Risks of Endosulfan Use to the Federally Threatened

     California Red-legged Frog (Rana aurora draytonii),
   Bay Checkerspot Butterfly (Euphydryas editha bayensis),
Valley Elderberry Longhorn Beetle (Desmocems caiifomicus
  dimorphus), and California Tiger Salamander (Ambystoma
                          californiense)

             And the Federally Endangered
 San Francisco Garter Snake (Thamnophis sirtalis tetrataenia),
       San Joaquin Kit Fox (vuipes macrotis mutica), and
   Salt Marsh Harvest Mouse (Reithrodontomys raviventris)
                Pesticide Effects Determinations

            Environmental Fate and Effects Division
                 Office of Pesticide Programs
                   Washington, D.C. 20460

                        June 18,2009
Primary Authors:
Ecological Effects
     Keith Sappington, M.S., Biologist, OPP/EFED
     Glen Thursby, Ph.D, Biologist, ORD/NHEERL/AED
     Sandy Raimondo, Ph.D., Ecologist, ORD/NHEERL/GED

Environmental Fate
     Mohammed Ruhman, Ph.D, Agronomist, OPP/EFED

Secondary Review:
     Mah Shamim, Ph.D., Branch Chief, OPP/EFED
     Environmental Risk Assessment Branch 5

-------
                                   Table of Contents
1            Executive Summary	11
2            Problem Formulation	25
  2.1 Purpose	25
  2.2 Scope	27
  2.3 Previous Assessments	29
  2.4 Stressor Source and Distribution	30
     2.4.1 Environmental Fate Properties	32
     2.4.2 Environmental Transport Mechanisms	36
     2.4.3 Mechanism of Action	36
     2.4.4 Use Characterization	37
  2.5 Assessed Species	48
  2.6 Designated Critical Habitat	53
  2.7 Action Area	55
  2.8 Assessment Endpoints and Measures of Ecological Effect	58
     2.8.1 Assessment Endpoints	58
     2.8.2 Assessment Endpoints for Designated Critical Habitat	62
  2.9 Conceptual Model	64
     2.9.1 Risk Hypotheses	64
     2.9.2 Diagram	65
  2.10 Analysis Plan	66
     2.10.1  Measures to Evaluate the Risk Hypothesis and Conceptual Model	66
       2.10.1.1 Measures  of Exposure	67
       2.10.1.2 Measures  of Effect	69
       2.10.1.3 Integration of Exposure and Effects	70
     2.10.2 Data Gaps	70
3            Exposure Assessment	71
  3.1 Label Application Rates, Intervals and Buffers	71
  3.2 Aquatic Exposure Assessment	73
     3.2.1 Modeling Approach	73
     3.2.2 Model Inputs	75
     3.2.3 Results	76
     3.2.4 Existing Monitoring Data	81
       3.2.4.1 Surface Water Monitoring Data	81
       3.2.4.2 Groundwater Monitoring Data	91
       3.2.4.3 Sediment Monitoring Data	91
       3.2 A A Atmospheric Monitoring Data	92
  3.3 Terrestrial Animal Exposure Assessment	99
  3.4 Terrestrial Plant Exposure Assessment	103
4            Effects Assessment	104
  4.1 Toxicity of Endosulfan to Aquatic Organisms	106
     4.1.1 Toxicity to Freshwater Fish and Aquatic-Phase Amphibians	108
       4.1.1.1 Freshwater Fish: Acute Exposure (Mortality) Studies -Registrant submitted ..108
       4.1.1.2 Freshwater Fish: Acute Exposure (Mortality) Studies—Open Literature	108
       4.1.1.3 Freshwater Fish: Chronic Exposure (Growth/Reproduction) Studies	108

-------
       4.1.1.4 Freshwater Fish: Sublethal Effects and Additional Open Literature Information
       	109
       4.1.1.5 Aquatic-phase Amphibian: Acute and Chronic Studies	110
       4.1.1.6 Freshwater Fish: Endosulfan Sulfate	110
     4.1.2 Toxicity to Freshwater Invertebrates	110
       4.1.2.1 Freshwater Invertebrates: Acute Exposure Studies	110
       4.1.2.2 Freshwater Invertebrates: Acute Exposure Studies—Open Literature	110
       4.1.2.3 Freshwater Invertebrates: Chronic Exposure Studies	Ill
       4.1.2.4 Freshwater Invertebrates: Sub-lethal Effects and Additional Open Literature
       Information	Ill
       4.1.2.5 Freshwater Invertebrates: Endosulfan Sulfate	Ill
       4.1.2.6 Freshwater Invertebrates: Sediment Toxicity	112
     4.1.3 Toxicity to Estuarine/Marine Fish	112
       4.1.3.1 Estuarine/Marine Fish: Acute Exposure (Mortality) Studies	112
       4.1.3.2 Estuarine/Marine Fish: Acute Exposure (Mortality) Studies—Open Literature 112
       4.1.3.3 Estuarine/Marine Fish: Chronic Exposure (Growth/Reproduction) Studies	112
     4.1.4 Toxicity to Estuarine/Marine Invertebrates	113
       4.1.4.1 Estuarine/Marine Invertebrates: Acute Exposure (Mortality) Studies	113
       4.1.4.2 Estuarine/Marine Invertebrates: Acute Exposure (Mortality) Studies—Open
       Literature	114
       4.1.4.3 Estuarine/Marine Invertebrates: Chronic Exposure (Growth/Reproduction)
       Studies	114
       4.1.4.4 Estuarine/Marine Invertebrates: Open Literature Data	115
       4.1.4.5 Estuarine/Marine Invertebrates: Endosulfan Sulfate	115
     4.1.5 Toxicity to Aquatic-phase Amphibians—Open Literature	115
     4.1.6 Toxicity to Aquatic Plants	117
     4.1.7 Freshwater Field/Mesocosm Studies	118
  4.2 Toxicity of Endosulfan to Terrestrial Organisms	119
     4.2.1 Toxicity to Birds, Reptiles, and Terrestrial-Phase Amphibians	120
       4.2.1.1 Birds: Acute Exposure (Mortality) Studies	120
       4.2.1.2 Birds: Chronic Exposure (Growth, Reproduction) Studies	120
       4.2.1.3 Birds: Open literature Studies	121
       4.2.1.4 Birds: Endosulfan Sulfate	121
     4.2.2 Toxicity to Mammals	121
       4.2.2.1 Mammals: Acute Exposure (Mortality) Studies	121
       4.2.2.2 Mammals: Chronic Exposure (Growth, Reproduction) Studies	122
       4.2.2.3 Mammals: Endosulfan Sulfate	122
     4.2.3 Toxicity to Terrestrial Invertebrates	122
       4.2.3.1 Terrestrial Invertebrates: Acute Exposure (Mortality) Studies	123
       4.2.3.2 Terrestrial Invertebrates: Open Literature Studies	123
     4.2.4 Toxicity to Terrestrial Plants	124
  4.3 Use of Probit Slope Response Relationship to Provide Information on the Endangered
  Species Levels of Concern	124
  4.4 Ecological Incident Summary	125
5             Risk Characterization	127
  5.1 Risk Estimation	127

-------
  5.1.1 Exposures in the Aquatic Habitat	128
     5.1.1.1 Freshwater Fish and Aquatic-phase Amphibians	128
     5.1.1.2 Freshwater Invertebrates	129
     5.1.1.3 Non-vascular Aquatic Plants	131
     5.1.1.4 Aquatic Vascular Plants	131
  5.1.2 Exposures in the Terrestrial Habitat	131
     5.1.2.1 Birds (surrogate for Reptiles and Terrestrial-phase amphibians)	131
     5.1.2.2 Mammals	135
     5.1.2.3 Terrestrial Invertebrates	138
     5.1.2.4 Terrestrial Plants	140
  5.1.3 Primary Constituent Elements of Designated Critical Habitat	140
  5.1.4 Spatial Extent of Potential Effects	140
     5.1.4.1 Spray Drift	140
     5.1.4.2 Downstream Dilution Analysis	141
     5.1.4.3 Overlap between CRLF, CTS, SFGS, BCB, VELB, SMHM, and SJKF habitat
     and  Spatial Extent of Potential Effects	141
5.2 Risk Description	143
  5.2.1 California Red-Legged Frog	163
     5.2.1.1 Direct Effects	163
     5.2.1.2 Indirect Effects to CRLF (via Potential Loss of Prey)	178
     5.2.1.3 Indirect Effects (via Habitat Effects)	185
     5.2.1.4 Effects to Designated Critical Habitat	186
  5.2.2 California tiger salamander	187
  5.2.2.1 Direct Effects	187
     5.2.2.2 Indirect Effects (via Reductions in Prey Base)	191
     5.2.2.3 Modification to Designated Critical Habitat	193
  5.2.3 San Francisco Garter Snake	194
     5.2.3.1 Direct Effects	194
     5.2.3.2 Indirect Effects (via Reductions in Prey Base)	196
     5.2.3.3 Indirect Effects (via Habitat Effects)	198
  5.2.4 Salt Marsh Harvest Mouse	199
     5.2.4.1 Direct Effects	199
     5.2A.2 Indirect Effects (via Reductions in Prey Base)	202
     5.2.4.3 Indirect Effects (via Habitat Effects)	203
  5.2.5 San Joaquin Kit Fox	204
     5.2.5.1 Direct Effects	204
     5.2.5.2 Indirect Effects (via Reductions in Prey Base)	206
     5.2.5.3 Indirect Effects (via Habitat Effects)	208
  5.2.6 Bay Checkerspot Butterfly	208
     5.2.6.1 Direct Effects	208
     5.2.6.2 Indirect Effects (via Reduction in Prey Base & Habitat Effects)	210
     5.2.6.3 Modification to Designated Critical Habitat	210
  5.2.7 Valley Elderberry Longhorn Beetle	211
     5.2.7.1 Direct Effects	211
     5.2.7.2 Indirect Effects (via Reduction in Prey Base & Habitat Effects)	212
     5.2.7.3 Modification to Designated Critical Habitat	213

-------
6            Uncertainties	214
  6.1 Exposure Assessment Uncertainties	214
     6.1.1 Maximum Use Scenario	214
     6.1.2 Aquatic Exposure Modeling of Endosulfan	214
       6.1.2.1 Potential Groundwater Contributions to Surface Water Chemical Concentrations
       	216
     6.1.3 Multiple Growing Seasons per Year	217
     6.1.4 Usage Uncertainties	217
     6.1.5 Terrestrial Exposure Modeling of Endosulfan	217
     6.1.6 Spray Drift Modeling	219
  6.2 Effects Assessment Uncertainties	220
     6.2.1 Age Class and Sensitivity of Effects Thresholds	220
     6.2.2 Use of Surrogate Species Effects Data	220
     6.2.3 Sublethal Effects	221
     6.2.4 Location of Wildlife Species	221
     6.2.5 Endocrine Disruption	221
7            Summary and Conclusions	222
8            References	232

-------
Appendices

   Appendix A      Endosulfan Label Analysis and Memorandum
   Appendix B      AgDrift Modeling
   Appendix C      Aquatic Exposure Modeling Approach for Endosulfan
   Appendix D      RQ Method and LOCs
   Appendix E      GIS Maps
   Appendix F      T-REX Example Output
   Appendix G      Bibliography of ECOTOX Open Literature Not Evaluated
   Appendix H      Accepted ECOTOX Data Table (sorted by effect) and Bibliography
   Appendix I      KABAM Model Description and Output
   Appendix J      Aquatic ECOTOX Data
   Appendix K      Terrestrial ECOTOX Data
   Appendix L      Endosulfan Incidents
   Appendix M     HED Effects Table
   Appendix N      Earthworm Fugacity Modeling
                                   Attachments

Attachment 1: Status and Life History for the CRLF
Attachment 2: Baseline Status and Cumulative Effects for the CRLF
Attachment 3: Status and Life History for the San Francisco Bay Species
Attachment 4: Baseline Status and Cumulative Effects for the San Francisco Bay Species
                                   List of Tables

Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS,
         SMHM, SJKF, BCB, AND VELB                                             15
Table 1.2 Effects Determination Summary for the Critical Habitat Impact Analysis           22
Table 2.1 Physiochemical properties of the multi-chemical stressor                         32
Table 2.2 Summary of fate and transport properties for endosulfan isomers and sulfate        33
Table 2.3 Endosulfan labeled uses assessed for California (No single application rate over 1.5 Ibs
         except for orchard crops and strawberries)                                      37
Table 2.4 California labeled non-crop use for endosulfan                                  40
Table 2.5 Top use patterns for endosulfan for selected years from 1990 to 2002              41
Table 2.6 Historical usage of endosulfan for 1994 (1st period) and 1995-1997 (2nd period)     43
Table:
2.7 Historical usage of endosulfan for 1998-2004 (3rd period) and 2005-2006 (4th period)
                                                                                   44
Table 2.8 Summary of Current Distribution, Habitat Requirements, and Life History Information
         for the Assessed Listed Species                                               49
Table 2.9 Designated Critical Habitat PCEs for the CRLF, CTS, BCB and VELB             53
Table 2.10 Taxa Used in the Analyses of Direct and Indirect Effects of Endosulfan for the
         Assessed Listed Species.                                                     60
Table 2.11 Taxa and Assessment Endpoints Used to Evaluate the Potential for the Use of
         Endosulfan  to Result in Direct and Indirect Effects to the Assessed Listed Species.  61

-------
Table 2.12 Summary of Assessment Endpoints and Measures of Ecological Effect for Primary
         Constituent Elements of Designated Critical Habitat for CRLF, BCB, VELB, and CIS.
                                                                                     62
Table 3.1 Summary of labeled endosulfan uses, scenarios, and application information        71
Table 3.2 Calculation of application rates for a- and p-endosulfan and endosulfan sulfate) for use
         in modeling (rates are for almond)                                              74
Table 3.3 Summary of PRZM/EZAMS environmental fate and transport data of a- and P-
         endosulfan and endosulfan sulfate used as aquatic exposure inputs for this endangered
         species assessment                                                            75
Table 3.4 Aquatic EECs (ug/L) for endosulfan used in California (a- and P-endosulfan)1       76
Table 3.5 Aquatic EECs (ug/L) for endosulfan used in California (endosulfan sulfate and total
         stressor)                                                                     77
Table 3.6 Sediment pore water EECs (ug/L) for endosulfan used in California (a- and P-
         endosulfan)                                                                  78
Table 3.7 Sediment pore water EECs (ug/L) for endosulfan used in California (endosulfan
         sulfate and total stressor)                                                      78
Table 3.8 Sediment EECs (ug/L) for endosulfan used in California (a- and P-endosulfan)      79
Table 3.9 Sediment EECs (ug/L) for endosulfan used in California (endosulfan sulfate and total
         stressor)                                                                     79
Table 3.10 Total soil and pore water concentrations for selected crop scenarios               81
Table 3.11 Summary of reported detection limits for the CDPR surface water data            83
Table 3.12 Summary of CPDR surface water monitoring data for California Counties collected
         for the first monitoring period (1991 to 1996)                                    83
Table 3.13 Summary of CPDR surface water monitoring data for California (the 2nd monitoring
         period)                                                                      85
Table 3.14 Summary of STORET surface water monitoring data for California               87
Table 3.15 A summary of NAWQA monitoring data for surface water from California        89
Table 3.16 Surface water data summary                                                  90
Table 3.17 A summary of monitoring data for ground water from California                 91
Table 3.18 A summary for NAWQA monitoring data to base sediment in California           91
Table 3.19 Summary of bottom sediment monitoring data for California                     92
Table 3.20 Frequency and concentrations of endosulfan species in ambient air                94
Table 3.21 Frequency and concentrations of endosulfan species in ambient air                96
Table 3.22 Concentration of pesticides (ng/L) in surface water samples collected at the
         Tablelands (Sequoia National Park), and the Sixty Lake Basin (Kings Canyon National
         Park) California, USA (Fellers et al. 2004).                                      98
Table 3.23 Summary of Dose and Dietary-based EECs Used for Estimating Dietary Risks to
         Terrestrial Organisms using T-REX ver. 1.4.1.                                 100
Table 4.1 Aquatic Toxicity Profile for Endosulfan (TGAI)                               107
Table 4.2 Categories of Acute Toxicity for Fish and Aquatic Invertebrates                 108
Table 4.3 Terrestrial Toxicity Profile for Endosulfan                                    119
Table 4.4. Categories of Acute Toxicity for Avian and Mammalian Studies                 119
Table 4.5 Comparison of Acute Toxicity of Endosulfan and Endosulfan Sulfate To Birds.    120
Table 5.1 Acute and Chronic RQs for freshwater fish based on EECs for use categories used to
         represent all endosulfan uses in CA.                                          128

-------
Table 5.2 Acute and Chronic RQs for freshwater invertebrates based on EECs for use categories
         used to represent all endosulfan uses in CA.                                    130
Table 5.3 Summary of the acute and chronic dose- and dietary-based RQs for birds (20 g)
         estimated based on the maximum endosulfan foliar spray applications using T-REX
         version 1.4.1.                                                               132
Table 5.4 Summary of the acute and chronic dose- and dietary-based RQs for mammals
         estimated based on the maximum endosulfan foliar spray applications using T-REX
         version 1.4.1.                                                               136
Table 5.5 Summary of the acute and chronic dose- and dietary-based RQs terrestrial insects
         estimated based on the maximum endosulfan foliar spray applications using T-REX
         version 1.4.1.                                                               139
Table 5.6 Summary of the highest RQ to LOG ratios for the two GIS use categories for
         endosulfan                                                                 141
Table 5.7 Summary of average endosulfan usage data and occurrence of the CRLF and the San
         Francisco Bay species at the county level                                      141
Table 5.8 Extent and location of overlap between areas  of species occurrence and areas of
         potential endosulfan usage                                                   142
Table 5.9 Risk Estimation Summary for endosulfan - Direct and Indirect Effects           144
Table 5.10  Risk Estimation  Summary for Endosulfan - Effects to Designated Critical Habitat
         (PCEs)                                                                    145
Table 5.11 Summary of the chance of individual acute effects to freshwater animals based on
         acute RQs, the acute listed species LOG, acute toxicity data, and probit slope response
         relationships(U).                                                            148
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on
         acute RQs, the acute listed species LOG, acute toxicity data, and probit slope response
         relationships.                                                               150
Table 5.13 Summary of the chance of individual acute effects to herpetofauna based on acute
         RQs, the acute listed species LOG, acute toxicity data, and probit slope response
         relationships.                                                               158
Table 5.14 Comparison of maximum detected concentrations of total endosulfan in surface
         waters of California to Agency LOG for freshwater fish                         166
Table 5.15 Assumed dietary preferences of small (1.4g), medium  (37g) and large (238g) CRLF
         feeding on aquatic biota of the model ecosystem.                               169
Table 5.16 Endosulfan chemical characteristics used as  input to KABAM                  169
Table 5.17 Estimated concentrations of endosulfan in ecosystem components for two crop
         exposure scenarios                                                         170
Table 5.18 Calculation of EECs for CRLF consuming aquatic prey contaminated by endosulfan
                                                                                   171
Table 5.19 Calculation of RQ values for CRLF consuming aquatic prey contaminated by
         endosulfan.                                                                171
Table 5.20 Summary of the acute and chronic dose- and dietary-based RQs for herpetofauna
         estimated based on the maximum endosulfan foliar spray applications using T-HERPS
         version 1.0.                                                                174
Table 5.21 Acute and Chronic RQs for freshwater sediment invertebrates based on EECs for use
         categories used to represent all endosulfan uses in CA                           182

-------
Table 5.22 Estimated concentrations of total endosulfan in earthworms and avian risk quotients
         for two crop exposure scenarios.                                              190
Table 5.23 Estimated concentrations of total endosulfan in earthworms and mammalian risk
         quotients for two crop exposure scenarios                                      201
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS,
         SMHM, SJKF, BCB, AND VELB                                            222
Table 7.2 Effects Determination Summary for the Critical Habitat Impact Analysis          229

                                    List of Figures

Figure 2.1 Chemical structures of endosulfan and related compounds	31
Figure 2.2 Endosulfan Use in Total Pounds per County (note vlues are different in chosen
         categories)	42
Figure 2.3 Historical usage data for endosulfan	43
Figure 2.4 Endosulfan usage and acreage for the top five crop use patterns (CA-PUR data;
         Average for 2005-2006)	47
Figure 2.5 Distribution of endosulfan usage for the top usage counties of Fresno, Kings, and
         Imperial	48
Figure 2.6 Initial area of concern, or "footprint" of potential use, for endosulfan	57
Figure 2.7 Conceptual Model for Endosulfan Effects on the Assessed Species and their Critical
         Habitat in the Aquatic Environment	65
Figure 2.8. Conceptual Model for Endosulfan Effects on the Assessed Species and their Critical
         Habitat in the Terrestrial Environment	66
Figure 3.1 Detected concentrations of alpha and beta-endosulfan and common degradate in
         California surface waters from 1991 to 2006 (No data were collected from 1996 to
         2000)	82
Figure 3.2 Detected concentrations of the total endosulfan species in California surface waters
         from 1991 to 2006	82
Figure 3.3 CPDR Monitoring data for Imperial County (11 sites, varied  times of the year)	84
Figure 3.4 CDPR Monitoring data for Merced County (4 sites), Stanislaus County (6 sites), San
         Joaquin County (2 sites) and Sacramento County (1 site) at varied times of the year.. 85
Figure 3.5 Detected concentrations of a- and p-endosulfan and endosulfan sulfate common
         degradate in California surface waters from 2001 to 2008 (STORET data)	88
Figure 3.6 Detected concentrations of the total endosulfan species in California surface waters
         from 2001 to 2008 (STORET data)	88
Figure 3.7 Detected concentrations of a- and P-endosulfan and endosulfan sulfate common
         degradate in surface waters of five California Counties (Counties with detections over
         theLOQ; 2001 to 2008 STORET data)	89
Figure 3.8 Concentration profiles for a- and P- endosulfan at the edge and near-field in ambient
         air following application in San Joaquin County, California	94
Figure 3.9 Concentration profiles for a- endosulfan in ambient air during a period of five weeks
         at 4 sites in  Fresno County, California	97
Figure 5.1 Temporal distribution of endosulfan application in California for 2006 (Source:
         CDPR PUR))	164
Figure 5.2 CRLF Reproductive Events by Month; Adults and juveniles can be present all year.
         	164

-------
Figure 5.3 Species sensitivity distribution of the acute toxicity of endosulfan to freshwater fish
         	167
Figure 5.4 Species sensitivity distribution of the acute toxicity of endosulfan to freshwater
         invertebrates via water column exposure	180
Figure 5.5 Species sensitivity distribution of the acute contact toxicity of endosulfan TGAI to
         terrestrial invertebrates based on species mean values	183
Figure 5.6 CTS Reproductive Events by Month; Adults and juveniles can be present all year. 188
Figure 5.7 Average Endosulfan Use (2005-2006) in Counties with Reported Occurrence of
         Listed Species, Occupied Core Areas and/or Critical Habitat	189
Figure 5.8 SFGS Reproductive Events by Month; Adults can be present all year	195
Figure 5.9 General Annual Life-History Parameters for the BCB	209
                                           10

-------
  1   Executive Summary
The purpose of this assessment is to evaluate potential direct and indirect effects on the
following species arising from FIFRA regulatory actions regarding use of endosulfan on
agricultural sites:
       California red-legged frog (CRLF), Rana aurora draytonii
       California Tiger Salamander (CTS), Ambystoma californiense,
       San Francisco Garter Snake (SFGS), Thamnophis sirtalis tetrataenia,
       Salt Marsh Harvest Mouse (SMHM), Reithrodontomys raviventris,
       San Joaquin Kit Fox (SJKF),  Vulpes macrotis mutica,
       Bay Checkerspot Butterfly (BCB), Euphydryas editha bayensis, and
       Valley Elderberry Longhorn Beetle  (VELB), Desmocerus californicus dimorphus.

In addition, this assessment evaluates whether these actions can be expected to result in effects to
designated critical habitat for the CRLF, CTS, BCB, and VELB; noting that critical habitat has
not been designated for the SFGS, SMBTM and SJKF. This assessment was completed in
accordance with the U.S. Fish and Wildlife Service (USFWS) and National Marine Fisheries
Service (NMFS) Endangered Species Consultation Handbook (USFWS/NMFS, 1998 and
procedures  outlined in the Agency's Overview Document (U.S. EPA, 2004).

The listing date and a general description of the range of each assessed species are as follows.

   •   The CRLF was listed as a threatened species by USFWS in 1996. The species is endemic
       to California and Baja California (Mexico) and inhabits both coastal and interior
       mountain ranges. A total of 243 streams or drainages are believed to be currently
       occupied by the species, with the greatest numbers in Monterey, San Luis Obispo, and
       Santa Barbara counties (USFWS,  1996) in California.
   •   The CTS was listed as threatened by the Fish and Wildlife  Service on August 4, 2004 and
       designated critical habitat was established on November 24, 2004. The two distinct
       population segments (one in Santa Barbara County and the other in Sonoma County)
       inhabit freshwater pools or ponds, grassland or oak savannah communities in low foothill
       regions and small mammal burrows.
   •   The SFGS was listed as an endangered  species by the USFWS in 1967 and was
       grandfathered under the Endangered Species Act (ESA) when it was signed into law in
       1973. The SFGS is endemic to the San Francisco Peninsula and San Mateo County and
       historically inhabited densely vegetated ponds or shallow marshlands near open hillsides
       found from San Francisco to Santa Cruz, including the San Francisco Peninsula. The
       current distribution of the SFGS is unknown because most of their historic range is now
       privately owned; however, it appears that the SFGS can still be found in much of its
       historic range.
   •   The SMHM was listed as an endangered species by the USFWS in 1970. The SMHM is
       currently found in tidal and non-tidal salt marshes in San Francisco, San Pablo, and
       Suisun bays.
   •   The SJKF was listed as endangered  by the USFWS on March 11, 1967. Its current range
       includes Alameda, Contra Costa, Fresno, Kern, Kings, Madera, Merced, Monterey, San
                                          11

-------
       Benito, San Joaquin, San Luis Obispo, Santa Barbara, Santa Clara, Stanislaus, Tulare and
       Ventura counties in California. It inhabits a variety of habitats, including grasslands,
       scrublands, vernal pool areas, oak woodland, alkali meadows and playas, and an
       agricultural matrix of row crops, irrigated pastures, orchards, vineyards, and grazed
       annual grasslands.
    •   The BCB was listed as a threatened species in 1987 by the USFWS. The species
       primarily inhabits native grasslands on serpentine outcrops around the San Francisco Bay
       Area in California. The distribution of known BCB populations is currently limited to
       Santa Clara and San Mateo Counties.
    •   The VELB was listed as threatened on August 8, 1980 by the USFWS. The species
       inhabits the Central Valley of California (from Shasta County to Fresno County in the
       San Joaquin Valley). The VELB is completely dependent on its host plant, elderberry
       (Sambucus species), which is a common component of the remaining riparian forests and
       adjacent upland habitats of California's Central Valley.

Endosulfan is a dioxathiepin broad spectrum contact insecticide and acaricide that is broadly
classified as a chlorinated hydrocarbon.  The chemical is used on major agricultural crops to
control insect pests such as aphids, fruitworms, beetles, leafhoppers, moth larvae, and whiteflies.
Labeled uses of endosulfan include a wide variety of vegetables, fruits, nuts, outdoor tree
nurseries, and cotton. The following uses are considered as part of the federal action evaluated
in this assessment: nut trees, non-bearing citrus trees, cole crops, fresh market sweet corn, cotton,
pome/stone fruits, leafy vegetables, cucurbits, outdoor nurseries for ornamentals and shade trees,
potato, sweet potato, dry beans and peas, carrot, and fruiting vegetables.  Endosulfan is applied as
liquid spray prepared from emulsifiable concentrates and wettable powder formulations using
aircraft or ground equipment. The pesticide is mostly applied as foliar spray with single rates
ranging from 0.5 to 2.5 Ibs a.i/A at minimum intervals ranging from 5 to 10 days and maximum
total rate from 0.5 to 3.0 Ib a.i/season or year.

Endosulfan is a 70:30 mixture of two biologically-active isomers (a and  P). The isomers differ
in their physiochemical and fate properties.  The chemical is relatively persistent and semi-
volatile.  The P -isomer is generally more persistent and the a-isomer is more volatile. Under
sterlie neutral/basic conditions, hydrolysis is an important degradation route; however,
endosulfans are persistent under acidic conditions. With the exception of hydrolysis in alkaline
conditions, microbial degradation in soils is the predominant route of endosulfan degradation.
The major transformation products found in the fate studies are endosulfan diol (hydrolysis) and
endosulfan sulfate (soil metabolism). Both the diol and sulfate transformation products have the
backbone structures of the parent compound.

Based on a review of the environmental fate data and use characterization for endosulfan,
endosulfan sulfate is considered the major degradate of concern that is addressed in this risk
assessment.  Based on limited data, endosulfan sulfate appears to have similar acute toxicity to
aquatic and terrestrial species (i.e., within an order of magnitude) and is assumed to have similar
chronic toxicity. Endosulfan sulfate is also more persistent than the parent isomers (alpha, beta
endosulfan) and tends to be a major component of total endosulfan residues found in biologically
active environmental media (e.g., aerobic soil).  Endosulfan diol appears to form in only minor
                                            12

-------
quantities in the soil system and is reported to be substantially less toxic than either the parent or
its major degradation product, endosulfan sulfate.

Endosulfans have a high affinity for sorption onto soil and are not expected to be highly mobile.
Endosulfan can move via spray drift and run-off causing contamination of surface waters where
it is expected to reside in sediments. As a result of endosulfan volatility, the chemical can move
to targets beyond its use area through atmospheric transport (volatilization and/or transport on
dust particles).  In this assessment, spray drift and runoff are considered quantitatively while
atmospheric transport and volatilization are considered qualitatively as the potential transport
mechanisms for endosulfan.

Available monitoring data from California suggest that endosulfan is present in important
environmental compartments, although endosulfan use and concentrations in surface water
appear to be declining.  The chemical isomers and/or their common sulfate degradate have been
detected in a variety of off-target locations in California, including rain and snow in the
mountains (McConnell et al, 1998), ground water (Fellers et al., 2004), surface water (STORET
database, CDPR, 2000), stream sediment (Weston et al., 2004; U.S. EPA, 1997), tadpole and
adult frog tissues (Sparling et al., 2001), and ambient air (Air Resource Board "ARB" of
California ( refer to Section 3.2.4.4 Atmospheric Monitoring Data).

The effects determinations for each listed species assessed is based on a weight-of-evidence
method that relies heavily on an evaluation of risks to each taxon relevant to assess both direct
and indirect effects to the listed species and the potential for effects to their designated critical
habitat (i.e., a taxon-level approach).  Since the assessed species exist within aquatic and
terrestrial habitats, exposure of the listed species, their prey and their habitats to endosulfan are
assessed separately for the two habitats.  Tier-II aquatic exposure models are used to estimate
high-end exposures of endosulfan in aquatic habitats resulting from runoff and spray drift from
different uses. Peak model-estimated environmental concentrations, resulting from different
endosulfan uses, ranged from 0.72 to 5.88 jig/L for the total a- and P- endosulphan and
endosulfan sulfate. These estimates are supplemented with analysis of available California
surface water monitoring data from U. S. Geological Survey's National Water Quality
Assessment (NAWQA) program and the California Department of Pesticide Regulation.  The
maximum concentration of endosulfan reported by the California Department  of Pesticide
Regulation surface water database (0.95 |ig/L from the 1991 to 2006 time period) is roughly 6
times lower than the highest peak model-estimated environmental concentration.
The concentrations of endosulfan reported by NAWQA for California surface waters were below
the limits of analytical detection (ranging from  0.0047 to 0.022 |ig/L  depending  on the
endosulfan compound measured).

To estimate endosulfan exposures to terrestrial  species resulting from uses involving endosulfan
foliar applications, the T-REX model is used. This model provides estimates of endosulfan
residues on forage items (e.g., short grass, tall grass, insects, and seeds) that result from direct
deposition on the field. A refinement of this model  (T-HERPS) is also used to enable further
characterization of dietary exposures to terrestrial-phase amphibians and reptiles, based on food
consumption and dietary preferences specific to herptivores. AgDRIFT and AGDISP (Teske and
Curbishley, 2003) models are also used to estimate deposition of endosulfan on terrestrial and
                                            13

-------
aquatic habitats from spray drift. For consumption of contaminated terrestrial prey (e.g., soil-
dwelling invertebrates), a fugacity-based earthworm bioaccumulation model is also used to
characterize direct or indirect effects on listed species.  To address consumption of contaminated
aquatic prey (e.g., fish and aquatic invertebrates), the KABAM model1 is used to estimate
endosulfan exposures to listed species via the aquatic food web.

The effects determination assessment endpoints for the listed species include direct toxic effects
on the survival, reproduction, and growth of the listed species itself, as well as indirect effects,
such as reduction of the prey base or modification of its habitat.  If appropriate data are not
available, toxicity data for birds are generally used as a surrogate for reptiles and terrestrial-
phase amphibians and toxicity data from fish are used as a surrogate for aquatic-phase
amphibians. For endosulfan,  equal toxicity was assumed between the alpha, beta and endosulfan
sulfate compounds, based on similarity in structures, mode of action, and limited data on the
comparative acute toxicity of these compounds. Therefore, modeled or measured environmental
exposures of total endosulfan (sum of alpha, beta and endosulfan sulfate) were compared with
toxicity data for the Technical Grade Active Ingredient (TGAI).  For sediment-borne exposures,
toxicity data were available only for endosulfan sulfate, and thus, comparisons of total
endosulfan EECs were made to toxicity estimates for endosulfan sulfate.

Risk quotients (RQs) are derived as quantitative estimates of potential high-end risk. Acute and
chronic RQs are compared to the Agency's levels of concern (LOCs) to identify instances where
endosulfan use within the action area has the potential to adversely affect the assessed species
and designated critical habitat if applicable, via direct toxicity or indirectly based on direct
effects to its food supply or habitat.  When RQs for each particular type of effect are below
LOCs, the pesticide is determined to have "no effect" on the listed species being assessed.
Where RQs exceed LOCs, a potential to cause  adverse effects is identified, leading to a
conclusion of "may affect." If a determination is made that use of endosulfan "may affect" the
listed species being assessed and/or its designated critical habitat (if applicable), additional
information is  considered to refine the potential for exposure and effects. Best available
information is  used to distinguish those actions that "may affect, but are not likely to adversely
affect" (NLAA) from those actions that are "likely to adversely affect" (LAA) for each  listed
species assessed. For designated critical habitat, distinctions are made for actions that are
expected to have 'no effect' on a designated critical habitat from those actions that may have an
effect on designated critical habitat.

Based on the best available information, the Agency makes a May Affect and Likely to
Adversely Affect (LAA) determination for the  CRLF, CTS, SFGS, SMHM,  SJKF, BCB, and
VELB from the use of endosulfan.  A summary of the risk conclusions and effects
determinations for each listed species assessed here and their designated critical habitat (if
applicable) is presented in Table 1.1 and Table 1.2 Further information on the results  of the
effects determination is included as part of the  Risk Description  in Section 5.2. Additionally, the
Agency has determined that there is the potential for effects to critical habitat of CRLF  and CTS
from the use of the chemical.  This LAA determination is based on the potential for direct effects
(to both aquatic and terrestrial-phase CRLF, SFGS, CTS, BCB, VELB, SMHM, SJKF), indirect
effects due to potential decreases in aquatic and terrestrial prey items, and the potential  for
  http://www.epa.gov/oppefedl/models/water/index.htm#KABAM
                                            14

-------
effects to designated critical habitat due to the potential loss of aquatic and terrestrial prey items.
Given the LAA determination for the CRLF, CIS, SFGS, SMHM, SJKF, BCB, and VELB and
potential modification of designated critical habitat for CRLF and CTS, a description of the
baseline status and cumulative effects for the CRLF is provided in Attachment 2 and the
baseline status and cumulative effects for the SFGS, CTS, BCB, VELB, SMHM and SJKF is
provided in Attachment 4.
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species
California red-
legged frog
(Rana aurora
draytonii)
(CRLF)
Effects
Determination 1
LAA1
Basis for Determination
POTENTIAL FOR DIRECT EFFECTS
Aquatic-phase CRLF (Eggs, Larvae, and Adults):
Freshwater Fish RQ
- Acute RQs for freshwater fish (used as a surrogate for aquatic-phase CRLFs) exceed
the listed species acute risk LOC for all 20 modeled crop scenarios.
- Chronic RQs for freshwater fish (used as a surrogate for aquatic-phase CRLFs)
exceed the chronic risk LOC for all 20 modeled crop scenarios.
Likelihood of Individual Mortality
-The chance of individual effects (i.e., mortality) for freshwater fish (surrogate for
aquatic -phase CRLFs) is as high as ~1 in 1.
Ecological Incident Reports
- 67 of the 83 endosulfan-associated ecological incidents reported to the Agency
involve fish and 53 of these are classified as 'probable' or 'highly probable;' 18 of the
20 incident reports associated with 'registered uses' involved mortality to fish.
Species Sensitivity Differences
- Model-based EECs exceed the LOC for approximately 80 to 97% of the tested
freshwater fish species
Surface Water Monitoring Data
- The highest reported values of total endosulfan in California equal or exceed the
acute LOC for freshwater fish.
Bioaccumulation in Aquatic Prey
- Based on consumption of aquatic prey that is predicted to bioaccumulate
endosulfan, acute dose-based RQs exceed the acute listed LOC in all 20 crop
exposure scenarios modeled for small, medium and large CRLF.
Temporal and Spatial Overlap
- Based on current endosulfan use data and potential endosulfan use on agricultural
crops/orchards/vineyard areas, there appears to be a potential for both temporal and
spatial overlap between aquatic-phase CRLF distribution and endosulfan agricultural
use (Appendix E).
Terrestrial-phase CRLF (Juveniles and Adults):
Direct Deposition on Forase Items: Avian RQ
- Acute-dose based RQs for 20g birds (surrogate for terrestrial-phase amphibians)
consuming large and small insects contaminated with endosulfan from direct
deposition exceed LOC for all 20 crop scenarios modeled.
- Chronic RQs for the small and large insect categories exceed the Agency LOC of 1
for 20 and 3 crop exposure scenarios, respectively.
                                         15

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
    Effects
Determination 1
Basis for Determination
                                  Direct Deposition on Forage Items: Refined Herpetofauna Modeling
                                  - Using refined modeling, acute listed LOG was exceeded in 39% or more of the
                                  acute dose-based; and acute dietary-based species/diet combinations modeled. The
                                  chronic LOG was exceeded in 46% of the chronic dietary-based species/diet
                                  combinations modeled.

                                  Likelihood of Individual Mortality
                                  The chance of individual effects (i.e., mortality) for birds (surrogate for terrestrial -
                                  phase CRLFs) and herpetofauna based on direct deposition onto food items is as high
                                  as~l in 1.

                                  Temporal and Spatial Overlap
                                  - Based on current endosulfan use data and potential endosulfan use on agricultural
                                  crops/orchards/vineyard areas, there appears to be a potential for both temporal and
                                  spatial overlap between terrestrial -phase CRLF distribution and endosulfan
                                  agricultural use (Appendix E).	
                                                    POTENTIAL FOR INDIRECT EFFECTS
                                  CRLF Aquatic Prey Items, Aquatic Habitat, Cover and/or Primary Productivity
                                  Freshwater fish and aquatic-phase amphibians'.
                                  - Acute and chronic RQs exceed the listed species acute and chronic LOCs; high
                                  likelihood of individual mortality, EECs exceed the LOC for 80% or more of the
                                  tested fish species and a large number of aquatic incidents involving fish, as
                                  described above for CRLF (see "Potential Direct Effects;  Aquatic Phase CRLF
                                  [eggs, larvae, adults])"

                                  .Freshwater Invertebrates'.
                                  -Acute and chronic RQs for freshwater invertebrates exceed the listed species acute
                                  and chronic risk LOC for all 20 crop exposure scenarios modeled.
                                  -The chance of individual effects (i.e., mortality) for freshwater invertebrates is as
                                  high as ~1 in 1.
                                  - Model-based EECs exceed the LOC for approximately 45% to 60% of the tested
                                  freshwater invertebrate species
                                  - The highest reported values of total endosulfan in California equal or exceed the
                                  listed acute  LOC  and chronic LOC for freshwater invertebrates.
                                  CRLF Terrestrial Prey Items, Riparian Habitat
                                  Terrestrial-phase Amphibians:
                                  - Exceedence of acute and chronic LOCs for terrestrial-phase amphibians as
                                  described above for CRLF (see '•'•Potential Direct Effects; Terrestrial Phase CRLF
                                  [juveniles and  adults])"

                                  Terrestrial Invertebrates:
                                  - Acute RQs exceed the Agency's interim LOC for listed terrestrial invertebrates in
                                  all 20 crop exposure scenarios modeled.
                                  - EECs exceed for large and small insects exceed the LD50 values for 80% and 99%
                                  of the tested terrestrial invertebrate species, respectively.
                                  -The chance of individual effects (i.e., mortality) for terrestrial invertebrates resulting
                                  from direct deposition in application sites is as high as ~1 in 1.
                                  - When indirect effects are considered via exposure of terrestrial insects to spray drift,
                                  spatial overlap  between potential endosulfan use  on agricultural crops, orchards, and
                                  vineyard areas  can extend up to 2 miles from the  source (Appendix E).
                                                      16

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
    Effects
Determination 1
        Basis for Determination
                                 Small Mammals: Direct Deposition on Forage Items
                                 - Acute and chronic dose-based RQs for small mammals foraging on food items
                                 receiving direct deposition of applied endosulfan exceed the acute listed LOG for
                                 mammals in all 20 crop scenarios modeled
                                 - Chronic diet-based RQs for small mammals foraging on food items receiving direct
                                 deposition of applied endosulfan exceed the chronic LOG for mammals in 13 of the
                                 20 crop scenarios modeled
                                 - The chance of individual effects (i.e., mortality) for small mammals is as high as ~1
                                 inl.

                                 Small Mammals: Bio accumulation in Terrestrial Prey
                                  - RQs exceed of acute listed and chronic LOCs in all 20 exposure scenarios modeled
                                 based on small mammals eating earthworms that are predicted to bioaccumulate
                                 endosulfan from soil.
California
tiger
salamander
 (Ambystoma
 californiense
LAA1
POTENTIAL FOR DIRECT EFFECTS
                  Direct Deposition on Forage Items: Avian RQ (Terrestrial-Phase CTS)
                  - Acute-dose based RQs for 20g birds (surrogate for terrestrial-phase amphibians)
                  consuming large and small insects contaminated with endosulfan from direct
                  deposition exceed LOG for all 20 crop scenarios modeled.
                  - Chronic RQs for the small and large insect categories exceed the Agency LOG of 1
                  for 20 and 3 crop exposure scenarios, respectively.

                  Direct Deposition on Forage Items: Refined Herpetofauna Modeling
                  - Using refined modeling, acute listed LOG was exceeded in 39% or more of the
                  acute dose-based; acute dietary-based species/diet combinations modeled. The
                  chronic LOG was exceeded in 46% of the chronic dietary-based species/diet
                  combinations modeled.

                  Likelihood of Individual Mortality
                  The chance of individual effects (/'. e., mortality) for birds (surrogate for reptiles and
                  terrestrial phase amphibians) and herpetofauna based on direct deposition onto food
                  items is as high as ~1 in 1.

                  Bioaccumulation in Aquatic Prey
                  - Based on consumption of aquatic prey that is predicted to bioaccumulate
                  endosulfan, acute dose-based RQs exceed the acute listed LOG in all 20 crop
                  exposure scenarios modeled for small CRLF (used as surrogate for CTS exposure
                  potential).

                  Bioaccumulation in Terrestrial Prey
                  - Based on consumption of terrestrial prey (earthworm) that is predicted to
                  bioaccumulate endosulfan, acute dose-based RQs and chronic diet-based RQs for
                  small birds (surrogate for terrestrial-phase amphibians) exceed the acute listed and
                  chronic LOCs in all 20 crop exposure scenarios modeled.

                  Freshwater Fish (Aquatic-phase CTS):
                  - Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
                  high likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
                  tested fish species and a large number of aquatic incidents involve fish, as described
                  above for CRLF (see "Potential Direct Effects; Aquatic Phase CRLF [eggs, larvae,
                  adults])"	
                                                     17

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
     Effects
Determination 1
        Basis for Determination
                                  Temporal and Spatial Overlap
                                  - Temporal overlap between endosulfan application and all life stages of CTS was
                                  identified.
                                  - A large amount of spatial overlap between CTS occurrence and current (and
                                  potential future use) of endosulfan was identified (Appendix E).	
                                                   POTENTIAL FOR INDIRECT EFFECTS
                                  Freshwater Fish:
                                  - Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
                                  high likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
                                  tested fish species and a large number of aquatic incidents involve fish, as described
                                  above for CRLF (see "Potential Direct Effects; Aquatic Phase CRLF [eggs, larvae,
                                  adults])"

                                  Freshwater Invertebrates:
                                  - Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
                                  high likelihood of individual mortality, and exceedence of LOCs for 45-60% of
                                  freshwater invertebrate species tested,  as described above for CRLF (see "Potential
                                  Indirect Effects; CRLF Aquatic Prey Items, Aquatic Habitat,  Cover and/or Primary
                                  Productivity)"

                                  Terrestrial Invertebrates'.
                                  - Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
                                  all 20 crop exposure scenarios modeled, there is a high likelihood of individual
                                  mortality, and EECs exceed LD50 values for vast majority of tested species, as
                                  described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
                                  Items, Riparian Habitat)	
 San Francisco
 garter snake
 (Thamnophis
    sirtalis
 tetrataenia)
LAA1
POTENTIAL FOR DIRECT EFFECTS
                  Direct Deposition on Forage Items: Avian RQ
                  - Acute-dose based RQs for 20g birds (surrogate for terrestrial-phase amphibians)
                  consuming large and small insects contaminated with endosulfan from direct
                  deposition exceed LOG for all 20 crop scenarios modeled.
                  - Chronic RQs for the small and large insect categories exceed the Agency LOG of 1
                  for 20 and 3 crop exposure scenarios, respectively.

                  Direct Deposition on Forage Items: Refined Herpetofauna Modeling
                  - Using refined modeling, acute listed LOG was exceeded in 39% or more of the
                  acute dose-based; and acute dietary-based species/diet combinations modeled.  The
                  chronic LOG was exceeded in 46% of the chronic dietary-based species/diet
                  combinations modeled.

                  Likelihood of Individual Mortality
                  The chance of individual effects (/'. e., mortality) for birds (surrogate for reptiles and
                  terrestrial phase amphibians) and herpetofauna based on direct deposition onto food
                  items is as high as ~1 in 1.

                  Temporal and Spatial Overlap
                  - Temporal overlap between endosulfan application and all life stages of SFGS was
                  identified.
                  - No spatial overlap identified at the County level based on current endosulfan use
                  reported for California. If cultivated crop/orchard/vineyard uses are considered areas
                                                     18

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
     Effects
Determination 1
        Basis for Determination
                                  of potential endosulfan use in the future, then a substantial amount of spatial overlap
                                  can be observed relative to the total area of SFGS occurrence sections (Appendix E).
                                                   POTENTIAL FOR INDIRECT EFFECTS
                                 Prey Items, Habitat, Cover And/Or Primary Productivity
                                 Freshwater Fish'.
                                 - Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
                                 high likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
                                 tested fish species and a large number of aquatic incidents involve fish, as described
                                 above for CRLF (see "Potential Direct Effects; Aquatic Phase CRLF [eggs, larvae,
                                 adults])"

                                 Freshwater Invertebrates:
                                 - Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
                                 high likelihood of individual mortality, and exceedence of LOCs for 45-60% of
                                 freshwater invertebrate species tested, as described above for CRLF (see "Potential
                                 Indirect Effects; CRLF Aquatic Prey Items, Aquatic Habitat, Cover and/or Primary
                                 Productivity)"

                                 Terrestrial Invertebrates:
                                 - Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
                                 all 20 crop exposure scenarios modeled, there is a high likelihood of individual
                                 mortality, and EECs exceed LD50 values for vast majority of tested species, as
                                 described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
                                 Items, Riparian Habitat)

                                 Small Terrestrial Vertebrates
                                 - Acute and chronic LOG exceedence for birds (surrogate for reptiles and terrestrial-
                                 phase amphibians) and herpetofauna for direct deposition on food items, a high
                                 likelihood of individual mortality, and RQ exceedence of acute listed LOG and
                                 chronic LOG based on consumption of terrestrial and aquatic prey by birds and
                                 mammals as described above for the CRLF (see "Potential Direct Effects;
                                 Terrestrial Phase CRLF [juveniles and adults] and "Potential for Indirect Effects:
                                 CRLF Terrestrial Prey Items, Riparian Habitat) "	
  Salt marsh
harvest mouse
(Reithrodonto
     mys
  raviventris)
   (SMHM)
LAA1
POTENTIAL FOR DIRECT EFFECTS
                  Direct Deposition on Forage Items: Mammalian RO
                  - Acute and chronic dose-based RQs for small mammals foraging on food items
                  receiving direct deposition of applied endosulfan exceed the acute listed LOG for
                  mammals in all 20 crop scenarios modeled
                  - Chronic diet-based RQs for small mammals foraging on food items receiving direct
                  deposition of applied endosulfan exceed the chronic LOG for mammals in 13 of the
                  20 crop scenarios modeled

                  Likelihood of Individual Mortality
                  - The chance of individual effects (i.e., mortality) for small mammals resulting from
                  direct deposition onto forage items is as high as ~1 in 1.

                  Bioaccumulation in Terrestrial Prey
                  - RQs exceed of acute listed and chronic LOCs in all 20 exposure scenarios modeled
                  based on small mammals eating earthworms that are predicted to bioaccumulate
                  endosulfan from soil.

                  Temporal and Spatial
                                                     19

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species

San Joaquin
kit fox
(Vulpes
macrotis
muticd)
Effects
Determination 1

LAA1
Basis for Determination
- Temporal overlap between endosulfan application and all life stages of the SMHM
was identified.
- A small amount of spatial overlap identified at the County level based on current
endosulfan use reported for California and based on areas of potential endosulfan use
in the future (i.e., cultivated crop, orchard and vineyard use areas).
- The potential for endosulfan to be transported long distances in the atmosphere may
increase SMHM exposure, particularly in areas of potential endosulfan use west of
SMHM occurrence sections in Solano County (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
Prey items, habitat, cover and/or primary productivity
Terrestrial invertebrates:
— Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled, there is a high likelihood of individual
mortality, and EECs exceed LD50 values for vast majority of tested species, as
described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
Items, Riparian Habitat)
Small Birds and Mammals (rearing sites):
- Acute and chronic RQ for birds and mammals (whose rearing sites are potentially
used by SMHM) exceed listed acute and chronic LOCs as a result of direct deposition
onto forage items and bioaccumulation in terrestrial prey as described above for the
CRLF (Potential for Direct Effects: Terrestrial-phase CRLF [Juveniles and Adults])
and for the SMHM (see "Potential for Direct Effects ")
POTENTIAL FOR DIRECT EFFECTS
Direct Deposition on Forage Items: Mammalian RO
- Acute and chronic dose-based RQs exceed the Agency acute listed and chronic LOG
in all modeled scenarios (N-20) for large mammals. Chronic dietary -based RQs
exceed the chronic LOG in all 20 modeled scenarios for large mammals that feed on
short grass, tall grass, and broadleaf plants/small insects (chronic dietary-based RQs
for 13/20 scenarios exceed the LOG for large mammals consuming
fruits/pods/seeds/large insects).
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of SJKF was
identified.
- A moderate amount of spatial overlap between SJKF occurrence and current (and
potential future use) of endosulfan was identified (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
Terrestrial Invertebrates:
- Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled, there is a high likelihood of individual
mortality, and EECs exceed LD50 values for vast majority of tested species, as
described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
Items, Riparian Habitat)
Small Birds and Mammals:
- Acute and chronic RQ for birds and mammals serving as prey to the SJKF exceed
listed acute and chronic LOCs as a result of direct deposition onto forage items and
bioaccumulation in terrestrial prey as described above for the CRLF (Potential for
Direct Effects: Terrestrial-phase CRLF [Juveniles and Adults]) and for the SMHM
(see "Potential for Direct Effects ")
20

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species
Bay
checkerspot
butterfly
(Euphydryas
editha
bayensis)
(BCB)
Valley
elderberry
longhorn
beetle
(Desmocems
californicus
dimorphus)
Effects
Determination 1
LAA1
LAA1
Basis for Determination
POTENTIAL FOR DIRECT EFFECTS
Terrestrial Invertebrate ROs:
- Acute RQs exceed the Agency's interim LOC for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled.
- EECs exceed for large and small insects exceed the LD50 values for 80% and 99%
of the tested terrestrial invertebrate species, respectively.
Likelihood of Individual Mortality
-The chance of individual effects (i.e., mortality) for terrestrial invertebrates resulting
from direct deposition in application sites is as high as ~1 in 1.
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of BCB was
identified.
- No spatial overlap identified at the County level based on current endosulfan use
reported for California. If cultivated crop/orchard/vineyard uses are considered areas
of potential endosulfan use in the future, then a substantial amount of spatial overlap
can be observed relative to the total area of BCB occurrence sections.
- When exposure of terrestrial insects via spray drift is considered, spatial overlap
between potential endosulfan use on agricultural crops, orchards, and vineyard areas
can extend up to 2 miles from the source.
- The potential for endosulfan to be transported long distances in the atmosphere may
increase BCB exposure, particularly in areas of potential endosulfan use west of BCB
occurrence sections in Santa Clara County (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
-Although effects to terrestrial plants cannot be quantified due to the lack of data,
aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
- No studies demonstrating significant adverse effects of endosulfan to any vascular
aquatic or terrestrial plant have been identified in the open literature. The one toxicity
study found to be acceptable for quantitative use suggests that aquatic nonvascular
plants are insensitive relative to aquatic animals.
- No ecological incidents have been reported to the Agency that involve any plants
and this was classified with a certainty index of "probable" or higher linking
endosulfan as a cause, despite that it is regularly directly applied on or near a very
wide variety of agricultural plants.
POTENTIAL FOR DIRECT EFFECTS
Terrestrial Invertebrate ROs:
- Acute RQs exceed the Agency's interim LOC for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled.
- EECs exceed for large and small insects exceed the LD50 values for 80% and 99%
of the tested terrestrial invertebrate species, respectively.
Likelihood of Individual Mortality
-The chance of individual effects (i.e., mortality) for terrestrial invertebrates resulting
from direct deposition in application sites is as high as ~1 in 1.
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of VELB was
identified.
- A small amount of spatial overlap identified at the County level based on current
21

-------
Table 1.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species

Effects
Determination 1

Basis for Determination
endosulfan use reported for California. If cultivated crop/orchard/vineyard uses are
considered areas of potential endosulfan use in the future, then a moderate amount of
spatial overlap can be observed relative to the total area of VELB occurrence sections
(Appendix E).
- When exposure of terrestrial insects via spray drift is considered, spatial overlap
between potential endosulfan use on agricultural crops, orchards, and vineyard areas
can extend up to 2 miles from the source
- The potential for endosulfan to be transported long distances in the atmosphere may
increase VELB exposure, particularly in areas of potential endosulfan use west of
VELB occurrence sections.
POTENTIAL FOR INDIRECT EFFECTS
-Although effects to terrestrial plants cannot be quantified due to the lack of data,
aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
- No studies demonstrating significant adverse effects of endosulfan to any vascular
aquatic or terrestrial plant have been identified in the open literature. The one toxicity
study found to be acceptable for quantitative use suggests that aquatic nonvascular
plants are insensitive relative to aquatic animals.
- No ecological incidents have been reported to the Agency that involve any plants
and this was classified with a certainty index of "probable" or higher linking
endosulfan as a cause, despite that it is regularly directly applied on or near a very
wide variety of agricultural plants.
   1  No effect (NE); May affect, but not likely to adversely affect (NLAA); May affect, likely to adversely  affect
   (LAA)
Table 1.2 Effects Determination Summary for the Critical Habitat Impact Analysis
   Designated
 Critical Habitat
       for:
    Effects
Determination 1
                     Basis for Determination
      CRLF
  May Affect
-There is a potential for direct effects to the aquatic-phase CRLF and
indirect effects via reduction of aquatic-phase prey items (aquatic
invertebrates, fish, and aquatic-phase amphibians) as described in Table 1.1
above.
- There is a potential for direct effects to the terrestrial-phase CRLF and
indirect effects via reduction of terrestrial-phase prey items (mammals,
amphibians, and terrestrial invertebrates) as described in Table 1.1 above.
       CTS
  May Affect
There is a potential for direct effects to the CTF and indirect effects via
reduction of aquatic-phase prey items (aquatic invertebrates and fish) as
described in Table 1.1 above.
- There is a potential for direct effects to the CTF and indirect effects via
reduction of terrestrial-phase prey items  (terrestrial invertebrates) as
described in Table 1.1 above.
      BCB
      NE1
-Although effects to terrestrial plants cannot be quantified due to the lack of
data, aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
-No studies demonstrating significant adverse effects of endosulfan to any
vascular aquatic or terrestrial plant have been identified in the open
literature. The one toxicity study found to be acceptable for quantitative use
                                                      22

-------

VELB

NE1
suggests that aquatic nonvascular plants are insensitive relative to aquatic
animals.
- No ecological incidents have been reported to the Agency that involve any
plants and this was classified with a certainty index of "probable" or higher
linking endosulfan as a cause, despite that it is regularly directly applied on
or near a very wide variety of agricultural plants.
-Although effects to terrestrial plants cannot be quantified due to the lack of
data, aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
-No studies demonstrating significant adverse effects of endosulfan to any
vascular aquatic or terrestrial plant have been identified in the open
literature. The one toxicity study found to be acceptable for quantitative use
suggests that aquatic nonvascular plants are insensitive relative to aquatic
animals.
- No ecological incidents have been reported to the Agency that involve any
plants and this was classified with a certainty index of "probable" or higher
linking endosulfan as a cause, despite that it is regularly directly applied on
or near a very wide variety of agricultural plants.
1 No effect (ME)
The Agency has determined that there is No Effect to BCB and VELB designated critical habitat
from the use of endosulfan. Although there were no reliable data to quantitatively evaluate the
effects and the potential risks of endosulfan to terrestrial plants, aquatic non-vascular plants are
not particularly sensitive to endosulfan, endosulfan has a neural toxic mode of action, and no
studies demonstrating significant adverse effects of endosulfan on growth,  survival, or
reproduction of any vascular aquatic or terrestrial plants have been identified in the open
literature. Furthermore, of the 83 ecological incidents reported to the Agency involving
endosulfan, only one was associated with plants (unspecified plant damage) and its certainty
index was only considered "possible." Lastly, a review of efficacy studies  for endosulfan
indicates that adverse effects on crops at registered application rates is not likely.

Based on the conclusions of this assessment, a formal consultation with the U. S. Fish and
Wildlife Service under Section 7 of the Endangered Species Act should be  initiated.

When evaluating the significance of this risk assessment's direct/indirect and habitat effects
determinations, it is important to note that pesticide exposures and predicted risks to the species
and its resources (i.e., food and habitat) are not expected to be uniform across the action area. In
fact, given the assumptions of drift and downstream transport (i.e., attenuation with distance),
pesticide exposure and associated risks to the species and its resources are expected to decrease
with increasing distance away from the treated field or  site of application.  Evaluation of the
implication of this non-uniform distribution of risk to the species would require information and
assessment techniques that are not currently available.  Examples of such information and
methodology required for this type of analysis would include the following:

          •   Enhanced information on the density and distribution of CRLF,  SFGS, CIS,
              BCB,  VELB,  SMHM, and SJKF life stages within the action area and/or
              applicable designated critical habitat. This information would allow for
              quantitative extrapolation of the present risk assessment's predictions of
              individual  effects to the proportion of the population extant  within geographical
                                           23

-------
areas where those effects are predicted. Furthermore, such population
information would allow for a more comprehensive evaluation of the significance
of potential resource impairment to individuals of the assessed species.
Quantitative information on prey base requirements for the assessed species.
While existing information provides a preliminary picture of the types of food
sources utilized by the assessed species, it does not establish minimal
requirements to sustain healthy individuals at varying life stages. Such
information could be used to establish biologically relevant thresholds of effects
on the prey base, and ultimately establish geographical limits to those effects.
This information could be used together with the density data discussed above to
characterize the likelihood of adverse effects to individuals.
Information on population responses of prey base organisms to the pesticide.
Currently, methodologies are limited to predicting exposures and likely levels of
direct mortality, growth or reproductive impairment immediately following
exposure to the pesticide.  The degree to which repeated exposure events and the
inherent demographic characteristics of the prey population play into the extent to
which prey resources may recover is not predictable.  An enhanced understanding
of long-term prey responses to pesticide exposure would allow for a more refined
determination of the magnitude  and duration of resource impairment, and together
with the information described above, a more complete prediction of effects to
individual species and potential  modification to critical habitat.
                              24

-------
  2  Problem Formulation

Problem formulation provides a strategic framework for the risk assessment. By identifying the
important components of the problem, it focuses the assessment on the most relevant life history
stages, habitat components, chemical properties, exposure routes, and endpoints. The structure
of this risk assessment is based on guidance contained in U.S. EPA's Guidance for Ecological
Risk Assessment (U.S. EPA 1998), the Services' Endangered Species Consultation Handbook
(USFWS/NMFS 1998) and is consistent with procedures and methodology outlined in the
Overview Document (U.S. EPA 2004) and reviewed by the U.S. Fish and Wildlife Service and
National Marine Fisheries Service (USFWS/NMFS 2004).

2.1  Purpose

The purpose of this endangered species assessment is to evaluate potential direct and indirect
effects on individuals of the following Federally Threatened  species:
    •   California Red-legged Frog (CRLF), Rana aurora draytonii,
    •  Bay Checkerspot Butterfly (BCB), Euphydryas editha bayensis,
    •  Valley Elderberry Longhorn Beetle (VELB), Desmocerus californicus dimorphus, and
    •   California Tiger Salamander (CTS), Ambystoma californiense,

and the following Federally Endangered species:
    •   San Francisco Garter Snake (Thamnophis sirtalis tetrataenia),
    •   San Joaquin Kit Fox (Vulpes macrotis mutica),  and
    •   Salt Marsh Harvest Mouse (Reithrodontomys raviventris)

These effects arise from FIFRA regulatory actions regarding use of endosulfan on tree crops
(nuts, citrus, and other fruits), field crops (cotton, potatoes, and row crops), cucurbits, leafy
vegetables (lettuce), fruiting vegetables (tomato, strawberry and eggplant), and tree outdoor
nurseries.  In addition, this assessment evaluates whether use on these use sites is expected to
result in effects to designated critical habitat for the CRLF CTS, BCB, and VELB. This
ecological risk assessment has been prepared consistent with the settlement agreement in Center for
Biological Diversity (CBD) vs. EPA etal. (Case No. 02-1580-JSW(JL)) which addresses the CRLF
and was entered in Federal District Court for the Northern District of California on October 20, 2006.
This assessment also addresses the SFGS, SMHM, BCB,  VELB, SJKF, and CTS for which
endosulfan was alleged to be of concern in a separate suit (Center for Biological Diversity (CBD) vs.
EPA et al. (Case No. 07-2794-JCS)).

In this assessment, direct and indirect effects to the CRLF, CTS, SFGS, SMHM, SJKF, BCB,
and VELB and potential effects to designated critical habitat for the CRLF, CTS, BCB, and
VELB are evaluated in accordance with the methods described in the Agency's Overview
Document (U.S. EPA 2004).  The effects determinations for  each listed species assessed is based
on a weight-of-evidence method  that relies heavily on an evaluation of risks to each taxon
relevant to assess both direct and indirect effects to the listed species and the potential for effects
to their designated critical habitat (i.e., a taxon-level approach).  Screening level methods include
use of standard models such as PRZM-EXAMS, T-REX, TerrPlant, AgDRIFT, and AGDISP, all
of which are described at length in the Overview Document.  Additional refinements include:
                                          25

-------
    1.  Use of T-HERPS model for estimating exposure to terrestrial phase amphibians and
       reptiles
    2.  Use of the KABAM model for estimating endosulfan accumulation in aquatic food webs
       and subsequent risks to terrestrial wildlife consuming aquatic prey
    3.  Use of earthworm fugacity modeling to predict concentrations in terrestrial food items for
       the SMHM, SJKF, and CIS.

Use of such information is consistent with the methodology described in the Overview
Document (U.S. EPA 2004), which specifies that "the assessment process may, on a case-by-
case basis, incorporate additional methods, models, and lines of evidence that EPA finds
technically appropriate for risk management objectives" (Section V, page 31 of U.S. EPA 2004).

