs from November through January, depending on the occurrence of fall and
winter storms. Fry incubate for four to eight weeks before emergence in the spring. Newly
emerged fry rear in backwater, side channels, and shallow channel edges for up to 18 months.
During winter, juveniles inhabit side channels, sloughs, backwater, and other protected channel
features. The off-channel and floodplain habitats provide feeding and growth opportunities that
are important before smoltification and seaward migration in spring. The three-year life cycle of
DRAFT 463
-------
DRAFT
coho salmon, with limited exchange between cohorts, makes them more vulnerable to
environmental perturbations than other salmonids.
Agricultural and urban land uses overlap substantially with this ESU, particularly in the Interior
Rogue and Klamath rivers. Because of this overlap, we expect that the proposed uses of
carbaryl, carbofuran, and methomyl will result in both individual fitness and population level
effects. In several streams, one or more B Ys are at the verge of extinction. Exposure to the three
a.i.s can result in loss of weak BYs. Given the low population abundance levels and the discrete
cohort life history of the coho salmon, the risk to the survival and recovery of the SONCC ESU
from the stressors of the action is high.
Oregon Coast Coho Salmon
The OC coho salmon ESU includes 11 naturally spawned populations in Oregon coastal streams
south of the Columbia River and north of Cape Blanco and one hatchery stock. While none of
the populations have become extinct, the ESU's current abundance levels are less than 10% of
historic populations. OC coho salmon abundance estimates from 2000 to 2007 ranged from a
low of 51,875 to a high of 260,000 naturally produced spawners. Long-term trends in ESU
productivity remain strongly negative.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections are habitat degradation from logging, road construction, urban development,
mining, agriculture, and recreation n. Within the various types of habitat, elevated temperature,
water diversions, and poor water quality also affect the status of OC coho salmon. The
cumulative impacts from these multiple threats continue to affect OC coho salmon.
Crop rotation patterns and crop types influence the distribution and frequency of pesticides
within an area. Based on limited agricultural activities, we expect that carbaryl and existing
stocks of carbofuran may be applied on a limited scale throughout the growing season.
Registered 24(c) uses in Oregon include carbofuran application to potatoes, nursery stock, sugar
beets, and watermelons. We also expect carbaryl use within urban and residential areas
throughout the entire year. However, we expect no to very low applications of methomyl
DRAFT 464
-------
DRAFT
throughout the growing season. Although carbaryl and carbofuran may be applied to pine
seedlings, we expect no to low applications in these forests. As runoff from urban and
agricultural areas may drain into adjacent streams, ESU exposure to these pesticides is likely.
However, given the above land uses and expected low application of the three carbamates, we
expect lower incidences offish exposure to these compounds in surface waters following
application.
OC coho salmon enter rivers in September or October; spawning occurs in December. Fry
emerge between March and July, then move to shallow areas near the stream banks. Juvenile
coho salmon are often found in small streams less than five ft wide, and may migrate
considerable distances to rear in lakes and off-channel ponds. Generally, coho salmon spend 18
months rearing in freshwater before moving out into the ocean. Given this duration spent in
shallow freshwater habitats, they are more likely to experience higher pesticide exposure and
contaminants.
Given the limited applications of the three carbamates and low incidences within land uses that
overlap with the range of OC coho, effects to individual fitness may occur. However, we do not
expect that exposure will have effects at the population level. Therefore, the risk to this species'
survival and recovery from the stressors of the action is low.
Ozette Lake Sockeye Salmon
This ESU is made up of only one historic population. Natural spawning aggregations remain on
two beaches of Ozette Lake. Two tributary spawning groups were initiated in!992 through
hatchery programs. Peak run size in the 1940s has been estimated to be between 3,000 and
18,000 fish, and actual production (i.e., including harvest) may have been as high as 50,000.
Recent estimates put the population at 3,600 - 4,600 individuals (Haggerty, Ritchie et al. 2007).
The supplemental hatchery program began with out-of-basin stocks and make up an average of
10% of the run. The proportion of beach spawners originating from the hatchery is unknown but
likely low. Uncertainty in past population counts coupled with poorly documented historical
abundance prevents calculation of population growth rates and trends.
DRAFT 465
-------
DRAFT
Major threats to this population identified in the Status of Listed Resources and Environmental
Baseline sections are siltation of spawning habitat from logging activities within the watershed
and genetic effects from past interbreeding with kokanee. Almost 80% of the land cover for this
ESU is evergreen forest. Between 1940 and 1984, 85% of the basin was clear-cut logged (Blum
1988). Roughly 77% of the land in Ozette Basin is managed for timber production (Jacobs,
Larson et al. 1996). The extent to which pesticide products are currently used by these
companies is unknown.
Ozette Lake is in a sparsely populated area, with less than 1% of land developed. No crop land
was identified in NLCD data (Table 32). The land use and environmental data indicate that the
Ozette Lake sockeye salmon may be exposed to carbaryl or carbofuran if applied in the
watershed. Carbaryl is registered for a number of non-agricultural applications which may take
place in forested and residential areas. Additionally, the area has a small population, making
residential use of carbaryl within the ESU a possibility. We do not expect 24(c) registrations of
carbofuran will be used within this ESU, though it may be applied to pine seedlings. However,
there are few data available on use and no monitoring data are currently available. We do not
expect methomyl to be used within the boundaries of this ESU.
Ozette Lake sockeye salmon enter the lake between April and August, and spawning occurs late
October through February. Fry emerge from gravel redds in the spring and emigrate to the open
waters of the lake where they remain for a full year. They then smolt as 1-year olds and migrate
to the open ocean. The majority of Ozette Lake sockeye salmon return to spawn as four-year old
fish after spending two full years at sea.
Given the life history of the Ozette Lake sockeye, we expect that that the proposed uses of
carbaryl and carbofuran pesticide products may contaminate aquatic habitats used by sockeye in
a way that might lead to individual fitness level consequences. While the uses of these materials
may lead to some overlap with the Ozette Lake sockeye, the existing and likely future land uses
should limit the applications of carbaryl, carbofuran, and methomyl containing pesticides.
Consequently, the risk posed by the proposed action to Ozette Lake sockeye salmon's survival
and recovery is low.
DRAFT 466
-------
DRAFT
Snake River Sockeye Salmon
The SR sockeye salmon ESU historically includes populations in five Idaho lakes as well as
artificially propagated sockeye salmon from the Redfish Lake Captive Broodstock Program.
Only one hatchery-sustained population remains and is found in Redfish Lake. This population
is listed as endangered and has an extremely high risk of extinction. Current smolt-to-adult
survival of sockeye originating from the Stanley Basin lakes is rarely greater than 0.3% (Hebdon,
Kline et al. 2004). No natural origin adults have returned to Redfish Lake to spawn since 1998;
the population is maintained entirely by propagation efforts. Around 30 fish of hatchery origin
return to spawn each year (FCRPS 2008).
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include impaired tributary flow and passage, migration barriers, degraded
water quality, and hydromodification of the Columbia and Snake Rivers. Like the Ozette Lake
ESU, the SR sockeye occupy a relatively undeveloped area with very little cropland (Table 370
However, the SR sockeye have the longest migration of any sockeye salmon, traveling 900 miles
inland. These waters are contaminated by drift and runoff from both agricultural and urban
areas. Exposure during migration likely adds to the low survivorship of smolts. The land use
and environmental data indicate that the SR sockeye may be exposed to carbaryl, carbofuran, and
methomyl during migration. Registered 24(c) uses within this ESU include carbofuran
application to potatoes and spinach grown for seed in Washington and potatoes and sugar beets
in Idaho.
Historically, sockeye salmon entered the Columbia River system in June and July, and arrived at
Redfish Lake between August and September (FCRPS 2008). Spawning occurred in lakeshore
gravel and generally peaked in October. Fry emerged in the spring (April and May) then
migrated to open waters of the lake to feed. Juvenile sockeye remained in the lake for one to
three years before migrating through the Snake and Columbia Rivers to the ocean. Adult
sockeye spent two or three years in the open ocean before returning to Redfish Lake to spawn.
During adult and juvenile migrations the sockeye are at their greatest risk of exposure to the
stressors of the action. Sockeye salmon making the 900 mile journey each way pass along many
DRAFT 467
-------
DRAFT
miles where agricultural crops are at the river's edge. Drift and runoff occurring in conjunction
with sockeye salmon migration is expected to cause adverse effect.
Given the life history of the SR sockeye, we expect the proposed uses of carbaryl, carbofuran,
and methomyl pesticide products may lead to individual fitness level consequences. However,
we do not expect fry to be exposed to the a.i.s, although some exposure may occur during adult
migration and lead to acute lethality or temporary AChE inhibition. However, this exposure is
not expected to occur at a frequency that would cause effects at the population level. Therefore,
the risk to this species' survival and recovery from the stressors of the action is low.
Central California Coaststeelhead
The CCC steelhead DPS includes all naturally spawned steelhead in streams from the Russian
River south to Aptos Creek. This area includes streams entering the San Francisco Bay. The
DPS consists of nine historic independent populations. There is limited information on the
abundance of these populations, but all are in severe decline.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections are dams and other migration barriers, urbanization and channel modification,
agricultural activities, predators, hatcheries, and water diversions. Throughout the species'
range, habitat conditions and quality have been degraded by a lack of channel complexity,
eroded banks, turbid and contaminated water, low summer flow and high water temperatures, an
array of contaminants found at toxic levels, and restricted access to cooler head waters from
migration barriers. There is limited monitoring data from streams within this DPS. The
cumulative impacts of these threats continue to affect the CCC steelhead.
Crop farming is concentrated in low laying areas and floodplains along the estuaries and stream
valleys of the Russian River and Santa Cruz drainage basins. Christmas trees production occurs
in these areas in addition to a variety of food crops. Carbaryl and methomyl are registered for a
variety crops grown in the Russian River valley and in the San Mateo and Santa Cruz counties.
Registered 24(c) uses for carbaryl in California include application to fruits and nuts, prickly
pear cactus, ornamental plants, and non-food crops. Methomyl has 24(c) registrations for insect
DRAFT 468
-------
DRAFT
control on ornamentals, beans, soybeans, radishes, sweet potatoes, Chinese broccoli, broccoli
raab, and pumpkins. The a.i.s are therefore expected to be used throughout the growing season.
Existing stocks of carbofuran are also expected to be applied throughout the growing season.
Further, 16% of the DPS' range consists of developed urban and residential land, mostly within
the San Francisco Bay and Russian River basin. Carbaryl is expected to be used the entire year
within the urban/residential areas.
The CCC steelhead populations are all of the winter-type. Fry emerge in spring and rear in
smaller tributaries and off-channel habitats. During periods with high flows, they move into
floodplain habitat. Juvenile steelhead remain in freshwater for one or more years before
migrating downstream to smolt.
The registered uses of the three carbamates indicate substantial overlap with the CCC steelhead
populations, especially in the San Francisco Bay, Russian River, and Santa Cruz area. We
expect that the proposed uses of carbaryl, carbofuran, and methomyl may contaminate spawning
and rearing habitats and lead to both individual and population level consequences. Thus, the
risk to this species' survival and recovery from the stressors of the proposed action is high.
California Central Valley steelhead
The CCV steelhead DPS includes all naturally spawned steelhead in the Sacramento River, San
Joaquin River, and their tributaries. This area includes streams entering the Sacramento-San
Joaquin Delta (Delta) east of Chipps Island. All populations rear and migrate through the
Sacramento River, San Joaquin River, Delta, and San Francisco Bay. The current distribution is
severely reduced and fragmented compared to historical distributions. About 6,000 river miles
of river habitat have been reduced to 300 miles. Historical returns within the DPS may have
approached two million adults annually. Current annual run size for the entire Sacramento-San
Joaquin system today is estimated at less than 10,000 returning adults.
The Status of Listed Resources and Environmental Baseline sections indicate that dams and other
migration barriers, urbanization and channel modification, agricultural activities, introduced non-
DRAFT 469
-------
DRAFT
native predators, hatcheries, and large scale water management and diversions negatively affect
this DPS. Steelhead habitat has been highly degraded by reduced channel complexity, eroded
banks, increased water temperature, migration barriers restricting access to cooler head waters,
and decreased water quality from contaminants. Numerous NAWQA, CDPR, and other
assessments found high concentration of contaminants in both the San Joaquin and Sacramento
Rivers and their tributaries. In the San Joaquin Basin, seven pesticides exceeded EPA criteria for
aquatic life. These pesticides include diuron, trifluralin, azinphos-methyl, carbaryl, chlorpyrifos,
diazinon, and malathion. The cumulative impacts from these threats continue to affect the CCV
steelhead.
Approximately 27% of the DPS is developed for cultivation of crops (Table 38). High densities
of crop farming occur throughout the San Joaquin Basin, in the Sacramento-San Joaquin Delta,
and along lower Sacramento River. Further, 9.2% of the DPS consists of urban development.
Based on the crop types, we expect carbaryl, carbofuran, and methomyl are commonly applied in
the Central Valley throughout the growing season. In urban and residential areas, we expect
carbaryl to be applied throughout the year. Monitoring in the San Joaquin basin found 48 of 83
pesticides tested for. Seven pesticides (diuron, trifluralin, azinphos-methyl, carbaryl,
chlorpyrifos, diazinon, and malathion) exceeded criteria for the protection of aquatic life.
Pesticides detected in the Sacramento River included thiobencarb, carbofuran, molinate,
simazine, metolachlor, dacthal, chlorpyrifos, carbaryl, and diazinon. Carbofuran was most often
detected in agricultural sites while carbaryl was also found in all urban sites tested.
All of the steelhead populations within this DPS exhibit the winter-type life history, though
detailed information about the CCV steelhead life history is not available. Juvenile steelhead
remain in freshwater for one or more years before migrating downstream to enter the ocean. The
CCV steelhead use the lower Sacramento River and the Delta for rearing and as migration
corridor. Some may utilize tidal marshes, non-tidal freshwater marshes, and other shallow areas
in the Delta for rearing areas for short periods during outmigration to the ocean.
The registered uses of the three carbamates indicate substantial overlap with the CCV steelhead
populations, especially in the lower Sacramento River and Delta. We expect that the proposed
DRAFT 470
-------
DRAFT
uses of carbaryl, carbofuran, and methomyl may contaminate spawning and rearing habitats and
lead to both individual and population level consequences. Thus, the risk to this species'
survival and recovery from the stressors of the proposed action is high.
Lower Columbia River Steelhead
The LCR Steelhead DPS includes 23 historical, naturally-spawned steelhead populations in
Columbia River tributaries on the Washington side between the Cowlitz and Wind Rivers in
Washington and on the Oregon side between the Willamette and Hood Rivers. Historical counts
from the Cowlitz, Kalama, and Sandy Rivers suggest the ESU probably exceeded 20,000 fish.
During the 1990s, fish abundance dropped to 1,000 to 2,000 fish. Many of the populations in
this DPS are small, and the long- and short-term trends in abundance of all individual
populations are negative. The annual population growth rate known for nine independent
populations ranged from 0.945 to 1.06.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections include dams; water diversion; destruction or degradation of riparian habitat;
and land use practices (logging, agriculture, and urbanization). Tributary hydropower
development has created barriers and reduced fish access, ultimately affecting the spatial
structure within the independent populations. Within the various types of habitat, water
diversions, elevated temperature, and poor water quality also affect the status of LCR steelhead.
Pesticides have also been detected in LCR steelhead habitats. NAWQA sampling from 1991-
1995 in surface waters within the DPS range detected more than 50 pesticides in streams. Ten
pesticides exceeded EPA's criteria for the protection of aquatic life from chronic toxicity. These
pesticides include carbaryl, carbofuran, atrazine, chlorpyrifos, and malathion. The cumulative
impacts from these multiple threats continue to affect this DPS.
Most of the highly developed land and agricultural areas in this DPS's range are adjacent to
salmonid habitat. Registered uses of carbaryl, carbofuran, and methomyl include applications to
crop agricultural sites, residential sites, and urban sites. Registered 24(c) uses in Oregon include
carbofuran application to potatoes, nursery stock, sugar beets, and watermelons. Based on the
crop types, we expect that carbaryl and methomyl are commonly applied in the Lewis River
DRAFT 471
-------
DRAFT
subbasin throughout the growing season and within urban and residential areas throughout the
entire year for carbaryl. Existing stocks of carbofuran are also expected to be applied throughout
the growing season. Thus, DPS exposure to these pesticides is likely.
This DPS includes winter- and summer-run types. Summer-run steelhead return to freshwater
between May and November. They enter the Columbia River in a sexually immature condition
and require several months in freshwater before spawning. Winter-run steelhead enter
freshwater from November to April. These fish are close to sexual maturation and spawn shortly
after arrival in their natal streams. Steelhead fry typically rear in off-channel habitats associated
with their natal rivers and streams for more than a year. Given this duration, juveniles are likely
to experience pesticide exposure.
Given the life history of LCR steelhead, we expect the proposed uses of carbaryl, carbofuran,
and methomyl pesticide products may contaminate rearing habitats and lead to individual fitness
and subsequent population level consequences. The risk to this species' survival and recovery
from the stressors of the action is high.
Middle Columbia River Steelhead
The MCR steelhead DPS includes 19 populations in Oregon and Washington subbasins upstream
of the Hood and Wind River systems to and including the Yakima River. Historical run
estimates for the Yakima River imply that annual species abundance may have exceeded 300,000
returning adults (Busby, Wainwright et al. 1996), whereas only 1,000 - 4,000 currently spawn.
The most recent 10-year period indicated trends in abundance were positive for approximately
half of the independent populations and negative for the remainder. Growth rates ranged from
0.97 to 1.02. Two historical populations are considered extinct.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections include barriers preventing steelhead migration above dams and fish
mortalities from the Columbia River hydroelectric system. Additionally, agricultural practices,
especially grazing, water diversions, and withdrawals, negatively affect this DPS. Elevated
temperature and poor water quality from contaminants also impact the status of MCR steelhead.
DRAFT 472
-------
DRAFT
In the Yakima River, 72 streams and river segments are listed as impaired waters and 83%
exceed temperature standards. Pesticide use and detections are also well document in MCR DPS
waters. Within the Yakima River Basin, 76 pesticide compounds were detected. They include
38 herbicides, 17 insecticides (such as carbaryl, diazinon, chlorpyrifos, and malathion), 15
breakdown products, and 6 others. The median and maximum numbers of chemicals in a
mixture were six and eight, respectively (Fuhrer, Morace et al. 2004). Many AChE-inhibiting
pesticides were also detected. They include azinphos-methyl, dimethoate, ethoprop, disulfuton,
aldicarb, aldicar sulfone, and carbaryl. In the Granger drainage of the lower Yakima, atrazine,
simazine, chlorpyrifos, diazinon, and malathion were also detected. The co-occurrence of
atrazine with carbaryl, and other OPs in aquatic habitats increases the likelihood of adverse
responses in salmonids and their aquatic prey. The cumulative impacts from these multiple
threats continue to affect MCR steelhead.
Agricultural development is high within the range of MCR steelhead. Based on crop types and
allowable uses, we expect that carbaryl and methomyl are commonly applied in the Yakima
River basin, Walla Walla River basin, Deschutes River basin, and the John Day River basin
throughout the growing season. Existing stocks of carbofuran are also expected to be applied
throughout the growing season. In urban and residential areas, carbaryl will likely be applied
throughout the entire year. Registered 24(c) uses within this ESU include carbofuran application
to potatoes, nursery stock, sugar beets, and watermelons in Oregon and to potatoes and spinach
grown for seed in Washington. Given the high amount of agriculture in these areas, we also
expect other AChE inhibiting pesticides will co-occur with carbaryl, carbofuran, and methomyl
and exacerbate adverse effects from AChE inhibition. Thus, DPS exposure to these pesticides is
expected to be high
Mature adults (three to five years old) may enter rivers any month of the year and spawn in late
winter or spring. Swim-up fry usually inhabit shallow water along banks of streams or aquatic
habitats on stream margins. Steelhead rear in a variety of freshwater habitats and most remain in
freshwater for two to three years. Some individuals, however, have stayed for as many as six to
seven years. Most MCR steelhead smolt at two years and spend one to two years in the ocean
prior to re-entering the freshwater to spawn.
DRAFT 473
-------
DRAFT
Intense agricultural development has a high degree of overlap with the spawning and rearing
areas of the MCR steelhead DPS. Given the long residency periods in freshwater, we expect that
the proposed uses of carbaryl, carbofuran, and methomyl may lead to both individual and
population level consequences. Therefore, the risk to this species' survival and recovery from
the stressors of the action is high.
Northern California steelhead
The NC steelhead DPS includes all naturally spawned steelhead in California coastal river basins
from Redwood Creek southward to, but not including, the Russian River, as well as two artificial
propagation programs. Historical estimates of annual production for this DPS are upwards of
200,000 fish. Exact information on abundance is lacking for most of the streams within the DPS,
though most populations are in decline. The DPS includes 15 historically independent
populations of winter steelhead and 10 populations of summer steelhead. None of the winter
steelhead populations are viable due to low abundance and production.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections are timber harvest, forest roads, vineyard developments, and road crossings
and dams. Stressors to the NC steelhead include lack of large instream woody debris, reduced
riparian vegetation, elevated water temperature, increased predation, and barriers that limit
access to tributaries. The cumulative impacts from these multiple threats continue to affect the
NC steelhead.
About 85% of the DPS consists of forests and chaparral (Table 26). Pesticide exposure within
the range of this DPS is expected to be low. There are areas of Mendocino County with a high
density of vineyards, and there is some crop farming in low laying areas and floodplains along
the estuaries and lower reaches of coastal streams. Coastal communities are located at the
mouths of several streams within the NC steelhead DPS. Based on the crop types, we expect
carbaryl, carbofuran, and methomyl to be commonly applied throughout the growing season in
these areas. Carbaryl is expected to be used the entire year within urban and residential areas.
However, because agricultural and urban development is not concentrated in any one basin, the
DRAFT 474
-------
DRAFT
risk of exposure to any given population is decreased. Additionally, the majority of steelhead
juveniles rear at higher elevations in the watersheds with less agricultural influence. The fry,
then, are less likely to be exposed during critical life stages.
The winter-type steelhead enters rivers as mature adults between November and April to spawn.
The summer-type enters the stream in immature condition between May and October. Early
arriving summer steelehead move high into the upper watersheds where they hold in deep pools
throughout the summer before spawning in fall. In northern California, juvenile steelhead
remain in freshwater for two or more years before migrating downstream to smolt. Juvenile
steelheads spend a variable amount of time in the estuary before entering the open ocean.
We expect the proposed uses of carbaryl, carbofuran, and methomyl pesticides products within
this DPS may lead to individual fitness effects, ranging from acute lethality to temporary AChE
inhibition. However, exposure of NC steelhead populations to the three carbamates will be
limited given the amount of overlap between agriculture, residential and urban development and
spawning and rearing habitat. As such, we do not expect exposure to result in population level
consequences. In conclusion, we expect the risk to survival and recovery of NC steelhead from
the proposed action is low.
Puget Sound Steelhead
The Puget Sound steelhead is comprised of 21 populations, some of which have both summer
and winter runs. Steelhead occur in all major watersheds within the Sound and out to the Elwha
River. Of the winter runs, 17 had declining and four had increasing population trends. Summer
run abundance trends are not available. Estimated run size for Puget Sound steelhead in the
early 1980s is approximately 100,000 winter-run steelhead and 20,000 summer-run steelhead.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections are habitat degradation from logging, road construction, urban development,
mining, agriculture, and recreation; water diversions; and poor water quality. In particular,
elevated temperature, water diversions, and poor water quality have the most significant
influences on the status of Puget Sound steelhead. Furthermore, there has been extensive
DRAFT 475
-------
DRAFT
urbanization in this region. Well over two million people inhabit the area, with most
development occurring along rivers and coastline.
Pesticide use and detections in the DPS's watershed are also well documented. 2006 NAWQA
sampling in the Puget Sound basin detected 26 pesticides and 74 other synthetic organic
chemicals in streams and rivers, with different mixtures of chemicals linked to agricultural and
urban settings. Urban streams sampled in Puget Sound showed the highest detections for
carbaryl, diazinon, and malathion. Diazinon was also frequently detected in urban streams at
concentrations that exceeded EPA guidelines for protecting aquatic life (Bortleson and Ebbert
2000).
Registered uses of carbaryl, carbofuran, and methomyl include applications to crop agricultural
sites, residential sites, and urban sites. In addition, Washington has 24(c) registrations for
carbofuran use on spinach grown for seed and potatoes. Land classes indicate high use of the
three a.i.s in the following drainage basins: Nooksack, Skagit River, Stillaguamish, Skykomish,
Snoqualimie, Snohomish, Lake Washington, Duwamish, and Puyallup. Roughly 7% of the
lowland areas (below 1,000 ft elevation) in the Puget Sound region are covered by impervious
surfaces, which increase urban runoff containing pollutants and contaminants into streams.
Pollutants carried into streams from urban runoff include pesticides, heavy metals, PCBs,
PBDEs, PAHs, Pharmaceuticals, nutrients, and sediments. Thus, we expect salmonid
populations within the ESU to be exposed to carbaryl, carbofuran, and methomyl. Further,
monitoring data shows that we can expect concurrent exposure with other AChE inhibiting
chemicals, leading to a pronounced severity of effects.
Summer steelhead enter freshwater between May and October, while winter steelhead enter
between November and April. Spawning generally occurs in late winter or spring. Immediately
after leaving the gravel, swim-up fry usually inhabit shallow water along banks of stream or
aquatic habitats on stream margins. Steelhead rear in a wide variety of freshwater habitats,
generally for two to three years. However, they may possibly reside up to six to seven years in
freshwater environments. Following the rearing period, they smolt and migrate to sea in the
spring.
DRAFT 476
-------
DRAFT
We expect the proposed uses of carbaryl, carbofuran, and methomyl pesticide products that
contaminate aquatic habitats may lead to both individual fitness level and subsequent population
level consequences. The risk to this species' survival and recovery from the stressors of the
action is high.
Snake River Basin Steelhead
The SR steelhead DPS includes 23 naturally spawned populations below impassable natural and
man-made barriers in streams in the Columbia River Basin upstream from the Yakima River,
Washington, to the U.S.-Canada border. The Snake River supports about 63% of the natural-
origin production of steelhead in the Columbia River Basin. The 10-year average for natural-
origin steelhead passing Lower Granite Dam between 1996 and 2005 is 28,303 adults. Annual
population growth rates show mixed long- and short-term trends in abundance and productivity.
The annual population growth rate is known for eight independent populations and range 0.89 to
1.08. One historical population is likely extirpated.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections include hydrosystem mortality, water diversions, excessive sediment, and
degraded water quality. Elevated temperature also impacts the status of SR steelhead. Pesticides
have been detected in SR steelhead freshwater habitats. NAWQA sampling in 1992-1995 in the
DPS's watersheds detected Eptam, atrazine, desethylatrazine, metolachlor, and alachlor.
Carbaryl and carbofuran were detected in only 1% of samples. The cumulative impacts from
these multiple threats continue to affect SR steelhead.
Some spawning and rearing habitat of the Snake River steelhead is in forested areas, where
carbaryl may be applied. However, the likelihood of exposure in forested areas is unknown.
Registered 24(c) uses within this ESU include carbofuran application to potatoes and spinach
grown for seed in Washington and potatoes and sugar beets in Idaho. Dryland agriculture occurs
in lower Clearwater Basin, with crops including wheat, peas, and lentils. Alfalfa, hay, and
grasses are also grown in the lower Clearwater as well as the lower Salmon River basin. All
three a.i.s are registered for use on alfalfa. Agricultural activities, urban communities, and
DRAFT 477
-------
DRAFT
industries are concentrated along the Snake River and near the mouths of major tributary valleys,
thus stream water quality and biological communities in the downstream portion of the upper
Snake River basin are degraded. Given that urban and agricultural areas are located adjacent to
streams in the Lower Clearwater River drainage, DPS exposure to these pesticides is likely.
Sexually immature adult Snake River summer steelheads enter the Columbia River from late
June to October. Adults migrate upriver until they reach Snake River tributaries where they
spawn between March and May of the following year. Emergence occurs by early June in low
elevation streams and as late mid-July at higher elevations. After hatching, juvenile SR
steelhead typically select off-channel habitats associated with their natal rivers and streams for
rearing. They spend two to three years in freshwater before they smolt and migrate to the ocean.
Juvenile steelhead are more likely to be exposed to pesticides and other contaminants because of
their long freshwater residency period.
Given the life history of SR steelhead, we expect the proposed uses of carbaryl and carbofuran
pesticide products that contaminate aquatic habitats will lead to individual fitness level
consequences and subsequent population level consequences. The uses of these materials in
some natal areas and along migratory routes may cause acute lethality, temporary AChE
inhibition, and reduced growth. We expect that exposure to methomyl may lead to individual
fitness consequences, but not to an extent that would affect population growth rates. The risk to
SR steelhead survival and recovery from the stressors of the action is high for carbaryl and
carbofuran, but low for methomyl.
South-Central California Coast steelhead
The S-CCC steelhead DPS includes all naturally spawned steelhead in streams from the Pajaro
River to the Santa Maria River. The major basins in the S-CCC steelhead range are the Pajaro
River system and the Salinas River system. Carmel River, a smaller basin, also supports a stable
run of steelhead. Historic abundance estimates for the DPS imply an annual return may have
been upwards of 20,000 fish. The estimated production in five of the major rivers indicates a
return of less than 500 adults.
DRAFT 478
-------
DRAFT
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections are dams and other migration barriers, urbanization and channel modification,
agricultural activities, and wildfires. These activities has resulted in reduced channel
complexity, eroded banks, increased water temperature, migration barriers restricting access to
cooler head waters, and decreased water quality from contaminants.
About 7% of the DPS is developed for cultivation of crops; a large portion of this land is in the
Salinas River valley. Crops are concentrated in low laying areas and floodplains along the
estuaries and lower reaches of streams. Based on the crop types, we expect carbaryl, carbofuran,
and methomyl are commonly applied within the DPS throughout the growing season.
Approximately 8% of the DPS has been developed for urban and residential purposes.
Developed areas are located at the mouth of several streams within the S-CCC steelhead DPS.
We expect carbaryl use in urban/residential areas throughout the entire year.
All S-CCC steelhead are a winter-run life history. Juvenile steelhead remain in freshwater for
one or more years before migrating downstream to smolt. Steelhead in this area consists of two
life history groups: one where the juveniles rear for one or two years mainly in freshwater and
another where the juveniles rear at the upper end of coastal lagoons for the first or second
summer.
In conclusion, based on the long freshwater residence time of steelhead and the considerable
overlap between S-CCC steelhead distribution and expected pesticide use, we expect that the S-
CCC may experience high levels of exposure. We expect that proposed uses of carbaryl,
carbofuran, and methomyl that contaminate spawning and rearing habitats may lead to both
individual and population level consequences. Therefore, the risk to the survival and recovery of
S-CCC from the proposed action is high.
Southern California Steelhead
The SC steelhead DPS includes populations in five major and several small coastal river basins
in California from the Santa Maria River to the U.S. - Mexican border. It is estimated that the
species' current distribution constitutes about 1% of the historical distribution. Current
DRAFT 479
-------
DRAFT
abundance is considerably reduced with an estimated escapement of 500 fish for four of the
larger rivers. Long-term estimates and population trends are lacking for the streams within the
DPS.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections are dams and other migration barriers, urbanization and channel modification,
agricultural activities, and wildfires. As a result of these activities, the stream substrate contains
a high proportion of fines, stream channels lack complexity, banks are eroding, migration
barriers restrict fish access to cooler head waters and tributaries, and the water is turbid and
contaminated.
Farming is concentrated in low laying areas and floodplains along the estuaries and lower
reaches of streams. Carbaryl, carbofuran, and methomyl are expected to be commonly used for
several crops during the growing season. Large areas of dense development exist in the counties
of Santa Barbra to San Diego. Carbaryl is expected to be used within urban/residential areas
throughout the entire year. The NAWQA analysis detected more than 58 pesticides in ground
and surface waters within the heavily populated Santa Ana basin, including multiple AChE
inhibitors. The invertebrate community in the basin is heavily altered by pesticide influence.
Carbaryl was detected in 42% or urban samples but few detections exceeded standards for
protection of aquatic life. Carbofuran was also detected but not in levels exceeding standards.
The application of other AChE inhibiting pesticides is expect to co-occur with the a.i.s, and
exacerbate adverse effects from AChE inhibition.
All steelhead populations in this DPS are winter-type. Juvenile steelheads remain in freshwater
for one or more years before migrating downstream to smolt. SC steelhead consists of two life
history groups: one where the juveniles rear for one or two years mainly in freshwater and
another where the juveniles rear at the upper end of coastal lagoons for the first or second
summer.
We expect the proposed uses of carbaryl, carbofuran, and methomyl pesticides products that
contaminate aquatic habitat will lead to both individual fitness level consequences and
DRAFT 480
-------
DRAFT
subsequent population level consequences. There is substantial overlap between urban and
agricultural areas and the spawning and rearing habitats of the SC steelhead populations. Given
the long residency times in off-channel habitats, the risk to this species' survival and recovery
from the stressors of the proposed action is high.
Upper Columbia River Steelhead
The UCR steelhead DPS includes all naturally spawned populations below natural and man-
made impassable barriers in the Columbia River basin upstream from the Yakima River,
Washington, to the U.S.-Canada border. The DPS is comprised of four naturally-spawned
populations and six artificial propagation programs in Washington State. The historical
independent populations are located within the Wenatchee, Entiat, Methow, and Okanogan river
systems. Abundance data indicate that all populations are below the minimum threshold for
recovery. The annual population growth rates range between 1.067 and 1.086. Overall adult
returns are dominated by hatchery fish.
The major threats to this DPS identified in the Status of Listed Resources and Environmental
Baseline sections include dams that block fish migration and alter river hydrology; water
diversions; destruction or degradation of riparian habitat; and land use practices (logging,
agriculture, urbanization). Elevated water temperature and poor water quality also affect the
status of UCR steelhead. Pesticides have also been detected in UCR steelhead freshwater
habitats. NAWQA sampling from 1992-1995 in the Central Columbia Plateau detected
numerous pesticides in surface water (Williamson et al. 1998). Carbaryl was detected in 5% of
samples, and carbofuran was detected in 6%. Methomyl was not detected. While detections of
these three chemicals did not exceed water quality standards, concentrations of six other
pesticides exceeded EPA criteria for the protection of aquatic life. The co-occurrence of atrazine
with carbaryl and other OPs in aquatic habitats increases the likelihood of adverse responses in
salmonids and their aquatic prey. The cumulative impacts from these multiple threats continue
to affect UCR steelhead.