In accordance with the Overview Document, provisions of the ESA, and the Services'
Endangered Species Consultation Handbook, the assessment of effects associated with
registrations of endosulfan is based on an action area. The action area is the area directly or
indirectly affected by the federal action, as indicated by the exceedence of the Agency's Levels
of Concern (LOCs). It is acknowledged that the action  area for a national-level FIFRA
regulatory decision associated with a use of endosulfan may potentially involve numerous areas
throughout the United States and its Territories. However, for the purposes of this assessment,
attention will be focused on relevant sections of the action area including those geographic areas
associated with locations of the CRLF, CTS, SFGS, SMHM, SJKF, BCB, and VELB and their
designated critical habitat within the state of California. As part of the "effects determination,"
one of the following three conclusions will be reached separately for each of the assessed species
regarding the potential use of endosulfan in accordance with current labels:

          •  "No effect";
          •  "May affect,  but not likely to adversely affect"; or
          •  "May affect and likely to adversely affect".

The CRLF, BCB, CTS, and VELB have designated critical habitat associated with them.
Designated critical habitat identifies specific areas that have the physical  and biological features,
(known as primary constituent elements or PCEs) essential to the conservation of the listed
species. A brief summary of PCEs for these  species being assessed is as follows.

       CRLF.  The PCEs for CRLF are aquatic and upland areas where suitable breeding and
       non-breeding  aquatic habitat is located, interspersed with upland foraging and dispersal
       habitat.

       BCB. The PCEs for  BCB include the presence of annual or perennial grasslands with
       little to no overstory and N-S or E-W slopes with a tilt of >7 degrees, areas of serpentinite
       ultramafic rock or similar soils that support the primary larval host plant (i.e., dwarf
       plantain) and at least one of the species' secondary host plants, the presence of adult
       nectar sources, aquatic features that provide moisture during the spring drought, and the
       presence stable holes and cracks in the soil, and surface rock outcrops  that provide shelter
       during the summer diapause.
                                           26

-------
       CTS. The PCEs for CIS include standing bodies of fresh water (e.g., ponds, vernal pools
       and dune ponds, and other ephemeral or permanent water bodies), barrier-free uplands
       adjacent to breeding ponds that contain small mammal burrows (used for habitat), and
       upland areas between breeding locations and areas with small mammal burrows that
       allow for dispersal among such sites

       VELB. The PCEs for VELB include areas containing the host plant of this species [i.e.,
       elderberry trees (Sambucus sp.)].

If the results of initial screening-level assessment methods show no direct or indirect effects (no
LOG exceedances) upon individuals or upon the PCEs of the species' designated critical habitat,
a "no effect" determination is made for use of endosulfan as it relates to each species and its
designated critical habitat.  If, however, potential direct or indirect effects to individuals of the
species are anticipated or effects may impact the PCEs of the designated critical habitat, a
preliminary  "may affect" determination is made for the FIFRA regulatory action regarding
endosulfan.

If a determination is made that use of endosulfan "may affect" a listed  species  or its designated
critical habitat, additional information is considered to refine the potential for exposure and for
effects to each species and other taxonomic groups upon which these species depend (e.g., prey
items). Additional information, including spatial analysis (to determine the geographical
proximity of the assessed species' habitat and endosulfan use sites) and further evaluation of the
potential impact of endosulfan on the PCEs is also used to determine whether effects to
designated critical habitat may occur. Based on the refined information,  the Agency uses the
best available information to distinguish those actions that "may affect, but are not likely to
adversely  affect" from those actions that "may affect and are likely to adversely affect" the
assessed listed species and/or have "no effect"  on or "may affect" the PCEs of its designated
critical habitat. This information is presented as part of the Risk Characterization in Section 5 of
this document.

The Agency believes that the analysis of direct and indirect effects to listed species provides the
basis for an  analysis of potential effects on the designated critical habitat. Because endosulfan is
expected to  directly impact living organisms within the action area (defined in Section 2.7),
critical habitat analysis for endosulfan is limited in a practical sense to those PCEs of critical
habitat that are biological or that can be reasonably linked to biologically mediated processes
(i.e., the biological resource requirements for the listed species associated with the critical habitat
or important physical aspects of the habitat that may be reasonably influenced  through biological
processes).  Activities that may affect critical habitat are those that alter the PCEs and
appreciably  diminish the value of the habitat.  Evaluation of actions related to use of endosulfan
that may alter the PCEs of the assessed species' critical habitat form the basis of the critical
habitat impact analysis. Designated critical is discussed further in Section 2.6

2.2 Scope

Endosulfan is mainly used as an agricultural insecticide to control insect pests  such as aphids,
fruit worms, beetles, leafhoppers,  moth larvae, and whiteflies. Endosulfan is applied as liquid
                                            27

-------
spray prepared from emulsifiasble concentrate (EC) and wettable powder (WP) formulations
using aircraft or ground equipment. The pesticide is mostly applied as foliar spray with single
rates ranging from 0.5 to 2.5 Ibs a.i/A at minimum intervals ranging from 5 to 10 days and
maximum total  rate from 0.5 to 3.0 Ib a.i/season or year. The following uses are considered as
part of the federal action evaluated in this assessment: nut trees (almonds, hazelnuts, and
walnuts), non-bearing citrus trees, cole crops, fresh market sweet corn, cotton, pome fruits
(apple, and pear), stone fruits (apricot, nectarine, peach, cherry,, plum,  and prune), leafy
vegetables (lettuce, Brussels sprouts, and celery), cucurbits, outdoor nurseries for ornamentals
and shade trees, potato, sweet potato, dry beans and peas, carrot, and fruiting vegetables (pepper,
strawberry, eggplant, and tomatoes). Non-agriculture is a minor use as it is used impregnated in
cattle ear tags.

The end result of the EPA pesticide registration process (i.e., the FIFRA regulatory action) is an
approved product label. The label is a legal document that stipulates how and where a given
pesticide may be used. Product labels (also known as end-use labels) describe the formulation
type (e.g., liquid or granular), acceptable methods of application, approved use sites, and any
restrictions on how applications may be conducted. Thus, the use or potential use of endosulfan
in accordance with the approved product labels for California is "the action" relevant to this
ecological risk assessment.

Although current registrations of endosulfan allow for use nationwide, this ecological risk
assessment and effects determination addresses currently registered uses of endosulfan in
portions of the action area that are reasonably assumed to be biologically relevant to the CRLF,
CTS, SFGS, SMHM, SJKF, BCB, and VELB and their designated critical habitat. Further
discussion of the action area for the CRLF, CTS, SFGS, SMHM, SJKF, BCB, and VELB and
their critical habitat, if any,  is provided in Section 2.7.

The two isomers of endosulfan, alpha (a) and beta (P), degrade at different rates into a common
degradate, the endosulfan sulfate. Although other degradates such as endosulfan diol and ether
form, expected quantities in the environment are low as they may form  as a result of hydrolysis
of spray drift reaching biologically active water bodies where the sulfate is expected to form
instead. In this assessment, the highly persistent endosulfan sulfate is included with both isomers.
This is because endosulfan sulfate is the major degradate forming in the soil system to which the
major amount of the pesticide is applied. Additionally, endosulfan sulfate is highly persistent in
the  major compartments of the environment. Exposure was determined for both parents and the
common endosulfan sulfate degradate using PRZM/EXAMS modeling.

The Agency does not routinely include, in its risk assessments, an evaluation of mixtures of
active ingredients, either those mixtures of multiple active ingredients in product formulations or
those in the applicator's tank. In the case of the  product formulations of active ingredients (that
is, a registered product containing more than one active ingredient), each active ingredient is
subject to an individual risk assessment for regulatory decision regarding the active ingredient on
a particular use site. If effects data are available for a formulated product containing more than
one active ingredient, they  may be used qualitatively  or quantitatively in accordance with the
Agency's Overview Document and the Services' Evaluation Memorandum (U.S., EPA 2004;
USFWS/NMFS 2004).
                                           28

-------
Endosulfan does not have any registered products that contain multiple active ingredients.

2.3 Previous Assessments

In February, 2002, the USEPA completed an Environmental Fate and Ecological Risk
Assessment of endosulfan in support of the re-registration eligibility decision for endosulfan use
on both food and non food crops (USEPA, 2002; DP Barcode D238673). Based on a screening
level assessment for terrestrial impacts and a refined assessment for aquatic impacts, it was
concluded that endosulfan (maximum application rate of 3 Ib a.i./A) was likely to result in acute
and chronic risk to both terrestrial and aquatic organisms. Based on the tomato crop scenario that
yielded the largest EECs, acute and chronic RQ values ranged from about 23 and 44 for
freshwater fish, respectively, to about 190 and 490 for estuarine/marine fish, respectively. Acute
and chronic RQ values for invertebrates ranged from approximately3.3 and 93 for freshwater
and from 42 to 130 for estuarine/marine fish, respectively. Monitoring data reported in the
assessment demonstrated widespread contamination of surface water and ecological incident data
indicated that the use pattern of endosulfan represented a substantial risk of mortality for non-
target aquatic species. Based on the 2002 Environmental Fate and Ecological Risk Assessment
RQs for birds and mammals exceeded the Agency's acute and chronic risk LOCs and ranged up
to a maximum of 2.7  for birds and 40 for mammals.

The refined aquatic risk assessment used probabilistic exposure assessment techniques and was
based on  actual reported application rates in California coupled with a 300-ft spray drift buffer.
The refined assessment concluded that on sites prone to  runoff within the endosulfan use area,
mortality to non-target fish is probable in any given year. Furthermore, based on "non-
conservative" assumptions (i.e., "typical" application rates and a 300-foot spray drift buffer),
EFED concluded that the modeled endosulfan use rates on 88% of the crops modeled would
exceed acute high risk LOCs more than 99% of the time. The assessment's conclusions were
supported by incident data indicating multiple fish kill incidents attributable to endosulfan use
are reported each year. While repopulation of aquatic communities was considered likely in the
assessment (via migration from unaffected areas), the intermediate effect on food chains was
considered uncertain.

In 2007, USEPA published an addendum to the 2002 environmental fate and ecological risk
assessment in order to evaluate the potential impact of new information related to endosulfan
toxicity, bioaccumulation, monitoring and transport, and ecological incidence (USEPA 2007;
D346213).  A number of aspects of the 2007 addendum  differed from the previous assessment in
2002:
   •   Aquatic exposure modeling considered both the parent isomers and the primary degradate
       of concern (endosulfan sulfate)
   •   A review of new monitoring data including the long-range transport of endosulfan was
       conducted
   •   Bioaccumulation potential of endosulfan in aquatic organisms was evaluated using
       empirical data and a food web model
   •   New data on the toxicity of endosulfan sulfate was included
   •   Risk quotients were based on combined exposure to parent and degradate compounds.
                                          29

-------
Risk quotient (RQ) predicted for aquatic organisms resulting from water column exposure to
endosulfan were about 20% to 40% higher in the 2007 addendum compared to the 2002 ERA.
This increase reflected the addition of endosulfan sulfate to the exposure modeling and data
indicating it is of similar acute toxicity to the parent isomers (a and P). Based on the tomato crop
scenario that yielded the largest EECs, acute and chronic RQ values ranged from about 30 to 60
for freshwater fish, respectively (compared to 23 and 44 from the 2002 ERA) to about 230 and
680 for estuarine/marine fish, respectively (compared to 190 and 490 from the 2002 ERA).
Acute and chronic RQ values for invertebrates ranged from approximately 4 to  130 for
freshwater (compared to 3.3 and 93 from the 2002 ERA) and from 50 to 190 for estuarine/marine
(compared to 42 to 130 from the 2002 ERA), respectively.  Findings from ecological incidents
generally supported the findings of endosulfan risk to fish and invertebrates.

Risks to freshwater and estuarine/marine invertebrates resulting from sediment exposure to
endosulfan are evident from the integration of exposure and effect characterization. The RQ
values  for sediment-dwelling invertebrates range form 0.9 to 2.8 depending on species. These
RQ values are based on predicted total endosulfan residues  compared to endosulfan sulfate
toxicity values, which assumes similar toxicity of endosulfan sulfate and total endosulfan
residues.

Based on preliminary results from an aquatic food web bioaccumulation model, risks to
piscivorous wildlife appeared relatively modest with mean predicted acute RQ values exceeding
the Agency acute LOG of 0.1 for one of eight species  modeled (0.15 for river otter) and 90th
percentile estimates exceeding the LOG for three of eight species modeled (0.18, 0.39, 0.20  for
mink, river otter and belted kingfisher, respectively).  Predicted  chronic RQ values did not
exceed the Agency LOG for any of the eight species modeled.

Risks to non-target terrestrial wildlife did not change from the 2002 ERA as a result of this
addendum, because the currently available terrestrial exposure model could not address total
residue exposure.  Based on the 2002 ERA, RQs for birds and mammals exceeded the Agency's
acute and chronic risk LOCs and range up to a maximum of 2.7  for birds and 40 for mammals.

In summary, the 2007 Addendum concluded there is a concordance of evidence that endosulfan
is undergoing long-range transport and has moved to sites distant from use areas. Furthermore,
the sulfate degradate was considered persistent and a significant source for endosulfan to enter
aquatic and terrestrial food chains. Endosulfan was not expected to biomagnify  appreciably in
aquatic food webs (i.e., increasing concentrations with increasing trophic level), although it was
found that endosulfan can bioaccumulate substantially, with BCFs ranging from approximately
1,000 to 3,000 L/kg w.w. Risks to both aquatic and terrestrial animals were identified using the
latest information available. The Agency is currently responding to public comments received on
the 2007 Addendum.

2.4 Stressor Source and Distribution

This assessment considers the stressor to consist of multiple compounds, namely, endosulfan
alpha, endosulfan beta and their common degradate endosulfan sulfate.  The chemical name of
endosulfan is 6,7,8,10,10-hexachloro-l,5,51,6,9,9a-hexahydro-6,9-methano-2,4,3-
                                           30

-------
benzadioxathiepin 3-oxide. Endosulfan has two stereo-isomers, endosulfan alpha and beta (a-
and p-endosulfan). According to the agricultural label, the maximum amount of a is 70% and
is 30%. Figure 2.1 provides the structures of a. and P-endosulfan and major transformation
products.
     ot-endosulfan (959-98-8)
  P-endosulfan (3213-65-9)
  Endosulfan Sulfate (1031-07-8)
                            The Multi-Chemical Stressor of Endosulfan
Endosulfan Diol (2157-19-9)
Endosulfan Ether (3369-52-6)
Endosulfan Lactone (3868-612-9)
                                 Other Degradates of Endosulfan
Figure 2.1 Chemical structures of endosulfan and related compounds

The source of the stressor is foliar application of endosulfan formulated products including
wettable powders (WP) and emulsifiable concentrates (EC) by air, ground and airblast
application methods. In an agricultural setting, major quantities of the applied pesticide active
ingredient reach target foliage and eventually end-up in the soil system. Smaller quantities of the
applied pesticide reach nearby field or water bodies initially by drift and followed by
runoff/erosion and by volatilization. Biologically mediated transformation of a and P-endosulfan
is expected to occur in the soil system producing endosulfan  sulfate (maximum levels >52%), a
common degradate of toxicological concern. Endosulfan sulfate is added to parent(s) as a
stressor because it is a major toxic and persistent soil transformation product.  In contrast,
endosulfan diol is not included as a component of the stressor because it is less toxic and
expected to form in biologically inactive neutral and alkaline systems. Endosulfan diol was a
major transformation product in sterile aqueous media (maximum levels >82%)  and was only a
                                            31

-------
   minor transformation product in aerobic soil systems in the laboratory and the field (maximum
   levels <10 %) (USEPA, 2002; DP Barcode D238673).

   For the purpose of this assessment, the term "endosulfan" refers to the technical grade active
   ingredient (TGAI) unless otherwise indicated.  The term "total endosulfan" or "endosulfans"
   refers to the combination of a -endosulfan, p-endosulfan and endosulfan sulfate unless
   otherwise noted.

          2.4.1 Environmental Fate Properties

   The physiochemical properties of the multi-chemical stressor are summarized in Table 2.1. Fate
   and transport data for endosulfan isomers and common degradate endosulfan sulfate (the multi-
   chemical stressor) are summarized in Table 2.2. Major and minor degradates detected in
   submitted environmental fate and transport studies are also included in the same Table 2.2.

   Table 2.1 Physiochemical properties of the multi-chemical stressor
Property
Formula
Molecular Weight
CAS Number
Water Solubility
Vapor Pressure
(torr@25°C)
Henry's Law Constant
(atm-m3mor1@25°Q
Log Kaw (air/water)
Log Km (octanol/water)
Corresponds to Kow of:
LogKoa
a -endosulfan
C9H6C1603S
406.9 g/mole
959-98-8
530 ug/L
4.6x10 5; 1.5xlO-5&3.0xlO-6
3.03xlO'6
-3.56
4.93
95,499
6.41 & 8.64
p-endosulfan
C9H6C1603S
406.9 g/mole
33213-65-9
280 ug/L
2.4x10 5; 6.9 & 7.2 x 10'7
l.SSxlO'6
-4.75
4.78
60,256
6.41 & 8.64
Endosulfan sulfate
C9H6C16S04
422.9 g/mole
1031-07-8
330 ug/L
9.75xl06
1.64X10'5
-4.78
3.71
5,129
8.45 & No value
 Vapor pressure for a and P-endosulfan (MRIDs 414215-01 and 400606-01) and for endosulfan sulfate from: Hinckley et
al (1990)2
 All three values were calculated from the 1st vapor pressure values above (in Bold) using molecular weight, solubility &
25°C temperature
 Source of data for partitioning coefficients: Log Kaw and Kow as reported (Schenker et al, 2005)3 while Log Koais
estimated by EPI Suite v4.04; 1st value= EPI Suite estimate & 2nd value= from EPI suite database	
    Daniel A. Hinckley, Terry F. Bidleman, William T. Foreman, Jack R. Tuschall. 1990. Determination of vapor
   pressure for non-polar and semi-polar organic compounds from gas chromatography retention data J. Chem. Eng.
   Data, 1990, 35 (3), pp 232-237
   3 Schenker, U., MacLeod, M, Scheringer, M, Hungerbuhler K., 2005. Improving data quality for environmental
   fate models: A least-square adjustment procedure for harmonizing physiochemical properties of organic compounds.
   Environmental Science & Technology 39, 8434-8441.
   4 The Estimation Programs Interface EPI Suite: http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm
                                                 32

-------
    Table 2.2 Summary of fate and transport properties for endosulfan isomers and sulfate
Study
Value (units)
Major & Minor Degradates
Reference &
Study Status *
Hydrolysis
t!/2 @ pH 5: stable (>200 days) for both a
andp
t!/2 @ pH 7: 19 days for a;l 1 days for (3
endosulfan sulfate: 184 days **
Major: Endosulfan diol
414129-01(A)for
only pH 5 and 7
Aqueous Photolysis
Stable, based on absorption spectrum
None reported
Registrant Data
Soil Photolysis
Stable
Minor: Endosulfan diol
414307-01 (A)
Aerobic Soil
Metabolism
(Five soils)
a-endosulfan: t!/2 35-67 days; 90th %= 57
days
(3-endosulfan: 104-265 days,; 90th %=208
days
endosulfan sulfate: Shows no clear decline
pattern	
Major: Endosulfan Sulfate
Minor: Endosulfan diol &
lactone
438128-01 (S)
Anaerobic Soil
Metabolism
 (Two soils, Two phase
study)	
a-endosulfan: 105 & 124 days; 90th%=115
days
p-endosulfan: 136 &161 days; 90th%=149
days
endosulfan sulfate: 120 & 165  days; 90th
%=143 days
All calculated from the anaerobic phase of
the experiment	
Major: Endosulfan Sulfate
Formed during the aerobic
phase
Minor: Endosulfan diol &
lactone
414129-04 (A)
Organic Carbon based
Adsorption Coefficient
(Koc)	
a-endosulfan: Koc 10,600 ml/g (Average for
4 soils)
P-endosulfan: Koc 13,500 ml/g (Average for
4 soils)
endosulfan sulfate & diol: Qualitatively,
when compared to parents sulfate is similar
or slightly more mobile while diol is
substantially more mobile
Not Applicable
                                                                                              414129-06 (S)
414129-05 (U)
No aerobic aquatic metabolism study was submitted; Submitted anaerobic aquatic metabolism study was Un-acceptable
449178-01 (U)	
Terrestrial Field
Dissipation
a-endosulfan: t!/2 46 days (GA tomato), 70
days (CA cotton), and 6-11 days (CA
cotton)
P-endosulfan: t!/2 90 days (GA tomato), 103
days (CA cotton) and 19-63 days (CA
cotton)
Important Note: tl/2 in soil surface layer,
encompassing movement as well as
degradation	
Endosulfan-sulfate and
endosulfan-diol. The sulfate
was dominant and appeared
to be more persistent	
413097-02 (A)
414686-01 (A)
430697-01 (A)
Accumulation in Non-
target Aquatic
Organisms	
600x in mussels with depuration t!/2 of 34 hr
2,429x for edible tissue and 2,755x for
whole body of mullet with depuration within
48 hr
Not Applicable
05003053 (A)

05005824 (A)
* Reference Numbers indicated are MRID or Access Numbers, Study Status: A= Acceptable; S= Supplemental and
                                                     33

-------
Study
Value (units)
Major & Minor Degradates
Reference &
Study Status *
U=Unacceptable
** An Extrapolated Hydrolysis at pH 7 value reported by Becker et al (2008)5 From Guerin & Kennedy 1992

    The physiochemical data for endosulfan isomers and sulfate indicate that the three chemicals are
    practically insoluble in water. Vapor pressure and Henry's law constant values for the three
    chemicals suggest that volatility is important in the dissipation of the stressor.  Endosulfan
    isomers and common degradate can be considered as semi-volatile compounds and can readily
    partition into air in measurable amounts as indicated by their air/water partition coefficients
    (from -4.78 to -3.56).  Air/octanol and octanol/water coefficients suggest the three chemicals
    have high potential for bio-accumulation in aquatic and terrestrial organisms, respectively (Table
    2.1).

    Based on laboratory data, the two biologically-active a- and -P stereo isomers of endosulfan
    differ in physiochemical and fate properties (Table 2.2). The -P isomer is generally more
    persistent and the a-isomer is more volatile.  In sterile conditions, a- and -P isomers are highly
    persistent in aqueous acid conditions (t/^>200 days) but highly vulnerable to hydrolysis in
    aqueous neutral conditions (t/^= 11 and 19 days for a-  and -p-endosulfan, respectivelly)
    producing a major degradate, the endosulfan diol.  Abiotic transformation by photolysis in
    aqueous media or on soil is not important.  Both a- and -P endosulfan showed no significant
    absorption peaks in the visible light region of the spectra (290-800 nm). Additionally, the
    isomers had a similar degradation profile in soil exposed to light or kept in a dark control.

    In contrast to abiotic transformation, biotic mediated oxidation in the soil system appears to be
    the dominant pathway for degradation of both endosulfan isomers into endosulfan sulfate. In
    several aerobic soil studies, endosulfan  isomers degraded into endosulfan sulfate with half-lives
    ranging from one to two months for a -endosulfan and from three to nine months for P -
    endosulfan.  Persistence of endosulfan appears to increase under anaerobic conditions. In two
    anaerobic soil metabolism studies, a- and P-endosulfan degraded with half-lives ranging from
    105 to 124 days for a-endosulfan and 136 to 161 days for P-endosulfan. Endosulfan sulfate was
    the major degradate formed during  the aerobic phase of the study then degraded with a half-life
    of 120 to  165 days.

    In field dissipation studies, endosulfan persisted in the  surface soil for weeks to months after
    application.  Field dissipation rates  capture a combination of degradation, transport, and uptake
    processes whereas laboratory studies capture only degradation processes.  Therefore, field
    dissipation rates can only be  compared to laboratory rates when these other transport and uptake
    processes are taken into account.

    The major transformation products  found in fate studies are endosulfan diol  (hydrolysis in sterile
    conditions) and endosulfan sulfate (soil  metabolism). Another minor transformation product is
    endosulfan lactone which was detected  at 4-6% in the anaerobic soil  system. It is noted that all
     L. Becker, U. Schenker and M. Scheringer. 2008. Overall persistence and long range transport potential of
    endosulfan and its transformation products. Swiss Institute of Technology, ETH Zurich, Switzerland
                                                34

-------
three degradation products have backbone structure of the parent compounds (Figure 2.1).

When considering how to account for degradates, the two major degradates were considered; that
is endosulfan sulfate and endosulfan diol. Endosulfan sulfate was identified for inclusion as a
component of the stressor for the following reasons:

    (1) Endosulfan sulfate forms by oxidation in the aerobic soil system (maximum levels >52%
       of the applied). Only minor amounts of endosulfan diol and lactone (<10%) are formed
       in the aerobic soil system. Also, the sulfate is expected to be more persistent than its
       parents in the aerobic soil system. In aerobic soil studies, endosulfan sulfate was the
       major persistent degradate forming from a- and p-endosulfan;

    (2) Endosulfan diol was reported to be less toxic than parent or endosulfan sulfate (Verma et
       a\ 2006, siddique et al 2003, and others6). The persistent organic pollutants review
       committee stated "Endosulfan is a very toxic chemical for nearly all kind of organisms.
       Metabolism occurs rapidly, but the oxidised metabolite endosulfan sulfate  shows an acute
       toxicity similar to that of the parent compound. In contrast, endosulfan-diol, which is
       another metabolite of endosulfan, is found substantially less toxic to fish by about three
       orders of magnitude"7

    (3) The bulk of applied pesticide  reaches the soil system with relatively small  amounts
       reaching water bodies by drift (estimated to be 2.4% with a 300 ft. buffer).

    (4) Endosulfan diol is assumed to form from parent isomers reaching water by drift only in
       abiotic conditions. As most environmentally  relevant water bodies are biologically active,
       endosulfan diol is not expected to form in substantial amounts in natural aquatic systems.
       However, no acceptable aerobic or anaerobic aquatic water sediment systems are
       available to confirm this assumption.

Endosulfan isomers can be considered as hardly mobile in the soil system. The two chemical
species have relatively high affinity to soil particles (Koc  ranges from 10,600 to 13,500 ml/g)
(Table 2.2). This property causes chemicals to partition  into the soil and reduces readiness for
leaching into ground water. Endosulfan sulfate appear to be similar to parent isomers in mobility
although the endosulfan diol is substantially more mobile (Table 2.2).

Laboratory measured KOW values (95,499 for a -endosulfan, 60,256 for P-endosulfan, and 5,129
for endosulfan sulfate) suggest a relatively high potential to bioconcentrate in aquatic organisms
such as fish. However, supplemental studies suggest that endosulfan isomers did not bio-
concentrate at the high levels suggested by K  (Table 2.1). In one study conducted with a -
 http://umbbd.msi.umn.edu/end/end map.html
http://www.ces.clemson.edu/ecl/caseStudy/case2.pdf

7 http://www.unon.org/confss/doc/unep/pops/POPRC  04/POPRC 4 14/K0841692%20POPRC-4-14.pdf
Persistent Organic Pollutants Review Committee; Fourth meeting; Geneva, 13-17 October 2008; Item 7 (a) of the
provisional agenda: Consideration of chemicals newly proposed for inclusion in nnexes A, B or C of the
Convention: endosulfan
                                            35

-------
endosulfan on mussels, the bio-concentration factor was approximately 600X, with a depuration
half-life of 34 hours (Ace. No. 05003053). In another study, conducted on striped mullet, the bio-
concentration factor was 2,400X for combined isomers in edible tissues (Ace. No. 05005824).
In this study, endosulfan depurated after 48 hours (Table 2.2). Tissue analysis in one study
revealed the presence of endosulfan sulfate rather than a- and p-endosulfan, suggesting the
potential accumulation of this degradate.

       2.4.2 Environmental Transport Mechanisms

Potential transport mechanisms or routes of pesticide exposure include surface water
runoff/erosion, spray drift, and secondary drift of volatilized or soil-bound residues leading to
deposition onto nearby or more distant ecosystems. In addition, pesticides can reach shallow
ground water through leaching when the chemical  is characterized by relative high mobility  and
moderate persistence. Low mobility pesticides may also leach into ground water if they can
persist long enough for such movement to occur.  For endosulfan, surface water run-off/erosion,
drift, and atmospheric transport are expected to be the major routes of exposure.

A number of studies have documented atmospheric transport and re-deposition of pesticides
from the Central Valley to the Sierra Nevada Mountains (Fellers et al., 2004, Sparling et al.,
2001, LeNoir et al., 1999, and McConnell et al., 1998).  Prevailing winds blow across the
Central Valley eastward to the Sierra Nevada Mountains, transporting airborne industrial and
agricultural pollutants into the Sierra Nevada ecosystems (Fellers et al., 2004, LeNoir et al.,
1999, and McConnell et al.,  1998).  Several sections of the range and/or critical habitat for the
listed species being evaluated in this assessment are located east of areas of potential endosulfan
use (Appendix E). The magnitude of transport via secondary  drift depends on the ability of
endosulfan to be mobilized into air and its eventual removal through wet and dry deposition  of
gases/particles and photochemical reactions in the  atmosphere.  Therefore, physicochemical
properties of endosulfan that describe its potential to enter the air from water or soil (e.g.,
Henry's Law constant and vapor pressure), pesticide use data, modeled estimated concentrations
in water and air, and available air monitoring data from the Central Valley and the Sierra Nevada
are  considered in evaluating the potential  for atmospheric transport of endosulfan to locations
where it could impact the listed species included in this assessment.

In general,  deposition of drifting or volatilized pesticides is expected to be greatest close to the
site of application. Computer models of spray drift (AgDRIFT and/or AGDISP) are used to
determine potential exposures to aquatic and terrestrial organisms via spray drift.

The distance of potential impact away from the use sites is determined by the distance required
to fall below the LOG for the taxonomic groups with the highest RQ to LOG ratio (for
endosulfan, this consists of terrestrial insects for transport via spray drift and freshwater fish for
transport via downstream flow).

       2.4.3 Mechanism of Action

Endosulfan is a dioxathiepin (broadly classified as a chlorinated hydrocarbon) insecticide and
acaricide. Technical endosulfan is made up of a mixture of two molecular forms (stereo isomers)
of endosulfan - the alpha (a) and beta (P) isomers. This arylheterocycle acts as a poison to a
                                           36

-------
  wide variety of insects and mites on contact through blockage of GAB A- (gamma amino butyric
  acid) gated chloride channels, which in turn induces the uptake of chloride ions by neurons. The
  blockage of the GAB A neurotransmitter activity by cyclodiene insecticides like endosulfan
  results in only partial repolarization of the neuron and a state of uncontrolled excitation (Klassen
  & Watkins, 1999). Stimulation of the central nervous system is the major characteristic of
  endosulfan poisoning  (Ecobichon 1991).

  Dissociation studies (Rauh etal. 1997) showed that dieldrin, ketoendrin, toxaphene, heptachlor
  epoxide and a- and p-endosulfan competitively bind with GABA receptors.  Additionally, animal
  data indicate that toxicity may also be influenced by species and by level of protein in the diet;
  rats which have been deprived of protein are nearly twice as susceptible to the toxic effects of
  endosulfan. Solvent and/or emulsifiers used with endosulfan in formulated products may
  influence its absorption into the system via all routes: technical endosulfan is slowly and
  incompletely absorbed into the body whereas absorption is more rapid in the presence of
  alcohols, oils, and emulsifiers (Gupta and Gupta 1979).

         2.4.4 Use Characterization

  Current labeled use

  (a) Nationally

  Current registered uses of endosulfan include a wide variety of food crops, non-food crops and
  ornamentals.  The list includes nut trees (almonds, hazelnuts, and walnuts), non-bearing citrus
  trees, cole crops (including Kohlrabi) , fresh market sweet corn, cotton, pome fruits (apple, and
  pear), stone fruits (apricot, nectarine, peach,  cherry, plum, and prune), leafy vegetables (lettuce,
  Brussels sprouts), celery, cucurbits, outdoor nurseries for ornamentals and shade trees, potato,
  sweet potato, dry beans and peas, carrot, and fruiting vegetables (pepper, strawberry, eggplant,
  and tomatoes), tobacco (field, seed/plant beds), pineapple (fresh market only),  and blueberry. No
  residential or public health uses are labeled for endosulfan although impregnated ear-tags are
  labeled for use on cattle.

  (b) California (CA)

  Labeled endosulfan uses that could potentially be used in California were analyzed in detail
  because it is the  critical first step in evaluating the federal action. The current  label for
  endosulfan represents the FIFRA regulatory  action; therefore, labeled use and  application rates
  specified on the label form the basis of this assessment. A summary of these labeled uses for
  endosulfan is included in Table 2.3. Application parameters included in Table 2.3 cover the
  most conservative application parameters and method of application for labeled crops.

Table 2.3 Endosulfan labeled uses assessed for California (No single application rate over 1.5 Ibs except for
orchard crops and strawberries)
Use (Application Method) 1
Almond, Hazelnut & Walnut
Application: Maximum Rate, Number and Total 2
Single Rate (Ib
a.i./A)
2.00
Number
1
Total/Season Or
Year
2.00
Minimum
Application Intervals
(day)3
NA
                                              37

-------
Table 2.3 Endosulfan labeled uses assessed for California (No single application rate over 1.5 Ibs except for
orchard crops and strawberries)
Use (Application Method) '""
Citrus, non-bearing trees only
Broccoli, Cabbage, Chinese cabbage &
Cauliflower
Collards, Kale & Mustard Green
Corn, Sweet for fresh market only
Cotton (Ground)
Cotton (Aerial)
Apple
Apricot, Nectarine, Peach, Cherry, Pear,
Plum & Prune
Lettuce (head & Leaf) 4 & Brussels
Sprouts
Cucumber, Melons, Pumpkin & Squash 5
Eggplant
Ornamentals or shade Trees 6
Potato
Sweet Potato
Dry Beans, Dry Peas & Pepper
Carrot
Celery
Application: Maximum Rate, Number and Total 2
Single Rate (Ib
a.i./A)
2.50
0.50+2.00
1.00
0.75
1.50
1.00
0.50+1.50
0.75
2.50
0.50+0.75+1.25
2.50
0.50+2.00
1.00
1.00
0.50+0.50+1.0
0.50
0.25
0.50
1.00
0.50
0.50
1.00
1.00
1.00
0.5
Number
1
2
2
1
1
2
2
2
1
3
1
2
2
2
3
1
2
6
2
4
3
2
1
1
2
Total/Season Or
Year
2.50
2.50
2.00
0.75
1.50
2.00
2.00
1.50
2.50
2.50
2.50
2.50
2.00
2.00
2.00
0.50
0.50
3.00
2.00
2.00
1.50
2.00
1.00
1.00
1.00
Minimum
Application Intervals
(day)3
NA
10
7
NA
NA
7
7
7
10
10
10
10
5
7
7
NA
7
10
5
5
5
5
NA
NA
5
                                                   38

-------
Table 2.3 Endosulfan labeled uses assessed for California (No single application rate over 1.5 Ibs except for
orchard crops and strawberries)
Use (Application Method) '""
Strawberry
Tomato
Application: Maximum Rate, Number and Total 2
Single Rate (Ib
a.i./A)
2.00
1.00
1.00
0.5
Number
1
2
2
4
Total/Season Or
Year
2.00
2.00
2.00
2.00
Minimum
Application Intervals
(day)3
NA
15
7
7
1 Unless specified, the application method can be any of aerial or ground or airblast for tree crops only
2 In some cases, the number of applications was adjusted in order to always include the maximum single rate and at the
same time abide by the maximum total applied/season or year.  An additional application entry was included for specific
crops to abide by both the maximum number of applications and the maximum total applied/season or year. Total Ibs a.i
applied/Acre was specified to be either per season or per year, therefore, multiple cropping may occur as follows: Two
crops for Broccoli & Lettuce; Three crops for Cabbage, Chinese cabbage, Cauliflower, Collards, Mustard Green &
Sweet Corn; and Four crops for Kale.
3 Application Intervals in italic were assumed as it was not specified on the Label. Assignments were based on previous
assessments (EPA, 2002) or on similarities in crops and target pest(s); NA= Not Applicable
4 Lettuce including: head, leaf (black seeded Simpson, and salad. Do not exceed 2 applications after thinning)
5 Squash: including summer, winter, and Hubbard
6 Outdoor Nursery for Shade trees and Shrubs only
** Although pineapple was one of the crops labeled for CA, the use was not considered as pineapple is not grown in CA.
  Data for Table 2.3 were obtained from the newly revised and USEPA approved 2007-2008
  labels which included the following 2002 RED mitigation measures:

      (1) Deletion of use on succulent beans, succulent peas, spinach, grapes, and pecans;
      (2) Reduction of maximum seasonal application rates (Ibs./a.i./A):
             (a) From 3.0 to 2.5 Ibs for pome fruit, stone fruit, and citrus;
             (b) From 3.0 to 2.0 Ibs for melons, cucurbits, lettuce, tomatoes, sweet potatoes,
                cotton (ground), broccoli, cauliflower, cabbage, kohlrabi, brussels sprouts,
                strawberries, filberts, walnuts, almonds, macadamia nuts, peppers, eggplant,
                potatoes, carrots, dry beans, dry peas, and tobacco;
             (c) From 3.0 to 1.5 Ibs for sweet corn, cotton (aerial) and blueberries;
             (d) From 3.0 to 1.0 Ib for celery.
      (3) Requirement of 100 ft. spray buffer for ground applications between a treated area and
         water bodies;
      (4) Requirement of 30 ft. maintained vegetative buffer strip between a treated area and water
         bodies;
      (5) Requirement of all products to be Restricted Use;
      (6) Restriction of use on:
             (a) Cotton to AZ, CA, NM, OK and TX only; and
             (b) Tobacco to IN, KY,  OH, PA,  TN and WV only.

  Similar mitigation measures apply for California with the following exceptions:
                                                39

-------
    (1) Maximum seasonal rates in Ibs a.i. /A/season were reduced to 1.5 Ibs for sweet corn, 1.0
       Ib for carrots and 0.5 Ib for eggplant;
    (2) Deletion of use on tobacco, blueberries, and kohlrabi.

These measures were included in Table 2.3 above. The complete analysis of labeled use along
with registration numbers  is included in Appendix A along with use verification memo of SRRD
and RD.  No 24(c) or sections 18s are relevant to California.

Finally, slightly different application restrictions are included for use in CA to further reduce
drift and protect surface waters. For example, at the national level buffer distance is 300 ft for
aerial application and  100 ft for ground application while the distance in CA is 300 ft for both
aerial and ground application. A 30 ft vegetative buffer is also included in CA. Other
restrictions also include: nozzle height maximum of 4 ft above ground or crop canopy; a
maximum wind speed 10 mph; and the use fine-medium or coarse spray ASAE 573.

Endosulfan is also labeled for use in ear tags for cattle. The ear tag is impregnated with
endosulfan (30% a.i.) and  is to be attached to animal ears (Table  2.4).

Table 2.4 California labeled non-crop use for endosulfan
Use Pattern*
Beef Range Feeder Cattle (Meat, All or
unspecified)
Dairy Cattle (Lactating or Unspecified)
Application Information
Two ear tags per animal (0
198 Ib a.i./Animal) for fly control (e.g
., horn fly)
* Avenger insecticide cattle ear tag (CA registration number 61483- 65-AA; November, 05, 2007)

In this assessment, the potential use of endosulfan in cattle ear tags within California is not
evaluated.  Most of the endosulfan released from cattle ear tags is expected to volatilize, adsorb
to the cow and/or wash off into the soil. Given that this particular formulation is not expected to
be subject to extensive transport, exposure is expected to be deminimus and was not
quantitatively assessed.
 http://www.cdpr.ca.gov/cgi-bin/label/labrep.pl
                                            40

-------
Historical usage
(a) Nationally

National data show a decline in the estimated endosulfan usage. USGS data show a decreased in
total usage from 1.79 million Ibs in 1992 to 1.52 million Ibs in 1997 and 0.79 million Ibs in
20029. Reported average annual endosulfan usage was estimated at approximately 1.38 million
Ibs, based on available pesticide survey data for the years from 1990 to 1999 (EPA, 2002).  A
summary of data, from both sources, is included in Table 2.5 showing the distribution of Ibs
used for various crops and in Figure 2.2 showing the general spatial distribution of usage
through out the United States.

Table 2.5 Top use patterns for endosulfan for selected years from 1990 to 2002
Crop
Cotton
Fruits (apples & pears)
Potatoes
Tomatoes
Pecan
Grapes
Cucurbits: cantaloupe, squash, watermelon & cucumber
Tobacco
Lettuce
Alfalfa
Green beans
Horticulture
Others**
Totals in millions of Ibs a.i
% of Total for the Year(s)
1992
25%
17%
10%
8%
7%
7%
6%
***
5%
***
***
***
15%
1.79
1997
29%
12%
11%
3%
3%
**
8%
11%
3%
3%
***
***
17%
1.52
1990-1999*
21%
8%
9%
4%
4%
**
6%
5%
4%
***
***
4%
35%
1.38
2002
20%
14%
11%
11%
**
**
8%
7%
4%
**
4%
***
21%
0.79
   * Average for the years from 1990 to 1999
Others** = sum of all uses donated as "*
Usage data indicate that the top three use patterns were for cotton, certain fruits, and potatoes
during the period from 1992 to 2002.  The 4th and 5th use patterns include combinations of
tomatoes, tobacco, and cucurbits or pecan, grapes, and cucurbits. As suggested by the 2002
RED, observed changes in the use pattern may be related to shifts in use in response to the
dynamics of the agricultural system (e.g., changes in crop area), pest populations (e.g., pest
outbreaks) changes in pesticide availability (e.g., new pesticides registered and restrictions on
old pesticides), and the variability in using data from various information sources (EPA, 2002).
 ' Sources of Data for 1992, 1997: URLs
                                            41

-------
 u
 D
       1992
      Average annual use of active ingredient (pounds per square mile of agricultural land in county
                 1992                       1997                        2002
             No estimate use             No estimated use              No estimated use
0.001 to 0.007
0.008 to 0.028
0.029 to 0.108
0.109 to 0.454
   > 0.455
0.001 to 0.006
0.007 to 0.027
0.028 to 0.105
0.106 to 0.480
   > 0.455
0.001 to 0.005
0.006 to 0.018
0.019 to 0.064
0.065 to 0.259
   > 0.260
Figure 2.2 Endosulfan Use in Total Pounds per County (note vlues are different in chosen categories)

(a) California

The Agency's Biological and Economic Analysis Division (BEAD) provided an analysis of both
national- and county-level usage information using state-level usage data obtained from USDA-
NASS10, Doane11 and the California's Department of Pesticide Regulation Pesticide Use
Reporting (CDPR PUR) database12. CDPR PUR is considered a more comprehensive source of
  United States Depart of Agriculture (USDA), National Agricultural Statistics Service (NASS) Chemical Use
Reports provide summary pesticide usage statistics for select agricultural use sites by chemical, crop and state. See
http://www.usda. gov/nass/pubs/estindxl .htm#agchem.
11 www.doane.com (the full dataset is not provided due to its proprietary nature)
12 The California Department of Pesticide Regulation's Pesticide Use Reporting database provides a census of
pesticide applications in the state. See http://www.cdpr.ca.gov/docs/pur/purmain.htm.
                                               42

-------
usage data than the USDA-NASS or EPA proprietary databases, and thus the usage data reported
for endosulfan by county in this California-specific assessment were generated using CDPR PUR
data.  Thirteen years (1994-2006) of usage data were included in this analysis. Data from CDPR
PUR were obtained for every pesticide application made on every use site at the section level
(approximately one square mile) of the public land survey system.  BEAD summarized these
data to the county level by site, pesticide, and unit treated. Calculating county-level usage
involved summarizing across all applications made within a section and then across all sections
within a county for each use site and for each pesticide.  Figure 2.3 summarizes available
historical usage data for the application years from 1994 to 2006.
600,000n

500,000
•c
g 400,000
8
£ 300,000
O
™ 200,000
SI
_l
100,000











<=







































D Total
• Total










Lbs





















of Endosulfan a.i









Applied
Acres Treated



















^



p





P^
_










i=








F

1

1994 1995 1996 1997 1998

=





_

^



=

^

1999 2000 2001
Application Year




=

2002




=

2003


^

•

2004

n
2005

c










2006
Figure 2.3 Historical usage data for endosulfan

Data show a gradual decrease in crop acreage treated and Ibs of a.i. used for the treatment from
1995 to 2006.  Four distinct periods can be identified, namely, 1994, 1995-1997, 1998-2004, and
2005-2006. Usage during these periods decreased from nearly 1A million Ibs (1st period), to 1A
million Ibs (2nd period), to 155 thousand Ibs (3rd period), to 88 thousand Ibs (4th period). BEAD
usage data included average annual pounds applied; average annual area treated, and average 95
and 99th percentile and maximum application rates for only two periods from 1994 to 1998 and
from 1999 to 2006. Therefore, it was necessary to recalculate this data to reflect the previously
identified four historical usage periods (Figure 2.3).  A summary of this recalculated data is
provided in Table 2.6 for the  1st and 2nd periods and Table 2.7 for the 4th and 5th periods. Details
are provided only for currently registered endosulfan uses.
th
   Table 2.6 Historical usage of endosulfan for 1994 (1st period) and 1995-1997 (2na period)
Crop Use Pattern1
1st Period: 1994
Total
Lbs a.i.
%
Rate (Ibs a.i./A)
Average
Maximum
2nd Period: 1995-1997 Averages For:
Total
Lbs a.i
%
Rate (Ibs a.i./ A)
Average
Maximum
                                           43

-------
    Table 2.6 Historical usage of endosulfan for 1994 (1st period) and 1995-1997 (2na period)
Crop Use Pattern1
Nut trees
Citrus (Oranges
only)
Cole crops 1
Cole crops 2
Corn
Cotton
Apple
Fruit trees
Lettuce & Brussels
sprouts
Cucurbits2
Eggplant
Ornamentals or
shade trees
Potato (No sweet
potatoes)
Dry Beans/Peas &
Pepper
Carrot
Celery
Strawberry
Tomato
Currently Labeled
Total
Not currently labeled
Total2
All Crops
1st Period: 1994
Total
Lbs a.i.
1,430
0
6,145
28
5,528
142,259
656
3,582
37,570
189,485
213
433
3,697
9,278
726
557
540
10,145
412,289
64,760
477,049
%
0.35%
0.00%
1%
0.01%
1%
35%
0.16%
1%
9%
46%
0.05%
0.11%
1%
2%
0.18%
0.14%
0.13%
2%
100%
Rate flbs O.I/A)
Average
0.79

0.22
0.06
0.61
0.88
0.54
0.81
0.81
0.57
0.41
0.59
0.55
0.35
0.35
0.5
0.59
0.67
0.55
0.29
0.52
Maximum
1.72

1.01
0.50
1.51
1.10
1.50
3.0013
1.28
2.64
0.54
3.00
0.89
2.00
0.75
0.75
2.00
2.39
2.64
1.84
2.64
2nd Period: 1995-1997 Averages For:
Total
Lbs a.i.
2,480
o
3
7,181
311
3,199
81,639
1,122
2,004
31,687
40,125
88
307
1,808
5,996
819
482
283
16,640
196,190
38,373
234,563
%
1%
0.00%
4%
0.16%
2%
42%
1%
1%
16%
20%
0.04%
0.16%
1%
3%
0.42%
0.25%
0.14%
8%
100%
Rate flbs a.l/A)
Average
1.52
0.75
0.71
0.76
0.82
0.79
1.13
1.27
0.9
0.6
0.45
0.64
0.76
0.61
0.88
0.75
0.86
0.64
0.82
0.32
0.76
Maximum
1.99
0.75
1.7
0.94
1.19
0.93
2.13
2.5
1
2.15
0.72
2.82
0.96
1.32
1.04
1
1.5
1.07
2.82
1.54
2.82
^^^^^^a
    1 Nut Trees= Almond, Hazelnut & Walnut; Cole Crops 1= Broccoli, Cabbage & Cauliflower; Cole Crops 2=
    Collards, Kale & Mustard Green; Fruit Trees= Apricot, Nectarine, Peach, Cherry, Pear, Plum & Prune; and
    Cucurbits= Cucumber, Melons, Pumpkin & Squash
    2 Crops not currently labeled were 14% for the 1st period of which 98% Grapes, Alfalfa & Sugar beet; and 16%
    for the 2nd period of which 95% Grapes, Alfalfa & Sugar beet. Other crops include artichoke, spinach, pecan,
    and canola
Table 2.7 Historical usage of endosulfan for 1998-2004 (3rd period) and 2005-2006 (4th period)
Crop Use Pattern1
Nut trees
Citrus (Oranges
3rd Period: 1998-2004
Total
Lbs a.i.
530
37
%
0.47%
0.03%
Rate flbs a.i./A)
Avg
0.96
1.87
Max
1.58
3.00
4* Period: 2005-2006 Averages For:
Total
Lbs a.i.
41
0

%
0.06%
0.00%
Rate flbs a.i./A)
Avg
0.42
0
Max
0.79
0
95*%
0.82
0
99*%
1.5
0
  This rate of 3.00 Ibs a.i/A (reported in Tables 2.5 and 2.6) donates possible misreporting. This value is excluded
from calculated statistics because it contains one or more possible misreporting. Misreporting is assumed on a case
by case basis but it is considered when calculated rates are less than 0.1, less than 1 Ib is applied, less than 1 acre is
treated, area units missing or questionable, and number of records for pounds and area do not match.
                                                   44

-------
Table 2.7 Historical usage of endosulfan for 1998-2004 (3rd period) and 2005-2006 (4th period)
Crop Use Pattern1
only)
Cole crops 1
Cole crops 2
Corn
Cotton
Apple
Fruit trees
Lettuce &
Brussels sprouts
Cucurbits2
Eggplant
Ornamentals or
shade trees
Potato (No sweet
potatoes)
Dry Beans/Peas &
Pepper
Carrot
Celery
Strawberry
Tomato

Crops Not
currently labeled2
All Crops
3rd Period: 1998-2004
Total.
Lbs a.i.

3,748
80
1,027
40,335
395
2,520
25,179
14,332
44
248
1,158
2,703
387
108
83
19,842
112,755
41,887
154,642
%

3%
0.07%
1%
36%
0.35%
2%
22%
13%
0.04%
0.22%
1%
2%
0.34%
0.10%
0.07%
18%
100%
Rate flbs O.I./A)
Avg

0.85
0.9
1.07
0.89
1.26
1.33
0.92
0.92
0.73
1.11
0.92
0.78
0.85
0.67
1.14
0.91
1.00
0.75
0.99
Max

1.48
1.19
1.45
1.35
1.79
2.50
1.25
3.00
0.93
2.94
1.00
1.00
0.93
1.17
2.00
1.40
2.94
3.00
2.94
4* Period: 2005-2006 Averages For:
Total
Lbs a.i.

764
1
1,143
17,977
84
281
27,031
7,787
41
53
643
2,471
59
73
2,128
13,343
73,920
13,672
87,592

%

1%
0.00%
2%
24%
0.11%
0.38%
37%
11%
0.06%
0.07%
1%
3%
0.08%
0.10%
3%
18%
100%
Rate flbs a.i/A)
Avg

0.69
0.23
0.93
0.83
1.98
1.54
0.38
0.86
0.5
0.29
0.95
0.85
0.93
0.39
1.2
0.85
0.77
0.93
0.78
Max

0.82
0.79
0.93
0.83
1.98
2.06
0.72
0.98
0.5
0.55
0.95
0.89
0.93
0.74
1.2
0.85
2.06
1.38
2.06
95* %

0.83
0.9
0.93
0.83
1.98
2.08
0.75
1
0.5
0.57
0.95
0.89
0.93
0.77
1.2
0.85
2.00
1.46
1.99
99*%

0.98
0.93
1
1.01
2
2.5
0.99
1
0.55
1
0.96
0.99
0.93
0.77
1.51
1.01
2.42
1.48
2.41
1 Nut Trees= Almond, Hazelnut & Walnut; Cole Crops 1= Broccoli, Cabbage & Cauliflower; Cole Crops 2=
Collards, Kale & Mustard Green; Fruit Trees= Apricot, Nectarine, Peach, Cherry, Pear, Plum & Prune; and
Cucurbits= Cucumber, Melons, Pumpkin & Squash
2 Crops not currently labeled were 27% for the 3rd period of which 97% Alfalfa, Grapes & Sugar beet; and 16% for
the 4th period of which 98% Alfalfa, Grapes & Pecans. Other crops include artichoke, spinach, safflower, canola,
Garlic & Oats
Based on currently registered crops, California usage data can be summarized as follows:

    (1) Most recent usage data suggest that the top five use patterns for endosulfan are lettuce,
       cotton, tomatoes, cucurbits, and dry beans (Table 2.7, 2005-2006 data). The same five
       crops are also the dominant usage for earlier years with cotton being on the top for most
       of the times followed by cucurbits, lettuce, tomatoes, and dry beans (Table 2.6 and Table
       2.7, 1994, 1995-1997 and 1998-2004 data).

    (2) The top five crops constitute between 90 to 94% of the current labeled crop usage,
       therefore special distribution of these five crops can be taken to reflect current most
       important use areas of endosulfan.

    (3) Statistical comparison between current maximum rates for currently registered crops and
       the most recent PUR data suggests that reported actual use rates (i.e., typical rates) are
                                             45

-------
       consistently in the low side. Current labeled rates (Ibs a.i./A) compared to reported actual
       use rates are as follows: average 1.83 compared to 0.77 Ibs a.i./A; maximum 2.50
       compared to 2.06 Ibs a.i./A, ; 95th % 2.55 compared to 2.00 Ibs a.i./A, and 99th %
       maximum 2.91compared to 2.42 Ibs a.i./A. With few exceptions14, current labeled rates
       are generally higher than the most recent typical use rates.

    Figure 2.4 provides a summary of the most recent endosulfan usage along with the general
    habitat areas of the CRLF and the San Francisco Bay species. Figure 2.4 shows clearly that
    the majority (nearly 82%) of the recent endosulfan use was in three counties: Fresno, Kings,
    and Imperial. Top ten use category constitutes nearly 97% of the use which includes the
    three counties just mentioned in addition to Riverside, Kern, Yolo, Siskiyou, Tulare, and
    Solano counties. In an additional 14 counties, (San Benito, Madera, Monterey, Colusa,
    Merced, Santa Clara, San Joaquin, Ventura, Sonoma, Los Angeles, Placer, Napa, Glenn, and
    Santa Barbar).usage is very small ranging from 1 Ib a.i to 647 Ibs a.. Maps of species
    occurrence and their designated critical habitat (where applicable) in relation to agricultural
    areas in California are provided in Appendix E.
14 Statistics exclude data where reported rates are exceptionally high due to possible misreporting. Any rate reported
in Tables 2.5 and 2.6 to equal 3 Ibs a.i. /A are excluded because it contains one or more possible misreporting.
Misreporting is assumed on a case by case basis but is considered when calculated rates are less than 0.1, less than 1
Ib is applied, less than 1 acre is treated, area units missing or questionable and number of records for pounds and
area do not match.
                                             46

-------
                                                  Endosulfan Related Crops
                                                  Top 5 Crop Usage Patterns "

                                                         (OOO's Acres)

                                                         10-50

                                                         50-100

                                                         100-150

                                                         150-200
                                                         200 - 378
                                                              CRLF Habitat
                                                              Sacramento Valley
                                                              Central Coast
                                                              Central Valley (San Joaquin)
                                                              Southern Coast
                                                              Southern Desert Valley
       Endosulfan Use
       Ibs a.i/County (County Name(s)% of total Ibs a.i used 2005^06
           <650 (Other" 3%)
           650-1,300 (Sutler 1% & Solano 1%)
        H  1,300-2,500 (Yolo 2%, Slskiyou 2% & Tulare 2%)
        +  2,500-3,000 (Riverside 3% & Kern 3%)
        +  7,005 (Imperial 8%)
        +  7,500-20,000 ((Kings 22%)
        +  44,938 (Fresno 51%)
Figure 2.4 Endosulfan usage and acreage for the top five crop use patterns (CA-PUR data; Average for 2005-
2006)

Finally, the top usage crops in the three top ranked usage counties are lettuce, tomatoes,
cucurbits, dry beans, and corn for Fresno County, cotton, tomatoes, and cole crops for Kings
County,  and cucurbits, cotton, and lettuce for Imperial County (Figure 2.5). Alfalfa was the major
crop used in the "not currently registered category".
                                              47

-------
                                      Fresno  • Kings  D Imperial
                                            Crop
                         (Notes: Not currently registered crop was mainly alfalfa)
Figure 2.5 Distribution of endosulfan usage for the top usage counties of Fresno, Kings, and Imperial

    2.5 Assessed Species

Table 2.8 provides a summary of the current distribution, habitat requirements, and life history
parameters for the listed species being assessed.  More detailed life-history and distribution
information can be found in Attachments 1 and 3.
                                             48

-------
Table 2.8 Summary of Current Distribution, Habitat Requirements, and Life History Information for the Assessed Listed Species
Assessed
Species1

California red-
legged frog
(Rana aurora
draytonif)






California tiger
salamander
(Ambystoma
californiense)







San Francisco
garter snake
(Thamnophis
sirtalis
tetrataenia)







Salt marsh harvest
mouse
Size


Adult
(85-138 cm
in length),
Females -
9-238 g,
Males -
13-163 g;
Juveniles
(40-84 cm
in length)
50 g










Adult
(46-13 1cm
in length),
Females -
227 g,
Males -
H3g;
Juveniles
(18-20 cm
in length)


Adult
8-14g
Current Range


Northern CA coast, northern
Transverse Ranges, foothills of
Sierra Nevada, and in southern CA
south of Santa Barbara






There are two distinct population
segments; one in Santa Barbara
County and the other in Sonoma
County.







San Mateo County











Northern subspecies can be found
in Marin, Sonoma, Napa, Solano,
Habitat Type


Freshwater perennial
or near-perennial
aquatic habitat with
dense vegetation;
artificial
impoundments;
riparian and upland
areas


Freshwater pools or
ponds (natural or
man-made, vernal
pools, ranch stock
ponds, other fishless
ponds); Grassland or
oak savannah
communities, in low
foothill regions;
Small mammal
burrows
Densely vegetated
freshwater ponds
near open grassy
hillsides; emergent
vegetation; rodent
burrows






Dense, perennial
cover with preference
Designated
Critical
Habitat?
Yes









Yes










No











No

Reproductive
Cycle

Breeding: Nov. to Apr.
Tadpoles: Dec. to Mar.
Young juveniles: Mar. to
Sept.






Emerge from burrows and
breed: fall and winter
rains
Eggs: laid in pond Dec. -
Feb., hatch: after 10 to 14
days
Larval stage: 3-6 months,
until the ponds dry out,
metamorphose late spring
or early summer, migrate
to small mammal burrows
Oviparous Reproduction3
Breeding: Spring (Mar.
and Apr.) and Fall (Sept.
to Nov.)
Ovulation and Pregnancy:
Late spring and early
summer
Young: Born 3-4 months
after mating



Breeding: March -
November
Diet


Aquatic -phase2: algae,
freshwater aquatic
invertebrates
Terrestrial-phase:
aquatic and terrestrial
invertebrates, small
mammals, fish and
frogs


Aquatic Phase: algae,
snails, zooplankton,
small crustaceans, and
aquatic larvae and
invertebrates, smaller
tadpoles of Pacific tree
frogs, CRLF, toads;
Terrestrial Phase:
terrestrial invertebrates,
insects, frogs, and
worms
Juveniles: frogs
(Pacific tree frog,
CRLF, and bullfrogs
depending on size) and
insects
Adults: primarily frogs
(mainly CRLFs; also
bullfrogs, toads); to a
lesser extent newts;
freshwater fish and
invertebrates; insects
and small mammals
Leaves, seeds, and
plant stems; may eat
49

-------
Table 2.8 Summary of Current Distribution, Habitat Requirements, and Life History Information for the Assessed Listed Species
Assessed
Species1

(Reithrodontomys
raviventris)








San Joaquin kit
fox
(Vulpes macro tis
mutica)






















Size












Adult
~2kg
























Current Range


and northern Contra Costa
counties. The southern subspecies
occurs in San Mateo, Alameda,
and Santa Clara counties with
some isolation populations in
Marin and Contra Costa counties.




Alameda, Contra Costa, Fresno,
Kern, Kings, Madera, Merced,
Monterey, San Benito, San
Joaquin, San Luis Obispo, Santa
Barbara, Santa Clara, Stanislaus,
Tulare and Ventura counties




















Habitat Type


for habitat in the
middle and upper
parts of the marsh
dominated by
pickleweed and
peripheral halophytes
as well as similar
vegetation in diked
wetlands adjacent to
the Bay
A variety of habitats,
including grasslands,
scrublands (e.g.,
chenopod scrub and
sub-shrub scrub),
vernal pool areas, oak
woodland, alkali
meadows and playas,
and an agricultural
matrix of row crops,
irrigated pastures,
orchards, vineyards,
and grazed annual
grasslands. Kit foxes
dig their own dens,
modify and use those
already constructed
by other animals
(ground squirrels,
badgers, and
coyotes), or use
human-made
structures .(culverts,
abandoned pipelines,
or banks in sumps or
roadbeds). They
Designated
Critical
Habitat?










No, but has
designated
core areas























Reproductive
Cycle

Gestation period: 21 - 24
days








Mating and conception:
late December - March.
Gestation period: 48 to 52
days.
Litters born: February -
late March

Pups emerge from their
dens at about 1 -month of
age and may begin to
disperse after 4-5
months usually in Aug. or
Sept.













Diet


insects; prefers "fresh
green grasses" in the
winter and pickleweed
and saltgrass during the
rest of the year; drinks
both salt and fresh
water



Small animals
including blacktailed
hares, desert
cottontails, mice,
kangaroo rats, squirrels,
birds, lizards, insects
and grass. It satisfies its
moisture requirements
from prey and does not
depend on freshwater
sources.















50

-------
Table 2.8 Summary of Current Distribution, Habitat Requirements, and Life History Information for the Assessed Listed Species
Assessed
Species1






Bay checkerspot
butterfly (BCB)
(Euphydryas
editha bayensis)















Valley elderberry
longhorn beetle
(Desmocerus
californicus
dimorphus)







Size







Adult
butterfly - 5
cm in length
















Males:
1.25-2.5 cm
length
Females:
1.9-2.5 cm
length






Current Range







Santa Clara and San Mateo
Counties [Because the BCB
distribution is considered a
metapopulation, any site with
appropriate habitat in the vicinity
of its historic range (Alameda,
Contra Costa, San Francisco, San
Mateo, and Santa Clara counties)
should be considered potentially
occupied by the butterfly (USFWS
1998, p. 11-177)].








Central Valley of California (from
Shasta County to Fresno County in
the San Joaquin Valley)









Habitat Type


move to new dens
within their home
range often (likely to
avoid predation by
coyotes)
1) Primary habitat -
native grasslands on
large serpentine
outcrops;
2) Secondary habitat
- 'islands' of smaller
serpentine outcrops
with native grassland;
3) Tertiary habitat -
non-serpentine areas
where larval food
plants occur







Completely
dependent on its host
plant, elderberry
(Sambucus species),
which is a common
component of the
remaining riparian
forests and adjacent
upland habitats of
California's Central
Valley

Designated
Critical
Habitat?





Yes


















Yes











Reproductive
Cycle






Larvae hatch in March -
May and grow to the 4th
instar in about two weeks.
The larvae enter into a
period of dormancy
(diapause) that lasts
through the summer. The
larvae resume activity
with the start of the rainy
season. Larvae pupate
once they reach a weight
of 300 -500 milligrams.
Adults emerge within 15
to 30 days depending on
thermal conditions, feed
on nectar, mate and lay
eggs during a flight
season lasting 4 -6 wksl
from late Feb.- early May
The larval stage may last
2 years living within the
stems of an elderberry
plant. Then larvae enter
the pupal stage and
transform into adults.
Adults emerge and are
active from March to June
feeding and mating, when
the elderberry produces
flowers.

Diet







Obligate with dwarf
plantain. Primary diet
is dwarf plantain plants
(may also feed on
purple owl's-clover or
exserted paintbrush if
the dwarf plantains
senesce before the
larvae pupate). Adults
feed on the nectar of a
variety of plants found
in association with
serpentine grasslands






Obligates with
elderberry trees
(Sambucus sp). Adults
eat the elderberry
foliage until about June
when they mate. Upon
hatching the larvae
tunnel into the tree
where they will spend
1-2 years eating the
interior wood which is
their sole food source.
51

-------
1  For more detailed information on the distribution, habitat requirements, and life history information of the assessed listed species, see Attachment 3
2  For the purposes of this assessment, tadpoles and submerged adult frogs are considered "aquatic" because exposure pathways in the water are considerably
different than those that occur on land.
3  Oviparous = eggs hatch within the female's body and young are born live.
                                                                        52

-------
2.6 Designated Critical Habitat
Critical habitat has been designated for the CRLF, BCB, CIS, and VELB. 'Critical
habitat' is defined in the ESA as the geographic area occupied by the species at the time
of the listing where the physical and biological features necessary for the conservation of
the species exist, and there is a need for special management to protect the listed species.
It may also include areas outside the occupied area at the time of listing if such areas are
'essential to the conservation of the species.' Critical habitat receives protection under
Section 7 of the ESA through prohibition against destruction or adverse modification
with regard to actions carried out, funded, or authorized by a federal Agency.  Section 7
requires consultation on federal actions that are likely to result in the destruction or
adverse modification of critical habitat.

To be included in a critical habitat designation, the habitat must be 'essential to the
conservation of the species.' Critical habitat designations identify, to the extent known
using the best  scientific and commercial data available, habitat areas that provide
essential life cycle needs of the species or areas that contain certain primary constituent
elements (PCEs) (as  defined in 50 CFR 414.12(b)). PCEs include, but are not limited to,
space for individual and population growth and for normal behavior; food, water, air,
light, minerals, or other nutritional or physiological requirements; cover or shelter; sites
for breeding, reproduction, rearing (or development) of offspring; and habitats that are
protected from disturbance or are representative of the historic geographical and
ecological distributions of a species. Table 2.9 describes the PCEs for the critical habitats
designated for the CRLF, CIS, BCB, and VELB.
Table 2.9 Designated Critical Habitat PCEs for the CRLF, CTS, BCB and VELB
Species
CRLF
PCEs
Alteration of channel/pond morphology or geometry and/or increase
in sediment deposition within the stream channel or pond.
Alteration in water chemistry /quality including temperature,
turbidity, and oxygen content necessary for normal growth and
viability of juvenile and adult CRLFs and their food source.
Alteration of other chemical characteristics necessary for normal
growth and viability of CRLFs and their food source.
Reduction and/or modification of aquatic-based food sources for pre-
metamorphs (e.g., algae)
Elimination and/or disturbance of upland habitat; ability of habitat to
support food source of CRLFs: Upland areas within 200 ft of the
edge of the riparian vegetation or dripline surrounding aquatic and
riparian habitat that are comprised of grasslands, woodlands, and/or
wetland/riparian plant species that provides the CRLF shelter,
forage, and predator avoidance
Elimination and/or disturbance of dispersal habitat: Upland or
riparian dispersal habitat within designated units and between
occupied locations within 0.7 mi of each other that allow for
movement between sites including both natural and altered sites
which do not contain barriers to dispersal
Reduction and/or modification of food sources for terrestrial phase
juveniles and adults
Alteration of chemical characteristics necessary for normal growth
Reference
50CFR414.12(b),
2006
                                        53

-------
Table 2.9 Designated Critical Habitat PCEs for the CRLF, CTS, BCB and VELB
Species

California tiger
salamander
Bay
Checkerspot
Butterfly
Valley
Elderberry
Longhorn
Beetle
PCEs
and viability of juvenile and adult CRLFs and their food source.
Standing bodies of fresh water, including natural and man-made
(e.g., stock) ponds, vernal pools, and dune ponds, and other
ephemeral or permanent water bodies that typically become
inundated during winter rains and hold water for a sufficient length
of time (i.e., 12 weeks) necessary for the species to complete the
aquatic (egg and larval) portion of its life cycle2
Barrier-free uplands adjacent to breeding ponds that contain small
mammal burrows. Small mammals are essential in creating the
underground habitat that juvenile and adult California tiger
salamanders depend upon for food, shelter, and protection from the
elements and predation
Upland areas between breeding locations (PCE 1) and areas with
small mammal burrows (PCE 2) that allow for dispersal among such
sites
The presence of annual or perennial grasslands with little to no
overstory that provide north/south and east/west slopes with a tilt of
more than 7 degrees for larval host plant survival during periods
of atypical weather (e.g., drought).
The presence of the primary larval host plant, dwarf plantain
(Plantago erecta) (a dicot) and at least one of the secondary host
plants, purple owl's-clover or exserted paintbrush, are required for
reproduction, feeding, and larval development.
The presence of adult nectar sources for feeding.
Aquatic features such as wetlands, springs, seeps, streams, lakes, and
ponds and their associated banks, that provide moisture during
periods of spring drought; these features can be ephemeral, seasonal,
or permanent.
Soils derived from serpentinite ultramafic rock (Montara, Climara,
Henneke, Hentine, and Obispo soil series) or similar soils
(Inks, Candlestick, Los Gatos, Pagan, and Barnabe soil series)
that provide areas with fewer aggressive, normative plant species for
larval host plant and adult nectar plant survival and reproduction.2
The presence of stable holes and cracks in the soil, and surface rock
outcrops that provide shelter for the larval stage of the bay
checkerspot butterfly during summer diapause.2
Areas that contain the host plant of this species [/'. e., elderberry trees
(Sambucus sp.)] (a dicot)
Reference

FR Vol. 69 No. 226
CTS, 68584, 2004
66 FR 21449 21489,
2001
43 FR 35636 35643,
1978
1 These PCEs are in addition to more general requirements for habitat areas that provide essential life cycle
needs of the species such as, space for individual and population growth and for normal behavior; food,
water, air, light, minerals, or other nutritional or physiological requirements; cover or shelter; sites for
breeding, reproduction, rearing (or development) of offspring; and habitats that are protected from
disturbance or are representative of the historic geographical and ecological distributions of a species.
2 PCEs that are abiotic, including, physico-chemical water quality parameters such as salinity, pH, and
hardness are not evaluated because these processes are not biologically mediated and, therefore, are not
relevant to the endpoints included in this assessment.

More detail on the designated critical habitat applicable to this assessment can be found
in Attachment 1  (for the CRLF) and Attachment 3 (for CTS, BCB, and VELB).
Activities that  may destroy or adversely modify critical habitat are those that alter the
                                              54

-------
PCEs and jeopardize the continued existence of the species. Evaluation of actions related
to use of endosulfan that may alter the PCEs of the designated critical habitat for the
CRLF, CTS, BCB, and VELB form the basis of the critical habitat impact analysis.

As previously noted in Section 2.1, the Agency believes that the analysis of direct and
indirect effects to listed species provides the basis for an analysis of potential effects on
the designated critical habitat. Because endosulfan is expected to directly impact living
organisms within the action area, critical habitat analysis for endosulfan is limited in a
practical sense to those PCEs of critical habitat that are biological or that  can be
reasonably linked to biologically mediated processes.

2.7 Action Area

For listed species assessment purposes, the action area is considered to be the area
affected directly or indirectly by the federal action and not merely the immediate area
involved in the action (50 CFR 402.02). It is recognized that the overall action area for
the national registration of endosulfan is likely to encompass considerable portions of the
United States based on the large array of agricultural uses.  However, the scope of this
assessment limits consideration of the overall action area to those portions that may be
applicable to the protection of the CRLF, CTS, SFGS, SMHM, SJKF, BCB, and VELB
and their designated critical habitat within the state of California. Although the
watershed for the San Francisco Bay extends northward into the very southwestern
portion of Lake County, Oregon, and westward into the western edge of Washoe County,
Nevada, the non-California portions of the watershed are small and very rural with little,
if any, agriculture. Therefore, no use of endosulfan is expected in these areas, and they
are not considered as part of the action area applicable to this assessment.

The definition of action area requires a stepwise approach that begins with an
understanding of the  federal  action. The federal action is defined by the currently labeled
uses for endosulfan.  An analysis of labeled uses and review of available product labels
was completed.  Several of the currently labeled uses are special  local needs (SLN) uses
or are restricted to specific states and are excluded from this assessment.  In addition, a
distinction has been made between food use crops and those that are non-food/non-
agricultural uses.  For those uses relevant to the assessed species, the analysis indicates
that, for endosulfan, the following agricultural uses are considered as part of the federal
action evaluated in this assessment:

       (1)  Terrestrial Food Crops
       Apricot, Cherry, Nectarine, Peach, Pear, Plum & Prune
       Cabbage & Cauliflower
       Carrot & Celery
       Collards& Kale
       Cucurbits (Cucumber, Melons, Pumpkin & Squash)
       Hazelnuts & Walnuts
       Lettuce & Brussels Sprouts
       Pepper
                                        55

-------
       Strawberry

       (2) Terrestrial + Greenhouse Food Crops
       Tomato

       (3) Terrestrial Food + Feed Crops
       Almonds
       Apple
       Beans & Peas (dry)
       Broccoli
       Cotton
       Eggplant
       Mustard Green
       Potato & Sweet Potato
       Sweet corn for fresh market only

       (4) Terrestrial Non-Food Crops
       Citrus
       Ornamentals or Shade Trees

Following a determination of the assessed uses, an evaluation of the potential "footprint"
of endosulfan use patterns (i.e., the area where pesticide application occurs) is
determined.  This "footprint" represents the initial area of concern, based on an analysis
of available land cover data for the state of California.  The initial area of concern is
defined as all land cover types and the stream reaches within the land cover areas that
represent the labeled uses described above. A map representing all the land cover types
that make up the initial area of concern for endosulfan is presented in Figure 2.6.
                                        56

-------
                 Endosulfan Use - Initial Area of Concern
           Orchard vineyard use
           Cultivated crop use
           County boundaries
                   i Kilometers
     0 2040   SO  120 1 BO
    Compiled from California County boundaries (ESRI, 2002),
    USQA, Gap Analysis Program Orchard/Vineyard Landcover (GAP)
    National Land Cover Database (NLCD) (MRLC, 2001)
    Map created by US Environmental Protection Agency, Office
    of Pesticides Programs, Environmental Fate and Effects Division.
    Projection: Alters Equal Area Conic US OS. North American
    Datum of 1983 (NAD 1 983).                                                       4/9/2009
Figure 2.6 Initial area of concern, or "footprint" of potential use, for endosulfan

Once the initial area of concern is defined, the next step is to define the potential
boundaries of the action area by determining the extent of offsite transport via spray drift
and runoff where exposure of one or more taxonomic groups to the pesticide exceeds the
listed species LOCs.

The Agency's approach to defining the action area under the provisions of the Overview
Document (USEPA 2004) considers  the results of the risk assessment process to establish
                                           57

-------
boundaries for that action area with the understanding that exposures below the Agency's
defined Levels of Concern (LOCs) constitute a no-effect threshold.  Deriving the
geographical extent of this portion of the action area is based on consideration of the
types of effects that endosulfan may be expected to have on the environment, the
exposure levels to endosulfan that are associated with those effects,  and the best available
information concerning the use of endosulfan and its fate and transport within the state of
California. Specific measures of ecological effect for the assessed species that define the
action area include any direct and indirect toxic effect to the assessed species and any
potential modification of its critical habitat, including reduction in survival, growth, and
fecundity as well as the full suite of sublethal effects available in the effects literature.
Therefore, the action area extends to a point where environmental exposures are below
any measured lethal or sublethal effect threshold for any biological entity at the whole
organism, organ, tissue, and cellular level of organization.  In situations where it is not
possible to determine the threshold for an observed effect, the action area is not spatially
limited and is assumed to be the entire state of California.

Based on a review of the available toxicity data in ECOTOX and toxicity studies
submitted to the Agency, several studies were identified that lacked a defined NOAEC
(i.e., a concentration were no observable adverse effects were documented; see Appendix
J).  For example, Hansen and Cripe (1991; ECOTOX # 115297) reported a LOAEC of <
0.27 ug a.i./L for sheepshead minnow resulting from 28-d, flow through exposures. Due
to the lack of a defined NOAEC (i.e., no observed adverse effect concentration), the
spatial  extent of the action area (i.e., the boundary where exposures  and potential effects
are less than the Agency's LOG) cannot be determined for endosulfan.  Therefore, it is
assumed that the action area encompasses the entire state of California, regardless of the
spatial  extent (i.e., initial area of concern or footprint) of the pesticide use(s).

2.8 Assessment Endpoints and Measures of Ecological Effect

Assessment endpoints are defined as "explicit expressions of the actual environmental
value that is to be protected."15  Selection of the assessment endpoints is based on valued
entities (e.g., CRLF, CTS, SFGS, SMHM, SJKF, BCB, and VELB), organisms important
in the life cycle of the assessed species, and the PCEs of its designated  critical habitat),
the ecosystems potentially at risk (e.g., waterbodies, riparian vegetation, and upland and
dispersal habitats), the migration pathways of endosulfan (e.g., runoff, spray drift, etc.),
and the routes by which ecological receptors are exposed to endosulfan (e.g., direct
contact, etc.).

2.8.1 Assessment Endpoints

Assessment endpoints for the CRLF, CTS, SFGS, SMHM, SJKF, BCB, and VELB
include direct toxic  effects on the survival, reproduction, and growth of individuals, as
well as indirect effects, such as reduction of the prey base or modification of its habitat.
In addition, potential modification of critical habitat is assessed by evaluating potential
effects to PCEs, which are components of the habitat areas that provide essential life
 ; From U.S. EPA (1992). Framework for Ecological Risk Assessment. EPA/630/R-92/001.
                                        58

-------
cycle needs of the assessed species.  Each assessment endpoint requires one or more
"measures of ecological effect," defined as changes in the attributes of an assessment
endpoint or changes in a surrogate entity or attribute in response to exposure to a
pesticide. Specific measures of ecological effect are generally evaluated based on acute
and chronic toxicity information from registrant-submitted guideline tests that are
performed on a limited number of organisms. Additional ecological effects data from the
open literature are also considered. It should be noted that assessment endpoints are
limited to direct and indirect effects associated with survival, growth, and fecundity, and
do not include the full suite of sublethal effects used to define the action area.  According
to the Overview Document (USEPA 2004), the Agency relies on acute and chronic
effects endpoints that are either direct measures of impairment of survival, growth, or
fecundity or endpoints for which there is a scientifically robust, peer reviewed
relationship that can quantify the impact of the measured effect endpoint on the
assessment endpoints of survival, growth, and fecundity. A complete discussion of all
the toxicity data available for this risk assessment, including resulting measures of
ecological effect selected for each taxonomic group of concern, is included in Section 4
of this document.

As described in the Agency's Overview Document (U.S. EPA, 2004), the most sensitive
endpoint for each taxon is used for risk estimation.  For this assessment, evaluated taxa
include aquatic-phase amphibians, freshwater fish, freshwater invertebrates, aquatic
plants, birds (surrogate for terrestrial-phase amphibians and reptiles), mammals,
terrestrial invertebrates, and terrestrial plants. Acute (short-term) and chronic (long-term)
toxicity information is characterized based on registrant-submitted studies and a
comprehensive review of the open literature on endosulfan.

Table 2.10 identifies the taxa used to assess the potential for direct and indirect effects
from the  uses of endosulfan for each listed species assessed here. The specific
assessment endpoints used to assess the potential for direct and indirect effects to each
listed species are provided in Table 2.11.
                                        59

-------
Table 2.10 Taxa Used in the Analyses of Direct and Indirect Effects of Endosulfan for the Assessed Listed Species.
Listed Species
California red-
legged frog
California
tiger
salamander
San Francisco
garter snake
Salt marsh
harvest mouse
San Joaquin
kit fox
Bay
checkerspot
butterfly
Valley
elderberry
longhorn
beetle
Birds
Direct
Indirect
(prey)
Direct
Direct
Indirect
(prey)
Indirect
(rearing
sites)
Indirect
(prey)
N/A
N/A
Mammals
Indirect
(prey)
N/A
Indirect
(prey)
Direct
Indirect
(rearing
sites)
Direct
Indirect
(prey)
N/A
N/A
Terr.
Plants
Indirect
(habitat)
Indirect
(habitat)
Indirect
(habitat)
Indirect
(food,
habitat)
Indirect
(food/
habitat)
Indirect
(food/
habitat)
*
Indirect
(food/
habitat)
*
Terr.
Inverts.
Indirect
(prey)
Indirect
(prey)
Indirect
(prey)
Indirect
(prey)
Indirect
(prey)
Direct
Direct
FW Fish
Direct
Indirect
(prey)
Direct
Indirect
(prey)
Indirect
(prey)
N/A
N/A
N/A
N/A
FW
Inverts.
Indirect
(prey)
Indirect
(prey)
Indirect
(prey)
N/A
N/A
N/A
N/A
Estuarine
/Marine
Fish
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Estuarine
/Marine
Inverts.
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Aquatic
Plants
Indirect
(food/
habitat)
Indirect
(food/
habitat)
Indirect
(habitat)
Indirect
(habitat)
N/A
N/A
N/A
N/A = Not applicable
Terr. = Terrestrial
Invert. = Invertebrate
FW = Freshwater
* = obligate relationship
                                                  60

-------
Table 2.11 Taxa and Assessment Endpoints Used to Evaluate the Potential for the Use of Endosulfan to Result in Direct
and Indirect Effects to the Assessed Listed Species.	
Taxa Used to Assess
Direct and/or Indirect
Effects to Assessed
Species	
Assessed Listed
Species
Assessment Endpoints
Measures of Ecological Effects
1. Freshwater Fish and
Aquatic-phase
Amphibians
Direct Effect -
-Aquatic-phase CRLF
- Aquatic-phase CTS
Survival, growth, and
reproduction of individuals
via direct effects
la.  Common carp acute LC50
Ib.  Common carp NOAEC (estimated
based on acute-chronic ratio)
                        Indirect Effect (prey')
                        -Aquatic-phase and
                        Terrestrial-phase CRLF
                        -SFGS
                        -CTS
                         Survival, growth, and
                         reproduction of individuals
                         via indirect effects on
                         aquatic prey food supply
                         (i.e., fish and aquatic-phase
                         amphibians)	
2. Freshwater
Invertebrates
Indirect Effect (prey)
-Aquatic-phase and
Terrestrial-phase CRLF
-SFGS
-CTS
Survival, growth, and
reproduction of individuals
via indirect effects on
aquatic prey food supply
(i.e., freshwater
invertebrates)	
2a.  Mayfly LC50
2b.  Mayfly NOAEC  (estimated based on
acute-chronic ratio)
3. Aquatic Plants
(freshwater/marine)
Indirect Effect
(food/habitat)
- Aquatic-phase CRLF
-SFGS
-SMHM
-CTS
Survival, growth, and
reproduction of individuals
via indirect effects on
habitat, cover, food supply,
and/or primary productivity
(i.e., aquatic plant
community)	
5a.  No data for Vascular aquatic plants
5b.  Green alga EC50
6. Birds
Direct Effect
-Terrestrial-phase CRLF
-SFGS
-CTS
Survival, growth, and
reproduction of individuals
via direct effects
6a.  Mallard LD50
6b.  Mallard NOAEC
                        Indirect Effect
                        (prey/habitat from nests)
                        -Terrestrial-phase CRLF
                        -SFGS
                        -SMHM
                        -SJKF
                         Survival, growth, and
                         reproduction of individuals
                         via indirect effects on
                         terrestrial prey (birds)
7. Mammals
Direct Effect
-SMHM
-SJKF
Survival, growth, and
reproduction of individuals
via direct effects
la.  Laboratory rat acute LD50
7b.  Laboratory rat chronic NOAEC
                        Indirect Effect
                        (prey/habitat from
                        burrows)
                        -Terrestrial-phase CRLF
                        -SFGS
                        -SMHM
                        -SJKF
                         Survival, growth, and
                         reproduction of individuals
                         via indirect effects on
                         terrestrial prey (mammals)
8. Terrestrial
Invertebrates
Direct Effect
-BCB
-VELB
Survival, growth, and
reproduction of individuals
via direct effects
8a. Beet webworm acute contact LD5
                        Indirect Effect (prey)
                         Survival, growth, and
                                                      61

-------
Table 2.11 Taxa and Assessment Endpoints Used to Evaluate the Potential for the Use of Endosulfan to Result in Direct
and Indirect Effects to the Assessed Listed Species.	
                      -Terrestrial-phase CRLF
                      - SFGS
                      -SMHM
                      -SJKF
                      -CTS
                      reproduction of individuals
                      via indirect effects on
                      terrestrial prey (terrestrial
                      invertebrates)
9. Terrestrial Plants
Indirect Effect
(food/habitat) (non-
obligate relationship)
-Terrestrial-phase CRLF
-SFGS
-SMHM
-SJKF
-CTS
Survival, growth, and
reproduction of individuals
via indirect effects on food
and habitat (i.e., riparian
and upland vegetation)
9a. No toxicity data for terrestrial plants
                      Indirect Effect
                      (food/habitat) (obligate
                      relationship)
                      -BCB
                      -VELB
       2.8.2 Assessment Endpoints for Designated Critical Habitat

       As previously discussed, designated critical habitat is assessed to evaluate actions related
       to the use of endosulfan that may alter the PCEs of the assessed species' designated
       critical habitat.  PCEs for the assessed species were previously described in Section 2.6.
       Actions that may affect critical habitat are those that alter the PCEs and jeopardize the
       continued existence of the assessed species. Therefore, these actions are identified as
       assessment endpoints.  It should be noted that evaluation of PCEs as assessment
       endpoints is limited to those of a biological nature (i.e., the biological resource
       requirements for the listed species associated with the critical habitat) and those for
       which endosulfan effects data are available.

       Some components of these PCEs are associated with physical abiotic features (e.g.,
       presence and/or depth of a water body, or distance between two sites), which are not
       expected to be measurably altered by use of pesticides.  Measures of ecological effect
       used to assess the potential for effects to the critical habitat of the CRLF, BCB, VELB,
       and  CTS are described in Table 2.12.
Table 2.12 Summary of Assessment Endpoints and Measures of Ecological Effect for Primary
Constituent Elements of Designated Critical Habitat for CRLF, BCB, VELB, and CTS.
Taxon Used to
Assess
Modification of
PCE
1 . Freshwater Fish
and Aquatic-phase
Amphibians
Assessed Listed
Species Associated
with the PCE
Direct Effect -
-Aquatic -phase
CRLF
Assessment
Endpoints
Survival, growth, and
reproduction of
individuals via direct
effects
Measures of Ecological Effects
la. Common carp acute LC50
Ib. Common carp NOAEC
(estimated based on acute-chronic
ratio)
                                                 62

-------
Table 2.12 Summary of Assessment Endpoints and Measures of Ecological Effect for Primary
Constituent Elements of Designated Critical Habitat for CRLF, BCB, VELB, and CTS.
Taxon Used to
Assess
Modification of
PCE

2. Freshwater
Invertebrates
5. Aquatic Plants
(freshwater/marine)
6. Birds
7. Mammals
8. Terrestrial
Invertebrates
9. Terrestrial Plants
Assessed Listed
Species Associated
with the PCE
Indirect Effect
(prey)
-Aquatic and
terrestrial-phase
CRLF
Indirect Effect
(prey)
-Aquatic -phase and
- Terrestrial-phase
CRLF
Indirect Effect
(food/habitat)
-Aquatic -phase
CRLF
-CTS
Direct Effect
-Terrestrial-phase
CRLF
Indirect Effect
(prey)
- Terrestrial-phase
CRLF
Indirect Effect
(prey/habitat from
burrows)
-Terrestrial-phase
CRLF
-CTS
Direct Effect
-BCB
-VELB
Indirect Effect
(prey)
-Terrestrial-phase ~
CRLF
-CTS
Indirect Effect
(food/habitat) (non-
obligate
relationship)
Assessment
Endpoints
Modification of critical
habitat via change in
aquatic prey food
supply (i.e., fish and
aquatic -phase
amphibians)
Survival, growth, and
reproduction of
individuals via indirect
effects on aquatic prey
food supply (i.e.,
freshwater
invertebrates)
Modification of critical
habitat via change in
habitat, cover, food
supply, and/or primary
productivity (i.e.,
aquatic plant
community)
Survival, growth, and
reproduction of
individuals via direct
effects
Modification of critical
habitat via change in
terrestrial prey (birds)
Modification of critical
habitat via change in
terrestrial prey
(mammals)
Survival, growth, and
reproduction of
individuals via direct
effects
Modification of critical
habitat via change in
terrestrial prey
(terrestrial
invertebrates)
Modification of critical
habitat via change in
food and habitat (i.e.,
riparian and upland
Measures of Ecological Effects

2a. Mayfly LC50
2b. Mayfly NOAEC (estimated
based on acute-chronic ratio)
5a. No data for Vascular aquatic
plants
5b. Green alga EC50
6a. Mallard LD50
6b. Mallard NOAEC
7a. Laboratory rat acute LD50
7b. Laboratory rat chronic
NOAEC
8a. Beet webworm acute contact
LD50
9a. No toxicity data for terrestrial
plants
63

-------
Table 2.12 Summary of Assessment Endpoints and Measures of Ecological Effect for Primary
Constituent Elements of Designated Critical Habitat for CRLF, BCB, VELB, and CTS.
Taxon Used to
Assess
Modification of
PCE

Assessed Listed
Species Associated
with the PCE
-Terrestrial-phase
CRLF
-CTS
Indirect Effect
(food/habitat)
(obligate
relationship)
-BCB
-VELB
Assessment
Endpoints
vegetation)
Measures of Ecological Effects

2.9 Conceptual Model

2.9.1 Risk Hypotheses

Risk hypotheses are specific assumptions about potential adverse effects (i.e., changes in
assessment endpoints) and may be based on theory and logic, empirical data,
mathematical models, or probability models (U.S. EPA, 1998). For this assessment, the
risk is stressor-linked, where the stressor is the release of endosulfan to the environment.
The following risk hypotheses are presumed in this assessment:

The labeled use of endosulfan within the action area may:

•      Directly affect the CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CTS by
causing mortality or by adversely affecting growth or fecundity;
•      Indirectly affect the CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CTS and/or
modify their designated critical habitat by reducing or changing the composition of food
supply;
•      Indirectly affect the CRLF, SFGS, SMHM, and CTS and/or modify their
designated critical habitat by reducing or changing the composition of the aquatic plant
community in the species' current range, thus affecting primary productivity and/or
cover;
•      Indirectly affect the CRLF,  SFGS, SMHM, BCB, VELB, SJKF, and CTS and/or
modify their designated critical habitat by reducing or changing the composition of the
terrestrial plant community in the species' current range;
•      Indirectly affect the CRLF SFGS, SMHM, SJKF, and CTS  and/or modify their
designated critical habitat by reducing or changing aquatic habitat in their current range
(via modification of water quality parameters, habitat morphology, and/or sedimentation).
                                       64

-------
2.9.2 Diagram

The conceptual model is a graphic representation of the structure of the risk assessment.
It specifies the endosulfan release mechanisms, biological receptor types, and effects
endpoints of potential concern. The conceptual models for aquatic and terrestrial phases
of the CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CTS and the conceptual models
for the aquatic and terrestrial PCE components of critical habitat are shown in Figure 2.7
and Figure 2.8. Although the conceptual models for direct/indirect effects and
modification of designated critical habitat PCEs are shown on the same diagrams, the
potential for direct/indirect effects and modification of PCEs will be evaluated separately
in this assessment. Exposure routes shown in dashed lines are considered insignificant
routes of exposure (e.g., groundwater seepage into surface water).
Stressor


Source
Exposure
Media
               |     a- and (3- Endosulfan Applied to Use Sites + Sulfate Formation      |

1 1
Spray drift 1 1 Runoff 1
11
Surface water/
Sediment
1
\
\ Soil | 	










T
Long range
atmospheric
transport


                 T
          Uptake/gills
          or integument
Receptors
Attribute
Change
                Uptake/gills
                or integument
              Aquatic Animals
              Invertebrates
              Vertebrates
                             Ingestion
 Fish/aquatic-
 phase amphibians
  Piscivorous
 mammals and
 birds   i
Individual
organisms
Reduced survival
Reduced growth
              Uptake/cell,
              roots, leaves
          Aquatic Plants
          Non-vascular
          Vascular
                                   Ingestion
Food chain
Reduction in algae
Reduction in prey
Modification of PCEs
related to prey
availability
Habitat integrity
Reduction in primary productivity
Reduced cover
Community change
Modification of PCEs related to
nabitat
Figure 2.7 Conceptual Model for Endosulfan Effects on the Assessed Species and
their Critical Habitat in the Aquatic Environment
                                         65

-------
Stressor
Source
Exposure
Media
a- and (3- Endosulfan Applied to Use Sites + Sulfate Formation

1
1


                           -I  Spray drift]	>  \ Runoff |
                          t ">—t ">—Dermal uptake/lngestion-^—
                                        Long range
                                        atmospheric
                                         transport
                         Terrestrial/riparian plants
                         grasses/forbs, fruit, seeds
                             (trees, shrubs)
                                        Root uptake.4!
                                                 .Wet/d1
                                       Ingestion
Receptors
 Attribute
 Change
 Birds/terrestrial-
 phase
 amphibians/
 reptilesfmammals
                 3s[rr
Individual
organisms
Reduced survival
Reduced growth
                                I
           deposition^.
Food chain
Reduction in prey
Modification of
PCEs related to
prey availability
Habitat integrity
Reduction in primary productivity
Reduced cover
 ommunity change
Modification of PCEs related to
nabitat
Figure 2.8. Conceptual Model for Endosulfan Effects on the Assessed Species and
their Critical Habitat in the Terrestrial Environment
2.10 Analysis Plan

In order to address the risk hypothesis, the potential for direct and indirect effects to the
CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CIS, prey items, and habitat is estimated
based on a taxon-level approach.  In the following sections, the use, environmental fate,
and ecological effects of endosulfan are characterized and integrated to assess the risks.
This is accomplished using a risk quotient (ratio of exposure concentration to effects
concentration) approach. Although risk is often defined as the likelihood and magnitude
of adverse ecological effects, the risk quotient-based approach does not provide a
quantitative estimate of likelihood and/or magnitude of an adverse effect.  However,  as
outlined in the Overview Document (U.S. EPA, 2004), the likelihood of effects to
individual organisms from  particular uses of endosulfan is estimated using the probit
dose-response slope and either the level of concern (discussed below) or actual calculated
risk quotient value.

2.10.1 Measures to Evaluate the Risk Hypothesis and Conceptual Model
                                         66

-------
2.10.1.1 Measures of Exposure

The environmental fate properties of endosulfan along with available monitoring data
indicate that runoff, spray drift, and long-range atmospheric transport are the principle
potential transport mechanisms of endosulfan to the aquatic and terrestrial habitats of the
CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CIS. In this assessment, transport of
endosulfan through runoff and spray drift is considered in deriving quantitative estimates
of CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CIS exposure to endosulfan , in
addition to exposure of their prey and habitats. Assessment of long-range atmospheric
transport is addressed qualitatively through a review of available monitoring data from
California but not explicitly modeled due to the lack of models that quantitatively predict
far-field pesticide concentrations resulting from near-field loadings.

Measures of exposure are based on aquatic and terrestrial models that predict estimated
environmental concentrations (EECs) of endosulfan using maximum labeled application
rates and methods of application.  The models used to predict aquatic EECs are the
Pesticide Root Zone Model coupled with the Exposure Analysis Model System
(PRZM/EXAMS). The model used to predict terrestrial EECs on food items is T-REX.
These models are parameterized using relevant reviewed registrant-submitted
environmental fate data.

PRZM (Carsel et al 1997) v3.12.2, May 2005 and EXAMS (Burns, 1997) v2.98.4.6,
April 2005 are screening simulation models coupled with the input shell pe5.pl (Aug
2007) to generate daily exposures and l-in-10 year EECs of endosulfan that may occur in
surface water bodies adjacent to application sites receiving endosulfan through runoff and
spray drift. PRZM simulates pesticide application, movement and transformation on an
agricultural field and the resultant pesticide loadings to a receiving water body via runoff,
erosion and spray drift. EXAMS simulates the fate of the pesticide and resulting
concentrations in the water body.  The standard scenario used for ecological pesticide
assessments  assumes application to a 10-hectare agricultural field that drains into an
adjacent 1-hectare water body, 2-meters deep (20,000 m3 volume) with no outlet.
PRZM/EXAMS was used to estimate screening-level exposure of aquatic organisms to
endosulfan.  The measure of exposure for aquatic species is the l-in-10 year return peak
or rolling mean concentration.  The l-in-10-year  60-day mean is used for assessing
chronic exposure to fish; the l-in-10-year 21-day mean is used for assessing chronic
exposure for aquatic invertebrates.

For endosulfan, aquatic exposures were modeled separately for the parent isomers (d and
P) and the primary degradate of concern (endosulfan sulfate). The resulting daily
concentrations were then summed to form a 30-year time series for total endosulfan (sum
of d, P and endosulfan sulfate). Due to similar acute toxicity and structure of the TGAI
and endosulfan sulfate, modeled or measured environmental exposures of total
endosulfan (sum  of d, P and endosulfan sulfate) were compared with toxicity data for the
TGAI.  For sediment-borne exposures, toxicity data were available only for endosulfan
sulfate, and thus, comparisons of total endosulfan EECs were made to toxicity estimates
for endosulfan sulfate.
                                       67

-------
Exposure estimates for the terrestrial animals assumed to be in the target area or in an
area exposed to spray drift are derived using the T-REX model (version 1.4.1, Oct. 9,
2008). This model incorporates the Kenega nomograph, as modified by Fletcher et al.
(1994), which is based on a large set of actual field residue data. The upper limit values
from the nomograph represented the 95th percentile of residue values from actual field
measurements (Hoerger and Kenega, 1972). For the purposes of this assessment, upper-
bound Kenaga nomogram estimates reported by T-REX are used for derivation of the
EECs for the terrestrial-phase CRLF, SFGS, SMHM, BCB, VELB,  SJKF, CTS and their
potential prey.

Birds are currently used as surrogates for terrestrial-phase amphibians and reptiles.
However, amphibians and reptiles are poikilotherms (body temperature varies with
environmental temperature) while birds are homeotherms (temperature is regulated,
constant, and largely independent of environmental temperatures). Therefore,
amphibians and reptiles tend to have much lower metabolic rates and lower caloric intake
requirements than birds or mammals. As a consequence, birds are likely to consume
more food than amphibians and reptiles on a daily dietary intake basis, assuming similar
caloric content of the food items. Therefore, the use of avian food intake allometric
equation as a surrogate to amphibians and reptiles is likely to result  in an over-estimation
of exposure and risk for reptiles and terrestrial-phase amphibians. Therefore, T-REX
(version 1.4.1) has been refined to the T-HERPS model (v. 1.0), which allows for an
estimation of food intake for poikilotherms using the same basic procedure as T-REX to
estimate avian food intake.

Because there is some evidence of the potential for bioaccumulation of endosulfan in
aquatic organisms, an additional exposure pathway that will be considered in this
assessment is the consumption of contaminated fish or aquatic invertebrates that have
bioaccumulated endosulfan dissolved in water and their aquatic diet. The potential risk
from this pathway will be evaluated and discussed further in the "Risk Description"
section of the document. Multiple lines of evidence will be used to  evaluate
bioaccumulation potential, including measured bioconcentration factors (BCF),
bioaccumulation factors (BAF) and a food web bioaccumulation model (Kow-Based
Aquatic Bioaccumulation Model or KABAM). The bioaccumulation assessment relies on
predicted water and sediment concentrations from PRZM/EXAMS to estimate
concentrations of endosulfan in  aquatic organisms. These estimated tissue concentrations
will be compared to toxicity values for various taxonomic groups that may eat aquatic
organisms in order to evaluate potential risk.

The potential for exposure of listed species through consumption of contaminated
terrestrial prey will be evaluated by fugacity modeling for the earthworm based on
predicted concentrations of endosulfan in soil and soil pore water using PRZM.
                                       68

-------
Spray drift models, AGDISP and/or AgDRIFT are used to assess exposures of terrestrial
animals to endosulfan deposited on terrestrial habitats by spray drift.  AGDISP (version
8.13; dated 12/14/2004) is used to simulate aerial and ground applications using the
Gaussian far-field extension. In addition to the buffered area from the spray drift
analysis, the downstream  extent of endosulfan that exceeds the LOG for the effects
determination is also considered.

2.10.1.2 Measures of Effect

Data identified in Section 2.8 are used as measures of effect for direct and indirect effects
to the CRLF, SFGS, SMHM, BCB, VELB, SJKF, and CIS. Data were obtained from
registrant submitted studies or from literature studies identified by ECOTOX.  The
ECOTOXicology database (ECOTOX) was searched in order to provide more ecological
effects data and in an attempt to bridge existing data gaps. ECOTOX is a source for
locating single chemical toxicity data for aquatic life, terrestrial plants, and wildlife.
ECOTOX was created and is maintained by the USEPA, Office of Research and
Development, and the National Health and Environmental Effects Research Laboratory's
Mid-Continent Ecology Division.

The assessment of risk for direct effects to the terrestrial-phase CRLF, CTS, and SFGS
makes the assumption that toxicity of endosulfan to birds is similar to or less than the
toxicity to terrestrial-phase amphibians and reptiles (this also applies to potential prey
items).  The same assumption is made for fish and aquatic-phase regarding direct effects
on CRLF and CTS (again, this also applies to potential prey items).

The acute measures of effect used for animals in this screening level assessment are the
LDso, LCso and ECso. LD stands for "Lethal Dose", and LDso is the amount of a material,
given all at once, that is estimated to cause the  death of 50% of the test organisms.  LC
stands for "Lethal Concentration" and LCso is the concentration of a chemical that is
estimated to kill 50% of the test organisms. EC stands for "Effective Concentration" and
the ECso is the concentration of a chemical that is estimated to produce a specific effect in
50% of the test organisms. Endpoints for chronic measures of exposure for listed and
non-listed animals are the NOAEL/NOAEC  and NOEC. NOAEL stands for "No
Ob served-Adverse-Effect-Level" and refers to  the highest tested dose of a substance that
has been reported to have no harmful (adverse) effects on test organisms. The NOAEC
(i.e., "No-Observed-Adverse-Effect-Concentration") is the highest test concentration at
which none of the observed effects were statistically different from the control. The
NOEC is the No-Observed-Effects-Concentration. For non-listed plants, only acute
exposures are assessed (i.e., EC25 for terrestrial plants and ECso for aquatic plants).

It is important to note that the measures of effect for direct and indirect effects to the
assessed species and their designated critical habitat are associated with impacts to
survival, growth, and fecundity, and  do not include the full suite of sublethal effects used
to define the action area.  According  the Overview Document (USEPA 2004), the
Agency relies on effects endpoints that are either direct measures of impairment of
survival, growth, or fecundity or endpoints for  which there is a scientifically robust, peer
                                       69

-------
reviewed relationship that can quantify the impact of the measured effect endpoint on the
assessment endpoints of survival, growth, and fecundity.

2.10.1.3 Integration of Exposure and Effects

Risk characterization is the integration of exposure and ecological effects characterization
to determine the potential ecological risk from agricultural uses of endosulfan and the
likelihood of direct and indirect effects to CRLF, SFGS, SMHM, BCB, VELB, SJKF,
and CIS in aquatic and terrestrial habitats. The exposure and toxicity effects data are
integrated in order to evaluate the risks of adverse ecological effects on non-target
species. For the assessment of endosulfan risks, the risk quotient (RQ) method is used to
compare exposure and measured toxicity values. EECs are divided by acute and chronic
toxicity values. For calculation of RQs for aquatic organisms, EECs for total endosulfan
(sum of d, P, and endosulfan sulfate) were divided by the appropriate toxicity value for
the TGAI (or in the case of sediment toxicity, endosulfan sulfate since acceptable data
were only available for this degradate). The resulting RQs are then compared to the
Agency's levels of concern (LOCs) (USEPA, 2004) (see Appendix D).

For this endangered species assessment, listed species LOCs are used for comparing RQ
values for acute and chronic exposures of endosulfan directly to the CRLF, SFGS,
SMHM, BCB, VELB, SJKF, and CTS.  If estimated exposures directly to the assessed
species of endosulfan resulting from a particular use are sufficient to exceed the listed
species LOG, then the effects determination for that use is "may affect". When
considering indirect effects to the assessed species due to effects to prey, the listed
species LOCs are also used.  If estimated exposures to the prey of the assessed species of
endosulfan resulting from a particular use are sufficient to exceed the listed species LOG,
then the effects determination for that use is a "may affect."  If the RQ being considered
also exceeds the non-listed species acute risk LOG, then the effects determination is a
LAA. If the acute RQ is between the listed species LOG and the non-listed acute risk
species LOG, then further lines of evidence (i.e. probability of individual effects, species
sensitivity distributions) are considered in distinguishing between a determination of
NLAA and a LAA.  If the RQ being considered for a particular use exceeds the non-listed
species LOG for plants, the effects determination is "may affect". Specifically, any
exceedances of the listed species LOG for dicots would also result in an LAA for the
BCB and VELB due to their obligate relationship with dicots. Further information on
LOCs is provided in Appendix D.

2.10.2 Data Gaps

Data gaps were identified for several taxa for ecological effects.  These include aquatic
vascular plants, terrestrial monicots, and terrestrial dicots.  In addition, no data acceptable
for quantitative use in this risk assessment was identified for amphibians (aquatic or
terrestrial phases) and reptiles.
                                        70

-------
  3  Exposure Assessment
Endosulfan is formulated as wettable powder and emulsifiable concentrate and
applied as liquid spray only. Application equipment includes various types of ground
sprayers, airblast sprayers, and aerial sprayers. Aerial application is permitted in all
crops and therefore was chosen to evaluate exposure because aerial application gave
the highest exposure because of higher spray drift levels Ground application is also
considered for cotton as it gave higher exposure values because it is applied at a
higher rate than aerial application.  Furthermore, aquatic EECs for a range of ground
application scenarios were evaluated and found to be within a factor of two of those
from the  corresponding aerial application scenario (most within 20%). Risk quotients
from the  ground application scenario yielding the lowest EEC (Eggplant) did not
change appreciably compared to the aerial application eggplant scenario and still
exceeded the acute and chronic LOCs for aquatic organisms. Terrestrial EECs and
RQs calculated using T-REX and T-HERPS models do not distinguish ground spray
from aerial application methods.  Therefore, the risk profile from aerial application is
expected to be applicable to that for ground spray applications, except for some
reduction in the spray drift distance modeled using AgDRIFT.

3.1 Label Application Rates, Intervals and Buffers

Endosulfan labels may be categorized into two types: labels for manufacturing uses
(including technical grade endosulfan and its formulated products) and end-use products.
While technical products, which contain endosulfan of high purity, are not used directly
in the environment, they are used to make formulated products, which can be applied in
specific areas to control insects.  The formulated product labels legally limit endosulfan's
potential  use to only those sites that are  specified on the labels.  Uses being assessed are
summarized in Table 3.1.

Table 3.1 Summary of labeled endosulfan uses, scenarios, and application information
Scenario
CA almond
CA citrus
CA cole
crops
CAcorn
CA cotton
Crop Represented
Almonds, Hazelnuts &
Walnuts
Citrus
Broccoli, Cabbage &
Cauliflower
Collards, Kale &
Mustard Green
Sweet corn for fresh
market only
Cotton (ground
application)
Application Rate
(Ibs/A)
2.00
2.50
0.5+2.00
1.00
0.75
1.5
1.00
0.50+1.50
Number of
Applications
1
1
2
2
1
1
2
2
Application Window (Interval)
(first Application Date"dd-mm")
01-02 to31-08 (NA) (07-02)
01-06 to 31-10 (10 days) (01-06)
30-01 to 31-08 (7 days)
(13-02 for Broccoli and 27-03
for Collards)
01-05 to 01-09 (NA) (15-05)
01-06 to 01-09 (7 days)
Mainly in September
(15-06 same results as 01-09)
                                        71

-------
Scenario

CA fruits
CA lettuce
CA melon
CA nursery
CA potato
CArow
crops
CA
strawberry
CA tomato
Crop Represented
Cotton (aerial
application)
Apple
Apricot, Cherry,
Nectarine, Peach,
Pear, Plum & Prune
Lettuce &
Brussels Sprouts
Cucurbits (Cucumber,
Melons, Pumpkin &
Squash)
Eggplant
Ornamentals or Shade
Trees
Potato
Sweet Potato
Beans & Peas (dry) &
Pepper
Carrot & Celery
Celery
Strawberry
Tomato
Application Rate
flbs/A)
0.75
2.50
0.540.75+1.25
2.50
0.50+2.00
1.00
1.00
0.50+0.50+1.00
0.50
0.25
0.50
1.00
0.50
0.50
1.00
1.00
0.50
2.00
1.00
1.00
0.50
Number of
Applications
2
1
3
1
2
2
2
3
1
2
6
2
4
3
2
1
2
1
2
2
4
Application Window (Interval)
(first Application Date"dd-mm")

15-04 to 30-05 (10 days) (15-04)
15-02 to 01-07 (10 days) (15-04)
01-02 to 01-04; and
15-09to 30-10 (5 days) (14-02)
30-04 to 01-08 (7 days) (14-05)
01-05 to 31-07 (7 days) (15-05)
15-02 to 15-07 (10 days) (21-02)
01-04 to 01-10 (5 days) (14-04)
01-05 to 07-06 (5 days) (14-05)
01/07 to 01-08 (5 days) (15-07)
01-03 to 01-10 (5 days) (15-03)
15-02 to 01-09 (15 days) (21-02)
01-04 to 30-07 (7 days) (15-04)
In ecological exposure modeling the drift component is usually set to a default value
of 5% (input parameters guidance, 2002).  For endosulfan, the principal mitigation
measure for reduction of drift is a 300 ft. buffer from surface water bodies.  This
buffer distance is the same for all types of applications in California. In order to arrive
at an estimate for drift, AgDRIFT was used (a computer model of spray drift). For
aerial application, AgDRIFT estimated drift beyond the 300 ft buffer to be 2.38% for
aerial and 0.15% for ground application (Appendix B).
                                        72

-------
3.2 Aquatic Exposure Assessment

3.2.1 Modeling Approach

Aquatic exposures are quantitatively estimated for all of assessed uses using scenarios
that represent high exposure sites for endosulfan use. Each of these sites represents a 10
hectare field that drains into a 1-hectare pond that is 2 meters deep and has no outlet.
Exposure estimates generated using the standard pond are intended to represent a wide
variety of vulnerable water bodies that occur at the top of watersheds including prairie
pot holes, playa lakes, wetlands, vernal pools,  man-made and natural ponds, and
intermittent and first-order streams. As a group, there are factors that make these water
bodies more or less vulnerable than the standard surrogate pond. Static water bodies that
have larger ratios of drainage area to water body volume would be expected to have
higher peak EECs than the standard pond.  These water bodies will be either shallower or
have large drainage areas (or both). Shallow water bodies tend to have limited additional
storage capacity, and thus, tend to overflow and carry pesticide in the discharge whereas
the standard pond has no discharge. As watershed size increases beyond 10 hectares, at
some point, it becomes unlikely that the entire watershed is planted to  a single crop,
which is all treated with the pesticide.  Headwater streams can also have peak
concentrations higher than the  standard pond, but they tend to persist for only short
periods of time and are then carried downstream. For uncertainties related to modeling
EECs in  estuarine/marine environments, refer the reader to the Section 6.

Application of endosulfan is expected to result in movement of its isomers and common
degradate into surface water by spray drift, run-off, and volatilization.  For aquatic
systems, estimated environmental concentrations (exposure EECs) are obtained by
PRZM and EXAMS modeling which represent EECs resulting from drift and run-
off/erosion. Although volatilization is simulated, as a loss, by PRZM, wet and dry re-
deposition, of this loss, into aquatic systems is not included in the EXAMS simulation
and resultant EECs. For endosulfan, modeling for EECs  should include species that
constitutes the multi-chemical  stressor, a- and p-endosulfan and endosulfan sulfate
(referred to in modeling as the total toxic residues or TTR).  Therefore, a method should
be used to determine EECs for the TTR.  Among the three methods suggested by EFED
to assess exposure to parent (here two parents) and degradation products, the SAP
recommended the use of formation and decline kinetics (FD) and in some cases the
Residue  Summation (RS) or Total Residue (TR) methods (SAP, 2009)16.  Currently, the
FD method can not be used as it requires the use of EXPRESS shell which is the only
shell capable of facilitating the use of parent/daughter PRZM/EXAMS modeling
feature17. Therefore, the RS method was used to arrive at exposure EECs for a- and P-
endosulfan and endosulfan sulfate separately and in total. The procedure is explained in
16 SAP (The FIFRA Scientific Advisory Panel). 2009. Minutes for the SAP Meeting on selected issues
associated with risk assessment process for pesticides with persistent, bio-accumulative and toxic
characteristics. October 28-31, 2008. Arlington, VA. (Document No. 2009-01 dated Jan 29, 2009).
17 At the time of executing modeling, the EXPRESS shell was not formally approved and parent to
daughter modeling feature of PRZM/EXAMS can not be utilized when PE-5 shell is used.
                                        73

-------
detail in Appendix C. In summary, the method requires assignment of application rates
for each of the three species of the stressor.  These rates are included in Table 3.2. The
application rates for the individual residues were normalized according to the percentage
of maximum residue detected in aerobic soil metabolism studies and the molecular
weight ratio of degradate and parent compounds.

Table 3.2 Calculation of application rates for a- and p-endosulfan and endosulfan sulfate) for use in
modeling (rates are for almond)	
 Chemical
 Species
% Maximum in Aerobic Soil Study
               Observed
                            Normalized
          Application Rate
          (kg/ha)
            Molecular
            Ratio
         Required Application
         Rate (kg/ha)
 Alpha
  70%
45%
1.009
               1.009
 Beta
  30%
19%
0.426
              0.426
 SO4
  57%
36%
0.807
1.039
0.839
 Total
 157%
100%
2.242
Example calculation given in Table 3.2 is for almonds which is applied at a single rate of
2 Ibs a.i./A (2.242 kg a.i/ha). Rates used in the simulations were 1.009 kg a.i/ha of a- and
0.426 kg a.i/ha for P-endosulfan and 0.839 kg a.i/ha. All of the three chemicals were
assumed to be applied at time zero. In this case, it is assumed that the maximum
degradation of the two parents and the maximum formation of the common degradate
occur at application. Three PRZM/EXAMS modeling runs were executed using PE-5.
The first modeling run gives l-in-10 years exposure EECs for a-endosulfan, the second
for P-endosulfan, and the third for endosulfan sulfate. EECs for the total residue of
concern, or the stressor, were calculated by combining daily concentrations from the
three chemicals (refer to Appendix C for a complete example calculation).

Use-specific management practices for all of the assessed uses of endosulfan were used
for modeling, including application rates, number of applications per year, application
intervals, and buffer widths and resulting spray drift values (modeled from AgDRTFT)
and the first application date for each use.  Date of first application was developed based
on several sources of information including data provided by BEAD, a summary of
individual applications from the CDPR PUR data,  pesticide label (crop stage and pest),
and Crop Profiles maintained by the USDA18. A sample of the distribution of endosulfan
applications to selected crops in Appendix C based on CDPR PUR data for 2006.
18
  URL: http://pestdata.ncsu.edu/cropprofiles/cropprofiles.cfm
                                        74

-------
      Data from Appendix C was used to pick two application windows for lettuce from early
      February to early April and from mid September to late October. In contrast, data in
      Appendix C for cotton indicates one major application window from late August to mid
      September.

      3.2.2 Model Inputs

      Environmental fate and transport and other data used for generating the parameters for
      the three PRZM/EXAMS runs are listed in Table 3.3.

      Table 3.3 Summary of PRZM/EZAMS environmental fate and transport data of a- and p-endosulfan
      and endosulfan sulfate used as aquatic exposure inputs for this endangered species assessment
Input Parameter (Unit)
Molecular Weight g/mole
Henry's constant (atm-m3 mol"1 @25 °C)
Vapor Pressure ( torr @ 25 °C)
Solubility in Water(mg/L)
Photolysis in Water (t!/2 in days @ pH 7)
Aerobic Soil Metabolism (90th % t!/2 in
days)
Hydrolysis (90th % t'/2 in days)
Aerobic Aquatic Metabolism (water column
t!/2 in days)
Anaerobic Aquatic Metabolism (benthic t!/2
in days)
Koc (Average in ml/g)
Application rate and frequency
Application intervals
Chemical Application Method (CAM)
Application Efficiency
Spray Drift Fraction1
Value *
a-endosulfan
406.9
3.0E-6
4.6E-5
0.53
Stable
57
11
114
230
10,600
fi-endosulfan
406.9
1.4E-6
2.4E-5
0.28
Stable
208
19
416
298
13,500
Endosulfan Sulfate
422.9
1.6E-5
9.8E-6
0.33
Stable
Stable
184
Stable
286
12,050 4
Depends on the crop (refer to Table 3.2 for an example)
2
95% for aerial and 99% for ground
0.024 for aerial and 0.002 for ground
* Source of data refer to Table 2.1 and Table 2.2 in the environmental fate properties section above
1 Inputs determined in accordance with EFED "Guidance for Chemistry and Management Practice Input Parameters for
Use in Modeling the Environmental Fate and Transport of Pesticides" dated February 28, 2002
2 Estimated 2 X aerobic soil metabolism half-life;
3 Estimated 2 X anaerobic soil metabolism half-life
4 Assumed to be equal to the average for its parents= 10,600 ml/g and 13,500 ml/g= 12,050 ml/g	
                                                75

-------
    PRZM/EXAMS simulation involved execution of three runs, the first run for a-
    endosulfan, the second run is for p-endosulfan, and the third run for endosulfan sulfate.

    3.2.3 Results

    Aquatic EECs

    The aquatic EECs for the various scenarios and application practices are listed in
    Table 3.4 and Table 3.5. The first table contains aquatic exposure EECs resulting from
    the first and second run while the second table contains EECs resulting from the third run
    and the post processed EECs for the total stressor from results of the three runs. An
    example of how the combined EECs were estimated is presented in Appendix C. In
    summary, the process includes combining daily EECs (simple addition for daily
    concentrations) followed by calculating the l-in-10 year averages using the 90th
    percentile procedure.  Although more than one run was executed for some crops, the
    more conservative results were chosen. For example, citrus was simulated at 2.5 Ib a.i/A
    with one application and 0.50+2.00 Ibs a.i. /A in two applications. In
    Table 3.4 and Table 3.5, the highest exposure EEC values were chosen.

    Table 3.4 Aquatic EECs (ng/L) for endosulfan used in California (a- and P-endosulfan)1
Crop
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Ornamentals or Shade Trees (Northern Central coast))
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run No. *
1
1
1
2
1
2
3
1
1
1
3
1
Add
1
Add
3
1
2
1
1
a-endosulfan
Peak
1.44
1.50
1.21
0.70
0.92
0.39
0.67
1.50
1.40
0.86
0.30
2.24
1.95
0.87
1.18
0.55
0.81
0.63
1.27
0.77
21-day
0.402
0.371
0.553
0.247
0.259
0.199
0.310
0.413
0.585
0.429
0.076
0.704
0.682
0.322
0.410
0.258
0.292
0.186
0.520
0.309
60-day
0.213
0.181
0.345
0.136
0.141
0.162
0.206
0.204
0.356
0.249
0.037
0.427
0.489
0.208
0.208
0.150
0.148
0.110
0.311
0.165
P-endosulfan
Peak
0.60
0.63
0.55
0.30
0.41
0.22
0.27
0.63
0.76
0.31
0.13
1.19
1.02
0.34
0.49
0.21
0.33
0.27
0.59
0.31
21-day
0.143
0.127
0.231
0.101
0.118
0.083
0.112
0.142
0.258
0.114
0.029
0.294
0.321
0.109
0.162
0.089
0.110
0.072
0.209
0.111
60-day
0.079
0.062
0.151
0.056
0.069
0.067
0.079
0.072
0.159
0.064
0.016
0.174
0.207
0.074
0.102
0.052
0.059
0.045
0.137
0.060
1 aerial application is used in runs for all crops. For cotton, an additional run is shown for ground application because
ground application to cotton is at a higher rate compared to aerial application.
* Add= an additional run using weather station to represent crop growing in areas different than the default areas assigned
                                             76

-------
Crop
by the scenario.
Run No. *

a-endosulfan
Peak

21-day

60-day

fi-endosulfan
Peak

21-day

60-day

    Table 3.5 Aquatic EECs (jig/L) for endosulfan used in California (endosulfan sulfate and total
    stressor)
Crop
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Ornamentals or Shade Trees (Northern Central coast))
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run No. *
1
1
1
2
1
2
3
1
1
1
3
1
Add
1
Add
3
1
2
1
1
Endosulfan Sulfate
Peak
1.41
1.27
3.82
1.22
2.60
0.73
0.66
1.33
4.86
1.19
0.35
2.14
1.82
0.74
1.01
0.49
1.59
1.55
2.27
0.78
21-day
0.483
0.344
1.742
0.515
0.897
0.354
0.357
0.434
1.845
0.490
0.122
0.744
0.645
0.313
0.413
0.233
0.561
0.499
1.054
0.375
60 day
0.329
0.186
1.250
0.367
0.658
0.287
0.284
0.242
1.381
0.370
0.092
0.484
0.456
0.202
0.270
0.151
0.384
0.437
0.750
0.227
Total (as parent)
Peak
3.35
3.35
4.81
1.67
2.88
1.24
1.56
3.40
5.88
1.55
0.72
3.43
4.38
1.89
2.63
1.18
2.00
1.72
3.64
1.79
21-day
1.026
0.900
2.603
0.775
1.137
0.596
0.824
1.013
2.575
0.815
0.237
1.357
1.728
0.770
1.011
0.622
0.851
0.735
1.714
0.808
60 day
0.617
0.444
1.757
0.501
0.774
0.521
0.593
0.532
1.854
0.613
0.131
1.004
1.130
0.492
0.561
0.358
0.508
0.538
1.108
0.450
* Add= an additional run using weather station to represent crop growing in areas different than the default areas assigned
by the scenario.

    For total stressor, the highest EECs are associated with lettuce for the peak (5.88 ppb),
    non-leafy cole crops for the 21-day average (2.60 ppb) and lettuce for the 60-day average
    (1.85 ppb).  In contrast, the lowest EECs for all time averages are for eggplant (peak=
    0.72 ppb, 21-day= 0.24 ppb, and 60-day= 0.13 ppb).  Data suggest that exposure is the
    highest for lettuce and  non-leafy cole crops (peak range= 3.43-5.88 ppb), followed by
    Strawberry, fruit/nut trees, citrus, and potatoes grown in the northern part of the central
    coast (peak range from 2.0 to 3.43 ppb), and all other crops except eggplant (1.18 to 1.89
    ppb). Eggplant has the lowest exposure EECS (<1 ppb).

    As expected, the major contributor to exposure is endosulfan sulfate (53% on the
    average) followed by a-endosulfan (34% on the average), and p-endosulfan (14% on the
    average). The results reflect the isomer ratio (70 % for a-endosulfan and 30% for by P-
    endosulfan) and persistence of the common degradate endosulfan sulfate.
                                             77

-------
Exposure to sediment pore water in aquatic systems was also obtained for the same runs,
this data are listed in Table 3.6 and Table 3.7.
Table 3.6 Sediment pore water EECs (ng/L) for endosu It'an used in California (a- and ft-endosulfan)
Crop
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Ornamentals or Shade Trees (Northern Central coast))
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run No. *
I
I
I
2
I
2
3
1
1
1
3
1
Add
1
Add
3
1
2
1
1
a-endosulfan
Peak
0.11
0.09
0.22
0.08
0.08
0.10
0.10
0.11
0.22
0.14
0.02
0.33
0.35
0.10
0.11
0.08
0.08
0.06
0.17
0.08
21-day
0.110
0.093
0.214
0.083
0.082
0.097
0.105
0.105
0.219
0.134
0.020
0.323
0.345
0.100
0.110
0.075
0.082
0.063
0.166
0.083
60-day
0.108
0.086
0.203
0.079
0.076
0.095
0.102
0.098
0.207
0.127
0.018
0.302
0.328
0.094
0.104
0.070
0.077
0.060
0.158
0.077
fi-endosulfan
Peak
0.05
0.03
0.11
0.04
0.05
0.05
0.04
0.04
0.11
0.04
0.01
0.13
0.15
0.04
0.07
0.03
0.04
0.03
0.08
0.03
21-day
0.046
0.032
0.107
0.036
0.051
0.046
0.043
0.041
0.104
0.038
0.009
0.131
0.147
0.036
0.068
0.027
0.039
0.027
0.080
0.033
60-day
0.045
0.030
0.102
0.034
0.047
0.045
0.042
0.038
0.098
0.035
0.009
0.122
0.139
0.034
0.064
0.025
0.038
0.026
0.076
0.031
* Add= an additional run using weather station to represent crop growing in areas different than the default
areas assigned by the scenario

Table 3.7 Sediment pore water EECs (jig/L) for endosu It'an used in California (endosulfan sulfate
and total stressor)
Crop
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Run No. *
1
1
1
2
1
2
3
1
1
1
o
J
1
Endosulfan Sulfate
Peak
0.22
0.10
0.89
0.27
0.39
0.21
0.18
0.16
0.90
0.26
0.06
0.35
21-day
0.215
0.101
0.881
0.265
0.390
0.208
0.183
0.158
0.892
0.254
0.064
0.352
60 day
0.213
0.095
0.850
0.261
0.383
0.206
0.181
0.150
0.873
0.242
0.060
0.341
Total (as parent) * *
Peak
0.37
0.23
1.18
0.38
0.51
0.35
0.33
0.30
1.19
0.42
0.09
0.80
21-day
0.363
0.222
1.168
0.375
0.508
0.344
0.324
0.298
1.181
0.416
0.090
0.793
60 day
0.357
0.207
1.123
0.365
0.492
0.339
0.319
0.280
1.145
0.394
0.085
0.753
                                            78

-------
Crop
Ornamentals or Shade Trees (Northern Central coast))
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run No. *
1 Add
1
1 Add
3
1
2
1
1
Endosulfan Sulfate
Peak
0.35
0.11
0.20
0.08
0.28
0.31
0.49
0.17
21-day
0.346
0.113
0.198
0.082
0.279
0.304
0.484
0.163
60 day
0.332
0.110
0.189
0.078
0.267
0.290
0.466
0.155
Total (as parent) * *
Peak
0.84
0.25
0.37
0.18
0.39
0.39
0.72
0.28
21-day
0.825
0.245
0.369
0.181
0.390
0.382
0.712
0.274
60 day
0.786
0.234
0.351
0.170
0.373
0.364
0.682
0.257
* Add= an additional run using weather station to represent crop growing in areas different than the default areas assigned
by the scenario
** Important Note: This total is a product of simple addition of the 1-10 year data and was not calculated from
combining daily concentrations (currently PE-5 do not produce a file containing daily concentrations for the pore water)


      Exposure to benthic sediment in aquatic systems was also obtained for selected runs, this
      data are listed in Table 3.8 and Table 3.9.
      Table 3.8 Sediment EECs (ng/L) for endosulfan used in California (a- and ft-endosulfan)
Crop
Broccoli, Cabbage & Cauliflower
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Strawberry
Tomato
Run No.
1
3
1
1
3
1
1
1
a-endosulfan
Peak
92
45
45
94
9
139
71
36
21-day
91
44
45
93
8
137
70
35
60-day
86
43
42
88
8
129
67
33
fi-endosulfan
Peak
59
24
22
57
5
72
44
18
21-day
58
23
22
56
5
71
43
18
60-day
55
23
20
53
5
66
41
17
      Table 3.9 Sediment EECs (jig/L) for endosulfan used in California (endosulfan sulfate and total
      stressor)
Crop
Broccoli, Cabbage & Cauliflower
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Strawberry
Tomato
Run No.
1
o
J
1
1
3
1
1
1
Endosulfan Sulfate
Peak
429
89
77
434
31
171
235
80
21-day
424
88
76
430
31
170
233
79
60 day
410
88
72
422
29
165
225
75
Total (as parent) *
Peak
580
158
144
585
45
382
350
134
21-day
573
155
143
579
44
378
346
132
60 day
551
154
134
563
42
360
333
125
* Important Note: This total is a product of simple addition of the 1-10 year data and was not calculated from combining
daily concentrations (currently PE-5 do not produce a file containing daily concentrations for the pore water)	
                                                    79

-------
Bioaccumulation Assessment

Considering that log KOW values for the parent endosulfan isomers and endosulfan sulfate
either approach or exceed a log Kow of 4, and that endosulfan can persist for relatively
long periods of time in aquatic ecosystems, the potential for direct effects on aquatic-
phase CRLF through consumption of contaminated aquatic prey was investigated. The
KABAM model (Kow (based) Aquatic Bio Accumulation Model) version 1.0 was used to
evaluate the potential exposure and risk of direct effects to aquatic-phase CRLF via
bioaccumulation and biomagnification in aquatic food webs. KABAM is used to
estimate potential bioaccumulation of hydrophobic organic pesticides in freshwater
aquatic ecosystems and risks to mammals and birds consuming aquatic organisms which
have bioaccumulated these pesticides. The bioaccumulation portion of KABAM is based
upon work by Arnot and Gobas (2004) who parameterized a bioaccumulation model
based on PCBs and some pesticides (e.g., lindane, DDT) in freshwater aquatic
ecosystems. KABAM relies on a chemical's octanol-water partition coefficient (Kow) to
estimate uptake and elimination constants through respiration and diet of organisms in
different trophic levels. Pesticide tissue residues are calculated for different levels of an
aquatic food web. The model then uses pesticide tissue concentrations in aquatic animals
to estimate dose- and dietary-based exposures and associated risks to mammals and birds
consuming aquatic organisms. Previous analyses using an earlier version of the KABAM
model indicate relatively close agreement between its predicted bioconcentration factor
(BCF) and those reported from experimental studies for endosulfan (USEPA 2007;
D346213).