Land uses within this DPS range include agricultural, urban and residential development. Based
on the crop types, we expect that carbaryl and methomyl are commonly applied in the
DRAFT 481
-------
DRAFT
Wenatchee, Methow, and Entiat subbasins throughout the growing season. Existing stocks of
carbofuran are also expected to be applied throughout the growing season. As runoff containing
pesticides and other contaminants from the above land uses drain into the Wenatchee, Entiat, and
Methow Rivers, DPS exposure to these pesticides is likely. We also expect other AChE
inhibiting pesticides will co-occur with carbaryl, carbofuran, and methomyl in the waters of the
Wenatchee, Entiat, Methow, and Lower Crab Creek subbasins and exacerbate adverse effects
from AChE inhibition.
UCR adults return to the Columbia River in the late summer and early fall. UCR steelhead
spawn and rear in the major tributaries to the Columbia River between Rock Island and Chief
Joseph dams. Adults reach spawning areas in late spring of the calendar year following entry
into the river. Newly emerged fry move about considerably and seek suitable rearing habitat,
such as stream margins or cascades. Steelhead fry typically select off-channel habitats
associated with their natal rivers and stream for extended periods of rearing. Fry move
downstream in the fall in search of suitable overwintering habitat (Chapman, Hillman et al.
1994). Most juvenile steelhead spend two or three years in freshwater before migrating to the
ocean. Some juvenile steelhead may spend up to seven years rearing in freshwater before
migrating to sea. Given the long duration in shallow freshwater habitats, juveniles are more
likely to experience higher pesticide exposure, contaminants, and elevated temperature.
We expect that the proposed uses of carbaryl, carbofuran, and methomyl may contaminate the
rearing habitat of UCR steelhead and lead to individual and population level effects. Therefore,
the risk to this species' survival and recovery from the stressors of the action is high.
Upper Willamette River Steelhead
The UWR steelhead DPS includes all naturally spawned late-fall populations below natural and
man-made impassable barriers in the Willamette River, Oregon, and its tributaries upstream from
Willamette Falls to the Calapooia River. The DPS is comprised of four historical populations.
Steelhead populations within this DPS have been declining, on average, since 1971 and have
exhibited large fluctuations in abundance. Long-term trends in the annual population growth rate
DRAFT 482
-------
DRAFT
are less than one. The annual population growth rate of the four independent populations ranged
from 0.97 to 1.023.
The major threats to this ESU identified in the Status of Listed Resources and Environmental
Baseline sections include habitat loss due to blockages from hydroelectric dams and irrigation
diversions, and degraded water quality within the Willamette mainstem and the lower reaches of
its tributaries. Additionally, pesticide use and detections in UWR steelhead freshwater habitats
are well documented. Fifty pesticides were detected in streams that drain both agricultural and
urban areas. Forty-nine pesticides were detected in streams draining agricultural land, while 25
pesticides were detected in streams draining urban areas. Ten of these pesticides, including
carbaryl and carbofuran, exceeded EPA criteria for the protection of freshwater aquatic life for
chronic toxicity (Wentz, Bonn et al. 1998). The cumulative impacts from these multiple threats
continue to affect UWR steelhead.
Based on the crop types, we expect that carbaryl and methomyl are commonly applied in the
Willamette Valley throughout the growing season. Existing stocks of carbofuran are also
expected to be applied throughout the growing season. Registered 24(c) uses in Oregon include
carbofuran application to potatoes, nursery stock, sugar beets, and watermelons. The Willamette
Basin is the largest agricultural area in Oregon. In 1992 the Willamette Basin accounted for 51%
of Oregon's total gross farm sales and 58% of Oregon's crop sales. About one-third of the
agricultural land is irrigated and most of it is adjacent to the mainstem Willamette River. Urban
developments are also located primarily in the valley along the mainstem Willamette River. We
expect carbaryl to be used throughout the year in urban and residential areas. Given that major
urban and agricultural areas are located adjacent to the mainstem Willamette, DPS exposure to
these pesticides is likely. We also expect that other AChE inhibiting pesticides will co-occur
with carbaryl, carbofuran, and methomyl in the waters of the Willamette Valley and exacerbate
adverse effects from AChE inhibition.
UWR steelhead enter the Willamette River in January and February and ascend to their spawning
areas in late March or April. After emergence, steelhead fry typically rear in off-channel habitats
associated with their natal rivers and streams for two to three years. Smolt migration past
DRAFT 483
-------
DRAFT
Willamette Falls begins in early April and extends through early June, with peak migration in
early to mid-May. Most spend two years in the ocean before re-entering freshwater to spawn.
Much of the Willamette Valley has been converted to agricultural purposes. Based on the co-
occurrence of high use land classes with spawning and rearing habitats, we expect that the
proposed uses of carbaryl, carbofuran, and methomyl may contaminate spawning and rearing
habitats and lead to both individual and population level effects. Therefore, the risk to this
species' survival and recovery from the stressors of the action is high.
Summary of Species-Level Effects
In the preceding section NMFS described expected population level effects in terms of
reductions in annual growth rate, productivity (reproduction), and abundance (numbers of
salmonids). We concluded that all but Ozette Lake sockeye salmon, Snake River sockeye
salmon, Northern California steelhead, Columbia River chum salmon, Hood Canal summer-run
chum salmon, and Oregon Coast coho salmon populations will likely show reductions in
viability due to the reregi strati on of carbaryl and carbofuran. We also concluded that all but
Ozette Lake sockeye salmon, Snake River sockeye salmon, Northern California steelhead,
Columbia River chum salmon, Hood Canal summer-run chum salmon, Oregon Coast coho
salmon, Snake River fall-run Chinook salmon, Snake River spring-summer-run Chinook salmon,
California Coastal Chinook salmon, and Snake River steelhead populations will likely show
reductions in viability due to the reregi strati on of methomyl. The effects of EPA's proposed
action are first manifested at the individual level where reductions in individual fitness are
expected. We showed that an individual's survival, migration, and growth are all significantly
reduced by the proposed action. We also showed that these reductions are likely intensified by
co-occurring stressors in the action area including the presence of other carbamates and OP
insecticides in the action area.
Therefore, given the severity of expected changes in the annual population growth rate for
affected populations, it is likely that California Coastal Chinook salmon, Central Valley spring-
run Chinook salmon, LCR Chinook salmon, Puget Sound Chinook salmon, Sacramento River
winter-run Chinook salmon, Snake River fall-run Chinook salmon, Snake River spring/summer-
DRAFT 484
-------
DRAFT
run Chinook salmon, UCR spring-run Chinook salmon, Upper Willamette River Chinook
salmon, Central California Coast coho salmon, Southern Oregon and Northern Coastal California
coho salmon, Central California Coast steelhead, California Central Valley steelhead, LCR
steelhead, MCR steelhead, Snake River Basin steelhead, South-Central California Coast
steelhead, Southern California steelhead, UCR steelhead, and Upper Willamette River steelhead
will experience reductions in viability, which ultimately reduces the likelihood of survival and
recovery of these species. The Ozette Lake sockeye salmon, Snake River sockeye salmon,
Northern California steelhead, Columbia River chum salmon, Hood Canal summer-run chum
salmon, and Oregon Coast coho salmon will not likely experience reductions in viability.
Effects of the Proposed Action to Designated Critical Habitat
NMFS' critical habitat analysis determines whether the proposed action will destroy or adversely
modify critical habitat for ESA-listed species by examining any change in the conservation value
of the essential features of critical habitat. Our analysis does not rely on the regulatory definition
of 'adverse modification or destruction' of critical habitat. Instead, this analysis focuses on
statutory provisions of the ESA, including those in Section 3 that define "critical habitat" and
"conservation," those in Section 4 that describe the designation process, and those in Section 7
setting forth the substantive protections and procedural aspects of consultation.
NMFS has designated critical habitat for all listed Pacific salmonids except for LCR coho
salmon and Puget Sound steelhead. The action area encompasses all designated critical habitat
areas considered in this Opinion. The PCEs for each listed species, where they have been
designated, are described in the Status of Listed Resources section of this Opinion. The PCEs
identify those physical or biological features that are essential to the conservation of the species
that may require special management considerations or protections. As the species addressed in
this Opinion have similar life history characteristics, they share many of the same PCEs. These
PCEs include sites essential to support one or more life stages (sites for spawning, rearing,
migration, and foraging) and contain physical or biological features essential to the conservation
oftheESU/DPS, such as:
1. freshwater spawning sites with water quantity and quality conditions and substrate
supporting spawning, incubation and larval development;
DRAFT 485
-------
DRAFT
2. freshwater rearing sites with water quantity and floodplain connectivity to form and
maintain physical habitat conditions and support juvenile growth and mobility; water
quality and forage supporting juvenile development; and natural cover such as shade,
submerged and overhanging large wood, logjams and beaver dams, aquatic vegetation,
large rocks and boulders, side channels, and undercut banks;
3. freshwater migration corridors free of obstruction, along with water quantity and quality
conditions and natural cover such as submerged and overhanging large wood, aquatic
vegetation, large rocks and boulders, side channels, and undercut banks supporting
juvenile and adult mobility and survival;
4. estuarine areas free of obstruction, along with water quality, water quantity, and salinity
conditions supporting juvenile and adult physiological transitions between fresh and
saltwater; natural cover such as submerged and overhanging large wood, aquatic
vegetation, large rocks and boulders, and side channels; and juvenile and adult forage,
including aquatic invertebrates and fishes, supporting growth and maturation;
5. nearshore marine areas free of obstruction with water quality and quantity conditions and
forage, including aquatic invertebrates and fishes, supporting growth and maturation; and
natural cover such as submerged and overhanging large wood, aquatic vegetation, large
rocks and boulders, and side channels; and
6. offshore marine areas with water quality conditions and forage, including aquatic
invertebrates and fishes, supporting growth and maturation.
At the time that each habitat area was designated as critical habitat, that area contained one or
more PCEs within the acceptable range of values required to support the biological processes for
which the species use that habitat. Based on our Effects Analysis, the proposed action will affect
characteristics of three PCEs: freshwater spawning habitat, freshwater rearing habitat, and
migration corridors. Of particular concern are the effects of EPA's proposed registration of
carbaryl, carbofuran, and methomyl on salmonid prey and water quality in these areas.
Direct exposure to carbaryl, carbofuran, and methomyl and the other chemical stressors of the
action within freshwater will have an effect on Pacific salmonid critical habitat that overlaps with
intense agricultural and residential/urban land uses. The Environmental Baseline discusses the
extent of anthropogenic alteration of salmonid habitat within the action area. As established in
Effects of the Action to Listed Species, agricultural, urban, and residential areas overlap with
salmonids' geographic range to differing degrees between ESUs/DPSs. As noted in the Effects
Analysis, pesticides most often occur in the aquatic environment as mixtures. Carbaryl,
carbofuran, and methomyl are found in environmental mixtures. Based on evidence of additive
DRAFT 486
-------
DRAFT
and synergistic effects of these compounds, we expect mortality of large numbers and types of
aquatic insects, which are prey items for salmon. A reduction in salmonid prey abundance poses
subsequent impacts on the overall growth of salmonid juveniles, especially during their first year
of survival.
Reduction in Prey
Freshwater rearing habitats must provide forage areas that support juvenile development. A
reduction in the abundance of prey items will decrease the conservation value of these habitats,
as they will support fewer individuals, especially during their first year of survival. Given the
environmental baseline conditions of the aquatic systems, the existing invertebrate taxa
community is already depauperate and requires longer periods for recovery following each
pesticide application event. The a.i.s in surface runoff and pesticide drift entering freshwater
rearing habitats will exacerbate reductions in insect communities that are salmonid prey items in
these degraded systems. We expect that rearing areas will be contaminated by the proposed uses
of carbaryl, carbofuran, and methomyl to the extent that the habitat is precluded from serving its
intended role in the survival and recovery of listed salmonids.
Reduction in Water Quality
Runoff from areas of intensive urban and agricultural development will likely contain carbaryl,
carbofuran, and methomyl in addition to other pesticides - particularly other AChE-inhibiting
pesticides, chemical pollutants, and sediments that also degrade water quality. Depending on the
available water flow, amount of shade from LWD and intact riparian zones, and water
temperature in aquatic habitats, the toxicity of carbaryl, carbofuran, and methomyl in tributary
and stream waters may become more pronounced. Reductions in water quality may reduce the
conservation value of designated spawning, rearing, and migratory habitat. We expect that
proposed uses may contaminate these areas, thereby precluding habitat from its intended purpose
in supporting the survival and recovery of listed Pacific salmonids.
The precise change in the conservation value of critical habitat within the ESU/DPS from the
proposed action cannot be quantified and will likely vary according to the specific designated
critical habitat. However, based on the effects described above, it is reasonably likely that the
DRAFT 487
-------
DRAFT
proposed action will have a large, local, negative reduction in that conservation value of
designated critical habitat within highly developed agricultural, residential, and urban areas. The
duration, frequency, and severity of these reductions will vary according to overall numbers and
volume of applications of carbaryl, carbofuran, and methomyl in areas of designated critical
habitat, among other variables.
Therefore, we expect that the registration of carbaryl and carbofuran will adversely affect
designated critical habitat of all listed Pacific salmonids except for that of Ozette Lake sockeye
salmon, Snake River sockeye salmon, Northern California steelhead, Columbia River chum
salmon, Hood Canal summer-run chum salmon, and Oregon Coast coho salmon. We also expect
that the reregi strati on of methomyl will adversely affect designated critical habitat of all listed
Pacific salmonids, except for that of Ozette Lake sockeye salmon, Snake River sockeye salmon,
Northern California steelhead, Columbia River chum salmon, Hood Canal summer-run chum
salmon, Oregon Coast coho salmon, Snake River fall-run Chinook salmon, Snake River spring-
summer-run Chinook salmon, California Coastal Chinook salmon, and Snake River steelhead.
Conclusion
Carbaryl and Carbofuran
After reviewing the current status of California Coastal Chinook salmon, Central Valley spring-
run Chinook salmon, LCR Chinook salmon, Puget Sound Chinook salmon, Sacramento River
winter-run Chinook salmon, Snake River fall-run Chinook salmon, Snake River spring/summer-
run Chinook salmon, UCR spring-run Chinook salmon, Upper Willamette River Chinook
salmon, Central California Coast coho salmon, LCR coho salmon, Southern Oregon and
Northern Coastal California coho salmon, California Central Valley steelhead, Central California
Coast steelhead, LCR steelhead, MCR steelhead, Puget Sound steelhead, Snake River Basin
steelhead, South Central California coast steelhead, Southern California steelhead, UCR
steelhead, and Upper Willamette River steelhead, the environmental baseline for the action area,
the effects of the proposed action, and the cumulative effects, it is NMFS' Opinion that the
registration of carbaryl and carbofuran is likely to jeopardize the continued existence of these
endangered or threatened species (Table 81).
DRAFT 488
-------
DRAFT
It is NMFS' Opinion that the registration of carbaryl and carbofuran is not likely to jeopardize
the continued existence of Ozette Lake sockeye salmon, Snake River sockeye salmon, Northern
California steelhead, Columbia River chum salmon, Hood Canal summer-run chum salmon, and
Oregon Coast coho salmon (Table 81).
After reviewing the current status of designated critical habitat for California Coastal Chinook
salmon, Central Valley spring-run Chinook salmon, LCR Chinook salmon, Puget Sound
Chinook salmon, Sacramento River winter-run Chinook salmon, Snake River fall-run Chinook
salmon, Snake River spring/summer-run Chinook salmon, UCR spring-run Chinook salmon,
Upper Willamette River Chinook salmon, Columbia River chum salmon, Central California
Coast coho salmon, Southern Oregon/Northern Coastal California coho salmon, California
Central Valley steelhead, Central California Coast steelhead, LCR steelhead, MCR steelhead,
Snake River Basin steelhead, South-Central California coast steelhead, Southern California
steelhead, UCR steelhead, and Upper Willamette River steelhead, the environmental baseline for
the action area, the effects of the proposed action, and the cumulative effects, it is NMFS'
Opinion that the registration of carbaryl and carbofuran, is likely to result in the destruction or
adverse modification of critical habitat of these endangered and threatened species (Table 82).
It is NMFS' Opinion that the registration of carbaryl and carbofuran is not likely to result in the
destruction or adverse modification of critical habitat of Ozette Lake sockeye salmon, Snake
River sockeye salmon, Northern California steelhead, Columbia River chum salmon, Hood
Canal summer-run chum salmon, and Oregon Coast coho salmon (Table 82).
Methomyl
After reviewing the current status of Central Valley spring-run Chinook salmon, LCR Chinook
salmon, Puget Sound Chinook salmon, Sacramento River winter-run Chinook salmon, UCR
spring-run Chinook salmon, Upper Willamette River Chinook salmon, Central California Coast
coho salmon, LCR coho salmon, Southern Oregon and Northern Coastal California coho salmon,
California Central Valley steelhead, Central California Coast steelhead, LCR steelhead, MCR
steelhead, Puget Sound steelhead, South Central California coast steelhead, Southern California
DRAFT 489
-------
DRAFT
steelhead, UCR steelhead, and Upper Willamette River steelhead, the environmental baseline for
the action area, the effects of the proposed action, and the cumulative effects, it is NMFS'
Opinion that the registration of methomyl is likely to jeopardize the continued existence of these
endangered or threatened species (Table 81).
It is NMFS' Opinion that the registration of methomyl is not likely to jeopardize the continued
existence of California Coastal Chinook salmon, Snake River fall-run Chinook salmon, Snake
River spring/summer-run Chinook salmon, Ozette Lake sockeye salmon, Snake River sockeye
salmon, Northern California steelhead, Columbia River chum salmon, Hood Canal summer-run
chum salmon, Oregon Coast coho salmon, and Snake River steelhead (Table 81).
After reviewing the current status of designated critical habitat for Central Valley spring-run
Chinook salmon, LCR Chinook salmon, Puget Sound Chinook salmon, Sacramento River
winter-run Chinook salmon, UCR spring-run Chinook salmon, Upper Willamette River Chinook
salmon, Central California Coast coho salmon, LCR coho salmon, Southern Oregon and
Northern Coastal California coho salmon, California Central Valley steelhead, Central California
Coast steelhead, LCR steelhead, MCR steelhead, Puget Sound steelhead, South Central
California coast steelhead, Southern California steelhead, UCR steelhead, and Upper Willamette
River steelhead, the environmental baseline for the action area, the effects of the proposed
action, and the cumulative effects, it is NMFS' Opinion that the registration of methomyl is
likely to result in the destruction or adverse modification of critical habitat of these endangered
and threatened species (Table 82).
It is NMFS' Opinion that the registration of methomyl is not likely to result in the destruction or
adverse modification of critical habitat of California Coastal Chinook salmon, Snake River fall-
run Chinook salmon, Snake River spring/summer-run Chinook salmon, Ozette Lake sockeye
salmon, Snake River sockeye salmon, Northern California steelhead, Columbia River chum
salmon, Hood Canal summer-run chum salmon, Oregon Coast coho salmon, and Snake River
steelhead (Table 82).
DRAFT 490
-------
DRAFT
Table 81. Jeopardy and Non-Jeopardy Determinations for Listed Species
Common Name
(Scientific Name)
Chinook salmon
Oncorhynchus tshawytscha
Chum salmon
Oncorhynchus keta
Coho salmon
Oncorhynchus kisutch
Sockeye salmon
Oncorhynchus nerka
Steelhead
Oncorhynchus mykiss
Distinct Population Segment or
Evolutionary Significant Unit
California Coastal
Central Valley Spring-run
Lower Columbia River
Upper Columbia River Spring-run
Puget Sound
Sacramento River Winter-run
Snake River Fall-run
Snake River Spring/Summer-run
Upper Willamette River
Columbia River
Hood Canal Summer-run
Lower Columbia River
Oregon Coast
Southern Oregon & Northern California Coast
Central California Coast
Ozette Lake
Snake River
Central California Coast
California Central Valley
Lower Columbia River
Middle Columbia River
Northern California
Puget Sound
Snake River
South-Central California Coast
Southern California
Upper Columbia River
Upper Willamette River
Carbaryl Carbofuran
Jeopardy
Jeopardy
Jeopardy
Jeopardy
Methomyl
Non-Jeopardy
Non-Jeopardy
Non-Jeopardy
Jeopardy
Non-Jeopardy
Jeopardy
Non-Jeopardy
Jeopardy
Non-Jeopardy
Jeopardy
Jeopardy
Jeopardy
Non-Jeopardy
DRAFT
491
-------
DRAFT
Table 82. Adverse Modification Determinations for Designated Critical Habitat of Listed Species
Common Name
(Scientific Name)
Chinook salmon
Oncorhynchus tshawytscha
Chum salmon
Oncorhynchus keta
Coho salmon
Oncorhynchus kisutch
Sockeye salmon
Oncorhynchus nerka
Steelhead
Oncorhynchus mykiss
Distinct Population Segment or
Evolutionary Significant Unit
California Coastal
Central Valley Spring-run
Lower Columbia River
Upper Columbia River Spring-run
Puget Sound
Sacramento River Winter-run
Snake River Fall-run
Snake River Spring/Summer-run
Upper Willamette River
Columbia River
Hood Canal Summer-run
Lower Columbia River
Oregon Coast
Southern Oregon & Northern California Coast
Central California Coast
Ozette Lake
Snake River
Central California Coast
California Central Valley
Lower Columbia River
Middle Columbia River
Northern California
Puget Sound
Snake River
South-Central California Coast
Southern California
Upper Columbia River
Upper Willamette River
Carbaryl Carbofuran
Adverse Modification
Methomyl
No Adverse
Modification
Adverse Modification
Adverse Modification
No Adverse
Modification
Adverse Modification
No Adverse Modification
No Designated Critical Habitat
No Adverse Modification
Adverse Modification
No Adverse Modification
Adverse Modification
No Adverse Modification
No Designated Critical Habitat
Adverse Modification
No Adverse
Modification
Adverse Modification
DRAFT
492
-------
DRAFT
This Opinion has concluded that EPA's proposed registration of pesticides containing carbaryl
and carbofuran is likely to jeopardize the continued existence of 22 endangered and threatened
Pacific salmonids and is likely to destroy or adversely modify designated critical habitat for 20
threatened and endangered salmonids. The Opinion further concluded that EPA's proposed
registration of pesticides containing methomyl is likely to jeopardize the continued existence of
18 endangered and threatened Pacific salmonids and is likely to destroy or adversely modify
designated critical habitat for 16 threatened and endangered salmonids. "Jeopardize the
continued existence of means "to engage in an action that reasonably would be expected,
directly or indirectly, to reduce appreciably the likelihood of both the survival and recovery of a
listed species in the wild by reducing the reproduction, numbers, or distribution of that species"
(50 CFR §402.02).
Regulations (50 CFR §402.02) implementing section 7 of the ESA define reasonable and prudent
alternatives as alternative actions, identified during formal consultation, that: (1) can be
implemented in a manner consistent with the intended purpose of the action; (2) can be
implemented consistent with the scope of the action agency's legal authority and jurisdiction; (3)
are economically and technologically feasible; and (4) NMFS believes would avoid the
likelihood of jeopardizing the continued existence of listed species or resulting in the destruction
or adverse modification of critical habitat.
NMFS reached this conclusion because measured and predicted concentrations of the three a.i.s
in salmonid habitats, particularly in off-channel habitats5, are likely to cause adverse effects to
listed Pacific salmonids including significant reductions in growth and survival. For carbaryl
and carbofuran, 22 ESUs/DPSs of listed Pacific salmonids are likely to suffer reductions in
5 Off-channel habitat - water bodies and/or inundated areas that are connected (accessible to salmonid juveniles)
seasonally or annually to the main channel of a stream including but not limited to features such as side channels,
alcoves, oxbows, ditches, andfloodplains.
Main channel -the stream channel that includes the thalweg (longitudinal continuous deepest portion of the channel.
DRAFT 493
-------
DRAFT
viability given the severity of expected changes in abundance and productivity associated with
the proposed action. Similarly for methomyl, 18 ESUs/DPSs of listed Pacific salmonids are
likely to suffer reductions in viability. These adverse effects are expected to appreciably reduce
the likelihood of both the survival and recovery of these listed Pacific salmonids. EPA's
proposed registration of carbaryl and carbofuran is also likely to result in the destruction or
adverse modification of critical habitat for 20 affected ESUs/DPSs because of adverse effects on
salmonid prey and water quality in freshwater rearing, spawning, and foraging areas. EPA's
proposed registration of methomyl is also likely to result in the destruction or adverse
modification of critical habitat for 16 affected ESUs/DPSs because of adverse effects on
salmonid prey and water quality in freshwater rearing, spawning, and foraging areas.
The Reasonable and Prudent Alternative (RPA) accounts for the following issues: (1) the action
will result in exposure to other chemical stressors in addition to the a.i. that may increase the risk
of the action to listed species, including unspecified inert ingredients, adjuvants, and tank mixes;
(2) exposure to chemical mixtures containing carbaryl, carbofuran, and methomyl and other
cholinesterase-inhibiting compounds result in additive and synergistic responses; and (3)
exposure to other chemicals and physical stressors (e.g., pH and temperature) in the baseline
habitat will likely intensify response to carbaryl, carbofuran, and methomyl.
The action as implemented under the RPA will remove the likelihood of jeopardy and of
destruction or adverse modification of critical habitat. In the proposed RPA, NMFS does not
attempt to ensure there is no take of listed species. NMFS believes take will occur, and has
provided an incidental take statement exempting that take from the take prohibitions, so long as
the action is conducted according to the RPA and reasonable and prudent measures (RPM).
Avoiding take altogether would most likely entail canceling registration, or prohibiting use in
watersheds inhabited by salmonids. NMFS recognizes that carbofuran's registration is currently
in the process of being canceled. However, NMFS is uncertain when and if cancelation will
occur. Furthermore, existing stocks under older labels would remain and are currently allowed
to be applied. The RPA and RPMs therefore apply to carbofuran as well. The goal of the RPA
is to reduce exposure to ensure that the action is not likely to jeopardize listed species or destroy
or adversely modify critical habitat.
DRAFT 494
-------
DRAFT
The RPA is comprised of seven required elements that must be implemented in its entirety
within one year of the receipt of the Opinion to ensure the proposed registration of these
pesticides is not likely to jeopardize endangered or threatened Pacific salmonids under the
jurisdiction of NMFS or destroy or adversely modify critical habitat that has been designated for
these species. These elements rely upon recognized practices for reducing drift and runoff of
pesticide products into aquatic habitats.
Specific Elements of the Reasonable and Prudent Alternative
Elements 1-5 shall be specified on FIFRA labels of all pesticide products containing carbaryl,
carbofuran, and methomyl used in California, Idaho, Oregon, and Washington. Alternatively,
the label could direct pesticide users to the EPA's Endangered Species Protection Program
(ESPP) bulletins that specify elements 1-5.
Element 1.
Do not apply pesticide products6 within specified buffers of salmonid habitats7 (Table 83).
Buffers only apply to those salmonid habitats where NMFS concluded jeopardy or the
destruction or adverse modification of designated critical habitat for listed Pacific salmonids.
Buffers shall be measured from the water's edge of salmonid habitat, including off-channel
habitat, to the point of deposition (below spray nozzle).
Pesticide buffers are recognized tools to reduce pesticide loading into aquatic habitats from drift.
EPA, USFWS, NMFS, courts, and state agencies routinely enlist buffers as pesticide load
reduction measures. EPA requires the use of buffers on end-use product labels for ground and/or
aerial applications for some products that pose risk to aquatic systems. For example, many
methomyl containing end-use products have mandated buffers of 25, 100, and 450 ft for ground,
6 Use of the term "pesticide products" in the Reasonable and Prudent Alternative section of the Opinion refers to
pesticide products containing carbaryl, carbofuran, and methomyl.
Salmonid habitats are defined as freshwaters, estuarine habitats, and nearshore marine habitats including bays
within the ESU/DPS ranges including migratory corridors. The freshwater habitats include intermittent streams and
other temporally connected habitats to salmonid-bearing waters. Freshwater habitats also include all known types of
off-channel habitats as well as drainages, ditches, and other man-made conveyances to salmonid habitats that lack
salmonid exclusion devices.
DRAFT 495
-------
DRAFT
aerial, and Ultra Low Volume applications, respectively. CDPR has pesticide use limitations of
120 and 600 ft buffers for carbaryl, carbofuran, and methomyl-containing pesticides when the
wind is blowing toward sensitive areas. On June 14, 1989, USFWS issued a Biological Opinion
for 165 listed species and 112 pesticide a.i.s. Prescribed buffers under species-specific RPAs
ranged from 60 ft (ground applications) to one half mile (aerial applications). Many of EPA's
historical county bulletins for endangered species referenced a 60 ft buffer for ground
applications and a 300 ft buffer for aerial spraying. One court decision prescribed mandatory 60
ft (ground) and 300 ft (aerial) buffers for applications within the ranges of ESA-listed Pacific
salmonids. NMFS has prescribed a range of buffers in ESA consultations for herbicide and
insecticide application actions by agencies such as the U.S. Forest Service and Bureau of Land
Management overlapping with ESA-listed salmonid habitats. Herbicide buffers ranged from 0 ft
to 500 ft depending on application type, rate, and frequency. Insecticide buffers ranged from 0 ft
to 200 ft depending on application type, rate, and frequency.
NMFS generated estimated environmental concentrations for the three TV-methyl carbamates for
off-channel habitats using the AgDrift model (set to EPA Tier 1 simulation defaults). NMFS
generated values for a range of buffer sizes in 100 ft increments for ground applications (0 -
1,000 ft), and aerial applications (0 - 1,000 ft). The dimensions of the off-channel habitat
modeled were 32.8 ft (10 m) wide and 0.328 ft (0.1 m) deep. Key model assumptions include:
- Drift as the only pathway of exposure;
A single application of each a.i.; and
- The estimates represent instantaneous average concentrations across the entire habitat
modeled.
DRAFT 496
-------
DRAFT
Table 83. Mandatory pesticide no application buffers for ground and aerial applications
Rate
Ibs/ acre
Carbaryl
Carbofuran
Methomyl
No Application Buffer (ft)
Ground Applications
0-1
>l-2
>3-5
>5-10
>10
200
300
400
500
600
200
300
400
500
600
50
NA
NA
NA
NA
Aerial Applications
All rates
1000
1000
600
The estimated concentrations decline as buffer size increases (Table 84). We note the disparity
between the concentrations predicted by the two models. For example, at 100 ft. the predicted
concentrations is 4.4 |ig/L from a ground application and from an aerial application is 92.9 |ig/L.
The two results are not directly comparable because the models use different methods to predict
amount of drift.
Table 84. Estimated environmental concentrations of carbaryl, carbofuran, and methomyl applied
at the rate or 1 Ib per acre for ground and aerial applications.
Ground application, low boom,
ASAE very fine-fine droplet distribution, 50th percentile estimates
EPA Tier 1 simulation
Buffer (ft)
0
10
100
200
300
400
500
600
700
800
900
997
Off-Channel (10 m * 0
1 m) (ug/L)
76.427
20.168
4.406
2.568
1.813
1.392
1.122
0.933
0.794
0.688
0.604
0.583
Aerial application, fine-medium droplet distribution. EPA Tier
Buffer (ft)
0
10
Off-Channel (10 m * 0
1 simulation
1 m) (ug/L)
333.566
260.482
DRAFT
497
-------
DRAFT
100
200
300
400
500
600
700
800
900
997
92.888
48.985
33.096
25.289
20.902
18.010
16.035
14.692
13.719
12.983
Based on the population growth modeling exercises, a four-day exposure to expected
concentrations of carbaryl, carbofuran, and methomyl will substantially reduce a population's
growth rate due to prey base reduction. All four life history types modeled demonstrated this
effect. As exposure duration and application rate increases, we expect more pronounced effects
on salmonid prey abundance and recovery timing leading to further reduced salmonid growth.
We expect that some juvenile salmonids are likely to experience reductions in growth. However,
the prescribed buffers are intended to avoid population level effects. We also expect that prey
items will die from these exposures. The likelihood of impacting salmonid prey availability, an
identified PCE, is substantially reduced by these buffers.
The majority of buffers described earlier are smaller than the buffers prescribed in this element.
Concentrations expected with smaller buffers would lead to a greater probability of affecting
populations and PCEs especially in habitats that are compromised from a variety of stressors
(described in the Environmental Baseline section).
The scenario we modeled with AgDrift in this RPA element is expected to occur when all of the
modeled variables are present e.g., specific wind speed, wind direction, release height, size of
off-channel habitat, droplet size distribution, etc. The input variables are relevant to field
conditions and the frequency of this exact scenario occurring remains unknown. We selected
this scenario to represent off-channel habitats used by a sensitive salmonid life stage i.e.,
juveniles. NMFS believes that these buffers will remove a substantial portion of risk attributed
to pesticide drift.
DRAFT
498
-------
DRAFT
Element 2. Do not apply when wind speeds are greater than or equal to 10 mph as measured
using an anemometer immediately prior to application. When applying pesticide products,
commence applications on the side nearest the aquatic habitat and proceed away from the aquatic
habitat.
Elements. For agricultural uses, provide a 20 ft (6.1 m) minimum strip of non-crop vegetation
(on which no pesticides shall be applied) on the downhill side of the application site immediately
adjacent to any surface waters that have a connection to salmonid-bearing waters. Vegetation
strips should be sufficiently dense to prevent any channelized flow through the strip. This
includes drainage systems that have salmonid exclusion devices, but drain to salmonid-bearing
waters.
Element 4. For all uses do not apply pesticide products when soil moisture is at field capacity,
or when a storm event likely to produce runoff from the treated area is forecasted by
NOAA/NWS, (National Weather Service) to occur within 48 h following application.
Element 5. Report all incidents offish mortality that occur within four days of application and
within the vicinity of the treatment area to EPA OPP (703-305-7695).
Element 6. In addition to the labeling requirements above, EPA shall develop and implement a
NMFS-approved effectiveness monitoring plan for off-channel habitats with annual reports. The
plan shall identify representative off-channel habitats within agricultural areas prone to drift and
runoff of pesticides. The number and locations of off-channel habitat sampling sites shall
include currently used off-channel habitats by threatened and endangered Pacific salmonids
identified by NMFS biologists and will include at least two sites for each general species (ESU,
DPS) i.e., coho salmon, chum salmon, steelhead, sockeye salmon, and ocean-type Chinook and
stream-type Chinook salmon. The plan shall collect daily surface water samples targeting at
least three periods during the application season for seven days. Collected water samples will be
analyzed for current-use OPs and carbamates following USGS schedule for analytical chemistry.
The report shall be submitted to NMFS OPR and will summarize annual monitoring data and
provide all raw data.