Details of the bioaccumulation assessment for endosulfan in relation to the CRLF are
provided in Section 5.2.1.1.

Soil EECs

Total soil  and pore water concentrations were calculated using PRZM simulations for
selected crop scenarios.  Crop scenarios were selected to include a range of
concentrations based on aquatic exposure EECs obtained for endosulfan use patterns.
Two PRZM runs were executed for six crop use patterns, one for aerial application and
the other for ground application. Results are summarized in Table 3.10 which includes
results for the ground applications only. Concentrations for aerial applications are not
included because values were either similar or slightly less.  It is noted that total soil and
soil pore water concentrations are represented by either the highest or the steady state
values. These values are obtained from plots of the total stressor daily concentrations
similar to those included in Figures C-l & C-2 (Appendix  C) for the total  soil
concentration and Figures C-3 & C-4 (Appendix C) for the soil pore water
concentration.  Figures C-l to C-4 suggest the following  for total stressor concentration
in the soil  system:
    (1) Concentration is determined by endosulfan sulfate because concentrations for the
       other two endosulfan species (alpha & beta) appear to reach steady state (at very
       low concentrations), rather quickly; and
                                        80

-------
         (2) Concentration does reach steady state for only two use patterns: cotton and
             nursery (Figures C-l & C-3) while steady state is not reached for the others
             (Figures C-2 & C-4)

      Table 3.10 Total soil and pore water concentrations for selected crop scenarios
Crop
Fruits
Cole crops
Lettuce
Nursery
Cotton
Eggplant
Bulk
Density
fe/^r
1.70
1.50
1.58
1.55
1.45
1.45
Total Soil Concentration
(mg/kg) *
13.20
14.00
12.00
9.80
4.61
3.59
(mg/m)**
22,400
21,000
18,600
15,190
6,685
5,206
(mole/m ) * *
0.055
0.052
0.046
0.037
0.016
0.013
Soil Pore Water Concentration
(mg/L)*
0.182
0.067
0.140
0.181
0.130
0.066
(mg/m )**
182
67
140
181
130
66
(mole/m )**
0.000447
0.000165
0.000344
0.000445
0.000319
0.000161
Concentration
Reached
Steady State?
No
No
No
Yes
Yes
No
* Bulk density from scenario for the top horizon (Top 10 cm); concentrations from PEZM simulation.
** Calculated values based on bulk density and molecular weight of endosulfan= 406.9 g/mole

      3.2.4 Existing Monitoring Data

      A critical  step in the process of characterizing EECs is comparing the modeled estimates
      with available monitoring data. Varied types of monitoring data exist for alpha and beta
      endosulfan and the common degradate endosulfan sulfate. Data cover important
      environmental compartments including surface water, ground water, sediment, and air.
      Included in this assessment are endosulfan data from the USGS NAWQA program, the
      EPA STORET, and the California Department of Pesticide Regulation (CDPR).

      3.2.4.1 Surface Water Monitoring Data

       CDPR Data
      Surface water was monitored from 1991 to 2006 for a- and p-endosulfan and/or
      endosulfan sulfate in thirty three counties in California (California Department of
      Pesticide regulations; CDPR)19. Reported data show relatively high frequency and
      concentrations from 1991 to 1996 particularly for Imperial County. In contrast,
      significantly lower frequency and concentrations were observed in data collected from
      2001 to 2006 for the three species (Figure 3.1) and their total (Figure 3.2).
      19California Department of Pesticide Regulations (CDPR) data was obtained from
      URL:http://www.cdpr.ca.gov/docs/emon/surfwtr/surfcont.htm
      http://www.cdpr.ca.gov/docs/emon/grndwtr/list_mon.htm (GW no data)
      http://www.cdpr.ca.gov/docs/emon/pubs/tac/endoslfn.htm (Air)
                                               81

-------
Figure 3.1 Detected concentrations of alpha and beta-endosulfan and common degradate in
California surface waters from 1991 to 2006 (No data were collected from 1996 to 2000).

0.2 -,
0.18 -

a U.14


S
0 04 -

0 -
Ja
9
+ • Alpha-endosulfan o Beta-endosulfan + endosulfan sulfate
+
+
+
*
£+
0 + •
± 0)-
* + i +
* *+ &»
ft *$ T § * t f t
n- Jan- Jan- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec-
0 91 92 92 93 94 95 96 97 98 99 00 01 02 03 04 05
Monitoring Date


Out of scale values were for SO4 only. These values occurred on 6/21/1993 and were: 0.562, 0.485, and 0.378 ppb
Figure 3.2 Detected concentrations of the total endosulfan species in California surface waters from
1991 to 2006

0.3000 -,
S"
£ 0 2000 -
c
o
1
a) o 1500
c
o
o
I
IE o 1000 -
0 0500
o.oooo -
Ja
9
_ • Endosulfan Species Total
•
*
• •
•••
• •
: •
• •
«*
• "1
• « • f
* • i%
jjSj * 	 ^ 	 ^ 	
n- Jan- Jan- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec- Dec-
0 91 92 92 93 94 95 96 97 98 99 00 01 02 03 04 05
Monitoring Date

Out of scale values occurred on 6/21/1993 and were 0.952, 0.825 & 0.495 ppb
In interpreting the data, the following should be taken in consideration:
                                            82

-------
(1) Not all endosulfan species were determined for all the samples suggesting that
observed total residue concentrations could be higher than that displayed in Figure 3.2
above.
(2) Reported detection limits for a- and p-endosulfan were mostly 0.005 ppb while it
ranged from 0.01 to 0.005 ppb for the common endosulfan sulfate degradate (Table
3.11)

Table 3.11 Summary of reported detection limits for the CDPR surface water data
Reported
Detection Limit
(Ppb)
0.0001
0.002
0.001
0.005
0.01
0.1
Proportion of Samples Analyzed at the Detection limit (in %)
Samples With Detections
alpha
2%
NA
0%
98%
NA
0%
Beta
0%
NA
NA
100%
NA
0%
SO4
1%
5%
NA
50%
44%
NA
Samples Without Detections
alpha
1%
NA
10%
74%
NA
15%
Beta
6%
NA
NA
70%
NA
24%
SO4
6%
11%
NA
29%
54%
NA
(NA= Not Applicable)

(3) During the first monitoring period (1991 to 1996), nearly 2,000 surface water samples
were collected from 83 sites in nine counties.  In contrast, only 500 surface water samples
were collected from 153 sites in 24 counties during the second monitoring period.

(4) During the second monitoring period, sampling was repeated for 7 counties, namely,
Imperial, Merced, Sacramento, San Joaquin, Sonoma, Stanislaus, and Sutler.

(5) During the second period rates and uses were reduced.

For the first monitoring period, data are summarized in Table 3.12.  In interpreting this
data, it is important to note the following:
    •  Endosulfan sulfate were determined for nearly 93% of the samples;
    •  a- and P-endosulfan were determined for only 31% of the samples; and
    •  All of the three species were determined for only 26% of the samples.

Table 3.12 Summary of CPDR surface water monitoring data for California Counties collected for
the first monitoring period (1991 to 1996)
County
Imperial
Merced
Monterey
Sacramento
San Joaquin
Santa Cruz
Sonoma
Stanislaus
Sutler
Monitoring Years
93-94
91-95
94-95
90- 92 & 94- 96
91-93
94-95
94-95
91-93
93-94
Total Number of
Sites
11
10
6
11
2
4
2
10
1
Samples
195
381
213
566
51
24
153
76
153
Detection Distribution
Detect
64%
2%
0%
<1%
4%
0%
0%
8%
0%
No Detect
36%
98%
100%
>99%
96%
100%
100%
92%
100%
                                        83

-------
Data in Table 3.12 indicate that at least one of the endosulfan species was observed in
surface waters from five out of the nine counties including Imperial (64%) followed by
Stanislaus (8%), San Joaquin (4%), Merced (2%), and Sacramento (<1%).  A summary of
all data is included in Figure 3.3 for Imperial County and Figure 3.4 for Stanislaus, San
Joaquin,  Merced, and Sacramento Counties.
Figure 3.3 CPDR Monitoring data for Imperial County (11 sites, varied times of the year)
                                                             Alpha Endosulfan • Beta Endosulfan  Sulfate
                                        4567
                                             Sampling Sites: All in Imperial County
                                                                                10
                         Out of scale values are: 0.56 ppb for 1; 0.39 ppb for 2 and 0.49 ppb for 3
                                         84

-------
Figure 3.4 CDPR Monitoring data for Merced County (4 sites), Stanislaus County (6 sites), San
Joaquin County (2 sites) and Sacramento County (1 site) at varied times of the year
n o
u.z
0.175-
.10
5 0.125-
Q.
1=
O
"-^ n ^
2 U.I
-t;
0)
o
5 0.075-
0.05-
0.025-








• Alpha Endosulfan • Beta Endosulfan • Sulfate






I I nil II „
||

JJjJJjJ



ll I





I
Jlh
llllll.llllll
JjJjJJJ
ME1 2 3 4 ST1 2 3 4 5
Samplind Sites: (Merced=ME 1-4; Stanislaus=ST 1-6; San Joaquin= SJ 1-2; and Sacramento=SA
111
6

_L
SJ 1







1
2 S




A

Observed ranges of concentrations were the highest for the common degradate
endosulfan sulfate (0.02 to 0.56 ppb) followed by a- and p-endosulfan (0.005 to 0.22 and
0.005 to 0.17 ppb, respectively). Total endosulfan residue ranged from 0.03 to 0.95 ppb.

For the second monitoring period (2001 to 2006), monitoring data are summarized in
Table 3.13
Table 3.13 Summary of CPDR surface water monitoring data for California (the 2   monitoring
period)
County
Alameda
Del Norte
Glenn
Humboldt
Kern
Imperial ***
Lake
Los Angeles **
Marin
Mendocino
Napa
Orange
Merced***
Monitoring
Years
01&02
02&03
06
02&03
06
02-04
06
01&03
01&02
02&03
02
02&03
06
Total Number Of:
Sites
2
1
2
6
2
12
1
30
2
3
1
8
1
Samples
4
4
16
20
4
67
4
62
4
17
1
36
1
Detection Distribution (%)
Detect
0%
0%
0%
0%
0%
9%
0%
3%
0%
0%
0%
0%
0%
No Detect
100%
100%
100%
100%
100%
91%
100%
97%
100%
100%
100%
100%
100%
Species Analyzed For *
Beta
No
No
Yes
No
Yes
No
Yes
No
No
No
No
No
No
                                         85

-------
County
Riverside
Sacramento***
San Bernardino
San Diego
SanJoaquin***
San Mateo
Santa Clara
Shasta
Siskiyou
Solano
Tehama
Trinity
Tulare
Sonoma***
Stanislaus***
Sutler***
Ventura **
Yolo
Monitoring
Years
02-04
05-06
02-04
02-04
04-05
02&03
02&03
06
02&03
05&04
06
02&03
05&06
02-03
04
06
01
05&06
Overall Summary
Total Number Of:
Sites
8
1
3
37
2
6
3
1
9
4
1
2
12
5
1
4
4
5
153
Samples
28
18
9
111
22
18
12
2
42
33
2
8
58
35
2
22
7
78
580
Detection Distribution (%)
Detect
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
86%
0%
3%
JV0 Detect
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
14%
100%
97%
Species Analyzed For *
Beta
No
No
No
No
No
No
No
Yes
No
Yes
Yes
No
Yes
No
Yes
Yes
No
Yes
Yes= 30%
 * a- endosulfan was not determined for all samples; p-endosulfan as indicated in the Table by Yes=
 determined or No= not determined, and 100% of the samples were analyzed for endosulfan sulfate
 ** These two counties were the location of the only six samples with endosulfan sulfate detections
 (samples were analyzed for endosulfan sulfate only). Concentrations were: 0.003 to 0.005 ppb from
 three locations in Los Angeles County; and 0.002 to 0.005 ppb from two locations in Ventura County
 *** Counties in which sampling is a repeat (i.e., these counties were also sampled during the first
 period)	

Data for the second monitoring period indicate two important points:

(1) For all counties: observed frequency of detection and ranges of concentrations were
    very low compared to earlier years (1991 to 1996). In fact the only five detections
    were reported for samples collected in 2001 with endosulfan sulfate concentrations in
    the range of 0.002 to 0.005 ppb (a- and P-endosulfan were not determined). Again,
    endosulfan sulfate were determined for nearly 100% of the samples, a- and P-
    endosulfan were determined for only 0% and 30% of the samples, respectively; and
(2) For Imperial, Stanislaus, San Joaquin, and  Merced Counties: Second monitoring
    (2002 to 2006) showed a significant decrease from the first monitoring in earlier
    years. During these recent years, detections were recorded in Imperial County only
    (9% of the samples) with no detections in Stanislaus, San Joaquin, and Merced
    Counties.

EPA STORET Data
In STORET20, data are reported on monitored surface water from 2001 to 2008 (mostly
from 2001 to 2003), for a- and P-endosulfan and/or endosulfan sulfate in  twenty one
counties in California (Table 3.14).
  STORET Data base URL: http://www.epa.gov/storet/dw home.html
                                         86

-------
Table 3.14 Summary of STORET surface water monitoring data for California
County
Alameda
Contra Costa
Del Norte
Humboldt
Imperial
Los Angeles
Marin
Mendocino
Napa
Orange
Riverside
San Bernardino
San Diego
San Luis Obispo
San Mateo
Santa Barbara
Santa Clara
Siskiyou
Solano
Sonoma
Ventura
Monitoring month- Year
Sep-01; Apr-02
Sep-01; Apr & Jun-02
Feb, Mar & Jun-02
Feb, Mar & Jun-02
May, Sep & Oct-02
Oct&Nov-Ol; Jan-03
Oct-01; Apr-02
Feb, Mar & Jun-02
39966
Oct-02; Jan, Apr & May-03
May, Sep & Oct-02; Apr & May-03
May & Oct-02
Mar, Apr, Jun & Sep-02; Jan, Apr & May-03
Jul-03; Jan & July-04/05/06/07 & Jan-08
Apr & Jun-02; Jan-03
Jun-06; Jan-08
Apr & Jun-02; Jan-03
Feb, Mar & Jun-02
Oct-01; Apr & Jun-02
Feb, Mar, Apr & Jun-02
Nov-02
Overall Summary
Total Number Of:
Sites
2
2
3
3
20
4
3
3
1
10
7
3
15
1
4
3
4
2
2
3
5
100
Samples
6
5
8
5
44
10
4
10
1
38
13
6
70
20
15
6
13
6
3
15
6
304
Detects Distribution (%)
>LOQ
33%
0%
0%
0%
0%
40%
0%
0%
0%
0%
0%
0%
9%
0%
0%
0%
0%
0%
0%
0%
50%
6%

-------
Figure 3.5 Detected concentrations of a- and p-endosulfan and endosulfan sulfate common degradate
in California surface waters from 2001 to 2008 (STORET data)	
n nin
n nnp
S*
Q_
Q_
*"" ' n nnfi
o
1
+j
o
o
n nco
o.ooo -
Jan

•
o
o
•
* o
*
» 0


• Alpha o Beta +SO4



4tf* • *
	 ^ 	 A 	 A 	 A 	 A 	 A 	 -AA 	 A 	 ^
^W rvlH . " W W W
-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07
Monitoring Date





	 1
Jan
-08
Data points >0 and O.002 represent samples with detections over the LOD but lower than the LOQ. These were
assumed to equal 0.001 ppb because LOD and LOQ were not reported. However, data suggest that the LOQ is
probably equal to 0.002 ppb.


Figure 3.6 Detected concentrations of the total endosulfan species in California surface waters from
2001 to 2008 (STORET data)
0.010 -
_ 0.008 -
S"
a.
r 0.006 -
o
2
S 0.004 -
O
O
0.002 -
0.000 -





• Endosulfan Species Total




	 tfMA— fllHBI^ 	 A 	 A 	 A 	 A 	 A 	
P rVl ^ iV WW~w w w w w w
Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06
Monitoring Date


-+• 	 • 	 • 	



Jan-07 Jan-08

Data points >0 and <0.002 represent samples with detections over the LOD but lower than the LOQ. These were
assumed to equal 0.001 ppb because LOD and LOQ were not reported. However, data suggest that the LOQ is probably
equal to 0.002 ppb.
                                            88

-------
Figure 3.7 Detected concentrations of a- and p-endosulfan and endosulfan sulfate common degradate
in surface waters of five California Counties (Counties with detections over the LOQ; 2001 to 2008
STORET data)
0.009
0.008-
_ 0.007-
si
g; 0.006-
o 0.005-
'•a
•g 0.004-
01
(J
c 0.003-
o
O
0.002-
0.001-
































• Alpha
• Beta
• S04


I 1
1
IN Ml
itltl

ii



i

i







n
.1
Alameda San Diego San Diego Los Angeles
Counties with Detections Over LOQ




1



Ventura

(1) Data points >0 and O.002 represent samples with detections over the LOD but lower than the LOQ.
These were assumed to equal 0.001 ppb because LOD and LOQ were not reported.  However, data suggest
that the LOQ is probably equal to 0.002 ppb.
(2) Samples with no detect are not included. The number of samples with no detects is equal to: 4 for
Alameda Co.; 58 for San Diego CO.; 6 for Los Angeles Co. & 3 for Ventura Co.

STORET data appear to be comparable to date reported by  CDPR for the years following
2001. For these years, STORET data reported maximum concentrations 0.008 ppb for a-
endosulfan, 0.0064  ppb for P-endosulfan, 0.005 ppb for endosulfan sulfate, and 0.008 ppb
for the total (Figure 3.5 and Figure 3.6). For the same period, CDPR data reported the
same maximum concentrations of 0.005 ppb for endosulfan sulfate (a- and P-endosulfan
were not determined).

USGSNAQWA Data
California surface water was monitored by the NAWQA program21 during 1995  and
from 2001 to 2007 for endosulfan residue including a-, P-, ether- endosulfan, and
endosulfan sulfate in fifteen sites in five counties.  A total of nearly 200 samples were
collected with no detection reported for any of the  endosulfan residue including a-, P-,
ether- endosulfan, and endosulfan sulfate. Detection limits varied for the various
endosulfan residues: 0.0047 to 0.011 ppb for a-endosulfan, 0.0142 for P-, endosulfan,
0.0041to 0.0066 ppb for ether- endosulfan, and 0.0058 to 0.022 ppb for endosulfan
sulfate (Table 3.15).

Table 3.15 A summary of NAWQA monitoring data for surface water from California
County
Merced
Site
Site 1
Site 2
Sampling Dates
Start
10-Jan-95
9-Oct-Ol
Finish
10-Jan-95
27-Apr-05
Number
of
Samples
1
23
Endosulfan Species Detection Limits (ppb)
Alpha
0.0100
0.0047
Beta
ND
0.0142
Ether
ND
0.0041
SO4
ND
0.0058
21 NAQWA data was obtained on January 17, 2008: URL:
http://infotrek.er.usgs.gov/traverse/f?p=NAWQA:HOME:3748645897450568
                                         89

-------
County
Riverside
Sacramento
San Joaquin
Stanislaus
Site
Site 1
Sitel
Site 2
Sitel
Sitel
Site 2
SiteS
Site 4
Site5
Site 6
Site?
SiteS
Site 9
Sampling Dates
Start
22-Jun-05
12-Apr-06
13-Jun-05
21-Jun-06
20-Apr-06
15-Jun-05
22-Nov-05
25-Apr-06
16-Jun-05
31-Oct-05
31-Oct-05
7-Dec-05
27-Apr-06
10-Jan-95
16-Oct-02
19-Oct-04
22-Jun-05
26-Apr-06
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
9-Oct-Ol
22-Jun-05
26-Apr-06
10-Jan-95
10-Jan-95
Finish
15-Feb-06
9-Aug-06
7-Feb-06
21-Jun-06
6-Sep-07
27-Oct-05
15-Mar-06
24-Sep-07
28-Sep-05
31-Oct-05
31-Oct-05
23-Feb-06
2-Aug-06
10-Jan-95
3-Sep-03
20-Apr-05
15-Feb-06
27-Sep-07
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
10-Jan-95
20-Apr-05
15-Feb-06
9-Aug-06
10-Jan-95
10-Jan-95
Number
of
Samples
5
3
5
1
20
7
8
26
8
1
1
2
3
1
18
4
5
25






4
20
5
3
1
1
Endosulfan Species Detection Limits (ppb)
Alpha
0.0047
0.0110
0.0047
0.0110
0.0110
0.0047
0.0047
0.0110
0.0047
0.0047
ND
0.0047
0.0110
0.0100
0.0047
0.0047
0.0047
0.0110
0.0100
0.0100
0.0100
0.0100
0.01
0.01
0.01
0.0047
0.0047
0.0110
0.0100
0.0100
Beta
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.0142
0.0142
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.0142
ND
ND
ND
ND
Ether
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.0041
0.0066
ND
ND
ND
ND
ND
ND
ND
ND
ND
0.0041
ND
ND
ND
ND
SO4
0.0138
0.0220
0.0138
0.0122
0.0220
0.0138
0.0138
0.0220
0.0138
ND
0.0200
0.0138
0.0220
ND
0.0058
0.0138
0.0138
0.0220
ND
ND
ND
ND
ND
ND
ND
0.0058
0.0138
0.0220
ND
ND
In conclusion, surface water monitoring data are summarized in the following Table
3.16
Table 3.16 Surface water data summary
                        Cnty= No of counties; Site= No of Sites; Smpl= No of Samples
Data Source
Total Number of:
Endosulfan Species
a (Max ppb)
P (Max ppb)
Sulfate (Max ppb)
CDPR1991-1996
Cnty
9

Site
83
Smpl
2,000
0.22
0.17
0.56
CDPR 2001-2006
Cnty
24
Site
153
Smpl
500
Not Determined
Not Determined
0.005 ppb
STORET 2001-2008
Cnty
21

Site
100
Smpl
304
0.008
0.0064
0.005
NAQWA (1995 & 2001-2007)
Cnty
5

Site
15
O.011
Smpl
200

O.0142
O.022
                                           90

-------
3.2.4.2 Groundwater Monitoring Data
No ground water monitoring data were reported for California by the CDPR for
endosulfan isomers and the common degradate. However, ground water monitoring data
were reported by NAWQA for 11 California counties from 149 ground water wells. Data
showed no detections of a-, P- and ether- endosulfan, and/or endosulfan sulfate.
Detection limits varied for the various endosulfan species: 0.0047 to 0.011 ppb for a-
endosulfan, 0.0142 ppb for P-  endosulfan, 0.0041 ppb for ether- endosulfan and 0.0058 to
0.022 ppb for endosulfan sulfate (Table 3.17).  Both P- endosulfan and ether- endosulfan
were not determined for nearly 39% of the samples.

Table 3.17 A summary of monitoring data for ground water from California
County
Fresno
Kem
Kings
Madera
Merced
Orange
Sacramento
San Bernardino
San Joaquin
Stanislaus
Tulare
Number
of
WeUs
1
26
1
3
6
5
13
2
31
5
1
14
3
18
3
10
Well Depths (ft)
Min
114
68
300
55
165
96
56
942
37
234
175
44
110
33
182
118
Max
300
390
300
244
265
230
265
1,020
200
900
175
250
195
195
310
313
Endosulfan Species Detection Limits (ppb)
Alpha
0.011
0.0047
0.011
0.0047
0.0047
0.011
0.0047
0.011
0.0047
0.011
0.011
0.0047
0.011
0.0047
0.011
0.0047
Beta
ND
0.0142
ND
0.0142
0.0142
ND
0.0142
ND
ND
ND
ND
0.0142
ND
0.0142
ND
0.0142
Ether
ND
0.0041
ND
0.0041
0.0041
ND
0.0041
ND
ND
ND
ND
0.0041
ND
0.0041
ND
0.0041
S04
0.022
0.0058
0.022
0.0058
0.0058
0.022
0.0058
0.022
0.0138
0.022
0.022
0.0058
0.022
0.0058
0.022
0.0058
Sampling Dates
Jun-06; Aug-06
Jul-01; Aug-01; Sep-01; Jun-02
Aug-06
Jul-02
Sep-01; Jul-02
Jun-06; Aug-06
Oct-01; Jul-02; Aug-02
Jun-06
Jul-05; Aug-05; Sep-05; Aug-06
Jun-06; Aug-06
Aug-06
Sep-01; Oct-01; Sep-02
Jun-06; Aug-06
Oct-01; Nov-01; Jun-02; Aug-02
Jun-06; Aug-06
Sep-01; Jun-02; Aug-02
* ND= Chemical species was not determined

In conclusion, ground water monitoring data are summarized as follows:

   •   No CDPR data; and
   •   NAQWA (2001-2006): 11 counties; 149 wells: Alpha <0.011 ppb; Beta<0.0142
       ppb; and Sulfate <0.022 ppb

3.2.4.3 Sediment Monitoring Data

Bottom sediment monitoring data were reported by NAWQA for 18 California counties
from 49 surface water sites. No detections were observed in the 53 samples which were
analyzed for a-endosulfan only.  Reported values were all less than the limit of detection
of 0.2 to 2 jig/Kg of a-endosulfan (Table 3.18)

Table 3.18 A summary for NAWQA monitoring data to base sediment in California
County
Sampling Dates
Stanislaus (Sept, 01)
Number of
Sites
1
Samples
1
Concentration
Alpha (ng/Kg)
0.2
                                       91

-------
County Sampling Dates
Alpine (Sept, 92); Colusa (Oct, 95); El Dorado (Sept, 92); Fresno
(Oct, 92); Kings (Oct, 92) and Orange (Sept, 98)
Nevada (Sept, 92 & Oct, 96)
Shasta (Oct, 95) and Yuba (Nov, 95)
San Joaquin (Oct, 92 & Nov, 97)
Riverside (Sept, 98); Sacramento (Oct & Nov, 95); Sutler (Nov &
Oct, 95); Tehama (Oct, 95) and Yolo (Oct, 95)
Merced (Oct, 92);
San Bernardino (Sept & Nov, 98)
Stanislaus (Oct, 92 & Nov, 97)
Riverside (Sept, 98); and San Bernardino Sept, 98
Number of
Sites
1 each
1
2 each
2
3 each
4
6
8
1 each
Samples
1 each
2
2 each
o
J
3 each
4
6
10
1 each
Concentration
Alpha (fig/Kg)
1
2
In STORET22, data are reported on monitored sediment (2 meter deep) from September,
2001 to January, 2003 for a- and p-endosulfan and/or endosulfan sulfate in eleven
counties in California.  Reported data show a single detection (one sample out of 84
samples) with a concentration of 9.01ng/g for a- and P-endosulfan (Table 3.19).
Examination of the surface water data associated with this sediment sample shows no
detection of any endosulfan species

Table 3.19 Summary of bottom sediment monitoring data for California
County
Alameda
Contra Costa
Imperial
Los Angeles
Marin
Riverside
San Bernardino
San Mateo
Santa Clara
Solano
Ventura
Overall
Monitoring Year(s)
September, 2001
September, 2001 & June, 2002
May, September & October, 2002
January, 2003
October, 2001 & June 2002
May, September & October, 2002
May & October, 2002
June, 2002
June, 2002
June, 2002
November, 2001
September, 2001 to January, 2003
No of Sites
2
2
20
1
3
6
2
2
2
1
1
42
No of Samples *
4
3
46
1
4
13
4
3
3
1
2
84
* Number of samples may not match the number of sites because of sampling the same site at different dates

In conclusion, sediment data are summarized as follows:
    •   No CDPR data;
    •   NAQWA (1992-1998):  18 counties, 49 sites, and 58 samples: Alpha <2 ppb;
       Others Not determined; and
    •   STORET (2002-2003) 11 Counties; 42 sites, and 84 samples: Only single sample
       with 9.01 ng/g concentration for alpha endosulfan

3.2.4.4 Atmospheric Monitoring Data
22 STORET Data base URL: http://www.epa.gov/storet/dw home.html
                                       92

-------
a-, p-endosulfan and endosulfan sulfate may be considered as semi-volatile chemicals.
Vapor pressures for the three chemicals are 4.56xlO"5 torr for a-endosulfan, 2.40xlO"5 torr
for P-endosulfan, and 9.75xlO"6 torr for endosulfan sulfate.  Vapor pressure values
suggest that the three chemicals may volatilize into the air from dry soil/plant surfaces.
Calculated Henry's Law constants are 3.03xlO"6 atm m3 mol^for a-endosulfan, 1.38xlO"6
atm m3 mol"1 for P-endosulfan, and 1.64xlO"5 atm m3 mol^for endosulfan sulfate. Values
indicate that the three chemicals may also volatilize from water or wet soils.  The three
endosulfan chemical species appear to have the ability to partition into air in relatively
important amounts and persist long enough to be transported by air into distant areas.
Reported half lives in the air were 2.7 days for a-endosulfan and 15 days for P-
endosulfan (Schenker et al. 2005), and 2.6 days for endosulfan sulfate as estimated by
EpiSuit V 4.023 . Chemicals with half-life of >2 days in  the air can be considered to be
persistent in this media (SAP, 2009)24.

Based on physiochemical and fate properties,  endosulfan may move from the  application
site to edge and near-field by drift and can be  carried away by short and long-range
transport. Air monitoring and modeling may be used to evaluate the importance of these
two dissipation routes for exposure to edge/near field and nearby and far away areas.

Drift Monitoring (Edge & Near-field)

A 24-hour drift monitoring study (referred to in the study description as application site
monitoring) was conducted during the month of April in San Joaquin County (CDPR,
1998)25. The study included four air sampling stations placed on north, south, east, and
west sides of an 8.5 acre apple orchard field (at 54, 27, 21 and 33 ft from field edge,
respectively). Endosulfan was applied by ground equipment at a rate of 1.5 Ibs a.i. /A to
represent application to apples and cherries (current rate  can be up to 2.5 Ibs a.i. /A). At
each site, samples were collected at seven intervals: 1.00, 1.90, 4.00,  8.10, 9.50, 23.75,
and 24.00 hours following application.

A total of 28 sample events were collected within the 24-hour period at the four field
locations. This study represents a specific pesticide application followed by drift
monitoring at specific times/locations for endosulfan leaving the application field. Data
show relatively high detects/concentrations for alpha endosulfan compared to low
detect/concentrations for beta endosulfan and  no detects  for endosulfan sulfate (
Table 3 20)
23 The Estimation Programs Interface EPI Suite: http://www.epa.gov/oppt/exposure/pubs/episuitedl.htm
24 SAP (The FIFRA Scientific Advisory Panel). 2009. Minutes for the SAP Meeting on selected issues
associated with risk assessment process for pesticides with persistent, bio-accumulative and toxic
characteristics. October 28-31, 2008. Arlington, VA. (Document No. 2009-01 dated Jan 29, 2009).

25 CDPR, 1998 Report for the Air Monitoring of Endosulfan in Fresno County (Ambient) and in San
Joaquin County (Application), Air Resources Board, California Environmental Protection Agency,
Monitoring and Laboratory Division, Engineering and Laboratory Branch, Project No. C96-034
Dated April 17, 1998
                                         93

-------
 Table 3.20 Frequency and concentrations of endosulfan species in ambient air
Endosulfan Species
Alpha
Beta
Sulfate
% of Total Samples In various Detection Categories *
"No Detect"
4%
36%
100%
Low Concentration
Detects
4%
14%
0%
High concentration
Detects
92%
50%
0%
Totals
100%
100%
100%
 * Detection Categories (no categories specified for endosulfan sulfate "SO4"):
 "None": <3.1 ng/m3 for alpha, and <6.2 ng/m3 for beta
 "Low Detects" :   3.1 to <10 ng/m3 for alpha, and 6.2 to <20 ng/m3 for beta
 "High Detects":  >10 ng/m3 for alpha; > ng/m3 for beta. These detects ranged from 4.1-140 ng/m3 for
 alpha endosulfan and were only three detects at  13, 13 and 26 ng/ m3 for beta endosulfan

 Detection/Quantification limits (LOD= Limit of detection; LOQ= Limit of quantification
 Assuming 0.96 m3 samples (8 hours sampling period @ 2 Liters/Min): Alpha: LOD=3.1 ng/m3 and LOQ=
 10ng/in3,Beta.LOD=6^andLOQ=20ng/^


 Figure 3.8 summarizes monitoring data reported for a- and P- endosulfan (endosulfan
 sulfate was not detected over the detection limit).
                  Alpha
                  (East)
Alpha
(South)
 Beta
(South)
Alpha
(West)
 Beta
(West)
                     ]VI orate red Endosulfan Species and direction of the station location from field
Note 1: 4 out of scale concentrations are 1,800; 3,800; 1,200 & 1,200 ng/m3 from left to right, respectively
 Figure 3.8 Concentration profiles for a- and P- endosulfan at the edge and near-field in ambient air
 following application in San Joaquin County, California

 Additionally, data suggest that applied 70:30 mixture of a- and P- endosulfan drifted
 from the application site in disproportional quantities in the four main directions.
 Endosulfan sulfate was not found above the LOQ in any of the application samples. On
 the average, the ratio between a- and P- endosulfan in drifted material was in the range of
 7-23:1 compared to nearly 2:1 in the applied formulation. High concentration of a-
 endosulfan (up to 2xp- endosulfan) may be explained by the disproportional quantities in
 the applied parent.  However, observed enrichment of drifted material with a- endosulfan,
 up to 23xp~ endosulfan, could not be fully explained. Higher vapor pressure of a-
 compared to P- endosulfan may have contributed to observed result.
                                           94

-------
Results indicate that concentrations of a- and P- endosulfan were highest in drift moving
through the east edge of the field followed by the north, then the south with lowest
concentrations moving to the west. Additionally, peak concentrations in the drift were
generally observed within the first four hours after application. The relative drift
distribution was partially related to wind speed and direction.  Wind rose data throughout
the first five periods were from the west and north-west and the south west at the end of
these periods.  Wind directions during the last two periods were from the west, north,
north-west with additional wind from the north-east at the start of the last period. Other
factors may have also contributed to observed results such as application direction
(application started at the north-west corner of the plot with the rows oriented east/west),
average width of area downwind, distance of sampling station from field edge.  Other
factors may have also contributed to observed variations including temperature, relative
humidity and sun light.  Changes in these parameters are expected because the
experiment started early in the morning of 4/8/97 at 5:30 AM and ended in the morning
of 4/10/97  at 8.00 AM.

Short-Range Transport Monitoring

Levels of twelve pesticides, including total endosulfan26, were determined in ambient air
collected at three residential locations within the vicinity of Salinas, California (CDPR,
1985)27. The study coincided with expected time of pesticide applications but was not
designed to coincide with any specific pesticide application sites or times.  At each site, a
three day sampling period was used with four six hour sampling times starting from mid
night, early morning (6:00 AM), noon, and night (6:00 PM).  Endosulfan was detected in
two out of the twelve air samples in a site located 1,200 ft from any agricultural field and
in four out of 12 samples in another site located at only  190 ft from any agricultural field.
All of the six detections were within a range of 37- 55 ng/m3 (31-46 ppt)28. Positive
detections were in samples collected early in the day, coinciding with the time when
pesticides were applied to nearby fields, although wind speeds were lowest.

In another  ambient air monitoring  study, a- and P- endosulfan and endosulfan sulfate
were determined in ambient air collected every 24-hours at five sampling sites in
relatively high-population areas in Fresno County (CDPR,  1998)29. A total of 125
samples were collected within five weeks at four sites. Sites were in proximity to cotton
and grape growing areas (two sites were within 50-100 yards and two were within 0.75-2
miles).  Sampling periods were during the months of July and August to coincide with the
26 Note: It was not clear from the study report whether concentrations were for total endosulfan but it
appears that at least a- and (3- endosulfan were monitored.
27 CDPR, 1985 Monterey County Residential Air Monitoring By RJ. Sava, et al, Environmental Hazards
Assessment Program, Department of Pesticide Regulation, California Environmental Protection Agency,
Sacramento, CA 95814-5624 dated December 1985, Reprinted December 1995
28 pptt= ng/m3 multiplied by 0.833
29 CDPR, 1998 Report for the Air Monitoring of Endosulfan in Fresno County (Ambient) and in San
Joaquin County (Application), Air Resources Board, California Environmental Protection Agency,
Monitoring and Laboratory Division, Engineering and Laboratory Branch, Project No. C96-034
Dated April 17, 1998
                                         95

-------
application of endosulfan on cotton and grapes. Again, no specific pesticide application
sites or times were associated with the study.  Data show relatively high
detects/concentrations for alpha endosulfan compared to low detect/concentrations for
beta endosulfan and no detects for endosulfan sulfate (Table 3.21).

Table 3.21 Frequency and concentrations of endosulfan species in ambient air
Endosulfan Species
Alpha
Beta
Sulfate
% of Total Samples in various Detection Categories *
"None"
18%
68%
100%
Low Concentration
Detects
9%
29%
0%
High Concentration
Detects
73%
3%
0%
Totals
100%
100%
100%
* Detection Categories (no categories specified for endosulfan sulfate "SO4"):
"None": O.58 ng/m3 for alpha, and <1.91 ng/m3 for beta
"Low Detects" :   0.58 to <2.49 ng/m3 for alpha, and 1.91 to <8.16 ng/m3 for beta
"High Detects":  >2.49 ng/m3 for alpha; >8.16 ng/m3 for beta. These detects ranged from 4.1-140 ng/m3
for alpha endosulfan and were only three detects at 13, 13 and 26 ng/ m3 for beta endosulfan

Detection/Quantification limits (LOD= Limit of detection; LOQ= Limit of quantification
Assuming 2.88 m3 samples (24 hours sampling period @ 2 Liters/Min): Alpha: LOD=1.1 ng/m3 and
LOQ= 3.8 ng/ m3; Beta: LOD= 3.8 and LOQ= 12 ng/ m3; and LOQ= 12.8 ng/ m3 for SO4 (endosulfan
sulfate)


Data suggest that application of endosulfan may have caused partitioning of a- and P-
endosulfan into ambient air to support maximum concentrations of 140 ng/m3 for a-
endosulfan (the 70% isomer in the formulation) and 26 ng/m3 for P- endosulfan (the 30%
isomer in the formulation).  The same conclusion applies for detection frequencies of
low/high concentrations.  These frequencies were 82% for a- endosulfan and 31% for P-
endosulfan (refer to Table 3.21: 82%= 9%+73%; and 31%= 29%+3%).  Figure 3.9
summarizes data for the main species detected, a- endosulfan.
                                          96

-------
Figure 3.9 Concentration profiles for a- endosulfan in ambient air during a period of five weeks at 4
sites in Fresno County, California.	
            50 n
            40
            30
            20
            10
                  Site 1
                                      Site 2               Site 3

                                  Time: From zero to 5 weeks for each site
                                                                             Site 4
Note 1: Distance from Cotton fields: sites 1&2 = 150-300 ft.; Site 3= 0.75-1.0 mile; Site 4= 1-2 miles.
Note 2: The 4 out of scale concentrations are 130, 120, 140, and 70 ng/m3 from left to right, respectively

Based on data shown in Figure 3.9, a- endosulfan appears to be present in the ambient
air of near field areas where its parent is applied to cotton and grapes. Although
relatively high spikes of concentrations exist, concentration levels of the chemical appear
to be mainly in the range of 5 to 35 ng/m3 (4- 29 ppt) which is lower than concentrations
monitored in 1985 (37- 55 ng/m3 = 31- 46 ppt). These near field areas were in the range
of 50 yards to 2 miles from possible application sites suggesting that endosulfan had
partitioned into the air and moved within short-range distances.  Dissipation of
endosulfan into and within the air mass is probably related to a combination of drift and
volatilization.

Long-Range Transport Monitoring

Atmospheric transport from CA central valley and re-deposition in the Sierra Nevada
Mountains were documented for a- and P- endosulfan and endosulfan sulfate. As a result
of atmospheric transport, the three chemicals have been found in the mountainous regions
of California in air surface water, sediment, precipitation/snow-pack, and biota.

Endosulfans were  detected in air samples taken from Sequoia and King Canyon National
Parks (Landers et al. 2008) in average concentrations of approximately  1,250 pg/g dry
XAD. Levels of endosulfan  detected in the California National Parks were greater than
those measured in  any  other park in the study (Sequoia/Kings Canyon, Rocky Mountain,
Mount Rainier, Glacier, Denali, Noatak, and Gates of the Arctic).

In surface water, residues of a-, p-endosulfan and endosulfan sulfate were present in
water from both Tablelands  and Sixty Lakes in the Sierra Nevada  Mountains of
California (Fellers et al. 2004).  Similar concentrations were detected for a- and P-
endosulfan from the Tablelands and Sixty Lakes sampling areas, however endosulfan
sulfate concentrations in Tablelands were almost an order of magnitude higher than in
                                         97

-------
Sixty Lakes and almost four times greater than P- endosulfan concentrations from the
same locations (Table 3.22). The a-isomer was also detected at two locations and at
various depths in Lake Tahoe in concentrations ranging from < 0.004 to 0.26 ng/L which
are similar to that found in snow in the surrounding basin (McConnell et al. 1998).
Endosulfan was one of the most common current use pesticide detected in sediment
collected from lakes in Sequoia and Kings Canyon National Parks with an average flux of
approximately 130 ng/m2/y

Table 3.22 Concentration of pesticides (ng/L) in surface water samples collected at the
Tablelands (Sequoia National Park), and the Sixty Lake Basin (Kings Canyon National
Park) California, USA (Fellers et al. 2004).
Chemical
a-endosulfan
P-endosulfan
Endosulfan sulfate
Compound
Detection limit
0.03
0.05
0.03
Tablelands
G049
0.78
0.40
2.9
G054
1.0
0.42
2.2
Sixty Lakes
S545
0.30
1.8
0.33
S471
0.37
0.17
0.40
In precipitation/snow pack, endosulfan and endosulfan sulfate were detected in seasonal
snow pack samples at seven national parks in the Western United States, including
mountainous regions in California (Sequoia, Rocky Mountain, Mount Rainier, Glacier,
Denali, Noatak, and Gates of the Arctic; Hageman et al. 2006). In this study,
concentrations of total endosulfan measured in seasonal snow pack ranged from < 0.0040
to 1.5 ng/L and the percent contribution of endosulfan sulfate to the total endosulfan
concentration averaged 24.0%. McConnell et al. (1998) reported endosulfan was present
in snow and rain samples from two elevations in Sequoia National Park and the southern
Sierras,  a region adjacent to California's Central Valley which is among the heaviest
pesticide use areas in the U.S. Levels of a-endosulfan found in precipitation range from <
0.035 to 6.5  ng/L while P-endosulfan ranged from < 0.012 to 1.4 ng/L (McConnell et al.
1998).

Endosulfan was also found in mountain yellow-legged frogs (Rana muscosa) from two
areas in the Sierra Nevada Mountains in California (Sixty Lakes Basin, Kings Canyon
National Park and Tablelands, Sequoia national Park; Fellers et al. 2004). Of the two
endosulfan isomers and the sulfate degradate, only the a-isomer was observed at levels
above quantitation.  Concentrations of a-endosulfan averaged 0.56 (± 0.36) and 0.45 (±
0.24) ng/g wet weight at Tablelands and Sixty Lakes Basin, respectively.  These
Concentrations were not significantly different from the two sites (Fellers et al 2004).
Endosulfan sulfate was measured in fish from Sequoia and Kings Canyon National Parks
at average concentrations of approximately 30 ng/g lipid, which was one of the highest
concentrations reported in fish throughout the western parks (Landers et al. 2008).

Total Endosulfan (sum of a, P, and sulfate) was measured in lichen and conifer samples
in Sequoia and Kings Canyon National Parks at average concentrations of 488 and 192
ng/g, respectively (Landers et al. 2008). In this study endosulfans, along with dacthal,
dominated the total pesticide laoding in vegetation samples from the lower 48 states.
                                       98

-------
3.3 Terrestrial Animal Exposure Assessment

A primary concern of endosulfan application is that birds and mammals may be exposed
through oral or dietary exposure to vegetative plant material or insects when foraging in
treated fields for nesting material or food. Estimation of endosulfan concentrations in
wildlife food items focuses on quantifying possible dietary ingestion of residues on
vegetative matter and insects. The EFED terrestrial exposure model T-REX (T-REX,
Version 1.4.1, dated October 9, 2008) simulates a one-year time period and is used to
estimate exposures and risks to avian (surrogate for terrestrial-phase amphibians and
reptiles) and mammalian species. Avian and mammalian toxicity, chemical application,
and foliar dissipation half-life data are input for the model, which provides estimates of
exposure concentrations (EECs) and risk quotients (RQs). Specifically, the model
provides estimates of concentrations (upper-bound and mean) of chemical residues on the
surface of different types of foliage and insects that may be dietary sources of exposure to
avian, mammalian, reptilian, or terrestrial-phase amphibian receptors. The surface residue
concentration (ppm) is estimated by multiplying the application rate (pounds active
ingredient per acre) by a value specific to each food item (termed the Hoerger-Kenaga
estimates).  For multiple applications, the EEC is determined by adding the mass on the
surface immediately following the application to the mass of the  chemical still present on
the surfaces on the day of application (determined based on first order kinetics using the
foliar half-life as the rate constant). The Hoerger-Kenaga estimates and a more  detailed
discussion of the methodology implemented by T-REX can be found at
http://www.epa.gov/oppefedl/models/terrestrial/.

T-REX is also used to calculate EECs for terrestrial insects exposed to endosulfan via
spray applications. Dietary-based EECs calculated by T-REX for small (broadleaf
plants/small insects dietary category) and large (fruits/pods/seeds/large insects dietary
category) invertebrates (units of a.i./g) are used to bound  an estimate of exposure to
terrestrial invertebrates. Available acute contact toxicity data for  bees (or other  terrestrial
invertebrates) exposed to endosulfan (in units of jig a.i./organism), are converted to jig
a.i./g (of organism) by dividing by the body weight of an individual organism used in the
test.  For honey bees, the body weight of a bee is assumed to be 0.128 g.  The EECs are
later compared to the adjusted acute contact toxicity data for terrestrial invertebrates in
order to derive RQs for terrestrial invertebrates.

Terrestrial EECs were derived for use categories (apples, cotton,  grapes, lettuce, pecans,
potatoes, tobacco, and tomatoes) using current application rates and intervals between
applications. Uncertainties in the terrestrial EECs are primarily associated with a lack of
data on interception and subsequent dissipation from foliar surfaces. Use-specific input
values, including number of applications, application rate, foliar half-life and application
interval are provided in Table 3.23.  The maximum number of applications per year, the
minimum application intervals, and the maximum application rates for each crop selected
as the representative crop for a "Crop Category" were derived from the product labels
(Table 2.3) and Willis and  McDowell (1987) summarized seven studies which evaluated
the foliar persistence of endosulfan on a variety of crops.  Foliar half-lives ranged from  1
                                        99

-------
to 5 days on a variety of crops (cotton, grapes, pears, tobacco, alfalfa, beets, and leafy
vegetables) in studies conducted in California, Arizona, Kentucky, Canada, and
Australia. The mean of the 13 reported half-life values was 3.2 days (standard deviation
of 1.4 days). The upper 90th percent confidence interval value for the mean (4 days) was
used as the foliar dissipation rate for modeling purposes.

The model was run for agricultural uses with the maximum single application rate and
number of applications as proposed on the labels; however, it should be noted that in
some instances recommended label use may include either a  single application of the
maximum allowable seasonal rate or multiple applications of lower rates totaling the
season maximum. The scenario which yielded the highest EEC and RQ values were used
for each use category where multiple application methods were possible. The upper-
bound Kenaga nomogram estimates reported by T-REX have been used for derivation of
the EECs for the terrestrial organisms and their potential prey (Table 3.23). An example
output from T-REX is available in Appendix F.
Table 3.23 Summary of Dose and Dietary-based EECs Used for Estimating Dietary Risks to Terrestrial
Organisms using T-REX ver. 1.4.1.
Use
Category
Almonds,
hazelnut &
walnut
Citrus
Broccoli,
cabbage,
Chinese
cabbage,
cauliflower,
kohlrabi
Kale,
Collards &
Mustard
Green
Sweet corn
App
Rate (Ib
a.i./A, #
Apps,
Interval
(da))
2, 1,
2.5, 1,
1,2,7
0.75, 1,
1.5, 1,
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Dose-Based EECs
Avian size class (grams)
20
546.67
250.56
307.50
34.17
7.59
683.34
313.20
384.38
42.71
9.49
354.60
162.52
199.46
22.16
4.92
205.00
93.96
115.31
12.81
2.85
410.00
187.92
230.63
25.63
100
311.74
142.88
175.35
19.48
4.33
389.67
178.60
219.19
24.35
5.41
202.21
92.68
113.74
12.64
2.81
116.90
53.58
65.76
7.31
1.62
233.80
107.16
131.51
14.61
1000
139.57
63.97
78.51
8.72
1.94
174.46
79.96
98.13
10.90
2.42
90.53
41.49
50.92
5.66
1.26
52.34
23.99
29.44
3.27
0.73
104.68
47.98
58.88
6.54
Mammalian size class
(grams)
15
457.64
209.75
257.42
28.60
6.36
572.05
262.19
321.78
35.75
7.95
296.85
136.06
166.98
18.55
4.12
171.62
78.66
96.53
10.73
2.38
343.23
157.31
193.07
21.45
35
316.29
144.97
177.91
19.77
4.39
395.37
181.21
222.39
24.71
5.49
205.16
94.03
115.40
12.82
2.85
118.61
54.36
66.72
7.41
1.65
237.22
108.73
133.44
14.83
1000
73.33
33.61
41.25
4.58
1.02
91.67
42.01
51.56
5.73
1.27
47.57
21.80
26.76
2.97
0.66
27.50
12.60
15.47
1.72
0.38
55.00
25.21
30.94
3.44
Dietary
EECs
480.00
220.00
270.00
30.00

600.00
275.00
337.50
37.50

311.35
142.70
175.14
19.46

180.00
82.50
101.25
11.25

360.00
165.00
202.50
22.50
                                       100

-------
Table 3.23 Summary of Dose and Dietary-based EECs Used for Estimating Dietary Risks to Terrestrial
Organisms using T-REX ver. 1.4.1.
Use
Category

Cotton
(ground)
Cotton
(Areal)
Apples
Apricot,
nectarine,
peach,
cherry, pear,
plum &
prune
Lettuce &
Brussels
sprouts
Cucumber,
melons,
pumpkin &
squash
Eggplant
Ornamentals
& shade
trees
App
Rate (Ib
a.i./A, #
Apps,
Interval
(da))

1,2,
0.75, 2, 7
2.5, 1,
2.5, 1,
1,2,5
1,2,7
0.5, 1,
0.5,6, 10
Dietary Category
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Dose-Based EECs
Avian size class (grams)
20
5.69
273.34
125.28
153.75
17.08
3.80
265.95
121.89
149.60
16.62
3.69
683.34
313.20
384.38
42.71
9.49
683.34
313.20
384.38
42.71
9.49
388.26
177.95
218.40
24.27
5.39
354.60
162.52
199.46
22.16
4.92
136.67
62.64
76.88
8.54
1.90
166.01
76.09
93.38
10.38
2.31
100
3.25
155.87
71.44
87.68
9.74
2.16
151.66
69.51
85.31
9.48
2.11
389.67
178.60
219.19
24.35
5.41
389.67
178.60
219.19
24.35
5.41
221 .40
101.48
1 24.54
13.84
3.08
202.21
92.68
113.74
12.64
2.81
77.93
35.72
43.84
4.87
1.08
94.67
43.39
53.25
5.92
1.31
1000
1.45
69.78
31.98
39.25
4.36
0.97
67.90
31.12
38.19
4.24
0.94
174.46
79.96
98.13
10.90
2.42
174.46
79.96
98.13
10.90
2.42
99.12
45.43
55.76
6.20
1.38
90.53
41.49
50.92
5.66
1.26
34.89
15.99
19.63
2.18
0.48
42.38
19.43
23.84
2.65
0.59
Mammalian size class
(grams)
15
4.77
228.82
104.88
128.71
14.30
3.18
222.64
102.04
125.23
13.91
3.09
572.05
262.19
321.78
35.75
7.95
572.05
262.19
321.78
35.75
7.95
325.03
148.97
182.83
20.31
4.51
296.85
136.06
166.98
18.55
4.12
114.41
52.44
64.36
7.15
1.59
138.97
63.70
78.17
8.69
1.93
35
3.29
158.15
72.48
88.96
9.88
2.20
153.87
70.52
86.55
9.62
2.14
395.37
181.21
222.39
24.71
5.49
395.37
181.21
222.39
24.71
5.49
224.64
102.96
126.36
14.04
3.12
205.16
94.03
115.40
12.82
2.85
79.07
36.24
44.48
4.94
1.10
96.05
44.02
54.03
6.00
1.33
1000
0.76
36.67
16.81
20.63
2.29
0.51
35.68
16.35
20.07
2.23
0.50
91.67
42.01
51.56
5.73
1.27
91.67
42.01
51.56
5.73
1.27
52.08
23.87
29.30
3.26
0.72
47.57
21.80
26.76
2.97
0.66
18.33
8.40
10.31
1.15
0.25
22.27
10.21
12.53
1.39
0.31
Dietary
EECs

240.00
110.00
135.00
15.00

233.51
107.03
131.35
14.59

600.00
275.00
337.50
37.50

600.00
275.00
337.50
37.50

340.91
156.25
191.76
21.31

311.35
142.70
175.14
19.46

120.00
55.00
67.50
7.50

145.76
66.81
81.99
9.11

101

-------
Table 3.23 Summary of Dose and Dietary-based EECs Used for Estimating Dietary Risks to Terrestrial
Organisms using T-REX ver. 1.4.1.
Use
Category
Potato
Sweet
potato
Dry beans
(except
Lima), peas
& pepper
Carrot
Celery
Strawberry
Tomato
App
Rate (Ib
a.i./A, #
Apps,
Interval
(da))
1,2,5
0.5,3,5
1,2,5
1,1,
1, 1,
1,2, 15
1,2,7
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Dose-Based EECs
Avian size class (grams)
20
388.26
177.95
218.40
24.27
5.39
218.29
100.05
122.79
13.64
3.03
388.26
177.95
218.40
24.27
5.39
273.34
125.28
153.75
17.08
3.80
273.34
125.28
153.75
17.08
3.80
293.65
134.59
165.18
18.35
4.08
354.60
162.52
199.46
22.16
4.92
100
221 .40
101.48
1 24.54
13.84
3.08
124.48
57.05
70.02
7.78
1.73
221 .40
101.48
1 24.54
13.84
3.08
155.87
71.44
87.68
9.74
2.16
155.87
71.44
87.68
9.74
2.16
167.45
76.75
94.19
10.47
2.33
202.21
92.68
113.74
12.64
2.81
1000
99.12
45.43
55.76
6.20
1.38
55.73
25.54
31.35
3.48
0.77
99.12
45.43
55.76
6.20
1.38
69.78
31.98
39.25
4.36
0.97
69.78
31.98
39.25
4.36
0.97
74.97
34.36
42.17
4.69
1.04
90.53
41.49
50.92
5.66
1.26
Mammalian size class
(grams)
15
325.03
148.97
182.83
20.31
4.51
182.74
83.76
102.79
11.42
2.54
325.03
148.97
182.83
20.31
4.51
228.82
104.88
128.71
14.30
3.18
228.82
104.88
128.71
14.30
3.18
245.83
112.67
138.28
15.36
3.41
296.85
136.06
166.98
18.55
4.12
35
224.64
102.96
126.36
14.04
3.12
126.30
57.89
71.04
7.89
1.75
224.64
102.96
126.36
14.04
3.12
158.15
72.48
88.96
9.88
2.20
158.15
72.48
88.96
9.88
2.20
169.90
77.87
95.57
10.62
2.36
205.16
94.03
115.40
12.82
2.85
1000
52.08
23.87
29.30
3.26
0.72
29.28
13.42
16.47
1.83
0.41
52.08
23.87
29.30
3.26
0.72
36.67
16.81
20.63
2.29
0.51
36.67
16.81
20.63
2.29
0.51
39.39
18.05
22.16
2.46
0.55
47.57
21.80
26.76
2.97
0.66
Dietary
EECs
340.91
156.25
191.76
21.31

191.67
87.85
107.81
11.98

340.91
156.25
191.76
21.31

240.00
110.00
135.00
15.00

240.00
110.00
135.00
15.00

257.84
118.18
145.03
16.11

311.35
142.70
175.14
19.46

N/A = Non-applicable

Upper-bound Kenega nomogram values reported by T-REX are used for derivation of
dietary EECs for evaluating direct and/or indirect effects to terrestrial phase CRLF, CIS,
SFGS, SMHM, SJKF, BCB, and VELB.  Potential direct acute and chronic effects of
endosulfan to the terrestrial-phase CRLF, SFGS and CTS are derived by considering
                                      102

-------
dietary-based exposures modeled in T-REX for a small bird (20g) consuming small
invertebrates.

Potential direct acute and chronic effects specifically to the SMHM are derived by
considering dose- and dietary-based EECs modeled in T-REX for a small mammal (15 g)
consuming a variety of dietary items (Table 3.23). Potential direct acute and chronic
effects specifically to the  SJKF are derived by considering dose- and dietary-based EECs
modeled in T-REX for a large mammal (1,000 g) consuming a variety of dietary items
(Table 3.23).

3.4 Terrestrial Plant Exposure Assessment

No terrestrial plant toxicity data were available for endosulfan that were considered
acceptable for quantitative use in risk estimation. Therefore, risks to terrestrial plants
could not be quantitatively assessed in this risk assessment but instead were qualitatively
considered.
                                       103

-------
4     Effects Assessment

This assessment evaluates the potential for endosulfan (and its primary degradate of
concern, endosulfan sulfate) to directly or indirectly affect the CRLF, CIS, SFGS,
SMHM, SJKF, BCB, and VELB or modify their designated critical habitat.  As
previously discussed in Section 2.7, assessment endpoints for the effects determination
for each assessed species include direct toxic effects on the survival, reproduction, and
growth, as well as indirect effects, such as reduction of the prey base or modification of
its habitat. In addition, potential modification of critical habitat is assessed by evaluating
effects to the PCEs, which are components of the critical habitat areas that provide
essential life cycle needs of each assessed species.  Direct effects to the aquatic-phase
CRLF and CTS are based on toxicity information for freshwater fish, while terrestrial-
phase CRLF, CTS and SFGS effects are based on avian toxicity data, given that birds are
generally used as a surrogate for terrestrial-phase amphibians and reptiles when suitable
data for terrestrial-phase amphibians are not available.

As described in the Agency's Overview Document (U.S. EPA, 2004), the most sensitive
endpoint for each taxon is used for risk estimation. For this assessment, evaluated taxa
include  freshwater fish (used as a surrogate for aquatic-phase amphibians), freshwater
invertebrates, estuarine/marine fish, estuarine/marine invertebrates, aquatic plants (non-
vascular only;  no data were available for vascular aquatic plants), birds (used as a
surrogate for terrestrial-phase amphibians and reptiles), mammals, terrestrial
invertebrates, and terrestrial plants (plants are only qualitatively evaluated due to lack of
appropriate data). Acute (short-term) and chronic (long-term) toxicity information is
characterized based on registrant-submitted studies and a comprehensive review of the
open literature on endosulfan

Toxicity endpoints are established based on data generated from guideline studies
submitted by the registrant,  and from open literature studies that meet the criteria for
inclusion into the ECOTOX database maintained by EPA/Office of Research and
Development (ORD; http://www.epa.gov/med/Prods_Pubs/ecotox.htm). Open literature
data presented in this assessment were obtained from ECOTOX information obtained on
November 5, 2008.  In order to be included in the ECOTOX database, papers must meet
the following minimum criteria:

       (1)    the toxic effects are related to single chemical exposure;
       (2)    the toxic effects are on an aquatic or terrestrial plant or animal species;
       (3)    there is a biological effect on live, whole  organisms;
       (4)    a concurrent environmental chemical concentration/dose or application
              rate is reported; and
       (5)    there is an explicit duration of exposure.

Data that pass  the ECOTOX screen are evaluated along with the registrant-submitted
data, and may be incorporated qualitatively or quantitatively into this endangered species
assessment.  In general, effects data in the open literature that are more conservative (i.e.,
                                        104

-------
show a lower toxicity endpoint) than the registrant-submitted data are evaluated further.
The degree to which open literature data are quantitatively or qualitatively characterized
for the effects determination is dependent on whether the information is relevant to the
assessment endpoints (i.e.., survival, reproduction, and growth) identified in Section 2.8.
For example, endpoints such as biochemical modifications are likely to be qualitatively
evaluated, because quantitative relationships between biochemical modifications in the
laboratory and reduction in species survival, reproduction, and/or growth in the field are
generally not available. Although this effects determination relies on endpoints that are
relevant to the assessment endpoints of survival, growth, or reproduction, it is important
to note that the full suite of sublethal endpoints potentially available in the effects
literature (regardless of their significance to the assessment endpoints) are considered to
define the action area for endosulfan.

The sublethal endpoint used to define the Action Area was based on reduction in growth
for the sheepshead minnow (Hansen and Cripe, 1991; ECOTOX reference # 14143), who
reported a LOAEC of < 0.27 ug a.i./L resulting from 28-d, flow through exposures.

Citations of all open literature not evaluated as part of this assessment because they were
either rejected by the ECOTOX screen or accepted by ECOTOX but not used (e.g.., the
endpoint is less sensitive) are included in Appendix G. Appendix G also includes a
rationale for rejection of those  studies that did not pass the ECOTOX screen and those
that were not evaluated as part of this endangered species risk assessment.

A detailed spreadsheet of the available ECOTOX open literature data, including the full
suite of lethal and sublethal endpoints is presented in Appendix H. Appendix J contains
a review of all the toxicity data considered in this assessment for aquatic species, while
Appendix K contains the same information for terrestrial species. Appendix M also
includes a summary of the human health effects data for endosulfan. Because the stability
of endosulfan TGAI in aquatic tests is a concern due to its volatility, preference was
given to those studies that quantified endosulfan exposure via analytical measurements
and where appropriate, used a flow-through exposure regime. In some cases, however,
these data were lacking (e.g., aquatic nonvascular plants), and toxicity values based on
nominal concentrations were used to fulfill a data gap. In general, studies from the open
literature do not provide sufficient documentation for evaluating all or most of the
OPPTS Guideline evaluation criteria. Consistent with the Overview Document (U.S.EPA
2004), if sufficient documentation was provided to judge that the open literature studies
were scientifically valid and acceptable endpoints were determined that were below the
'benchmark' values, they were typically classified as supplemental—but acceptable for
quantitative use. If sufficient documentation was not provided to comprehensively  judge
the scientific validity of the study but no information existed that suggested the study
would be invalid, these studies were typically classified  as supplemental-but acceptable
for qualitative use. This classification was often applied to studies with nominal exposure
concentrations.

In addition to registrant-submitted and open literature toxicity information, other sources
of information, including use of the acute  probit dose response relationship to establish
                                        105

-------
the probability of an individual effect and reviews of the Ecological Incident Information
System (EIIS), are conducted to further refine the characterization of potential ecological
effects associated with exposure to endosulfan. A summary of the available aquatic and
terrestrial ecotoxicity information, use of the probit dose response relationship, and the
incident information for endosulfan are provided in Sections 4.1 through 4.4,
respectively.

As discussed in Section 2.4.1, endosulfan sulfate is the primary degradate of
toxicological concern for endosulfan (a+P). Available data suggest that the acute toxicity
of endosulfan sulfate can be similar to the parent isomers, or on occasion, somewhat less
toxic, depending on the species tested. For example, registrant-submitted studies of the
TGAI and endosulfan sulfate indicate similar acute toxicity to bluegill (1.7 vs. 3.8 ug/L,
respectively; MRID 38806 and 46382604). Comparison among two decapod crustaceans
(grass shrimp and mysid shrimp) indicates the acute toxicity of parent isomers and the
sulfate degrade are within the same  order of magnitude (1.3 and 7.9 ug/L, respectively;
MRID 5005824 and 46406401). For avian species, registrant-submitted data indicate
similar acute toxicity of endosulfan  TGAI and endosulfan sulfate based on the avian
acute oral study with bobwhite quail and subacute dietary studies with mallard and
bobwhite quail (MRID  137189; 136998; 160000; 22923; 46382604). Comparative acute
toxicity data available in the open literature forDaphnia magna, Hyalella azteca and
Oncorhynchus mykiss also support the finding of similar acute toxicity of the sulfate
degradate and parent TGAI; however, the exposure regime used in these tests is uncertain
due to lack of confirmatory analytical measurements (Wan et al., 2005; ECOTOX
reference # 87973).

Acceptable studies comparing the chronic toxicity of the TGAI and sulfate degradate
were not identified in this review. Therefore, consistent with the available data and the
previous endosulfan risk assessment (U.S.  EPA, 2007; D346213), the acute and chronic
toxicity of endosulfan (TGAI) and endosulfan sulfate will be assumed to be equivalent. A
detailed summary of the available ecotoxicity information for all endosulfan degradates
and formulated products can be found in Appendices H, J and K.