DRAFT 499
-------
DRAFT
Element 7. This element is specific to Washington State's 24(c) for carbaryl applications on
estuarine mudflats. As this use is specific for application on mudflats within estuaries, elements
1, 3, and 4 are not applicable for this use. Elements 2 and 5 are required. A monitoring program
approved by NMFS OPR shall be established in Willapa Bay and Grays Harbor to determine the
presence, AChE activity, and genetic source populations of captured juvenile salmonids. For
example, a monitoring program shall be implemented to determine salmonid presence and health
at five locations using beach seines with the first incoming tide after carbaryl application.
Additionally, fyke nets placed in five channels proximate to sprayed mudflats shall be used to
determine salmonid presence and health at first outgoing tide after carbaryl application. All dead
or dying salmonids, if any, shall be collected and genetically analyzed to determine source
population origin. Additionally, muscle and brain AChE activity shall be measured in collected
dead or dying salmonids. A subset of the live salmonids captured in seines and nets shall be
genetically analyzed to determine source population origin. Annual reports shall be submitted to
NMFS and include the raw data and summaries of the raw data including number offish caught,
species of salmonids, genetic analysis results, AChE activity results of brain and muscle samples,
and number of dead or dying salmonids collected. The sampling effort shall be conducted
annually as long as carbaryl is allowed for use in Willapa Bay and Grays Harbor.
Although NMFS has concluded that EPA's action for carbaryl and carbofuran is likely to
jeopardize 22 listed ESUs/DPSs and destroy or adversely modify 20 designated critical habitats,
NMFS does not believe that the effects of the action will attain this level in the year between
issuance of this Opinion and EPA's implementation of the RPA. NMFS has further concluded
that EPA's action for methomyl is likely to jeopardize 18 listed ESUs/DPSs and destroy or
adversely modify 16 designated critical habitats. As with carbaryl and carbofuran, NMFS does
not believe that the effects of the action for methomyl will attain this level in the year between
issuance of this Opinion and EPA's implementation of the RPA. Products containing these three
a.i.s have been in use for some time. NMFS believes that these products have contributed to
ESU/DPS declines, but not to the extent that one year of additional use as now authorized would
lead to likely jeopardy or adverse modification.
DRAFT 500
-------
DRAFT
Based on the above conclusions on the effects of EPA's proposed registration of pesticide
products containing carbaryl, carbofuran, and methomyl, the EPA is required to notify NMFS
OPR of its final decision on implementation of the reasonable and prudent alternative.
", . . • • ;•
Section 9 of the ESA and federal regulation pursuant to section 4(d) of the ESA prohibits the
take of endangered and threatened species, respectively, without special exemption. Take is
defined as to harass, harm, pursue, hunt, shoot, wound, kill, trap, capture or collect, or to attempt
to engage in any such conduct. Harm is further defined by NMFS to include significant habitat
modification or degradation that results in death or injury to listed species by significantly
impairing essential behavioral patterns, including breeding, feeding, or sheltering. Incidental
take is defined as take that is incidental to, and not the purpose of, the carrying out of an
otherwise lawful activity. Under the terms of section 7(b)(4) and section 7(o)(2), taking that is
incidental to and not intended as part of the agency action is not considered to be prohibited
taking under the ESA provided that such taking is in compliance with the terms and conditions of
this Incidental Take Statement.
Amount or Extent of Take Anticipated
As described earlier in this Opinion, this is a consultation on the EPA's registration of pesticide
products containing carbaryl, carbofuran, and methomyl, and their formulations as they are used
in the Pacific Northwest and California and the impacts of these applications on listed
ESUs/DPSs of Pacific salmonids. The EPA authorizes use of these pesticide products for pest
control purposes across multiple landscapes. The goal of this Opinion is to evaluate the impacts
to NMFS' listed resources from the EPA's broad authorization of applied pesticide products.
This Opinion is a partial consultation because pursuant to the court's order, EPA sought
consultation on only 26 listed Pacific salmonids under NMFS' jurisdiction. However, even
though the court's order did not address the two more recently listed ESUs and DPSs, NMFS
analyzed the impacts of EPA's actions to them because they belong to the same taxon and the
analysis requires consideration of the same information. Consultation with NMFS will be
DRAFT 501
-------
DRAFT
completed when EPA makes effect determinations on all remaining species under NMFS'
jurisdiction and consults with NMFS as necessary.
For this Opinion, NMFS anticipates the general direct and indirect effects that would occur from
EPA's registration of pesticide products across the states of California, Idaho, Oregon, and
Washington to 28 listed Pacific salmonids under NMFS' jurisdiction. Recent and historical
surveys indicate that listed salmonids occur in the action area, in places where they will be
exposed to the stressors of the action. The RPAs are designed to reduce this exposure but not
eliminate it. Pesticide runoff and drift of carbaryl, carbofuran, and methomyl are most likely to
reach streams and other aquatic sites when they are applied to crops and other land use settings
located adjacent to wetlands, riparian areas, ditches, off-channel habitats, and intermittent
streams. These inputs into aquatic habitats are especially high when rainfall immediately follows
applications. The effects of pesticides and other contaminants found in urban runoff especially
from areas with a high degree of impervious surfaces may also exacerbate degraded water
quality conditions of receiving waters used by salmon. Urban runoff is also generally warmer in
temperature and elevated water temperature pose negative effects on certain life history phases
for salmon, increasing susceptibility to chemical stressors. The range of effects of the three a.i.s
on salmonids include reductions in growth, prey capture, and swimming ability, impaired
olfaction affecting homing and reproductive behaviors, and increased susceptibility to predation
and disease. Thus, we expect some exposed fish will respond to these effects by changing
normal behaviors. In some cases, fish may die, be injured, or suffer sublethal effects. These
results are not the purpose of the proposed action. Therefore, incidental take of listed salmonids
is reasonably certain to occur over the 15-year duration of the proposed action.
Given the variability of real-life conditions, the broad nature and scope of the proposed action,
and the migratory nature of salmon, the best scientific and commercial data available are not
sufficient to enable NMFS to estimate a specific amount of incidental take associated with the
proposed action. As explained in the Description of the Proposed Action and the Effects of the
Proposed Actions sections, NMFS identified multiple uncertainties associated with the proposed
action. Areas of uncertainty include:
DRAFT 502
-------
DRAFT
1. Incomplete information on the proposed action (i.e., no master label summarizing all
authorized uses of pesticide products containing carbaryl, carbofuran, and methomyl);
2. Limited use and exposure data on stressors of the action for non-agricultural uses of these
pesticides;
3. Minimal information on exposure and toxicity for pesticide formulations, adjuvants, and
other/inert ingredients within registered formulations;
4. No information on permitted tank mixtures and associated exposure estimates;
5. Limited data on toxicity of environmental mixtures;
6. No known method to predict synergistic responses from exposure to combinations of the
three a.i.s;
7. Annual variable conditions regarding land use, crop cover, and pest pressure;
8. Variable temporal and spatial conditions within each ESU, especially at the
population level; and
9. Variable conditions of water bodies in which salmonids live.
NMFS therefore identifies as a surrogate for the allowable extent of take the ability of this action
to proceed without any fish kills attributed to the use of carbaryl, carbofuran, or methomyl or any
compounds, degradates, or mixtures in aquatic habitats containing individuals from any
ESU/DPS. Because of the difficulty of detecting salmonid deaths, the fishes killed are not to be
listed salmonids. Salmonids appear to be more sensitive to these compounds, so that if there are
kills of other freshwater fishes attributed to use of these pesticides, it is likely that salmonids
have also died, even if no dead salmonids can be located. In addition, if stream conditions due
to pesticide use kill less sensitive fishes in certain areas, the potential for lethal and non-lethal
takes downstream areas increases. A fish kill is considered attributable to one of these three
ingredients, its metabolites, or degradates, if measured concentrations in surface waters are at
levels expected to kill fish, if AChE measurements were taken of the fish carcass and correlate to
fish death, if pesticides were applied in the general area, and if pesticide drift or runoff was
witnessed or apparent.
NMFS notes that with increased monitoring and study of the impact of these pesticides on water
quality, particularly water quality in off-channel habitats, NMFS will be able to refine this
incidental take statement, and future incidental take statements, to allow other measures of the
extent of take.
DRAFT 503
-------
DRAFT
Reasonable and Prudent Measures
The measures described below are non-discretionary, and must be undertaken by the EPA so that
they become binding conditions of any grant or permit issued to the applicant(s), as appropriate,
for the exemption in section 7(o)(2) to apply. The EPA has a continuing duty to regulate the
activity covered by this incidental take statement. If the EPA (1) fails to assume and implement
the terms and conditions or (2) fails to require the applicant(s) to adhere to the terms and
conditions of the incidental take statement through enforceable terms that are added to the permit
or grant document, the protective coverage of section 7(o)(2) may lapse. In order to monitor the
impact of incidental take, the EPA must report the progress of the action and its impact on the
species to NMFS OPR as specified in the incidental take statement. [50 CFR§402.14(i)(3)].
To satisfy its obligations pursuant to section 7(a)(2) of the ESA, the EPA must monitor (a) the
direct, indirect, and cumulative impacts of its long-term registration of pesticide products
containing carbaryl, carbofuran, and methomyl; (b) evaluate the direct, indirect, or cumulative
impacts of pesticide misapplications in the aquatic habitats in which they occur; and (c) the
consequences of those effects on listed Pacific salmonids under NMFS' jurisdiction. The
purpose of the monitoring program is for the EPA to use the results of the monitoring data and
modify the registration process in order to reduce exposure and minimize the effect of exposure
where pesticides will occur in salmonid habitat.
The EPA shall:
1. Minimize the amount and extent of incidental take from use of pesticide products
containing carbaryl, carbofuran, and methomyl by reducing the potential of chemicals
reaching the water.
2. Monitor any incidental take or surrogate measure of take that occurs from the action, and
3. Report annually to NMFS OPR on the monitoring results from the previous season.
DRAFT 504
-------
DRAFT
Terms and Conditions
To be exempt from the prohibitions of section 9 of the ESA, the EPA must comply with the
following terms and conditions, which implement the reasonable and prudent measure described
above. These terms and conditions are non-discretionary.
1. EPA shall include the following instructions requiring reporting offish kills either on the
labels for all products containing carbaryl, carbofuran, and methomyl or in ESPP
Bulletins:
NOTICE: Incidents where salmon appear injured or killed as a result of pesticide
applications shall be reported to NMFS OPR at 301-713-1401 and EPA at 703-305-7695.
The finder should leave the fish alone, make note of any circumstances likely causing the
death or injury, location and number offish involved, and take photographs, if possible.
Adult fish should generally not be disturbed unless circumstances arise where an adult
fish is obviously injured or killed by pesticide exposure, or some unnatural cause. The
finder may be asked to carry out instructions provided by NMFS OPR to collect
specimens or take other measures to ensure that evidence intrinsic to the specimen is
preserved.
2. EPA shall report to NMFS OPR any incidences regarding carbaryl, carborfuran, or
methomyl effects on aquatic ecosystems added to its incident database that it has
classified as probable or highly probable.
3. Do not apply pesticide products when wind speeds are greater than or equal to 10 mph as
measured using an anemometer immediately prior to application. When applying
pesticide products, commence applications on the side nearest the aquatic habitat and
proceed away from the aquatic habitat.
4. For all uses do not apply pesticide products when soil moisture is at field capacity, or
when a storm event likely to produce runoff from the treated area is forecasted by
NOAA/NWS, (National Weather Service) to occur within 48 h following application.
Conservation Recommendations
Section 7(a) (1) of the ESA directs federal agencies to use their authorities to further the
purposes of the ESA by carrying out conservation programs for the benefit of endangered and
threatened species. Conservation recommendations are discretionary agency activities to
DRAFT 505
-------
DRAFT
minimize or avoid adverse effects of a proposed action on listed species or critical habitat, to
help implement recovery plans, or to develop information.
The following conservation recommendations would provide information for future
consultations involving future authorizations of pesticide a.i.s that may affect listed species:
1. Conduct mixture toxicity analysis in screening-level and endangered species biological
evaluations;
2. Develop models to estimate pesticide concentrations in off-channel habitats; and
3. Develop models to estimate pesticide concentrations in aquatic habitats associated with
non-agricultural applications, particularly in residential and industrial environments.
In order for NMFS to be kept informed of actions minimizing or avoiding adverse effects or
benefiting listed species or their habitats, the EPA should notify NMFS OPR of any conservation
recommendations it implements in the final action.
Reinitiation Notice
This concludes formal consultation on the EPA's proposed registration of pesticide products
containing carbaryl, carbofuran, and methomyl and their formulations to ESA-listed Pacific
salmonids under the jurisdiction of the NMFS. As provided in 50 CFR 402.16, reinitiation of
formal consultation is required where discretionary federal agency involvement or control over
the action has been retained (or is authorized by law) and if: (1) the extent of take specified in the
Incidental Take Statement is exceeded; (2) new information reveals effects of this action that
may affect listed species or designated critical habitat in a manner or to an extent not previously
considered in this biological opinion; (3) the identified action is subsequently modified in a
manner that causes an effect to the listed species or critical habitat that was not considered in this
Opinion; or (4) a new species is listed or critical habitat designated that may be affected by the
identified action. If reinitiation of consultation appears warranted due to one or more of the
above circumstances, EPA must contact NMFS OPR. If none of these reinitiation triggers are
met within the next 15 years, then reinitiation will be required because the Opinion only covers
the action for 15 years.
DRAFT 506
-------
DRAFT
I!" ' :"-; " ; :•,
(NMFS), N. M. F. S. (2006). Draft Columbia River Estuary Recovery Plan Module, Lower
Columbia River Estuary Partnership under contract to NOAA Fisheries.
(NMFS), N. M. F. S. (2008). Endangered Species Act Section 7 Consultation: Biological
Opinion on Environmental Protection Agency Registration of Pesticides Containing
Chlorpyrifos, Diazinon, and Malathion
482.
Abbas, R., I. R. Schultz, et al. (1996). "Toxicokinetics of parathion and paraoxon in rainbow
trout after intravascular administration and water exposure." Toxicology and Applied
Pharmacology 136(1): 194-199.
Abdel-Halim, K. Y., A. K. Salama, et al. (2006). "Organophosphorus pollutants (OPP) in aquatic
environment at Damietta Governorate, Egypt: Implications for monitoring and biomarker
responses." Chemosphere 63(9): 1491-1498.
Adams, P. (2000). "Memorandum 19 January to Rodney Mclnnis: Status review update for the
steelhead Northern California evolutionarily significant unit. ." available from Southwest
Fisheries Science Center, 110 Shaffer Road, Santa Cruz, California 95060.
Ajawani, S. (1956). "A review of Lake Washington watershed, historical, biological, and
limnological." M.S. thesis. University of Washington, Seattle, Washington.
Allan, J. D. (1995). Stream Ecology: Structure and Function of Running Waters. London, Alden
Press.
Anderson, B. S., J. W. Hunt, et al. (2003). "Integrated assessment of the impacts of agricultural
drainwater in the Salinas River (California, USA)." Environmental Pollution 124(3): 523-532.
Anderson, B. S., J. W. Hunt, et al. (2003). "Ecotoxicologic impacts of agricultural drain water in
the Salinas River, California, USA." Environmental Toxicology and Chemistry 22(10): 2375-
2384.
Anderson, B. S., B. M. Phillips, et al. (2006). "Identifying primary stressors impacting
macroinvertebrates in the Salinas River (California, USA): Relative effects of pesticides and
suspended particles." Environmental Pollution 141(3): 402-408.
Anderson, P. D., D. Dugger, et al. (2007). Surface water monitoring program for pesticides in
salmonid-bearing streams, 2006 monitoring data summary. Washington State Department of
Ecology. Publication No. 07-03-016.: 43 p.
DRAFT 507
-------
DRAFT
Anderson, S. E. (1999). Use of off-channel freshwater wetlands by juvenile chinook and other
salmonids: Potential for habitat restoration in Puget Sound., The Evergreen State College.
Masters: 73
Anderson, T. D. and M. J. Lydy (2002). "Increased toxicity to invertebrates associated with a
mixture of atrazine and organophosphate insecticides." Environmental Toxicology and
Chemistry 21(7): 1507-1514.
Antwi, L. A. K. (1985). "Effects of aerial spraying of chlorphoxin on the brain
acetylchloinesterase activity offish from three rivers in the Ivory Coast, West Africa."
Environmental Pollution 39: 151-159.
Arsenault, J. T. M., W. L. Fairchild, et al. (2004). "Effects of water-borne 4-nonylphenol and 17
beta-estradiol exposures during parr-smolt transformation on growth and plasma IGF-I of
Atlantic salmon (Salmo salar L.)." Aquatic Toxicology 66(3): 255-265.
Arukwe, A. and K. Roe (2008). "Molecular and cellular detection of expression of vitellogenin
and zona radiata protein in liver and skin of juvenile salmon (Salmo salar) exposed to
nonylphenol." Cell and Tissue Research 331(3): 701-712.
Arunachalam, S., K. Jeyalakshmi, et al. (1980). "Toxic and sublethal effects of carbaryl on a
freshwater catfish, Mystus vittatus (B:loch)." Archives of Environmental Contamination and
Toxicology 9: 307-316.
Aspelin, A. L. and A. H. Grube (1999). Pesticides industry sales and usage-1996 and 1997
market estimates: U.S. Environmental Protection Agency, Pesticide Industry Sales and Usage
Report. U.S. Environmental Protection Agency, Biological and Economic Analysis Division.: 39
P-
Bacchetta, R., P. Mantecca, et al. (2008). "Axial-skeletal defects caused by Carbaryl in Xenopus
laevis embryos." Science of the Total Environment 392(1): 110-118.
Bakkala, R. G. (1970). "Synopsis of biological data on the chum salmon, Oncorhynchus keta
(Walbaum) 1792." FAQ Fish. Synop. 41: U.S. Fish. Wildl. Serv. Circ. 315: 89 p.
Balint, T., J. Ferenczy, et al. (1997). "Similarities and differences between the massive eel
(Anguilla anguilla L) devastations that occurred in Lake Balaton in 1991 and 1995."
Ecotoxicology and Environmental Safety 37(1): 17-23.
Barnhart, R. A. (1986). "Species profiles: life histories and environmental requirements of
coastal fishes and invertebrates (Pacific Southwest): steelhead. ." Technical Report, TR-EL-82-
4/872-11-60. Humboldt State University. Arcada. California.
Barren, M. G. andK. B. Woodburn (1995). "Ecotoxicology of chlorpyrifos." Reviews of
Environmental Contamination and Chemistry 144: 1-93.
DRAFT 508
-------
DRAFT
Bax, N. J. (1983). "The early marine migration of juvenile chum salmon (Oncorhynchus ketd)
through Hood Canal - its variability and consequences." Ph.D. thesis. University of
Washington, Seattle, Washington: 196 p.
Baxter, C. V., K. D. Fresh, et al. (2007). "Invading rainbow trout usurp a terrestrial prey subsidy
from native charr and reduce their growth and abundance." Oceologia 153: 461-470.
Beamish, R. J. and C.-E. M. Neville (1995). "Pacific salmon and Pacific herring mortalities in
the Fraser River plume caused by river lamprey (Lampetra ayresi)." Can. J. Fish. Aquat.
ScL(52): 644-650.
Beamish, R. J., B. L. Thomson, et al. (1992). "Spiny dogfish predation on chinook and coho
slamon and the potential effects on hatchery-produced salmon." Trans. Am. Fish. Soc. 121(4):
444-445.
Beechie, T. and S. Bolton (1999). "An approach to restoring salmonid habitat-forming processes
in Pacific Northwest watersheds." Fisheries 24: 6-15.
Beechie, T. J., M. Liermann, et al. (2005). "A classification of habitat types in a large river and
their use by juvenile salmonids." Transactions of the American Fisheries Society 134: 717-729.
Belden, J. B., R. J. Gilliom, et al. (2007). "How well can we predict the toxicity of pesticide
mixtures to aquatic life? ." Integrated Environmental Assessment and Management 3: 364-372.
Belitz, K., S. N. Hamlin, et al. (2004). "Water Quality in the Santa Ana Basin, California, 1999-
2001." U.S. Department of the Interior, U.S. Geological Survey Circular 1238, Reston Virginia.
Available: http://pubs.usgs.gov/circ/2004/1238/pdf/circularl238.pdf (February 2008).
Berg, L. (2001). "Yakima subbasin summary (draft)." Northwest Power Planning Council
Portland, Oregon.
Berman, C. H. (1990). "The effect of elevated holding temperatures on adult spring chinook
reproductive success." M.S. thesis. University of Washington.
Beyers, D. W., M. S. Farmer, et al. (1995). "Effects of rangeland aerial application of Sevin-4-oil
on fish and aquatici invertbrate drift in the Little Missouri River, North Dakota." Archives of
Environmental Contamination and Toxicology 28: 27-34.
Beyers, D. W. and P. J. Sikoski (1994). "Acetylcholinesterase inhibition in federally endangered
Colorado squawfish exposed to carbaryl and malathion." Environmental Toxicology and
Chemistry 13: 935-939.
Bigler, B. (1985). "Kotzebue Sound chum salmon (Oncorhynchus ketd) escapement and return
data, 1962-1984." Alaska Pep. Fish Game ADF&G Tech Data Rep. 149: 112 p.
DRAFT 509
-------
DRAFT
Bilby, R. E. and P. A. Bisson (2001). Function and distribution of large woody debris. River
Ecology and Management Lessons from the Pacific Coastal Ecoregion. R. J. Naiman and R. E.
Bilby. New York, Springer-Verlag New York, Inc.: 324-346.
Bilby, R. E., B. R. Fransen, et al. (2001). "Preliminary evaluation of the use of nitrogen stable
isotope ratios to establish escapement levels for pacific salmon." Fisheries 26(1): 6-14.
Research over the last decade has established the ecological significance of the nutrients
and organic matter deposited by Pacific salmon in the freshwater habitats where they
spawn. A large proportion of the nitrogen in plants and animals in streams where salmon
are abundant may be derived from spawning fish, and juvenile salmonids exhibit higher
growth rates at locations where carcasses are available. Currently, no method is available
to establish salmon escapement goals that meet the nutritional needs of streams. We
examined the relationship between the abundance of spawning coho salmon
(Oncorhynchus kisutch) and the nitrogen stable isotope ratio of coho salmon parr to
determine whether a saturation level for salmon-derived nitrogen could be identified.
Coho parr were collected from 26 sites in western Washington in late winter. The isotope
ratio in the coho parr was related to the abundance of salmon spawning at that site the
previous autumn. The amount of carcass-derived nitrogen increased with increasing
abundance of carcass tissue up to 0.15 kg of carcass/ m(2) of streambed area but
exhibited no increase above this level. These preliminary data suggest that relationships
between stable isotope values and carcass abundance may provide a useful supplement to
traditional methods of establishing escapement goals for Pacific salmon.
Binelli, A., F. Ricciardi, et al. (2005). "Screening of POP pollution by AChE and EROD
activities in Zebra mussels from the Italian Great Lakes." Chemosphere 61(8): 1074-1082.
Bird, S. L., S. G. Perry, et al. (2002). "Evaluation of the AgDisp aerial spray algorithms in the
AgDrift model." Environmental Toxicology and Chemistry 21(3): 672-681.
Bisson, P. A. and R. E. Bilby (2001). Organic matter and trophic dynamics. River Ecology and
Management Lessons from the Pacific Coastal Ecoregion. R. J. Naiman and R. E. Bilby. New
York, Springer-Verlag New York, Inc.: 373-398.
Bisson, P. A., B. E. Rieman, et al. (2003). "Fire and aquatic ecosystems of the western USA:
current knowledge and key questions." Forest Ecology and Management(178): 213-229.
Bisson, P. A., K. Sullivan, et al. (1988). "Channel hydraulics, habitat use, and body form of
juvenile coho salmon, steelhead, and cutthroat trout in streams." Transactions of the American
Fisheries Society 117: 262-273.
Bjorkstedt, E., B. C. Spence, et al. (2005). "An analysis of historical population structure for
evolutionarily signficant units of Chinook salmon, coho salmon, and steelhead in the north-
Central California Coast Recovery Domain." NOAA-TM-NMFS-SWFSC-382.
Bjornn, T., D. Craddock, et al. (1968). "Migration and survival of Redfish Lake, Idaho, Sockeye
salmon, Oncorhynchus nerka" Transactions of the American Fisheries Society 97: 360-373.
DRAFT 510
-------
DRAFT
Bjornn, T. C. and N. Horner (1980). "Biological criteria for classification of Pacific salmon and
steelhead as threatened or endangered under the Endangered Species Act." Idaho Cooperative
Fisheries Research Unit for NMFS.
Blum, J. P. (1988). "Assessment of factors affecting sockeye salmon (Oncorhynchus nerkd)
production in Ozette Lake, WA." Masters Thesis, University of Washington, Seattle,
Washington.
Bocquene, G., C. Bellanger, et al. (1995). "JOINT ACTION OF COMBINATIONS OF
POLLUTANTS ON THE ACETYLCHOLINESTERASE ACTIVITY OF SEVERAL MARINE
SPECIES." Ecotoxicology 4(4): 266-279.
Boomer, R. (1995). "Letter to R. Gustafson, NMFS, from R. Boomer, Project Leader, Western
Washington Fishery Resource Office, Fish and Wildlife Service Quilcene National Fish
Hatchery Log Books, 1920-1958." West coast sockeye salmon administrative record.
Environmental and Technical Services Division, National Marine Fisheries Service, Portland,
Oregon.
Bortleson, G. C., M. J. Chrzastowski, et al. (1980). "Historical changes of shoreline and wetland
at eleven major deltas in the Puget Sound region, Washington." U.S. Geological Survey,
Hydrologic Investigations Atlas HA-617, U.S. Department of Justice and Bureau of Indian
Affairs, Reston, Virginia.
Bortleson, G. C. and J. C. Ebbert (2000). "Occurrence of pesticides in streams and ground water
in the Puget Sound basin, Washington, and British Columbia, 1996-98." United States
Geological Survey, Water-Resources Investigations Report 00-4118, Tacoma, Washington.
Bouldin, J. L., J. L. Farris, et al. (2007). "Assessment of diazinon toxicity in sediment and water
of constructed wetlands using deployed Corbicula fluminea and laboratory testing." Archives of
Environmental Contamination and Toxicology 53(2): 174-182.
Bowman, K. E. and G. W. Minshall (2000). "Assessment of short- and mid-term effects of
wildlife on habitat structure of the Payette National Forest." Prepared by the Stream Ecology
Center, Idaho State University for the Payette National Forest: 45 p.
Bradford, M. J. (1995b). "Partitioning mortality in Pacific salmon." in Emmett, R.L. and M.H.
Schiewe (editors). NOAA-NMFS-NWFSC TM-29: Estuarine and Ocean Survival of
Northeastern Pacific Salmon: Proceedings of the workshop. U.S. Pep. Commer.: 4 p.
Brazner, J. C. and E. R. Kline (1990). "Effects of chlorpyrifos on the diet and growth of larval
fathead minnows, Pimephalespromelas, in littoral enclosures." Canadian Journal of Fisheries
and Aquatic Sciences 47: 1157-1165.
DRAFT 511
-------
DRAFT
Brennan, J. S., K. F. Higgins, et al. (2004). "Juvenile salmon composition, timing, distribution,
and diet in the marine nearshore waters of central Puget Sound in 2001-2002 " King County
Department of Natural Resources and Parks, Seattle, Washington: 164 p.
Bretaud, S., P. Saglio, et al. (2002). "Biochemical and behavioral effects of carbofuran in
goldfish (Carassius auratus)." Environmental Toxicology and Chemistry 21(1): 175-181.
Brett, J. R. (1979). "Environmental factors and growth." in W.S. Hoar. D.J. Randall and J.R.
Brett (editors) Fish Physiology. Academic Press, New York.
Brett, J. R. (1995). "Energetics." Physiological ecology of Pacific salmon. Edited by C.Groot L.
Margolis, and W.C. Clarke. Unversity of British Columbia Press, Vancouver, B.C.: pp. 3-68.
Brewer, S. K., E. E. Little, et al. (2001). "Behavioral dysfunctions correlate to altered physiology
in rainbow trout (Oncorynchus mykiss) exposed to cholinesterase-inhibiting chemicals "
Archives of Environmental Contamination and Toxicology 40: 70-76.
Brodeur, R. D., J. P. Fisher, et al. (2004). "Juvenile salmonid distribution, growth, condition,
origin, and environmental and species associations in the Northern California Current." Fish.
Bull 102: 25-46.
Brown, L. R., P. B. Moyle, et al. (1994). "Historical decline and current status of coho salmon in
California." North American Journal of Fisheries Management 14(2): 237-261.
Brown, R. J., M. Conradi, et al. (1999). "Long-term exposure to 4-nonylphenol affects sexual
differentiation and growth of the amphipod Corophium volutator (Pallas, 1766)." Science of the
Total Environment 233(1-3): 77-88.
Brown, S. B. and W. L. Fairchild (2003). "Evidence for a causal link between exposure to an
insecticide formulation and declines in catch of Atlantic salmon." Human and Ecological Risk
Assessment 9(1): 137-148.
Buchwalter, D. B., J. J. Jenkins, et al. (2003). "Temperature influences on water permeability and
chlorpyrifos uptake in aquatic insects with differing respiratory strategies." Environmental
Toxicology and Chemistry 22(11): 2806-2812.
Buchwalter, D. B., J. F. Sandahl, et al. (2004). "Roles of uptake, biotransformation, and target
site sensitivity in determining the differential toxicity of chlorpyrifos to second to fourth instar
Chironomous riparius (Meigen)." Aquatic Toxicology 66(2): 149-157.
Burgner, R. L. (1991). "The life history of sockeye salmon (Oncorhynchus nerka)." in Groot C.
and L. Margolis (editors). Life history of Pacific salmon. University of British Columbia Press,
Vancouver, British Columbia, Canada.: 3-117.
DRAFT 512
-------
DRAFT
Burke, C., P. Anderson, et al. (2006). Surface water monitoring program for pesticides in
salmonid-bearing streams, 2003-2005. Washington State Department of Ecology. Publication
No. 06-03-036.: 114 p.
Burton, C. A., J. A. Izbicki, et al. (1998). "Water-quality trends in the Santa Ana River at MWD
Crossing and below Prado Dam, Riverside County, California." U.S. Geological Survey Water-
Resources Investigations Report 97-4173. Sacramento, California.
Busack, C. (1990). "Yakima/Klickitat production project genetic risk assessment." Genetics
Unit Washington Department of Fisheries, Olympia, Washington.
Busby, P. J., T. C. Wainwright, et al. (1996). "Status review of steelhead from Washington,
Oregon, and California." U.S. Department of Commerce, National Marine Fisheries Service,
Northwest Fisheries Science Center, Seattle, Washington, NOAA Technical Memorandum
NMFS-NWFSC-27.
Caffrey, J. M. (1996). "Glyphosate in fisheries management." Hydrobiologia 340: 259-263.
Carlson, A. R. (1971). "Effects of long-term exposure to carbaryl (Sevin) on survival, growth,
and reproduction of the fathead minnow (Pimephales promelas)." Journal of Fisheries Research
Board of Canada 29: 583-587.
Carlson, R. W., S. P. Bradbury, et al. (1998). "Neurological effects on the startle response and
escape from predation by medaka exposed to organic chemicals." Aquatic Toxicology 43: 51-68.
Carter, J. L. and V. H. Resh (2005). "Pacific Coast rivers of the coterminous United States."
Pages 541-590 in A. C. Benke and C. E. Gushing, editors. Rivers of North America. Elsevier
Academic Press, Burlington, Massachusetts. Available:
http://books.google.com/books?id=faOUlwkiYFIC&pg=RA3-PA541&lpg=RA3-
PA541&dq=pacific+coast+rivers+of+the+coterminous+united+states&source=web&ots=-
pMpvECFaA&sig=FkGrIiwgkfDvHxXCWXRalK XSvU#PPRl,Ml (February 2008).
Casillas, E. (1999). "Role of the Columbia River estuary and plume in salmon productivity." in
Ocean conditions and the management of the Columbia River salmon, proceeding of a
symposium, July 1, 1999.
Castillo, L. E., E. Martinez, et al. (2006). "Water quality and macroinvertebrate community
response following pesticide applications in a banana plantation, Limon, Costa Rica." Science of
the Total Environment 367(1): 418-432.
CBFWA (1990). "Review of the history, development, and management of anadromous fish
production facilities in the Columbia River basin." Columbia Basin Fish and Wildlife Authority,
Portland, Oregon.
DRAFT 513
-------
DRAFT
CDFG (1965). "California fish and wildlife plan. ." California Department of Fish and Game,
Sacramento, California III supporting data: Part B, inventory salmon-steelhead and marine
resources.
CDFG (1994). "Petition to the California Board of Forestry to list coho salmon (Oncorhynchus
kisutch) as a sensitive species." California Department of Fish and Game, Sacramento,
California.
CDFG (1995). "Letter dated 30 March 1995 to M. Schiewe for the ESA Administrative Record
for West Coast Steelhead." available from Environmental and Technical Services Division,
National Marine Fisheries Service, 525 NE Oregon St., Suite 500, Portland, Oregion 97232.
CDFG (1998). "Report to the Fish and Game Commission: a status review of the spring-run
chinook salmon (Oncorhynchus tshawytschd) in the Sacramento River drainage. California
Department of Fish and Game Candidate Species Status Report 98-01."
CDFG (2003). "letter dated 20 March 2003 from Robert C. Hight to Dr. Michael F. Tillman,
containing review of February 2003 draft of "Preliminary conclusions regarding the updated
status of listed ESUs of West Coast salmon and steelhead."." available from Southwest Fisheries
Science Center, 110 Shaffer Road, Santa Cruz, CA 95060.
CDFG (2007). "Final 2006 California Commercial Landings." Available:
http://www.dfg.ca.gov/marine/landings06.asp (February 2008).
CDPR (2007). Summary of pesticide use report data 2006 indexed by commodity. Sacramento,
CA, California Department of Pesticide Regulation.
CDPR (2007). Summary of Pesticide Use Report Data 2006, Indexed by Commodity.
Sacramento, CA, California Department of Pesticide Regulation: 508.
CDPR (2008). California Department of Pesticide Regulation's Surface Water Database.
http://www.cdpr. ca. gov/docs/emon/surfwtr/surfdes .htm.
CDPR (2009). Active pesticide product label database query.
CEPA (2007). "California EPA TMDL Program - California's 2006 Section 303(d) List of Water
Quality Limited Segments. 2007. Downloaded September 26, 2008. Full dataset available at
http://www.swrcb.ca.gov/water issues/programs/tmdl/303d Iists2006 approved.shtml."
CEPA (2007). "California EPA TMDL Program - Summary statistics for 2006 303(d) List.
Download September 26, 2008. Available at
http://www.swrcb.ca.gov/water issues/programs/tmdl/303d Iists2006 approved.shtml."
Chapman, D., G. T. Hillman, et al. (1994). "Status of summer/fall Chinook salmon in the mid-
Columbia region." Report to Chelan, Douglas, and Grant County PUDs, Boise, Idaho.
DRAFT 514
-------
DRAFT
Chapman, D. and K. Witty (1993). "Habitat of weak salmon stocks in the Snake River basin and
feasible recovery measures." Report to the Bonneville Power Administration, DOE/BP-99654-1,
Portland, Oregon.
Chapman, D. W. and T. C. Bjornn (1969). "Distribution of salmonids in strams, with special
refernce to food and feeding." pages 153-176 m Northcote, T.G. (editor). Symposium on salmon
and trout in streams H.R. MacMillan Lectures in Fisheries, University of British Columbia,
Institute of Fisheries, Vancouver, British Columbia, Canada.
Chapman, R. A. and C. M. Cole (1982). "Observations on the influence of water and soil pH on
the persistence of insecticides." Journal of Environmental Science and Health B17: 487-504.
CIG (2004). Overview of climate change impacts in the U.S. Pacific Northwest. Univeristy of
Washington, Seattle, Washington, Climate Impacts Group.
Clarke, G. H. (1929). "Sacramento-San Joaquin salmon (Oncorhynchus tschawytscha) fishery of
California. California Fish and Game Bulletin." 17(73).
Cold, A. and V. E. Forbes (2004). "Consequences of a short pulse of pesticide exposure for
survival and reproduction of Gammarus pulex." Aquatic Toxicology 67(3): 287-299.
Colgrove, D. J. and J. W. Wood (1966). "Occurrence and control of Chondrococcus columnaris
as related to Fraser River sockeye salmon." Int. Pac. Salmon Fish. Comm. Prog. Rep. No. 15.
Collier, T. K., J. E. O'Neill, et al. (2006). "Toxic chemical contaminants and Puget Sound."
NOAA Fisheries, Northwest Fisheries Science Center, Environmental Conservation Division,
Seattle, Washington and Washington Department of Fish and Wildlife, Fish Program, Marine
Resources, Olympia, Washington.
Collins, B. D. and A. J. Sheikh (2005). "Historical reconstruction, classification, and change
analysis of Puget Sound tidal marshes." Final report to Washington Department of Natural
Resources Aquatic Resources Division, Olympia, Washington. Available:
http://riverhi story.ess.Washington.edu/project_reports/fmalrpt_rev_augl2_2005.pdf (February
2008).
Collis, K. (2007). "Research, monitoring, and evaluation of avian predation on salmonid smolts
in the lower and mid-Columbia River - 2006 season summary." U.S. Geological survey report to
Bonneville Power Administration and the U.S. Army Corps of Engineers.
Colville, A., P. Jones, et al. (2008). "Effects of chlorpyrifos on macroinvertebrate communities
in coastal stream mesocosms." Ecotoxicology 17(3): 173-180.
Conaway, C. H., S. Squire, et al. (2003). "Mercury speciation in the San Francisco Bay estuary."
Marine Chemistry 80(T2-3V): 199-225.
DRAFT 515
-------
DRAFT
Connor, W. P., J. G. Sneva, et al. (2005). "Two alternative juvenile life history types for fall
Chinook salmon in the Snake River basin." Transactions of the American Fisheries Society 134:
291-304.
Coppage, D. L. and E. Matthews (1974). "Short-term effects of organophosphate pesticides on
cholinesterases of estuarine fishes and pink shrimp." Bulletin of Environmental Contamination
and Toxicology 11: 483-488.
Courtemanch, D. L. and K. E. Gibbs (1980). "Short- and long-term effects of forest spraying of
carbaryl (sevin-4-oil) on stream invertebrates." Canadian Entomology 112: 2741-276.
Creekman, L. L. and E. F. Hurlburt (1987). "Control of burrowing shrimp on oyster beds in
Willapa Bay and Grays Harbor. 1985. Washington State Department of Fisheries. Special
Shellfish Report No. 3. ."27.
Crepeau, K. L. and K. M. Kuivila (2000). "Rice pesticide concentrations in the Colusa Basin
Drain and the Sacramento River, California, 1990-1993." Journal of Environmental Quality
29(3): 926-935.
Cuffney, T. F., J.B. Wallace and J.R. Webster (1984). "Pesticide manipulation of a headwater
stream: invertebrate responses and their significance for ecosystem processes." Freshwater
Invertebrate Biology 3(4): 153-171.
Cuffney, T. F., M. R. Meador, et al. (1997). Distribution offish, benthic invertebrate, and algal
communities in relation to physical and chemical conditions, Yakima River Basin, Washington,
1990. U. S. G. Survey: 1-94.
Davies, P. E. and L. S. J. Cook (1993). "Catastrophic macroinvertebrate drift and sublethal
effects on brown trout, Salmo trutta, caused by cypermethrin spraying on a Tasmanian stream."
Aquatic Toxicology 27: 201-224.
de Vlaming, V., C. DiGiorgio, et al. (2004). "Irrigation runoff insecticide pollution of rivers in
the Imperial Valley, California (USA)." Environmental Pollution 132(2): 213-229.
Demko, D. B. and S. P. Cramer (2000). "Effects of pulse flows on juvenile Chinook migration in
the Stanislaus River." Report of S.P. Cramer & Associates, Inc. to South San Joaquin Irrigation
District Manteca, California and Oakdale Irrigation District, Oakdale, California.
Detra, R. L. and W. J. Collins (1986). "Characterization of Cholinesterase Activity in Larval
Chironomus-Riparius Meigen (=Thummi Kiefer)." Insect Biochemistry 16(4): 733-739.
Detra, R. L. and W. J. Collins (1991). "The Relationship of Parathion Concentration, Exposure
Time, Cholinesterase Inhibition and Symptoms of Toxicity in Midge Larvae (Chironomidae,
Diptera)." Environmental Toxicology and Chemistry 10(8): 1089-1095.
DRAFT 516
-------
DRAFT
Dimick, R. E. and F. Merryfield (1945). "The fishes of the Willamette River system in relation to
pollution." Oregon State College Engineering Experiement Station. Bulletin Series No. 20.
Dineen, G. S., S. C. Harrison, et al. (2007). "Growth, production and bioenergetics of brown
trout in upland streams with contrasting riparian vegetation." Freshwater Biology 52: 771-783.
Dodson, J. and C. Mayfield (1979). "Modification of the rheotropic response of rainbow trout
(Salmon gairdneri) by sublethal doses of the aquatic herbicides diquat and simizine."
Environmental Pollution 18: 147-157.
Domagalski, J. (2000). "Pesticides in Surface Water Measured at Select Sites in the Sacramento
River Basin, California, 1996-1998." U.S. GEOLOGICAL SURVEY Water-Resources
Investigations Report 00-4203 National Water-Quality Assessment Program.
Doty, D. C., D. A. Armstrong, et al. (1990). "Comparison of carbaryl pesticide impacts on
dungeness crab (Cancer magister) versus benefits of habitat derived from oyster culture in
Willapa Bay, Washington. Report to the Willapa Bay/Grays Harbor Oyster Growers Association.
FRI-UW-9020. University of Washington, Seattle Washington.".
Douglas, W. S., A. Mclntosh, et al. (1993). "TOXICITY OF SEDIMENTS CONTAINING
ATRAZINE AND CARBOFURAN TO LARVAE OF THE MIDGE CHIRONOMUS-
TENTANS." Environmental Toxicology and Chemistry 12(5): 847-853.
Dubrovsky, N. M., C. R. Kratzer, et al. (1998). "Water Quality in the San Joaquin-Tulare Basins,
California, 1992-1995." U.S. Department of the Interior. U.S. Geological Survey circular 1159.
Reston, Virginia. Available: http://pubs.usgs.gov/circ/circll59/circll59.pdf (February 2008).
Dumbauld, B. R., K. M. Brooks, et al. (2001). "Response of an estuarine benthic community to
application of the pesticide carbaryl and cultivation of Pacific oysters (Crassostrea gigas) in
Willapa Bay, WA." Marine Pollution Bulletin 42(10): 826-844.
Ebbert, J. and S. Embry (2001). "Pesticides in surface water of the Yakima River Basin,
Washington, 1999-2000 - their occurrence and an assessment of factors affecting concentraitons
and loads." U.S. Geological Survey, Water Investigations Report 01-4211, Portland Oregon.
Ebbert, J. C., S. S. Embrey, et al. (2000). "Water quality in the Puget Sound basin, Washington
and British Columbia, 1996-98." U.S. Department of the Interior. U.S. Geological Survey
Circular 1216, Reston, Virginia. Available: http://pubs.usgs.gov/circ/circl216/pdf/circl216.pdf
(February 2008).
Eder, K. J., H. R. Kohler, et al. (2007). "Pesticide and pathogen: Heat shock protein expression
and acetylcholinesterase inhibition in juvenile Chinook salmon in response to multiple stressors."
Environmental Toxicology and Chemistry 26(6): 1233-1242.
Ehlke, R. D. and K. Keller (2003). "2002 Chum salmon spawnign ground surveys on the
mainstem Columbia River and its Washington tributaries." Prepared for the Pacific States Marine
DRAFT 517
-------
DRAFT
Fisheries Commission, Vancouver, Washington by the Bonneville Power Administration,
Portland, Oregon
EPA (1998). "Guidelines for ecological risk assessment." EPA/630/R-95/002F published on May
14. 1998. Federal Register 63(93): 26846-26924.
EPA (1998). "Reregistration eligibility decision (RED) - Methomyl." 248.
EPA (2003). "Carbaryl analysis of risk to endangered and threatened salmon and steelhead." 69.
EPA (2003). "Methomyl analysis of risk to endangered and threatened salmon and steelhead."
247.
EPA (2004). "Carbofuran analysis of risks to endangered and threatened salmon and steelhead."
69.
EPA (2004). Overview of the Ecological Risk Assessment Process in the Office of Pesticide
Programs, U.S. Environmental Protection Agency - Endangered and Threatened Species Effects
Determinations. O. o. P. Resources.
EPA (2006). "Interim reregi strati on eligibility decision - Carbofuran." 38.
EPA (2007). Environmental Protection Agency, Fish and Wildlife Service, and National Marine
Fisheries Service Pesticide Coordination Meeting. December 10 - 12, 2007. National
Conservation Training Center, Shepherdstown, WV
EPA (2007). "Reregistration eligibility decision (RED) for carbaryl." 43.
EPA (2007). "Revised N-Methyl Carbamate Cumulative Risk Assessment. ." 277.
EPA (2007). "Risks of carbaryl use to the federally-listed California red legged frog (Rana
aurora draytonii)." 160.
EPA (2007). "Risks of methomyl use to the federally listed California redLegged
frog (Rana aurora draytonii)." 270.
EPA (2008). "Amended reregi strati on eligibility decision for Carbaryl." 55.
EPA (2008). Inert Ingredients Permitted for Use in Nonfood Use Pesticide Products: Last
Updated January 7, 2008. U.S. Environmental Protection Agency, Office of Prevention,
Pesticides and Toxic Substance: 93 p.
DRAFT 518
-------
DRAFT
Fairchild, W. L., E. O. Swansburg, et al. (1999). "Does an association between pesticide use and
subsequent declines in catch of Atlantic salmon (Salmo salar) represent a case of endocrine
disruption?" Environmental Health Perspectives 107(5): 349-357.
FCRPS (2008). "Endangered Species Act Section 7(a)(2) Consultation Biological Opinion and
Magnuson-Stevens Fishery Conservation and Management Act Essential Fish Habitat
Consultation Consultation on remand for operaiton of the Federal Columbia River Poser System,
11 Bureau of Reclamation Projects in the Columbia Basin and ESA Section 10(a)(l)(A) permit
for juvenile fish transportaion program."
Feldman, K. L., D. A. Armstrong, et al. (2000). "Oysters, crabs, and burrowing shrimp: review
of an environmental conflict over aquatic resources and pesticide use in Washington State's
(USA) Coastal Estuaries." Estuaries 23(2): 141-176.
Ferguson, J. (2006). "Estimated survival of juvenile salmonids through the lower Columbia
River and estuary, and mortality due to avian predation " NOAA Fisheries, Northwest Fisheries
Science Center, Seattle, Washington: 2 p.
Flagg, T. A., F. W. Waknitz, et al. (1995). "The effect of hatcheries on native coho salmon
populations in the lower Columbia River." Pages 366-375 in H. Schramm and R. Piper, editors.
Uses and effects of cultured fishes in aquatic systems. American Fisheries Society, Bethesda,
Maryland.
Flickinger, E. L., C. A. Mitchell, et al. (1986). "BIRD POISONING FROM MISUSE OF THE
CARBAMATE FURADAN IN A TEXAS RICE FIELD." Wildlife Society Bulletin 14(1): 59-
62.
Foott, J. S., R. Harmon, et al. (2003). "Ceratomyxosis resistance in juvenile Chinook salmon and
steelhead trout from the Klamath River, 2002 Investigational Report." U.S. Fish and Wildlife
Service, California-Nevada Fish Health Center, Anderson, California. [768 Kbl at KRIS Web
site.
Forbes, V. E. and A. Cold (2005). "Effects of the pyrethroid esfenvalerate on life-cycle traits and
population dynamics of Chironomus riparius - Importance of exposure scenario." Environmental
Toxicology and Chemistry 24(1): 78-86.
Ford, J. K. B. and G. M. Ellis (2006). "Selective foraging by fish-eating killer whales Orcinus
orca in British Columbia." Marine Ecology Progress Series(316): 185-199.
Foss, S. F., P. R. Ode, et al. (2007). "Non-indigenous aquatic organisms in the coastal waters if
California. ." California Fish and Game 93(3): 111-129.
Fresh, K. D. (1997). The role of competition and predation in the decline of Pacific salmon and
steelhead. Pacific salmon and their ecosystems: status and future options D. J. Stouder, P. A.
Bisson and R. J. Naiman. New York, Chapman and Hall: 245-275.
DRAFT 519
-------
DRAFT
Fresh, K. L., E. Casillas, et al. (2005). "Role of the estuary in the recovery of Columbia River
basin salmon and steelhead and evaluation of the effects of selected factors on salmonid
population viability." U.S. Dept. Commer.. NOAA Tech. Memo. NMFS-NWFSC-69: 105.
Friesen, T. A., J. S. Vile, et al. (2004). "Migratory behavior, timing, rearing, and habitat use of
juvenile salmnids in the lower Willamette River." Oregon Department of Fish and Wildlife,
Columbia River Investigations, Clackamas, Oregon: 63-119.
Friesen, T. A. and D. L. Ward (1999). "Management of the northern pikeminnow and
implcations for juvenile salmonid survival in the lower Columbia and Snake rivers.
." North American Journal of Fisheries Management 19): 406-420.
Fry, D. H., Jr., (1961). "King salmon spawning stocks of the California Central Valley, 1940-
1959." California Department of Fish and Game 47(1): 55-71.
Fuhrer, G. J., J. L. Morace, et al. (2004). "Water quality in the Yakima Basin, Washington, 1999-
2000." U.S. Department of the Interior, U.S. Geological Survey Circular 1237, water research
investigations report 03-4026, Portland, Oregon. Available:
http://pubs.usgs.gov/wri/wri034026/pdf/wri034026.pdf (February 2008).
Fulton, L. A. (1968). "Spawning areas and abundance of Chinook salmon, Oncorhynchus
tshawytscha, in the Columbia River basin - past and present." U.S. Fish and Wildlife Service
Spec. Sci. Rep. Fish. 571.
Fulton, M. H. and P. B. Key (2001). "Acetylcholinesterase inhibition in estuarine fish and
invertebrates as an indicator of organophosphorus insecticide exposure and effects."
Environmental Toxicology and Chemistry 20(1): 37-45.
Geisy, J. P., K. R. Solomon, et al. (1999). Chlorpyrifos: Ecological Risk Assessment in North
American Aquatic Environments. New York City, Springer-Verlag.
Giddings, J. M., L. W. Hall, et al. (2000). "Ecological risks of diazinon from agricultural use in
Sacramento - San Joaquin River Basins, CA." Risk Analysis 20(5): 545-572.
Giesy, J. P., K. R. Solomon, et al. (1998). Ecological risk assessment of chlorpyrifos in North
American aquatic environments: Report of an expert panel. Williamston, Dow AgroSciences:
212.
Gilliom, R. J., J. E. Barbash, et al. (2006). The Quality of Our Nation's Waters-Pesticides in the
Nation's Streams and Ground Water, 1992-2001: 172.
Gilliom, R. J., J. E. Barbash, et al. (2006). "The quality of our nation's waters; pesticides in the
nation's streams and groundwater, 1992-2001." U.S. Geological Survey circular 1291: 52 p.
Glick, P. (2005). "Fish out of water: a guide to global warming and Pacific Northwest rivers."
National Wildlife Federation, Reston, Virginia: 25 p.
DRAFT 520
-------
DRAFT
Good, T. P., R. S. Waples, et al. (2005). "Updated status of federally listed ESUs of West Coast
salmon and steelhead." U.S. Department of Commerce, NOAA Technical Memorandum NMFS-
NWFSC-66.
Good, T. P., R. S. Waples, et al. (2005). Updated status of federally listed ESUs of West Coast
salmon and steelhead. NOAA Technical Memorandum. Seattle, WA, U.S. Department of
Commerce. NMFS-NWFSC-66: 1-598.
Grange, G. F. (2002). Organophosphate insecticides and in-situ inhibition of acetylcholine
esterase activity in steelhead (Oncorhynchus mykiss). Environmental Sciences and Resources.
Portland, Portland State University. Master of Science: 104.
Greco, L., E. Capri, et al. (2007). "Biochemical responses in Salmo salar muscle following
exposure to ethynylestradiol and tributyltin." Chemosphere 68(3): 564-571.
Gregory, S. V. and P. A. Bisson (1997). "Degradation and loss of anadromous salmonid habitat
in the Pacific Northwest." pages 277-314 in Stouder D.J, Bisson, P. A, Naiman, R.J. (editors)
Pacific salmon and their ecoystems. New York: Chapman and Hall.
Greswell, R. E. (1999). "Fire and aquatic ecosystems in forested biomes of North America."
Transactions of the American Fisheries Society 128(2): 193-221.
Groot, C. and T. P. Quinn (1987). "Homing migration of sockeye salmon, Oncorhynchus nerka,
to the Fraser River." Fishery Bulletin 85(3).
Guillen, G. (2003). "Klamath River fish die-off, September 2002: causative factors of mortality."
Report number AFWO-F-02-03. U.S. Fish and Wildlife Service. Arcata Fish and Wildlife
Office. Arcata, California 128 p.
Gustafson, R. G., T. C. Wainwright, et al. (1997). "Status review of sockeye salmon from
Washington and Oregon." NOAA Techical Memorandum NMFS NWFSC: 282p.
Haggerty, M. J., A. C. Ritchie, et al. (2007). "Lake Ozette sockeye limiting factors analysis: draft
8_1." Prepared for the Makah Indian Tribe and NOAA Fisheries in cooperation with the Lake
Ozette Sockeye steering Committee, Port, Angeles, Washington.
Haines, T. A. (1981). "Effects of an application of carbaryl on brook trout (Salvellinus
fontinalls)" Bulletin of Environmental Contamination and Toxicology 27: 534-542.
Hall, L. W., W. D. Killen, et al. (2006). "Characterization of benthic communities and physical
habitat in the Stanislaus, Tuolumne, and Merced Rivers, California." Environmental Monitoring
and Assessment 115(1-3): 223-264.
Hanazato, T. and M. Yasuno (1989). "EFFECTS OF CARBARYL ON THE SPRING
ZOOPLANKTON COMMUNITIES IN PONDS." Environmental Pollution 56(1): 1-10.
DRAFT 521
-------
DRAFT
Hanazato, T. and M. Yasuno (1990). "Influence of time of application on an insecticide on
recovery patterns of a zooplanlton community in experimental ponds." Archives of
Environmental Contamination and Toxicology 19: 77-83.
Hanson, B., R. W. Baird, et al. (2005). "Focal behavioral observations and fish-eating killer
whales: improving our understanding of foraging behaviour and prey selection." Abstract at the
16th Biennial Conference on the Biology of Marine Mammals, San Diego, California.
Hard, J., R. P. Jones, et al. (1992). "Pacific salmon and artificial propagation under the
endangered species act." U.S. Department of Commerce, National Oceanic and Atmospheric
Administration, National Marine Fisheries Service, Northwest Fisheries Science Center,
Technical Memorandum NMFS-NWFSC-2, Seattle, Washington. Available:
http://www.nwfsc.noaa.gov/publications/techmemos/tm2/tm2.html (February 2008).
Hardersen, S. and S. D. Wratten (1998). "The effects of carbaryl exposure of the penultimate
larval instars of Xathocnemis zealandica on emergence and fluctuating asymmetry."
Ecotoxicology 7(5): 297-304.
Hare, S. R., N. J. Mantua, et al. (1999). "Inverse production regimes: Alaskan and west coast
salmon." Fisheries 24(1): 6-14.
Hartt, A. C. and M. B. Dell (1986). "Early oceanic migrations and growth of juvenile Pacific
salmon and steelhead trout." International North Pacific Fisheries Commission Bulletin 47.
Hatakeyama, S., H. Shiraishi, et al. (1990). "Effects of aerial spraying of insecticides on
nontarget macrobenthos in a mountain stream." Ecotoxicology and Environmental Safety 19:
254-270.
Havens, K. E. (1995). "INSECTICIDE (CARBARYL, 1-NAPTHYL-N-METHYL
CARBAMATE) EFFECTS ON A FRESH-WATER PLANKTON COMMUNITY -
ZOOPLANKTON SIZE, BIOMASS, AND ALGAL ABUNDANCE." Water Air and Soil
Pollution 84(1-2): 1-10.
Healey, M. C. (1991). "Life history of Chinook salmon (Oncorhynchus tshawytscha)" pages
311-394 in Groot C. and Margolis (editors). Pacific salmon life histories. University of British
Columbia Press. Vancouver, Canada.
Hebdon, J. L., P. Kline, et al. (2004). "Evaluating reintroduction strategies for Redfish Lake
sockeye salmon captive broodstock strategy." pages 401-413 in Nickum, M.J., P.M. Mazik, J.G.
Nickum, and P.P. MacKinlay (editors). Propogated fish in resource management. American
Fisheries Society, Symposium 44, American Fisheries Society, Bethesda, Maryland.
Heckmann, L. H. and N. Friberg (2005). "Macroinvertebrate community response to pulse
exposure with the insecticide lambda-cyhalothrin using in-stream mesocosms." Environmental
Toxicology and Chemistry 24(3): 582-590.
DRAFT 522
-------
DRAFT
Hellou, J., J. Leonard, et al. (2009). "Comparison of the partitioning of pesticides relative to the
survival and behaviour of exposed amphipods." Ecotoxicology 18(1): 27-33.
Hemmer, M. J., C. J. Bowman, et al. (2002). "Vitellogenin mRNA regulation and plasma
clearance in male sheepshead minnows, (Cyprinodon variegatus) after cessation of exposure to
17[beta]-estradiol and p-nonylphenol." Aquatic Toxicology 58(1-2): 99-112.
Henning, J. A., R. E. Gresswell, and I. A. Fleming (2006). "Juvenile salmonid use of freshwater
emergent wetlands in the floodplain and its implications for conservation management." North
American Journal of Fisheries Management 26: 367-376.
Henry, M. and G. Atchison (1984). "Behavioral effects of methyl-parathionon social groups of
bluegill (Lepomis macrochirus)." Environmental Toxicology and Chemistry 3: 399-408.
Higgs, D. A., J. S. Macdonald, et al. (1995). Nutrition and feeding habits in relation to life
history stage. Physiological Ecology of Pacific Salmon. C. Groot, L. Margolis and W. C. Clarke.
Vancouver, Canada, University of British Columbia Press: 159-315.
Hinck, J. E., C. J. Schmitt, et al. (2004). "Biomonitoring of environmental status and trends
(BEST) program: environmental contaminants and their effects on fish in the Columbia River
basin." U.S. Department of the Interior, U.S. Geological Survey, Columbia Environmental
Research Center, scientific investigation report 2004-5154, Columbia, Missouri.
Hirsch, R. M., W.M. Alley, and W.G. Wilbur (1988). Concepts for a national water-quality
assessment program.
Holland, H. T., D. L. Coppage, et al. (1967). "Use offish brain acetylcholinesterase to monitor
pollution of organophosphate pesticides." Bulletin of Environmental Contamination and
Toxicology 2: 156-162.
Howard, T. E. (1975). "Swimming performance of juvenile coho salmon (Oncorhynchus kisutch)
exposed to bleached kraft pulp mill effluent." Journal of Fisheries Research Board of Canada 32:
789-793.
Howell, P., K. Jones, et al. (1985). "Stock assessment of Columbia River anadromous
salmonids." Final report to the Bonneville Power Administration for Contract No. DE-A179-
84BP12737, Bonneville Power Administration, Portland, Oregon II: Steelhead stock
summaries, stock transfer guidelines, and information needs.
Hoy, T., T. E. Horsberg, et al. (1991). "Inhibition of Acetylcholinesterase in Rainbow-Trout
Following Dichlorvos Treatment at Different Water Oxygen Levels." Aquaculture 95(1-2): 33-
40.
DRAFT 523
-------
DRAFT
Hunt, J. W., B. S. Anderson, et al. (2003). "Ambient toxicity due to chlorpyrifos and diazinon in
a central California coastal watershed." Environmental Monitoring and Assessment 82(1): 83-
112.
Hurlburt, E. F. (1986). "Control of burrowing shrimp on oyster beds in Willapa Bay and Grays
Harbor. 1984. Washington State Department of Fisheries. Special Shellfish Report No. 2. ." 40.
Hutchinson, T. H., G. T. Ankley, et al. (2006). "Screening and testing for endocrine disruption in
fish - Biomarkers as "signposts," not "traffic lights," in risk assessment." Environmental Health
Perspectives 114: 106-114.
ICBTRT (2003). "Independent populations of chinook, steelhead, and sockeye for listed
evolutionarily significant units within the Interior Columbia River Domain." Northwest Fisheries
Science Center, Seattle, Washington.
ICBTRT (2005). "Viability criteria for application to Interior Columbia Basin salmonid ESUs."
112 p.
IEPSPWT (1999). "Monitoring, assessment, and research on central valley steelhead: status of
knowledge, review of existing programs, and assessments of needs in comprehensive
monitoring, assessment and research program plan." Technical Application VII-A-11.
Incardona, J. P., T. K. Collier, et al. (2004). "Defects in cardiac function precede morphological
abnormalities in fish embryos exposed to poly cyclic aromatic hydrocarbons." Toxicology and
Applied Pharmacology 196: 191205.
IPCC (2001). "Climate change 2001: impacts adaptation and vulnerability, summary for policy
makers and technical summary. IPCC, Geneva, Switzerland."
ISG (1996). "Return to the river: restoration of salmonid fishes in the Columbia River
ecosystem." Northwest Power Planning Council, Independent Science Group report 96-6,
Portland, Oregon. Available:
http://www.ecy.wa.gov/programs/WR/wstf/images/pdf/normrivr.pdf (February 2008).
Ishibashi, H., M. Hirano, et al. (2006). "Reproductive effects and bioconcentration of 4-
nonylphenol in medaka fish (Oryzias latipes)." Chemosphere 65(6): 1019-1026.
Jacobs, R., G. Larson, et al. (1996). "The sockeye salmon Oncorhynchus nerka population in
Lake Ozette, Washington, USA." Tech. Rep. NPS/CCSOSU/NRTR-96/04. Available from
Denver Service Center, Technical Information Center, P. O. Box 25287, Denver, Colorado
80225-0287.: 104 p.
Jardine, T. D., D. L. MacLatchy, et al. (2005). "Development of a short-term in situ caging
methodology to assess long-term effects of industrial and municipal discharges on salmon
smolts." Ecotoxicology and Environmental Safety 62(3): 331-340.
DRAFT 524
-------
DRAFT
Jarrard, H. E., K. R. Delaney, et al. (2004). "Impacts of carbamate pesticides on olfactory
neurophysiology and cholinesterase activity in coho salmon (Oncorhynchus kisutch)" Aquatic
Toxicology 69: 133-148.
Jenkins, J., P. Jepson, et al. (2004). Watershed-based Ecological Risk Assessment of Pesticide
Use in Western Oregon: A Conceptual Framework. Report to Oregon Department of Agriculture.
Available online at:
http://www.ipmnet.org/Pesticide Toxicology/Reports Other/PDA watershed eco-
risk final report 04-28-04.pdf
Jin-Clark, Y., M. J. Lydy, et al. (2002). "Effects of atrazine and cyanazine on chlorpyrifos
toxicity in Chironomus tentans (Diptera : Chironomidae)." Environmental Toxicology and
Chemistry 21(3): 598-603.
JISAO (2007). "Pacific Northwest impacts of climate change." accessed at
http://www.jisao.washington.edu/main.html accessed on 6/22/08: 9 p.
Johnson, A. (2001). "Carbaryl concentrations in Willapa Bay and recommendations for water
quality guidelines. Washington Department of Ecology, Environmental Assesssment Program.
Publication No. 01-03-005."
Johnson, A. and J. Cowles (2003). Washington state surface water monitoring program for
pesticides in salmon habitat for two index watersheds: A study for the Washington State
Department of Agriculture conducted by the Washington State Department of Ecology.
Publication No. 03-03-104. : 38 p.
Johnson, A. and A. Newman (1983). "Water quality in the gap-to-gap reach of the Yakima
River, June - October 1982." Washington Department of Ecology, Olympia, Washington.
Available: http://www.ecy.wa.gov/biblio/83el7.html (February 2008).
Johnson, B. T. (1986). "Potential impact of selected agricultural chemical contaminants pn a
norhtern prairie wetland: A microcosm evaluation." Environmental Toxicology and Chemistry 5:
473-485.
Johnson, K. R., P. C. Jepson, et al. (2008). "Esfenvalerate-induced case-abandonment in the
larvae of the caddisfly (Brachycentrus americanus)." Environmental Toxicology and Chemistry
27(2): 397-403.
Johnson, L., T. K. Collier, et al. (2002). "An analysis in support of sediment qulaity thresholds
for poly cyclic aromatic hydrocarbons (PAHs) to protect estuarine fish." Aquatic Conservation:
Marine Freshwater Ecosystems 12: 517-538.
Johnson, M. L. (1964). "San Lorenzo River System, Santa Cruz County." CDFG unpublished
file report. Administrative Record for coho salmon in Scott and Waddell Creeks. National
Marine Fisheries Service, Southwest Regional Office, Long Beach, California.
DRAFT 525
-------
DRAFT
Johnson, O. W., W. S. Grant, et al. (1997). "Status review of chum salmon from Washington,
Oregon, and California." National Marine Fisheries Service, Northwest Fisheries Science Center,
NOAA Technical Memorandum NMFS-NWFSC-32, Seattle. Washington: 280 p.
Jones, R. D. (2009). Comments on carbofuran in draft description of the federal action. February
3, 2009 email from R. David Jones, EPA Office of Pesticide Programs to Tony Hawkes, NMFS
Office of Protected Resources.
Joy, J. (2002). "Upper Yakima River basin suspended sediment and organochlorine pesticide
total maximum daily load evaluation." Washington Department of Ecology, publication number
02-30-012. Olympia, Washington. Available: http://www.ecv.wa.gov/biblio/0203012.html
(February 2008).
Kagan, J. S., J. C. Hak, et al. (1999). "Oregon gap analysis project final report:a geographic
approach to planning for biological diversity." Oregon Natural Heritage Program, Portland,
Oregon.
Kallander, D. B., S. W. Fisher, et al. (1997). "Recovery following pulsed exposure to
organophosphorus and carbamate insecticides in the midge, Chironomus riparius" Archives of
Environmental Contamination and Toxicology: 29-33.
Kalleberg, H. (1958). "Observations in a stream tank of territoriality and competition in juvenile
salmon and trout (Salmo solar L. and Salmo trutta L.)." Report of the Institute of Freshwater
Research, Drottningholm 39: 55-98.
Kammerer, J. C. (1990). "Largest rivers in the United States." Water Fact Sheet. U.S.
Department of the Interior, U.S. Geological Survey, open file report 87-242, Reston, Virginia.
Available: http://pubs.usgs.gov/of/1987/ofr87-242/pdf/ofr87242.pdf (February 2008).
Karnak, R. E. and W. J. Collins (1974). "The susceptibility to selected insecticides and
acetylcholinesterase activity in a laboratory colony of midge larvae, Chironmus tentans (Diptera:
Chironomidae)." Bulletin of Environmental Contamination and Toxicology 12: 62-69.
Kassler, T. W. and A. Marshall (2004). GSI Estimate of stock group contributions to the 2003
Willapa Bay
Chinook Fishery. February 10, 2004 memorandum to Tim Flint. Washington Department of Fish
and Wildlife.
Kemmerich, J. (1945). "A review of the artificial propagation and transplantation of the sockeye
salmon of the Puget Sound area in the State of Washington conducted by the federal government
from 1896 to 1945." Report to the Regional Director, U.S. Fish and Wildlife Service.
Kiely, T., D. Donaldson, et al. (2004). Pesticides Industry Sales and Usage
2000 and 2001 Market Estimates. U.S. Environmental Protection Agency, Biological and
Economic Analysis Division: 33 p.
DRAFT 526
-------
DRAFT
Kier Associates (1991). "Long range plan for the Klamath River Basin Conservatin Area Fishery
Restoration Program." U.S. Fish and Wildlife Service Klamath River Fishery Resource Office.
Yreka. California. [8.5 MB1 at KRIS Web site: 403 p.
King County (2002d). "Literature review of endocrine disrupters in secondary treated effluent:
toxicological effects in aquatic organisms. King County Department of Natural Resources,
Seattle, Washington 98104."
Koplin, D. W., E. T. Furlong, et al. (2002). "Pharmaceuticals, hormones, and other organic
wastewater contaminants in U.S. Streams, 1999-2000: A national reconnaissance."
Environmental Science and Technology 36(6): 1202-1211.
Koski, K. V. (1975). "The survival and fitness of two stocks of chum salmon (Oncorhynchus
keta) from egg deposition to emergence in a controlled-stream environment at Big Beef Creek."
PhD, thesis. University of Washington, Seattle, Washington: 212 p.
Kreutzweiser, D. P. and P. K. Sibley (1991). "INVERTEBRATE DRIFT IN A HEADWATER
STREAM TREATED WITH PERMETHRIN." Archives of Environmental Contamination and
Toxicology 20(3): 330-336.