4.1 Toxicity of Endosulfan to Aquatic  Organisms

Table 4.1 summarizes the most sensitive aquatic toxicity endpoints, based on an
evaluation of both the submitted studies  and the open literature, as previously discussed.
A brief summary of submitted and open literature data considered relevant to this
ecological risk assessment for the CRLF, CTS. SFGS, SMHM,  SJKF,  BCB, and VELB is
presented below. Additional information on species life history is provided in
Attachments 1 and 3.
                                       106

-------
Table 4.1 Aquatic Toxicity Profile for Endosulfan (TGAI)
Assessment
Endpoint
Purity
(% a.i.)
Surrogate
Species
Toxicity Value Used
for Quantitative
Risk Estimates
(jig a.i./L)
Effects
Reference/Acceptability
Freshwater Fish and Aquatic-phase Amphibians (1)
Survival
Reproduction and Growth
96.2
Not
Available
Common Carp
(Cyprinus carpio)
Common Carp
(Cyprinus carpio)
LC50 = 0.1
NOAEC = 0.023
Mortality
Not Available
Sunderam et al. 1992; (
ECOTOX 5850)/ Supplemental
Estimated
Freshwater Invertebrates
Survival
Reproduction and Growth
Sediment Toxicity
96
Not
Available
99
Mayfly (Atalophlebia
australis)
Mayfly (Atalophlebia
australis))
Midge,
Chironomus tentans
LC50 = 0.6
NOAEC = 0.01
20-d NOAEC = 0.35
Mortality
Not Available
Mortality and
Emergence
Leonard, et al. 1999 (ECOTOX
20012)/Supplemental
Estimated (3)
47318101 /Acceptable
Estuarine/Marine Fish
Survival
Reproduction and Growth
96
Not
Available
Spot (Leiostomus
xanthurus)
Spot (Leiostomus
xanthurus)
LC50 = 0.09
NOAEC = 0.045
Mortality
Not Available
MRID 5005824/ Acceptable
(4)
Estimated
Estuarine/Marine Invertebrates
Survival
Reproduction and Growth
Sediment Toxicity
96
Not
Available

Pink shrimp (Penaeus
dourarum)
Pink shrimp (Penaeus
dourarum
Amphipod
(Leptocheirus
plumulosus)
LC50 = 0.04
NOAEC = 0.013
28-d NOAEC = 1.58
Mortality
Not Available
Growth
MRID 5005824/ Acceptable
Estimated
46929001 / Supplemental
Aquatic Non Vascular Plants
Survival and Growth
non-vascular
Green alga
(Pseudokirchneriella
subcapitatum)
EC50 = 428
Growth (Based on
cell counts)
DeLorenzo et al. 2002
(ECOTOX 65915)/Supplemental
Aquatic Vascular Plants
Survival and Growth
No acceptable data indentified
(1) Freshwater fish are used as a surrogate for aquatic-phase amphibians because no acceptable data for
quantitative use currently exist for endosulfan to aquatic-phase amphibians.
(2) The fathead minnow (Pimephalespromelas) acute to chronic ratio (ACR = 4.3) was used to estimate a
common carp chronic toxicity value because it is the most acutely sensitive freshwater fish species and no
chronic toxicity data are available for it.
(3) The daphnid (Daphnia magna) acute to chronic ratio (ACR = 61.5) was used to estimate a freshwater
shrimp chronic toxicity value because it is the most acutely sensitive freshwater invertebrate species and no
chronic toxicity data are available for it.
(4) The sheepshead minnow (Cyprinodon variegatus) acute to chronic ratio (ACR = 2.0) was used to
estimate a spot chronic toxicity value because it is the most acutely sensitive saltwater fish species and no
chronic toxicity data are available for it.
(5)The mysid shrimp (Americanmysis bahia) acute to chronic ratio (ACR =3.1) was used to estimate a spot
chronic toxicity value because it is the most acutely sensitive saltwater invertebrate species and no chronic
toxicity data are available for it.
(6) Sediment toxicity values are for endosulfan sulfate in sediment pore water.

Toxicity to  fish and aquatic invertebrates is categorized using the system shown in Table
4.2 (U.S. EPA, 2004). Toxicity categories for aquatic plants have not been defined.
                                              107

-------
Table 4.2 Categories of Acute Toxicity for Fish and Aquatic Invertebrates
LC50 (ppm)
<0.1
>0.1- 1
>1-10
> 10 - 100
>100
Toxicity Category
Very highly toxic
Highly toxic
Moderately toxic
Slightly toxic
Practically nontoxic
4.1.1 Toxicity to Freshwater Fish and Aquatic-Phase Amphibians

A summary of acute and chronic toxicity of endosulfan to freshwater fish and aquatic-
phase amphibians, including the toxicity of the primary degradate of concern (endosulfan
sulfate), is provided below in Sections 4.1.1.1 through 4.1.1.6.

4.1.1.1 Freshwater Fish: Acute Exposure (Mortality) Studies - Registrant submitted

The acute toxicity studies available to the Agency demonstrate that endosulfan can be
classified as very highly toxic to freshwater fish, with LC50 values ranging from 0.37 to
3.3 ug a.i./L (see Appendix J for details). The studies also demonstrate that the
formulated end use products with at least 33% a.i. are of similar acute toxicity to
freshwater fish (LC50 values from 0.47 to 5.6 ug a.i./L) as the TGAI. One formulation
containing only 4% endosulfan was substantially less acutely toxic to freshwater fish
(LC50 of 28 ug a.i./L).

4.1.1.2 Freshwater Fish: Acute Exposure (Mortality) Studies—Open Literature

In addition to the acute toxicity studies discussed above, several acute toxicity studies
with freshwater fish were identified from the open literature (see Appendix J) which
report more sensitive values. The  lowest of these more sensitive acute LC50 values is 0.1
ug a.i./L (Sunderam et al. 1992; ECOTOX reference #5850) for the common carp,
Cyprinus carpio. This study was evaluated and considered supplemental but appropriate
for quantitative use in this risk assessment. Therefore, an acute LC50 value of 0.1 ug
a.i./L will be used to quantitatively estimate acute risk to freshwater fish. Since no acute
mortality studies with aquatic-phase amphibians in the open literature have been
identified as acceptable for quantitative use within the context of this risk assessment, the
most sensitive LC50 value for freshwater fish will also be used  to estimate acute risks to
aquatic-phase amphibians.

4.1.1.3 Freshwater Fish: Chronic Exposure (Growth/Reproduction) Studies

Two acceptable life-cycle studies on fathead minnows (Pimephalespromelas) were
available to the Agency to evaluate the effects of chronic exposure to endosulfan (99%
a.i.) on freshwater fish (MRID 05008271, 45868601). The first  study (MRID 05008271)
demonstrated that chronic exposure to concentration as low as 0.40 ug a.i./L has the
potential to cause reproductive toxicity. Only 1% of the eggs hatched at this
concentration (see Appendix J for details). The NOAEC and LOAEC for this study are
0.20 and 0.40 ug a.i./L. The second  study (MRID 45868601) demonstrated a reduction in
                                       108

-------
the reproduction and growth of fathead minnows at 0.11 ug a.i./L and aNOAEC at 0.056
ug a.i./L. Although these studies are considered acceptable, a fathead minnow acute to
chronic ratio (ACR = 4.3) was used to estimate a common carp chronic toxicity value
because carp is the most acutely sensitive freshwater fish species (i.e., about a factor of
10 more sensitive than fathead minnow) and no chronic toxicity data for endosulfan are
available for it. The  estimated chronic NOAEC value for the common carp was
calculated as follows:

        Acute fathead minnow/chronic fathead minnow= acute carp/chronic carp

Where:

       the acute fathead minnow value is based on the LC50 value 0.86 ug a.i./L (MRID
       05008271),

       the chronic fathead minnow value is based on the NOAEC of 0.20 ug a.i./L from
       the same study and,

       the acute common carp value is based on the LC50 is 0.1 ug a.i./L as described
       previously (Sunderam et al. 1992; ECOTOX reference #5850).

Therefore, 0.86/0.20 = 0.1/X

       and,

       Estimated common carp NOAEC = (0.20 x 0.1)70.86 = 0.023 ug a.i./L.

This estimated NOAEC of 0.023 ug a.i./L for common carp will be used to quantitatively
estimate chronic risk of endosulfan to freshwater fish.  Since no chronic studies with
aquatic-phase amphibians in the open literature have been identified as acceptable for
quantitative  use within the context of this risk assessment, the estimated NOAEC of
0.023 ug a.i./L for freshwater fish will also be used to  estimate chronic risk to aquatic-
phase amphibians.

4.1.1.4 Freshwater  Fish: Sublethal Effects and Additional Open Literature
Information

No additional acceptable studies from the open literature were identified for freshwater
fish that: established more sensitive acute or chronic endpoints than the data listed above;
filled critical data gaps; presented a toxicity profile for under-represented taxa (e.g.,
toxicity data for amphibians); or provided information on sub-lethal effects that could be
clearly and reasonable linked to relevant assessment endpoints (i.e., survival,
reproduction, and growth) at concentrations lower than the most sensitive  endpoints used
to quantitatively evaluate risk.
                                       109

-------
4.1.1.5 Aquatic-phase Amphibian: Acute and Chronic Studies

No toxicity data for aquatic-phase amphibians were identified among the submitted
studies or that were considered acceptable for quantitative use in this risk assessment.

4.1.1.6 Freshwater Fish: Endosulfan Sulfate

Available data indicates that endosulfan sulfate is of similar acute toxicity to freshwater
fish as the TGAI.  Specifically, the 96-hour LC50 of endosulfan sulfate on bluegill was
3.8 ug/L and is considered very highly toxic (MRID 46382604; Appendix J). Acute
toxicity testing of endosulfan sulfate (99.9% radiopurity) was also tested with common
carp (MRID 45421402), which also indicated that the degradate is very highly toxic (96-
hr LC50 = 2.2 ug/L) to freshwater fish.  The study was classified as supplemental since it
failed to comply with EPA's recommended species and testing conditions. However,
results from the submitted and open literature studies are consistent with studies of parent
endosulfan TGAI where 96-hr LC50 values ranged from 0.37 to 3.3 ug/L.

4.1.2 Toxicity to Freshwater Invertebrates

A summary of acute and chronic freshwater invertebrate toxicity data, including data
published in the open literature is provided below in Sections 4.1.2.1 through 4.1.2.6.

4.1.2.1 Freshwater Invertebrates: Acute Exposure Studies

The acute toxicity studies available to the Agency demonstrate that endosulfan can be
classified as very highly toxic or highly toxic to freshwater invertebrates, with EC50
values ranging from 2.3 to 166  ug a.i./L. (see Appendix J). There were no studies
submitted to the Agency on  the effect of commercial end use products on freshwater
invertebrates.

4.1.2.2 Freshwater Invertebrates: Acute Exposure Studies—Open Literature

In addition to the aforementioned acute mortality studies with freshwater invertebrates,
several additional studies (see Appendix J) reported more sensitive values. Two species
of mayfly nymphs and one genus (species not indicated) of caddisfly larvae were tested
in static tests in which the concentrations of endosulfan were measured at the beginning
of each test (Leonard, et al.  1999; ECOTOX reference #20012. The EC/LC50s using
technical grade endosulfan in this study ranged from 0.4 to 1.8 ug/L, varying based on
the size of the organisms and the duration of the specific test. The lowest value was for
caddisfly larvae (Cheumatopsyche sp.); however, since this organism was not identified
to species this datum will not be used. The next lowest value was for the mayfly
Atalophlebia australis. The  0.6 ug/L value for A australis is considered the lowest acute
value for freshwater invertebrates that is acceptable for quantitative use in this risk
assessment. There was one study from the open literature that indicated for at least two
species (Daphnia magna and Hyalella azteca) commercial  end use products are  not more
                                       110

-------
toxic than endosulfan applied as the TGAI (Wan et al. 2005; ECOTOX reference #
87973).

4.1.2.3 Freshwater Invertebrates: Chronic Exposure Studies

One chronic exposure study with endosulfan TGAI involving the freshwater invertebrate,
Daphnia magna, was submitted to the Agency (MRID 5008271). The study examined the
effects of endosulfan on three consecutive generations of daphnids and noted that the
poor survival of third generation in the control and the lowest treatment group (2.7 ug/L)
precluded drawing valid conclusions about the cumulative effects of exposure. However,
the first generation portion of this study is considered acceptable for quantitative use in
this risk assessment, and generally coincides with the duration and design of the current
OPPTS Guideline (OPPTS 850.1300). The NOAEC and LOAEC from the first
generation of this study are 2.7 and 7.0 ug a.i./L based on reduced survival. Although this
study is considered acceptable, a daphnia acute to chronic ratio (ACR = 61.5) was used to
estimate a chronic value for the more acutely-sensitive mayfly, Atalophlebia australis.
No chronic toxicity data are available for this species. The estimated chronic NOAEC
value for the mayfly was calculated similarly to that for the common carp above:

        Estimated freshwater mayfly NOAEC = (0.6 x  2.7)7166 = 0.01 ug a.i./L.

This estimated NOAEC of O.Olug a.i./L for mayflies will be used to quantitatively
estimate chronic risk to freshwater invertebrates.

4.1.2.4 Freshwater Invertebrates: Sub-lethal Effects  and Additional Open
Literature Information

No additional acceptable studies from the open literature, beyond the one already
discussed, were identified for freshwater invertebrates that: established more sensitive
acute or chronic endpoints than the data listed above; filled critical data gaps; presented a
toxicity profile for  under-represented taxa; or provided information on sub-lethal effects
that could be clearly and reasonably linked to relevant assessment endpoints (i.e.,
survival, reproduction, and growth) at concentrations lower than the most sensitive
endpoints  used to quantitatively evaluate  risk to freshwater invertebrates.

4.1.2.5 Freshwater Invertebrates: Endosulfan Sulfate

Acute toxicity testing endosulfan sulfate using Daphnia magna (99.4% active ingredient)
indicated that the degradate is highly toxic (EC50 = 300 ug/L) to water fleas (MRID
45421403). The study was classified as supplemental since it was conducted using a
water hardness (160 - 180 mg/L as CaCO3) outside of the recommended range (40 - 49
mg/L as CaCOs). These results are similar to the acute toxicity of the parent TGAI to D.
magna (166 ug/L; MRID 5008271; Appendix J).

No data were identified on the chronic toxicity of endosulfan sulfate to freshwater
invertebrates that were considered acceptable for quantitative use in this risk assessment.
                                       Ill

-------
4.1.2.6 Freshwater Invertebrates: Sediment Toxicity

Endosulfan sulfate sediment toxicity tests were conducted for the midge, Chironomus
tentans, in two studies. The 20-day NOAEC and LOAEC determined from pore water
were measured as 0.35 and 1.2 ug/L respectively, based on mortality and percent
emergence (MRID 47318101). In a second study classified as supplemental, the most
sensitive NOAEC and LOAEC were determined as 2.7 and 3.8 ug/L, respectively based
on growth (MRID 46382605).

4.1.3 Toxicity to Estuarine/Marine Fish

Although direct and indirect effects to the listed species are not dependent on
estuarine/marine fish, toxicity data for estuarine/marine were reviewed in order to
provide a broader characterization of endosulfan risks to fish. A summary of acute and
chronic estuarine/marine fish data, including data published in the open literature is
provided below in Sections 4.1.3.1 through 4.1.3.4.

4.1.3.1 Estuarine/Marine Fish: Acute Exposure (Mortality) Studies

The acute toxicity studies submitted to the Agency demonstrate that endosulfan can be
classified as very highly toxic to estuarine/marine fish, with LC50 values ranging from
0.09 to 0.38 ug a.i./L for four species offish (striped bass, striped mullet, pinfish and
spot; see Appendix J). No acceptable acute toxicity studies were available for
estuarine/marine fish using formulated end-use products. The most sensitive acute LC50
value of 0.09  ug a.i./L for spot, Leiostomus xanthurus (MRID 5005824) will be used to
quantitatively estimate acute risk to estuarine/marine fish..

4.1.3.2 Estuarine/Marine Fish: Acute Exposure (Mortality) Studies—Open
Literature

In addition to the aforementioned acute mortality studies with estuarine/marine fish an
additional  study was identified in the open literature which reported several LC50 values
for sheepshead minnows (Cyprinodon variegatus), the preferred test species (Schimmel
1981; ECOTOX reference # 3740).  This study was a report listing the results of an
interlaboratory comparison of acute toxicity tests using estuarine animals—sheepshead
minnow was one of the species tested. There were six acute LC50 values based on
measured, flow-through tests ranging from 0.34 to 1.15  ug a.i./L (see Appendix J).
While none of these values is lower than the LC50 for the  spot listed above, their
geometric mean (0.76 ug a.i./L) will be used to establish the  acute to chronic toxicity
ratio for estuarine/marine fish used in the next section.

4.1.3.3 Estuarine/Marine Fish: Chronic Exposure (Growth/Reproduction) Studies

No acceptable chronic toxicity data were available to the Agency for estuarine/marine
fish; however, there was a study from the open literature (Hansen and Cripe 1991;
ECOTOX 115297) summarizing an interlaboratory comparison of the early-life stage
                                       112

-------
toxicity test using the sheepshead minnow Cyprinodon variegatus. Ten tests conducted
with measured, flow-through exposures were reported with NOAECs ranging from 0.17
to 1.2 ug a.i./L. The corresponding LOAECs are based on either survival or growth (see
Appendix J). The geometric mean of these ten values is 0.38 ug a.i./L. This study fulfills
a data gap for a chronic test with the preferred estuarine/marine fish. Although these tests
are considered acceptable for quantitative use in this risk assessment, a sheepshead
minnow acute to chronic ratio (ACR = 2.0) was used to estimate a spot chronic toxicity
value because it is the most acutely sensitive estuarine/marine fish species by
approximately a factor of four compared to sheepshead minnow and no chronic toxicity
data for endosulfan are available for it.  The estimated chronic NOAEC value for the spot
was calculated similarly to that demonstrated above for the common carp (except with
the ACR from the sheepshead minnow).

            Estimated spot NOAEC = (0.09 x 0.38)/0.76 = 0.045 ug a.i./L.

This estimated NOAEC of 0.045 ug a.i./L for spot will be used to  quantitatively estimate
chronic risk to estuarine/marine fish.

4.1.3.4 Estuarine/Marine Fish: Endosulfan Sulfate

One submitted study was available that evaluated the acute toxicity of endosulfan sulfate
on estuarine/marine fish.  This study was conducted with sheepshead minnow and
resulted in a 96-hr LC50 of 3.1 ug/L for endosulfan sulfate (MRID 46382603).  This
value is comparable to the range of LCSOs for the TGAI with sheepshead minnow
discussed above.

4.1.4 Toxicity to Estuarine/Marine Invertebrates

A summary of acute and chronic estuarine/marine invertebrate data, including data
published in the open literature, is provided below in Sections 4.1.4.1 through 4.1.4.5.

4.1.4.1 Estuarine/Marine Invertebrates: Acute Exposure (Mortality) Studies

The acute toxicity studies available to the Agency demonstrate that endosulfan can be
classified as highly toxic to very highly toxic to estuarine/marine invertebrates, with
LC50 values ranging from 0.04 to 790 ug a.i./L (see  Appendix J). There was one study
available that used a formulated end-use product and demonstrated that for the brown
shrimp that formulated endosulfan can be classified as very highly toxic with an LC50 of
0.24 ug a.i./L (MRID 1328). The most sensitive LC50 value of 0.04 ug a.i./L (MRID
5005824) for pink shrimp (Penaeus douraruni) is considered acceptable for quantitative
use to estimate acute risks to estuarine/marine invertebrates.
                                       113

-------
4.1.4.2 Estuarine/Marine Invertebrates: Acute Exposure (Mortality) Studies—Open
Literature

In addition to the aforementioned acute mortality studies with estuarine/marine
invertebrates, an additional study was identified in the open literature which reported
several LC50 values for mysids (Americamysis bahia; Schimmel 1981; ECOTOX
reference # 3740). This study reported the results of an interlaboratory comparison of
acute toxicity tests using estuarine animals. There were five acute LC50 values based on
measured, flow-through tests ranging from 0.38 to 1.29 ug a.i./L (see Appendix J).
While none of these values is lower than the LC50 for the pink shrimp listed above, their
geometric mean (0.83 ug a.i./L) will be used to establish the acute to chronic toxicity
ratio for estuarine/marine invertebrates used in the next section.

4.1.4.3 Estuarine/Marine Invertebrates: Chronic Exposure (Growth/Reproduction)
Studies

No acceptable chronic toxicity data were  available to the Agency for estuarine/marine
invertebrates with the endosulfan TGAI; however, one study was identified from the
open literature (McKenney 1982; ECOTOX reference #3736) that summarized an
interlaboratory comparison of chronic toxicity tests using the mysid Americamysis bahia.
Five tests were conducted under measured, flow-through conditions with reported
NOAECs ranging from 0.14 to 0.52 ug a.i./L. The corresponding LOAECs were based
primarily on survival (see Appendix J). The geometric mean of these five values is 0.27
ug a.i./L. The study fulfills a data gap for a chronic test with a preferred estuarine/marine
invertebrate. Although these tests are considered acceptable for quantitative use in this
risk assessment, a mysid acute to chronic ratio (ACR = 3.1) was used to estimate a pink
shrimp chronic NOAEC because it is the  most acutely sensitive estuarine/marine
invertebrate species (i.e., by approximately a factor of 10 compared to Americamysis
bahia) and no chronic toxicity data for endosulfan are available for pink shrimp. The
estimated chronic NOAEC value for the pink shrimp was calculated similarly to that
demonstrated above for the common carp (except with the ACR from the mysid).

         Estimated pink shrimp NOAEC = (0.04 x 0.27)/0.83 = 0.013 ug a.i./L.

This estimated NOAEC of 0.013 ug a.i./L for pink shrimp will be used to quantitatively
estimate chronic risk to estuarine/marine  invertebrates.
                                       114

-------
4.1.4.4 Estuarine/Marine Invertebrates: Open Literature Data

No additional acceptable studies from the open literature were identified for
esturarine/marine invertebrates that: establish more sensitive acute or chronic endpoints
than the data above; filled critical data gaps; presented a toxicity profile for under-
represented taxa; or provided information on sub-lethal effects that could be clearly and
reasonably linked to relevant assessment endpoints (i.e., survival, reproduction, and
growth) at concentrations lower than the most sensitive endpoints used to quantitatively
evaluate risk.

4.1.4.5 Estuarine/Marine Invertebrates: Endosulfan Sulfate

A 28-d flow-through, measured chronic toxicity study of the effects of endosulfan sulfate
was conducted with the mysid, Americamysis bahia (MRID 46781601).  Endpoints
measured included survival, growth (length and dry weight), and reproduction
(offspring/female/day). The most sensitive endpoint from this study was dry weight of
male mysids, with a NOAEC and LOAEC of 0.38 and 0.73 ug ai/L, respectively. This
NOAEC is similar to those reported by McKenney et al 1982 (ECOTOX reference #
3736) described above (0.14 to 0.52 ug a.i./L). This study is considered supplemental
because raw data were not provided, terminal growth measurements were not conducted
and a minor discrepancy was identified in the reported survival of control organisms.

4.1.5 Toxicity to Aquatic-phase Amphibians—Open Literature

A number of studies involving amphibian toxicity testing and endosulfan were identified
in the open literature and are summarized below. None of these studies provide reliable
estimates of toxicity that may be used quantitatively in this risk assessment; however,
they do provide some information regarding the hazard of endosulfan to amphibians. For
a comprehensive consideration of all potential effects data and additional information for
amphibians please refer to the detailed spreadsheet of the available ECOTOX open
literature data that can be found in Appendix H.

Bernabo, et al. (2008 ; ECOTOX reference #  103223) and Brunelli,  et al. (2009)
evaluated the effect of endosulfan (99% purity) on tadpoles of the common toad Bufo
bufo. Four-day-old tadpoles were exposed for 96 hr in a static test. The 96 hr LC50 value
based on nominal concentrations was 430 ug/L (Bernabo, et al. (2008). In a separate test,
exposure to 200 ug/L endosulfan resulted in an apparent increase in mucus production by
the gills, as well as inflammation of the gills. These changes occurred after only 24 hr and
became more pronounced as the exposure continued (Bernabo, et al. 2008). The same
group continued their work with endosulfan and this toad species looking for behavior
and morphological differences, after much longer (43-53 days) exposure durations
(Brunelli, et al. 2009). Three concentrations and a control were used. There were no
statistically significant differences between the control and 10 ug/L for survival, the rate
of deformity or the number of toads metamorphosed (into Gosner stage 46—complete tail
resorption). However, there was 77% and 97% mortality at 50 and 100 ug/L,
                                       115

-------
respectively. In addition, 100% of the surviving tadpoles exhibited various degrees of
deformity and none of the survivors reach stage 46.

Several papers evaluated the effect of endosulfan on tadpoles of Australian frogs, but all
were unmeasured tests.  Broomhall (2002; ECOTOX reference # 65732) exposed
tadpoles ofLitoria citropa to a single 0.8 ug/L nominal concentration of endosulfan
(Thiodan) under two different temperature regimes, one more variable than the other over
the 96 hr test. At ambient temperatures (ca. 20 C) the survival was 89% in the endosulfan
treatment (100% in the control). Under a similar average temperature, but more variable,
only 58% of the tadpoles survived, compared to 92% in the controls. After termination of
the above test, all surviving tadpoles were transferred to clean water and held at 19 C for
24 days. These tadpoles were then evaluated for their ability to avoid predation by
odonates. The endosulfan treated animals were more vulnerable to predation, measured
by a significant decrease in the number of minutes elapsed until capture.

Broomhall and Shine (2003; ECOTOX reference # 71866) exposed tadpoles of the
Australian treefrog Litoria freycineta to two different concentrations of endosulfan
(source not given, but assume was same  as Broomhall (2002)—Thiodan).  There was no
effect on survival after 96 hr exposure to 0.03  or 1.3 ug/L a.i. relative to the controls.
However, exposed tadpoles grew more slowly than controls—although not statistically
significant. And the survivors at the higher concentration were captured and  consumed by
odonate predators (dragon fly larvae) significantly sooner when tested 15 days after
transfer to clean water. Broomhall (2004; ECOTOX reference #73400) completed
another series of experiments similar to the above, using the same exposure
concentrations with a different species, Limnodynastesperonii, with similar results.

Canadian frogs were evaluated based on unmeasured concentrations. Green frog eggs and
larvae (Rana clamitans) were exposed to two different scenarios using Thiodan 50WP
(Harris, et al.  1998; ECOTOX reference  #19300). In the first test series, Gosner stage 8
embryos  through to stage 25 tadpoles were continuously exposed to a treatment for 13
days (two-thirds renewal of treatment solutions every second day). The  96 hr LC50 was
greater than the highest concentration tested (11,750 ug/L a.i.); however, the 13 day
LC50 was 15  ug  a.i. /L—suggesting that the earlier embryonic stages were not as
sensitive to endosulfan. Note, hatching occurred around day eight of the test. A second
set of exposures in this study consisted of the same exposure for the first 4 days, then a
period of 7.5 days in reference water, followed by  a second 4-day exposure to the same
concentrations. The 96 hr LC50 was again a greater than, and the 16 day LC50 was the
same as that in the first test. The frequency of deformities at hatch also was calculated in
the second test, resulting in an EC50 of 2,430  ug a.i./L.

The toxicity of Thiodan 4EC (40% emulsified concentrate) was evaluated (Berrill et al.
1998; ECOTOX  reference # 19467) with newly hatched and two-week-old tadpoles of
the wood frog (Rana sylvatica), the American toad (Bufo americanus), and the green frog
(Rana clamitans). Embryos of the wood frog were also tested, but were not as sensitive
as tadpoles; therefore embryos of the other two species were not tested.  New hatched
tadpoles of each species were exposed for 96 hr followed by a 10 day
                                       116

-------
recovery/observation period. Two-week of tadpoles were likewise exposed for 96 hr, and
followed for and additional 7 day recovery period in clean water. Water concentrations of
endosulfan were measured, but only after the first hour of the exposure period and the
treatments were apparently not renewed. Exposure concentrations ranged from 32 to 364
ug a.i. /L , depending on the life stage and species. There was essentially no mortality for
either life stage for all three species during the first 96 hr; however, there was extensive
postexposure mortality. Also, two-week old tadpoles were more sensitive than newly
hatched tadpoles—no LCSOs were calculated.

Early stage tadpoles ofBufo melaanostictus, Limnonectes limnocharis and Microphyla
ornata were exposed to a commercial end-use product of endosulfan (3 SEC) in 96-h
acute toxicity tests (Dey and Gupta 2002) The authors used 10 animals per container and
three containers per treatment. The animals were collected from the field and acclimated
for 3 to 4 days. They did not describe the chambers or the volume of water used. The
source of the water was dechlorinated tapwater. Control mortality also was not described.
The 96-h LCSOs were based on nominal concentrations and are 20, 1.3  and 0.16 ug/L,
respectively, for the three species listed above. The relative sensitivity of M. ornate is
likely due to its much smaller size. This was the likely explanation for the sensitivity of
this species to the pesticide dichlorvos and the herbicide butachlor (Geng et al. 2005).

Finally, there is a preprint available that evaluates the growth and survival of the northern
leopard frog Ranapipiens tadpoles (Shenoy et al. inpress). They exposed four-day-old
tadpoles to Endosulfan 3EC (33% a.i.) for seven weeks. Each individual was placed into
a separate 1.5 L of exposure medium with the water changed every 3 days. There were 15
individuals used for each treatment. Besides the controls, treatments were 0.2, 1 and 5
ug/L of the active ingredient. Concentrations were not measured. Growth rate (length)
was not affected in surviving tadpoles in any treatment relative to the controls. No LC50
values were calculated; however, there was 100% mortality by day 12 and day 28 in the 5
and 1 ug/L treatments, respectively. Based on the authors' Figure 1 the control survival
was approximately 80% at the end of the seven weeks and that in the 0.2 ug/L treatment
approximately 40%. Although no 96 hr data were presented, the figure suggests that there
was little of no mortality during the first 4 days of the experiment. Although not
calculated by the authors the above data suggest that the chronic LC50 would be close to
0.2 ug/L.
4.1.6 Toxicity to Aquatic Plants

For the purposes of this assessment, plant acute EC50 values, rather than NOAEC values,
are to be used to assess the potential for effects to the aquatic-phase CRLF, CTS, SFGS,
BCB, VELB, SMHM and SJKF via indirect effects on habitat, cover, food supply, and/or
primary productivity (i.e., aquatic plant community), because there are no obligate
relationships between the assessed species and any aquatic plant species. There were no
acceptable or supplemental data on the toxicity of endosulfan to aquatic  vascular or non-
vascular plants submitted to the Agency. There were several references available from
                                       117

-------
the open literature: however, only one of these could be considered as following methods
suitable to be considered acceptable (DeLorenzo et al 2002; ECOTOX reference #
65915). This study reported the 96 hr EC50 for the green alga Pseudokirchneriella
subcapitatum (formerly Selenastrum capricoratum) using a standard ASTM protocol.
This species is one of the OPP preferred species. The EC50 is 428 ug/L based on nominal
concentrations of technical grade endosulfan (98.6% a.i.). There were no acceptable data
identified in the open literature for aquatic vascular plants.

4.1.7 Freshwater Field/Mesocosm Studies

There were no acceptable data submitted to the Agency on  freshwater field or mesocosm
studies with endosulfan. There were; however, two studies  identified from the open
literature (Hose et al. 2002; ECOTOX 62267; Hose et al. 2003; ECOTOX 72555). Both
studies used the same artificial stream experimental design  using flow-though river water.
The flow was stopped during the exposure phase of the experiments. In the first study the
endosulfan (96% purity) was applied as contaminated sediment slurries. The stream flow
was blocked for 12 h after the initial exposure, then flow was restored and the
macroinvertebrate community sampled after an additional 96 h. There was no effect on
the structure of the macroinvertebrate community at the highest concentration measured
(6.14 ug/L—interstitial waters of the gravel substrate). The authors state that laboratory
studies have shown that 12-h exposures to this concentration (without sediment) are
toxic. However, they further conclude that the presence of fine sediment in the interstitial
water is likely to reduce the toxicity of endosulfan, presumably by binding the pesticide
and reducing its bioavailability. Although there were no effects on the community
structure due to endosulfan exposure, there were differences in the number of individuals
in the downstream drift of the mesocosms for some of the taxa. For example, the mayfly
Jappa kutera had increased numbers in the high-dose streams relative to the controls,
possibly as an avoidance mechanism.

These same authors conducted a similar study, but with aqueous endosulfan (Hose, et al
2003). Two separate experiments were carried out; a 12-h and a 48-h exposure
(endosulfan was measured in the streams for each experiment). The  first was a 12-h
exposure, followed by and additional 96-h of observation. During the 12-h exposure the
stream flow was stopped. Fifty taxa were recorded in the benthic assemblages during this
first  experiment. The data were analyzed using principle response curves.  The
assemblages at 8.69 ug/L and lower were not statistically different from the controls;
however, the assemblages at the highest concentration tested (48.87  ug/L) were different
from the controls—largely due to decreases in the  abundance of several key taxa. In the
48-h study endosulfan was added every 6 h for 48  h. During each 6-h exposure the stream
flow was stopped. The water was renewed at the beginning of each 6-h  exposure. The
assemblages at 1.00 ug/L were not statistically different from the controls; however, the
assemblages at the higher concentration tested (6.87 and 30.70 ug/L) were different from
the controls—as above this was largely due to decreases in  the abundance of several key
taxa. In this latter experiment the authors noted the appearance of blooms  of the
filamentous alga Spyrogyra sp. in the 6.87 and 30.70 ug/L treatments after 13  d. No algae
were noticed prior to the start of the experiment and none were present in  the controls or
                                       118

-------
1.00 ug/L treatments on day 13. The authors suggested that this was likely associated
with an apparent reduction in the number of tadpoles at these higher endosulfan
concentrations—however, the numbers of tadpoles were not quantified.

4.2 Toxicity of Endosulfan to Terrestrial Organisms

Table 4.3 summarizes the most sensitive terrestrial toxicity endpoints, based on an
evaluation of both the submitted studies and the open literature. A brief summary of
submitted and open literature data considered relevant to this ecological risk assessment
is presented below.
Table 4.3 Terrestrial Toxicity Profile for Endosulfan
Assessment
Endpoint
Purity
(% a.i.)
Surrogate
Species
Toxicity Value Used
for Quantitative
Risk Estimates
(a.i.)
Effects
Reference/Acceptability
Birds, Terrestrial-phase Amphibians and Reptiles"
Survival
Survival
Reproduction and Growth
97.2
96
96
Mallard duck
(Anas platyrhynchos)
Northern bobwhite
quail
Mallard duck
(Anas platyrhynchos)
LD50 = 28 mg/kg
LC50 = 805 ppm
NOAEC = 30 mg/kg-diet
LOAEC = 64 mg/kg-diet
Mortality
Mortality
Eggs laid,
eggs set, embryo
viability
MRID 136998/ Acceptable
MRID 22923/Acceptable
MRID 40335001/Acceptable
Mammals
Survival
Reproduction and Growth


Rat (Rattus norvegicus)
Rat (Rattus norvegicus)
LD50 = 10 mg/kg (female)
LD50 = 40 mg/kg (male)
NOAEC = 15 mg/kg diet
LOAEC = 75 mg/kg diet
Mortality
Decrease body
weight
MRID 00038307/Acceptable
MRID 00148264/Acceptable
Terrestrial Invertebrates
Survival
100
Beet webworm
(Pyrausta sticticalis)
Id
LD50 = 0.15uga.i./g
Mortality
Leonova and Slunko
2004/ECOTOX 100430
Terrestrial Plants
Survival and Growth
No acceptable data identified
a Birds are used as a surrogate for terrestrial-phase amphibians and reptiles because no acceptable data for
quantitative use currently exist for endosulfan toxicity to these taxonomic groups.
Acute toxicity to terrestrial animals is categorized using the classification system shown
in Table 4.4 (U.S. EPA, 2004). Toxicity categories for terrestrial plants have not been
defined.
Table 4.4. Categories of Acute Toxicity for Avian and Mammalian Studies
Toxicity Category
Very highly toxic
Highly toxic
Moderately toxic
Slightly toxic
Practically non-toxic
Oral LD50
< 10 mg/kg
10-50 mg/kg
51 -500 mg/kg
501 -2000 mg/kg
> 2000 mg/kg
Dietary LCSO
< 50 ppm
50 - 500 ppm
501 - 1000 ppm
1001 - 5000 ppm
> 5000 ppm
                                         119

-------
   4.2.1 Toxicity to Birds, Reptiles, and Terrestrial-Phase Amphibians

   As specified in the Overview Document, the Agency uses birds as a surrogate for reptiles
   and terrestrial-phase amphibians when toxicity data for each specific taxon are not
   available (U.S. EPA, 2004). A summary of acute and chronic bird data, including data
   published in the open literature is provided below in Sections 4.2.1.1 through 4.2.1.4.

   4.2.1.1 Birds: Acute Exposure (Mortality) Studies

   The  results of the acute oral toxicity studies (MRID  137189, 136998, and 160000)
   available for endosulfan (see Table 4.5 and Appendix K) indicate that endosulfan can be
   classified as moderately toxic to highly toxic to avian species on an acute oral basis with
   LD50 values for Northern bobwhite quail (Colinus virginianus)., Mallard duck {Anas
   platyrhynchos), and Ring-necked pheasant (Phasianus colchicus) ranging from 28 to >
   320  mg a.i./kg-body weight. Based  on all of the acute oral toxicity data, mallard ducks
   were the most acutely sensitive with LD50 values ranging from 28  to 45 mg a.i./kg-body
   weight. The LD50 value of 28 mg a.i./kg-body weight will be used in the quantitative
   estimate of risk.

   The  results of the single sub-acute dietary toxicity study (MRID 22923) available for
   endosulfan indicate that it can be classified as slightly toxic to moderately toxic with
   LC50 values for Northern bobwhite quail (Colinus virginianus), Mallard duck (Anas
   platyrhynchos), Ring-necked pheasant (Phasianus colchicus), and Japanese quail
   (Coturnix japonicd) ranging from 805 to 1275 mg a.i./kg-diet. Northern bobwhite quail
   was  the species tested with the lowest value. The LC50 value of 805 mg a.i./kg-diet will
   be used in the quantitative estimate  of risk.

Table 4.5 Comparison of Acute Toxicity of Endosulfan and Endosulfan Sulfate To Birds.
Species
Northern bobwhite
quail
Colinus virginianus
Mallard duck
Anas platyrhynchos
Acute Oral
Endosulfan
Toxicity
LD50 Category
(ppm) (MRID)
--
highly
28 toxic
(136998)
Toxicity

Endosulfan Sulfate
Toxicity
LD50 Category
(ppm) (MRID)
44
—
highly
toxic
(464305-
01)
—

Acute Dietary
Endosulfan
5-day Toxicity
LC50 Category
(ppm) (MRID)
805
1053
moderately
toxic
(22923)
slightly toxic
(22923)
Toxicity

Endosulfan Sulfate
5-day Toxicity
LC50 Category
(ppm) (MRID)
>3528
1642
(46430502)
(a)
slightly
toxic
(463826-
01)
  Study classified as supplemental.

   4.2.1.2 Birds: Chronic Exposure (Growth, Reproduction) Studies

   Three avian reproduction studies with mallard ducks (Anasplatyrhynchos) and Northern
   bobwhite quail (Colinus virginianus) have been submitted to the Agency for endosulfan
   (MRID 146843, 40335001, and 40335002). Only one test established both NOAEC and
   LOAEC values (see Appendix K). This test was a 22 week study with mallard ducks
                                          120

-------
(Anas platyrhynchos). The NOAEC was 30 ppm and the LOAEC was 64 ppm (using
measured dietary concentrations), based on number of eggs laid (53% of control) and
embryo viability (52% of control) (MRID 40335001). The other mallard duck test had an
NOAEC  < 30 ppm (MRID 146843) and was not used because the measured dietary
concentrations were not presented. The bobwhite quail test showed an NOAEC = 64 ppm
and an LOAEC = 134 ppm. The results of these studies indicate that endosulfan may
have adverse effects on avian reproduction at higher levels of exposure, and the most
sensitive NOAEC of 30 ppm (mg a.i./kg-diet) will be used to quantitatively estimate the
risks to birds (and thus, terrestrial-phase amphibians and reptiles) resulting from chronic
exposure to endosulfan.

4.2.1.3 Birds:  Open literature Studies

No additional acceptable studies from the open literature were identified for birds (or
terrestrial-phase amphibians or reptiles) that:  established more sensitive acute or chronic
endpoints than existing data; filled critical data gaps; presented a toxicity profile for
under-represented taxa (e.g., toxicity data for amphibians or reptiles); or provided
information on sub-lethal effects that could be clearly and reasonably linked to relevant
assessment endpoints (i.e., survival, reproduction, and growth) at concentrations lower
than the most sensitive endpoints used to quantitatively evaluate risk.

4.2.1.4 Birds:  Endosulfan Sulfate

Based on the comparison of parent endosulfan TGAI with the endosulfan sulfate
degradate using acute  dietary LCSOs for bobwhite quail and mallard duck, endosulfan
sulfate appears about equal in toxicity to waterfowl (mallard) and at least a factor of 4
less toxic to game birds (quail; Table 4.5).  No chronic avian studies were identified for
endosulfan sulfate that were considered acceptable for quantitative use in this risk
assessment.

4.2.2 Toxicity to Mammals

Typically, mammalian toxicity data from the  Agency's Health Effects Division (HED)
are used to approximate toxicity to mammals. However, wild mammals toxicity tests
may be required on a case-by-case basis, depending on the results of the lower tier studies
such as acute and sub-acute testing, intended use pattern, and pertinent environmental
fate characteristics. No studies evaluating toxicity to wild mammal species have been
submitted by the registrants for endosulfan. A summary of acute and chronic mammalian
data, including data published in the open literature, is provided below in Sections 4.2.2.1
through 4.2.1.3.

4.2.2.1 Mammals: Acute Exposure (Mortality) Studies

The acute oral LD50 values for the laboratory rat (Rattus norvegicus) were  10 and 40
mg/kg  for females and males, respectively (MRID 38307). Since  acute toxicity estimates
fall in the range of 10 to 50 mg/kg, endosulfan is classified as highly toxic on an acute
                                       121

-------
exposure basis (see Appendix K for more detail). Additional data from the open
literature support the conclusion that female rats are more sensitive to endosulfan than
males (Gupta 1976; ECOTOX 103467). LDSOs for males were greater than those for
females by about a factor of 2. The LDSOs from the submitted data are still the lowest
numbers. Gupta (1976) also presented LC50 data for mice (Mus musculus). Unlike rats,
there were no differences in LDSOs between male and female mice. However, mice were
more sensitive than rats in there study. Values ranged from 6.5 to 13.5 mg/kg body
weight for mice (and 22.1 to 89.4 for rats), depending on the vehicle used to administer
the endosulfan (see Appendix K). The acute oral LD50 value 10 and 40 mg/kg for
females and male rats, respective will be used to estimate the risk to mammals from acute
exposure to endosulfan. The lower value for the mouse was not used because the
exposure was via intraperitoneal injection rather than through diet.

4.2.2.2 Mammals: Chronic Exposure (Growth, Reproduction) Studies

A two-generation rate reproduction study measured the NOAEC and LOAEL as 15 and
75 ppm, respectively (MRTD 148264), with decreased body weight as the most sensitive
endpoint (see Appendix K for more detail). An additional study from the open literature
supports the conclusion from Gupta (1976; ECOTOX 103467) that mice are more
sensitive to endosulfan than rats. Hack et al. (1995; ECOTOX 103384) exposed groups of
rats and mice for 24 months to various concentrations of endosulfan in their diet. These
authors concluded that the NOAEC and LOAEC for rats are 15 and 75 ppm,
respectively—based on final body weight for both males and females (there was no effect
of endosulfan on rat survival). There was no effect of dietary endosulfan on the growth of
either male or female mice; however the authors' data would list the NOAEC and
LOAEL as 6 and 18 ppm, respectively—based on survival of females.

Two other open literature studies using oral intubation showed that spermatogenesis in
rats is reduced significantly at endosulfan doses as low as 2.5 mg/kg body weight after 70
days of exposure (Sinha et al. 1995; ECOTOX 103592), and that fetal survival and
development can be impaired at concentrations as low as 1.0 mg/kg body weight fed to
sexually mature female rats on days 6 to 20 of their pregnancies (Singh et al. 2007;
ECOTOX 103238).

4.2.2.3 Mammals: Endosulfan Sulfate

No acute or chronic mammalian studies were identified for endosulfan sulfate that were
considered acceptable for quantitative use in this risk assessment.

4.2.3 Toxicity to Terrestrial Invertebrates

A summary of terrestrial invertebrate data, including data published in the open literature,
is provided below in Sections 4.2.3.1 and 4.2.3.2.
                                      122

-------
4.2.3.1 Terrestrial Invertebrates: Acute Exposure (Mortality) Studies

The only Agency guideline acute study for terrestrial invertebrates is for the honey bee
(Apis melliferd). A total of four honey bee studies (see Appendix K) are available that
include acute contact, acute oral and acute contact with treated foliage with TGAI and
formulated end use product (MRIDs 1999, 05004151, 05008936, and 05012881). The
acute contact LD50 values are 4.5 and 7.1 ug a.i./bee, both with TGAI. The sole acute
oral LD50 is 6.9 ug a.i./bee, also with TGAI. Two studies reported exposures based on
foliar residues. Both studies observed little to no mortality at the highest concentrations
tested,  making the LCSOs > 7.1 and>1.10 Ib/acre. The acute contact LD50 of 4.5 ug
a.i./bee is multiplied by 1 bee/0.128g, which is based on the weight of an adult honey
bee, in order to estimate the toxicity in terms of ppm (ug a.i./g of bee).

No additional data on the effects of endosulfan to non-target terrestrial invertebrates was
submitted to the Agency.

4.2.3.2 Terrestrial Invertebrates: Open Literature Studies

The ECOTOX database was examined for toxicity data using non-target species with
endpoints expressed in terms similar to those for the standard test with honey bees. The
stingless bee and the Indian honey bee were both more sensitive than the standard honey
bee in comparable tests based on per bee exposure (see Appendix K). However, both the
stingless bee and the Indian honey bee are smaller than the standard honey bee so the
conversion used for the honey bee in the last section (1/0.128 g/bee) should not be used.
Data from the test using the stingless bee Trigona spinipes (Madeira and Heblling-
Beraldo 1989; ECOTOX reference #51755) is the lowest LC50 (0.21 ug a.i/bee). Three
different foliar residue studies using three different species of parasitic wasps also are
listed in Appendix K. Significant mortality occurred at the reported application rates
(0.50tol.501b/a).

Several studies listed in the ECOTOX database used earthworms as the  test organism.
Mosleh et al (2003; ECOTOX reference # 86741; note this is the same data that the
authors also report in Mosleh et al. 2002; ECOTOX reference #87129) report LCSOs
using a commercial end-use product mixed with natural  soil for the earthworm Lumbricus
terrestris. The LCSOs declined as the duration of exposure increased, and ranged from
3.36 to 12.29 mg endosulfan/kg soil.

There are numerous studies available from the open literature relative to targeted insect
studies. The data from studies which calculated contact, oral or foliar residue acute
LDSOs are presented in Appendix K. Most of these data are larvae of several moth
species. Two studies (Leonova and Slynko 1996, 2004; ECOTOX references #103049
and 100430) showed that at least for the cotton bollworm and the beet webworm adult
moths more sensitive that the larvae of these species. As expected, overall these studies
show that applications of formulated endosulfan are likely to reduce the numbers and
possibly eliminate populations of arthropods. The lowest acute contact LD50 or 0.15
ug/g (ppm) for the adult moth of the beet webworm (Pyrasta sticticalis) will be used to
                                       123

-------
estimate the risk to terrestrial invertebrates from acute exposure to endosulfan.

4.2.4 Toxicity to Terrestrial Plants

No data have been submitted to the Agency to evaluate the effects of endosulfan on
terrestrial plants because historically, terrestrial plant toxicity studies and associated risk
analysis of plants were not required for registration of a pesticide unless it met specific
use and pesticide classification criteria which would trigger potential concerns. In
addition to the lack of registrant-submitted data, no studies demonstrating significant
adverse effects of endosulfan to any terrestrial plant have been identified in the open
literature. Although a number of studies involving terrestrial plants and endosulfan were
identified in the open literature, none of these studies provide reliable estimates of
toxicity that may be used in this risk assessment. Reasons that these studies were deemed
unacceptable for use were primarily because these studies were associated with efficacy
determinations in which observations were confounded by the presence of an insect pest
complex. As such, plants in these tests generally did not demonstrate any adverse effects
at any test levels, but did not test up to the maximum allowable rate.  For a comprehensive
consideration of all potential effects data and additional information  for terrestrial plants
please refer to the detailed spreadsheet of the available ECOTOX open literature data that
can be found in Appendix H.
4.3 Use of Probit Slope Response Relationship to Provide Information on the
Endangered Species Levels of Concern

The Agency uses the probit dose response relationship as a tool for providing additional
information on the potential for acute direct effects to individual listed species and
aquatic animals that may indirectly affect the listed species of concern (U.S. EPA, 2004).
As part of the risk characterization, an interpretation of acute RQs for listed species is
discussed. This interpretation is presented in terms of the chance of an individual event
(i.e., mortality or immobilization) should exposure at the EEC actually occur for a species
with sensitivity to endosulfan on par with the acute toxicity endpoint selected for RQ
calculation. To accomplish this interpretation, the Agency uses the slope of the dose
response relationship available from the toxicity study used to establish the acute toxicity
measures of effect for each taxonomic group that is relevant to this assessment. The
individual effects probability associated with the acute RQ is based on the mean  estimate
of the slope and an assumption of a probit dose response relationship. In addition to a
single effects probability estimate based on the mean, upper and lower estimates of the
effects probability are also provided to account for variance in the slope, if available.

Individual effect probabilities are calculated based on an Excel spreadsheet tool IECV1.1
(Individual Effect Chance Model Version 1.1) developed by the U.S. EPA, OPP,
Environmental Fate and Effects Division (June 22, 2004). The model allows for such
calculations by entering the mean slope estimate (and the 95% confidence bounds of that
                                        124

-------
estimate) as the slope parameter for the spreadsheet. In addition, the acute RQ is entered
as the desired threshold.

4.4 Ecological Incident Summary

A review of the Ecological Incident Information System (EIIS) database for ecological
incidents involving endosulfan was completed on March 25, 2009. This database consists
of exposure incident reports submitted to the EPA from 1994 to present. A summary of
ecological incidents involving endosulfan are listed in Table L.I in Appendix L.  This
table is divided into incidents involving aquatic organisms only, terrestrial organisms
only, and both aquatic and terrestrial organisms. Within each of these sections of the
table, incidents are ordered by date beginning with the earliest incident.

Incidents listed in EIIS are categorized by the likelihood that a particular pesticide is
associated with that particular incident. These classifications include highly probable,
probable, possible, unlikely or unrelated. "Highly probable" incidents usually require
carcass residues or clear circumstances regarding the exposure. "Probable" incidents
include those where residue information was not available or circumstances were  less
clear than those for "highly probable." "Possible" incidents occur when multiple
chemicals may have been involved and the contribution of an individual chemical is not
obvious. An "unlikely" incident classification is given when a given chemical is
considered nontoxic to the type of organism involved or the chemical was analyzed and
not detected in samples. The "unrelated" category is used  for incidents  confirmed  not to
involve pesticides. No unrelated incidents were listed for endosulfan.

The number of reports listed in the EIIS database is believed to be only a fraction  of the
total incidents involving organismal mortality and damage caused by pesticides. Few
resources are assigned to incident reporting. Reporting by states is only voluntary, and
individuals discovering incidents may not be informed on the procedure of reporting
these occurrences. Additionally, much of the database is generated from registrant-
submitted incident reports. Registrants are legally required to provide detailed reports  of
only "major" ecological incidents involving pesticides, while "minor" incidents are
reported aggregately. Because of these logistical difficulties, EIIS is most likely a
minimal representation of all pesticide-related ecological incidents.

The EIIS database contained 83 incident reports involving endosulfan. Most of the
incidents involve aquatic ecosystems (75 or 90% of the total incidents). Seven incident
reports involve terrestrial ecosystems and one involves a combined aquatic/terrestrial
ecosystem. California was most represented among all 50 states (26 reports) followed by
North Carolina (9), Louisiana and South Carolina (5 each) and Washington State (4).
Additional characterization of these incidents is provided below.

Incident Certainty and Legality

Of the  83 incidents reported, 23 (28%) are categorized as  'highly probable' and 40 (48%)
are categorized as 'probable.' Collectively the 'highly probable' and 'probable'
                                        125

-------
categories represent 76% of the reported incidents.  Regarding the legal status, the
'unknown' and 'misuse accidental' represent the largest legality categories with 36% and
35% of the incident reports, respectively. Approximately 25% of the reports consist of
registered uses. Lastly, only three of the incident reports involved intentional misuse.

Of the 15 'highly probable' and 'probable' incident reports that involved only 'registered
uses,' most described pesticide runoff following periods of heavy rainfall as the likely
event that led to the reported incident. The majority of the 'highly probable' and
'probable' incidents classified as 'accidental misuse' involved aerial application too close
to bodies of water as stipulated by the label, spills and equipment washing.

Aquatic Incidents

The vast majority of the aquatic incident reports involved mortality to fish (67), a highly
sensitive taxonomic group. Only three incidents reportedly involved aquatic
invertebrates, but the likelihood of observing impacts to aquatic invertebrates is low
compared to fish. Of the 67 aquatic incidents involving fish, 53 (80%) are classified  as
either 'highly probable' or 'probable' in the context of endosulfan use. A wide variety of
fresh and estuarine species were reportedly affected (e.g., carp, catfish, largemouth bass,
shad, menhaden, mullet, spot, bluegill sunfish, gar and trout).

Terrestrial Incidents

Of the seven terrestrial incidents, none are classified as 'highly probable'  and two are
classified as 'probable.' The two 'probable' incidents involve birds (blue jay, crow, owl),
mammals (squirrel, opossum, red fox) and an amphibian (unidentified frog).

Plant Incidents

Only one reported incident involved plants and this was classified as 'possible,' but the
plant species was not identified.
                                        126

-------
  5  Risk Characterization

Risk characterization is the integration of the exposure and effects characterizations.
Risk characterization is used to determine the potential for direct and/or indirect effects to
the CRLF, CIS, SFGS, SMHM, SJKF, BCB, and VELB or for modification to its
designated critical habitat from the use of endosulfan in CA.  The risk characterization
provides an estimation (Section 5.1) and a description (Section 5.2) of the likelihood of
adverse effects; articulates risk assessment assumptions, limitations, and uncertainties;
and synthesizes an overall conclusion regarding the likelihood of adverse effects to the
CRLF and other SFB species or their designated critical habitat (i.e., "no effect," "likely
to adversely affect," or "may affect, but  not likely to adversely affect").

5.1 Risk Estimation

Risk is estimated by calculating the ratio of exposure to toxicity. This ratio is the risk
quotient (RQ), which is then compared to pre-established acute and chronic levels of
concern (LOCs) for each category evaluated (Appendix D). For acute exposures to the
aquatic animals, as well as terrestrial invertebrates, the LOG is 0.05. For acute exposures
to the birds (and, thus, reptiles and terrestrial-phase amphibians) and mammals, the LOG
is 0.1. The LOG for chronic exposures to animals, as well  as acute exposures to plants is
1.0.

Acute and chronic risks to aquatic organisms are estimated by calculating the ratio of
exposure to toxicity using l-in-10 year EECs, based on the label-recommended
endosulfan usage scenarios, summarized in Table 3.5 and the appropriate aquatic toxicity
endpoint from Table 4.1. As described in Section 2.10.1 (Problem Formulation), aquatic
exposures were modeled separately for the parent isomers (d and P) and the primary
degradate of concern (endosulfan sulfate). The resulting daily concentrations were then
summed to form a 30-year time series for total endosulfan (sum of d, P and endosulfan
sulfate). Due to similar acute toxicity and structure of the TGAI and endosulfan sulfate,
modeled or measured environmental exposures of total endosulfan (sum of d, P and
endosulfan sulfate) were compared with toxicity data for the TGAI. For sediment-borne
exposures, toxicity data were available only for endosulfan sulfate, and thus, comparisons
of total endosulfan EECs were made to toxicity estimates for endosulfan sulfate.

Acute and chronic risks to terrestrial animals are estimated based on exposures resulting
from applications of endosulfan (Table 3.23)  and the appropriate toxicity endpoint from
Table 4.3. Exposures were not derived  for terrestrial plants, as described in  Section 3.4.
                                       127

-------
5.1.1 Exposures in the Aquatic Habitat
5.1.1.1 Freshwater Fish and Aquatic-phase Amphibians

Acute risk to freshwater fish and aquatic-phase amphibians, and the potential for direct
effects to CRLF and CTS specifically, is based on peak EECs in the standard
PRZM/EXAMS pond and the lowest acute toxicity value for freshwater fish. Currently
registered agricultural uses of endosulfan within California are listed in Table 3.1. Based
on freshwater fish toxicity data (LCso value of 0.1 jig a.i./L for common carp) and
modeled aquatic peak EECs for various use scenarios used to represent all of the
agricultural uses of endosulfan in CA, all acute RQs for freshwater fish range from 7.2
(eggplant) to 58.8 (lettuce and brussels sprouts); therefore, the entire set of 20 modeled
scenarios used to represent all of the agricultural uses of endosulfan in CA, resulted in an
exceedance of the Agency's acute listed species LOG (RQ>0.05)(see Table 5.1).

Chronic risk to freshwater fish and aquatic-phase amphibians, and the potential for direct
effects to CRLF specifically, is based on 60-day EECs  and the lowest chronic toxicity
value for freshwater fish. However, in the case  of endosulfan, the fathead minnow acute
to chronic ratio (ACR=4.3) was used to estimate a common carp chronic toxicity
NOAEC value of 0.023 jig a.i./L because it is the most acutely sensitive freshwater fish
species and no chronic toxicity  data are available for it. Based on 60-day EECs for
various use scenarios used to represent all of the agricultural uses of endosulfan in CA,
and the estimated NOAEC of 0.023 jig a.i./L, chronic RQs for freshwater fish range from
5.7 (eggplant) to  80.6 (lettuce and brussels sprouts. All 20 of the modeled scenarios used
to represent all of the agricultural uses of endosulfan in CA resulted in an exceedance of
the Agency's chronic risk LOG (RQ>1) for freshwater  fish (see Table 5.1).

Based on exceedances of the Agency's acute listed species LOG (RQ>0.05) for the entire
set of 20 modeled scenarios used to represent all of the agricultural uses of endosulfan in
CA, and exceedances of the Agency's chronic risk LOG (RQ>1) for all 20 of the modeled
scenarios, endosulfan does have the potential to directly affect the CRLF and CTS.
Additionally, since the acute and chronic RQs are exceeded,  there is a potential for
indirect effects to those listed species that rely on freshwater fish (and/or aquatic-phase
amphibians) during at least some portion of their life-cycle (i.e., CRLF, SFGS, and CTS).
Table 5.1 Acute and Chronic RQs for freshwater fish based on EECs for use
categories used to represent all endosulfan uses in CA.
Use Category'1'
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Run
No.
1
1
1
2
1
Peak
EEC
(ug/L)
3.35
3.35
4.81
1.67
2.88
60-
day
EEC
(ug/L)
0.62
0.44
1.76
0.50
0.77
Acute
RQ(1)
33.50
33.50
48.10
16.70
28.80
Chronic
RQ(1)
26.83
19.30
76.39
21.78
33.65
                                       128

-------
Table 5.1 Acute and Chronic RQs for freshwater fish based on EECs for use
categories used to represent all endosulfan uses in CA.
Use Category'1'
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Ornamentals or Shade Trees (Northern Central
coast)
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run
No.
2
3
1
1
1
3
1
1 Add
1
1 Add
3
1
2
1
1
Peak
EEC
(ug/L)
1.24
1.56
3.40
5.88
1.55
0.72
3.43
4.38
1.89
2.63
1.18
2.00
1.72
3.64
1.79
60-
day
EEC
(ug/L)
0.52
0.59
0.53
1.85
0.61
0.13
1.00
1.13
0.49
0.56
0.36
0.51
0.54
1.11
0.45
Acute
RQ(1)
12.40
15.60
34.00
58.80
15.50
7.20
34.30
43.80
18.90
26.30
11.80
20.00
17.20
36.40
17.90
Chronic
RQ(1)
22.65
25.78
23.13
80.61
26.65
5.70
43.65
49.13
21.39
24.39
15.57
22.09
23.39
48.17
19.57
111 RQ values in bold indicate exceedence of listed species acute LOG (0.05) and chronic LOG (1 .0). Acute RQs were
calculated based on the peak EEC divided by the LC50 value of 0.1 ug/L for the most sensitive freshwater fish (common
carp, Cyprinus carpio). Chronic RQs were based on the 60 day average EEC divided by the estimated chronic NOAEC for
common carp of 0.023 ug/L.
              5.1.1.2 Freshwater Invertebrates

Acute risk to freshwater invertebrates is based on peak EECs in the standard
PRZM/EXAMS pond and the lowest acute toxicity value for freshwater invertebrates.
Based on freshwater invertebrate toxicity data (ECso value of 0.6 jig a.i./L for mayflies)
and modeled aquatic peak EECs for various use scenarios used to represent all of the
agricultural uses of endosulfan in C A, all acute RQs for freshwater invertebrates range
from 1.20 (eggplant) to 8.02 (broccoli, cabbage, and cauliflower). Therefore, the entire
set of 20 modeled scenarios used to represent all of the agricultural uses of endosulfan in
CA, resulted in an exceedance of the Agency's acute listed species LOG (RQ>0.05)(see
Table 5.2)

Chronic risk is based on 21-day EECs and the lowest chronic toxicity value for
freshwater invertebrates. However, in the case of endosulfan, waterflea acute to chronic
ratio (ACR=61.5) was used to estimate a mayfly chronic toxicity NOAEC value of 0.01
jig a.i./L because it is the most acutely sensitive freshwater invertebrate species and no
chronic toxicity data are available for it. Based on 21-day EECs for various use scenarios
used to represent all of the agricultural uses of endosulfan in CA, and the estimated
NOAEC of 0.01 jig a.i./L, all chronic RQs for freshwater invertebrates range from 23.7
(eggplant) to 260.3 (broccoli, cabbage, and cauliflower); therefore, the entire set of 20
                                       129

-------
modeled scenarios used to represent all of the agricultural uses of endosulfan in CA,
resulted in an exceedance of the Agency's chronic risk LOG (RQ>1) for freshwater
invertebrates (see Table 5.2).

Based on exceedances of the Agency's acute listed species LOG (RQ>0.05) and chronic
risk LOG (RQ>1) for the entire set of 20 modeled scenarios used to represent all of the
agricultural uses of endosulfan in CA, there is a potential for direct effects on the CRLF
and indirect effects CRLF,  SFGS, and CTS, the listed species that rely on freshwater
invertebrates during at least some portion of their life-cycle.
Table 5.2 Acute and Chronic RQs for freshwater invertebrates based on EECs for use categories
used to represent all endosulfan uses in CA.
Use Category'1'
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Ornamentals or Shade Trees (Northern Central
coast)
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run
No.
1
1
1
2
1
2
3
1
1
1
3
1
1 Add
1
1 Add
3
1
2
1
1
Peak
EEC
(ug/L)
3.35
3.35
4.81
1.67
2.88
1.24
1.56
3.4
5.88
1.55
0.72
3.43
4.38
1.89
2.63
1.18
2
1.72
3.64
1.79
21 -day
EEC
(ug/L)
1.03
0.90
2.60
0.78
1.14
0.60
0.82
1.01
2.58
0.82
0.24
1.36
1.73
0.77
1.01
0.62
0.85
0.74
1.71
0.81
Acute
RQ(1)
5.58
5.58
8.02
2.78
4.80
2.07
2.60
5.67
9.80
2.58
1.20
5.72
7.30
3.15
4.38
1.97
3.33
2.87
6.07
2.98
Chronic
RQ<1)
102.60
90.00
260.30
77.50
113.70
59.60
82.40
101.30
257.50
81.50
23.70
135.70
172.80
77.00
101.10
62.20
85.10
73.50
171.40
80.80
|n) RQ values in bold indicate exceedence of listed species acute LOG (0.05) and chronic LOG (1 .0). Acute RQs were
calculated based on the peak EEC divided by the LC50 value of 0.6 ug/L for the most sensitive freshwater invertebrate
(mayfly, Atalophlebia australis). Chronic RQs were based on the 21 day average EEC divided by the estimated chronic
NOAEC for mayfly of 0.01 ug/L.
                                       130

-------
              5.1.1.3 Non-vascular Aquatic Plants

Acute risk to aquatic non-vascular plants is based on peak EECs in the standard pond and
the lowest acute toxicity value. Based on the only aquatic non-vascular plant toxicity
datum (EC50 = 428 jig a.i./L for the freshwater green alga Pseudokirchneriella
subcapitatum) and the maximum aquatic peak EEC of all use scenarios representing all
of the agricultural uses of endosulfan in CA (5.88 for lettuce and brussels sprouts) all
RQs for aquatic non-vascular plants are < 0.014. Since the RQs do not exceed the
Agency's LOG (1) for aquatic non-vascular plants, endosulfan is determined to also have
no indirect effects to those listed species that rely on non-vascular aquatic plants during at
least some portion of their life-history (i.e., aquatic-phase CRLF, CIS, SFGS, and
SMHM).