Kruckeberg, A. R. (1991). "The natural history of Puget Sound Country." University of
Washington Press, Seattle, Washington.
La Riviere, M. G. (1991). "The Ozette Lake sockeye salmon enhancement program." Makah
Fisheries Management Department uppubl. Rep.: 9 p.
Labenia, J. S., D. H. Baldwin, et al. (2007). "Behavioral impairment and increased predation
mortality in cutthroat trout exposed to carbaryl." Marine Ecology-Progress Series 329: 1-11.
Laetz, C. A., D. H. Baldwin, et al. (In Press). "The synergistic toxicity of pesticide mixtures:
Implications for ecological risk assessment and the conservation of threatened Pacific salmon."
Environmental Health Perspectives ehponline.org: doi: 10.1289/ehp.0800096.
Laufle, J. C., G. B. Pauley, et al. (1986). "Species profiles: life histories and environmental
requirements of coastal fishes and invertebrates (Pacific northwest): coho salmon." National
Coastal Ecosystem Team, Division of Biological Services, Research and Development Fish and
Wildlife Service. U.S. Department of Interior, Vicksburg, Mississippi.
LCFRB (2004). "Lower Columbia Salmon Recovery and Fish and Wildlife Subbasin Plan.".
Leeuwangh, P. (1994). Comparison of chlorpyrifos fate and effects in outdoor aquatic micro- and
mesocosmsof various scale and construction. Freshwater Field Tests for Hazard Assessment of
Chemicals. I. R. Hill, F. Heimbach, P. Leeuwangh and P. Matthiessen. Boca Raton, Lewis
Publishers: 217-248.
DRAFT 527
-------
DRAFT
Lerner, D. T., B. T. Bjornsson, et al. (2007). "Aqueous exposure to 4-nonylphenol and 17 beta-
estradiol increases stress sensitivity and disrupts ion regulatory ability of juvenile Atlantic
salmon." Environmental Toxicology and Chemistry 26(7): 1433-1440.
Lerner, D. T., B. T. Bjornsson, et al. (2007). "Larval exposure to 4-nonylphenol and 17 beta-
estradiol affects physiological and behavioral development of seawater adaptation in atlantic
salmon smolts." Environmental Science & Technology 41(12): 4479-4485.
Li, H. X., H. Jiang, et al. (2008). "Acute toxicity of the pesticide methomyl on the topmouth
gudgeon (Pseudorasbora parva): mortality and effects on four biomarkers." Fish Physiology and
Biochemistry 34(3): 209-216.
Li, S. N. and D. F. Fan (1996). "Correlation between biochemical parameters and susceptibility
of freshwater fish to malathion." Journal of Toxicology and Environmental Health 48(4): 413-
418.
Lichatowich, J. A. (1999). "Salmon without rivers. A history of the Pacific salmon crisis." Island
Press, Washington D.C.
Liess, M. and R. Schulz (1999). "Linking insecticide contamination and population response in
an agricultural stream." Environmental Toxicology and Chemistry 18: 1948-1955.
Liess, M. and P. C. Von der Ohe (2005). "Analyzing effects of pesticides on invertebrate
communities in streams." Environmental Toxicology and Chemistry 24(4): 954-965.
Lin, J., j. Hetrick, and R.D. Jones (1998). PRZM-EXAMS Overview for the July 29, 2008
Science Advisory Panel meeting. O. o. P. Programs.
Little, E. and S. Finger (1990). "Swimming behavior as an indicator of sublethal toxicity in fish."
Environmental Toxicology and Chemistry 9: 13-19.
Little, E. E., R. D. Archeski, et al. (1990). "Behavioral indicators of sublethal toxicity in rainbow
trout." Archives of Environmental Contamination and Toxicology 19: 380-385.
Liu, H. X., M. Q. Yi, et al. (2007). "Substrate specificity of brain acetylcholinesterase and its
sensitivity to carbamate insecticides in Carassius auratus." Fish Physiology and Biochemistry
33(1): 29-34.
Lohner, T. W. and S. W. Fisher (1990). "Effects of pH and temperature on the acute toxicity and
uptake of carbaryl in the midge, Chironomus riparius." Aquatic Toxicology 16: 335-354.
Luo, Q., M. Ban, et al. (2005). "Distinct effects of 4-nonylphenol and estrogen-17 beta on
expression of estrogen receptor alpha gene in smolting sockeye salmon." Comparative
Biochemistry and Physiology C-Toxicology & Pharmacology 140(1): 123-130.
DRAFT 528
-------
DRAFT
Lydy, M. J. and K. R. Austin (2005). "Toxicity assessment of pesticide mixtures typical of the
Sacramento-San Joaquin Delta using Chironomus tentans." Archives of Environmental
Contamination and Toxicology 48(1): 49-55.
Macek, K. J. (1975). "Acute toxicity of pesticide mixtures to bluegills." Bulletin of
Environmental Contamination and Toxicology 14(6): 648-652.
Madsen, S. S., S. Skovbolling, et al. (2004). "17-beta estradiol and 4-nonylphenol delay smolt
development and downstream migration in Atlantic salmon, Salmo salar." Aquatic Toxicology
68(2): 109-120.
Majewski, M. S. and D. S. Baston (2002). Atmospheric transport of pesticides in the
Sacramento, California, Metropolitan area, 1996-1997. U.S. Geological Survey Water Resources
Investigations Report 02-4100
64.
Major, W. W. I, C. E. Grue, et al. (2005). Research Update: Exposure of salmonids to carbaryl
following applications to control burrowing shrimp in Willapa Bay, Washington. Seattle
Washington: 12.
Makah Fisheries Management (2000). "Lake Ozette sockeye hatchery and genetic management
plan. Biological assessment Section 7 consultation." available from Makah Fisheries, P. O. Box
115. Neah Bay. Washington 98357.
Mantua, N. J., S. R. Hare, et al. (1997). "A pacific interdecadal climate oscillation with impacts
on salmon production." Bulletin of the American Meteorological Society(78): 1069-1079.
Markle, J., O. Kalman, et al. ( 2005). Mitigating Diazinon Runoff into Waterways from Fruit and
Nut Orchards in the Sacramento Valley. Coalition for Urban/Rural Environmental Stewardship.
Report funded by CALFED.: 23.
Marr, J. C. (1943). "Age, length, and weight studies of three species of Columbia River salmon
(Oncorhynchus keta, O. gorbuscha, and O. kisutch)." Stanford Ichthyology Bulletin 2: 157-197.
Marshall, D. E. and E. W. Britton (1990). "Carrying capacity of coho salmon streams." Canadian
Reportns on Fisheries and Aquatic Science(2058): 1-32.
Matthews, G. M. and R. S. Waples (1991). "Status review for Snake River spring and summer
Chinook salmon." U.S. Department of Commerce, NOAA Technical Memorandum NMFS-
F/NWC-200.
Matthiessen, P., D. Sheahan, et al. (1995). "USE OF A GAMMARUS-PULEXBIOASSAY TO
MEASURE THE EFFECTS OF TRANSIENT CARBOFURAN RUNOFF FROM
FARMLAND." Ecotoxicology and Environmental Safety 30(2): 111-119.
DRAFT 529
-------
DRAFT
Matthiessen, P., D. Shearan, et al. (1995). "Use of Gammarus pulex bioassay to measure the
effects of transient carbofuran runoff from farmland." Ecotoxicology and Environmental Safety
30: 111-119.
Mayer, F. L. and M. R. Ellersieck (1986). Manual of acute toxicity: Interpretation and data base
for 410 chemicals and 66 species of freshwater animals. USFWS. Resource Publication 160:
579.
McClure, M. M., E. E. Holmes, et al. (2003). "A large-scale, multispecies status assessment:
Anadromous salmonids in the Columbia Basin." Ecological Applications 13(4): 964-989.
McCormick, S. D., M. F. O'Dea, et al. (2005). "Endocrine disruption of parr-smolt
transformation and seawater tolerance of Atlantic salmon by 4-nonylphenol and 17 beta-
estradiol." General and Comparative Endocrinology 142(3): 280-288.
McCullough, D. (1999). "A review and synthesis of effects of alterations to the water
temperature regime on freshwater life stages of salmonids, with special referene to Chinook
salmon." Columbia Intertribal Fisheries Commission, Portland, OR. Prepared for the U.S.
Environmental Protection Agency Region 10. Published as EPA 910-R-99-010.
McElhaney, P., M. Chilcote, et al. (2007). "Viability status of Oregon salmon and steelhead
populations in the Willamette and lower Columbia basins." Draft report by NOAA Fisheries and
Oregon Department of Fish and Wildlife.
McElhaney, P., M. H. Ruckleshaus, et al. (2000). "Viable salmonid populations and the recovery
of evolutionarily significant units." U.S. Dept. Commerce., NOAA Tech. Memo. NMFS-
NWFSC-42: 156 p.
McEwan, D. and T. A. Jackson (1996). "Steelhead restoration and management plan for
California." California Department of Fish and Game, Sacramento, California.
McEwan, D. R. (2001). "Central Valley steelhead." pages 1-43 in Brown. R.L. (editor).
Contributions to the biology of the central valley salmonids. California Department of Fish and
Game, Sacramento, California.
McPhail, J. D. and C. C. Lindsey (1970). "Freshwater fishes of northwestern Canada and
Alaska." Fisheries Research Board of Canada, Bulletin 173. Ottawa, Canada.
Meador, J., J. Stein, et al. (1995). "A review of bioaccumulation ofpolycyclic aromatic
hydrocarbons by marine organisms." Environmental Contamination and Toxicology 143: 79-
165.
Meehan, W. R. and T. R. Bjornn (1991). "Salmonid distributions and life histories: cutthroat
trout. Influence of forest and rangeland management on salmonid fishes and their habitats. ."
Bethesda, American Fisheries Society Special Publication 19: 66-67.
DRAFT 530
-------
DRAFT
Merritt, R. W. and K. W. Cummins, Eds. (1995). An Introduction to the Aquatic Insects of North
America. Dubuque, Iowa, Kendall Hunt Publishing Company.
Metcalfe, N. B., N. H. C. Fraser, et al. (1999). "Food availability and the nocturnal vs. diurnal
foraging trade-off in juvenile salmon." Journal of Animal Ecology 68: 371-381.
Meyers, J. M., R. G. Kope, et al. (1998). "Status review of Chinook salmon from Washington,
Idaho, Oregon, and California." U.S. Department of Commerce, NOAA Technical
Memorandum. NMFS-NWFSC-35.
Miller, R. J. and E. L. Brannon (1982). "The origin and development of life history patterns in
Pacific salmonids." pages 296-309 in Brannon, E.L. and E.O. Salo (editors). Proceedings of the
Salmon and Trout Migratory Behavior Synposium. School of Fisheries, University of
Washington, Seattle, Washington.
Miller, R. R., J. D. Williams, et al. (1989). "Extinctions of North American fishes during the last
century." Fisheries 14(6): 22-38.
Mills, S. K. and J. H. Beatty (1979). "The propensity interpretation of fitness." Philosophy of
Science 46: 263-286.
Mineau, P. (1991). Cholinesterase-inhibiting Insecticides: Their Impact on Wildlife and the
Environment. Amsterdam, Elsevier.
Minshall, G. W., T. V. Royer, et al. (2001). "Response of the Cache Creek marcroinvertebrates
during the first 10 years following disturbance by the 1988 Yellowstone wildfires." Canadian
Journal of Fisheries and Aquatic Sciences(58): 1077-1088.
Moffett, J. W. (1949). "The first four years of king salmon maintenance below Shasta Dam,
Sacramento River, California." California Department of Fish and Game, Sacramento, California
35: 77-102.
Monserrat, J. M., J. S. Yunes, et al. (2001). "Effects of Anabaena spiroides (cyanobacteria)
aqueous extracts on the acetylcholinesterase activity of aquatic species." Environmental
Toxicology and Chemistry 20(6): 1228-1235.
Montgomery, D. R., E. M. Beamer, G. R. Pess, and T. P. Quinn (1999). "Channel type and
salmonid spawning distribution and abundance." Canadian Journal of Fisheries and Aquatic
Sciences 56: 377-387.
Moore, M. M. (1980). "Factors influencing survival of juvenile steelhead rainbow trout (Salmo
Gairdnerii gairdnerif) in the Ventura River." M.S. thesis, Humboldt State University, Arcata,
California: 82 p.
DRAFT 531
-------
DRAFT
Morley, S. A., P. S. Garcia, et al. (2005). "Juvenile salmonid (Oncorhynchus spp.) use of
constructed and natural side channels in Pacific Northwest rivers." Canadian Journal of Fisheries
and Aquatic Sciences 62: 2811-2821.
Morley, S. A. and J. R. Karr (2002). "Assessing and restoring the health of urban streams in the
Puget Sound Basin." Conservation Biology 16(6): 1498-1509.
Mount, J. F. (1995). "California rivers: the conflict between fluvial process and land use."
University of California Press, Berkeley and Los Angeles, California.
Moyle, P. B. (1976). "The decline of anadromous fishes in California." Conservation Biology
8(3): 869-870.
Muirhead-Thomson, R. C. (1987). Pesticide Impact on Stream Fauna, Cambridge University
Press.
Mullie, W. C., P. J. Verwey, et al. (1991). "THE IMPACT OF FURADAN-3G
(CARBOFURAN) APPLICATIONS ON AQUATIC MACROINVERTEBRATES IN
IRRIGATED RICE IN SENEGAL." Archives of Environmental Contamination and Toxicology
20(2): 177-182.
Myers, J., C. Busack, et al. (2006). "Historical population structure of Pacific salmonids in the
Willamette River and lower Columbia River basins." NOAA Techical Memorandum NMFS
NWFSC NMFS-NWFSC-73: Slip.
Nickelsen, T. E., J. W. Nicholas, et al. (1992). "Status of anadromous salmonids in Oregon
coastal basins." Oregon Department of Fish and Wildlife, Resource Development Section and
Ocean Salmon Management Oregon Department of Fish and Wildlife, Portland,Oregon.
Nicosia, S., N. Carr, et al. (1991). "OFF-FIELD MOVEMENT AND DISSIPATION OF SOIL-
INCORPORATED CARBOFURAN FROM 3 COMMERCIAL RICE FIELDS." Journal of
Environmental Quality 20(3): 532-539.
NLCD (2001). "National Land Cover Data. Multi-Resolution Land Characteristics Consortium.
Downloaded May 2008. Available at: http://www.mrlc.gov/index.php ".
NMA (2007). "2004 State Mining Statistics." Available:
http://www.nma.org/statistics/state statistics 2004.asp# (February 2008).
NMFS (2005b). "Final assessment of NOAA Fisheries' critical habitat analytical review teams
for 12 evolutionarily significant units of West Coast salmon and steelhead." available online at
http://www.nwr.noaa.gov/Salmon-Habitat/Critical-Habitat/2005 Biological -Teams-Report.
accessed July 2008.
NMFS (2007). Request for endangered species act section 7 informal consultation on the EPA's
reregi strati on and use of atrazine in the Chesapeake Bay watershed, September 1, 2006.
DRAFT 532
-------
DRAFT
NMFS (2007d). "Biological Opinion for U.S. Department of Agriculture, Forest Service aerial
application of eight long-term retardants on all forest service lands."
NMFS (2008). National Marine Fisheries Service endangered species act section 7 consultation:
Environmental Protection Agency registration of pesticides containing chlorpyrifos, diazinon,
and malathion.: 482.
Norberg-King, T. J., E. J. Durhan, et al. (1991). "Application of toxicity indentification
evaluation procedures to the ambient waters of the Colusa Basin Drain, California."
Environmental Toxicology and Chemistry 10: 891-900.
Norbergking, T. J., E. J. Durhan, et al. (1991). "APPLICATION OF TOXICITY
IDENTIFICATION EVALUATION PROCEDURES TO THE AMBIENT WATERS OF THE
COLUSA BASIN DRAIN, CALIFORNIA." Environmental Toxicology and Chemistry 10(7):
891-900.
NRC (1996). "Upstream: Salmon and society in the Pacific northwest." National Academy Press,
Washington, D.C. Available: http://books.nap.edu/openbook.php?record_id=4976&page=381
(February 2008).
NRC (2004). "Managing the Columbia River: Instream flows, water withdrawals, and salmon
survival." National Academy Press, Washington D.C. Available:
http://www.nap.edu/catalog.php7record id=10962#toc (February 2008).
NWPPC (1986). "Compilation of information on salmon and steelhead losses in the Columbia
River basin." Report to the Northwest Power Planning Council, Portland, Oregon. Available:
http://www.nwcouncil.org/library/1986/Compilation.htm (February 2008)
ODA (2008). "Oregon Department of Agriculture- Pesticide Use Reporting System 2007 annual
report. http://www.oregon.gov/ODA/PEST/docs/pdf/pursreportweb2007.pdf"
ODFW (2007). "Oregon Department of Fish and Wildlife. Fish Hatchery Management Plan.
Available online at: http://www.dfw.state.or.us/fish/nfcp/hatchery_mgmt.pdf".
Opperman, J. J. and A. M. Merenlender (2004). "The effectiveness of riparian restoration for
improving instream fish habitat in four hardwood-dominated California streams." North
American Journal of Fisheries Management 24: 822-834.
Pablo, F., F. R. Krassoi, et al. (2008). "Comparison of the fate and toxicity of chlorpyrifos -
Laboratory versus a coastal mesocosm system." Ecotoxicology and Environmental Safety 71(1):
219-229.
DRAFT 533
-------
DRAFT
Palmisano, J. F., R. H. Ellis, et al. (1993). "The impact of environmental and management
factors on Washington's wild anadromous salmon and trout." Washington Forest Protection
Association and Washington Department of Natural Resources, Olympia, Washington.
Paul, M. J. and J. L. Meyer (2001). "Streams in the urban landscape." Annual Review of Ecology
and Systematics 32: 333-365.
Pearcy, W. G. (1992). "Ocean ecology of North Pacific salmonids." University of Washington.
University of Washington Press, Seattle, Washington: 179 p.
Pearcy, W. G. (1997). "What have we learned in the last decade? What are research priorities?"
in Emmett R.L. and M.H. Schiewe (editors). NOAA-NMFS-NWFSC TM-29: Estuarine and
Ocean Survival of Northeastern Pacific Salmon: Proceedings of the workshop. U.S. Pep.
Commer.: 5 p.
Peterson, J. L., P. C. Jepson, et al. (2001). "Effect of varying pesticide exposure duration and
concentration on the toxicity of carbaryl to two field-collected stream invertebrates, Calineuria
californica (Plecoptera : Perlidae) and Cinygma sp (Ephemeroptera : Heptageniidae)."
Environmental Toxicology and Chemistry 20(10): 2215-2223.
Peterson, J. L., P. C. Jepson, et al. (2001). "Effect of varying pesticide exposure duration and
concentration on the toxicity of carbaryl to two field-collected stream invertebrates, Calineuria
californica (Plecoptera: Perlidae) and Cinygma sp. (Emphemeroptera: Heptagaenidae)."
Environmental Toxicology and Chemistry 20: 2215-2223.
Peterson, J. L., P. C. Jepson, et al. (2001). "A test system to evaluate the susceptibility of
Oregon, USA, native stream invertebrates to triclopyr and carbaryl." Environmental Toxicology
and Chemistry 20(10): 2205-2214.
Pitcher, T. J. (1986). "Functions of shoaling in teleosts." pages 294-337 in Fisher, T.J. (editor).
The behavior of teleost fishes. John Hopkins University Press, Baltimore, Maryland.
Post, G. and R. Leasure (1974). "Sublethal effect of malathion to three salmonid species."
Bulletin of Environmental Contamination and Toxicology 12(3): 312-319.
PS AT (2004). "State of the Sound 2004." State of Washington. Office of the Governor. Olvmpia.
Washington. Available: http://www.psat.wa.gov/Publications/StateSound2004/141963_81 l.pdf
(February 2008).
PS AT (2005). "State of the Sound 2004. Puget Sound Action Team, Office of the Governor,
Olympia, WA."
PS AT (2007). "State of the Sound 2007." State of Washington. Office of the Governor.
Publication Number PSAT 07-01, Olvmpia, Washington. Available:
http://www.psat.wa.gov/Publications/state sound07/2007 stateofthesound.pdf (February 2008).
DRAFT 534
-------
DRAFT
Pusey, B. 1, A. H. Arthington, et al. (1994). "EFFECTS OF A PULSED APPLICATION OF
CHLORPYRIFOS ON MACROINVERTEBRATE COMMUNITIES IN AN OUTDOOR
ARTIFICIAL STREAM SYSTEM." Ecotoxicology and Environmental Safety 27(3): 221-250.
Quigley, T. M., S. J. Arbelbide, et al. (1997). "Assessment of ecosystem components in the
interior Columbia River Basin and portions of the Klamath and Great Basins: an introduction."
Pages 1-32 in T. M. Quigley and S. J. Arbelbide, editors. An assessment of ecosystem
components in the interior Columbia River Basin and portions of the Klamath and Great Basins:
an introduction. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research
Station. General Technical Report PNW-GTR 405.
Rabeni, C. F. and J. G. Stanley (1975). "Operational sprating of acephate to suppress spruce
budworm has minor effects on stream fishes and invertebrates." Bulletin of Environmental
Contamination and Toxicology 23: 327-334.
Reavis, B. (1991). "Status of California steelhead stocks." in Pacific States Marine Fisheries
Commission and Association of Northwest Steelheaders (organizers). International Symposium
on Steelhead Trout Management 3-5 January 1991, Portland, Oregon.
Reisenbichler, R. R. (1997). Genetic factors contributing to declines of anadromous salmonids in
the Pacific Northwest. Pacific salmon and their ecosystems: status and future options. D. J.
Stouder, P. A. Bisson and R. J. Naiman. New York, Chapman and Hall: 223-244.
Relyea, R. A. (2005). "The impact of insecticides and herbicides on the biodiversity and
productivity of aquatic communities." Ecological Applications 15(2): 618-627.
Rich, W. H. (1942). "The salmon runs of the Columbia River in 1938." Fishery Bulletin 50(37):
103-147.
Richter, J. E. (2002). "Urban runoff water quality: a salmonid's perspective." Available online at
http://www.4sos.org/wssupport/ws_rest/assess.asp accessed on 7/7/08: 5 p.
Riggs, L. A. (1990). "Principles for genetic conservation and production quality: results of a
scientific and technical clarification and revision." Unpublished report prepared for the
Northwest Power Planning Council, prepared by Genetic Resource Consultants.
Rinne, J. N. (2004). "Forests, fish and fire: relationships and management implications for fishes
in the southwestern USA." pages 151-156 in G.J. Scrimgeour, G.Eisler, B. McCulloch, U. Silins,
and M. Monita (editors). Forest Land - Fish Conference II - Ecosystem Stewardship through
Collaboration. Proceedings of the Forest-Land-Fish Conference II, April 26-28, 2004.
Edmonton, Alberta, Canada.
Roby, D. D., K. Colis, et al. (2008). Research, Monitoring, and Evaluation of Avian Predation on
Salmonid Smolts in the Lower and Mid-Columbia River
Corvallis, Oregon, USGS - Oregon Cooperative Fish and Wildlife Research Unit Department of
Fisheries and Wildlife
DRAFT 535
-------
DRAFT
Roby, D. D., K. Collis, et al. (2006). "Research, monitoring, and evaluation of avian predation on
salmonid smolts in the lower and mid-Columbia River." Draft Season Summary submitted to the
Bonneville Power Administratin and the U.S. Army Corps of Engineers, Portland, Oregon: 130
P-
Roni, P. (2002). "Habitat use by fishes and pacific giant salamanders in small western Oregon
and Washington streams. Transactions of the American Fisheries Society." 131: 743-761.
Rosetta, T. and D. Borys (1996). "Identification of sources of pollutants to the lower Columbia
River basin. Draft report. Prepared for the Lower Columbia River Bi-State Program." Oregon
Department of Environmental Quality. Available: http://www.lcrep.org/pdfs/58.%2002750.pdf
(February 2008).
Roy, S., A. Chattoraj, et al. (2006). "Arsenic-induced changes in optic tectal histoarchitecture
and acetylcholinesterase-acetylcholine profile in Channa punctatus: Amelioration by selenium."
Comparative Biochemistry and Physiology C-Toxicology & Pharmacology 144(1): 16-24.
Ruckelshaus, M. H. and M. M. McClure (2007). "Sound Science: synthesizing ecological and
socioeconomic information about the Puget Sound ecosystem." Prepared in Cooperation with the
Sound Science collaborative team. U.S. Department of Commerce, National Oceanic and
Atmospheric Administration, National Marine Fisheries Service, Northwest Fisheries Science
Center, Seattle, Washington. Available:
http://www.nwfsc.noaa.gov/research/shared/sound_science/index.cfm (February 2008).
Rutter, C. (1904). "Natural history of the quinnat salmon. Investigation on Sacramento River,
1896-1901." Bulletin. U.S. Department of Commerce Bureau of Fisheries 22: 65-141.
Ryan, B. A., M. Carper, et al. (2006). "Alternative barging strategies to improve survival of
transported juvenile salmonids." Draft report sent to U.S. Army Corps of Engineers, Walla
Walla, Washington.
Saglio, P., S. Trijasse, et al. (1996). "Behavioral effects of waterborne carbofuran in goldfish."
Archives of Environmental Contamination and Toxicology 31(2): 232-238.
Sakamoto, M., K. H. Chang, et al. (2006). "Inhibition of development of anti-predator
morphology in the small cladoceran Bosmina by an insecticide: impact of an anthropogenic
chemical on prey-predator interactions." Freshwater Biology 51(10): 1974-1983.
Salo, E. O. (1991). "Life history of chum salmon " in Pacific Salmon Life Histories. C.Groot and
L.Margolis (editors). University of British Columbia Press, Vancouver, British Columbia,
Canada
DRAFT 536
-------
DRAFT
Sandahl, J. F., D. H. Baldwin, et al. (2004). "Odor-evoked field potentials as indicators of
sublethal neurotoxicity in juvenile coho salmon (Oncorhynchus kisutch) exposed to copper,
chlorpyrifos, or esfenvalerate." Canadian Journal of Fisheries Aquatic Sciences 64: 404-413.
Sandahl, J. F., D. H. Baldwin, et al. (2005). "Comparative thresholds for acetylcholinesterase
inhibition and behavioral impairment in coho salmon exposed to chlorpyrifos." Environmental
Toxicology and Chemistry 24: 136-145.
Schoenfuss, H. L., S. E. Bartell, et al. (2008). "Impairment of the reproductive potential of male
fathead minnows by environmentally relevant exposures to 4-nonylphenol." Aquatic Toxicology
86(1): 91-98.
Scholz, N. L., N. K. Truelove, et al. (2000). "Diazinon disrupts antipredator and homing
behaviors in chinook salmon (Oncoryhnchus tshawytscha)." Canadian Journal of Fisheries
Aquatic Sciences 57: 1911-1918.
Scholz, N. L., N. K. Truelove, et al. (2006). "Dose-additive inhibition of chinook salmon
acetylcholinesterase activity by mixtures of organophosphate and carbamate insecticides."
Environmental Toxicology and Chemistry 25(5): 1200-1207.
Schriever, C. A., M. H. Ball, et al. (2007). "Agricultural intensity and landscape structure:
Influences on the macroinvertebrate assemblages of small streams in northern Germany."
Environmental Toxicology and Chemistry 26(2): 346-357.
Schroder, S. L. (1977). "Assessment of production of chum salmon fry from the Big Beef Creek
spawning channel." Completion report. Univ. Wash.Fish. Res. Inst. FRI-UW 7718: 77 p.
Schroder, S. L., K. V. Koski, et al. (1974). "Big Beef Creek studies." in: Research in fisheries
1973. Univ. Wash. Coll. Fish. Contrib. 390: p. 26-27.
Schulz, R. (2004). "Field studies on exposure, effects, and risk mitigation of aquatic nonpoint-
source insecticide pollution: A review." Journal of Environmental Quality 33(2): 419-448.
Schulz, R. and M. Liess (2001). "Acute and chronic effects of particle-associated fenvalerate on
stream macroinvertebrates: A runoff simulation study using outdoor microcosms." Archives of
Environmental Contamination and Toxicology 40(4): 481-488.
Schulz, R. and M. Liess (2001). "Runoff simulation with particle-bound fenvalerate in
multispecies stream microcosms: Importance of biological interactions." Environmental
Toxicology and Chemistry 20(4): 757-762.
Seaber, P. R., F. P. Kapinos, et al. (1987). Hydrologic Unit Maps, U. S. Geologic Survey.
Segner, H. (2005). "Developmental, reproductive, and demographic alterations in aquatic
wildlife: Establishing causality between exposure to endocrine-active compounds (EACs) and
effects." Acta Hydrochimica Et Hydrobiologica 33(1): 17-26.
DRAFT 537
-------
DRAFT
Shaffer, A., P. Grain, et al. (submitted). "Genetic assessment of juvenile Chinook salmon
associated with nearshore habitats of the central and western Strait of Jaun de Fuca,
Washington." North American Journal of Fisheries Management: 22.
Shapovalov, L. and A. C. Taft (1954). "The life histories of the steelhead rainbow trout (Salmo
gairdneri gairdnerf) and silver salmon (Oncorhynchus kisutch) with special reference to Waddell
Creek, California, and recommendations regarding their management.
." California Department of Fish and Game Bulletin 98. Available:
http://content.cdhb.org/xtf/view?docld=kt9x0nb3v6&brand=calisphere&doc.vie=entiret
ext (May 2008).
Sharom, M. S., J. R. W. Miles, et al. (1980). "Persistence of 12 insecticides in water." Water
Resources 14: 1089-1093.
Shea, T. B. and E. S. Berry (1983). "Toxicity of carbaryl and 1-naphthol to goldfish (Carassius
auratus) and Killifish (Fundulus heteroclitus)" Bulletin of Environmental Contamination and
Toxicology 31: 526-529.
Shumann, T. (1994). "Letter to the ESA administrative record for west coast steelhead form the
Steelhead Listing Project, Santa Cruz, California, dated 5 May 1994, re: steelhead return
estimates for Santa Cruz County." Environmental and Technical Services Division, National
Marine Fisheries Service, Portland, Oregon.
Simpson, I. C., P. A. Roger, et al. (1994). "EFFECTS OF NITROGEN-FERTILIZER AND
PESTICIDE MANAGEMENT ON FLOODWATER ECOLOGY IN A WETLAND
RICEFIELD .2. DYNAMICS OF MICROCRUSTACEANS AND DIPTERAN LARVAE."
Biology and Fertility of Soils 17(2): 138-146.
Smith, G. J. (1993). toxicology and Pesticide Use in relation to Wildlife: Organophosphorus and
Carbamate Compounds. Boca Raton, U.S. Department of the Interior.
Snover, A. K., P. W. Mote, et al. (2005). "Uncertain future: climate change and its effects on
Puget Sound." A report for the Puget Sound Action Team by the Climate Impacts Group (Center
for Science in the Earth System Joint Institute for the Study of the Atmosphere and Oceans,
University of Washington, Seattle, Washington)
Sommer, T. R., W. C. Harrell, et al. (2005). "Habitat use and stranding risk of juvenile Chinook
salmon on a seasonal floodplain." North American Journal of Fisheries Management 25: 1493-
1504.
Although juvenile Chinook salmon Oncorhynchus tshawytscha are known to use a
variety of habitats, their use of seasonal floodplains, a highly variable and potentially
risky habitat, has not been studied extensively. Particularly unclear is whether a seasonal
floodplain is a net "source" or a net "sink" for salmonid production. To help address
this issue, we studied salmon habitat use in the Yolo Bypass, a 24,000-ha floodplain of
the Sacramento River, California. Juvenile salmon were present in the Yolo Bypass
DRAFT 538
-------
DRAFT
during winter-spring; fish were collected in all regions and substrates of the floodplain in
diverse habitats. Experimental releases of tagged hatchery salmon suggest that the fish
reared on the floodplain for extended periods (mean 5 33 d in 1998, 56 d in 1999, and 30
d in 2000). Floodplain rearing and associated growth are also supported by the
significantly larger size of wild salmon at the floodplain outlet than at the inlet during
each of the study years. Several lines of evidence suggest that although the majority of
young salmon successfully emigrated from the floodplain, areas with engineered water
control structures had comparatively high rates of stranding. Adult ocean recoveries of
tagged hatchery fish indicate that seasonal floodplains support survival at least
comparable with that of adjacent perennial river channels. These results indicate that
floodplains appear to be a viable rearing habitat for Chinook salmon, making floodplain
restoration an important tool for enhancing salmon production.
Sommer, T. R., M. L. Nobriga, et al. (2001). "Floodplain rearing of juvenile chinook salmon:
evidence of enhanced growth and survival." Canadian Journal of Fisheries and Aquatic Sciences
58: 325-333.
Spence, B. C., G. A. Lomnicky, et al. (1996). "An ecosystem approach to salmonid
conservation." Man Tech Environmental Research Services Corps, TR-4501-96-6057, Corvallis,
Oregon.
St. Aubin, C. (2004). The use of biological indicators to measure the indirect and direct effects to
threatened steelhead (Oncorhynchus mykiss) from azinphosmethyl in Hood River, Oregon.
Environmental Sciences and Resources. Portland, Portland State University. Master of Science:
60.
Stanford, J. A., F. R. Hauer, et al. (2005). "Columbia River basin." Pages 591-653 in A. C.
Benke and C. E. Gushing, editors. Rivers of North America. Elsevier Academic Press,
Burlington, Massachusetts. Available:
http://books.google.com/books?id=faOUlwkiYFIC&pg=RA3-PA541&lpg=RA3-
PA541&dq=pacific+coast+rivers+of+the+coterminous+united+states&source=web&ots=-
pMpyECFaA&sig=FkGrIiwgkfDvHxXCWXRalK XSvU#PPRLMl (February 2008).
Stearns, S. C. (1982). "The evolution of life histories." New York, New York, Oxford University
Press.
Steward, C. R. and T. C. Bjornn (1990). "Supplementation of salmon and steelhead stocks with
hatchery fish: a synthesis of published literature." Bonneville Power Administration, Technical
Report 90-1. Portland. Oregon.
Stone, I. (1874). "Report of operations during 1872 at the United States salmon hatching
establishment on the McCloud River. ." U.S. Commmission on Fish and Fisheries, Report for
1872 and 1873. Part II: 168-215. Washington. D.C. Available:
http://penbav.org/cof/cof 1872.html (April 2008).
DRAFT 539
-------
DRAFT
Swift III, C. H. (1979). Preferred stream discharges for salmon spawning and rearing in
Washington.USGS Open-File Report 77-422 USGS: 51p.
Sytsma, M. D., J. R. Cordell, et al. (2004). Lower Columbia River aquatic nonindigenous species
survey 2001-2004. Final Technical Report. October. Prepared for the US Coast Guard and the
US Fish and Wildlife Service.
Takahashi, H. and T. Hanazato (2007). "Synergistic effects of food shortage and an insecticide
on a Daphnia population: rapid decline of food density at the peak of popualtion density reduces
tolerance to the chemical and induces a large population crash." Limnology 8(45-51).
Teske, M. E. (2001). A user's guide for AgDrift 2.04: A tiered approach for the assessment of
spray drift of pesticides, Spray Drift Task Force: 128.
Tierney, K., M. Casselman, et al. (2007). "The relationship between cholinesterase inhibition and
two types of swimming performance in chlorpyrifos-exposed coho salmon (Oncorhynchus
kisutch)." Environmental Toxicology and Chemistry 26(5): 998-1004.
Tufts, D. F. (1989). "Control of burrowing shrimp on oyster beds in Willapa Bay and Grays
Harbor. 1986. Washington State Department of Fisheries. Special Shellfish Report No. 4. ." 44.
Tufts, D. F. (1990). "Control of burrowing shrimp on oyster beds in Willapa Bay and Grays
Harbor. 1987. Washington State Department of Fisheries. Special Shellfish Report No. 5. ." 46.
Tynan, T. (1997). Life history characteristics of summer chum solmon populations in the Hood
Canal and eastern Strait of Juan de Fuca Regions. Volume I. Washington Department of Fish and
Wildlife. Technical Report #H97-06.
USCB (2005). Interim State Population Projections, 2005. P. Division, U.S. Census Bureau.
USFS (2000). "Fire recharges native fisheries." U.S. Department of Agriculture Forest
Service - Northern Region: 3 p.
USFWS (2008). "Informal consultation on the effects of atrazine re-registration on the
endangered Alabama sturgeon and the endangered dwarf wedge-mussel."
USFWS and NMFS (1998). "Consultation Handbook. Procedures for conducting consultation
and conference activities under section 7 of the endangered species act. ."
Van den Brink, P. J., J. M. Baveco, et al. (2007). "An individual-based approach to model spatial
population dynamics of invertebrates in aquatic ecosystems after pesticide contamination."
Environmental Toxicology and Chemistry 26(10): 2226-2236.
Van den Brink, P. J., R. P. A. van Wijngaarden, et al. (1996). "Effects of the insecticide Dursban
4E (active ingredient chlorpyrifos) in outdoor experimental ditches: II. Invertebrate community
responses and recovery." Environmental Toxicology and Chemistry 15: 1143-1153.
DRAFT 540
-------
DRAFT
Van der Linde, L. H. (2005). The effects of organophosphates on the macroinvertebrate
community structure in relation to threatened salmonids in Hood River Basin, Oregon.
Environmental Sciences and Resources. Portland, Portland State University. Master of Science:
140.
Van Wijngaarden, R. P. A., T. C. M. Brock, et al. (2005). "Threshold levels for effects of
insecticides in freshwater ecosystems: A review." Ecotoxicology 14(3): 355-380.
Varanasi, U., J. Stein, et al. (1989). "Biotransformation and disposition of PAH in fish."
Metabolism of polycyclic aromatic hydrocarbons in the aquatic environment Varansi U. (ed).
CRC Press: Boca Raton. Florida: 93-149.
Varanasi, U., J. Stein, et al. (1992). "Chlorinated and aromatic hydrocarbons in bottom
sediments, fish and marine mammals in U.S. coastal waters; laboratory and field studies of
metabolism and accumulation." Persistent Pollutants in Marine Ecosystems. C.H. W alker and
D.R. Livingstone (eds). Permagon Press: New York: 83-115.
Walker, C. H. and H. M. Thompson (1991). Phylogenetic distribution of cholinesterases and
realted esterases. Chloinesterase-inhibiting insecticides: Their impact on wildlife and the
environment. P. Mineau. New York, Elsevier. 2: 1-17.
Walker, R. L. and J. S. Foott (1993). "Disease survey of Klamath River salmonid smolt
populations, 1992." U.S. Fish and Wildlife Service, California-Nevada Fish Health Center,
Anderson. California. [150 Kbl at KRIS Website: 46 p.
Wallace, J. B., A. D. Huryn, et al. (1991). "COLONIZATION OF A HEADWATER STREAM
DURING 3 YEARS OF SEASONAL INSECTICIDAL APPLICATIONS." Hydrobiologia
211(1): 65-76.
Wallace, J. B., G. J. Lugthart, et al. (1989). "THE IMPACT OF REPEATED INSECTICIDAL
TREATMENTS ON DRIFT AND BENTHOS OF A HEADWATER STREAM." Hydrobiologia
179(2): 135-147.
Ward, S., A. H. Arthington, et al. (1995). "THE EFFECTS OF A CHRONIC APPLICATION
OF CHLORPYRIFOS ON THE MACROINVERTEBRATE FAUNA IN AN OUTDOOR
ARTIFICIAL STREAM SYSTEM - SPECIES RESPONSES." Ecotoxicology and
Environmental Safety 30(1): 2-23.
Wayland, M. (1991). "EFFECT OF CARBOFURAN ON SELECTED
MACROINVERTEBRATES IN A PRAIRIE PARKLAND POND - AN ENCLOSURE
APPROACH." Archives of Environmental Contamination and Toxicology 21(2): 270-280.
Wayland, M. (1991). "Effect of carbofuran on selected macroinvertebrates in a prairie parkland
pond: an enclosure approach." Archives of Environmental Contamination and Toxicology 21:
270-280.
DRAFT 541
-------
DRAFT
Wayland, M. and D. A. Boag (1990). "TOXICITY OF CARBOFURAN TO SELECTED
MACROINVERTEBRATES IN PRAIRIE PONDS." Bulletin of Environmental Contamination
and Toxicology 45(1): 74-81.
WDF, WDW, et al. (1993). "1992 Washington state salmon and steelhead stock inventory
(SASSI) " available from Washington Deparmtnet of Fish and Wildlfie, 600 Captiol Way N.,
Olvmpia, Washington 98501.
WDFW (2005). "Washington Department of Fish and Wildlife.Methow River Steelhead
Hatchery and Genetic Management Plan. Available online at: http://wdfw.wa.gov/hat/hgmp."
Weis, P. and J. S. Weis (1974). "Schooling behavior of Menidia medidia in the presence of the
insecticide Sevin (Carbaryl)." Marine Biology 28: 261-263.
Weiss, C. M. and J. L. Botts (1957). "The response of some freshwater fish to isopropyl
methylphosphonofluoridate (Sarin) in water." Limnology and Oceanography 2: 363-370.
Weisskopf, C. P. and A. S. Felsot (1998). "Carbaryl adn imidacloprid concentrations in Willapa
Bay water and sediment following application for control of burrowing shrimp. A final report to
the Washington State Commission on Pesticide Registration and the Willapa-Grays Harbor
Oyster Growers Association. April 2, 1998.".
Weitkamp, L. A., T. C. Wainwright, et al. (1995). "Status review of coho salmon from
Washington, Oregon, and California
":258p.
Welch, D. W. (1996). "Growth and energetics of salmon in the sea." in Emmett R.L. and M.H.
Schiewe (editors). NOAA-NMFS-NWFSC TM-29: Estuarine and Ocean Survival of
Northeastern Pacific Salmon: Proceedings of the workshop. U.S. Pep. Commer. : 7 p.
Wentz, D. A., B. A. Bonn, et al. (1998). "Water quality in the Willamette Basin, Oregon 1991-
95." U.S. GEOLOGICAL SURVEY circular 1161.
Werner, L, L. A. Deanovic, et al. (2000). "Insecticide-caused toxicity to Ceriodaphnia dubia
(Cladocera) in the Sacramento-San Joaquin River Delta, California, USA." Environmental
Toxicology and Chemistry 19(1): 215-227.
Wheeler, A. P., P. L. Angermeier, et al. (2005). "Impacts of new highways and subsequent
landscape uranization on stream habitat and biota." Fisheries Science 13: 141-164.
Whitmore, R. W., J. E. Kelly, et al. (1992). National Home And Garden Pesticide Use
Survey. , Research Triangle Institute (RTI).
DRAFT 542
-------
DRAFT
Wilcove, D. S., D. Rothstein, et al. (1998). "Quantifying threats to imperiled species in the
United States: Assessing the relative importance of habitat distinction, alien species, pollution,
over exploitation, and disease." Bioscience 48(8): 607-615.
Wildish, D. J. and N. A. Lister (1973). "Biological effects of fenitrothion in the diet of brook
trout." Fisheries Research Board of Canada 10: 333-339.
Williams, A. K. and C. R. Sova (1966). "Acetylcholinesterase levels in brains of fishes from
polluted waters." Bulletin of Environmental Contamination and Toxicology 1: 198-204.
Williamson, A. K., M. D. Munn, et al. (1998). Water Quality in the Central Columbia Plateau,
Washington and Idaho, 1992-95: U.S. Geological Survey Circular 1144, on line at:
http://water.usgs.gov/pubs/circll4.
Wisby, W. J. and A. D. Hasler (1954). "Effect of occlusion on migrating silver salmon." Journal
of Fisheries Research Board of Canada 11: 472-478.
Woods, R. A. and E. B. Figueroa (2007). "Industry output and employment projections to 2016."
Monthly Labor Review 130: 53-85.
WSDA (2004). "Washington state carbofuran use summary. State of Washington. Department of
Agriculture. ." 9 pp. .
WSDE (2006). "Washington State Department of Ecology - Water quality standards for surfaces
waters of the state of Washington chapter 173-201A WAC. Publication Number 06-10-091.
Available at: http://www.ecv.wa.gov/pubs/0610091.pdf."
Wu, L. (2000). "TMDLs: a significant change in water quality regulation enforcement." accessed
at http://esce.ucr.edu/soilwater/winter2000.htm on 7/7/08: 5 p.
Wunderlich, R. C., B. D. Winter, et al. (1994). "Restoration of the Elwha River ecosystem."
Fisheries 19((8)): 11-19.
Wydoski, R. S. and R. R. Whitney (1979). "Inland fishes of Washington." University of
Washington. University of Washington Press, Seattle, Washington.
Xie, L. T., K. Thrippleton, et al. (2005). "Evaluation of estrogenic activities of aquatic herbicides
and surfactants using an rainbow trout vitellogenin assay." Toxicological Sciences 87(2): 391-
398.
Yi, M. Q., H. X. Liu, et al. (2006). "Inhibitory effects of four carbamate insecticides on
acetylcholinesterase of male and female Carassius auratus in vitro." Comparative Biochemistry
and Physiology C-Toxicology & Pharmacology 143(1): 113-116.
DRAFT 543
-------
DRAFT
Yoshiyama, R. M., E. R. Gerstung, et al. (1996). "Historical and present distribution of chinook
salmon in the Central Valley drainage of California." Department of Wildlife, Fish, and
Conservation Biology, University of California, Davis California.
Zinkl, J. G., P. J. Shea, et al. (1987). "Brain cholinesterase activity of rainbow trout poisoned by
carbaryl." Bulletin of Environmental Contamination and Toxicology 38: 29-35.
DRAFT 544
-------
DRAFT
Appendix 1: Population Modeling
DRAFT 545
-------
DRAFT
Introduction
To assess the potential for adverse impacts of the anticholinesterase insecticides on Pacific
salmon populations, a model was developed that explicitly links impairments in the
biochemistry, behavior, prey availability and somatic growth of individual salmon to the
productivity of salmon populations. More specifically, the model connects known effects of the
pesticides on salmon physiology and behavior with community-level effects on salmon prey to
estimate population-level effects on salmon. The model used here is an extension of one
developed for investigating the direct effects of pesticides on the biochemistry, behavior and
growth of ocean-type Chinook salmon (Baldwin et al., in press).
In the freshwater portion of their life, Pacific salmon may be exposed to insecticides that act by
inhibiting acetylcholinesterase (AChE). Acetylcholinesterase is a crucial enzyme in the proper
functioning of cholinergic synapses in the central and peripheral nervous systems of vertebrates
and invertebrates. Of consequence to salmon, anticholinesterase insecticides have been shown to
interfere with salmon swimming behavior (Beauvais et al. 2000, Brewer et al. 2001, Sandahl et
al. 2005), feeding behavior (Sandahl et al. 2005), foraging behavior (Morgan and Kiceniuk
1990), homing behavior (Scholz et al. 2000), antipredator behaviors (Scholz et al. 2000) and
reproductive physiology (Moore and Waring 1996, Waring and Moore 1997, Scholz et al. 2000).
Anticholinesterase insecticides have also been found to reduce benthic densities of aquatic
invertebrates and alter the composition of aquatic communities (Liess and Schulz 1999, Schulz
and Liess 1999, Schulz et al. 2002, Fleeger et al. 2003, Schulz 2004, Chang et al. 2005, Relyea
2005). Spray drift and runoff from agricultural and urban areas can expose aquatic invertebrates
to relatively low concentrations of insecticides for as little as minutes or hours, but populations
of many taxa can take months or even years to recover to pre-exposure or reference densities
(Wallace et al. 1991, Liess and Schulz 1999, Anderson et al. 2003, Stark et al. 2004). For
example, when an aquatic macroinvertebrate community in a German stream was exposed to
runoff containing parathion (an acetylcholinesterase inhibitor) and fenvalerate (another
commonly used insecticide), eight of eleven abundant species disappeared and the remaining
three were reduced in abundance (Liess and Schulz 1999). Long-term changes in invertebrate
densities and community composition likely result in reductions in salmon prey availability.
DRAFT 546
-------
DRAFT
Therefore, in addition to the direct impacts that acetylcholinesterase inhibitors have on salmon,
there may also be, independently, significant indirect effects to salmon via their prey (Peterson et
al. 2001a). Wild juvenile salmon feed primarily on invertebrates in the water column and those
trapped on the water's surface, actively selecting the largest items available (Healey 1991, Quinn
2005). Salmon are often found to be food limited (Quinn 2005), suggesting that a reduction in
prey number or size due to insecticide exposure may further stress salmon. For example, Davies
and Cook (1993) found that several months following a spray drift event, benthic and drift
densities were still reduced in exposed stream reaches. Consequently, brown trout in the exposed
reaches fed less and grew at a slower rate compared to those in unexposed stream reaches
(Davies and Cook 1993). Although the insecticide in their study was cypermethrin (a
pyrethroid), similar reductions in macroinvertebrate density and recovery times have been found
in studies with acetylcholinesterase inhibitors (Liess and Schulz 1999, Schulz et al. 2002),
suggesting indirect effects to salmon via prey availability may be similar.
One likely biological consequence of reduced swimming, feeding, foraging, and prey availability
is a reduction in food uptake and, subsequently, a reduction in somatic growth of exposed fish.
Juvenile growth is a critical determinant of freshwater and marine survival for Chinook salmon
(Higgs et al. 1995). Reductions in the somatic growth rate of salmon fry and smolts are believed
to result in increased size-dependent mortality (Healey 1982, West and Larkin 1987, Zabel and
Achord 2004). Zabel and Achord (2004) observed size-dependent survival for juvenile salmon
during the freshwater phase of their outmigration. Mortality is also higher among smaller and
slower growing salmon because they are more susceptible to predation during their first winter
(Healey 1982, Holtby et al. 1990, Beamish and Mahnken 2001). These studies suggest that
factors affecting the organism and reducing somatic growth, such as anticholinesterase
insecticide exposure, could result in decreased first-year survival and, thus, reduce population
productivity.
Changes to the size of juvenile salmon from exposure to carbaryl, carbofuran, and methomyl
were linked to salmon population demographics. We used size-dependent survival of juveniles
during a period of their first year of life. We did this by constructing and analyzing general life-
history matrix models for coho salmon (Oncorhynchus kisutch), sockeye salmon (O. nerkd) and
DRAFT 547
-------
DRAFT
ocean-type and stream-type Chinook salmon (O. tshawytscha). A steelhead (O. mykiss) life-
history model was not constructed due to the lack of demographic information relating to the
proportions of resident and anadromous individuals, the freshwater residence time of steelhead,
and rates of repeated spawning. The basic salmonid life history modeled consisted of hatching
and rearing in freshwater, smoltification in estuaries, migration to the ocean, maturation at sea,
and returning to the natal freshwater stream for spawning followed shortly by death. Differences
between the modeled strategies are lifespan of the female, time to reproductive maturity, and the
number and relative contribution of the reproductive age classes (Figure 1). The coho females
we modeled reach reproductive maturity at age 3 and provide all of the reproductive
contribution. Sockeye females reach maturity at age 4 or 5, but the majority of reproductive
contributions are provided by age 4 females. Chinook females can mature at age 3, 4 or 5, with
the majority of the reproductive contribution from ages 4 and 5. The primary difference between
the ocean-type and stream-type Chinook is the juvenile freshwater residence with ocean-type
juveniles migrating to the ocean as subyearlings and stream-type overwintering in freshwater and
migrating to the ocean as yearlings. The models depicted general populations representing each
life-history strategy and were constructed based upon literature data described below. Specific
populations were not modeled due to the difficulty in finding sufficient demographic and
reproductive data for a single population.
A separate acute toxicity model was constructed that estimated the population-level impacts of
juvenile mortality resulting from exposure to lethal concentrations of carbaryl, carbofuran, and
methomyl. These models excluded sublethal and indirect effects of the pesticide exposures and
focused on the population-level outcomes resulting from an annual exposure of juveniles to a
pesticide. The lethal impact was implemented as a change in first year survival for each of the
salmon life-history strategies.
The overall model endpoint used to assess population-level impacts for both the growth and
acute lethality models was the percent change in the intrinsic population growth rate (lambda, A,)
resulting from the pesticide exposure. Change in A, is an accepted population parameter often
used in evaluating population productivity, status, and viability. The National Marine Fisheries
Service uses changes in A, when estimating the status of species, conducting risk and viability
DRAFT 548
-------
DRAFT
assessments, developing Endangered Species Recovery Plans, composing Biological Opinions,
and communicating with other federal, state and local agencies (McClure et al. 2003). While
values of A,<1.0 indicate a declining population, negative changes in lambda greater than the
natural variability for the population indicate a loss of productivity. This can be a cause for
concern since the decline could make a population more susceptible to dropping below 1.0 due to
impacts from multiple stressors.
The following models were developed to serve as a means to assess the potential effects on ESA-
listed salmon populations from exposure to AChE inhibiting pesticides, including n-methyl
carbamates and organophosphorus insecticides. The growth model focuses on the impacts to
prey abundance and a salmon's ability to feed which are integrated into reductions in juvenile
growth. Assessing the results from different pesticide exposure scenarios relative to a control
(i.e. unexposed) scenario can indicate the potential for sublethal pesticide exposures to lead to
changes in the somatic growth and survival of individual subyearling salmon. Consequently,
subsequent changes in salmon population dynamics as indicated by percent change in a
population's intrinsic rate of increase assists us in forecasting the potential population-level
impacts to listed populations. Also, the model helps us understand the potential influence of life-
history strategies that might explain differential results within the species modeled.
Methods
The model consists of two parts, an organismal portion and a population portion. The organismal
portion of the model links AChE inhibition and reduced prey abundance due to insecticide
exposure to potential reductions in the growth of individual fish. The population portion of the
model links the sizes of individual subyearling salmon to their survival and the subsequent
growth of the population. Models were constructed using MATLAB 7.7.0 (R2008b) (The
MathWorks, Inc. Natick, MA).
Organismal Model
For the organismal model a relationship between AChE activity and somatic growth of salmonid
fingerlings was developed using a series of relationships between pesticide exposure, AChE
activity, feeding behavior, food uptake, and somatic growth rate (Figures 2-4). The model
DRAFT 549
-------
DRAFT
incorporates empirical data when available. Since growth and toxicity data are limited,
extrapolation from one salmon species to the others was done with the assumption that the
salmon stocks would exhibit similar physiological and toxicological responses. Sigmoidal dose-
response relationships based upon the AChE inhibition EC50 values and their slopes are used to
determine the level of AChE activity (Figure 2A, 2B, 2C) from the exposure concentration of
each pesticide exposure or pulse.
A linear relationship based on empirical data related AChE activity to feeding behavior (Sandahl
et al. 2005, Figure 2D). Feeding behavior was then assumed to be directly proportional to food
uptake, defined as potential ration (Figure 2E). The potential ration expresses the amount of food
the organism can consume when prey abundance is not limiting. Potential ration over time
(Figure 2F) depicts how the food intake of individual fish changes in response to the behavioral
effects of the pesticide exposure over the modeled growth period. Potential ration is equal to
final ration if no effects on prey abundance are incorporated (Figure 4). If effects of pesticide
exposure on prey abundance are incorporated, final ration is the product of potential ration
(relating to the fish's ability to capture prey, Figure 2) and the relative abundance of prey
available following exposure (Figure 3). Next, additional empirical data (e.g. Weatherley and
Gill 1995) defined the relationship between final ration and somatic growth rate (Figure 4C).
While the empirical relationship is more complex (e.g. somatic growth rate plateaus at rations
above maximum feeding), a linear model was considered sufficient for the overall purpose of
this model. Finally, the model combines these linear models relating AChE activity to feeding
behavior, feeding behavior to potential ration, and final ration to somatic growth rate to produce
a linear relationship between AChE activity and somatic growth rate (Figure 4D). One important
assumption of the model is that the relationships are stable, i.e. do not change with time. The
relationships would need to be modified to incorporate time as a variable if, for example, fish are
shown to compensate over time for reduced AChE activity to improve their feeding behavior and
increase food uptake.
Juvenile salmonids are largely opportunistic, feeding on a diverse community of aquatic and
terrestrial invertebrate taxa that are entrained in the water column or on the surface (Higgs et al.
1995). As a group, these invertebrates are among the more sensitive taxa for which there is
DRAFT 550
-------
DRAFT
toxicity data, but within this group, there is a wide range of sensitivities. To determine a single
effect concentration to use in the model analyses, a search was completed using the EPA's
Ecotox database for each pesticide (http://cfpub.epa.gov/ecotox/). Several criteria were used to
determine which reported effect concentrations were included in the final analysis. The data
included were from studies on taxa that are known to be salmonid prey (or are functionally
similar to salmonid prey); these include a diverse group of fresh and saltwater crustaceans,
aquatic insects and worms. Studies with exposures of at least 24 hours and not more than 96
hours were included. Studies examining shorter and longer exposure times are known to affect
invertebrates (e.g., Peterson et al. 2001b), but these were excluded so that estimated ECSOs
would be comparable to previous analyses (NMFS 2008). Studies reporting invertebrate LCSOs
and ECSOs in which mortality or immobilization of invertebrates was the recorded endpoint were
included; the term "EC50" will be used in this report to describe all of these included data. Data
derived for sublethal endpoints (e.g. growth or reproduction) were not included. If specific data
were represented more than once in the Ecotox output, duplicates were eliminated. Data from
several recent peer-reviewed studies that are not yet included in the Ecotox data base, but report
effect concentrations that caused mortality, were also included.
From the distributions of those data, a single effect concentration and slope were derived to best
represent the diverse community of prey available in juvenile salmonid freshwater and estuarine
habitats. The distributions of individual invertebrate ECSOs and the geometric means of ECSOs
by taxa were analyzed to estimate the 50th and 10th percentiles. Figures X show the analyses of
the distributions of geometric means of ECSOs by taxa. Specifically, for each pesticide, a
probability plot was used to graph the EC50 concentrations normalized to a normal probability
distribution. For each plot, the X axis is scaled in probability (between zero and 100%) and
shows the percentage of entire data whose value is less than the data point. The Y axis displays
the range of the data on a log scale. The results of a linear regression of the log-transformed
concentrations are shown and highlight the lognormal distribution of the data (cite Figs). In the
regression equation, the normsinvQ function returns the inverse of the standard normal
cumulative distribution. The standard normal distribution has a mean of zero and a standard
deviation of one. For example, given a percentile value of 50 (i.e. a probability of 0.5),
DRAFT 551
-------
DRAFT
normsinv(50) returns a value of zero. The plots and regressions were performed using
KaleidaGraph 4.03 (Synergy Software).
The decision to use the 10th percentile rather than the 50th percentile is consistent with previous
designations by EPA, and is reasonable because of the relative sensitivity of invertebrates that
are most likely consumed by juveniles. In addition, the 10th percentile is a reasonable threshold
because the data included in the meta-analysis were limited to concentrations that caused
mortality or immobilization within a short period of time (1-4 days). A growing number of
studies on a variety of insecticides have reported that concentrations well below LCSOs can cause
delayed mortality or sublethal effects that may scale up to affect populations, especially in
scenarios with multiple exposures and/or other stressors. Evidence for ecologically significant
sublethal or delayed effects includes reduced growth rates (Schulz and Liess 200Ib, Forbes and
Cold 2005), altered behavior (Johnson et al. 2008), reduced emergence (Schulz and Liess
200la), reduced reproduction (Cold and Forbes 2004, Sakamoto et al. 2006), and reduced
predator defenses (Sakamoto et al. 2006, Johnson et al. 2008). Finally, the available toxicity data
- and therefore the data included for these analyzes - are from studies using taxa hearty enough
to survive laboratory conditions. Studies specifically examining salmonid prey that are more
difficult to rear in the laboratory have documented relatively low LC50 or EC50 values when
exposed to current-use pesticides (e.g., Peterson et al. 2001a, Johnson et al. 2008). It is
noteworthy that the most relevant invertebrate study for carbaryl - a study using a diverse group
of seven salmonid prey taxa collected from streams in the Pacific Northwest - reports LCSOs for
carbaryl with a geometric mean of 26.28 |ig/L (range 11.1 - 61.0) (Peterson et al. 2001a).
The models allow exposures that can include multiple AChE-inhibiting pesticides over various
time pulses. Sigmoidal dose-response relationships, at steady-state, between each single pesticide
exposure and 1) AChE activity and 2) relative prey abundance are modeled using specific ECSOs
and ECSOs and slopes (Figure 2B and 3B). The timecourse for each exposure was built into the
model as a pulse with a defined start and end during which the exposure remained constant
(Figure 2A and 3 A). The timecourse for AChE activity, on the other hand, was modeled using
two single-order exponential functions, one for the time required for the exposure to reach full
effect and the other for time required for complete recovery following the end of the exposure
DRAFT 552
-------
DRAFT
(time-to-effectAChE activity and time-to-recovery ACHE activity, respectively; Figure 2C). The apparent
activity level was back-calculated to result in a relative concentration (concentration/ AChE
inhibition EC50) for each day of the growth period for each pulse. The relative concentration for
each day was summed across all the pulses to result in a total apparent concentration for each
day. The sigmoid slope used in the calculation of AChE activity using the apparent concentration
was the arithmetic mean of the sigmoid slopes for each pesticide present on each day. The
timecourse for relative prey abundance was modeled incorporating a one day spike in prey drift
relative to the toxicity and available prey base followed by a drop in abundance due to the toxic
impacts (Figure 3C). Recovery is assumed to be due to a constant influx of invertebrates from
connected habitats (aquatic and terrestrial) that are not exposed to the pesticide. Incoming
organisms are subject to toxicity if pesticides are still present and this alters the rate of recovery
during exposures. Incorporating dynamic effects and recovery variables allows the model to
simulate differences in the pharmacokinetics (e.g. the rates of uptake from the environment and
of detoxification) of various pesticides and simulate differences in invertebrate community
response and recovery rates (see below).
The relationship between final ration and somatic growth rate (Figure 4C) produces a
relationship representing somatic growth rate over time (Figure 4D), which is then used to model
individual growth rate and size over time. The growth models were run for 1000 individual fish,
with initial weight selected from a normal distribution with a mean of 1.0 g and standard
deviation of 0.1 g. The size of 1.0 g was chosen to represent subyearling size in the spring prior
to the onset of pesticide application. For each iteration of the model (one day for the organismal
model), the somatic growth rate is calculated for each fish by selecting the parameter values from
normal distributions with specified means and standard deviations (Table 1). The weight for each
fish is then adjusted based on the calculated growth rate to generate a new weight for the next
iteration. The length (days) to run the growth portion of the model was selected to represent the
time from when the fish enter the linear portion of their growth trajectory in the mid to late
spring until they change their growth pattern in the fall due to reductions in temperature and
resources or until they migrate out of the system. The outputs of the organismal model that are
handed to the population models consist of mean weights (with standard deviations) after the
DRAFT 553
-------
DRAFT
species-appropriate growth period (Table 2). A sensitivity analysis was run to determine the
influence of the parameter values on the output of the growth model.
The parameter values defining control conditions that are constant for all the modeled species are
listed in Table 1. Model parameters such as the length of the growth period and control daily
growth rate that are species specific are listed in Table 2. Each exposure scenario was defined by
a concentration and exposure time for each pesticide. The duration of time until full effect for the
pesticides was assumed to be within a few days (Ferrari et al. 2004), with a half-life of 0.5 days.
For prey, it is assumed there is a constant, independent influx of prey from upstream habitats that
will eventually (depending on the rate selected) return prey abundance to 1. As mentioned above,
however, these invertebrates are subject to exposure once added to the system, and therefore prey
recovery rate is a product of the influx rate as well as the exposure scenario. While recovery rates
reported in the literature vary, it is assumed a 1% recovery rate is ecologically realistic (Ward et
al. 1995, Van den Brink et al. 1996, Colville et al. 2008). It was also assumed that regardless of
the exposure scenario, relative prey abundance would not drop below a specific floor (Figure
3B). This assumption depends on a minimal yet constant terrestrial subsidy of prey and/or an
aquatic community with tolerant individuals that would be available as prey, regardless of
pesticide exposure and in addition to the constant recovery rate. No studies specify floors per se,
but studies quantifying invertebrate densities following highly toxic exposures indicate a floor of
0.2 is ecologically realistic (i.e. regardless of the exposure, 20% of a fish's ration will be
available daily; e.g., Cuffney et al. 1984). Finally, because prey availability has been found to
increase dramatically albeit briefly following pesticide exposures (due to immediate mortality
and/or emigration of benthic prey into the water column; Davies and Cook 1993, Schulz 2004), a
one-day prey spike is included for the day following an exposure. The relative magnitude of the
spike is calculated as the product of the standing prey availability the day prior to exposure
(minus the floor), the toxicity of the exposure, and a constant of 20. This calculation therefore
accounts for the potential prey that are available and the severity of the exposure. The spike will
be greater when more prey are available and/or the toxicity of the exposure is greater;
alternatively, the spike will be small when few prey are available and/or the exposure toxicity is
low.
DRAFT 554
-------
DRAFT
Below are the mathematical equations used to derive Figures 2, 3, and 4.
Figures 2A and 3 A use a step function:
time < start; exposure = 0
start < time < end; exposure = exposure concentration(s)
time > end; exposure = 0.
Figures 2B and 3B use a sigmoid function:
y = bottom + (top - bottom)/(l + (exposure concentration/EC50)Aslope).
For 2B, y = AChE activity, top = Ac, bottom = 0.
For Figure 3B, y = prey abundance, top = PC (in this case 1), bottom = Pf.
Figures 2D, 2E, and 4C use a linear function (the point-slope form of a line):
y = m*(x-xl) + yl.
For 2D, m = Mfa, xl = Ac, and yl = Fc.
For 2E, m = Mrf (computed as Rc/Fc), xl = Fc, and y 1 = Re.
For 4C, m = Mgr, xl = Re, and yl = Gc.
Figure 2C uses a series of exponential functions:
time < start; y = c
start < time < end; y = c - (c - i)*(l - exp(-ke*(time - start)))
time > end; ye = c - (c - i)*(l - exp(-ke*(end - start)))
y = ye + (c - ye)*(l - exp(-kr*(time - end))).
For Figure 2C, c = Ac, i = Ai, ke = ln(2)/AChE effect half-life, kr = ln(2)/AChE recovery
half-life. For Figure 2C the value of ye is calculated to determine the amount of inhibition
that is reached during the exposure time, which may not be long enough to reach the
maximum level of inhibition.
For Figure 3C, an exposure pulse would result in a 1-day spike followed by a decline to
the impacted level based upon the prey toxicity. During exposures resulting in low prey
DRAFT 555
-------
DRAFT
toxicity, toxicity-limited recovery can occur. After exposure ends a constant rate of
recovery proceeds until control drift is reached or another exposure occurs
preyavail=preydrift(day-l)-floor;
prey tox= 1 /(1 +(concentrati on)Aprey si ope);
preyrecrate=0.01;
preydriftrec = prey recrate* prey tox.
time=start; spike=(-l+10A(1.654*preyavail))*(l-preytox)
prey drift =preydrift+spike
start < time < end; preydrift=(preyavail*preytox)+preyrdriftrec+floor;
time>end; prey drift = preydrift(day-l)+preydriftrec
Figure 2F is generated by using the output of Figure 2C for a given time as the input for
2D and using the resulting output of 2D as the input for 2E. The resulting output of 2E
produces a single time point in the relationship in 2F. Performing this series of
computations across multiple days produces the entire relationship in 2F. 4D is generated
by taking the outputs of 4A and 4B for the same day. Note the relationship of 4A is
equivalent to 2F. The resulting outputs of 4A and 4B are multiplied to produce a final
ration for a given day. The prey abundance (4B) available for consumption during a prey
spike is capped at a maximum of 1.5*control drift to provide a limited benefit to the
individual fish. The final ration is used as input for 4C to generate 4D.
Population Model
The weight distributions from the organismal growth portion of the model are used to calculate
size-dependent first-year survival for a life-history matrix population model for each species and
life-history type. This incorporates the impact that reductions in size could have on population
growth rate and abundance. The first-year survival element of the transition matrix incorporates a
size-dependent survival rate for a three- or four-month interval (depending upon the species)
which takes the juveniles up to 12 months of age. This time represents the 4-month early winter
survival in freshwater for stream-type Chinook, coho, and sockeye models. For ocean-type
Chinook, it is the 3-month period the subyearling smolt spend in the estuary and nearshore
habitats (i.e. estuary survival). The weight distributions from the organismal model are converted
DRAFT 556
-------
DRAFT
to length distributions by applying condition factors from data for each modeled species (cf;
0.0095 for sockeye and 0.0115 for all others) as shown in Equation L.
Equation L: length(mm) = ((fish weight(g)/cf)A(l/3))*10
The relationship between length and early winter or estuary survival rate was adapted from Zabel
and Achord (2004) to match the survival rate for each control model population (Howell et al.
1985, Kostow 1995, Myers et al. 2006). The relationship is based on the length of a subyearling
salmon relative to the mean length of other competing subyearling salmon of the same species in
the system, Equation D, and relates that relative difference to size-dependent survival based upon
Equation S. The values for a and resulting size-dependent survival (survival (()) for control runs
for each species are listed in Table 2. The constant a is a species-specific parameter defined such
that it produces the correct control survival (|) value when Alength equals zero.