5.1.1.4 Aquatic Vascular Plants

Toxicity data have  not been  identified for quantitatively estimating  risk to vascular
aquatic  plants,  and  therefore, RQs cannot be calculated at this time for this taxonomic
group. Discussion regarding lines of evidence for the potential for indirect effects to those
listed species that rely on vascular plants during at least some portion of their life-history
(i.e., aquatic-phase  CRLF, CIS, SFGS, and  SMHM)  can be  found  in the "Risk
Description" portion of the chapter (Section 5.2).

5.1.2 Exposures in  the Terrestrial Habitat

5.1.2.1 Birds (surrogate for Reptiles and  Terrestrial-phase amphibians)

As previously discussed in Section 3.3, potential direct effects to terrestrial species are
based on foliar applications of endosulfan. Potential direct acute effects are derived by
considering  dose- and dietary-based EECs modeled in T-REX for  a small bird (20 g)
consuming a variety of dietary items (Table 3.23) and acute oral and subacute dietary
toxicity endpoints for avian species. Based  on the most sensitive bird acute data
(LD50=28 mg/kg b.w. for the mallard duck) adjusted for differences in body weight by
T-REX (LD50=14.54 mg/kg b.w.), and the modeled acute dose-based EECs for various
use scenarios and diet categories, all RQs for  small birds ranged from 0.13 (Eggplant,
Granivores)  to 47 (three different use categories for short grass diet). All 100 acute dose-
based RQs for uses of endosulfan in CA resulted in an exceedence of the Agency's acute
listed species risk LOG (RQ>0.1) for birds  (see Table 5.3).  Acute dietary-based EECs
and the most sensitive bird dietary toxicity value (805 mg/kg diet for bobwhite quail)
resulted in RQs that ranged from 0.01 (4 different use categories for
fruit/pods/seeds/large insects dietary category) to 0.75 (3 use categories for short grass).
Only 57 of the 80 acute dietary-based RQs  exceeded the Agency's acute  listed species
risk LOC (>0.1) for birds (Table 5.3).

Potential direct chronic effects to terrestrial birds (and reptiles and terrestrial-phase
amphibians) are derived by considering dietary-based EECs modeled in T-REX for a
small bird (20 g) consuming a variety of dietary items. Chronic effects are estimated
                                       131

-------
using the lowest available chronic dietary toxicity data for birds (NOEC= 30 mg/kg diet.
for mallard duck). Chronic dietary-based RQs ranged from 0.25 (eggplant for
fruit/pods/seeds/large insects) to 20 (3 different use categories for short grass). Sixty-
three of the 80 chronic dietary-based RQs exceeded the Agency's chronic risk LOG (>1)
for birds (Table 5.3).

Based on exceedences of the Agency's acute risk LOG (RQ>0.1) and chronic risk LOG
(RQ>1) endosulfan does have the potential to directly effect reptiles and terrestrial-phase
amphibians (i.e., CRLF, GTS and SFGS) during at least some portion of their life-history.
In addition, endosulfan also has the potential to indirectly affect the CRLF, SFGS,
SMHM and SJKF since they potentially rely on birds, reptiles or terrestrial-phase
amphibians during some portion of their life-history.
Table 5.3 Summary of the acute and chronic dose- and dietary-based RQs for
birds (20 g) estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category
Almonds, hazelnut &
walnut
Citrus
Broccoli, cabbage,
Chinese cabbage,
cauliflower, kohlrabi
Kale, CollardsS
Mustard Green
Sweet corn
Run
No.
1
1
1
2
1
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Acute Dose-
Based(1)
37.60
17.23
21.15
2.35
0.52
47.00
21.54
26.44
2.94
0.65
24.39
11.18
13.72
1.52
0.34
14.10
6.46
7.93
0.88
0.20
28.20
12.93
15.86
Sub Acute
Dietary-
Basec/1*
0.60
0.27
0.34
0.04

0.75
0.34
0.42
0.05

0.39
0.18
0.22
0.02

0.22
0.10
0.13
0.01

0.45
0.20
0.25
Chronic
Dietary-
Basec/1*
16.00
7.33
9.00
1.00

20.00
9.17
11.25
1.25

10.38
4.76
5.84
0.65

6.00
2.75
3.38
0.38

12.00
5.50
6.75
                                       132

-------
Table 5.3 Summary of the acute and chronic dose- and dietary-based RQs for
birds (20 g) estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category

Cotton (ground)
Cotton (Areal)
Apples
Apricot, nectarine,
peach, cherry, pear,
plum & prune
Lettuce & Brussels
sprouts
Cucumber, melons,
pumpkin & squash
Eggplant
Run
No.

1
3
1
1
1
1
3
Dietary Category
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Acute Dose-
Based(1)
1.76
0.39
18.80
8.62
10.58
1.18
0.26
18.29
8.38
10.29
1.14
0.25
47.00
21.54
26.44
2.94
0.65
47.00
21.54
26.44
2.94
0.65
26.71
12.24
15.02
1.67
0.37
24.39
11.18
13.72
1.52
0.34
9.40
4.31
Sub Acute
Dietary-
Base/1*
0.03

0.30
0.14
0.17
0.02

0.29
0.13
0.16
0.02

0.75
0.34
0.42
0.05

0.75
0.34
0.42
0.05

0.42
0.19
0.24
0.03

0.39
0.18
0.22
0.02

0.15
0.07
Chronic
Dietary-
Base/1*
0.75

8.00
3.67
4.50
0.50

7.78
3.57
4.38
0.49

20.00
9.17
11.25
1.25

20.00
9.17
11.25
1.25

11.36
5.21
6.39
0.71

10.38
4.76
5.84
0.65

4.00
1.83
133

-------
Table 5.3 Summary of the acute and chronic dose- and dietary-based RQs for
birds (20 g) estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category

Ornamentals & shade
trees
Potato
Sweet potato
Dry beans (except
Lima), peas & pepper
Carrot
Celery
Run
No.

1


1




3




1




2




2


Dietary Category
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Acute Dose-
Based(1)
5.29
0.59
0.13
11.42
5.23
6.42
0.71
0.16
26.71
12.24
15.02
1.67
0.37
15.01
6.88
8.45
0.94
0.21
26.71
12.24
15.02
1.67
0.37
18.80
8.62
10.58
1.18
0.26
18.80
8.62
10.58
1.18
0.26
Sub Acute
Dietary-
Base/1*
0.08
0.01

0.18
0.08
0.10
0.01

0.42
0.19
0.24
0.03

0.24
0.11
0.13
0.01

0.42
0.19
0.24
0.03

0.30
0.14
0.17
0.02

0.30
0.14
0.17
0.02

Chronic
Dietary-
Base/1*
2.25
0.25

4.86
2.23
2.73
0.30

11.36
5.21
6.39
0.71

6.39
2.93
3.59
0.40

11.36
5.21
6.39
0.71

8.00
3.67
4.50
0.50

8.00
3.67
4.50
0.50

134

-------
Table 5.3 Summary of the acute and chronic dose- and dietary-based RQs for
birds (20 g) estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category
Strawberry
Tomato
Run
No.


2




1


Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small
Insects
Fruits/Pods/ Seeds/ Large
Insects
Granivore
Acute Dose-
Based(1)
20.20
9.26
11.36
1.26
0.28
24.39
11.18
13.72
1.52
0.34
Sub Acute
Dietary-
Base/1*
0.32
0.15
0.18
0.02

0.39
0.18
0.22
0.02

Chronic
Dietary-
Base/1*
8.59
3.94
4.83
0.54

10.38
4.76
5.84
0.65

111 RQ values in bold indicate exceedence of acute LOG (0.1 ) and chronic LOG (1 .0). Acute dose-based RQs were
based on the dose-based EECs divided by the adjusted LD50 for a 20 g bird (14.54 mg/kg b.w.). Acute dietary-based
RQs were based on the dietary EECs divided by the avian acute dietary LC50 (805 mg/kg diet). Chronic RQs based
on the dietary EECs divided by the avian NOEC (30 mg/kg diet).
5.1.2.2 Mammals

Potential risks to mammals are derived using T-REX, acute and chronic rat toxicity data,
and a two body-size and dietary categories. Potential direct acute effects specifically to
the salt marsh harvest mouse (SMHM) are derived by considering dose- and dietary-
based EECs modeled in T-REX for a small mammal (15 g) consuming a variety of
dietary items (Table 5.4) and acute oral and subacute dietary toxicity endpoints for rats.
Potential direct acute effects specifically to the San Joaquin kit fox (SJKF) are derived by
considering dose- and dietary-based EECs modeled in T-REX for a large mammal (1,000
g) consuming a variety of dietary items (Table 5.4) and acute oral and subacute dietary
toxicity endpoints for rats. Based on the rat acute data (LD50=10 kg/kg b.w.) adjusted for
differences in body weight for a 15 g small mammal by T-REX (LD50=21.98) and the
modeled dose-based EECs for various use scenarios and diet categories, all RQs for small
mammals ranged from 0.07 to 26.03 with 96 of the 100 modeled scenarios exceeding the
Agency's acute listed species risk LOG (0.1). Based on the adjusted LD50 for a large
mammal (LD50=7.69 mg/kg b.w.), and the modeled EECs, the acute dose-based RQs for
large mammals ranged from 0.03 to 11.92, with 84 of the 100 modeled scenarios
exceeding the Agency's LOG (0.1).

Potential chronic risks to mammals are derived using a calculated NOAEL (0.75 mg/kg
b.w.) from T-REX which is based on the chronic dietary NOAEC of 15 mg/kg diet for
rats. This calculated toxicity value was further adjusted in T-REX based on the size class
of the mammals (1.65 mg/kg b.w for 15 g and 0.58 mg/kg b.w. for 1000 g). These
                                      135

-------
adjusted values, coupled with the dose-based EECs, resulted in RQs that range from 0.96
to 347.04 for the small mammal (15 g) and from 0.44 to 158.9 for the large mammal
(1000 g) (see Table 5.4). Ninety-nine of the 100 RQs for the small mammal, and 91 of
100 RQs for the large mammal exceeded the Agency's chronic risk LOG (RQ>1) for
mammals. Additional chronic risks to mammals are derived using the chronic dietary-
based EECs and the dietary chronic value for rats (15 mg/kg diet). These chronic dietary-
based RQs ranged from 0.50 to 40, with 72 of the 100 modeled scenarios resulting in
RQs exceeding the Agency's chronic risk LOG (RQ>1)

Based on exceedences of the Agency's acute risk LOG (RQ>0.1) and chronic risk LOG
(RQ>1) endosulfan has the potential to directly affect the SMHM and SJKF.
Additionally, since the acute and chronic RQs are exceeded, there is a potential for
indirect effects to those listed species that rely on mammals during at least some portion
of their life-cycle (i.e., CRLF, SFGS, SMHS, and SJKF).
Table 5.4 Summary of the acute and chronic dose- and dietary-based RQs for
mammals estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category
Almonds,
hazelnut & walnut
Citrus
Broccoli,
cabbage, Chinese
cabbage,
cauliflower,
kohlrabi
Kale, CollardsS
Mustard Green
Sweet corn
Cotton (ground)
Run
No.
1
1
1
2
1
1
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Acute Dose-
Based(1)
15 g
20.82
9.54
11.71
1.30
0.29
26.03
11.93
14.64
1.63
0.36
13.51
6.19
7.60
0.84
0.19
7.81
3.58
4.39
0.49
0.11
15.62
7.16
8.78
0.98
0.22
10.41
4.77
5.86
0.65
1000 g
9.53
4.37
5.36
0.60
0.13
11.92
5.46
6.70
0.74
0.17
6.18
2.83
3.48
0.39
0.09
3.58
1.64
2.01
0.22
0.05
7.15
3.28
4.02
0.45
0.10
4.77
2.18
2.68
0.30
Chronic Dose-
Based(1)
15 g
277.63
127.25
156.17
17.35
3.86
347.04
159.06
195.21
21.69
4.82
180.09
82.54
101.30
11.26
2.50
104.11
47.72
58.56
6.51
1.45
208.22
95.44
117.13
13.01
2.89
138.82
63.62
78.08
8.68
1000 g
127.12
58.26
71.51
7.95
1.77
158.90
72.83
89.38
9.93
2.21
82.46
37.79
46.38
5.15
1.15
47.67
21.85
26.82
2.98
0.66
95.34
43.70
53.63
5.96
1.32
63.56
29.13
35.75
3.97
Chronic
Dietary(1)-
Based
32.00
14.67
18.00
2.00

40.00
18.33
22.50
2.50

20.76
9.51
11.68
1.30

12.00
5.50
6.75
0.75

24.00
11.00
13.50
1.50

16.00
7.33
9.00
1.00
                                      136

-------
Table 5.4 Summary of the acute and chronic dose- and dietary-based RQs for
mammals estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category

Cotton (Areal)
Apples
Apricot, nectarine,
peach, cherry,
pear, plum &
prune
Lettuce &
Brussels sprouts
Cucumber,
melons, pumpkin
& squash
Eggplant
Ornamentals &
shade trees
Potato
Sweet potato
Run
No.

3
1
1
1
1
3
1
1
3
Dietary Category
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Acute Dose-
Based(1)
15 g
0.14
10.13
4.64
5.70
0.63
0.14
26.03
11.93
14.64
1.63
0.36
26.03
11.93
14.64
1.63
0.36
14.79
6.78
8.32
0.92
0.21
13.51
6.19
7.60
0.84
0.19
5.21
2.39
2.93
0.33
0.07
6.32
2.90
3.56
0.40
0.09
14.79
6.78
8.32
0.92
0.21
8.31
3.81
4.68
0.52
0.12
1000 g
0.07
4.64
2.13
2.61
0.29
0.06
11.92
5.46
6.70
0.74
0.17
11.92
5.46
6.70
0.74
0.17
6.77
3.10
3.81
0.42
0.09
6.18
2.83
3.48
0.39
0.09
2.38
1.09
1.34
0.15
0.03
2.90
1.33
1.63
0.18
0.04
6.77
3.10
3.81
0.42
0.09
3.81
1.74
2.14
0.24
0.05
Chronic Dose-
Based(1)
15 g
1.93
135.07
61.90
75.97
8.44
1.88
347.04
159.06
195.21
21.69
4.82
347.04
159.06
195.21
21.69
4.82
197.18
90.38
110.91
12.32
2.74
180.09
82.54
101.30
11.26
2.50
69.41
31.81
39.04
4.34
0.96
84.31
38.64
47.42
5.27
1.17
197.18
90.38
110.91
12.32
2.74
110.86
50.81
62.36
6.93
1.54
1000 g
0.88
61.84
28.35
34.79
3.87
0.86
158.90
72.83
89.38
9.93
2.21
158.90
72.83
89.38
9.93
2.21
90.29
41.38
50.79
5.64
1.25
82.46
37.79
46.38
5.15
1.15
31.78
14.57
17.88
1.99
0.44
38.60
17.69
21.71
2.41
0.54
90.29
41.38
50.79
5.64
1.25
50.76
23.27
28.55
3.17
0.71
Chronic
Dietary(1)-
Based

15.57
7.14
8.76
0.97

40.00
18.33
22.50
2.50

40.00
18.33
22.50
2.50

22.73
10.42
12.78
1.42

20.76
9.51
11.68
1.30

8.00
3.67
4.50
0.50

9.72
4.45
5.47
0.61

22.73
10.42
12.78
1.42

12.78
5.86
7.19
0.80

137

-------
Table 5.4 Summary of the acute and chronic dose- and dietary-based RQs for
mammals estimated based on the maximum endosulfan foliar spray applications
using T-REX version 1.4.1.
Use Category
Dry beans
(except Lima),
peas& pepper
Carrot
Celery
Strawberry
Tomato
Run
No.
1
2
2
2
1
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Acute Dose-
Based(1)
15 g
14.79
6.78
8.32
0.92
0.21
10.41
4.77
5.86
0.65
0.14
10.41
4.77
5.86
0.65
0.14
11.19
5.13
6.29
0.70
0.16
13.51
6.19
7.60
0.84
0.19
1000 g
6.77
3.10
3.81
0.42
0.09
4.77
2.18
2.68
0.30
0.07
4.77
2.18
2.68
0.30
0.07
5.12
2.35
2.88
0.32
0.07
6.18
2.83
3.48
0.39
0.09
Chronic Dose-
Based(1)
15 g
197.18
90.38
110.91
12.32
2.74
138.82
63.62
78.08
8.68
1.93
138.82
63.62
78.08
8.68
1.93
149.13
68.35
83.89
9.32
2.07
180.09
82.54
101.30
11.26
2.50
1000 g
90.29
41.38
50.79
5.64
1.25
63.56
29.13
35.75
3.97
0.88
63.56
29.13
35.75
3.97
0.88
68.29
31.30
38.41
4.27
0.95
82.46
37.79
46.38
5.15
1.15
Chronic
Dietary(1)-
Based
22.73
10.42
12.78
1.42

16.00
7.33
9.00
1.00

16.00
7.33
9.00
1.00

17.19
7.88
9.67
1.07

20.76
9.51
11.68
1.30

|1)RQ values in bold indicate exceedence of listed species acute LOG (0.1) and chronic LOG (1.0). Acute dose-based
RQs were based on dose-based EECs divided by adjusted rat LDSOs (21 .98 mg/kg b.w for 15 g and 7.69 mg/kg b.w. for
1 0OOg). Chronic dose-based RQs were based on dose-based EECs divided by adjusted rat NOAECs (1 .65 mg/kg b.w.
for 15 g and 0.58 mg/kg b.w. for 1000g). Chronic dietary-based RQs were based on dietary EECs divided by the chronic
dietary NOAEC for rats (15 mg/kg diet).
              5.1.2.3 Terrestrial Invertebrates

Potential risks to terrestrial invertebrates resulting from foliar spray applications of
endosulfan are derived using T-REX, and the most sensitive toxicity data available for
terrestrial invertebrates. In the case of endosulfan, the beet web worm was used as a
surrogate for evaluating risks of endosulfan to terrestrial invertebrates. The toxicity value
for terrestrial invertebrates is the lowest available acute contact LD50 (0.15 jig a.i./g)
EECs in ppm calculated by T-REX for small and large insects are divided by the above
toxicity value for terrestrial invertebrates. Larvae for both the Bay checkerspot butterfly
and the Valley elderberry longhorn beetle are considered 'small insects' in this
assessment, while the adults of these species are considered 'large insects'. Based on the
EECs and the above contact acute value, all 20 use category RQs for small insects ranged
from 450 to 2250, and those for large insects ranged from 50 to 250 (Table 5.5). All 20
                                        138

-------
use category RQs for both size classes exceeded the Agency's interim LOG for listed
terrestrial invertebrates (RQ>0.05). Based on the exceedences of the Agency's interim
LOG for listed terrestrial invertebrates, endosulfan use in CA does have the potential to
directly adversely affect the BCB and the VELB  (Table 5.5). Additionally, since the RQs
are exceeded, there is a potential for indirect effects to those listed species that rely on
terrestrial invertebrates during at least some portion of their life-cycle (i.e., CRLF, SFGS,
CIS, SMHM, and SJKF.
Table 5.5 Summary of the acute and chronic dose- and dietary-based RQs terrestrial insects estimated based on
the maximum endosulfan foliar spray applications using T-REX version 1.4.1.
Use Category
Almonds, hazelnut & walnut
Citrus
Broccoli, cabbage, Chinese
cabbage, cauliflower, kohlrabi
Kale, Collards & Mustard Green
Sweet corn
Cotton (ground)
Cotton (Areal)
Apples
Apricot, nectarine, peach, cherry,
pear, plum & prune
Lettuce & Brussels sprouts
Cucumber, melons, pumpkin &
squash
Eggplant
Ornamentals & shade trees
Potato
Sweet potato
Dry beans (except Lima), peas &
pepper
Carrot
Celery
Strawberry
Tomato
Run No.
1
1
1
2
1
1
3
1
1
1
1
3
1
1
3
1
2
2
2
1
Dietary Category
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Small insects'1'
1800

2250

1168

675

1350

900

876

2250

2250

1278

1168

450

547

1278

719

1278

900

900

967

1168

Large insects

200

250

130

75

150

100

97

250

250

142

130

50

61

142

80

142

100

100

107

130
                                       139

-------
 Table 5.5 Summary of the acute and chronic dose- and dietary-based RQs terrestrial insects estimated based on
 the maximum endosulfan foliar spray applications using T-REX version 1.4.1.
Use Category
Run No.
Dietary Category
Small insects
Large insects
 |2) RQ values in bold indicate exceedence of Agency's interim LOG for listed terrestrial invertebrates (0.05). Terrestrial insect RQs were
 based on diet category EECs for small and large insects divided by the contact acute value for beet webworm, Pyrausta sticticalis
 (LD50=0.15 ug/g).

5.1.2.4 Terrestrial Plants

Generally, for indirect effects, potential effects on terrestrial vegetation are assessed
using RQs from terrestrial plant seedling emergence and vegetative vigor EC25 data as a
screen. However, such toxicity data have not been identified for quantitatively estimating
risk to terrestrial plants (as described in Section 4.2.4) as a result of endosulfan use;
therefore, RQs cannot be calculated at this time for this taxonomic group. Discussion
regarding lines of evidence for the potential for indirect effects to those listed species that
rely on terrestrial plants during at least some portion of their life-history (i.e., terrestrial-
phase CRLF, SFGS, CIS, BCB, VELB, SMHM and SJKF) can be found in the "Risk
Description" portion of the chapter (Section 5.2).

5.1.3 Primary Constituent Elements of Designated Critical Habitat

For endosulfan use, the assessment endpoints for designated critical habitat PCEs involve
the same endpoints as those being assessed relative to the potential for direct and indirect
effects to the listed species assessed here. Therefore, the effects determinations for direct
and indirect effects are used as the basis of the effects determination for designated
critical habitat.

5.1.4 Spatial Extent of Potential Effects

In order to determine the spatial extent of effects on terrestrial and aquatic habitats due to
endosulfan exposures through spray drift, it is necessary to estimate the distance that
spray applications can drift from the treated area and still be present at concentrations that
exceed levels of concern. An analysis of spray drift distances was completed using all
available tools, including AgDrift, AGDISP and  the Gaussian extension to AGDISP.

       5.1.4.1 Spray Drift

Spray drift analysis determines  the additional distance from the treated area where listed
species LOCs  are  exceeded as a result of spray drift.   This distance is based  on the
taxonomic group that yields the largest RQ to  LOG ratio.   Both terrestrial and  aquatic
taxonomic groups are considered in this analysis.

For endosulfan, the results of the screening-level assessment indicate that the required
values are one value for LOC= 0.05 and two values for RQ (2,250 for apples and 1,350
for sweet corn). Using these values, AgDrift and AGDISP were not able to estimate
needed buffer distance, therefore the Gaussian extension to AGDISP was used. Buffer
distances were determined for two crops, an orchard and a field crop with varying
                                         140

-------
    application parameters as described in details in Appendix G. Spray drift results range
    from 12,266 to 12,775 feet for orchards and 3,383 to 11,279 feet for field crops.

           5.1.4.2 Downstream Dilution Analysis

    Down-stream dilution analysis is necessary to define the full extent of the effects area.
    This is because effects area may be larger than the initial area of concern or "footprint" of
    potential uses. This analysis determines downstream extent of exposure in streams and
    rivers where the EEC could potentially be above levels that would exceed the highest RQ
    to LOG ratio. Based on all aquatic RQs, the greatest RQ to LOG ratio for  all aquatic
    organisms (plants and animals) are determined for  the two major use patterns (Table
    5.6).

    Table 5.6 Summary of the highest RQ to LOG ratios for the two GIS use categories for endosulfan
Category*
Cultivated Crops
Orchards/Vineyards
Use Pattern
Lettuce/Brussels Sprouts
Pear
Species
Common Carp
RQ
58.8
33.5
we
0.05
    * Two GIS mapping categories.

    The ratios in Table 5.6 would be used to determine the downstream extent of the effects
    area.  Results  of this  analysis for  endosulfan  suggest that  the maximum  distance
    downstream is 285 km for cultivated crops and 120 km for orchards. Therefore, exposure
    and possible  effects  on most sensitive aquatic  species may extend to 120 to 285 km
    beyond the initial area of concern shown in Figure 2.6.

    5.1.4.3 Overlap between CRLF, CTS, SFGS, BCB, VELB, SMHM, and SJKF
    habitat and Spatial Extent of Potential Effects

    Table 5.7 shows a summary of recent usage data in counties where the CRLF and the
    San Francisco Bay species occur.
Table 5.7 Summary of average endosulfan usage data and occurrence of the CRLF and the San Francisco Bay
species at the county level
County
Alameda
Colusa
Contra Costa
Fresno
Glenn
Imperial
Kern
Kings
Average Usage (2005-2006)
Ibs
%
Rank2
No Use Reported
250.96
0.29%

No Use Reported
44,937.84
4.16
7,004.91
2,584.47
19,648.25
51%
0.00%
8%
3%
22%
1

3
5
2
CRLF and San Francisco Bay Species
CRLF
0

0
0


0

SFGS








SMHM
0

0





BCB
0

0





VELB








SJKF
a

0
0


0
0
CTS
****







                                           141

-------
Table 5.7 Summary of average endosulfan usage data and occurrence of the CRLF and the San Francisco Bay
species at the county level
County
Los Angeles
Madera
Matin
Merced
Monterey
Napa
Placer
Riverside
San Benito
San Bernardino
San Francisco
San Joaquin
San Luis Obispo
San Mateo
Santa Barbara
Santa Clara
Santa Cruz
Siskiyou
Solano
Sonoma
Stanislaus
Sutler
Tulare
Ventura
Yolo
Average Usage (2005-2006)
Ibs
40.13
468.2
%
0.05%
0.53%
Rank2


No Use Reported
230.2
396.45
4.56
9.09
2,926.87
646.58
0.26%
0.45%
0.01%
0.01%
3%
0.74%




4
11
No Use Reported
No Use Reported
137.55
0.16%

No Use Reported
0.25
1
173.33
0.00%
0.00%
0.20%



No Use Reported
2,063.91
885.72
55.75
2%
1%
0.06%
7
10

No Use Reported
1,027.73
1,879.55
63.38
2,151.40
1%
2%
0.07%
2%
9
8

6
CRLF and San Francisco Bay St
CRLF
0

0
0
0
0
0
0
0
0
s
0
0
0
0
0
0

0
0
0


0

SFGS













0











SMHM


0


H




0


0

0


0
0





. i
pecies
BCB










0


0

0









VELB


















0






SJKF

0

0

0


0



0
0
0
0




0

0


CIS

























Total          87,592.24   100%
    0  Species occurrence= Yes
 California red-legged frog (Rana aurora draytonii) (CRLF), San Francisco Garter Snake (Thamnophis sirtalis
tetrataenia) (SFGS), Salt Marsh Harvest Mouse (Reithrodontomys raviventris)
 (SMHM), Bay Checkerspot Butterfly (Euphydryas editha bayensis) (BCB), Valley Elderberry Longhorn Beetle
(Desmocerus californicus dimorphus) (VELB), San Joaquin Kit Fox (Vulpes macrotis mutica) (SJKFj, and California
Tiger Salamander (Ambystoma californiense) (CTS).
 Top 11 counties with usage over 500 Ib/county	


     In contrast, possible overlap of the species occurrence and potential usage are shown in
     maps included in Appendix E. Co-occurrence data are summarized in Table 5.8.

     Table 5.8 Extent and location of overlap between areas of species occurrence and areas
     of potential endosulfan usage
    Species
                   Extent of possible overlap/Counties of Occurrence
    CRLF
Low with high acreage of both cultivated crops and orchards in the following Counties: Santa
                                               142

-------

SFGS
SMHM
BCB
VELB
SJKF
CTS
Clara, Santa Barbara, Alameda, Contra Costa, Ventura, San Benito, Merced, Stanislaus, and San
Joaquin
High with low acreage of cultivated crops only in San Mateo County
Very low with low acreage of cultivated crops in only in Sonoma County
High with low acreage of cultivated crops and orchards only in Santa Clara County
Medium with cultivated crops and orchards in only Solano County
Low with high acreage of mainly cultivated crops in the western edge of the central valley
counties including: Contra Costa, Alameda, San Bento, Monterey, San Joaquin, Stanislaus,
Merced, Kings, Fresno and Kern.
Low with high acreage of mainly orchards in the eastern edge of the central valley counties
including: : Sacramento, Santa Clara, Stanislaus, Alameda, Contra Costa, Merced, Madera, and
Fresno
Data show the species occurrence may overlap with potential endosulfan usage although
the extent appears to be generally limited because it is either low or occurs on limited
acreage.  It is important to note however, the following contradicting factors:

   (1) Drift is expected to cause extension of the initial area of concern for most
       sensitive terrestrial species by 2.1 miles for cultivated crops to 2.4 miles for
       orchards.
   (2) Downstream dilution is expected to cause extension of the initial area of concern
       for most sensitive aquatic species by 75 miles (120 km) for cultivated crops to
       177 miles (285 km) for orchards.
   (3) Areas associated with endosulfan usage are expected to be much smaller than
       those mapped because endosulfan is not used on all cultivated crops or on all
       orchards and vines.

The first two factors would tend to increase the extent of the area affected while the third
factor would reduce the possibility of the overlap to occur.

5.2 Risk Description

The risk description synthesizes overall conclusions regarding the likelihood of adverse
impacts leading to an effects determination (i.e..,  "no effect," "may affect, but not likely
to adversely affect," or "likely to adversely affect") for the assessed species and the
potential for effects to their designated critical habitat.

If the RQs presented in the Risk Estimation (Section 5.1) show no direct or indirect
effects for the assessed species, and no effects to PCEs of the designated critical habitat,
a "no effect" determination is made, based on the use of endosulfan within the action
area. However, if LOCs for direct or indirect effect are exceeded or there may be effects
to the PCEs of the critical habitat, the Agency concludes a preliminary "may affect"
determination for the FIFRA regulatory action regarding endosulfan

Based on exceedances of LOCs, a potential to cause adverse effects to CRLF, SFGS,
SMHM, BCB, VELB, SJKF  and CTS and potential to cause effects designated critical
habitat (CRLF, BCB, VELB  and CTS) has been  identified. The Agency concludes a
                                       143

-------
      preliminary "may affect" determination for the CRLF, SFGS, SMHM, BCB, VELB,
      SJKF and CTS based on the currently labeled uses of endosulfan. A summary of the risk
      estimation results are provided in Table 5.9 for direct and indirect effects to the listed
      species assessed here and in Table 5.10 for the PCEs of their designated critical habitat.
Table 5.9  Risk Estimation Summary for endosulfan - Direct and Indirect Effects
      Taxa"
                      LOC Exceedances
                           (Y/N)
                                                   Description of Results of Risk Estimation
                                                                        Assessed Species
                                                                       Potentially Affected
Freshwater Fish and
Aquatic-phase
Amphibians
(Y)
                                       Acute RQs range from 7.2 to 58.8; exceeding the Agency's
                                       acute listed species LOC (0.05) in all 20 crop scenarios.

                                       Chronic RQs range from 5.7 to 80.6; exceeding the Agency's
                                       chronic species LOC (1.0) in all 20 crop scenarios.
                                                                     Direct Effects:

                                                                     Aquatic-phase
                                                                     CRLF,CTS
Acute RQs range from 1.2 to 9.8 for freshwater invertebrates;
exceeding the Agency's acute listed species LOC (0.05) in all
20 crop scenarios.

Chronic RQs range from 23.7 to 260; exceeding the Agency's
chronic species LOC (1.0) in all 20 crop scenarios.	
                                                                     Indirect Effects:

                                                                     Aquatic-phase CRLF,
                                                                     SFGS, CTS	
Freshwater
Invertebrates
                           (Y)
                                                                     Indirect Effects:

                                                                     Aquatic-phase CRLF,
                                                                     SFGS, CTS
Vascular Aquatic
Plants
                       (N; however
                     effects cannot be
                        quantified)
                                       There are no acceptable vascular aquatic plant data available
                                                                     Indirect Effects:

                                                                     Aquatic-phase CRLF,
                                                                     SFGS, SMHM, CTS
Non-Vascular
Aquatic Plants
                           (N)
            There are no exceedances of the Agency's LOC for aquatic
            plants (1.0)
                                                        Indirect Effects:

                                                        Aquatic-phase CRLF,
                                                        SFGS, SMHM, CTS
Birds, Reptiles, and
Terrestrial-Phase
Amphibians
                           (Y)
            Acute dose diet-based RQs range from 0.15 to 47.0 for short
            grass; 0.07 to 21.5 fortall grass; 0.08 to 26.4 forbroadleaf
            plants/small insects; 0.01 to 2.9 for fruits/pods/seeds/large
            insect, and 0.03 to 0.65 for the granivore dietary categories.
            For short-grass, RQs exceed the Agency's acute listed species
            LOC (0.1) in all 20 crop scenarios and at least one RQ in all
            five dietary categories exceeded the acute listed species LOC
            (0.1).

            Chronic RQs for birds range from 4.0 to 20.0 for short grass;
            1.8 to 9.2 fortall grass; 2.3 to 11.3 forbroadleaf plants/small
            insects; and 0.25 to 1.3 for fruits/pods/seeds/large insect
            dietary categories.  For short-grass, tall grass, and broadleaf
            plants/small insect dietary categories, RQs exceed the
            Agency's chronic listed species LOC (1.0) in all 20 crop
            scenarios and at least one RQ in all four dietary categories
            exceeded the acute listed species LOC (1.0).	
                                                                                                Direct Effects:
                                                                                                Terrestrial-phase
                                                                                                CRLF, SFGS, CTS
                                                                                                Indirect Effects:

                                                                                                Terrestrial-phase
                                                                                                CRLF, SFGS, SMHM,
                                                                                                SJKF
                           (Y)
Mammals
                                       Acute dose-based RQs for mammals range from 2.4 to 26.0
                                       for short grass; 1.1 to 11.9 fortall grass; 1.3 to 14.6 for
                                       broadleaf plants/ small insects; 0.15 to 1.6 for fruits/pods/
                                       seeds/large insects; and 0.03 to 0.36 for the granivore dietary
                                       categories.  Acute RQs exceed the Agency's LOC (0.1) for
                                       one or more crop scenarios for all five dietary categories.
                                                                     Direct Effects:

                                                                     SMHM; SJKF
                                                    144

-------
Table 5.9  Risk Estimation Summary for endosulfan  - Direct and Indirect Effects
       Taxam
 LOC Exceedances
      (Y/N)
                                                    Description of Results of Risk Estimation
                                                              Assessed Species
                                                             Potentially Affected
                                         Chronic dose-based RQs for mammals range from 31.8 to
                                         347 for short grass; 14.6 to 159 for tall grass; 17.9 to 195
                                         broadleaf plants/small insects; 2.0 to 21.7 for
                                         fruits/pods/seeds/large insect; and 0.44 to 4.8 for the
                                         granivore dietary categories. Chronic RQs exceed the
                                         Agency's chronic listed species LOC (1.0) in all 20 crop
                                         scenarios for four of the five dietary categories modeled.
                                                                            Indirect Effects:

                                                                            Terrestrial-phase
                                                                            CRLF, SFGS, SMHM,
                                                                            SJKF
Terrestrial
Invertebrates
                                                                                                  Direct Effects:
                                                                            BCB, VELB
                            (Y)
                   RQs based on EECs for small insects and large insects range
                   from 450 to 2250 and 50 to 250, respectively
                                                          Indirect Effects:

                                                          Terrestrial phase
                                                          CRLF, SFGS,
                                                          SMHM, SJKF, CTS
Terrestrial Plants •
Monocots
  (N; however
effects cannot be
   quantified)
There are no acceptable terrestrial plant data available
Indirect Effects:
CRLF, SFGS,
SMHM, BCB, VELB,
SJKF, CTS
Terrestrial Plants •
Dicots
  (N; however
effects cannot be
   quantified)
There are no acceptable terrestrial plant data available
Indirect Effects:
CRLF, SFGS,
SMHM, BCB, VELB,
SJKF, CTS
         Risks to estuarine/marine fish and invertebrates were not assessed because the listed species for
       endosulfan do not comprise this taxonomic group nor does available life history and critical habitat
       information indicate they would be affected indirectly by effects on estuarine/marine species or this
       habitat.
Table 5.10 Risk Estimation Summary for Endosulfan - Effects to Designated Critical Habitat (PCEs)
Taxa(1)
Freshwater Fish and
Aquatic -phase
Amphibians
Freshwater
Invertebrates
Vascular Aquatic
Plants
Non-Vascular
May Affect
Habitat
(Y/N)
(Y)
(Y)
(N; however
effects cannot
be quantified)
(N)
Description of Results of Risk Estimation
Acute RQs range from 7.2 to 58.8; exceeding the Agency's
acute listed species LOC (0.05) in all 20 crop scenarios.
Chronic RQs range from 5.7 to 80.6; exceeding the Agency's
chronic species LOC (1.0) in all 20 crop scenarios.
Acute RQs range from 1.2 to 9.8 for freshwater invertebrates;
exceeding the Agency's acute listed species LOC (0.05) in all
20 crop scenarios.
Chronic RQs range from 23.7 to 260; exceeding the Agency's
chronic species LOC (1.0) in all 20 crop scenarios.
There are no acceptable vascular aquatic plant data available
There are no exceedances of the Agency's LOC for aquatic
Species Associated
with Designated
Critical Habitat that
May be Modified
CRLF,CTS
CRLF,CTS
CRLF,CTS
CRLF,CTS
                                                     145

-------
Table 5.10  Risk Estimation Summary for Endosulfan - Effects to Designated Critical Habitat (PCEs)
      Taxa(1)
 May Affect
   Habitat
    (Y/N)
        Description of Results of Risk Estimation
 Species Associated
  with Designated
Critical Habitat that
  May be Modified
Aquatic Plants
                plants (1.0)
Birds, Reptiles, and
Terrestrial-Phase
Amphibians
     (Y)
Acute dose diet-based RQs range from 0.15 to 47.0 for short
grass; 0.07 to 21.5 for tall grass; 0.08 to 26.4 forbroadleaf
plants/small insects; 0.01 to 2.9 for fruits/pods/seeds/large
insect, and 0.03 to 0.65 for the granivore dietary categories.
For short-grass, RQs exceed the Agency's acute listed species
LOG (0.1) in all 20 crop scenarios and at least one RQ in all
five dietary categories exceeded the acute listed species LOG
(0.1).

Chronic RQs for birds range from 4.0 to 20.0 for short grass;
1.8 to 9.2 fortall grass; 2.3 to 11.3 forbroadleaf plants/small
insects; and 0.25 to 1.3 for fruits/pods/seeds/large insect
dietary categories. For short-grass, tall grass, and broadleaf
plants/small insect dietary categories, RQs exceed the
Agency's chronic listed species LOG (1.0) in all 20 crop
scenarios and at least one RQ in all four dietary categories
exceeded the acute listed species LOG (1.0).	
                                                                                              CRLF,CTS
                          (Y)
Mammals
                Acute dose-based RQs for mammals range from 2.4 to 26.0
                for short grass; 1.1 to 11.9 for tall grass; 1.3 to 14.6 for
                broadleaf plants/small insects; 0.15 to 1.6 for
                fruits/pods/seeds/large insects; and 0.03 to 0.36 for the
                granivore dietary categories.  Acute RQs exceed the Agency's
                LOG (0.1) for one or more crop scenarios for all five dietary
                categories.	
                                                                                              CRLF
                                     Chronic dose-based RQs for mammals range from 31.8 to 347
                                     for short grass; 14.6 to 159 for tall grass; 17.9 to 195
                                     broadleaf plants/small insects; 2.0 to 21.7 for
                                     fruits/pods/seeds/large insect; and 0.44 to 4.8 for the
                                     granivore dietary categories. Chronic RQs exceed the
                                     Agency's chronic listed  species LOG (1.0) in all 20  crop
                                     scenarios for four of the  five dietary categories modeled.	
Terrestrial
Invertebrates
                                                                        CRLF,CTS
                          (Y)
                RQs based on EECs for small insects and large insects range
                from 450 to 2250 and 50 to 250, respectively
Terrestrial Plants
Monocots
 (N; however
effects cannot
be quantified)
There are no acceptable terrestrial plant data available
                                                         CRLF,BCB, VELB,
                                                         CTS
Terrestrial Plants
Dicots
 (N; however
effects cannot
be quantified)
There are no acceptable terrestrial plant data available
CRLF,BCB, VELB,
CTS
     Following a "may affect" determination, additional information is considered to refine
     the potential for exposure at the predicted levels based on the life history characteristics
     (i.e., habitat range, feeding preferences, etc.) of the assessed species. Based on the best
     available information, the Agency uses the refined evaluation to distinguish those actions
                                                  146

-------
that "may affect, but are not likely to adversely affect" from those actions that are "likely
to adversely affect" the assessed species and its designated critical habitat.

The criteria used to make determinations that the effects of an action are "not likely to
adversely affect" the assessed species or its designated critical habitat include the
following:

    •   Significance of Effect: Insignificant effects are those that cannot be meaningfully
       measured, detected, or evaluated in the context of a level of effect where "take"
       occurs for even a single individual. "Take" in this context means to harass or
       harm, defined as the following:
            •   Harm includes significant habitat modification or degradation that
                results in death or injury to listed species by significantly impairing
                behavioral patterns such as breeding, feeding, or sheltering.
            •   Harass is defined as actions that create the likelihood of injury to listed
                species to such an extent as to significantly disrupt normal behavior
                patterns which include, but are not limited to, breeding,  feeding, or
                sheltering.
    •  Likelihood of the Effect Occurring:  Discountable effects are those that are
       extremely unlikely to occur.
    •  Adverse Nature of Effect:  Effects that are wholly beneficial without any adverse
       effects are not considered adverse.

 A description of the risk and effects determination for each of the established assessment
endpoints for the assessed species  and their designated critical habitat is provided in
Sections 5.2.1 through 5.2.7. The effects determination section for each  listed species
assessed will follow a similar pattern. Each will start with  a discussion  of the potential
for direct effects, followed by a discussion of the potential  for indirect effects.  The
likelihood of effects occurring will be informed by the probit-based chance of individual
acute effects for freshwater fish and invertebrates (Table 5.11), terrestrial birds,
mammals and insects (Table 5.12) and herpetofauna based on T-HERPS modeling
(Table 5.13).  For those listed species that have designated critical habitat, the section
will end with a discussion on the potential for effects to the critical habitat from the use
of endosulfan.
                                         147

-------
Table 5.11 Summary of the chance of individual acute effects to
acute RQs, the acute listed species LOC, acute toxicity data, and
relationships0'2-1.
freshwater animals based on
probit slope response
Use Category
N/A
Almonds, Hazelnuts &
Walnuts
Citrus
Broccoli, Cabbage &
Cauliflower
Collards, Kale &
Mustard Green
Sweet corn for fresh
market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels
Sprouts
Cucurbits
Eggplant
Ornamentals or Shade
Trees (Southern
Coast)
Ornamentals or Shade
Trees (Northern
Central coast)
Potato
Potato (Northern
Central coast)
Run
No.
N/A
1
1
1
2
1
2
3
1
1
1
3
1
1 Add
1
1 Add
Acute FW
Fish RQ
LOC =
0.05
33.50
33.50
48.10
16.70
28.80
12.40
15.60
34.00
58.80
15.50
7.20
34.30
43.80
18.90
26.30
Change of Individual
Effect at RQ or LOC
Based on Probit Slope &
Slope's 95% Cl
1 in4.18E+18
(1 in 216, 1 in 1.75E+31)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
1 in 1 .00
(1 in 1.05, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
1 in 1 .00
Acute FW
Invert. RQ
LOC = 0.05
5.58
5.58
8.02
2.78
4.80
2.07
2.60
5.67
9.80
2.58
1.20
5.72
7.30
3.15
4.38
Change of Individual
Effect at RQ or LOC
Based on Probit Slope
& Slope's 95% Cl
1 in4.18E+18
(1 in 21 6, 1 in 1.75E+31)
1 in 1.00
(1 in 1.07,1 in 1.00)
1 in 1.00
(1 in 1.07,1 in 1.00)
1 in 1.00
(1 in 1.04, 1 in 1.00)
1 in 1.02
(1 in 1.23, 1 in 1.00)
1 in 1.00
(1 in 1.09, 1 in 1.00)
1 in 1.08
(1 in 1.36, 1 in 1.00)
1 in 1.03
(1 in 1.26, 1 in 1.00)
1 in 1.00
(1 in 1.07, 1 in 1.00)
1 in 1.00
(1 in 1.02, 1 in 1.00)
1 in 1.03
(1 in 1.26, 1 in 1.00)
1 in 1.56
(1 in 1.78, 1 in 1.31)
1 in 1.00
(1 in 1.07, 1 in 1.00)
1 in 1.00
(1 in 1.04, 1 in 1.00)
1 in 1.01
(1 in 1.19, 1 in 1.00)
1 in 1.00
                                           148

-------
Use Category

Sweet Potato
Beans & Peas (dry) &
Pepper
Carrot & Celery
Strawberry
Tomato
Run
No.

3
1
2
1
1
Acute FW
Fish RQ

11.80
20.00
17.20
36.40
17.90
Change of Individual
Effect at RQ or LOG
Based on Probit Slope &
Slope's 95% Cl
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.02, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
1 in 1 .00
(1 in 1.00, 1 in 1.00)
1 in 1 .00
(1 in 1.01, 1 in 1.00)
Acute FW
Invert. RQ

1.97
3.33
2.87
6.07
2.98
Change of Individual
Effect at RQ or LOG
Based on Probit Slope
& Slope's 95% Cl
(1 in 1.11, 1 in 1.00)
1 in 1.10
(1 in 1.39, 1 in 1.00)
1 in 1.01
(1 in 1.17, 1 in 1.00)
1 in 1.02
(1 in 1.22,1 in 1.00)
1 in 1.00
(1 in 1.06, 1 in 1.00)
1 in 1.02
(1 in 1.21, 1 in 1.00)
   Bolded RQs exceed the Agency's acute listed species LOG (0.05 for aquatic animals). When acute RQs exceed the Agency's listed
species LOG, the chance of individual effects was calculated at the RQ and the LOG, whereas if there was no exceedance, then the
chance of individual effects was calculated only at the LOG.


   Although an LC50 or an EC50 has been established for all aquatic taxonomic groups, information is unavailable to estimate probit
slopes from  the studies from which the endpoints were derived. Therefore, a default slope of 4.5 with an assumed 95% confidence
interval of 2 and 9 has been assumed as per original Agency assumptions of typical slope cited in Urban and Cook (1986).
                                                          149

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category
N/A
Almonds,
hazelnut &
walnut
Citrus
Broccoli,
cabbage,
Chinese
cabbage,
Dietary Category
N/A
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
20 g
Bird
Acute
RQ
LOC =
0.1
37.60
17.23
21.15
2.35
0.52
47.00
21.54
26.44
2.94
0.65
24.39
11.18
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in1.05E+17
(1 in 6290, 1 in
3.31 E+40)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 121.33
(1 in 6.46, 1 in
1 .28E+04)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1 in 17.33
(1 in 3.96, 1 in
153.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
15 g
Mammal
Acute RQ
LOC = 0.1
20.82
9.54
11.71
1.30
0.29
26.03
11.93
14.64
1.63
0.36
13.51
6.19
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 2.94E+05
(1 in 44, 1 in
8.86E+18)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .03, 1 in 1 .00)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1 .44
(1 in 1.69, 1 in 1.18)
1 in 130.50
(1 in 7.11, 1 in
1.61E+06)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.21
(1 in 1.51, 1 in 1.03)
1 in 42.78
(1 in 5.31, 1 in
2.87E+04)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.00
(1 in 1 .06, 1 in 1 .00)
1000 g
Mammal
Acute
RQ
LOC =
0.1
9.53
4.37
5.36
0.60
0.13
11.92
5.46
6.70
0.74
0.17
6.18
2.83
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 2.94E+05
(1 in 44, 1 in
8.86E+18)
1 in 1.00
(1 in 1.03, 1 in 1.00)
1 in 1.00
(1 in 1.11, 1 in 1.00)
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 6.42
(1 in 3.06, 1 in 46.49)
1 in 2.57E+04
(1 in 25.29, 1 in
7.31 E+1 4)
1 in 1.00
(1 in 1.02, 1 in 1.00)
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 3.54
(1 in 2.51, 1 in 8.01)
1 in 4.55E+03
(1 in 16.92, 1 in
9.70E+11)
1 in 1.00
(1 in 1.06, 1 in 1.00)
1 in 1.02
(1 in 1.22, 1 in 1.00)
Insect
Acute
RQ
LOC =
0.05


1800
200



2250
250



Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl
1 in 29
(1 in 154, 1 in
1220)


1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



150

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category
cauliflower,
kohlrabi
Kale,
Collards &
Mustard
Green
Sweet corn
Dietary Category
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
20 g
Bird
Acute
RQ
13.72
1.52
0.34
14.10
6.46
7.93
0.88
0.20
28.20
12.93
15.86
1.76
0.39
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.06
(1 in 1.34, 1 in 1.01)
1 in3.10E+04
(1 in 22.08, 1 in
6.70E+09)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 3. 12
(1 in 2.37, 1 in 4.33)
1 in 1.14E+09
(1 in 185.19, 1 in
8.52E+20)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.02
(1 in 1 .23, 1 in 1 .00)
1 in 3.70E+03
(1 in 14.00, 1 in
4.07E+07)
15 g
Mammal
Acute RQ
7.60
0.84
0.19
7.81
3.58
4.39
0.49
0.11
15.62
7.16
8.78
0.98
0.22
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.00
(1 in 1 .04, 1 in 1 .00)
1 in 2.70
(1 in 2.26, 1 in 3.94)
1 in 1.86E+03
(1 in 13.69, 1 in
3.27E+10)
1 in 1.00
(1 in 1 .04, 1 in 1 .00)
1 in 1.01
(1 in 1.15, 1 in 1.00)
1 in 1.00
(1 in 1.11, 1 in 1.00)
1 in 12.43
(1 in 3.75, 1 in
396.26)
1 in 1.41E+05
(1 in 37.27, 1 in
5.18E+17)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1 in 1.00
(1 in 1 .03, 1 in 1 .00)
1 in 2.08
(1 in 2.03, 1 in 2. 16)
1 in 709.49
(1 in 10.85, 1 in
8.62E+08)
1000 g
Mammal
Acute
RQ
3.48
0.39
0.09
3.58
1.64
2.01
0.22
0.05
7.15
3.28
4.02
0.45
0.10
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.01
(1 in 1.1 6, 1 in 1.00)
1 in 31 .64
(1 in 4.89, 1 in
9.86E+03)

1 in 1.01
(1 in 1.1 6, 1 in 1.00)
1 in 1.20
(1 in 1.50, 1 in 1.03)
1 in 1.09
(1 in 1.37, 1 in 1.00)
1 in 587.41
(1 in 10.36, 1 in
4.25E+08)

1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 1.01
(1 in 1.18, 1 in 1.00)
1 in 1.00
(1 in 1.1 3, 1 in 1.00)
1 in 17.32
(1 in 4.1 3, 1 in
1 .22E+03)

Insect
Acute
RQ
1168
130



675
75



1350
150

Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl
1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)

151

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category
Cotton
(ground)
Cotton
(Areal)
Apples
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
20 g
Bird
Acute
RQ
18.80
8.62
10.58
1.18
0.26
18.29
8.38
10.29
1.14
0.25
47.00
21.54
26.44
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.38
(1 in 1.67, 1 in 1.21)
1 in 2.79E+06
(1 in 55.89, 1 in
3.62E+14)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1 .45
(1 in 1 .72, 1 in 1 .28)
1 in 4.72E+06
(1 in 62.15, 1 in
1.30E+15)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
15 g
Mammal
Acute RQ
10.41
4.77
5.86
0.65
0.14
10.13
4.64
5.70
0.63
0.14
26.03
11.93
14.64
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.00
(1 in 1 .07, 1 in 1 .00)
1 in 4.99
(1 in 2.82, 1 in
21.50)
1 in 1.27E+04
(1 in 21 .50, 1 in
4.90E+13)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.00
(1 in 1 .07, 1 in 1 .00)
1 in 5.38
(1 in 2.89, 1 in
27.03)
1 in 1.58E+04
(1 in 22.60, 1 in
1.12E+14)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1000 g
Mammal
Acute
RQ
4.77
2.18
2.68
0.30
0.07
4.64
2.13
2.61
0.29
0.06
11.92
5.46
6.70
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.07
(1 in 1.33,1 in 1.00)
1 in 1.03
(1 in 1.24, 1 in 1.00)
1 in 111.34
(1 in 6.83, 1 in
9.03E+05)

1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.08
(1 in 1.34, 1 in 1.00)
1 in 1.03
(1 in 1.25, 1 in 1.00)
1 in 128.84
(1 in 7.09, 1 in
1 .54E+06)

1 in 1.00
(1 in 1.02, 1 in 1.00)
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 1.00
(1 in 1.05, 1 in 1.00)
Insect
Acute
RQ


900
100



876
97



2250
Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl


1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
152

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category

Apricot,
nectarine,
peach,
cherry, pear,
plum & prune
Lettuce &
Brussels
sprouts
Cucumber,
melons,
pumpkin &
Dietary Category
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
20 g
Bird
Acute
RQ
2.94
0.65
47.00
21.54
26.44
2.94
0.65
26.71
12.24
15.02
1.67
0.37
24.39
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1 in 17.33
(1 in 3.96, 1 in
153.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1 in 17.33
(1 in 3.96, 1 in
153.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.03
(1 in 1 .27, 1 in 1 .00)
1 in 7.97E+03
(1 in 16.53, 1 in
2.55E+08)
1 in 1.00
(1 in 1.00, 1 in 1.00)
15 g
Mammal
Acute RQ
1.63
0.36
26.03
11.93
14.64
1.63
0.36
14.79
6.78
8.32
0.92
0.21
13.51
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.21
(1 in 1.51, 1 in 1.03)
1 in 42.78
(1 in 5.31, 1 in
2.87E+04)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.21
(1 in 1.51, 1 in 1.03)
1 in 42.78
(1 in 5.31, 1 in
2.87E+04)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1 in 1.00
(1 in 1 .03, 1 in 1 .00)
1 in 2.28
(1 in 2. 12, 1 in 2.64)
1 in 1.01E+03
(1 in 11.82, 1 in
3.25E+09)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1000 g
Mammal
Acute
RQ
0.74
0.17
11.92
5.46
6.70
0.74
0.17
6.77
3.10
3.81
0.42
0.09
6.18
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 3.54
(1 in 2.51,1 in 8.01)
1 in 4.55E+03
(1 in 16.92, 1 in
9.70E+11)
1 in 1.00
(1 in 1.02, 1 in 1.00)
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 3.54
(1 in 2.51, 1 in 8.01)
1 in 4.55E+03
(1 in 16.92, 1 in
9.70E+11)
1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 1.01
(1 in 1.19, 1 in 1.00)
1 in 1.00
(1 in 1.14, 1 in 1.00)
1 in 21 .54
(1 in 4.39, 1 in
2.57E+03)

1 in 1.00
(1 in 1.06, 1 in 1.00)
Insect
Acute
RQ
250



2250
250
0


1278
142


Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)


153

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category
squash
Eggplant
Ornamentals
& shade
trees
Dietary Category
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
20 g
Bird
Acute
RQ
11.18
13.72
1.52
0.34
9.40
4.31
5.29
0.59
0.13
11.42
5.23
6.42
0.71
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.06
(1 in 1.34, 1 in 1.01)
1 in3.10E+04
(1 in 22.08, 1 in
6.70E+09)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .01 , 1 in 1 .00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 40.31
(1 in 4.93, 1 in
1 .02E+03)
1 in3.53E+13
(1 in 1.37E+03, 1 in
9.08E+31)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 9.39
(1 in 3.34, 1 in 40.29)
15 g
Mammal
Acute RQ
6.19
7.60
0.84
0.19
5.21
2.39
2.93
0.33
0.07
6.32
2.90
3.56
0.40
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.00
(1 in 1 .06, 1 in 1 .00)
1 in 1.00
(1 in 1 .04, 1 in 1 .00)
1 in 2.70
(1 in 2.26, 1 in 3.94)
1 in 1.86E+03
(1 in 13.69, 1 in
3.27E+10)
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 1.05
(1 in 1.29, 1 in 1.00)
1 in 1.02
(1 in 1.21, 1 in 1.00)
1 in 70.91
(1 in 6.07, 1 in
1 .75E+05)
1 in 7.05E+06
(1 in 88.87, 1 in
2.03E+24)
1 in 1.00
(1 in 1 .06, 1 in 1 .00)
1 in 1.02
(1 in 1.22, 1 in 1.00)
1 in 1.01
(1 in 1.16, 1 in 1.00)
1 in 28.72
(1 in 4.76, 1 in
7.02E+03)
1000 g
Mammal
Acute
RQ
2.83
3.48
0.39
0.09
2.38
1.09
1.34
0.15
0.03
2.90
1.33
1.63
0.18
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.02
(1 in 1.22, 1 in 1.00)
1 in 1.01
(1 in 1.16, 1 in 1.00)
1 in 31 .64
(1 in 4.89, 1 in
9.86E+03)

1 in 1.05
(1 in 1.29, 1 in 1.00)
1 in 1.76
(1 in 1.88,1 in 1.57)
1 in 1 .40
(1 in 1.67, 1 in 1.14)
1 in 1.01E+04
(1 in 20.37, 1 in
2.02E+13)

1 in 1.02
(1 in 1.22, 1 in 1.00)
1 in 1.41
(1 in 1.67, 1 in 1.16)
1 in 1.21
(1 in 1.51, 1 in 1.03)
1 in 2.40E+03
(1 in 14.54, 1 in
8.48E+10)
Insect
Acute
RQ

1168
130



450
50



547
61
Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl

1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.01, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.01, 1 in
1.00)
154

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category

Potato
Sweet potato
Dry beans
(except
Lima), peas
& pepper
Dietary Category
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
20 g
Bird
Acute
RQ
0.16
26.71
12.24
15.02
1.67
0.37
15.01
6.88
8.45
0.94
0.21
26.71
12.24
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.88E+11
(1 in 501 .26, 1 in
2.34E+26)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.03
(1 in 1 .27, 1 in 1 .00)
1 in 7.97E+03
(1 in 16.53, 1 in
2.55E+08)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 2.46
(1 in 2. 17, 1 in 2.81)
1 in 2.79E+08
(1 in 140.36, 1 in
2.73E+19)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
15 g
Mammal
Acute RQ
0.09
14.79
6.78
8.32
0.92
0.21
8.31
3.81
4.68
0.52
0.12
14.79
6.78
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.00E+06
(1 in 57.77, 1 in
1.02E+21)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1 in 1.00
(1 in 1 .03, 1 in 1 .00)
1 in 2.28
(1 in 2. 12, 1 in 2.64)
1 in 1.01E+03
(1 in 11.82, 1 in
3.25E+09)
1 in 1.00
(1 in 1 .03, 1 in 1 .00)
1 in 1.00
(1 in 1.14, 1 in 1.00)
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 9.96
(1 in 3.51, 1 in
190.27)
1 in8.14E+04
(1 in 32.90, 1 in
6.17E+16)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.00
(1 in 1 .05, 1 in 1 .00)
1000 g
Mammal
Acute
RQ
0.04
6.77
3.10
3.81
0.42
0.09
3.81
1.74
2.14
0.24
0.05
6.77
3.10
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl

1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 1.01
(1 in 1.19, 1 in 1.00)
1 in 1.00
(1 in 1.14, 1 in 1.00)
1 in 21 .54
(1 in 4.39, 1 in
2.57E+03)

1 in 1.00
(1 in 1.14, 1 in 1.00)
1 in 1.16
(1 in 1.46, 1 in 1.02)
1 in 1.07
(1 in 1.34, 1 in 1.00)
1 in 398.57
(1 in 9.42, 1 in
9.99E+07)

1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 1.01
(1 in 1.1 9, 1 in 1.00)
Insect
Acute
RQ



1278
142



719
80



Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



155

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category

Carrot
Celery
Dietary Category
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
20 g
Bird
Acute
RQ
15.02
1.67
0.37
18.80
8.62
10.58
1.18
0.26
18.80
8.62
10.58
1.18
0.26
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.03
(1 in 1 .27, 1 in 1 .00)
1 in 7.97E+03
(1 in 16.53, 1 in
2.55E+08)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.38
(1 in 1.67, 1 in 1.21)
1 in 2.79E+06
(1 in 55.89, 1 in
3.62E+14)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.38
(1 in 1.67, 1 in 1.21)
1 in 2.79E+06
(1 in 55.89, 1 in
3.62E+14)
15 g
Mammal
Acute RQ
8.32
0.92
0.21
10.41
4.77
5.86
0.65
0.14
10.41
4.77
5.86
0.65
0.14
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.00
(1 in 1 .03, 1 in 1 .00)
1 in 2.28
(1 in 2. 12, 1 in 2.64)
1 in 1.01E+03
(1 in 11.82, 1 in
3.25E+09)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.00
(1 in 1 .07, 1 in 1 .00)
1 in 4.99
(1 in 2.82, 1 in
21.50)
1 in 1.27E+04
(1 in 21 .50, 1 in
4.90E+13)
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.00
(1 in 1 .07, 1 in 1 .00)
1 in 4.99
(1 in 2.82, 1 in
21.50)
1 in 1.27E+04
(1 in 21 .50, 1 in
4.90E+13)
1000 g
Mammal
Acute
RQ
3.81
0.42
0.09
4.77
2.18
2.68
0.30
0.07
4.77
2.18
2.68
0.30
0.07
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.14, 1 in 1.00)
1 in 21 .54
(1 in 4.39, 1 in
2.57E+03)

1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.07
(1 in 1.33, 1 in 1.00)
1 in 1.03
(1 in 1.24, 1 in 1.00)
1 in 111.34
(1 in 6.83, 1 in
9.03E+05)

1 in 1.00
(1 in 1.10, 1 in 1.00)
1 in 1.07
(1 in 1.33, 1 in 1.00)
1 in 1.03
(1 in 1.24, 1 in 1.00)
1 in 11 1.34
(1 in 6.83, 1 in
9.03E+05)

Insect
Acute
RQ
1278
142



900
100



900
100

Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl
1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)

156

-------
Table 5.12 Summary of the chance of individual acute effects to terrestrial animals based on acute RQs, the acute listed species LOC,
acute toxicity data, and probit slope response relationships.
Use
Category
Strawberry
Tomato
Dietary Category
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
Short Grass
Tall Grass
Broadleaf Plants/ Small Insects
Fruits/Pods/ Seeds/ Large Insects
Granivore
20 g
Bird
Acute
RQ
20.20
9.26
11.36
1.26
0.28
24.39
11.18
13.72
1.52
0.34
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1 .00, 1 in 1 .00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.24
(1 in 1.56, 1 in 1.10)
1 in 7.40E+05
(1 in 42.65, 1 in
1.44E+13)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.06
(1 in 1.34, 1 in 1.01)
1 in3.10E+04
(1 in 22.08, 1 in
6.70E+09)
15 g
Mammal
Acute RQ
11.19
5.13
6.29
0.70
0.16
13.51
6.19
7.60
0.84
0.19
Chance of
Individual Effect at
RQ or LOC Based
on Probit Slope &
Slope's 95% Cl
1 in 1.00
(1 in 1 .02, 1 in 1 .00)
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 1.00
(1 in 1 .06, 1 in 1 .00)
1 in 4.1 3
(1 in 2.65, 1 in
12.37)
1 in7.31E+03
(1 in 18.90, 1 in
5.91 E+12)
1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.00
(1 in 1 .06, 1 in 1 .00)
1 in 1.00
(1 in 1 .04, 1 in 1 .00)
1 in 2.70
(1 in 2.26, 1 in 3.94)
1 in 1.86E+03
(1 in 13.69, 1 in
3.27E+10)
1000 g
Mammal
Acute
RQ
5.12
2.35
2.88
0.32
0.07
6.18
2.83
3.48
0.39
0.09
Chance of Individual
Effect at RQ or LOC
Based on Probit
Slope & Slope's 95%
Cl
1 in 1.00
(1 in 1.08, 1 in 1.00)
1 in 1.05
(1 in 1.30, 1 in 1.00)
1 in 1.02
(1 in 1.22, 1 in 1.00)
1 in 76.94
(1 in 6.20, 1 in
2.36E+05)