Equation D: Alength = fish length(mm) - mean length(mm)
Equation S: Survival $ = (e(«+(°-0329*A1^h))) / (1 + e(a+(o.o329*Aiength)))
Randomly selecting length values from the normal distribution calculated from the organismal
model output size and applying equations 1 and 2 generates a size-dependent survival probability
for each fish. This process was replicated 1000 times for each exposure scenario and
simultaneously 1000 times for the paired control scenario and results in a mean size-dependent
survival rate for each population. The resulting size-dependent survival rates are inserted in the
calculation of first-year survival in the respective control and pesticide-exposed transition
matrices.
The investigation of population-level responses to pesticide exposures uses life-history
projection matrix models. Individuals within a population exhibit various growth, reproduction,
and survivorship rates depending on their developmental or life-history stage or age. These age
specific characteristics are depicted in the life-history graph (Figure 1A-D) in which transitions
are depicted as arrows. The nonzero matrix elements represent transitions corresponding to
reproductive contribution or survival, located in the top row and the subdiagonal of the matrix,
respectively (Figure IE). The survival transitions in the life-history graph are incorporated into
the n x n square matrix (A) by assigning each age a number (1 through n) and each transition
from age i to age j becomes the element a;j of matrix A (i = row, j = column) and represent the
proportion of the individuals in each age passing to the next age as a result of survival. The
reproductive element (aij) gives the number of offspring that hatch per individual in the
contributing age, j. The reproductive element value incorporates the proportion of females in
each age, the proportion of females in the age that are sexually mature, fecundity, fertilization
success, and hatch success.
DRAFT 557
-------
DRAFT
In order to understand the relative impacts of a short-term exposure of a single pesticide on
exposed vs. unexposed fish, we used parameters for an idealized control population that exhibits
an increasing population growth rate. All characteristics exhibit density independent dynamics.
The models assume closed systems, allowing no migration impact on population size. No
stochastic impacts are included beyond natural variability as represented by selecting parameter
values from a normal distribution about a mean each model iteration (year). Ocean conditions,
freshwater habitat, fishing pressure, and marine resource availability were assumed constant and
density independent.
In the model an individual fish experiences an exposure once as a subyearling (during its first
spring) and never again. The pesticide exposure is assumed to occur annually. All subyearlings
within a given population are assumed to be exposed to the pesticide. No other age classes
experience the exposure. The model integrates this as every brood class being exposed as
subyearlings and thus the vital demographic rates of the transition matrix are continually
impacted in the same manner. Regardless of the level of AChE inhibition due to the direct
exposure, only the sublethal effects are incorporated in the models at this time.
The model recalculates first-year survival for each run using a size-dependent survival value
selected from a normal distribution with the mean and standard deviation produced by Equation
S. Population model output consists of the percent change in lambda from the unexposed control
populations derived from the mean of two thousand calculations of both the unexposed control
population and the pesticide exposed population. Change in lambda, representing alterations to
the population productivity, was selected as the primary model output for reasons outlined
previously.
A prospective analysis of the transition matrix, A, (Caswell 2001) explored the intrinsic
population growth rate as a function of the vital rates. The intrinsic population growth rate, A,,
equals the dominant eigenvalue of A and was calculated using matrix analysis software
(MATLAB version 7.7.0 by The Math Works Inc., Natick, MA). Therefore A, is calculated
directly from the matrix and running projections of abundances over time is redundant and
DRAFT 558
-------
DRAFT
unnecessary. The stable age distribution, the proportional distribution of individuals among the
ages when the population is at equilibrium, is calculated as the right normalized eigenvector
corresponding to the dominant eigenvalue A,. Variability was integrated by repeating the
calculation of A, 2000 times selecting the values in the transition matrix from their normal
distribution defined by the mean standard deviation. The influence of each matrix element, a;j, on
A, was assessed by calculating the sensitivity values for A. The sensitivity of matrix element fly-
equals the rate of change in A, with respect to ay, defined by 8 A/ 8%. Higher sensitivity values
indicate greater influence on A. The elasticity of matrix element atj is defined as the proportional
change in A relative to the proportional change in ay, and equals (%/A) times the sensitivity ofay.
One characteristic of elasticity analysis is that the elasticity values for a transition matrix sum to
unity (one). The unity characteristic also allows comparison of the influence of transition
elements and comparison across matrices.
Due to differences in the life-history strategies, specifically lifespan, age at reproduction and first
year residence and migration habits, four life-history models were constructed. This was done to
encompass the different responses to freshwater pesticide exposures and assess potentially
different population-level responses. Separate models were constructed for coho, sockeye,
ocean-type and stream-type Chinook. In all cases transition values were determined from
literature data on survival and reproductive characteristics of each species.
A life-history model was constructed for coho salmon (O. kisutch) with a maximum age of 3.
Spawning occurs in late fall and early winter with emergence from March to May. Fry spend 14-
18 months in freshwater, smolt and spend 16-20 months in the saltwater before returning to
spawn (Pess et al. 2002). Survival numbers were summarized in Knudsen et al. (2002) as
follows. The average fecundity of each female is 4500 with a standard deviation of 500. The
observed number of males:females was 1:1. Survival from spawning to emergence is 0.3 (0.07).
Survival from emergence to smolt is 0.0296 (0.00029) and marine survival is 0.05 (0.01). All
parameters followed a normal distribution (Knudson et al. 2002). The calculated values used in
the matrix are listed in Table 3. The growth period for first year coho was set at 180 days to
represent the time from mid-spring to mid-fall when the temperatures and resources drop and
somatic growth slows (Knudson et al. 2002).
DRAFT 559
-------
DRAFT
Life-history models for sockeye salmon (O. nerka) were based upon the lake wintering
populations of Lake Washington, Washington, USA. These female sockeye salmon spend one
winter in freshwater, then migrate to the ocean to spend three to four winters before returning to
spawn at ages 4 or 5. Jacks return at age 2 after only one winter in the ocean. The age proportion
of returning adults is 0.03, 0.82, and 0.15 for ages 3, 4 and 5, respectively (Gustafson et al. 1997).
All age 3 returning adults are males. Hatch rate and first year survival were calculated from
brood year data on escapement, resulting presmolts and returning adults (Pauley et al. 1989) and
fecundity (McGurk 2000). Fecundity values for age 4 females were 3374 (473) and for age 5
females were 4058 (557) (McGurk 2000). First year survival rates were 0.737/month (Gustafson
et al. 1997). Ocean survival rates were calculated based upon brood data and the findings that
90% of ocean mortality occurs during the first 4 months of ocean residence (Pauley et al. 1989).
Matrix values used in the sockeye baseline model are listed in Table 3. The 168 day growth
period represents the time from lake entry to early fall when the temperature drops and somatic
growth slows (Gustafson et al. 1997).
A life-history model was constructed for ocean-type Chinook salmon (O. tshawytscha) with a
maximum female age of 5 and reproductive maturity at ages 3, 4 or 5. Ocean-type Chinook
migrate from their natal stream within a couple months of hatching and spend several months
rearing in estuary and nearshore habitats before continuing on to the open ocean. Transition
values were determined from literature data on survival and reproductive characteristics from
several ocean-type Chinook populations in the Columbia River system (Healey and Heard 1984,
Howell et al. 1985, Roni and Quinn 1995, Ratner et al. 1997, PSCCTC 2002, Green and Beechie
2004). The sex ratio of spawners was approximately 1:1. Estimated size-based fecundity of
4511(65), 5184(89), and 5812(102) was calculated based on data from Howell et al., 1985, using
length-fecundity relationships from Healy and Heard (1984). Control matrix values for the
Chinook model are listed in Table 3. The growth period of 140 days encompasses the time the
fish rear in freshwater prior to entering the estuary and open ocean. The first three months of
estuary/ocean survival are the size-dependent stage. Size data for determining subyearling
Chinook condition indices came from data collected in the lower Columbia River and estuary
(Johnson et al. 2007).
DRAFT 560
-------
DRAFT
An age-structured life-history matrix model for stream-type Chinook salmon with a maximum
age of 5 was defined based upon literature data on Yakima River spring Chinook from Knudsen
et al. (2006) and Fast et al. (1988), with sex ratios of 0.035, 0.62 and 0.62 for females spawning
at ages 3, 4, and 5, respectively. Length data from Fast et al. (1988) was used to calculate
fecundity from the length-fecundity relationships in Healy and Heard (1984). The 184-day
growth period produces control fish with a mean size of 96mm, within the observed range
documented in the fall prior to the first winter (Beckman et al. 2000). The size-dependent
survival encompasses the 4 early winter months, up until the fish are 12 months old.
Acute Toxicity Model
In order to estimate the population-level responses of exposure to lethal pesticide concentrations,
acute mortality models were constructed based upon the control life-history matrices described
above. The acute responses are modeled as direct reduction in the first year survival rate (SI).
All subyearling salmon are assumed to be exposed in each scenario. Exposures are assumed to
result in a cumulative reduction in survival as defined by the concentration and the dose-response
curve as defined by the LC50 and slope for each pesticide. A sigmoid dose-response relationship
is used to accurately handle responses well away from LC50 and to be consistent with other
does-response relationships. The model inputs for each scenario are the exposure concentration
and acute fish LC50, as well as the sigmoid slope for the LC50. For a given concentration a
pesticide survival rate (1-mortality) is calculated and is multiplied by the control first-year
survival rate, producing an exposed scenario first-year survival for the life-history matrix.
Variability is incorporated as described above using mean and standard deviation of normally
distributed survival and reproductive rates and model output consists of the percent change in
lambda from unexposed control populations derived from the mean of 1000 calculations of both
the unexposed control population and the pesticide exposed population. The percent change in
lambda is considered different from control when the difference is greater than the percent of one
standard deviation from the control lambda.
DRAFT 561
-------
DRAFT
Results
Sensitivity Analysis
A sensitivity analysis conducted on the organismal model revealed that changes in the control
somatic growth rate had the greatest influence on the final weights (Table 1). While this
parameter value was experimentally derived for another species (sockeye salmon; Brett et al.
1969), this value was adapted for each model species and is within the variability reported in the
literature for other salmonids (reviewed in Weatherley and Gill 1995). Other parameters related
to the daily growth rate calculation, including the growth to ration slope (Mgr) and the control
ration produced strong sensitivity values. Initial weight, the prey recovery rate and the prey floor
also strongly influenced the final weight values (Table 1). Large changes (0.5 to 2X) in the other
key parameters produced proportionate changes in final weight.
The sensitivity analysis of all four of the control population matrices predicted the greatest
changes in population growth rate (k) result from changes in first-year survival. Parameter values
and their corresponding sensitivity values are listed in Table 4. The elasticity values for the
transition matrices also corresponded to the driving influence of first-year survival, with
contributions to lambda of 0.33 for coho, 0.29 for ocean-type Chinook, 0.25 for stream-type
Chinook, and 0.24 for sockeye.
Model Output
Organismal and population model outputs for all scenarios are shown in Tables 5-16 and were
summarized as graphs in the main text. As expected, greater changes in population growth
resulted from longer exposures to the pesticides. The factors driving the level of change in
lambda were the Prey Drift and relative AChE Activity parameters determined by the toxicity
values for each pesticide (Table 3). The low Prey Abundance ECso values drive the effects for all
three compounds at most concentrations investigated since they have higher AChE EC50 values
(Tables 3 & 9-16).
DRAFT 562
-------
DRAFT
Output from the acute toxicity models was presented in the Risk Characterization section of the
main text. Increases in direct mortality during the first year of life produced large impacts on the
population growth rates for all the life-history strategies.
While strong trends in effects were seen for each pesticide across all four life-history
strategies modeled, some slight differences were apparent. The similarity in patterns likely stems
from using the same toxicity values for all four models, while the differences are consequences
of distinctions between the life-history matrices. The stream-type Chinook and sockeye models
produced very similar results as measured as the percent change in population growth rate. The
ocean-type Chinook and coho models output produced the greatest changes in lambda resulting
from the pesticide exposures. When looking for similarities in parameters to explain the ranking,
no single life history parameter or characteristic, such as lifespan, reproductive ages, age
distribution, lambda and standard deviation, or first-year survival show a pattern that matches
this consistent output. Combining these factors into the transition matrix for each life-history and
conducting the sensitivity and elasticity analyses revealed that changes in first-year survival
produced the greatest changes in lambda. In addition, the elasticity analysis can be used to
predict relative contribution to lambda from changes in first-year survival on a per unit basis. As
detailed by the elasticity values reported above, the same change in first-year survival will
produce a slightly greater change in the population growth rate for coho and ocean-type Chinook
than for stream-type Chinook and sockeye. While some life-history characteristics may lead a
population to be more vulnerable to an impact, the culmination of age structure, survival and
reproductive rates as a whole strongly influences the population-level response.
DRAFT 563
-------
DRAFT
Figure 1: Life-History Graphs and Transition Matrix for coho (A), sockeye (B) and Chinook (C)
salmon. The life-history graph for a population labeled by age, with each transition element
labeled according to the matrix position, a;j, i row and j column. Dashed lines represent
reproductive contribution and solid lines represent survival transitions. D) The transition matrix
for the life-history graph depicted in C.
Figure 2: Relationships used to link anticholinesterase exposure to the organism's ability to
acquire food (potential ration). See text for details. Relationships in B, C, and D utilize empirical
data. Closed circles represent control conditions. Open circles represent the exposed (inhibited)
condition. A) Representation of a constant level of anticholinesterase pesticide exposure (either a
single compound or mixtures). B) Sigmoidal relationship between exposure concentration and
steady-state acetylcholinesterase (AChE) activity showing a dose-dependent reduction defined
by control activity (horizontal line, Ac), sigmoidal (i.e. hille) slope (AChE slope), and the
concentration producing 50% inhibition (vertical line, EC50). C) Timecourse of
acetylcholinesterase inhibition based on modeling the time-to-effect and time-to-recovery as
single exponential curves with different time-constants. At the start of the exposure AChE
activity will be at control and then decline toward the inhibited activity (Ai) based on Panel B.
D) Linear model relating acetylcholinesterase activity to feeding behavior using a line that passes
through the feeding (Fc) and activity (Ac) control conditions with a slope of Mfa. E) The
relationship between feeding behavior and the potential ratio an organism could acquire (if not
food limited) used a line passing through the control conditions (Fc as in Panel D and the control
ration, Re) and through the origin producing a slope (Mrf) equal to Rc/Fc. F) Timecourse for
effect of exposure to anticholinesterase on potential ration produced by combining C & E.
Figure 3: Relationships used to link anticholinesterase exposure to the availability of prey. See
text for details. Relationships in B and C utilize empirical data. Closed circles represent control
conditions. Open circles represent the exposed (inhibited) condition. A) Representation of a
constant level of anticholinesterase pesticide exposure (either single compound or mixtures). B)
Sigmoidal relationship between exposure concentration and relative prey abundance showing a
dose-dependent reduction defined by control abundance (horizontal line at 1, PC), sigmoid (i.e.
hille) slope (prey slope), the concentration producing a 50% reduction in prey (vertical line,
DRAFT 564
-------
DRAFT
ECso), and a minimum abundance always present (horizontal line denoted as floor, Pf). C)
Timecourse of prey abundance including a 1-day spike in prey drift relative to the available prey
and the level of toxicity followed by a drop to the level of impact or the floor whichever is
greater. During extended exposures at low toxicity recovery can begin at the constant prey influx
rate multiplied by the current level of toxicity. After exposure recovery to control prey drift is at
the constant rate of influx from upstream habitats.
Figure 4: Relationships used to link anticholinesterase exposure to growth rate relating to long-
term weight gain of each fish. See text for details. Relationships in A, B, and C utilize empirical
data. Closed circles represent control conditions. Open circles (e.g. Ai) represent the exposed
(inhibited) condition. A&B) Relationships describing the Timecourse of the effects of
anticholinesterase exposure on the organisms ability to capture food (Panel A, potential ration)
and the availability of food to capture (Panel B, relative prey abundance). The figures are the
same as those in Figures 2F and 3C, respectively. For a given exposure concentration and time,
multiplying potential ration by relative prey abundance yields the final ration acquired by the
organism. C) A linear model was used to relate final ration to growth rate using a line passing
through the control conditions and through the maintenance condition with a slope denoted by
Mgr. D) Timecourse for effect of exposure to anticholinesterase on growth rate produced by
combining A, B, & C.
DRAFT 565
-------
DRAFT
Table 1. List of values used for control parameters to model organismal growth
and the model sensitivity to changes in the parameter.
Parameter
acetylcholinesterase activity (Ac)
feeding (Fc)
ration (Re)
feeding vs. activity slope (Mfa)
ration vs. feeding slope (Mrf)
growth vs. ration slope (Mgr)
growth vs. activity slope (Mga)
initial weight
control prey drift
AChE impact time-to-effect (t1/2)
AChE time-to-recovery (t1/2)
prey floor
prey recovery rate
somatic growth rate (Gc)
Value1
1.04'b
1.04'b
5% weight/day65
1.0s
5 (Rc/Fc)
0.35b
1 .75 (Mfa*Mrf*Mgr)
1 gram"
1.04
0.5 daya
0.25 dayslu
0.2011
0.011"
1.3"
Error"
0.06b
0.05b
0.05'
0.1b
-
0.02b
-
0.1b
0.0511
n/a
n/a
n/a
n/a
0.06b
Sensitivity"
-0.167
0.088
-0.547
-0.047
-
-0.547
-
1.00
0.116
0.005
-0.0001
0.178
0.323
2.531
mean value of a normal distribution used in the model or constant value when no corresponding
error is listed
2 standard deviation of the normal distribution used in the model
3 mean sensitivity when baseline parameter is changed over range of 0.5 to 2-fold
4 other values relative to control
5 derived from Sandahl et al., (2005)
6 derived from Brett et al., (1969)
7 data from Brett et al., (1969) has no variability (ration was the independent variable) so a
variability of 1% was selected to introduce some variability
8 consistent with field-collected data for juvenile Chinook (Nelson et al., 2004)
estimated from Ferrari et al., 2004
10
consistent with Labenia et al., 2007
11 estimated from Van den Brink et al., 1996
12 derived from Ward et al., 1995, Van den Brink et al., 1996, Colville et al., 2008
13 derived from Brett and adapted for ocean-type Chinook, used for sensitivity analysis
DRAFT
566
-------
DRAFT
Table 2. Species specific control parameters to model organismal growth and survival rates.
Growth period and survival rate are determined from the literature data listed for each species. Gc
and a were calculated to make the basic model produce the appropriate size and survival values
from the literature.
days to run organismal
growth model
growth rate
% body wt/day (Gc)
a from equation S
Control Survival §
Chinook
Stream-type1
184
1.28
-0.33
0.418
Chinook
Ocean-type2
140
1.30
-1.99
0.169
Coho3
184
0.90
-0.802
0.310
Sockeye4
168
1.183
-0.871
0.295
1 Values from data in Healy and Heard 1984, Fast et al., 1988, Beckman et al., 2000, Knudsen et
al., 2006
2 Values from data in Healey and Heard 1984, Howell et al., 1985, Roni and Quinn 1995, Ratner
et al.,1997, PSCCTC 2002, Green and Beechie, 2004, Johnson et al., 2007
3 Values from data in Pess et al., 2002, Knudsen et al., 2002
4 Values from data in Pauley et al., 1989, Gustafson et al., 1997, McGurk 2000
Table 3. Effects values (ug/L) and slopes for AChE activity, acute fish lethality, and prey
abundance dose-response curves.
compound
carbaryl
carbofuran
methomyl
AChE Activity
EC501 ug/L
145.8
58.4
213
AChE Activity
slope
0.95
0.95
0.95
Fish lethality
LC502ug/L
250
164
560
Fish
lethality
slope
3.63
3.63
3.63
Prey
Abundance
EC504ug/L
4.33
1.22
20.74
Prey
Abundance
Slope5
5.5
5.5
5.5
Values from Laetz et. al 2009
2 Values from EPA BEs
3 sigmoidal slope that produces responses with a probit slope of Peterson, see text.
4 Values from analysis of global search of reported LC50 and ECSOs reported in EPA's Ecotox
database. See text.
5 Values from Peterson et al., 2001
DRAFT
567
-------
DRAFT
Table 4. Matrix transition element and sensitivity (S) and elasticity (E) values for each model species. These control values are listed by
the transition element taken from the life-history graphs as depicted in Figure 1 and the literature data described in the method text.
Blank cells indicate elements that are not in the transition matrix for a particular species. The influence of each matrix element on A, was
assessed by calculating the sensitivity (S) and elasticity (E) values for A. The sensitivity of matrix element a/, equals the rate of change
in X with respect to the transition element, defined by 8X/ 5a. The elasticity of transition element a/, is defined as the proportional change
in X relative to the proportional change in a/,, and equals (a//X) times the sensitivity of a/,. Elasticity values allow comparison of the
influence of individual transition elements and comparison across matrices.
Transition
Element
S1
S2
S3
S4
R3
R4
R5
Chinook
Stream-type
Value1
0.0643
0.1160
0.17005
0.04
0.5807
746.73
1020.36
S
3.844
2.132
1.448
0.319
0.00184
0.000313
1.25E-05
E
0.247
0.247
0.246
0.0127
0.0011
0.233
0.0127
Chinook
Ocean-type
Value2
0.0056
0.48
0.246
0.136
313.8
677.1
1028
S
57.13
0.670
0.476
0.136
0.0006
0.000146
1.80E-05
E
0.292
0.292
0.106
0.0168
0.186
0.0896
0.0168
Coho
ValueJ
0.0296
0.0505
732.8
S
11.59
6.809
0.000469
E
0.333
0.333
0.333
Sockeye
Value4
0.0257
0.183
0.499
0.1377
379.57
608.7
S
9.441
1.326
0.486
0.322
0.000537
7.28E-05
E
0.239
0.239
0.239
0.0437
0.195
0.0437
1 Value calculated from data in Healy and Heard 1984, Fast et al., 1988, Beckman et al., 2000, Knudsen et al., 2006
2 Value calculated from data in Healey and Heard 1984, Howell et al., 1985, Roni and Quinn 1995, Ratner et al., 1997, PSCCTC 2002,
Green and Beechie, 2004, Johnson et al., 2007
3 Value calculated from data in Pess et al., 2002, Knudsen et al., 2002
4 Value calculated from data in Pauley et al., 1989, Gustafson et al., 1997, McGurk 2000
DRAFT
568
-------
DRAFT
Figure 1.
A. Coho
B. Sockeye
914
C. Chinook,
ocean-type
and stream-
type
D. Transition
Matrix
A=
0
S1=a2J
0
0
0
0
0
S2=a32
0
0
R3=a«
0
0
S3=a43
0
R4=aT4
0
0
0
S4=a54
R5=af5
0
0
0
0
DRAFT
569
-------
DRAFT
Figure 2.
A
CD_
3
CO
O
Q.
X
start
end
time
"o
CD
LLJ
^
O
control AChE (Ac)
.AChE slope
start
exposure
AChE time-to-effect
end AChE time-to-recovery
time
D
g
'>
03
_c
0)
.0
D)
c
T3
-------
A
-------
DRAFT
c
o
"CD
eg
'**-•
C
0
-t-»
O
Q.
start
time
B
ro
start
I
Pi
time
Figure 4.
final ration = potential ration X prey abundance
\
maintenance ... ,
final ration
D
CD
o
D)
end
time
DRAFT
572
-------
DRAFT
References
Anderson, B.S., Hunt, J.W., Phillips, B.M., Nicely, P.A., Gilbert, K.D., DeVlaming, V., Connor,
V., Richard, N., and Tjeerdemat, R.S. 2003. Ecotoxicologic impacts of agricultural drain
water in the Salinas River, California, USA. Environmental Toxicology and Chemistry
22:2375-2384.
Baldwin, D.B., Spromberg, J.A., Collier, T.K., and Scholz, N.L. 2009. A fish of many scales:
extrapolating sublethal pesticide exposures to the productivity of wild salmon
populations. Ecological Applications. In press.
Beamish, R.J., and Mahnken, C. 2001. A critical size and period hypothesis to explain natural
regulation of salmon abundance and the linkage to climate and climate change. Progress
in Oceanography 49: 423-437.
Beauvais, S.L., Jones, S.B., Brewer, S.K., and Little, E.E. 2000. Physiological measures of
neurotoxicity of diazinon and malathion to larval rainbow trout (Oncorhynchus mykiss)
and their correlation with behavioral measures. Environmental Toxicology and Chemistry
19:1875-1880.
Beckman, B.R., Larsen, D.A., Sharpe, C., Lee-Pawlak, B., Schreck, C.B., and Dickhoff, W.W.
2000. Physiological status of naturally reared juvenile spring Chinook salmon in the
Yakima River: seasonal dynamics and changes associated with smolting. Transactions of
the American Fisheries Society 129:727-753.
Brett, J.R., Shelbourn, I.E., and Shoop, C.T. 1969. Growth rate and body composition of
fingerling sockeye salmon, Oncorhynchus nerka, in relation to temperature and ration
size. Journal of the Fisheries Board of Canada 26:2363-2394.
DRAFT 573
-------
DRAFT
Brewer, S.K., Little, E.E., DeLonay, A.J., Beauvais, S.L., Jones, S.B. and Ellersieck, M.R. 2001.
Behavioral dysfunctions correlate to altered physiology in rainbow trout (Oncorynchus
mykiss) exposed to cholinesterase-inhibiting chemicals. Archives of Environmental
Contamination and Toxicology 40:70-76.
Caswell, H. 2001. Matrix population models: Construction, analysis, and interpretation.
Sundarland, MA, USA: Sinauer Assoc.
Chang, K.H., Sakamoto, M., and Hanazato, T. 2005. Impact of pesticide application on
zooplankton communities with different densities of invertebrate predators: An
experimental analysis using small-scale mesocosms. Aquatic Toxicology 72:373-382.
Cold, A., and Forbes, V.E. 2004. Consequences of a short pulse of pesticide exposure for
survival and reproduction ofGammaruspulex. Aquatic Toxicology 67: 287-299.
Colville, A., Jones, P., Pablo, F., Krassoi, F., Hose, G., and Lim, R. 2008. Effects of chlorpyrifos
on macroinvertebrate communities in coastal stream mesocosms. Ecotoxicology 17: 173-
180.
Cuffney, T.F., Wallace, J.B., and Webster, J.R. 1984. Pesticide manipulation of a headwater
stream: invertebrate responses and their significance for ecosystem processes. Freshwater
Invertebrate Biology 3: 153-171.
Davies, P.E., and Cook, L.S.J. 1993. Catastrophic macroinvertebrate drift and sublethal effects
on brown trout, Salmo trutta, caused by cypermethrin spraying on a Tasmanian stream.
Aquatic Toxicology 27(3-4):201-224.
Fast, D.E., Hubble, J.D., and Kohn, M.S. 1988. Yakima River Spring Chinook Enhancement
Study, Annual Report FY 1988.U.S. Department of Energy, Bonneville Power
Administration Division of Fish and Wildlife. Project No. 82-16. 101pp.
DRAFT 574
-------
DRAFT
Ferrari, A., Venturino, A., and Pechen de D'Angelo, A.M. 2004. Time course of brain
cholinesterase inhibition and recovery following acute and subacute azinphosmethyl,
parathion, and carbaryl exposure in the goldfish (Carassius auratus). Ecotoxicology and
Environmental Safety 57:420-425.
Fleeger, J.W., Carman, K.R., andNisbet, R.M. 2003. Indirect effects of contaminants in aquatic
ecosystems. Science of the Total Environment 317:207'-233.
Forbes, V.E., and Cold, A. 2005. Effects of the pyrethroid esfenvalerate on life-cycle traits and
population dynamics of Chironomus riparius - Importance of exposure scenario.
Environmental Toxicology and Chemistry 24: 78-86.
Greene, C.M., and Beechie, TJ. 2004. Consequences of potential density-dependent mechanisms
on recovery of ocean-type Chinook salmon (Oncorhynchus tshawytschd). Canadian
Journal of Fisheries and Aquatic Sciences 61:590-602.
Gustafson, R.G., Wainwright, T.C., Winans, G.A., Waknitz, F.W., Parker, L.T., and Waples,
R.S. 1997. Status review of sockeye salmon from Washington and Oregon. U.S. Dept.
Commerce, NOAA Tech. Memo. NMFS-NWFSC-33, 282 pp.
Healey, M.C. 1982. Timing and relative intensity of size-selective mortality of juvenile chum
salmon (Oncorhynchus ketd) during early sea life. Canadian Journal of Fisheries and
Aquatic Sciences 39:952-957.
Healey, M.C. 1991. Life history of Chinook salmon (Oncorhynchus tshawytschd). Pages 311-
394 in C. Groot and L. Margolis, editors. Pacific salmon life histories. University of
British Columbia Press, Vancouver, Canada.
Healey, M.C., and Heard, W.R. 1984. Inter- and intra-population variation in the fecundity of
Chinook salmon (Oncoryhchus tshawytschd) and its relevance to life history theory.
Canadian Journal of Fisheries and Aquatic Sciences 41:476-483.
DRAFT 575
-------
DRAFT
Higgs, D.A., MacDonald, J.S., Levings, C.D., and Dosanjh, B.S. 1995. Nutrition and feeding
habits in relation to life history stage. Pages 159-315. in C. Groot, L. Margolis, and W.C.
Clarke, editors. Physiological Ecology of Pacific Salmon. University of British Columbia
Press, Vancouver, Canada.
Holtby, L.B., Andersen, B.C., and Kadowak, R.K. 1990. Importance of smolt size and early
ocean growth to interannual variability in marine survival of coho salmon (Oncorhynchus
kisutch). Canadian Journal of Fisheries and Aquatic Sciences 47:2181-2194.
Howell, P., Jones, K., Scarnecchia, D., LaVoy, L., Kendra, W., Ortmann, D., Neff, C., Petrosky,
C., and Thurow, R. 1985. Stock assessment of Columbia River anadromous salmonids
Volume I: Chinook, coho, chum, and sockeye salmon stock summaries. Final Report to
Bonneville Power Administration. Bonneville Power Administration, P.O Box 3621,
Portland OR 97208, DE-AI79-84BP12737, Project No. 83-335. 579 p.
Johnson, K.R., Jepson, P.C. and Jenkins, J.J. 2008. Esfenvalerate-induced case-abandonment in
the larvae of the caddisfly (Brachycentrus americanus). Environmental Toxicology and
Chemistry 27: 397-403.
Johnson, L.L., Ylitalo, G.M., Arkoosh, M.R., Kagley, A.N., Stafford, C.L., Bolton, J.L.,
Buzitis, J., Anulacion, B.F., and Collier, T.K. 2007. Contaminant exposure in outmigrant
juvenile salmon from Pacific Northwest estuaries. Environmental Monitoring and
Assessment 124:167-194.
Knudsen, C.M., Schroder, S.L., Busack, C.A., Johnston, M.V., Pearsons, T.N., Bosch, W.J., and
Fast, D.E. 2006. Comparison of life history traits between first-generation hatchery and
wild upper Yakima River spring Chinook salmon. Transactions of the American
Fisheries Society 135:1130-1144.
DRAFT 576
-------
DRAFT
Knudsen, E.E., Symmes, E.W., and Margraf, FJ. 2002. Searching for an ecological life history
approach to salmon escapement management. American Fisheries Society Symposium
34:261-276.
Kostow, K. 1995. Biennial Report on the Status of Wild Fish in Oregon. Oreg. Dep. Fish Wildl.
Rep., 217 p. + app. (Available from Oregon Department of Fish and Wildlife, P.O. Box
59, Portland, OR 97207.)
Labenia, J.S., Baldwin, D.H., French, B.L., Davis, J.W., and Scholz, N.L. 2007. Behavioral
impairment and increased predation mortality in cutthroat trout exposed to carbaryl.
Marine Ecology Progress Series 329: 1-11.
Laetz, C.A., Baldwin, D.H., Collier, T.K., Hebert, V., Stark, J.D., and Scholz, N.L. 2009. The
synergistic toxicity of pesticide mixtures: implications for risk assessment and the
conservation of endangered Pacific salmon. Environmental Health Perspectives 117:348-
353.
Liess, M., and Schulz, R. 1999. Linking insecticide contamination and population response in an
agricultural stream. Environmental Toxicology and Chemistry 18(9): 1948-1955.
McClure, M.M., Holmes, E.E., Sanderson, B.L., and Jordan, C.E. 2003. A large-scale,
multispecies status assessment: anadromous salmonids in the Columbia River Basin.
Ecological Applications 13(4):964-989.
McGurk, M.D. 2000. Comparison of fecundity-length-latitude relationships between
nonanadromous (kokanee) and anadromous sockeye salmon (Oncorhynchus nerka).
Canadian Journal of Zoology 78: 1791-1805.
Moore, A., and Waring, C.P. 1996. Sublethal effects of the pesticide Diazinon on olfactory
function in mature male Atlantic salmon parr. Journal of Fish Biology 48:758-775.
DRAFT 577
-------
DRAFT
Morgan, M.J., and Kiceniuk, J.W. 1990. Effect of fenitrothion on the foraging behavior of
juvenile Atlantic salmon. Environmental Toxicology and Chemistry 9:489-495.
Myers, J., Busack, C., Rawding, D., Marshall, A., Teel, D., Van Doornik, D.M., and Maher,
M.T. 2006. Historical population structure of Pacific salmonids in the Willamette River
and lower Columbia River basins. U.S. Dept. Commerce, NOAA Tech. Memo. NMFS-
NWFSC-73, Slip.
Nelson, T. S., G. Ruggerone, H. Kim, R. Schaefer, and M. Boles. 2004. Juvenile Chinook
migration, growth and habitat use in the Lower Green River, Duwamish River and
Nearshore of Elliott Bay 2001-2003, Draft Report. King County Department of Natural
Resources and Parks. Seattle, Washington.
PSCCTC (Pacific Salmon Commission Chinook Technical Committee). 2002. Pacific Salmon
Commission Joint Chinook Technical Committee Report: Annual Exploitation Rate
Analysis and Model Calibration. Report TCCHINOOK (02)-3. Vancouver, British
Columbia, Canada.
Pauley, G.B., Risher, R., and Thomas, G.L. 1989. Species profiles: Life histories and
environmental requirements of coastal fishes and invertebrates (Pacific Northwest) -
sockeye salmon. U.S. Fish Wildl. Serv., Biol. Rep. 82(11.116). U.S. Army Corps of
Engineers, TR EL-82-4, 22 p.
Pess, G.R., Montgomery, D.R., Steel, E.A., Bilby, R.E., Feist, B.E., and Greenberg, H.M. 2002.
Landscape characteristics, land use, and coho salmon (Oncorhynchus kisutch) abundance,
Snohomish River, Wash., U.S.A. Canadian Journal of Fisheries and Aquatic Sciences
59:613-623.