1 in 1.00
(1 in 1.06, 1 in 1.00)
1 in 1.02
(1 in 1.22, 1 in 1.00)
1 in 1.01
(1 in 1.16, 1 in 1.00)
1 in 31 .64
(1 in 4.89, 1 in
9.86E+03)

Insect
Acute
RQ


967
107



1168
130

Chance of
Individual Effect
at RQ or LOC
Based on Probit
Slope & Slope's
95% Cl


1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)



1 in 1.00
(1 in 1.00, 1 in
1.00)
1 in 1.00
(1 in 1.00, 1 in
1.00)

(1)
   Bolded RQs exceed the Agency's acute listed species LOC (0.1 for terrestrial animals). When acute RQs exceed the Agency's listed species LOC, the chance of individual effects was calculated at the RQ and the LOC,
whereas if there was no exceedance, then the chance of individual effects was calculated only at the LOC.
   Although an LD50 has been established for all terrestrial taxonomic groups, information is unavailable to estimate probit slopes from the mammal studies from which the endpoints were derived. Therefore, a default slope
of 4.5 with an assumed 95% confidence interval of 2 and 9 has been assumed as per original Agency assumptions of typical slope cited in Urban and Cook (1986). The probit slope for the bird acute LD50 was 8.5 with 95%
confidence limits of 3.6 and 13.4. The probit slope for the insect was 1.91 with 95% confidence intervals of 1.40 and 2.42.
                                                                                                  157

-------
Table 5.13 Summary of the chance of individual acute effects to herpetofauna based on acute RQs, the acute listed species LOC, acute
toxicity data, and probit slope response relationships
Use Category
N/A
Almonds, hazelnut &
walnut
Citrus
Broccoli, cabbage,
Chinese cabbage,
cauliflower, kohlrabi
Kale, CollardsS
Mustard Green
Dietary Category
N/A
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
1.4 g
Herp.
Acute
RQ
LOC =
0.1
1.08
0.12



1.34
0.15



0.70
0.08



0.40
0.04


Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 1.05E+17
(1 in 6290, 1 in 3.31 E+40)
1 in 1 .65
(1 in 1.83, 1 in 1.51)
1 in4.56E+14
(1 in 2.24E+03, 1 in 4.94E+34)



1 in 1.16
(1 in 1.47, 1 in 1.04)
1 in8.99E+11
(1 in 678.21 , 1 in 1 .09E+28)



1 in 10.90
(1 in 3.49, 1 in 55.57)




1 in 2.50E+03
(1 in 12.86, 1 in1.60E+07)



35 g
Herp.
Acute
RQ
LOC =
0.1
0.65
0.07
11.64
0.73
0.02
0.81
0.09
14.55
0.91
0.03
0.42
0.05
7.55
0.47
0.01
0.24
0.03
4.36
0.27
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 1.05E+17
(1 in 6290, 1 in 3.31 E+40)
1 in 18.60
(1 in 4.04, 1 in 179.07)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 8.34
(1 in 3.23, 1 in 31 .32)

1 in 4.63
(1 in 2.71, 1 in 9.28)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 2.76
(1 in 2.27, 1 in 3.45)

1 in 1.49E+03
(1 in 11.47, 1 in4.68E+06)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 360.53
(1 in 8.33, 1 in1.63E+05)

1 in 1.18E+07
(1 in 74.77, 1 in1.21E+16)

1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 1.24E+06
(1 in 47.37, 1 in5.00E+13)
238 g
Herp.
Acute
RQ
LOC =
0.1
0.32
0.04
1.37
0.09
0.01
0.40
0.04
1.71
0.11
0.01
0.21
0.02
0.89
0.06
0.01
0.12
0.01
0.51
0.03
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 1.05E+17
(1 in 6290, 1 in 3.31 E+40)
1 in 7.49E+04
(1 in 26.57, 1 in5.61E+10)

1 in 1.14
(1 in 1 .45, 1 in 1 .04)


1 in 2.72E+03
(1 in 13.09, 1 in1.95E+07)

1 in 1 .02
(1 in 1.25, 1 in 1.00)
1 in 1.30E+16
(1 in 4.23E+03, 1 in 1 .90E+38)

1 in 2.98E+08
(1 in 142.18, 1 in3.21E+19)

1 in 3.03
(1 in 2.35, 1 in 4.10)


1 in3.81E+14
(1 in2.16E+03, 1 in3.17E+34)

1 in 145.19
(1 in 6.74, 1 in 1 .94E+04)

158

-------
Table 5.13 Summary of the chance of individual acute effects to herpetofauna based on acute RQs, the acute listed species LOC, acute
toxicity data, and probit slope response relationships.
Use Category

Sweet corn
Cotton (ground)
Cotton (Areal)
Apples
Apricot, nectarine,
peach, cherry, pear,
Dietary Category
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
1.4 g
Herp.
Acute
RQ

0.81
0.09



0.54
0.06



0.52
0.06



1.34
0.15



1.34
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl

1 in 4.68
(1 in 2.72, 1 in 9.50)




1 in 91 .07
(1 in 6.03, 1 in 6.58E+03)




1 in 119.41
(1 in 6.43, 1 in 1 .23E+04)




1 in 1.16
(1 in 1.47, 1 in 1.04)
1 in8.99E+11
(1 in 678.21 , 1 in 1 .09E+28)



1 in 1.16
(1 in 1.47, 1 in 1.04)
35 g
Herp.
Acute
RQ
0.01
0.48
0.05
8.73
0.55
0.02
0.32
0.04
5.82
0.36
0.01
0.31
0.03
5.66
0.35
0.01
0.81
0.09
14.55
0.91
0.03
0.81
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl

1 in 264.86
(1 in 7.76, 1 in 7.89E+04)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 79.1 9
(1 in 5.83, 1 in 4.76E+03)

1 in 6.52E+04
(1 in 25.81,1 in4.01E+10)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in1.06E+04
(1 in 17.58,1 in5.08E+08)

1 in1.02E+05
(1 in 28.34, 1 in 1.18E+11)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.60E+04
(1 in 19.18, 1 in 1.35E+09)

1 in 4.63
(1 in 2.71, 1 in 9.28)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 2.76
(1 in 2.27, 1 in 3.45)

1 in 4.63
(1 in 2.71, 1 in 9.28)
238 g
Herp.
Acute
RQ
0.004
0.24
0.03
1.03
0.06
0.01
0.16
0.02
0.68
0.04
0.01
0.16
0.02
0.67
0.04
0.01
0.40
0.04
1.71
0.11
0.01
0.40
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl

1 in 1.40E+07
(1 in 77.40, 1 in 1.85E+16)

1 in 1 .86
(1 in 1.94, 1 in 1.79)


1 in 1 .44E+1 1
(1 in 475.86, 1 in1.21E+26)

1 in 12.41
(1 in 3.62, 1 in 73.56)


1 in2.90E+11
(1 in 545.19, 1 in6.79E+26)

1 in 15.03
(1 in 3.81, 1 in 111.89)


1 in 2.72E+03
(1 in 13.09, 1 in1.95E+07)

1 in 1 .02
(1 in 1.25, 1 in 1.00)
1 in 1.30E+16
(1 in 4.23E+03, 1 in 1 .90E+38)

1 in 2.72E+03
(1 in 13.09, 1 in1.95E+07)
159

-------
Table 5.13 Summary of the chance of individual acute effects to herpetofauna based on acute RQs, the acute listed species LOC, acute
toxicity data, and probit slope response relationships.
Use Category
plum & prune
Lettuce & Brussels
sprouts
Cucumber, melons,
pumpkin & squash
Eggplant
Ornamentals &
shade trees
Dietary Category
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
1.4 g
Herp.
Acute
RQ
0.15



0.76
0.08



0.70
0.08



0.27
0.03



0.33
0.04

Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in8.99E+11
(1 in 678.21 , 1 in 1 .09E+28)



1 in 6.26
(1 in 2.97, 1 in 17.15)




1 in 10.90
(1 in 3.49, 1 in 55.57)




1 in 1.62E+06
(1 in 50.03, 1 in9.61E+13)




1 in 5.56E+04
(1 in 24.96, 1 in2.73E+10)


35 g
Herp.
Acute
RQ
0.09
14.55
0.91
0.03
0.46
0.05
8.26
0.52
0.02
0.42
0.05
7.55
0.47
0.01
0.16
0.02
2.91
0.18
0.01
0.20
0.02
3.53
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 2.76
(1 in 2.27, 1 in 3.45)

1 in 491 .37
(1 in 8.94, 1 in 3.37E+05)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 135.66
(1 in 6.63, 1 in1.66E+04)

1 in 1.49E+03
(1 in 11.47,1 in4.68E+06)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 360.53 (1 in 8.33, 1 in
1.63E+05)

1 in 1.16E+11
(1 in 456. 17,1 in7.08E+25)

1 in 1.00
(1 in 1.05, 1 in 1.00)
1 in 6.42E+09
(1 in 259.99, 1 in5.91E+22)

1 in1.07E+09
(1 in 183.02,1 in7.36E+20)

1 in 1.00
(1 in 1.02, 1 in 1.00)
238 g
Herp.
Acute
RQ
0.04
1.71
0.11
0.01
0.23
0.03
0.97
0.06
0.01
0.21
0.02
0.89
0.06
0.01
0.08
0.01
0.34
0.02
0.003
0.10
0.01
0.42
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl

1 in 1 .02
(1 in 1.25, 1 in 1.00)
1 in 1.30E+16
(1 in 4.23E+03, 1 in 1 .90E+38)

1 in 4.27E+07
(1 in 96.71, 1 in2.80E+17)

1 in 2. 18
(1 in 2.07, 1 in 2.30)


1 in 2.98E+08
(1 in 142.18, 1 in3.21E+19)

1 in 3.03
(1 in 2.35, 1 in 4.10)




1 in 2.67E+04
(1 in 21 .39, 1 in4.65E+09)




1 in 1.68E+03
(1 in 11. 78, 1 in6.23E+06)
160

-------
Table 5.13 Summary of the chance of individual acute effects to herpetofauna based on acute RQs, the acute listed species LOC, acute
toxicity data, and probit slope response relationships.
Use Category

Potato
Sweet potato
Dry beans (except
Lima), peas &
pepper
Carrot
Dietary Category
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
1.4 g
Herp.
Acute
RQ


0.76
0.08



0.43
0.05



0.76
0.08



0.54
0.06



Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl


1 in 6.26
(1 in 2.97, 1 in 17.15)




1 in1.11E+03
(1 in 10.74, 1 in2.32E+06)




1 in 6.26
(1 in 2.97, 1 in 17.15)




1 in 91 .07
(1 in 6.03, 1 in 6.58E+03)




35 g
Herp.
Acute
RQ
0.22
0.01
0.46
0.05
8.26
0.52
0.02
0.26
0.03
4.65
0.29
0.01
0.46
0.05
8.26
0.52
0.02
0.32
0.04
5.82
0.36
0.01
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 8.08E+07
(1 in 109.79, 1 in 1.32E+18)

1 in 491 .37
(1 in 8.94, 1 in 3.37E+05)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 135.66
(1 in 6.63, 1 in1.66E+04)

1 in 3.46E+06
(1 in 58.37, 1 in6.10E+14)

1 in 1.00
(1 in 1.01, 1 in 1.00)
1 in 3.99E+05
(1 in 37.58, 1 in3.22E+12)

1 in 491 .37
(1 in 8.94, 1 in 3.37E+05)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 135.66
(1 in 6.63, 1 in1.66E+04)

1 in 6.52E+04
(1 in 25.81, 1 in4.01E+10)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.06E+04(1 in 17.58, 1 in
5.08E+08)

238 g
Herp.
Acute
RQ
0.03
0.003
0.23
0.03
0.97
0.06
0.01
0.13
0.01
0.55
0.03
0.004
0.23
0.03
0.97
0.06
0.01
0.16
0.02
0.68
0.04
0.01
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl


1 in 4.27E+07
(1 in 96.71, 1 in2.80E+17)

1 in 2. 18
(1 in 2.07, 1 in 2.30)


1 in6.20E+13
(1 in 1.53E+03, 1 in 3.63E+32)

1 in 77.93
(1 in 5.80, 1 in 4.59E+03)


1 in 4.27E+07
(1 in 96.71, 1 in2.80E+17)

1 in 2. 18
(1 in 2.07, 1 in 2.30)


1 in 1 .44E+1 1
(1 in 475.86, 1 in 1.21E+26)

1 in 12.41
(1 in 3.62, 1 in 73.56)


161

-------
Table 5.13 Summary of the chance of individual acute effects to herpetofauna based on acute RQs, the acute listed species LOC, acute
toxicity data, and probit slope response relationships.
Use Category
Celery
Strawberry
Tomato
Dietary Category
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
1.4 g
Herp.
Acute
RQ
0.54
0.06



0.58
0.06



0.70
0.08



Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 91 .07
(1 in 6.03, 1 in 6.58E+03)




1 in 46.81
(1 in 5. 12, 1 in 1.43E+03)




1 in 10.90
(1 in 3.49, 1 in 55.57)




35 g
Herp.
Acute
RQ
0.32
0.04
5.82
0.36
0.01
0.35
0.04
6.25
0.39
0.01
0.42
0.05
7.55
0.47
0.01
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 6.52E+04
(1 in 25.81, 1 in4.01E+10)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 1.06E+04
(1 in 17.58, 1 in5.08E+08)

1 in2.11E+04
(1 in 20.35, 1 in 2.65E+09)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 3.84E+03
(1 in 14. 11,1 in4.44E+07)

1 in 1.49E+03
(1 in 11.47, 1 in4.68E+06)

1 in 1.00
(1 in 1.00, 1 in 1.00)
1 in 360.53
(1 in 8.33, 1 in 1.63E+05)

238 g
Herp.
Acute
RQ
0.16
0.02
0.68
0.04
0.01
0.17
0.02
0.74
0.05
0.01
0.21
0.02
0.89
0.06
0.01
Chance of Individual Effect at
RQ or LOC Based on Probit
Slope & Slope's 95% Cl
1 in 1 .44E+1 1
(1 in 475.86, 1 in 1.21E+26)

1 in 12.41
(1 in 3.62, 1 in 73.56)


1 in2.39E+10
(1 in 336.03, 1 in 1 .49E+24)

1 in 7.82
(1 in 3.1 7, 1 in 27.32)


1 in 2.98E+08
(1 in 142.18, 1 in3.21E+19)

1 in 3.03
(1 in 2.35, 1 in 4.10)


Bolded RQs exceed the Agency's interim herpetofauna acute listed species LOC (0.1 ). When acute RQs exceed the Agency's listed species LOC, the chance of individual effects was calculated at the RQ and the LOC,
whereas if there was no exceedance, then the chance of individual effects was calculated only at the LOC.
(2)
Because birds are a surrogate for reptiles and terrestrial-phase amphibian, the probit slopes for birds was used. The probit slope for the bird acute LD50 was 8.5 with 95% confidence limits of 3.6 and 13.4. The probit
slope for the insect was 1.91 with 95% confidence intervals of 1 .40 and 2.42.
The 1 .4 g size class wase used to represent small terrestrial phase CRLF; the 37 g size class was used to represent adult CTS and the 238 g size class was used to represent adult SFGS.
162

-------
5.2.1 California Red-Legged Frog

5.2.1.1 Direct Effects

The characterization of the likelihood of direct effects on the CRLF from registered uses of
endosulfan in California considers multiple lines of evidence. These lines of evidence include:

    •   The degree of temporal and spatial overlap between known CRLF habitat and
       endosulfan use (Appendix E).
    •   Estimated risks from exposure modeling in aquatic and terrestrial ecosystems
    •   The likelihood of impacts to individual CRLF using probit dose-response modeling
    •   Estimated risks using monitoring data from aquatic and terrestrial ecosystems
    •   Use of refined exposure and risk estimation methods (e.g., T-HERPS modeling for
       herpetofauna, earthworm fugacity modeling for contaminated terrestrial invertebrate
       prey, KABAM modeling for contaminated aquatic prey)
    •   Information from reported ecological incidents associated with endosulfan use.

In addition to these lines of evidence, the robustness of the risk estimation described in
Section 5.1 is also considered by characterizing the major uncertainties and sensitivity of risk
projections to underlying assumptions specific to the CRLF assessment. A discussion of each
of these lines of evidence is presented in the following sections.

Direct Effects: Aquatic-Phase CRLF

The aquatic-phase considers life stages of the CRLF that are obligatory aquatic organisms,
including eggs and larvae.  It also considers submerged terrestrial-phase juveniles and adults,
which spend a portion of their time in water bodies that may receive runoff and spray drift
containing  endosulfan.

       Temporal and Spatial Overlap

Endosulfan is currently registered for a wide variety of agricultural crops (e.g., cotton, lettuce,
brussels sprouts, cucurbits, dry beans, cole crops), orchards (e.g., citrus, almonds, walnuts),
nursery (ornamentals and shade trees) and only one non-agricultural use (ear tags for cattle).
These agriculture uses span a large variety of use  sites and geographical regions throughout
the entire state of California. Historically, however, the use of endosulfan has changed
substantially over the last decade, with overall use dropping from approximately 500,000 Ibs
a.i. in 1994 to about 90,000 Ibs a.i. in 2006 (Figure 2.3). Therefore, data from the most
recent reporting years is  used here for evaluating the degree of temporal and spatial overlap
between the CRLF and endosulfan application in California.

Based on data from the most recent reporting year (2006), the registered endosulfan uses
allow for the potential for nearly year-round use, with the exception of little or no use during
November, December and a portion of April (see Figure 5.1). Based on the likely occurrence
of aquatic life stages of the CRLF (eggs, tadpoles and young juveniles during January-
September; Figure 5.2) in combination with the CDPR PUR data from 2006 (Figure 5.1), it
                                        163

-------
is reasonably expected that the registered use of endosulfan could coincide with each of the
critical life-stages of the aquatic-phase CRLF.
           Time distribution for the total Lbs of endosulfan a.i applied during the year 2006
     600
       0
       9-Jan   9-Feb   9-Mar   9-Apr   9-May   9-Jun   9-Jul    9-Aug   9-Sep   9-Oct   9-Nov   9-Dec
                                            Date of Application
Figure 5.1 Temporal distribution of endosulfan application in California for 2006 (Source: CDPRPUR))

.........

........1........
Young Juveniles
Tadpoles (except those that over-
winter)
Breeding/Egg Masses
J
F
M
A

M


J


J


A


S



o



N



D

Figure 5.2 CRLF Reproductive Events by Month; Adults and juveniles can be present all year.
Table 5.7 summarizes information on the extent that endosulfan use and CRLF could
potentially coincide spatially based on reported use information from 2005-2006 (CDPR
PUR) and information on the occurrence of the CRLF or its critical habitat (Appendix E).
Since pesticide use information is only available at the county level, the degree that the CRLF
or its critical habitat actually coincides in the same or similar locations cannot be precisely
determined.  Nevertheless, there does appear to be a significant potential for overlap between
the use of endosulfan and the presence of the CRLF or its critical habitat at least at the county
level of resolution (Appendix E). Specifically, the CRLF is present in counties with the first,
                                        164

-------
fourth, fifth, tenth and eleventh highest reported usage from 2005-2006 (Table 5.7). The
CRLF is also present in most other counties with lower endosulfan reported use.

       Modeling—Based EECs

Acute RQs based on modeled EECs for freshwater fish (used as a surrogate for the aquatic
phase CRLF) range from 7.2 to 58.8, thus exceeding the Agency's LOG (0.05) in all 20 crop
scenarios modeled (Table 5.1).  Similarly, chronic RQs range from 5.7 to 80.6 and exceed
the Agency's LOG (1.0) in all 20 crop scenarios modeled. These exceedences of the
Agency's acute and chronic LOCs suggest the potential for direct effects on the CRLF.

As previously mentioned, EFED also estimates the chance of an individual event (i.e.,
mortality or immobilization) corresponding to the listed species acute LOCs and/or RQs
should exposure at the EEC actually occur for a species with sensitivity to endosulfan on par
with the acute toxicity endpoint selected for RQ calculation. To do this, the Agency uses the
EFED spreadsheet IEC (version 1.1.xls) and the slope of the dose response relationship
available from the toxicity  study used to establish the acute toxicity measures of effect for
each taxonomic group that  is relevant to this assessment.  The individual effects probability
associated with the listed species acute LOCs and/or RQs is based on the mean estimate of the
slope and an assumption of a probit dose response relationship. In addition to a single effects
probability estimate based on the mean, available information on the 95% confidence interval
of the slope is also used to estimate upper and lower estimates  of the effects probability to
account for variance in the  slope, if available.  If an LDso or LCso has been  established for a
particular taxonomic group, but information is unavailable to estimate  a slope from a study, a
default slope assumption of 4.5 (with upper and lower bounds of 2 and 9, respectively) is used
as per original  Agency assumptions of typical slope cited in Urban and Cook (1986).

Based on a default slope of 4.5  (with upper and lower bounds of 2 and 9), the LCso of the
most acutely sensitive freshwater fish (common carp; LCso = 0.1 jig a.i./L;  Sunderam et al.,
1992; ECOTOX ref # 5850) and the acute listed species LOC  of 0.05, the chance of an
individual mortality for aquatic-phase CRLF is ~ 1 in 4.18 xlO8 (with lower and upper bounds
of ~ 1 in 2.16 x!02to ~ 1 in 1.75xl031; Table 5.11). Based on an analysis  of the likelihood of
individual mortality using the highest acute RQ value for freshwater fish (RQ=58.8 for
Lettuce and Brussels  Sprouts) and a default slope of 4.5 (with upper and lower bounds of 2
and 9), the likelihood of individual mortality is ~ 1 in 1 (with lower and upper bounds of- 1
in 1.00 to 1 in  1.00). At the lowest RQ value (i.e., RQ = 7.2 for Eggplant), the likelihood of
individual mortality is - 1 in 1.00(with lower and upper bounds of - 1 in 1.00 to - 1 in 1.05).

Based on: (1) the high probability of an individual mortality occurrence using the highest
acute RQ (-1 in 1), (2) acute RQs that are above the listed species LOC (0.05) for all 20
modeled scenarios; (3) chronic  RQs that exceed the Agency's LOC for all 20 modeled
scenarios (1.0). and (4) the potential for spatial and temporal overlap of endosulfan usage with
each of the critical life-stages of the aquatic-phase CRLF, it appears that endosulfan is likely
to cause direct adverse effects to aquatic-phase CRLFs should  exposure at the predicted EECs
actually occur for a CRLF and assuming its sensitivity to endosulfan on par with the acute
freshwater fish LCso (common carp; LCso = 0.1  jig a.i./L) selected for RQ calculation.
                                       165

-------
       Monitoring-based EECs

Available data on the occurrence of endosulfan (and its primary degradate, endosulfan sulfate)
in surface water were evaluated as another line of evidence for potential risks to aquatic-phase
CRLF.  As described in Section 3.2.4.1, surface water data were compiled from three sources:
CDPR, EPA's STORE! database, and the USGS NAWQA. As an approximate screen of
these data, maximum detected values for total endosulfan were used to calculate acute and
chronic RQs for freshwater fish (used as a surrogate for aquatic-phase CRLF; Table 5.14).
As described in Section 3.2.4.1, available surface water monitoring data indicate a substantial
reduction in endosulfan concentrations after 2001 as compared to the early to mid 1990's.
Based on this comparison, acute and chronic RQs based on maximum reported concentrations
range from 9.5 to 41.3, respectively, in the 1991-1996 sampling period but drop to 0.05-0.08
and 0.22-0.35, respectively,  during the 2001-2008 sampling period. Acute RQs based on
maximum detected concentrations during the 2001 time periods still exceed the Agency's
LOG of 0.05 for listed species, but are much lower than in the 1990s.  Chronic RQs based on
maximum detected during the 2001-2008 sampling  period do not exceed the Agency's
chronic listed LOG of 1.0.

Several caveats should be considered when interpreting the monitoring data summarized in
Table 5.14 and in Section 3.2.1.4. First, these data do not represent targeted monitoring data
that are designed to capture peak concentrations of pesticides in surface waters. Therefore,
these data likely underestimate peak concentrations of endosulfan and are likely not directly
comparable to modeling-based peak EECs.  Second, the NAWQA data included limits of
detection that exceeded the concentration corresponding to the Agency listed acute LOG of
0.05 (i.e., 0.005 ug/L) and therefore are of limited value for estimating the potential for direct
effects to the CRLF. Despite these limitations, the available monitoring data during the most
recent sampling period for endosulfan in California surface waters (2001-2008) does suggest
a potential for acute risk to aquatic-phase CRLF assuming their acute sensitivity is
comparable to the most sensitive freshwater fish tested.

Table 5.14 Comparison of maximum detected concentrations of total endosulfan in surface waters of California
to Agency LOG for freshwater fish

Data Source(1)
CDPR

STORET
NAQWA

Sampling
Period
1991-96
2001-06
2001-08
1995; 01-07
Total
Number of
Samples
-2000
580
304
204
Maximum
Value Detected
(ug/L)
0.95
0.005
0.008
ND(2)

FW Fish
Acute RQ(3)
9.5
0.05
0.08

FW Fish
Chronic RQ(4)
41.3
0.22
0.35
  See Section 3.2.4.1 for a description of data sources.
(2) Limit of detection typically ranged from 0.01 to 0.02 ug/L for a- and (3-endosulfan and/or endosulfan sulfate,
which exceeded concentrations corresponding to the acute LOG (0.005 ug/L).
(3) RQ based on LC50 for freshwater fish (0.1 ug/L; Sunderam et al, 1992; ECOTOX 5850), which are used as a
surrogate for aquatic-phase amphibians. Bold values indicate equivalence or exceedence of Agency's acute listed
LOCofO.05.
(4) RQ is based on a NOAEC for freshwater fish of 0.023 ug/L (estimated using an acute-chronic ratio of 4.3 for
fathead minnow; see Table 4.1). Bold values indicate exceedence of Agency's chronic listed LOG of 1.0.
                                         166

-------
       Species Sensitivity Analysis

As indicate above, acute toxicity data for the most sensitive species of freshwater fish
(common carp, LCso = 0.1 jig a.i./L) is being used to represent the sensitivity of the aquatic
phase CRLF due to the absence of acceptable toxicity data for amphibians. To evaluate the
potential uncertainty of the risk estimation with respect to using the most sensitive fish
species, the sensitivity of the common carp is put into context of available data for all tested
freshwater fish (Figure 5.3). In this figure, the geometric mean of LCso values for  each
species of freshwater fish from both studies submitted to the Agency and those available in
ECOTOX are provided in the form of a species sensitivity distribution.  Geometric mean of
LC50 values for each species are indicated by a solid circle . Notably, the ECOTOX data that
were greater than the initial acute toxicity benchmark of 0.83 jig/L were not evaluated beyond
the initial OPP data screen described in Section 4. The thick blue line designates the LCso
value used in the RQ calculation (0.1 |ig/L).  The vertical blue lines represent the range of
predicted peak EECs from the 20 crop exposure scenarios modeled (i.e., 0.72 ug/L  for
Eggplant and 5.88 ug/L for Lettuce and Brussels sprouts scenarios).  The vertical red lines
indicate the range of the peak EECs (0.72 and 5.88 |ig/L) divided by the acute listed species
LOCofO.05.
                                 Freshwater Fish
1.00


0.90

0.80
•* 0.70
TO
_ 0.60
re
c
.2 0.50
0
2 0.40
0.30

o
LU
LU
to
_O







0 20 -il benchmark
0.10 J| •*
ii :
n on H

O
LU
LU
to
0
D)
,•*
9

g
•
,f
•
f








£
•











Q
A
•
f

8
D
LU
LU

to
0






t





O
0
o
LU
LU

W
0
D)






U .UU i i 	 i i 	 i i 	 i i 	
0.10 1.00 10.00 100.00 100
LC50 (ug/L)
Figure 5.3 Species sensitivity distribution of the acute toxicity of endosulfan to freshwater fish

Based on the range of modeled peak EECs of 0.72 to 5.88 ug/L determined in Section 3 (see
Table 5.1), the proportion of species affected would range from approximately the 15th to the
70th percentiles of the observed species sensitivity distribution blue lines on Figure 5.3).
                                         167

-------
Taking into account the acute listed LOG of 0.05, the assumed acute sensitivity of freshwater
fish would have to be near above the least sensitive species tested in order for the modeled
EECs (0.72 - 5.88 ug/L) to not exceed the agency LOG. Therefore, this analysis indicates
that within the context of the sensitivity of freshwater fish, RQ exceedences of the LOG of
0.05 are not sensitive to the selection of the most acutely sensitive fish species tested (i.e.,
LOG exceedences would occur even the  more acutely insensitive species was selected for RQ
calculation).

While the previous discussion address sensitivity differences within the context of freshwater
fish, it does not address potential differences in sensitivity between  freshwater fish and
aquatic-phase amphibians. Although no data were found in the open literature that were
considered acceptable for quantitative use in this risk assessment, data were found that can be
used qualitatively for describing the general sensitivity of aquatic-phase amphibians. The
lowest LC50 in the open literature for endosulfan using amphibians was 0.16 jig a.i./L for
tadpoles ofMicorohyla ornate (see Section 4.1.5), based on unmeasured values. This species
of amphibian is much smaller than CRLF (22-23 mm snout-vent length for adults—Geng  et
al. 2005). Adults of CRLF range from  85-138 mm SVL (see Attachment 3). It is reasonable
to assume that smaller animals would be more susceptible to the same toxicant exposure than
larger ones. However, even if the CRLF  has the same acute sensitivity asM orgata, the
freshwater fish acute benchmark (LC50 = 0.1  jig a.i./L) should still  be protective of the
aquatic-phase of CRLF. The lowest endosulfan effect concentration for longer exposure
durations (Ranapipiens—for 7 weeks) suggested a 49 da LC50 of 0.2 jig a.i./L (estimated
from Shenoy et al., in press). Again, if we assume that the CRLF had this same sensitivity, the
freshwater chronic benchmark (0.023 jig a.i./L) would likely be protective. At an LC50 of 0.2
jig a.i./L and the default probit slope of 4.5, the expected mortality at 0.023 jig a.i./L would be
close to 0%.

Consumption of Contaminated Aquatic Prey

Freshwater fish and aquatic invertebrates are a documented component of the diet of the
aquatic-phase CRLF. For persistent organic chemicals with log KOW values that exceed 4,
evidence from empirical and modeling studies indicates that biomagnification in aquatic food
webs can significantly increase exposure to organisms feeding on top trophic level aquatic
prey (USEPA, 2000; 2003; 2008; Fisk et al., 1998). Log Kow values for a-endosulfan have
been reported from 3.55 to 4.74 (McConnell et al 1998; MRID 414215-01, respectively).  Log
KOW values for p-endosulfan have been reported from 3.62 to 4.79 (McConnell et al 1998;
MRID 414215-01, respectively). A log K0w value of 3.71 has been reported for endosulfan
sulfate (Table 2.1).

Considering that log K0w values for the parent endosulfan isomers and endosulfan sulfate
either approach or exceed a log Kow of 4, and that endosulfan can persist for relatively long
periods of time in aquatic ecosystems,  the potential for direct effects on aquatic-phase CRLF
through consumption of contaminated  aquatic prey was investigated.  The KAB AM model
(Kow (based) Aquatic BioAccumulation Model) version 1.0 was used to evaluate the potential
exposure and risk of direct effects to aquatic-phase CRLF via bioaccumulation and
biomagnification in aquatic food webs. KAB AM is used to estimate potential
                                       168

-------
bioaccumulation of hydrophobic organic pesticides in freshwater aquatic ecosystems and risks
to mammals and birds consuming aquatic organisms which have bioaccumulated these
pesticides. The bioaccumulation portion of KABAM is based upon work by Arnot and Gobas
(2004) who parameterized a bioaccumulation model based on PCBs and some pesticides (e.g.,
lindane, DDT) in freshwater aquatic ecosystems. KABAM relies on a chemical's octanol-
water partition coefficient (K0w) to estimate uptake and elimination constants through
respiration and diet of organisms in different trophic levels. Pesticide tissue residues are
calculated for different levels of an aquatic food web. The model then uses pesticide tissue
concentrations in aquatic animals to estimate dose- and dietary-based exposures and
associated risks to mammals and birds consuming  aquatic organisms.  Previous analyses using
an earlier version of the KABAM model indicate relatively close agreement between its
predicted bioconcentration factor (BCF) and those reported from experimental studies for
endosulfan (2007; D346213).

For the aquatic-phase CRLF analysis, the ecosystem components for avian consumers were
modeled, as birds are considered surrogates for aquatic-phase amphibians. However, the
default avian species used in KABAM were altered to more accurately represent the size
range and dietary preferences of the CRLF.  Specifically, three size classes were modeled
(1.4g, 37g, and 238g) each with a different dietary composition that would be representative
of its body size and likely feeding pattern (Table 5.15).  For each of these size classes, two
different dietary preferences were selected to bound the range of trophic levels at which the
CRLF could potentially feed.
Table 5.15 Assumed dietary preferences of small (1.4g), medium (37g) and large (238g)
CRLF feeding on aquatic biota of the model ecosystem.

Trophic level in
diet
phytoplankton
Zooplankton
Benthic
invertebrates
filter feeders
small fish
Medium fish
large fish
Total
Diet for:
small CRLF
1
100.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
small
CRLF 2
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
100.0%
Med
CRLF1
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
100.0%
med
CRLF 2
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
0.0%
100.0%
large
CRLF1
0.0%
0.0%
100.0%
0.0%
0.0%
0.0%
0.0%
100.0%
large
CRLF 2
0.0%
0.0%
0.0%
0.0%
0.0%
100.0%
0.0%
100.0%
Other chemical and ecosystem-specific parameters selected for the KABAM analysis are
summarized in Table 5.16 and in Appendix I.
Table 5.16 Endosulfan chemical characteristics used as input to KABAM
Characteristic
Log KOW
Value
4.76
Source/Comments
Weighted mean of Kow for alpha and beta endosulfan based
on TGAI (70:30 alpha:beta). MRID 414215-01
                                       169

-------
    (L/kg OC)
10,600
Average value for alpha endosulfan used in PRZM/EXAMS
modeling (Table 3.3)
Estimated time to
steady state (Ts;
days)	
18
This value is calculated automatically from the Log K0w value
and is used to select the appropriate averaging period for the
surface water and pore water EEC	
Surface water EEC
(M9/L)
0.24-2.60
Range of 21-d surface water EECs for total endosulfan
(alpha, beta, sulfate) predicted using PRZM/EXAM modeling
(Table 3.5). Results from each of the 20 crop scenarios were
modeled separately.
Pore Water EEC
(M9/L)
0.09-1.18
Range of 21-d pore water EECs for total endosulfan (alpha,
beta, sulfate) predicted using PRZM/EXAM modeling (Table
3.7). Results from each of the 20 crop scenarios were
modeled separately.
Concentrations in aquatic prey predicted using KABAM for the two crop scenarios yielding
the lowest and highest 21-d average surface and pore water EECs (Eggplant and Broccoli,
Cabbage & Cauliflower, respectively) are summarized in Table 5.17.  Calculation of dose and
dietary-based EECs for these scenarios are provided in Table 5.18 and the resulting RQ
values are summarized in Table 5.19. For the scenario yielding the largest surface water and
pore water EECs (Broccoli, Cabbage & Cauliflower), acute, dose-based RQs range from 0.14
to 4.0, thus exceeding the acute listed LOG of 0.1 for all  six CRLF size class/feeding
preference combinations. Acute and chronic, dietary-based RQs do not  exceed the Agency
LOCs for the Broccoli, Cabbage & Cauliflower exposure scenario.  For  the scenario yielding
the smallest surface water and pore water EECs (Eggplant), acute, dose-based RQs range
from 0.01 to 0.37 and exceed the acute listed LOG of 0.1 for two of the six CRLF size
class/feeding preference combinations. Results from aquatic bioaccumulation modeling for all
20 crop exposure scenarios are summarized in Appendix I.

Based on these results, it appears that there is a potential for direct effects on aquatic-phase
CRLF due to the consumption of aquatic prey contaminated with  endosulfan, assuming
similar sensitivity to the most sensitive avian  species tested and surface water exposure
profiles predicted by PRZM/EXAMS.
Table 5.17 Estimated concentrations of endosulfan in ecosystem components for two crop exposure
scenarios





Ecosystem Component
Water (total)*
Water (freely dissolved)*
Sediment (pore water)*
Sediment (in solid)**
Phytoplankton
Zooplankton

Eggp

Total
concentration
(|jg/kg-ww)
0.237
0.237
0.090
38
622
468
ant(1)
Lipid
normalized
concentration
([jg/kg-lipid)
N/A
N/A
N/A
N/A
31,082
15,606
Broccoli, Cabbaae &
Caulifli

Total
concentration
(|jg/kg-ww)
2.603
2.603
1.181
501
6,828
5,142
Dwer^
Lipid
normalized
concentration
([jg/kg-lipid)
N/A
N/A
N/A
N/A
341,376
171,405
                                        170

-------
Benthic Invertebrates
Filter Feeders
Small Fish
Medium Fish
Large Fish
504
333
697
762
916
16,808
16,633
17,432
19,047
22,904
5,560
3,668
7,686
8,400
10,075
185,325
183,403
192,149
209,990
251,880
* Units: |jg/L; **Units: |jg/kg-dw
(1) Total concentrations in aquatic organisms calculated using KABAM v.1 .0. Lipid normalized
concentrations based on lipid fractions of 2% for phytoplankton and filter feeders, 3% for
zooplankton and benthic macroinvertebrates, and 4% for small, medium and large fish.
Table 5.18 Calculation of EECs for CRLF consuming aquatic prey contaminated by endosulfan
Wildlife
Species
Small
CRLF1
Small
CRLF 2
Med
CRLF1
Med
CRLF 2
Large
CRLF1
Large
CRLF 2
Biological Parameters
Body
Weight
(kg)
0.0014
0.0014
0.037
0.037
0.238
0.238
Dry Food
Ingestion
Rate
(kg-dry
food/kg-
bw/day)
0.577
0.577
0.184
0.184
0.096
0.096
Wet Food
Ingestion
Rate
(kg -wet
food/kg-
bw/day)
5.767
2.403
0.766
0.681
0.400
0.356
Drinking
Water
Intake
(L/d)
0.001
0.001
0.006
0.006
0.023
0.023
EECs (pesticide intake) (1>
Dose Based
(mg/kg-bw/d)
Eggplant
3.585
1.212
0.386
0.475
0.202
0.271
Broccoli,
Cabbage &
Cauliflower
39.37
13.36
4.261
5.236
2.225
2.988
Dietary Based (ppm)
Eggplant
0.62
0.50
0.50
0.70
0.50
0.76
Broccoli,
Cabbage &
Cauliflower
6.83
5.56
5.56
7.69
5.56
8.40
(1) EECs calculated using KABAM v. 1.0 using dietary preferences for the small (1.4g), medium (37g) and large
(238g) CRLF in Table 5.17 and predicted concentrations in aquatic prey from Table 5.19.
Table 5.19 Calculation of RQ values for CRLF consuming aquatic prey contaminated by endosulfan.
Wildlife Species
Acute
Dose Based'1 '
Dietary Based (2)
Chronic
Dietary Based (3)
Broccoli, Cabbage & Cauliflower
small CRLF 1
small CRLF 2
med CRLF 1
med CRLF 2
large CRLF 1
4.036
1.369
0.267
0.328
0.106
0.008
0.007
0.007
0.010
0.007
0.228
0.185
0.185
0.256
0.185
171

-------
Table 5.19 Calculation of RQ values for CRLF consuming aquatic prey contaminated by endosulfan.
Wildlife Species
large CRLF 2
Acute
Dose Based'1 '
0.142
Dietary Based (2)
0.010
Chronic
Dietary Based (3)
0.280
Eggplant
small CRLF 1
small CRLF 2
med CRLF 1
med CRLF 2
large CRLF 1
large CRLF 2
0.367
0.124
0.024
0.030
0.010
0.013
0.001
0.001
0.001
0.001
0.001
0.001
0.021
0.017
0.017
0.023
0.017
0.025
(1) Acute dose-based RQs calculated using the dose-based EECs from Table 5.20 and adjusted,
acute dose-based toxicity values of 9.76, 15.94, and 21.08 mg/kg-bw for the 1.4g, 37g, and 238g
CRLF, respectively. Unadjusted LD50 = 28 mg/kg-bw based on mallard duck (MRID 136998 ).
(2) Acute diet-based RQs calculated using diet-based EECs from Table 5.20 and a diet-based LC50
of 805 ppm for bobwhite quail (MRID 22923 )
(3) Chronic diet-based RQs calculated using the dietary EECs and an avian dietary NOAEC of 30
ppm for mallard duck (MRID 40335001)
       Ecological Incident Reports

As summarized in Section 4.4 and Appendix L, the vast majority of the 83 aquatic incident
reports associated with endosulfan involved mortality to fish (67). A wide variety of fresh and
estuarine species were reportedly affected (e.g., carp, catfish, largemouth bass, shad,
menhaden, mullet, spot, bluegill sunfish, gar and trout). These incident reports support the
findings from laboratory toxicity studies that fish are a highly  sensitive taxonomic group to
endosulfan exposure.  Of the 67 aquatic incidents involving fish, 53 (80%) are classified as
either 'highly probable' or 'probable' in the context of endosulfan use. Of all the incident
reports classified as highly probable or probable, about 15 (25%) are associated with a
'registered use' (most are accidental misuse or of unknown causes). Of those associated with
'registered uses,' most described pesticide runoff following periods of heavy rainfall as the
likely event that led to the reported incident. No incidents were reported for aquatic-phase
amphibians.  As noted in Appendix L, there are a number of uncertainties associated with
incident reports, including a likely underreporting of actual ecological incidents that occur and
uncertainty associated with attributing causality. Nevertheless, the information available in
from the incident reports are not inconsistent with the results from the modeling-based RQs
that suggest risks to freshwater fish (and by extension, the aquatic-phase CRLF) can exceed
Agency LOG.
                                        172

-------
Direct Effects: Terrestrial-Phase CRLF
       T-REX and T-HERPS Modeling

As described in Section 5.1, risks to terrestrial-phase CRLF were estimated using birds as a
surrogate using the Agency's T-REX model. (Table 5.3)  Risk quotients from the "broadleaf
plants/small insect" and "fruits/pods/seeds/large insect" categories were chosen for assessing
risks to terrestrial phase CRLF because it consumes terrestrial invertebrates and other prey
types were not represented in T-REX. The dose-based acute RQs calculated for 20g birds
consuming small and large insects exceed the Agency acute listed LOG (0.1) in all 20 crop
scenarios modeled (range: 5.3 to 26.4 for small insects; 0.6 to  2.9 for large insects,
respectively). The  acute diet-based RQs exceed the Agency's acute listed LOG in 19 of the
20 scenarios modeled for small insects but none of the 20 scenarios modeled for large insects
(range: 0.08 to 0.42 for small insects; 0.01 to 0.05 for large insects).  Chronic RQs for the
small and large insect categories exceed the Agency LOG of 1 for 20 and 3 crop exposure
scenarios,  respectively. Chronic RQs range from 2.3 to 11.3 for the small insect category and
0.3 to 1.3 for the large insect dietary category.

Direct effects on listed reptiles and terrestrial-phase amphibians were evaluated in the same
manner as above for birds except T-HERPS was used instead of T-REX. The model T-
FtERPS is a modified form of T-REX, which is designed to be more reflective of the food
requirements of amphibians  (including consumption of small herbivorous and insectivorous
mammals and small amphibians) and allow for an estimation of food intake for poikilotherms
using the same basic procedure as T-REX. This involves adjusting daily food intake with an
allometric model that accounts for the lower food intake of poikilothermic reptiles and
amphibians. The net effect of this approach is a reduction in pesticide exposure due to
reduced food consumption.
The same bird acute datum (LD50=28 mg/kg b.w. for mallard duck) was  used; however, the
adjusted LDSOs for the different  size classes were different.  These were 9.76 mg/kg b.w. for
small (1.4 g), 15.94 mg/kg b.w. for medium (37 g) and 21.08 for large (238 g) herpetofauna,
and were calculated by T-HERPS. All three size classes are  used to evaluate the potential for
direct acute effects on CRLF. Based on the modeled acute dose-based EECs for various use
scenarios and diet categories and the above acute toxicity data, the RQs for small
herpetofauna ranged from 0.03 to 1.34, with 24 of the 40 modeled scenarios resulting in
exceedences of the Agency's acute listed species LOG (>0.1). Based on the modeled acute
dose-based EECs for various use scenarios and diet categories and the above acute toxicity
data, the RQs for medium herpetofauna ranged from 0.006 to 14.55 with 60 of the 100
modeled scenarios resulting  in exceedences  of the Agency's acute listed species LOG (>0.1).
The acute dose-based RQs for large  herpetofauna range from 0.003 to 1.71 with 41 of the 100
modeled scenarios exceeding the Agency's LOG (>0.1). The acute dietary-based RQs ranged
from 0.003 to 0.71, and 39 of the 100 scenarios resulted in RQs that exceeded the agency's
LOG (>0.1). Finally, the chronic dietary-based RQs for herpetofauna ranged from 0.078 to
19.07 with 46 of these values exceeding the LOG for listed species (>1) (see Table 5.20).
Based on exceedences of the Agency's acute risk LOG (RQ>0.1) and chronic risk LOG
(RQ>1), it is concluded that endosulfan has  the potential to directly affect reptiles and
                                       173

-------
terrestrial-phase amphibians (i.e., CRLF, CIS and SFGS) during at least some portion of their
life-history.
Table 5.20 Summary of the acute and chronic dose- and dietary-based RQs for herpetofauna estimated
based on the maximum endosulfan foliar spray applications using T-HERPS version 1.0.
Use Category
Almonds,
hazelnut & walnut
Citrus
Broccoli,
cabbage,
Chinese
cabbage,
cauliflower,
kohlrabi
Kale, CollardsS
Mustard Green
Sweet corn
Cotton (ground)
Cotton (Areal)
Run
No.
1
1
1
2
1
1
3
Dietary Category
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Acute Dose Based
1.4 g 37 g 238 g
1.08
0.12



1.34
0.15



0.70
0.08



0.40
0.04



0.81
0.09



0.54
0.06



0.52
0.06


0.65
0.07
11.64
0.73
0.02
0.81
0.09
14.55
0.91
0.03
0.42
0.05
7.55
0.47
0.01
0.24
0.03
4.36
0.27
0.01
0.48
0.05
8.73
0.55
0.02
0.32
0.04
5.82
0.36
0.01
0.31
0.03
5.66
0.35
0.32
0.04
1.37
0.09
0.01
0.40
0.04
1.71
0.11
0.01
0.21
0.02
0.89
0.06
0.01
0.12
0.01
0.51
0.03
0.004
0.24
0.03
1.03
0.06
0.01
0.16
0.02
0.68
0.04
0.01
0.16
0.02
0.67
0.04
Sub
Acute
Dietary-
Based
0.34
0.04
0.57
0.04
0.01
0.42
0.05
0.71
0.04
0.01
0.22
0.02
0.37
0.02
0.01
0.13
0.01
0.21
0.01
0.004
0.25
0.03
0.43
0.03
0.01
0.17
0.02
0.28
0.02
0.01
0.16
0.02
0.28
0.02
Chronic
Dietary-
Based
9.00
1.00
15.25
0.95
0.31
11.25
1.25
19.07
1.19
0.39
5.84
0.65
9.90
0.62
0.20
3.38
0.38
5.72
0.36
0.12
6.75
0.75
11.44
0.72
0.23
4.50
0.50
7.63
0.48
0.16
4.38
0.49
7.42
0.46
                                       174

-------
Table 5.20 Summary of the acute and chronic dose- and dietary-based RQs for herpetofauna estimated
based on the maximum endosulfan foliar spray applications using T-HERPS version 1.0.
Use Category

Apples
Apricot,
nectarine, peach,
cherry, pear,
plum & prune
Lettuce &
Brussels sprouts
Cucumber,
melons, pumpkin
& squash
Eggplant
Ornamentals &
shade trees
Potato
Run
No.

1
1
1
1
3
1
1
Dietary Category
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Acute Dose Based
1.4 g 37 g 238 g

1.34
0.15



1.34
0.15



0.76
0.08



0.70
0.08



0.27
0.03



0.33
0.04



0.76
0.08



0.01
0.81
0.09
14.55
0.91
0.03
0.81
0.09
14.55
0.91
0.03
0.46
0.05
8.26
0.52
0.02
0.42
0.05
7.55
0.47
0.01
0.16
0.02
2.91
0.18
0.01
0.20
0.02
3.53
0.22
0.01
0.46
0.05
8.26
0.52
0.02
0.01
0.40
0.04
1.71
0.11
0.01
0.40
0.04
1.71
0.11
0.01
0.23
0.03
0.97
0.06
0.01
0.21
0.02
0.89
0.06
0.01
0.08
0.01
0.34
0.02
0.003
0.10
0.01
0.42
0.03
0.003
0.23
0.03
0.97
0.06
0.01
Sub
Acute
Dietary-
Based
0.01
0.42
0.05
0.71
0.04
0.01
0.42
0.05
0.71
0.04
0.01
0.24
0.03
0.40
0.03
0.01
0.22
0.02
0.37
0.02
0.01
0.08
0.01
0.14
0.01
0.003
0.10
0.01
0.17
0.01
0.004
0.24
0.03
0.40
0.03
0.01
Chronic
Dietary-
Based
0.15
11.25
1.25
19.07
1.19
0.39
11.25
1.25
19.07
1.19
0.39
6.39
0.71
10.83
0.68
0.22
5.84
0.65
9.90
0.62
0.20
2.25
0.25
3.81
0.24
0.08
2.73
0.30
4.63
0.29
0.09
6.39
0.71
10.83
0.68
0.22
175

-------
Table 5.20 Summary of the acute and chronic dose- and dietary-based RQs for herpetofauna estimated
based on the maximum endosulfan foliar spray applications using T-HERPS version 1.0.
Use Category
Sweet potato
Dry beans
(except Lima),
peas& pepper
Carrot
Celery
Strawberry
Tomato
Run
No.
3
1
2
2
2
1
Dietary Category
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Broadleaf Plants/Small Insects
Fruits/Pods/Seeds/Large Insects
Small Herbivore Mammals
Small Insectivore Mammal
Small Amphibians
Acute Dose Based
1.4 g 37 g 238 g
0.43
0.05



0.76
0.08



0.54
0.06



0.54
0.06



0.58
0.06



0.70
0.08



0.26
0.03
4.65
0.29
0.01
0.46
0.05
8.26
0.52
0.02
0.32
0.04
5.82
0.36
0.01
0.32
0.04
5.82
0.36
0.01
0.35
0.04
6.25
0.39
0.01
0.42
0.05
7.55
0.47
0.01
0.13
0.01
0.55
0.03
0.004
0.23
0.03
0.97
0.06
0.01
0.16
0.02
0.68
0.04
0.01
0.16
0.02
0.68
0.04
0.01
0.17
0.02
0.74
0.05
0.01
0.21
0.02
0.89
0.06
0.01
Sub
Acute
Dietary-
Based
0.13
0.01
0.23
0.01
0.005
0.24
0.03
0.40
0.03
0.01
0.17
0.02
0.28
0.02
0.01
0.17
0.02
0.28
0.02
0.01
0.18
0.02
0.31
0.02
0.01
0.22
0.02
0.37
0.02
0.01
Chronic
Dietary-
Based
3.59
0.40
6.09
0.38
0.12
6.39
0.71
10.83
0.68
0.22
4.50
0.50
7.63
0.48
0.16
4.50
0.50
7.63
0.48
0.16
4.83
0.54
8.19
0.51
0.17
5.84
0.65
9.90
0.62
0.20
|2) RQ values in bold indicate exceedence of listed species acute LOG (0.1) and chronic LOG (1 .0). Acute dose-base RQs were
based on dose-based EECs divided by adjusted avian acute LCSOs (9.76 mg/kg b.w. for small; 15.94 mg/kg b.w. for medium;
and 21 .08 mg/kg b.w. for large).
176

-------
       Likelihood of Effects

Based on an observed slope of 8.5 (with upper and lower bounds of 3.6 and 13.4,
respectively), the corresponding LD50 of the most sensitive avian species tested (mallard
duck; MRID 136998), and the acute listed species LOG of 0.1, the chance of an individual
mortality for terrestrial-phase CRLF is ~ 1 in 1.05 xlO17 (with upper and lower bounds of ~ 1
in 6.29 x!03to ~ 1 in 3.31 xlO40, respectively, Table 5.13). Using T-HERPS which considers
the feeding preferences and bioenergetics of amphibians and the observed slope described
previously, the likelihood of individual mortality to small  (1.4g) CRLF using the highest
acute RQ value (RQ=1.34 for Citrus) is ~ 1 in 1.16 (with lower and upper bounds of ~ 1 in
1.47 to 1 in 1.04, respectively). For medium size CRLF (37g), the likelihood of individual
mortality for the highest acute RQ (RQ=14.55 for Citrus)  is ~ 1 in 1.00 (with lower and upper
bounds of ~ 1 in 1.00 to ~ 1 in 1.00). Finally, for large size CRLF (238g), the likelihood of
individual mortality for the highest acute RQ (RQ=1.71 for Citrus) is ~ 1 in 1.02 (with lower
and upper bounds of ~ 1  in 1.25 to ~ 1 in 1.00, respectively). Based on the crop exposure
scenario that results in the smallest RQ values (Eggplant), the likelihood of individual
mortality for the highest acute RQs (0.27 for  1.4g, 2.91 for 37g and 0.34 for 238g CRLF) are:
1 in 1.62 xlO6 (bounds =  1 in 1 in 9.61  xlO13  to 1 in  50.0) for 1.4g CRLF; 1 in 1.00 (bounds=
lin l.OStol  in 1.00) for 37g CRLF; and 1 in 2.67 x!04(bounds=l in4.65x!09  to 1 in 21.4)
for 238g CRLF.

       Ecological Incident Reports

Very few ecological incidents involving endosulfan have been reported for terrestrial
organisms (Section 4.4 and Appendix L).  Of the seven terrestrial incidents that have been
reported for endosulfan, none are classified as 'highly probable' and two are classified as
'probable.' The two 'probable' incidents involve birds (blue jay, crow, owl), mammals
(squirrel, opossum, red fox) and an amphibian (unidentified frog). One of the two probable
incidents (incident ID: 1012626-001) involved morality to a blue jay, a crow, a red fox, and 12
gray squirrels apparently  poisoned from an unknown source.  A pooled sample of the
gastrointestinal tract from four of the gray squirrels was found to have 126 ppm endosulfan.
The other 'probable' ecological incident (ID 1010533-001) involved apparent human health
impacts in addition to anecdotal evidence of food chain transfer and secondary poisoning of
endosulfan (termite to frog to owl).

Conclusions Regarding Direct Effects on CRLF

Based on multiple lines of evidence, including:

       (1) RQ exceedence of freshwater fish acute and chronic LOCs using model and
       monitoring-based EEC estimates
       (2) RQ exceedence of acute listed LOG for for  small, medium and large CRLF
       consuming aquatic prey predicted to bioaccumulate endosulfan
       (3) RQ exceedence of acute and chronic LOCs for small, medium and large CRLF
       modeled using T-HERPS,
                                       177

-------
       (4) the high probability of an individual mortality occurrence using the highest acute
       RQ from T-HERPS and for freshwater fish (~1 in 1),
       (5) the spatial and temporal overlap among endosulfan use and CRLF occurrence and
       critical habitat described previously,

the Agency concludes that agricultural uses of endosulfan in California have a reasonable
likelihood to cause direct adverse effects to aquatic and terrestrial-phase CRLFs should
exposure at the predicted EECs actually occur for aquatic and terrestrial-phase CRLF and
assuming the sensitivity of CRLF to endosulfan is similar to the most sensitive freshwater fish
(for its aquatic phase) and avian species (for its terrestrial-phase) selected for RQ calculation.
5.2.1.2 Indirect Effects to CRLF (via Potential Loss of Prey)

As discussed previously in Section 2.5, the diet of aquatic-phase CRLF tadpoles is composed
primarily of unicellular aquatic plants (i.e., algae and diatoms) and detritus, while the diet of
terrestrial-phase CRLF includes terrestrial and aquatic invertebrates, mammals, frogs, and
fish. The main food source for juvenile aquatic- and terrestrial-phase CRLFs is thought to be
aquatic and terrestrial invertebrates. However, life history data for terrestrial-phase CRLFs
indicate that large adult frogs also consume terrestrial vertebrates, including mice and frogs.

Algae (non-vascular plants)

Based on the most sensitive surrogate aquatic non-vascular plant toxicity data (EC50 value of
428 jig a.i./L for the marine alga, Pseudokirchneriella subcapitatum, and the maximum
aquatic peak EEC of all use scenarios used to represent all of the agricultural uses  of
endosulfan in CA (5.88 ug/L for Lettuce and Brussels Sprouts), all RQs for aquatic non-
vascular plants are < 0.014 (see Section 5.1.1.3). Since the RQs do not exceed the Agency's
LOG (1) for aquatic non-vascular plants based on the most sensitive data available to the
Agency, labeled endosulfan use in California appears not likely to indirectly affect the
aquatic-phase CRLF via effects to aquatic non-vascular plant food sources. Although the lack
of risk to non-vascular aquatic plants is consistent with the insecticidal mode of action of
endosulfan, it should be considered that data were available for only one species of aquatic
non-vascular plant that was considered suitable for quantitative use in  this  risk assessment. In
order for potential effects to non-vascular plants to be identified, an untested species must be
at least 75 times more sensitive than the available data for P. subcapitatum.  Therefore, only
fish, aquatic invertebrates, and frogs will be characterized for potential indirect effects to the
aquatic-phase CRLF.

Freshwater Fish and Aquatic-phase Frogs

The potential for direct effects to listed fish and aquatic-phase frogs is discussed above in
Section 5.2.1.1. Because fish and frogs are also considered potential prey  items for the
aquatic-phase CRLF, indirect effects via potential prey item reduction are also considered
here.  As previously described, acute RQs based on modeled EECs for freshwater fish (used as
a surrogate for the aquatic phase CRLF) range from 7.2 to 58.8, thus exceeding the Agency's
listed LOG (0.05) in all 20 crop scenarios modeled (Table 5.1). Risk quotients based on
                                        178

-------
maximum reported concentrations of total endosulfan in California surface waters also exceed
the Agency's acute listed LOG.  Similarly, chronic RQs range from 5.7 to 80.6 and exceed the
Agency's LOG (1.0) in all 20 crop scenarios modeled.  Available information indicates that
the modeled EECs extend well into the cumulative species sensitivity distribution for
endosulfan (Figure  5.3). Probit analysis using the default slope of 4.5 for freshwater fish also
indicates a high probability of an individual mortality occurrence (~1 in 1), even with the RQ
associated with the smallest peek EEC (7.2 for Eggplant). The preponderance of reported
ecological incidents involves freshwater fish, which further supports findings of risk to this
taxonomic group. Therefore, considering the aforementioned lines of evidence and the
potential spatial and temporal overlap of reported endosulfan use with the occurrence of the
CRLF (Table 5.7 and Appendix E), there appears to be a reasonable potential for indirect
effects to the aquatic-phase CRLF from loss offish/aquatic amphibian prey as result of
labeled endosulfan use in California provided that exposures occur at or near modeled EECs
and that the sensitivity of tested fish is similar to those found in aquatic ecosystems from
which the CRLF obtains prey.

Freshwater Invertebrates (Water Column Exposure)

Based on surrogate freshwater invertebrate toxicity data (LCso value of 0.6 jig a.i./L for
mayfly) and modeled aquatic peak EECs for various use scenarios used to represent all of the
agricultural uses of endosulfan in CA, acute RQs for freshwater invertebrates range from 1.2
to 9.8 for freshwater invertebrates; exceeding the Agency's acute listed species LOG (0.05) in
all 20 crop scenarios modeled (Table 5.2). Because the acute RQs all substantially exceed the
acute listed LOG of 0.05, the likelihood of mortality to an individual freshwater invertebrate
is considered high (e.g., from ~1 in 1.56 to ~1 in 1; Table 5.11).

Based on the distribution of species-mean acute toxicity values for freshwater invertebrates
available from ECOTOX and submitted studies,  these  minimum and maximum peak acute
EECs would exceed the LOG for approximately 45% and 60% of the tested species,
respectively (vertical red lines in Figure 5.4). As illustrated in Figure 5.4, the LC50 chosen
for quantitative use in this risk assessment corresponds to approximately the 15th percentile
when all available data are included in the distribution (i.e., those considered appropriate only
for qualitative use in addition to those data considered acceptable for quantitative use for RQ
calculation). Although those data that fall below the LC50 used for RQ calculation (0.6 ug/L)
are not considered acceptable for quantitative use due to factors such as use of unmeasured
exposure concentrations, insufficient documentation of test procedures, inconsistency with the
results of other acceptable studies (see Appendix J), they do suggest that effects of
endosulfan might occur at concentrations of about l/6th that selected for RQ calculation. The
lowest LC50 value in Figure 5.4 is for the 2nd instar  larval stage of the southern house
mosquito. We did not consider this value usable because the authors of the study did not
provide adequate documentation on how the study was conducted. The exposure
concentrations were not listed or measured; the number of organisms used for each treatment
was not given; and the few details that are given are only listed for the 4th instar. One has to
assume that the procedures were the same for the other life stages.
                                        179

-------
                           Freshwater Invertebrates





c
ns
£

ns
c
O
is
ns
^
LL.





1.00 -
0.90 -

0.80 -

0.70 -
0.60 -


0.50 -


0.40 -
0.30 -
0.20 -

0.10 -
n nn
i i
I I
I I
i i
I I
I I
I I

I ?!
g> i _£J i
.0 I ~* |
i i
1 1
i i
i i
1 1 A
i |»
|U* ]
benchmark I!*
If
•*f i

* * ll 1
0
2
O
LU
LU
to
^

X

*
f
>







O - *
O A*
^ > *
^ •
LU ^T
Ml A
!i: A ^
c/) A
0 A
^ /
•= *.*
f*












          0.001     0.01      0.1      1       10     100    1000   10000   100000
                                     LC50
Figure 5.4 Species sensitivity distribution of the acute toxicity of endosulfan to freshwater invertebrates via
water column exposure

Similarly, chronic RQs calculated from the 21-d EECs exceed the Agency LOG of 1 for all 20
scenarios (Table 5.2). The 21-d EECs range from 0.24 to 2.6 ug/L which correspond to
chronic RQs 23.7 and 260, respectively, based on the estimated NOAEC of 0.01 ug/L for the
mayfly, Atalophlebia australis. This NOAEC was estimated using an acute-chronic ratio of
61.5 derived from Daphnia magna because the mayfly was a much more acutely sensitive
compared to D. magna.  However, use of an estimated NOAEC introduces uncertainty into
the calculation of the chronic RQ values. Since acceptable data on the chronic toxicity of
endosulfan to freshwater invertebrates  are limited to D. magna,  chronic toxicity data for
estuarine and marine invertebrates are considered here in order to evaluate the uncertainty
associated with the estimated NOAEC  of 0.01  ug/L. Specifically, five chronic tests are
available for the saltwater mysid, Americamysis bahia, that were conducted using flow-
through and measured exposures (McKenney 1982; ECOTOX reference #3736; Appendix J).
NOAECs from these five tests (0.14 to 0.52 ug a.i./L) are within an order of magnitude of the
estimated NOAEC of 0.01 ug/L for D.  magna, thus suggesting the estimated NOAEC is not
unreasonable within the limits of this comparison. Furthermore, if these measured NOAECs
were used directly for chronic RQ calculation, the resulting RQs would still exceed the
chronic LOG in all 20 exposure scenarios tested. On the other hand, if the measured chronic
NOAEC of 2.7 ug/L for D. magna was used for RQ calculation, none of the calculated RQs
would exceed the LOG of 1 (the highest RQ would approach 1).  However, use of this
NOAEC of 2.7 ug/L is not considered reasonable for estimating risks from chronic exposure
to freshwater invertebrate assemblages because it exceeds the LC50 values for about 30% of
                                       180

-------
the tested species (Figure 5.4), likely because D. magna is a relatively insensitive invertebrate
species.

With respect to measured concentrations of endosulfan in surface water, maximum
concentrations of total endosulfan (0.95 ug/L from 1991-1996; 0.05-0.08 from 2001-2008;
Table 5.14) correspond to chronic RQ values of 95 and 5 to 8 for the 1991-1996 and 2001-
2008 sampling periods, respectively, based on a NOAEC of 0.01 ug/L for the mayfly. Acute
RQ values based on an LC50 of 0.6 ug/L for mayfly and these maximum reported endosulfan
concentrations are 1.6 for the 1991-1996 sampling period and 0.08-0.13 from the 2001-2008
sampling period and exceed the acute listed LOG of 0.05. Due to the declining pattern of
endosulfan use in California, data from the 2001-2008 time period are considered more
representative of current and likely future use of endosulfan in California.  The acute and
chronic RQs based on the maximum reported concentrations from the 2001-2008 time period
suggest a potential risk to freshwater invertebrates from chronic exposure to endosulfan in
California.