DRAFT 578
-------
DRAFT
Peterson, J.L., Jepson, P.C., and Jenkins, JJ. 2001a. A test system to evaluate the susceptibility
of Oregon, USA, native stream invertebrates to triclopyr and carbaryl. Environmental
Toxicology and Chemistry 20:2205-2214.
Peterson, J.L., Jepson, P.C., and Jenkins, J.J. 2001b. Effect of varying pesticide exposure
duration and concentration on the toxicity of carbaryl to two field-collected stream
invertebrates, Calineuria californica (Plecoptera : Perlidae) and Cinygma sp
(Ephemeroptera : Heptageniidae). Environmental Toxicology and Chemistry 20: 2215-
2223.
Quinn, T.P. 2005. The behavior and ecology of Pacific salmon and trout. University of
Washington Press. Seattle, WA. 378pp.
Ratner, S., Lande, R., and Roper, B.B. 1997. Population viability analysis of spring Chinook
salmon in the South Umpqua River, Oregon. Conservation Biology 11:879-889.
Relyea, R.A. 2005. The impact of insecticides and herbicides on the biodiversity and
productivity of aquatic communities. Ecological Applications 15(2): 618-627.
Roni, P., and Quinn, T.P. 1995. Geographic variation in size and age of North American
Chinook salmon. North American Journal of Fisheries Management 15:325-345.
Sakamoto, M., Chang, K.H., and Hanazato, T. 2006. Inhibition of development of anti-predator
morphology in the small cladoceran Bosmina by an insecticide: impact of an
anthropogenic chemical on prey-predator interactions. Freshwater Biology 51: 1974-
1983.
Sandahl, J.F., Baldwin, D.H., Jenkins, J.J., and Scholz, N.L. 2005. Comparative thresholds for
acetylcholinesterase inhibition and behavioral impairment in coho salmon exposed to
chlorpyrifos. Environmental Toxicology and Chemistry 24:136-145.
DRAFT 579
-------
DRAFT
Scholz, N.L., Truelove, N.K., French, B.L., Berejikian, B.A., Quinn, T.P., Casillas, E., and
Collier, T.K. 2000. Diazinon disrupts antipredator and homing behaviors in Chinook
salmon (Oncorhynchus tshawytscha). Canadian Journal of Fisheries and Aquatic
Sciences 51'(9): 1911-1918.
Schulz, R. 2004. Field studies on exposure, effects, and risk mitigation of aquatic nonpoint-
source insecticide pollution: A review. Journal of Environmental Quality 33:419-448.
Schulz, R., and Liess, M. 1999. A field study of the effects of agriculturally derived insecticide
input on stream macroinvertebrate dynamics. Aquatic Toxicology 46:155-176.
Schulz, R., and Liess, M. 200la. Acute and chronic effects of particle-associated fenvalerate on
stream macroinvertebrates: A runoff simulation study using outdoor microcosms.
Archives of Environmental Contamination and Toxicology 40: 481-488.
Schulz, R., and Liess, M. 2001b. Toxicity of aqueous-phase and suspended particle-associated
fenvalerate: Chronic effects after pulse-dosed exposure of Limnephilus lunatus
(Trichoptera). Environmental Toxicology and Chemistry 20: 185-190.
Schulz, R., Thiere, G., and Dabrowski, J.M. 2002. A combined microcosm and field approach to
evaluate the aquatic toxicity of azinphosmethyl to stream communities. Environmental
Toxicology and Chemistry 21:2172-2178.
Stark, J.D., Banks, I.E., and Vargas, R. 2004. How risky is risk assessment: The role that life
history strategies play in susceptibility of species to stress. Proceedings of the National
Academy of Sciences, USA 101:732-736.
Van den Brink, P.J., van Wijngaarden, R.P.A., Lucassen, W.G.H., Brock, T.C.M. and
Leeuwangh, P. 1996. Effects of the insecticide Dursban 4E (active ingredient
chlorpyrifos) in outdoor experimental ditches: II. Invertebrate community responses and
recovery. Environmental Toxicology and Chemistry 15(7): 1143-1153.
DRAFT 580
-------
DRAFT
Wallace, J.B., Huryn, A.D., and Lugthart, GJ. 1991. Colonization of a headwater stream during
3 years of seasonal insecticidal applications. Hydrobiologia 211: 65-76.
Ward, S., Arthington, A.H., and Pusey, BJ. 1995. The effects of a chronic application of
chlorpyrifos on the macroinvertebrate fauna in an outdoor artificial stream system -
species responses. Ecotoxicology and Environmental Safety 30: 2-23.
Waring, C.P., and Moore, A.P. 1997. Sublethal effects of a carbamate pesticide on pheromonal
mediated endocrine function in mature male Atlantic salmon (Salmo salar) parr. Fish
Physiology and Biochemistry 17:203-211.
Weatherley, A.H., and Gill, H.S. 1995. Growth. Pages 103-158. in C. Groot, L. Margolis, and
W.C. Clarke, editors. Physiological Ecology of Pacific Salmon. University of British
Columbia Press, Vancouver, Canada.
West, C.J., and Larkin, P. A. 1987. Evidence of size-selective mortality of juvenile sockeye
salmon (Oncorhynchus nerka) in Babine Lake, British Columbia. Canadian Journal of
Fisheries and Aquatic Sciences 44: 712-721.
Zabel, R.W., and Achord, S. 2004. Relating size of juveniles to survival within and among
populations of Chinook salmon. Ecology 85:795-806.
DRAFT 581
-------
DRAFT
Appendix 2. Species and Population Annual Rates of Growth
DRAFT 582
-------
DRAFT
Chinook Salmon
ESU
California Coastal
Central Valley Spring - Run
(Good et al., 2005 -90% Cl)
Lower Columbia River
(Good et al., 2005) (# =
McElhany et al., 2007)
Population
Eel River
Redwood Creek
Mad River
Humboldt Bay tributaries
Bear River
Mattole River
Tenmile to Gualala
Russain River
Butte Creek - spring run
Deer Creek - spring run
Mill Creek - spring run
Youngs Bay
Grays River - fall run
Big Creek
Elochoman River - fall run
Clatskanie River #
Mill, Abernathy, Germany Creeks - fall run
Scappose Creek
Coweeman River - fall run
Lower Cowlitz River - fall run
Upper Cowlitz River - fall run
Toutle River - fall run
Kalamaha River - fall run
Salmon Creek / Lewis River - fall run
Clackamas River - fall run
Washougal River - fall run
Sandy River - fall run
Lower Gorge tributaries
Upper Gorge tributaries - fall run
Hood River - fall run
Big White Salmon River - fall run
Sandy River - late fall run
North Fork Lewis River - late fall run
Upper Cowlitz River - spring run
Cispus River
Tilton River
Toutle River - spring run
Kalamaha River - spring run
Lewis River - spring run
Sandy River - spring run #
Big White Salmon River - spring run
Hood River - spring run
A-H=0
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.300
1.170
1.190
N/A
0.944
N/A
1.037
0.990
0.981
N/A
1.092
0.998
N/A
N/A
0.937
0.984
N/A
1.025
N/A
N/A
0.959
N/A
0.963
0.943
0.968
N/A
N/A
N/A
N/A
N/A
N/A
0.961
N/A
N/A
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.060
1.040
1.000
N/A
0.739
N/A
0.813
0.824
0.769
N/A
0.855
0.776
N/A
N/A
0.763
0.771
N/A
0.803
N/A
N/A
0.751
N/A
0.755
0.715
0.756
N/A
N/A
N/A
N/A
N/A
N/A
0.853
N/A
N/A
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.600
1.350
1.470
N/A
1.204
N/A
1.323
1.189
1.252
N/A
1.393
1.282
N/A
N/A
1.242
1.256
N/A
1.308
N/A
N/A
1.224
N/A
1.229
1.243
1.204
N/A
N/A
N/A
N/A
N/A
N/A
1.083
N/A
N/A
DRAFT
583
-------
DRAFT
Chinook Salmon (continued)
ESU
Upper Columbia River
Spring - Run (FCRPS)
Puget Sound (only have A
where hatchery fish = native
fish), (Good et al., 2005)
Sacramento River Winter -
Run (Good, 2005 - 90% Cl))
Population
Methow River
Twisp River
Chewuch River
Lost / Early River
Entiat River
Wenatchee River
Chiawawa River
Nason River
Upper Wenatchee River
White River
Little Wenatchee River
Nooksack - North Fork
Nooksack - South Fork
Lower Skagit
Upper Skagit
Upper Cascade
Lower Sauk
Upper Sauk
Suiattle
Stillaguamish - North Fork
Stillaguamish - South Fork
Skykomish
Snoqualmie
North Lake Washington
Cedar
Green
White
Puyallup
Nisqually
Skokomish
Dosewallips
Duckabush
Hamma Hamma
Mid Hood Canal
Dungeness
Elwha
Sacramento River - winter run
A-H=0
1.100
N/A
N/A
N/A
0.990
1.010
N/A
N/A
N/A
N/A
N/A
0.750
0.940
1.050
1.050
1.060
1.010
0.960
0.990
0.920
0.990
0.870
1.000
1.070
0.990
0.670
1.160
0.950
1.040
1.040
1.170
N/A
N/A
N/A
1.090
0.950
0.970
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.680
0.880
0.960
0.990
1.010
0.890
0.900
0.930
0.880
0.970
0.840
0.960
1.000
0.920
0.610
1.100
0.890
0.970
1.000
1.070
N/A
N/A
N/A
0.980
0.840
0.870
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.820
0.990
1.140
1.110
1.110
1.130
1.020
1.050
0.960
1.010
0.900
1.040
1.140
1.060
0.730
1.220
1.010
1.110
1.080
1.270
N/A
N/A
N/A
1.200
1.060
1.090
DRAFT
584
-------
DRAFT
Chinook Salmon (continued)
ESU
Snake River Fall - Run
(Good, 2005)
Snake River Spring/Summer
- Run (FCRPS)
Upper Williamette River
(McElhanyetal.,2007)
Population
Lower Snake River
Tucannon River
Wenaha River
Wallowa River
Lostine River
Minam River
Catherine Creek
Upper Grande Ronde River
South Fork Salmon River
Secesh River
Johnson Creek
Big Creek Spring Run
Big Creek Summer Run
Loon Creek
Marsh Creek
Bear Valley /Elk Creek
North Fork Salmon River
Lemhi River
Pahsimeroi River
East Fork Salmon Spring Run
East Fork Salmon Summer Run
Yankee Fork Spring Run
Yankee Fork Summer Run
Valley Creek Spring Run
Valley Creek Summer Run
Upper Salmon Spring Run
Upper Salmon Summer Run
Alturas Lake Creek
Imnaha River
Big Sheep Creek
Lick Creek
Clackamas River
Molalla River
North Santiam River
South Santiam River
Calapooia River
McKenzie River
Middle Fork Williamette River
Upper Fork Williamette River
A-H=0
1.024
1.000
1.100
N/A
1.050
1.050
0.970
N/A
1.110
1.070
N/A
1.090
1.090
N/A
1.080
1.100
N/A
1.020
1.080
1.040
1.040
N/A
N/A
N/A
N/A
1.060
1.060
N/A
1.050
N/A
N/A
0.967
N/A
N/A
N/A
N/A
0.927
N/A
N/A
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.849
N/A
N/A
N/A
N/A
0.761
N/A
N/A
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.102
N/A
N/A
N/A
N/A
1.129
N/A
N/A
DRAFT
585
-------
DRAFT
Chum Salmon
ESU
Columbia River
Hood Canal Summer- Run
(only have A where hatchery
fish reproductive potential =
native fish; Good et. al.,
2005)
Population
Youngs Bay
Gray's River
Big Creek
Elochoman River
Clatskanie River
Mill, Abernathy and German Creeks
Scappose Creek
Cowlitz River
Kalama River
Lewis River
Salmon Creek
Clackamus River
Sandy River
Washougal River
Lower Gorge tributaries
Upper Gorge tributaries
Jimmycomelately Creek
Salmon / Snow Creeks
Big / Little Quilcene rivers
Lilliwaup Creek
Hamma Hamma River
Duckabush River
Dosewallips River
Union River
Chimacum Creek
Big Beef Creek
Dewetto Creek
A-H=0
N/A
0.954
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.984
N/A
0.850
1.230
1.390
1.190
1.300
1.100
1.170
1.150
N/A
N/A
N/A
95% Cl -lower
N/A
0.855
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.883
N/A
0.690
1.130
1.170
0.750
1.110
0.930
0.930
1.050
N/A
N/A
N/A
95% Cl - upper
N/A
1.064
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.096
N/A
1.010
1.330
1.610
1.630
1.490
1.270
1.410
1.250
N/A
N/A
N/A
DRAFT
586
-------
DRAFT
Coho Salmon
ESU
Central California Coast
Lower Columbia River
(Goodetal., 2005)
Population
Ten Mile River
Noyo River
Big River
Navarro River
Garcia River
Other Mendacino County Rivers
Gualala River
Russain River
Other Sonoma County Rivers
Martin County
San Mateo County
Santa Cruz County
San Lorenzo River
Youngs Bay
Grays River
Elochoman River
Clatskanie River
Mill, Abernathy, Germany Creeks
Scappose Creek
Cispus River
Tilton River
Upper Cowlitz River
Lower Cowlitz River
North Fork Toutle River
South Fork Toutle River
Coweeman River
Kalama River
North Fork Lewis River
East Fork Lewis River
Upper Clackamas River
Lower Clackamas River
Salmon Creek
Upper Sandy River
Lower Sandy River
Washougal River
Lower Columbia River gorge tributaries
White Salmon
Upper Columbia River gorge tributaries
Hood River
A-H=0
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.028
N/A
N/A
1.102
N/A
N/A
N/A
N/A
N/A
N/A
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.898
N/A
N/A
0.874
N/A
N/A
N/A
N/A
N/A
N/A
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.177
N/A
N/A
1.172
N/A
N/A
N/A
N/A
N/A
N/A
DRAFT
587
-------
DRAFT
Coho Salmon (continued)
ESU
Southern Oregon and
Northern California Coast
Oregon Coast
Population
Southern Oregon and Northern California
Coast
Necanicum
Nehalem
Tillamook
Nestucca
Siletz
Yaquima
Alsea
Siuslaw
Umpqua
Coos
Coquille
A-H=0
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Sockeye Salmon
ESU
Ozette Lake
Snake River
Population
Ozette Lake
Snake River
A-H=0
N/A
N/A
95% Cl -lower
N/A
N/A
95% Cl - upper
N/A
N/A
DRAFT
588
-------
DRAFT
Steelhead
DPS
Central California Coast
(Goodetal., 2005)
California Central Valley
(Goodetal., 2005)
Lower Columbia River
(Goodetal., 2005)
Population
Russain River
Lagunitas
San Gregorio
Waddell Creek
Scott Creek
San Vincente Creek
San Lorenzo River
Sequel Creek
Aptos Creek
Sacramento River
Cispus River
Tilton River
Upper Cowlitz River
Lower Cowlitz River
Coweeman River
South Fork Toutle River
North Fork Toutle River
Kalama River - winter run
Kalama River - summer run
North Fork Lewis River - winter run
North Fork Lewis River - summer run
East Fork Lewis River - winter run
East Fork Lewis River - summer run
Salmon Creek
Washougal River - winter run
Washougal River - summer run
Clackamas River
Sandy River
Lower Columbia gorge tributaries
Upper Columbia gorge tributaries
A-H=0
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.950
N/A
N/A
N/A
N/A
0.908
0.938
1.062
1.010
0.981
N/A
N/A
N/A
N/A
N/A
N/A
1.003
0.971
0.945
N/A
N/A
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.900
N/A
N/A
N/A
N/A
0.792
0.830
0.915
9.130
0.889
N/A
N/A
N/A
N/A
N/A
N/A
0.884
0.901
0.850
N/A
N/A
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.020
N/A
N/A
N/A
N/A
1.041
1.059
1.233
1.117
1.083
N/A
N/A
N/A
N/A
N/A
N/A
1.138
1.047
1.051
N/A
N/A
DRAFT
589
-------
DRAFT
Steelhead (continued)
DPS
Middle Columbia River
(Goodetal., 2005)
Northern California (Good et
al., 2005)
Puget Sound*
Snake River (Good et al.,
2005)
South-Central California
Coast
Southern California
Population
Klickitat River
Yakima River
Fifteenmile Creek
Deschutes River
John Day - upper main stream
John Day - lower main stream
John Day - upper north fork
John Day - lower north fork
John Day - middle fork
John Day - south fork
Umatilla River
Touchet River
Redwood Creek
Mad River- winter run
Eel River -summer run
Mattole River
Ten Mile river
Noyo River
Big River
Navarro River
Garcia River
Gualala River
Other Humboldt County streams
Other Mendocino County streams
Puget Sound
Tucannon River
Lower Granite run
Snake A run
Snake B run
Asotin Creek
Upper Grande Ronde River
Joseph Creek
Imnaha River
Camp Creek
South-Central California Coast
Santa Ynez River
Ventura River
Matilija River
Creek River
Santa Clara River
A-H=0
N/A
1.009
0.981
1.022
0.975
0.981
1.011
1.013
0.966
0.967
1.007
0.961
N/A
1.000
0.980
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.886
0.994
0.998
0.927
N/A
0.967
1.069
1.045
1.077
N/A
N/A
N/A
N/A
N/A
N/A
95% Cl -lower
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
0.930
0.930
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
95% Cl - upper
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
1.050
1.040
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
DRAFT
590
-------
DRAFT
Steelhead (continued)
DPS
Upper Columbia River
(Goodetal., 2005)
Upper Williamette River
(McElhanyetal.,2007)
Population
Wenatchee / Entiat Rivers
Methow / Okanogan Rivers
Molalla River
North Santiam River
South Santiam River
Calapooia River
A-H=0
1.067
1.086
0.988
0.983
0.976
1.023
95% Cl -lower
N/A
N/A
0.790
0.789
0.855
0.743
95% Cl - upper
N/A
N/A
1.235
1.231
1.114
1.409
DRAFT
591
-------
DRAFT
IX
7-DADMax
ACA
AChE
a.i.
APEs
APHIS
BE
BEAD
BLM
BMP
BOR
BOR
BPA
BRT
BY
CAISMP
CALFED
CBFWA
CBI
CC
CCC
CCV
CDPR
CHART
CIDMP
CFR
cfs
CDFG
Corps
' • :
7-day average of the daily maximum
Alternative Conservation Agreement
acetylcholinesterase
active ingredient
alkylphenol ethoxylates
U.S. Department of Agriculture Animal Plant and Health Inspection Service
Biological Evaluation
Biological and Economic Analysis Divsion
Bureau of Land Management
Best Management Practices
Bureau of Reclaimation
Bureau of Reclamation
Bonneville Power Administration
Biological Review Team (NOAA Fisheries)
Brood Years
California Aquatic Invasive Species Management Plan
CALFED Bay-Delta Program (California Resource Agency)
Columbia Basin Fish and Wildlife Authority
Confidential Business Information
California Coastal
Central California Coast
Central California Valley
California Department of Pesticide Regulation
Critical Habitat Assessment Review Team
Comprehensive Irrigation District Management Plan
Code of Federal Regulations
cubic feet per second
California Department of Fish and Game
U.S. Department of the Army Corps of Engineers
DRAFT
592
-------
DRAFT
CSOs combined sewer/stormwater overflows
CSWP California State Water Project
CURES Coalition for Urban/Rural Environmental Stewardship
CVP Central Valley Projects
CVRWQCB Central Valley Regional Water Quality Control Board
CWA Clean Water Act
d day
DCI Date Call-Ins
DDD Dichloro Diphenyl Dichloroethane
DDE Diphenyl Dichlorethylene
DDT Dichloro Diphenyl Trichloroethane
DER Data Evaluation Review
DEQ Oregon Department of Environmental Quality
DIP Demographically Independent Population
DOE Washington State Department of Ecology
DPS Distinct Population Segment
EC Emulsifiable Concentrate Pesticide Formulation
ECso Median Effect Concentration
EEC Estimated Environmental Concentration
EFED Environmental Fate and Effects Division
ELM Environmental Information Management
EPA U.S. Environmental Protection Agency
ESPP Endangered Species Protection Program
ESA Endangered Species Act
ESU Evolutionarily Significant Unit
EU European Union
EXAMS Tier II Surface Water Computer Model
FERC Federal Energy Regulatory Commission
FCRPS Federal Columbia River Power System
FFDCA Federal Food and Drug Cosmetic Act
FIFRA Federal Insecticide, Fungicide, and Rodenticide Act
DRAFT
593
-------
DRAFT
FQPA Food Quality Protection Act
ft feet
GENEEC Generic Estimated Exposure Concentration
h hour
HCP Habitat Conservation Plan
HSRG Hatchery Scientific Review Group
HUC Hydrological Unit Code
IBI Indices of Biological Integrity
ICTRT Interior Columbia Technical Recovery Team
ILWP Irrigated Lands Waiver Program
IPCC Intergovernmental Panel on Climate Change
IRED Interim Re-registration Decision
LCFRB Lower Columbia Fish Recovery Board
ISG Independent Science Group
ITS Incidental Take Statement
km kilometer
Lbs Pounds
LCso Median Lethal Concentration.
LCR Lower Columbia River
LOAEC Lowest Observed Adverse Effect Concentration.
LOEL Lowest Observed Adverse Effect level
LOG Level of Concern
LOEC Lowest Observed Effect Concentration
LOQ Limit of Quantification
LWD Large Woody Debris
m meter
MCR Middle Columbia River
mg/L milligrams per liter
MO A Memorandum of Agreement
MPG Major Population Group
MRID Master Record Identification Number
DRAFT
594
-------
DRAFT
MTBE Methyl tert-butyl ether
NASA National Aeronautics and Space Administration
NAWQA U.S. Geological Survey National Water-Quality Assessment
NC Northern California
NEPA National Environmental Protection Agency
NLCD Natural Land Cover Data
NP Nonylphenol
NPDES National Pollutant Discharge Elimination System
NFS National Parks Services
NRCS Natural Resources Conservation Service
NWS National Weather Service
NEPA National Environmental Policy Act
NMA National Mining Association
NMC TV-methyl carbamates
NMFS National Marine Fisheries Service
NOAA National Oceanic and Atmospheric Administration
NOAEC No Observed Adverse Effect Concentration
NPDES National Pollution Discharge Eliminating System
NPIRS National Pesticide Information Retrieval System
NRC National Research Council
OC Oregon Coast
ODFW Oregon Division of Fish and Wildlife
OP Organophosphates
Opinion Biological Opinion
OPP EPA Office of Pesticide Program
PAH polyaromatic hydrocarbons
PBDEs polybrominated diphenyl ethers
PCBs polychlorinated biphenyls
PCEs primary constituent elements
POP Persistent Organic Pollutants
ppb Parts Per Billion
DRAFT
595
-------
DRAFT
PPE Personal Protection Equipment
PSP Pesticide Stewardship Partnerships
PSAMP Puget Sound Assessment and Monitoring Program
PSAT Puget Sound Action Team
PRIA Pesticide Registration Improvement Act
PRZM Pesticide Root Zone Model
PUR Pesticide Use Reporting
QA/QC Quality Assurance/Quality Control
RED Re-registration Eligibility Decision
REI Restricted Entry Interval
RPA Reasonable and Prudent Alternatives
RPM reasonable and prudent measures
RQ Risk Quotient
SAP Scientific Advisory Panel
SAR smolt-to-adult return rate
SASSI Salmon and Steelhead Stock Inventory
SC Southern California
S-CCC South-Central California Coast
SONCC Southern Oregon Northern California Coast
SLN Special Local Need (Registrations under Section 24(c) of FIFRA)
SR Snake River
TCE Trichloroethylene
TCP 3,5,6-trichloro-2-pyridinal
TGAI Technical Grade Active Ingredient
TIE Toxicity Identification Evaluation
TMDL Total Maximum Daily Load
TRT Technical Recovery Team
UCR Upper Columbia River
USFS United States Forest Service
USC United States Code
USFWS United States Fish and Wildlife Service
DRAFT
596
-------
DRAFT
USGS United States Geological Survey
UWR Upper Willamette River
VOC Volatile Organic Compounds
VSP Viable Salmonid Population
WDFW Washington Department of Fish and Wildlife
WLCRTRT Willamette/Lower Columbia River Technical Review Team
WQS Water Quality Standards
WWTIT Western Washington Treaty Indian Tribes
WWTP Wastewater Treatment Plant
DRAFT
597
-------
DRAFT
4;
303(d) waters Section 303 of the federal Clean Water Act requires states to prepare a list of all
surface waters in the state for which beneficial uses - such as drinking, recreation, aquatic
habitat, and industrial use - are impaired by pollutants. These are water quality limited estuaries,
lakes, and streams that do not meet the state's surface water quality standards and are not
expected to improve within the next two years. After water bodies are put on the 303(d) list they
enter into a Total Maximum Daily Load Clean Up Plan.
Active ingredient The component(s) that kills or otherwise affects the pest. Active
ingredients are always listed on the label (FIFRA 2(a)).
Adulticide
A compound that kills the adult life stage of the pest insect.
Anadromous Fish Species that are hatched in freshwater migrate to and mature in salt water
and return to freshwater to spawn.
Adjuvant
A compound that aides the operation or improves the effectiveness of a
pesticide.
Alevin
Life history stage of a salmonid immediately after hatching and before the
yolk-sac is absorbed. Alevins usually remain buried in the gravel in or
near the egg nest (redd) until their yolk sac is absorbed when they swim up
and enter the water column.
Anadromy
The life history pattern that features egg incubation and early juvenile
development in freshwater migration to sea water for adult development,
and a return to freshwater for spawning.
Assessment Endpoint Explicit expression of the actual ecological value that is to be protected
(e.g., growth of juvenile salmonids).
DRAFT 598
-------
DRAFT
Bioaccumulation
Accumulation through the food chain (i.e.., consumption of food,
water/sediment) or direct water and/or sediment exposure.
Bioconcentration
Uptake of a chemical across membranes, generally used in reference to
waterborne exposures.
Biomagnification
Transfer of chemicals via the food chain through two or more trophic
levels as a result of bioconcentration and bioaccumulation.
Degradates
New compounds formed by the transformation of a pesticide by chemical
or biological reactions.
Distinct Population A listable entity under the ESA that meets tests of discreteness and
Segment significance according to USFWS and NMFS policy. A population is
considered distinct (and hence a "species" for purposes of conservation
under the ESA) if it is discrete fro an significant to the remainder of its
species based n factors such as physical, behavioral, or genetic
characteristics, it occupies an unusual or unique ecological setting, or its
loss would represent a significant gap in the species' range.
Escapement
The number offish that survive to reach the spawning grounds or
hatcheries. The escapement plus the number offish removed by harvest
form the total run size.
Evolutionarily
Significant Unit
A group of Pacific salmon or steelhead trout that is (1)
substantially reproductively isolated from other conspecific
units and (2) represent an important component of the evolutionary legacy
of the species.
Fall Chinook
This salmon stock returns from the ocean in late summer and early
DRAFT
599
-------
Salmon
DRAFT
fall to head upriver to its spawning grounds, distinguishing it from other
stocks which migrate in different seasons.
Fate
Dispersal of a material in various environmental compartments (sediment,
water air, biota) as a result of transport, transformation, and degradation.
Flowable
A pesticide formulation that can be mixed with water to form a suspension
in a spray tank.
Fry
Stage in salmonid life history when the juvenile has absorbed its yolk sac
and leaves the gravel of the redd to swim up into the water column. The
fry stage follows the alevin stage and in most salmonid species is followed
by the parr, fmgerling, and smolt stages. However, chum salmon
juveniles share characteristics of both the fry and smolt stages and can
enter sea water almost immediately after becoming fry.
Half-pounder
Hatchery
A life history trait of steelhead exhibited in the Rogue, Klamath, Mad, and
Eel Rivers of southern Oregon and northern California. Following
smoltification, half-pounders spend only 2-4 months in the ocean, then
return to fresh water. They overwinter in fresh water and emigrate to salt
water again the following spring. This is often termed a false spawning
migration, as few half-pounders are sexually mature.
Salmon hatcheries use artificial procedures to spawn adults and raise the
resulting progeny in fresh water for release into the natural environment,
either directly from the hatchery or by transfer into another area. In some
cases, fertilized eggs are outplanted (usually in "hatch-boxes"), but it is
more common to release fry or smolts.
Inert ingredients
"an ingredient which is not active" (FIFRA 2(m)). It may be toxic or
enhance the toxicity of the active ingredient.
DRAFT
600
-------
DRAFT
Iteroparous
Capable of spawning more than once before death
Jacks
Male salmon that return from the ocean to spawn one or more years before
full-sized adults return. For coho salmon in California, Oregon,
Washington, and southern British Columbia, jacks are 2 years old, having
spent only 6 months in the ocean, in contrast to adults, which are 3 years
old after spending 1 1A years in the ocean.
Jills
Female salmon that return from the ocean to spawn one or more years
before full-sized adult returns. For sockeye salmon in Oregon,
Washington, and southern British Columbia, Jills are 3 years old (age 1.1),
having spent only one winter in the ocean in contrast to more typical
sockeye salmon that are age 1.2, 1.32.2, or 2.3 on return.
Kokanee
The self-perpetuating, non-anadromous form of O. nerka that occurs in
balanced sex ration populations and whose parents, for several generations
back, have spent their whole lives in freshwater.
Lambda
Also known as Population growth rate, or the rate at which the abundance
offish in a population increases or decreases.
Major Population
Group (MPG)
A group of salmonid populations that are geographically and
genetically cohesive. The MPG is a level of organization between
demographically independent populations and the ESU.
Main channel
The stream channel that includes the thalweg (longitudinal continuous
deepest portion of the channel.
Metabolite
A transformation product resulting from metabolism.
DRAFT
601
-------
Mode of Action
DRAFT
A series of key processes that begins with the interaction of a pesticide
with a receptor site and proceeds through operational and anatomical
changes in an organisms that result in sublethal or lethal effects.
Natural fish
A fish that is produced by parents spawning in a stream or lake bed, as
opposed to a controlled environment such as a hatchery.
Nonylphenols
A type of APE and is an example of an adjuvant that may be present as an
ingredient of a formulated product or added to a tank mix prior to
application.
Off-channel habitat Water bodies and/or inundated areas that are connected (accessible to
salmonid juveniles) seasonally or annually to the main channel of a stream
including but not limited to features such as side channels, alcoves, ox
bows, ditches, and floodplains.
Parr
The stage in anadromous salmonid development between absorption of the
yolk sac and transformation to smolt before migration seaward.
Persistence
The tendency of a compound to remain in its original chemical form in the
environment.
Pesticide
Any substance or mixture of substances intended for preventing,
destroying, repelling or mitigating any pest.
Reasonable and Recommended alternative actins identified during formal
Prudent Alternative consultation that can be implemented in a manner consistent
(RPA) with the scope of the Federal agency's legal authority an jurisdiction, that
are economically an technologically feasible, an that the Services believes
would avoid the likelihood of jeopardizing the continued existence of the
DRAFT
602
-------
DRAFT
listed species or the destruction or adverse modification of designated
critical habitat.
Redd
A nest constructed by female salmonids in streambed gravels where eggs
are deposited and fertilization occurs.
Riparian area
Area with distinctive soils an vegetation between a stream or other body of
water and the adjacent upland. It includes wetlands and those portions of
flood plains an valley bottoms that support riparian vegetation.
Risk
The probability of harm from actual or predicted concentrations of a
chemical in the aquatic environment - a scientific judgement.
Salmonid
Fish of the family Salmonidae, including salmon, trout, chars, grayling,
and whitefish. In general usage, the term usually refers to salmon, trout,
and chars.
SASSI
A cooperative program by WDFW and WWTIT to inventory and evaluate
the status of Pacific salmonids in Washington State. The SASSI report is
a series of publications from this program.
Semelparous
The condition in an individual organism of reproducing only once in a
lifetime.
Smolt
A juvenile salmon or steelhead migrating to the ocean and undergoing
physiological changes to adapt from freshwater to a saltwater
environment.
Sublethal
Below the concentration that directly causes death. Exposure to sublethal
concentrations of a material may produce less obvious effect on behavior,
DRAFT
603
-------
DRAFT
biochemical, and/or physiological function of the organism often leading
to indirect death.
Surfactant A substance that reduces the interfacial or surface tension of a system or a
surface-active substance.
Synergism A phenomenon in which the toxicity of a mixture of chemicals is greater
than that which would be expected from a simple summation of the
toxicities of the individual chemicals present in the mixture.
Technical Grade Pure or almost pure active ingredient. Available to formulators.
Active Ingredient Most toxicology data are developed with the TGAI. The percent
(TGAI) AI is listed on all labels.
Technical Recovery Teams convened by NOAA Fisheries to develop technical products
Teams (TRT) related to recovery planning. TRTs are complemented by planning forums
unique to specific states, tribes, or reigns, which use TRT and other
technical products to identify recovery actions.
Teratogenic Effects produced during gestation that evidence themselves as altered
structural or functional processes in offspring.
Total Maximum defines how much of a pollutant a water body can tolerate (absorb)
Daily Load (TMDL) daily and remain compliant with applicable water quality standards. All
pollutant sources in the watershed combined, including non-point sources,
are limited to discharging no more than the TMDL.
Unique Mixture A specific combination of 2 or more compounds, regardless of the
presence of other compounds.
Viable Salmonid An independent population of Pacific salmon or steelhead trout
DRAFT 604
-------
Population
DRAFT
that has a negligible risk of extinction over a 100-year time frame.
Viability at the independent population scale is evaluated based on the
parameters of abundance, productivity, spatial structure, and diversity.
VSP Parameters
Abundance, productivity, spatial structure, and diversity. These describe
characteristics of salmonid populations that are useful in evaluating
population viability. See NOAA Technical Memorandum NMFS-
NWFSC-, "Viable salmonid populations and the recovery of
evolutionarily significant units," McElhany et al., June 2000.
WDFW
Washington Department of Fish and Wildlife is a co-manager of
salmonids and salmonid fisheries in Washington State with WWTIT and
other fisheries groups. The agency was formed in the early 1990s by the
combination of the Washington Department of Fisheries and the
Washington Department of Wildlife.
WWTIT
Western Washington Treaty Indian Tribes is an organization of Native
American tribes with treaty fishing rights recognized by the U.S.
government. WWTIT is a co-manager of salmonids and salmonid
fisheries in western Washington in cooperation with the WDFW and other
fisheries groups.
WQS
"A water quality standard defines the water quality goals of a waterbody,
or portion thereof, by designating the use or uses to be made of the water
and by setting criteria necessary to protect public health or welfare,
enhance the quality of water and serve the purposes of the Clean Water
Act." Each state is responsible for maintaining water quality standards.
DRAFT
605
------- |