In conclusion, based on: (1) the exceedances of listed species acute and chronic risk LOCs for
freshwater invertebrates for all of the modeled scenarios assessed, (2) the high probability of
an individual mortality occurrence using both the highest and lowest acute RQ for freshwater
invertebrates, (3) the spatial overlapping of patterns  of endosulfan use in California with
CRLF habitat (Table 5.7), and (4) the reported aquatic monitoring data that exceed acute
listed and chronic LOCs, it appears that there are adequate lines of evidence to conclude that
there is a potential for indirect effects to the aquatic-phase CRLF from loss of freshwater
invertebrate prey as result of labeled endosulfan use in California provided that exposures
occur at or near modeled EECs and that the sensitivity of tested freshwater invertebrates is
similar to those found in aquatic ecosystems from which the CRLF obtains prey.

Freshwater Invertebrates: Sediment Exposure
As summarized in Section 4.1.2.6, toxicity data are available for freshwater invertebrates via
exposure to contaminated sediments for endosulfan (as endosulfan sulfate). Exposure via
sediments is of interested because it can involve both uptake through respiration of pore and
overlying water in addition to ingestion of contaminated sediment.  Chronic risk to freshwater
sediment invertebrates is based on is based on 21-day EECs and the lowest chronic toxicity
value for freshwater sediment invertebrates. Based on freshwater sediment toxicity data for
endosulfan sulfate in porewater (NOAEC=0.35 jig a.i./L for midge) and 21-day EECs for
various use scenarios used to represent all  of the agricultural uses of endosulfan in CA, all
acute RQs for freshwater invertebrates range from 0.26  (eggplant) to 3.37 (lettuce and
brussels sprouts). Twelve of the set of 20 modeled scenarios used to represent all of the
agricultural uses of endosulfan in CA resulted in an exceedance of the Agency's chronic risk
LOG (RQ>1) for freshwater invertebrates (see Table 5.21).
Based on exceedances of the Agency's chronic risk LOG (RQ>1) for 12 of the set of 20
modeled scenarios used to represent all of the agricultural uses of endosulfan in CA, there is a
potential for indirect effects to the CRLF via reduction in its freshwater sediment invertebrate
prey base, provided that exposures occur at or near modeled EECs and that the sensitivity of
tested invertebrates via sediment exposure is similar to those found in aquatic ecosystems
upon which the CRLF relies for prey.
                                        181

-------
Table 5.21 Acute and Chronic RQs for freshwater sediment invertebrates based on EECs for use
categories used to represent all endosulfan uses in CA
Use Category
Almonds, Hazelnuts & Walnuts
Citrus
Broccoli, Cabbage & Cauliflower
Collards, Kale & Mustard Green
Sweet corn for fresh market only
Cotton (ground)
Cotton (aerial)
All fruit trees
Lettuce & Brussels Sprouts
Cucurbits
Eggplant
Ornamentals or Shade Trees (Southern Coast)
Ornamentals or Shade Trees (Northern Central coast)
Potato
Potato (Northern Central coast)
Sweet Potato
Beans & Peas (dry) & Pepper
Carrot & Celery
Strawberry
Tomato
Run No.
1
1
1
2
1
2
3
1
1
1
3
1
1 Add
1
1 Add
3
1
2
1
1
Peak
EEC
(ug/L)
0.37
0.23
1.18
0.38
0.51
0.35
0.33
0.30
1.19
0.42
0.09
0.80
0.84
0.25
0.37
0.18
0.39
0.39
0.72
0.28
21 -day
EEC
(ug/L)
0.36
0.22
1.17
0.38
0.51
0.34
0.32
0.30
1.18
0.42
0.09
0.79
0.83
0.25
0.37
0.18
0.39
0.38
0.71
0.27
Chronic
RQ(2)
1.04
0.63
3.34
1.07
1.45
0.98
0.93
0.85
3.37
1.19
0.26
2.27
2.36
0.70
1.05
0.52
1.11
1.09
2.03
0.78
|2) RQ values in bold indicate exceedence of chronic LOG (1 .0). Chronic RQs were based on the 21 day average EEC divided
by the chronic NOAEC for the midge Chironomustentans of 0.35 ug/L (MRID 47318101)
Terrestrial Invertebrates

Terrestrial invertebrates are one of the prey items for terrestrial phase CRLF. As such,
impacts to the terrestrial invertebrate prey base of the CRLF have the potential to indirectly
affect the CRLF. As described in Section 5.1.2.3, the potential risks to terrestrial
invertebrates resulting from foliar spray applications of endosulfan were derived using T-
REX, and the most sensitive toxicity data available for terrestrial invertebrates (acute contact
LD50 (0.15 jig a.i./g) for the beet web worm).  EECs from the large and small insect
categories ranged 37.5 to 337 ppm, respectively, for the crop exposure scenarios yielding the
lowest and highest EECs (eggplant and citrus fruits, respectively; Table 3.23).  The resulting
RQs for the small insect category ranged from 450 to 2250 and RQs calculated for large
insects ranged from 50 to 250 (Table 5.5).  All 20 use category RQs for both size classes
substantially exceeded the Agency's interim LOG for listed terrestrial invertebrates
(RQ>0.05), which is not unexpected given the insecticidal mode of action of this pesticide.
                                       182

-------
When compared to the acute contact toxicity of endosulfan of 35 ug/g based on the honey bee
(calculated from an LD50 of 4.5 ug/bee using a body weight of 0.128g; MRID 000199), the
adult beet web worm (a target insect) is approximately 230 times more sensitive (LD50 of
0.15 ug/g). If RQs were calculated based on the standard honey bee acute contact LD50 (35
ug/g), the smallest RQ (1.1 for eggplant) would still exceed the Agency's interim acute listed
LOG for terrestrial invertebrates by nearly 20-fold. When compared to the distribution of
acute contact toxicity  of endosulfan to terrestrial invertebrates (Figure 5.5), the lowest EEC
for small insects (337  ppm) and large insects (37.5 ppm) approximate the 99th and 80th
percentiles, respectively.
                                Terrestrial Insects
1.0 -
0.9 -
0.8 -
* 07-
c
TO
* 0.6 -
TO
| °-5 -
"o
= 0.4-
0.3 -

0.2 -
0.1 -

0 0
I °
I 0
I
• - — — H
1 O b
O 1 • b
0 rr
D | • rr

1
1
E O 1
i 1
• I
! I
1 1
0.1 1 10 100
LD50 (n,g/g)



oney Bee
eetle larva
eetle adult
oth larva
oth adult








1000

Figure 5.5 Species sensitivity distribution of the acute contact toxicity of endosulfan TGAI to
terrestrial invertebrates based on species mean values.

Given the large acute RQs, the probit-based likelihood estimates of individual mortality all
indicate a high probability for mortality to individual terrestrial insects (~1 in 1; Table 5.12).
No reported ecological incidents associated with endosulfan involved terrestrial insects;
however it is likely that impacts to terrestrial insects would be substantially under-reported in
the EIIS  database due to the difficulty in observing and documenting causal linkages between
endosulfan use and non-target terrestrial insects.
Based on the exceedences of the Agency's interim LOG for listed terrestrial invertebrates,
endosulfan use in CA does have the potential to indirectly affect the CRLF through reduction
in terrestrial prey base, provided that exposures occur at or near modeled EECs and that the
sensitivity of tested terrestrial invertebrates via acute contact exposure is similar to those
found in  terrestrial ecosystems upon which the CRLF utilizes as a prey source.
                                         183

-------
Small Terrestrial Vertebrates

For endosulfan spray applications, dietary-based and dose-based exposures of terrestrial-
phase CRLF to potential small terrestrial vertebrate prey (mammals and amphibians) are
assessed using small mammals (15 g) which consume short grass, and small birds (20g; they
represent terrestrial-phase amphibians that it may consume) which consume small (broadleaf
plants/small insects dietary category) and large (fruits/pods/seeds/large insects dietary
category) invertebrates using the T-REX model described earlier in Section 5.2.1.1.

RQ values representing direct exposures of endosulfan to terrestrial-phase CRLFs are also
used to represent exposures of endosulfan to amphibians in terrestrial habitats that may serve
as prey for the CRLF. For terrestrial-phase amphibians, as described in Section 5.2.1.1, the
Agency concluded that agricultural uses of endosulfan in California have a reasonable
likelihood to cause direct adverse effects to terrestrial-phase CRLFs should exposure at the
predicted EECs actually occur for a terrestrial-phase CRLF and assuming its sensitivity to
endosulfan is similar to the most sensitive avian species selected for RQ calculation.
Furthermore,  dose-based RQs determined from T-HERPS modeling for the  1.4g and 37g
amphibians are used to bracket the potential risks to a specific prey item (Pacific tree frog,
2.3g) of the terrestrial-phase CRLF. Please refer to Section 5.2.1.1 for additional discussion
and characterization of potential risks to terrestrial-phase amphibians that may serve as prey
for the terrestrial-phase CRLF. Therefore, that Agency concludes that endosulfan use in CA
does have the potential to indirectly adversely affect the CRLF through reduction in terrestrial
vertebrate prey base (birds and other amphibians), provided that exposures occur at or near
modeled EECs and that the sensitivity of tested avian and amphibian prey is similar to those
found in terrestrial ecosystems upon which the CRLF utilizes as a prey source.
Regarding the potential effects on the small mammal prey base of the CRLF, the Agency used
surrogate toxicity data (LD50=10 mg/kg-bw and NOAEL =15 mg/kg-diet), the maximum
allowable application rate (1 application, 2.5 Ibs a.i./acre/application), the foliar dissipation
half-life of 4 days for endosulfan from Willis and McDowell (1987), and upper bound Kenaga
values from T-REX, there is a potential for direct adverse effects on SMHM individuals from
foliar spray applications of endosulfan in CA (Table 5.4).  Although EECs for small
mammals feeding on short grass yield the highest risk estimates, it appears that the impacts of
endosulfan use to the SMHM potentially extend beyond those feeding on short grass.

For spray applications of endosulfan, the dose-based acute RQs range from 5.21 to 26.03,
from 2.39 to 11.93, from 2.93  to 14.64,  and from 0.33 to 1.63 for the short grass, tall grass,
broadleaf plants/small insects, and fruits/pods/seeds/large insects dietary categories,
respectively. For acute dose-based RQs, all modeled scenarios (N=20) resulted in an
exceedance of the Agency's acute listed species LOG (RQ>0.1) for mammals that feed on
short grass, tall grass, broadleaf plants/small insects, and fruits/pods/seeds/large insects.

The dose-based chronic RQs ranged from 69.41 to 347.04, from 31.81 to 159.06, from  39.04
to 195.21, and from 4.34 to 21.69, for the short grass, tall grass, broadleaf plants/small insects,
and fruits/pods/seeds/large insect dietary categories, respectively. For chronic dose-based
RQs, all modeled scenarios (N=20) resulted in an exceedance of the Agency's acute listed
species LOG (RQ>1) for mammals that feed on short grass, tall  grass, broadleaf plants/small
insects, and fruits/pods/seeds/large insects.
                                        184

-------
Estimates of the likelihood of mortality to individual 15g mammalian prey resulting from
predicted endosulfan EECs range from ~1 in 70 for large insects based on the eggplant crop
exposure scenario to ~1 in 1 for the citrus fruits exposure scenario.

The dietary-based chronic RQs range from 8 to 40.00, from 3.67 to 18.33, from 4.50 to 22.50,
and from 0.50 to 2.50, for the short grass, tall grass, broadleaf plants/small insects, and
fruits/pods/seeds/large insects dietary categories, respectively. For chronic dietary-based RQs,
all modeled scenarios (N=20) resulted in an exceedance of the Agency's acute listed species
LOG (RQ>1) for mammals that feed on short grass, tall grass, and broadleaf plants/small
insects. Thirteen out of 20 modeled scenarios resulted in an exceedance of the Agency's
chronic risk LOG (RQ>1) for mammals that feed on fruits/pods/seeds/large insects. Therefore,
RQs exceed the Agency's acute listed species LOG (RQ>0.1) and chronic risk LOG (RQ>1)
for mammals (see Table 5.4) for foliar spray applications of endosulfan.

Therefore, that Agency concludes that endosulfan use in CA does have the potential to
indirectly affect the CRLF through reduction in terrestrial vertebrate prey base (small
mammals), provided that exposures occur at or near modeled EECs and that the sensitivity of
tested mammalian prey is similar to those found in terrestrial ecosystems upon which the
CRLF  utilizes as a prey source.

5.2.1.3 Indirect Effects (via Habitat Effects)

Aquatic plants serve several important functions in aquatic ecosystems.  Non-vascular aquatic
plants  are primary producers and provide the autochthonous energy base for aquatic
ecosystems. Vascular plants provide structure, rather than energy, to the system, as
attachment sites for many aquatic invertebrates, and refugia for juvenile  organisms, such as
fish and frogs. Emergent plants help reduce sediment loading and provide stability to
nearshore areas and lower streambanks.  In addition, vascular aquatic plants are important as
attachment sites for egg masses of aquatic species. No toxicity data were available for
vascular aquatic plants that were considered appropriate for quantitative  or qualitative use in
this assessment. For non-vascular aquatic plants, the maximum RQ calculated is s < 0.014
(see Section 5.1.1.3). Since the RQs do not exceed the Agency's LOG (1) for aquatic non-
vascular plants based on the most sensitive data available to the Agency, labeled endosulfan
use in  California will not indirectly affect the aquatic-phase CRLF via effects to aquatic
habitat based on non-vascular aquatic plants.

Terrestrial plants serve several important habitat-related functions for the listed assessed
species. In addition to providing habitat and cover for invertebrate and vertebrate prey items
of the listed assessed species, terrestrial vegetation also provides shelter and cover from
predators while foraging. Upland vegetation including grassland and woodlands provides
cover during dispersal. Riparian vegetation helps to maintain the integrity of aquatic systems
by providing bank and thermal stability,  serving as a buffer to filter out sediment, nutrients,
and contaminants before they reach the watershed, and serving as  an energy source.  No data
on the  effects of endosulfan on terrestrial plants were considered acceptable for quantitative
use in this risk assessment. Based on efficacy studies, however, available information
                                        185

-------
indicates that agricultural crops are not adversely affected by endosulfan use rates that are
associated with pest control. This, in conjunction with the insecticidal mode of action of
endosulfan, suggests that terrestrial plants may not be sensitive to endosulfan compared to
aquatic and terrestrial animals.  However, use of efficacy studies can be confounded by lack
of experimental controls, multiple stressors (pesticide and insect pest), and plant species that
may not be representative of those upon which the CRLF may depend.

5.2.1.4 Effects to Designated Critical Habitat

       Aquatic-Phase PCEs

Three of the four assessment endpoints for the aquatic-phase primary constituent elements
(PCEs) of designated critical habitat for the CRLF are related to potential effects to aquatic
and/or terrestrial plants:

    •  Alteration of channel/pond morphology or geometry and/or increase in sediment
       deposition within the stream channel or pond: aquatic habitat (including riparian
       vegetation) provides for shelter, foraging, predator avoidance, and aquatic dispersal
       for juvenile and adult CRLFs.
    •  Alteration  in water chemistry/quality including temperature, turbidity,  and oxygen
       content necessary for normal growth and viability of juvenile and adult CRLFs and
       their food source.
    •  Reduction and/or modification of aquatic-based food sources for pre-metamorphs
       (e.g., algae).

Conclusions for potential indirect effects to the CRLF via direct effects to aquatic and
terrestrial plants are used to determine whether effects to critical habitat may occur.  As
discussed above for aquatic plants and terrestrial plants (Section 5.2.1.3), labeled endosulfan
use in California will not indirectly affect the CRLF via impacts to habitat and/or primary
production.

The remaining aquatic-phase PCE is "alteration  of other chemical characteristics necessary
for normal growth and viability of CRLFs and their food source."  Other than impacts to algae
as food items for tadpoles (discussed above), this PCE is assessed by considering direct and
indirect effects to the aquatic-phase CRLF via acute and chronic freshwater fish and
invertebrate toxicity  endpoints as measures of effects. Based on the analyses discussed above,
there is a potential for habitat effects via impacts to aquatic-phase CRLFs (Sections  5.2.1.1)
and effects to freshwater invertebrates and fish as food items (Sections 5.2.1.2) from
endosulfan use in California.

       Terrestrial-Phase PCEs

Two of the four assessment endpoints for the terrestrial-phase PCEs of designated critical
habitat for the CRLF are related to potential effects to terrestrial plants:

    •  Elimination and/or disturbance of upland habitat; ability of habitat to support food
       source of CRLFs: Upland areas within 200 ft of the edge of the riparian vegetation or
                                        186

-------
       drip line surrounding aquatic and riparian habitat that are comprised of grasslands,
       woodlands, and/or wetland/riparian plant species that provides the CRLF shelter,
       forage, and predator avoidance.

    •   Elimination and/or disturbance of dispersal habitat: Upland or riparian dispersal
       habitat within designated units and between occupied locations within 0.7 mi of each
       other that allow for movement between sites including both natural and altered sites
       which do not contain barriers to dispersal.

As discussed above for terrestrial plants (Section 5.2.1.3), labeled endosulfan use in
California will not indirectly affect the CRLF via impacts to habitat and/or primary
production.

The third terrestrial-phase PCE is "reduction and/or modification of food sources for
terrestrial phase juveniles and adults." To assess the impact of endosulfan on this PCE, acute
and chronic toxicity endpoints for terrestrial invertebrates, mammals, and terrestrial-phase
frogs are used as measures of effects. Based on the potential for a reduction in mammalian,
terrestrial invertebrate, and amphibious prey items from registered endosulfan use (Section
5.2.1.2), the Agency concludes there is a potential for habitat effects via indirect effects to
terrestrial-phase CRLFs via reduction in prey base.

The fourth terrestrial-phase PCE is based on alteration of chemical characteristics necessary
for normal growth and viability of juvenile and adult CRLFs and their food source. Based on
the  preceding discussions, the Agency concludes there is a potential for habitat effects via
direct (Section 5.2.1.1) and indirect effects (Section 5.2.1.2) to terrestrial-phase CRLFs.

5.2.2 California tiger salamander

5.2.2.1 Direct Effects

Endosulfan is currently registered for numerous diverse agricultural uses that span a large
variety of geographical regions throughout the entire state of California. Therefore, there is
the  potential for endosulfan use across the state to spatially and temporally coincide with the
CTS breeding season.

The CTS has aquatic phase eggs and larvae and terrestrial phase juveniles and adults.
Aquatic-phase CTS inhabits low elevation vernal pools and seasonal ponds while juvenile  and
adults are found in associated grassland, oak savannah, and coastal scrub plant communities.
Although aquatic phase CTS are adapted to natural  vernal pools and ponds, they are
frequently found in manmade or modified ephemeral and permanent ponds, including  stock
ponds. Terrestrial phase CTS prefer open grassland to areas of continuous woody vegetation
and may live for up to 10 years, spending the majority of their lives in the upland habitats.
The upland component typically consists of grassland savannah, but also can consist of
grasslands with scattered oak trees, and scrub and chaparral habitats. Juvenile and adult CTS
spend the dry summer and fall months in the burrows of California ground squirrels
(Spermophilus beecheyf) and Botta's pocket gopher (Thomomys bottae). California tiger
                                        187

-------
salamanders cannot dig their own burrows, and as a result their presence is associated with
active burrows of small mammals such as ground squirrels and pocket gophers.

The young aquatic  larvae eat small crustaceans, algae, and mosquito larvae. When they are
large enough, they  begin to take advantage of aquatic insects, invertebrates and tadpoles of
Pacific treefrogs, California red-legged frogs, western toads, and spadefoot toads. Once they
reach metamorphosis, which is usually by late spring or early summer, juveniles are ready to
roam to their terrestrial nesting habitat and may disperse up to two miles from their natal
ponds. As adults, CTS feed on terrestrial invertebrates, insects, frogs, and worms.

From southern Colusa County south to northern Kern County, CTS habitat is disjunct remnant
vernal pool complexes and isolated ponds scattered mainly along narrow strips of rangeland
on each side of the  Central Valley. From Suisun Bay south to the Temblor Range, CTS are
found in sag ponds and human-maintained stock  ponds in the coastal ranges. Populations of
CTS are also located  in Sonoma and Santa Barbara counties.

Temporal and Spatial Overlap

A comparison of Figure 5.1 and Figure 5.6 indicates that endosulfan use and CTS sensitive
life stages (larvae, young juveniles) could potentially overlap in time. A peak in endosulfan
application in Feb-March coincides with larval development. Application from May - October
coincides with young juvenile dispersal.


1 | Young Juveniles
Larvae
Breeding/Egg
Masses
J
F

M

A

M


J


J


A


S



o





N
D
Figure 5.6 CTS Reproductive Events by Month; Adults and juveniles can be present all year
Since pesticide use information is only available at the county level, the degree that the CTS
or its critical habitat actually coincides in the same or similar locations cannot be precisely
determined. With this limitation in mind, the spatial overlap of average endosulfan use for
2005-06 reported by CDPR/PUR and CTS distribution (including critical habitat) is
summarized at the county level in Table 5.7.  Figure 5.7 summarizes the overlap in average
endosulfan use and species/habitat occurrence for all 7 species assessed. For the CTS,
Approximately 85% of endosulfan use (74,196 Ibs) is applied in counties in which the CTS or
its critical habitat occurs.

While Figure 5.7 suggests a significant potential overlap in endosulfan use and the
occurrence of CTS or its critical habitat at the county level,  review of the overlap between
cultivated and  orchard crop use and the occurrence of the CTS (or its critical habitat) suggests
a much smaller amount of spatial overlap currently occurs (Figure E.7 of Appendix E).
Specifically, most of the CTS habitat is on the fringe of the cultivated crop/orchard land use
area of the San Joaquin valley.  However, due to the documented long-range transport of
                                        188

-------
endosulfan to the Sierra Nevada mountains of California (e.g., Fellers et al., 2004), the critical
habitat east of the cultivated crop/orchard use area may be subject to endosulfan exposure.
QH nnn -,


70,000
§" 60,000
Q 50,000
Jfl
3 40,000
j? 30,000
1 20,000
10,000






5





c





3,05





:RL


7


3




0.25 1,120 o.OO 4.56
I I " " I I I
F SFGS SMHM BOB VELB £
Species

1,10








SJKf

	 7
4








r

4,19









CTS

g. 	











Figure 5.7 Average Endosulfan Use (2005-2006) in Counties with Reported Occurrence of Listed Species,
Occupied Core Areas and/or Critical Habitat

Direct Exposure of Aquatic and Terrestrial Phase CTS

The RQ values representing exposures of endosulfan to freshwater fish and aquatic-phase
amphibians that described the direct effects to aquatic-phase CRLF are also used to represent
exposures of endosulfan to the aquatic-phase CTS. As described in Section 5.2.1.1, the
Agency determined a reasonable potential for direct effects on freshwater fish and aquatic-
phase amphibians based on LOG exceedences from model- and monitoring-based RQs, the
high probability of individual mortality, and ecological incident reports. Similarly, the
Agency determined a reasonable potential exists for direct effects on birds (surrogate for
terrestrial phase amphibians) based on exposure to food items receiving direct deposition
from endosulfan application (i.e., T-REX, T-HERPS modeling) and consumption of
contaminated aquatic prey (i.e., KABAM bioaccumulation modeling for small CRLF which
was used as a surrogate for estimating CTS exposure potential).

Consumption of Contaminated Terrestrial Prey

The CTS is also known to consume worms as part of its diet.  The T-REX model is useful for
assessing exposures of terrestrial animals to pesticides through consumption of foliar surfaces
of crops and insects on the treated site.  The model cannot be used to assess pesticide
exposures to terrestrial animals resulting from consumption of soil dwelling invertebrates
which have accumulated the pesticide in their tissues. In order to explore the potential
exposures of terrestrial-phase CTS to total residues of endosulfan that have accumulated in
earthworms inhabiting endosulfan-treatment sites, a simple fugacity approach was  employed
to estimate endosulfan concentrations in earthworms based on predicted concentrations in
soil. As with the T-REX modeling, birds were used as a surrogate for terrestrial-phase CTS.
Therefore, application rates, the foliar dissipation rate, toxicity data, body weights and
                                        189

-------
consumption weights were identical to that used for the avian T-REX modeling described in
Section 5.1. A log Kow for endosulfan of 4.76 was used based on the weighted mean Kow
for alpha and beta isomers of 70:30, respectively (MRID 414215-01). A soil distribution
coefficient (Kd) of 69 cm3/g was used based on the weighted mean Kd for alpha and beta
isomers of 70:30, respectively (MRID 414129-06). This Kd represents the lower range of
four Kd values available for different soils in this study, wherein an upper range of Kd were
320 and 428 cm3/g for alpha and beta endosulfan was reported. Using the lower value for Kd
represents a conservative assumption since for a given Kow, concentrations in earthworms
increase with decreasing Kd. For this analysis, the two crop exposure scenarios that produce
the highest and lowest dietary EECs (Citrus/Fruits and Eggplant, respectively) resulting from
soil exposure via fugacity modeling were evaluated in order to provide information on the
overall range of potential risks to terrestrial-phase CIS consuming contaminated earthworms.
Risks were evaluated for the 20g and lOOg avian size classes, which bracket the estimated
mass of adult CTS (50g).  Details on this approach are provided in Appendix N.

PRZM estimated endosulfan concentrations in soil and soil pore water of 22.4 and 0.183 g/m3,
respectively, for the Citrus and 5.3 and 0.066 g/m3, respectively, for the Eggplant crop
exposure scenarios. Based on these concentrations, the estimated concentration of endosulfan
in earthworms  is 223  and 64 ppm for Citrus and Eggplant, respectively (Table 5.22). These
values translate into dose-based EECs that range from 41.7 to 254 mg/kg-bw, depending on
body weight and the crop exposure scenario.  Acute dose-based RQs (2.3-17.5) exceed the
avian, acute listed species LOG of 0.1 for both crop exposure scenarios, regardless of body
weight. Acute dietary-based RQs (0.08-0.28) exceed the avian, acute listed LOG for the Citrus
crop scenario but approach the LOG for the Eggplant scenario.  Finally, chronic dietary-based
RQs (2.1-7.4) exceed the LOG of 1 for both crop scenarios. Sensitivity analysis indicates that
LOCs would be exceeded even if the higher Kd values were selected. Therefore, based on the
assumptions that: (1)  the sensitivity and exposures to small and medium size birds are
reasonable surrogates for the CTS and (2) the CTS receives 100% of its diet from
earthworms, there is  a potential for direct effects on terrestrial-phase CTS from the
consumption of contaminated earthworms from fields treated with endosulfan. It is noted that
even if the dose of endosulfan to the CTS was reduced to 50% of current estimate, (i.e.,
through lower earthworm  consumption rates or lower concentrations), exceedence of acute
and chronic LOG would still occur.
Table 5.22 Estimated concentrations of total endosulfan in earthworms and avian risk quotients for two
crop exposure scenarios.
Results*
Dose-based EEC (1)
(mg/kg-bw)
Acute Dose-based RQ (2'3)
Diet-based EEC (n)
(ppm w.w. in earthworm)
Acute Dietary-based RQ (2 4)
Chronic Dietary-based RQ (25)
Eggi
20g
73.1
5.0
3lant
100g
41.7
2.3
64
0.08
2.1
Citrus
20g
254
17.5
100g
145
7.8
223
0.28
7.4
                                       190

-------
 * Birds are used as a surrogate for terrestrial-phase amphibians
 (1) Based on fugacity modeling for earthworm using total endosulfan concentrations in soil and soil
 pore water of 22.4 and 0.183 g/m3, respectively for Citrus and 5.3 and 0.066 g/m3, respectively, for
 Eggplant exposure scenarios; See Appendix N for details.
 (2) RQ values in bold exceed the acute listed species LOG of 0.1 or the chronic LOG of 1.0
 (3) Acute dose-based RQs calculated using the  dose-based EECs and adjusted, acute dose-based
 toxicity values  of 14.5 and 18.5 mg/kg-bw for the 20g and 100g bird, respectively. Unadjusted LD50
 = 28 mg/kg-bw based on mallard duck (MRID 136998).
 (4) Acute diet-based RQs calculated using diet-based EECs and a diet-based LC50 of 805 ppm for
 bobwhite quail (MRID 22923)
 (5) Chronic diet-based RQs calculated using diet-based EECs and an  avian dietary NOAEC of 30
 ppm for mallard duck (MRID: 40335001)	
Spatial Overlap

For the CIS, a relatively small amount of overlap was identified based on GIS mapping of the
occurrence of CTS or its critical habitat and cultivated crop/orchard use areas (Figure E.7 of
Appendix E), suggesting a potential for spatial overlap in endosulfan use and subsequent
exposure. Therefore, assuming that the sensitivity of tested freshwater fish species reasonably
approximates the distribution of sensitivities of aquatic-phase CTS, there appears to be a
reasonable potential for direct effects to the aquatic phase CTS as result of labeled endosulfan
use in California, particularly for those areas where crop/orchard sites, endosulfan use, and
CTS occurrence overlap.

5.2.2.2 Indirect Effects (via Reductions in Prey Base)

Potential forage items of aquatic phase CTS includes algae, snails, zooplankton, small
crustaceans, aquatic larvae, aquatic invertebrates, and smaller tadpoles of Pacific tree frogs,
CRLF, and toads. Forage items for the terrestrial phase include terrestrial invertebrates,
insects, frogs, and worms

Freshwater Fish and Aquatic-phase Amphibians

RQ values representing exposures of freshwater fish and aquatic-phase amphibians to
endosulfan that may serve as prey for the aquatic-phase CRLF are also used to represent
exposures to freshwater fish and aquatic-phase amphibians that may serve as prey for the
CTS. As described in Section  5.2.1.2, the Agency determined a reasonable potential for
direct effects on freshwater fish and aquatic-phase amphibians based on RQ exceedences of
listed LOCs for freshwater fish and aquatic phase amphibians using modeled and monitoring
data-based EECs, the high probability of an individual mortality occurrence for all crop use
scenarios, the high percentage of freshwater fish that would be affected at model-based EECs
and the reported number of probable and highly probable aquatic incidents involving
endosulfan and fish.

For the CTS, a relatively small amount of overlap was identified based on GIS mapping of the
occurrence of CTS or its critical habitat and cultivated crop/orchard use areas (Figure E.7 of
Appendix E), suggesting a potential for spatial overlap in endosulfan use and subsequent
                                        191

-------
exposure. Therefore, assuming that the sensitivity of tested freshwater fish species reasonably
approximates the distribution of sensitivities of freshwater fish and aquatic-phase amphibians
used as prey by the CTS, there appears to be a reasonable potential for indirect effects to the
aquatic phase CTS as result of labeled endosulfan use in California, particularly for those
areas where crop/orchard sites, endosulfan use, and CTS occurrence overlap.

Freshwater Invertebrates

RQ values representing exposures of endosulfan to freshwater invertebrates that may serve as
prey for the aquatic-phase CRLF are also used to represent exposures of endosulfan to
freshwater invertebrates that may serve as prey for the CTS. As described in Section 5.2.1.2,
the Agency determined that based on the exceedances of non-listed species acute  and chronic
risk LOCs for freshwater invertebrates for all of the modeled scenarios assessed and the high
probability of an individual mortality occurrence using both the highest and lowest acute RQ
for freshwater invertebrates, there appears to be a potential for direct effects to freshwater
invertebrates as result of labeled endosulfan use in California.

For the CTS, a relatively small amount of overlap was identified based on GIS mapping of the
occurrence of CTS or its critical habitat and cultivated crop/orchard use areas (Figure E.7 of
Appendix E), suggesting a potential for spatial overlap in endosulfan use and subsequent
exposure. Therefore, assuming that the sensitivity of tested freshwater invertebrate species
reasonably approximates the distribution of sensitivities of freshwater invertebrates and used
as prey by the CTS, there appears to be a reasonable potential for indirect effects to the
aquatic phase CTS as result of labeled endosulfan use in California, particularly for those
areas where crop/orchard sites, endosulfan use, and CTS occurrence overlap.

Terrestrial Invertebrates

RQ values representing exposures of endosulfan to terrestrial invertebrates that may serve as
prey for the terrestrial-phase CRLF are also used to represent exposures of endosulfan to
terrestrial invertebrates that may serve as prey for the CTS.  As described in Section 5.2.1.2,
the Agency determined that based on the fact that endosulfan is a highly efficacious broad
spectrum insecticide, exceedances of the Agency's interim LOG for listed terrestrial
invertebrates for all of the modeled scenarios assessed, the high percentage of tested terrestrial
invertebrate species that would be affected at the modeled EECs and the high probability of
an individual mortality occurrence based on the RQs for terrestrial invertebrates,  there
appears to be a reasonable potential for direct effects to terrestrial invertebrates as a result of
labeled endosulfan use in California.

For the CTS, a relatively small amount of overlap was identified based on GIS mapping of the
occurrence of CTS or its critical habitat and cultivated crop/orchard use areas (Figure E.7 of
Appendix E), suggesting a potential for spatial overlap in endosulfan use and subsequent
exposure. Therefore, assuming that the sensitivity of tested terrestrial invertebrate species
reasonably approximates the distribution of sensitivities of terrestrial invertebrates and used as
prey by the CTS, there appears to be a reasonable potential for indirect effects to the terrestrial
                                         192

-------
phase CIS as result of labeled endosulfan use in California, particularly for those areas where
crop/orchard sites, endosulfan use, and CTS occurrence overlap.

Aquatic and Terrestrial Plants

As noted in Section 5.2.1.3, no data are available to reliably, quantitatively evaluate the
effects and the potential risks of endosulfan to vascular aquatic plants. No incidents have been
reported to the Agency that involve any aquatic plants, despite that it is regularly directly
applied on or near a very wide variety of agricultural plants. Only one reported incident
involved plants and this was classified as 'possible,' but the plant species was not identified
(Section 4.4 and Appendix L). No data on the effects of endosulfan on terrestrial plants
were considered acceptable for quantitative use in this risk  assessment. Based on efficacy
studies, available information indicates that agricultural crops are not adversely affected by
endosulfan use rates that are associated with pest control. This, in conjunction with the
insecticidal mode of action of endosulfan, suggests that terrestrial plants may not be sensitive
to endosulfan compared to aquatic and terrestrial animals. However, use of efficacy studies
can be confounded by lack of experimental controls, multiple stressors (pesticide and insect
pest), and plant species that may not be representative of those upon which the CTS may
depend.  Therefore, although effects to aquatic and terrestrial plants cannot be quantified due
to the lack of data, the available lines of evidence provide no compelling reason to believe
that there is a potential for indirect effects to the CTS from  loss of plant food items as result of
labeled endosulfan use in California.

5.2.2.3 Modification to Designated Critical Habitat
The primary constituent elements  (PCEs) of the CTS critical habitat include:
       •    (PCE1) Standing bodies of fresh water, including natural and man-made (e.g.,
           stock) ponds, vernal pools, and  dune ponds, and other ephemeral or permanent
           water bodies that typically become inundated during winter rains and hold water
           for a sufficient length of time (i.e., 12 weeks) necessary for the species to complete
           the aquatic (egg and larval) portion of its life cycle.
       •    (PCE2) Barrier-free uplands adjacent to breeding ponds that contain small
           mammal burrows. Small mammals are essential in creating the underground
           habitat that juvenile and adult California tiger salamanders depend upon for food,
           shelter, and protection  from the elements and predation.
       •    (PCES) Upland areas between breeding locations (PCE 1) and areas with small
           mammal burrows (PCE 2) that allow for dispersal among such sites.

Aquatic (Vascular and Non-vascular) and Terrestrial Plants

Aquatic plants serve several important functions in aquatic  ecosystems such as primary
production (non-vascular, vascular) and refugia structure (vascular plants). Rooted plants help
reduce sediment loading and provide stability to nearshore areas and lower streambanks. In
addition, vascular aquatic plants are important as attachment sites for egg masses of aquatic
species. Terrestrial plants serve several important functions such as providing habitat and
                                        193

-------
cover for invertebrate and vertebrate prey items of the CIS. Terrestrial plants also provide
energy to the terrestrial ecosystem through primary production. Upland vegetation including
grassland and woodlands provides cover during dispersal. Riparian vegetation helps to
maintain the integrity of aquatic systems by providing bank and thermal stability, serving as a
buffer to filter out sediment, nutrients, and contaminants before they reach the watershed, and
serving as an energy source.

As stated in Section 5.2.1.3 covering the evaluation of potential indirect effects to the CRLF
via habitat effects, although effects to vascular aquatic and terrestrial plants cannot be
quantified due to the lack of data, available lines of evidence provide no compelling reason to
believe that endosulfan will affect any type of plants to the extent that  it would affect the
habitat integrity of the CIS. Labeled endosulfan use in California appears not to indirectly
affect the CIS via impacts to habitat and/or primary production. Please refer to Section
5.2.1.3 for additional discussion and characterization of potential risks to plants that may be
foraged upon by the CTS.

5.2.3 San  Francisco Garter Snake

5.2.3.1 Direct Effects

Endosulfan is currently registered for numerous diverse agricultural uses that span a large
variety of geographical regions throughout the entire state of California. Therefore, there is
the potential for endosulfan use across the state to spatially and temporally coincide with the
SFGS breeding season.

Temporal and Spatial Overlap

SFGSs mate in the spring (March and April) and fall (September through November), with
mating being heavily concentrated in the first few warm days of March. Female  SFGS can
store the male's sperm over the winter and can retain viable sperm for periods ranging from 3
to 53 months. Ovulation in the common garter snake typically occurs in late spring with
pregnancy occurring in early summer. The young are typically born about three to four
months after successful mating. SFGS are ovoviviparous, and females give birth from June
through September with young typically born  in July or August; however, young can be born
as late as early September. Typically, neonate  snakes, 18 to 20 cm in length, are born in the
upland areas near the aquatic feeding habitats  and disperse immediately after they are born.
                                        194

-------
Figure 5.1 and Figure 5.8 show the extent that endosulfan use and SFGS sensitive life stages
(young juveniles) could potentially overlap. Endosulfan application from May - October
coincides with young juvenile birth.


J


F


Breeding/Egg
Masses
M
A


M
Young Juveniles (Neonates)

J

J

A


Breeding/Egg Masses
S
o
N


D
Figure 5.8 SFGS Reproductive Events by Month; Adults can be present all year
All known populations of the San Francisco garter snake occur in San Mateo County near
freshwater marshes, ponds, and slow-moving streams along the coast. Since pesticide use
information is only available at the county level, the degree that the SFGS and endosulfan
actually coincide cannot be precisely determined.  The spatial overlap of endosulfan use and
SFGS distribution is summarized at the county level in Table 5.7 and Figure 5.7. Only 0.25
Ib ai were reported used from 2005-2006 that coincided with San Mateo County, less than
0.001% of the total endosulfan use. Thus, the prospect of spatial overlap between current
endosulfan use and SFGS occurrence appears negligible. The long-range transport of
endosulfan would extend exposure beyond current crop/orchard/vineyard areas in other
counties, however, the prevailing winds would likely lead to dispersal eastward, away from
SFGS occurrence regions in San Mateo County. If cultivated crop/orchard/vineyard uses are
considered areas of potential endosulfan use in the future, then a substantial amount of spatial
overlap can be observed relative to the total area of SFGS occurrence sections (Figure E.2 of
Appendix E)

Direct Exposure

Direct acute and chronic exposures of the SFGS were evaluated using the same approaches
employed for estimating direct exposures to the terrestrial-phase CRLF (Section 5.2.1.1). In
addition, toxicity estimates for both listed species, the terrestrial-phase CRLF and the SFGS,
are based on the same surrogate avian toxicity data.  Therefore, RQ values representing the
potential for direct exposures and effects of endosulfan to the terrestrial-phase CRLF, are also
used to represent the potential for direct exposures and effects of endosulfan to the SFGS.

Based on the risk characterization presented in Section 5.2.1.1 for terrestrial-phase CRLF,
there is potential for labeled endosulfan use to cause direct adverse effects to the SFGS via
chronic toxicity based on the weight-of-evidence involving acute and chronic risk LOG
exceedances with T-REX and  the refined T-HERPS model. However, the average amount of
current endosulfan use at the county level and SFGS distribution do not coincide, and direct
effects to the SFGS appear to be unlikely based on current endosulfan use patterns (Figure
5.7). Although the potential for endosulfan to be transported long distances from the
application site could extend exposure well beyond current use  areas, such long-range
transport would need to overcome the direction of the prevailing winds (west to east) in order
to reach SFGS occurrence sections in San Mateo County, which appears unlikely. Refer to
Section 3.2.4.4 for more information on long-range transport of endosulfan. If cultivated
crop/orchard/vineyard uses are considered areas of potential endosulfan use in the future, then
                                        195

-------
a substantial amount of spatial overlap can be observed relative to the total area of SFGS
occurrence sections (Figure E.2 of Appendix E). Therefore, based on the potential future use
of endosulfan in cultivated crop/orchard/vineyard areas of San Mateo County, there appears to
be a reasonable potential for direct effects on SFGS from labeled uses of endosulfan.

              5.2.3.2 Indirect Effects (via Reductions in Prey Base)

Newborn and juvenile SFGS prey almost exclusively on Pacific tree frogs in temporary pools
during the spring and early summer to the point that the SFGS may be so dependent on their
anuran prey that they are not able to switch to other available prey sources if necessary to
survive. SFGS under 500 mm snout-to-vent length (SVL) require Pacific tree frogs in various
stages of metamorphosis, whereas individuals over 500 mm SVL can consume Pacific tree
frog, CRLF, and bullfrog tadpoles and adults.

The main diet of adult SFGS consists of CRLF. Adult SFGSs may also feed on  smaller
juvenile non-native bullfrogs (Rana catesbeiana).  Immature California newts (Taricha
torosa), California toads (Bufo boreas halophilus)., recently metamorphosed western toads
(Bufo boreas), threespine stickleback (Gasterosteus aculeatus), and non-native mosquito fish
(Gambusia affinis) are also known to be consumed by SFGS.  Small mammals,  reptiles,
amphibians, possibly invertebrates, and some fish species may also be consumed by the
SFGS.
Freshwater Fish and Aquatic-phase Amphibians
RQ values representing exposures of endosulfan to freshwater fish and aquatic-phase
amphibians that may serve as prey for the aquatic-phase CRLF are also used to represent
exposures of endosulfan to freshwater fish and aquatic-phase amphibians that may serve as
prey for the SFGS. As described in Section 5.2.1.2, the Agency determined a reasonable
potential for direct effects to freshwater fish and aquatic-phase amphibians based on RQ
exceedences of listed LOCs for freshwater fish and aquatic phase amphibians using modeled
and monitoring data-based EECs, the high probability of an individual mortality occurrence
for all crop use scenarios, the high percentage of freshwater fish that would be affected at
model-based EECs, and the reported number of probable and highly probable aquatic
incidents involving endosulfan and fish

As discussed in Section 5.2.3.1 for direct effects on the SFGS, there does appear to be a
potential for spatial overlap between SFGS distribution and cultivated/orchard/vineyard
agricultural areas where endosulfan could be used in the future, although overlap in current
use patterns appears negligible.  Therefore, based on the potential future use of endosulfan in
cultivated crop/orchard/vineyard areas of San Mateo County, there appears to be a reasonable
potential for indirect effects on the SFGS from loss of freshwater fish and amphibian prey
items.

Freshwater Invertebrates
RQ values representing exposures of endosulfan to freshwater invertebrates that may serve as
prey for the aquatic-phase CRLF are also used to represent exposures of endosulfan to
freshwater invertebrates that may serve as prey for the SFGS. As described in Section
                                        196

-------
5.2.1.2, the Agency determined that based on the exceedances of non-listed species acute and
chronic risk LOCs for freshwater invertebrates for all of the modeled scenarios assessed and
the high probability of an individual mortality occurrence using both the highest and lowest
acute RQ for freshwater invertebrates, there appears to be adequate lines of evidence to
conclude there is a potential for direct effects to freshwater invertebrates as result of labeled
endosulfan use in California.

As discussed in Section 5.2.3.1 for direct effects on the SFGS, there does appear to be a
potential for spatial overlap between SFGS distribution and cultivated/orchard/vineyard
agricultural areas where endosulfan could be used in the future, although overlap in current
use patterns appears negligible. Therefore, based on the potential future use of endosulfan in
cultivated crop/orchard/vineyard areas of San Mateo County, there appears to be a reasonable
potential for indirect effects on the SFGS from loss of freshwater invertebrate prey items.

Terrestrial Invertebrates

RQ values representing exposures of endosulfan to terrestrial invertebrates that may serve as
prey for the terrestrial-phase CRLF are also used to represent exposures of endosulfan to
terrestrial  invertebrates that may serve as prey for the SFGS. As described in Section 5.2.1.2,
the Agency determined that based on the fact that endosulfan is a highly efficacious broad
spectrum insecticide, exceedances of the Agency's interim LOG for listed terrestrial
invertebrates for all of the modeled scenarios assessed, the high percentage of tested terrestrial
invertebrate species that would be affected at the modeled EECs, the high probability of an
individual mortality occurrence based on the RQs for terrestrial invertebrates, there appears to
be a  reasonable potential for direct effects on terrestrial invertebrates from the labeled
endosulfan use in California.

As discussed in Section 5.2.3.1 for direct effects on the SFGS, there does appear to be a
potential for spatial overlap between SFGS distribution and cultivated/orchard/vineyard
agricultural areas where endosulfan could be used in the future, although overlap in current
use patterns appears negligible. Therefore, based on the potential future use of endosulfan in
cultivated crop/orchard/vineyard areas of San Mateo County, there appears to be a reasonable
potential for indirect effects on the SFGS from loss of terrestrial invertebrate prey items.

Small Terrestrial Vertebrates
RQ values representing exposures of endosulfan to small terrestrial vertebrates that may serve
as prey  for the terrestrial-phase CRLF are also used to represent exposures of endosulfan to
terrestrial  small terrestrial  vertebrates that may serve as prey for the  SFGS. Similarly, the
Agency determined a reasonable potential exists for direct effects on birds (surrogate for
terrestrial  phase amphibians) based on exposure to food items receiving direct deposition
from endosulfan application (i.e., T-REX, T-HERPS modeling); consumption of
contaminated aquatic prey (i.e., KABAM bioaccumulation modeling), and consumption of
contaminated terrestrial prey (i.e., earthworm fugacity modeling; Section 5.2.1.1).

As discussed in Section 5.2.3.1 for direct effects on the SFGS, there does appear to be a
potential for spatial overlap between SFGS distribution and cultivated/orchard/vineyard
                                         197

-------
agricultural areas where endosulfan could be used in the future, although overlap in current
use patterns appears negligible. Therefore, based on the potential future use of endosulfan in
cultivated crop/orchard/vineyard areas of San Mateo County, there appears to be a reasonable
potential for indirect effects on the SFGS from loss of small terrestrial vertebrate prey items.

              5.2.3.3 Indirect Effects (via Habitat Effects)

SFGS inhabit areas near densely vegetated ponds and in open hillsides where it can sun, feed,
and find cover in rodent burrows. It forages extensively in aquatic habitats. Fresh-water
habitats including natural and manmade (e.g.  stock) ponds, slow moving streams, vernal pools
and other ephemeral or permanent water bodies which typically support inundation during
winter rains and hold water for a minimum of 12 weeks in a year of average rainfall and
upland habitats within 200 ft of the mean high water mark of such aquatic habitat.

Aauatic (Vascular and Non-vascular) and Terrestrial Plants
Aquatic plants serve several important functions in aquatic ecosystems such as primary
production (non-vascular, vascular) and refugia structure (vascular plants). Rooted plants help
reduce sediment loading and provide stability to nearshore areas and lower streambanks. In
addition, vascular aquatic plants are important as attachment sites for egg masses of aquatic
species.

Terrestrial plants serve several important habitat-related functions for the SFGS. In addition
to providing habitat and cover for invertebrate and vertebrate prey items of the SFGS,
terrestrial vegetation also provides shelter for the SFGS and cover from predators while
foraging. Terrestrial plants also provide energy to the terrestrial ecosystem through primary
production.  Upland vegetation including grassland and woodlands provides cover during
dispersal. Riparian vegetation helps to maintain the integrity of aquatic systems by providing
bank and thermal stability, serving as a buffer to filter out sediment, nutrients, and
contaminants before they reach the watershed, and serving as an energy source.

As noted in Section 5.2.1.3, no data are  available to reliably, quantitatively evaluate the
effects and the potential risks of endosulfan to vascular aquatic plants. No incidents have been
reported to the Agency that involve any  aquatic plants, despite that it is regularly directly
applied on or near a very wide variety of agricultural plants. Only one reported incident
involved plants and this was classified as 'possible,' but the plant species was not identified
(Section 4.4 and Appendix L). No data on the effects of endosulfan on terrestrial plants
were considered acceptable for quantitative use in this risk assessment.  Based on efficacy
studies, available information indicates that agricultural crops are not adversely affected by
endosulfan use rates that are associated with pest control. This, in conjunction with the
insecticidal mode of action of endosulfan, suggests that terrestrial plants may not be sensitive
to endosulfan compared to aquatic and terrestrial animals. However, use of efficacy studies
can be confounded by lack of experimental controls, multiple stressors (pesticide and insect
pest), and plant species that may not be representative of those upon which the SFGS may
depend. Although effects to aquatic and terrestrial plants cannot be quantified due to the lack
of data, the available lines of evidence provide no compelling reason to  believe that there is a
                                         198

-------
potential for indirect effects to the SFGS from loss of plant food items as result of labeled
endosulfan use in California.

5.2.4 Salt Marsh Harvest Mouse

              5.2.4.1 Direct Effects

Endosulfan is currently registered for numerous diverse agricultural uses that span a large
variety of geographical regions throughout the entire state of California (Figure 2.6 and
Figure 2.5). Therefore, there is the potential for endosulfan use across the state to spatially
and temporally coincide with the SMHM breeding season.

The SMHM is a small, mostly nocturnal rodent that lives in tidal and diked salt marshes, only
around the San Francisco Bay and its tributaries. SMFDVI depend on dense, perennial cover
and prefer habitat in the middle and upper parts of the marsh dominated by pickleweed and
peripheral halophytes as well as similar vegetation in diked wetlands adjacent to the Bay. Salt
marsh harvest mice are cover-dependent species and only live under thick vegetation. Salt
marsh harvest mice use pickleweed (Salicornia virginicd) as their primary/preferred habitat as
long as they have non-submerged, salt-tolerant vegetation for escape during the highest tides.
Their diet consists of seeds, grasses, forbs and insects.

Temporal and Spatial Overlap

The SMFDVI breeds from spring through the fall and there is strong potential for periods of
breeding activity to overlap temporally with peak usage of endosulfan from May through
September (Figure 5.1). While male SMFDVI are described as reproductively active from
April through September, with some active throughout the year, female SMFDVI have a long
breeding season that extends from as early as March to November. In general, the northern
subspecies of SMFDVI breeds from May to November and the southern subspecies breeds from
March to November. Despite the long breeding season, the SMFDVI is characterized as having
a low reproductive potential, with each female typically having only one or two litters per
year with an average litter size of about three or four.

Since pesticide use information is only available at the county level, the degree that the
SMFDVI actually coincides in the  same or similar locations cannot be precisely determined.
The spatial overlap of endosulfan use and SMFDVI distribution is summarized at the county
level in Table 5.7 and Figure 5.7). Approximately 1.25% of endosulfan use (1,120 Ibs) is
applied in counties in which the SMHM occurs. These counties represent approximately 66%
of the SMFDVI range by county.  Thus, the potential for direct effects to the SMHM appears to
be low based on current endosulfan use patterns. The potential  for endosulfan to be
transported long distances from the application site could extend exposure well beyond
current use areas, which appears to be most relevant for SMHM occurrence sections in Solano
County (where current cultivated crop areas  are west of the occurrence sections).  Refer to
Section 3.2.4.4 for more information on long-range transport of endosulfan. If cultivated
crop/orchard/vineyard uses are considered areas of potential endosulfan use in the future, then
                                        199

-------
a very small amount of spatial overlap can be observed relative to the total area of SMHM
occurrence sections (Figure E.3 of Appendix E).

Direct Exposure: Deposition on Forage Items

Direct exposure of the SMHM and the resulting risks in terrestrial environments were
evaluated based on dose- and dietary-based EECs estimated using two approaches: (1) T-REX
for foliar spray applications and (2) fugacity modeling for predicting endosulfan
contamination of terrestrial prey.  For dose-based exposure calculations, a body weight of 15
g was used for small mammals which correspond to the body weight of the SHMH mouse (8-
14 g). All estimated EECs  (i.e., EECs for short grass, tall grass, broadleaf plants/small insects,
fruits/pods/seeds/large insects) were considered relevant for evaluation of direct effects
because the SMHM has been known to feed on leaves,  seeds, plant stems, insects, and
grasses.

Based on surrogate toxicity data (LD50=10 mg/kg-bw and NOAEL = 15 mg/kg-diet), the
maximum allowable application rate (1 application, 2.5 Ibs a.i./acre/application), the foliar
dissipation half-life of 4 days for endosulfan from Willis and McDowell (1987), and upper
bound Kenaga values from T-REX, there is a potential for direct adverse effects on SMHM
individuals from foliar spray applications of endosulfan in CA (see Table 5.4). Although
EECs for small mammals feeding  on short grass yield the highest risk estimates, it appears
that the potential impact of endosulfan use to the SMHM potentially extend beyond those
feeding on  short grass.

For spray applications of endosulfan, the dose-based acute RQs for 15 g mammal range from
5.21 to 26.03, from 2.39 to 11.93,  from 2.93 to 14.64, and from 0.33 to 1.63 for the short
grass, tall grass, broadleaf plants/small insects, and fruits/pods/seeds/large insects dietary
categories,  respectively  (Table 5.4).For  acute dose-based RQs, all modeled scenarios (N=20)
resulted in an exceedance of the Agency's acute listed species LOG (RQ>0.1) for mammals
that feed on short grass, tall grass,  broadleaf plants/small insects,  and fruits/pods/seeds/large
insects.

The dose-based chronic RQs ranged from 69.41 to 347.04, from 31.81 to 159.06, from 39.04
to 195.21, and from 4.34 to 21.69, for the short grass, tall grass, broadleaf plants/small insects,
and fruits/pods/seeds/large insect dietary categories, respectively  (Table 5.4). For chronic
dose-based RQs, all modeled scenarios (N=20) resulted in an exceedance of the Agency's
acute listed species LOG (RQ>1) for mammals that feed on short grass, tall grass, broadleaf
plants/small insects, and fruits/pods/seeds/large insects.

The dietary-based chronic  RQs range  from 8 to 40.00, from 3.67 to  18.33, from 4.50 to 22.50,
and from 0.50 to 2.50, for the short grass, tall grass, broadleaf plants/small insects, and
fruits/pods/seeds/large insects dietary  categories, respectively (Table 5.4). For chronic
dietary-based RQs, all modeled scenarios (N=20) resulted in an exceedance of the Agency's
acute listed species LOG (RQ>1) for mammals that feed on short grass, tall grass, and
broadleaf plants/small insects. Thirteen out of 20 modeled scenarios resulted in an exceedance
of the Agency's chronic risk LOG  (RQ>1) for mammals that feed on fruits/pods/seeds/large
                                        200

-------
insects. RQs exceed the Agency's acute listed species LOG (RQ>0.1) and chronic risk LOG
(RQ>1) for mammals for foliar spray applications of endosulfan.
Direct Exposure: Contaminated Soil

In addition to the consumption of food items subject to direct deposition of endosulfan
applied to agricultural areas discussed above, the potential exists for direct effects to the
SMHM via soil contaminated with endosulfan. In order to explore the potential exposures of
SMHM to total residues of endosulfan that have accumulated in earthworms inhabiting
endosulfan-treatment sites, a simple fugacity approach was employed to estimate endosulfan
concentrations in earthworms and subsequent exposures to the SMHM consuming
earthworms. Pesticide application rates, the foliar dissipation rate, toxicity data, body weights
and consumption weights were identical to that used for the mammalian T-REX modeling
described in Section 5.1. A log Kow for endosulfan of 4.76 was used based on the weighted
mean Kow for alpha and beta isomers of 70:30, respectively (MRID 414215-01). A soil
distribution coefficient (Kd) of 69 cm3/g was used based on the weighted mean Kd for alpha
and beta isomers of 70:30, respectively  (MRID 414129-06). This Kd represents the lower
range of four Kd values available for different soils in this study, wherein an upper range of
Kd were 320 and 428 cm3/g for alpha and beta endosulfan was reported.  Using the lower
value for Kd represents a conservative assumption since for a given Kow, concentrations in
earthworms increase with decreasing Kd.  For this analysis, the two crop exposure scenarios
that produce the highest and lowest dietary EECs  (Citrus/Fruits and Eggplant, respectively)
were evaluated in order to provide information on the overall range of potential risks to
SMHM consuming contaminated earthworms. Risks were evaluated for the 15g mammalian
size classes (Table 5.23; the lOOOg size class was included for evaluating risk to the SJKF
which is discussed in Section 5.2.5). Details on this approach are provided in Appendix N.
Results presented in Table 5.23 indicate dose-based RQs exceed the Agency's listed acute a
LOG for  15g mammal even when using the exposure scenario with the lowest soil EECs
(eggplant). Furthermore, chronic dose- and dietary-based RQs exceed the chronic LOG for
the eggplant exposure scenario.
Table 5.23 Estimated concentrations of total endosulfan in earthworms and mammalian risk quotients
for two crop exposure scenarios
Results
Dose-based EEC (1)
(mg/kg-bw)
Acute Dose-based RQ (2'3)
Chronic Dose-based RQ (24)
Diet-based EEC l"
(ppm w.w. in earthworm)
Chronic Dietary-based RQ (25)
Eggi
15g<6)
61.2
2.8
37.1
3lant
1000g(6)
9.8
1.3
17.0
64
4.3
Citrus
15g<6)
212.7
9.7
129.0
1000g(6)
34.1
4.4
59.08
64
4.3
(1) Based on fugacity modeling for earthworm using total endosulfan concentrations in soil and soil
pore water of 22.4 and 0.183 g/m3, respectively for Citrus and 5.3 and 0.066 g/m3, respectively, for
Eggplant exposure scenarios; See Appendix N for details.
(2) RQ values in bold exceed the acute listed species LOG of 0.1 or the chronic LOG of 1.0
(3) Acute dose-based RQs calculated using the dose-based EECs and adjusted, acute dose-based
                                       201

-------
 Table 5.23 Estimated concentrations of total endosulfan in earthworms and mammalian risk quotients
 for two crop exposure scenarios
                                      Eggplant
Citrus
 toxicity values of 21.98 and 7.69 mg/kg-bw for the 15g and 1000g mammal, respectively.
 Unadjusted LD50 = 10 mg/kg-bw based on rat (MRID 00038307).
 (4) Chronic dose-based RQs calculated using dose-based EECs and an adjusted, dose-based
 chronic NOAEC of 1.65 mg/kg-d for 15g mammals and 0.58 for 1000g mammals; Unadjusted dose-
 based NOAEC for rat calculated as 0.75 mg/kg-bw for the rat(MRID 00148264)
 (5) Chronic diet-based RQs calculated using diet-based EECs and a mammalian dietary NOAEC of
 15 ppm for rat (MRID: 00148264)
 (6) 15 g mammal used for estimating risks to SMHM; 10OOg used for SJKF	
In conclusion, based on:
    1.  exceedances of the Agency's listed species acute risk LOG (RQ>0.1) and chronic risk
       LOG (RQ>1) for small mammals feeding on a variety of terrestrial food items for the
       various endosulfan application scenarios considered,
    2.  the high probability of an individual mortality occurrence based on the highest acute
       RQ for mammals (~1 in 1; Table 5.12),
    3.  LOG exceedences via consumption of earthworms via soil contamination (Table
       5.23), evidence of similar sensitivity of rats and mice to endosulfan (Section 4 and
       Appendix K),
    4.  a significant degree of temporal overlap with endosulfan applications w/ SHMH life
       stages, and
    5.  some spatial overlap between SMHM occurrence sections and cultivated/
       orchard/vineyard crop use areas;

it appears there is a potential for direct effects to the SMHM as result of labeled endosulfan
use in California, although this potential is likely to be limited to small areas spatially.
              5.2.4.2 Indirect Effects (via Reductions in Prey Base)

Potential forage items of the SMHM include leaves, seeds, plant stems, and insects, although
seasonal variation has been observed in SMHM stomach contents with fresh green grasses
more prevalent in the winter, and pickleweed and saltgrass dominating during the rest of the
year.

Terrestrial Invertebrates

RQ values representing exposures of endosulfan to terrestrial invertebrates that may serve as
prey for the terrestrial-phase CRLF are also used to represent exposures of endosulfan to
terrestrial invertebrates that may serve as prey for the SMHM.  As described in Section
5.2.1.2, the Agency determined that based on the fact that endosulfan is  a highly efficacious
broad spectrum insecticide, exceedances of the Agency's interim LOG for listed terrestrial
invertebrates for all of the modeled scenarios assessed, the high percentage of tested terrestrial
invertebrate  species that would be affected at the modeled EECs, and the high probability of
an individual mortality occurrence based on the RQs for terrestrial invertebrates, there appears
                                        202

-------
to be a reasonable potential for direct effects on terrestrial invertebrates from the labeled
endosulfan use in California.

However, as discussed previously in Section 5.2.4.1 for directs effects on SMHM, the
potential overlap of current and future endosulfan use patterns with the occurrence of the
SMHM (and by extension, its terrestrial invertebrate prey) appears to be very small.
Nevertheless, in these areas of spatial overlap, there does appear to be sufficient lines of
evidence for a direct effect on the SMHM through reductions in terrestrial invertebrate prey,
assuming the sensitivity of tested terrestrial invertebrates is similar to the terrestrial
invertebrate prey being consumed by the SMHM.

Aquatic (Vascular and Non-vascular) and Terrestrial Plants

As noted in Section 5.2.1.3, no data are available to reliably, quantitatively evaluate the
effects and the potential risks of endosulfan to vascular aquatic plants or terrestrial plants.
Only one reported incident involved  plants (1014404-001, June 1, 1990 in Washington State),
but no damage was reported and the incident classified only as 'possible' (i.e., exposure
concentrations were not documented).  Although effects to vascular aquatic and terrestrial
plants cannot be quantified due to the lack of data, the available lines of evidence provide no
compelling reason to believe that there is a potential for indirect effects to the SMHM from
loss of plant food items as result of labeled endosulfan use in California.

              5.2.4.3 Indirect Effects (via Habitat Effects)

Aquatic (Vascular and Non-vascular) and Terrestrial Plants
Aquatic plants serve several important functions in aquatic ecosystems. Non-vascular aquatic
plants are primary producers and provide the autochthonous energy base for aquatic
ecosystems. Vascular plants provide structure as attachment sites and refugia for many
aquatic invertebrates, fish, and juvenile organisms, such as fish and frogs. In addition,
vascular plants also provide primary  productivity and oxygen to the aquatic ecosystem.
Rooted plants help reduce sediment loading and provide stability to nearshore areas  and lower
streambanks. In addition, vascular aquatic plants are important as attachment sites for egg
masses of aquatic species.

Terrestrial plants serve several important habitat-related functions for the SMHM. In addition
to providing habitat and cover for invertebrate prey items of the SMHM, terrestrial vegetation
also provides nesting material, shelter, and cover from predators while foraging. Terrestrial
plants also provide energy to the terrestrial ecosystem through primary production.  Upland
vegetation including grassland and woodlands provides cover during dispersal. Riparian
vegetation helps to maintain the  integrity of aquatic systems by providing bank and thermal
stability, serving as a buffer to filter out sediment, nutrients, and contaminants before they
reach the watershed, and serving as an energy  source.

As stated in Section 5.2.1.4 covering the evaluation of potential indirect effects to the CRLF
via habitat effects, although effects to vascular aquatic and terrestrial plants cannot be
quantified due to the lack of data, since the RQs do not exceed the Agency's LOG (1) for
                                         203

-------
aquatic non-vascular plants based on the most sensitive data available to the Agency, and
because available lines of evidence provide no compelling reason to believe that endosulfan
will affect any type of plants to the extent that it would affect the habitat integrity of the
SMHM, labeled endosulfan use in California appears not likely to indirectly affect the SMHM
via impacts to habitat and/or primary production. Please refer to Section 5.2.1.4 for additional
discussion and characterization of potential risks to plants that may be foraged upon by the
SMHM.

Small Terrestrial Vertebrates

SMHM do not burrow, but some winter nests may be constructed in burrows and small
crevices. SMHM nests are described as minimal, and the  SMHM may build over old birds'
nests or use nests built by Suisun shrews, after the young shrews have dispersed. Therefore,
the potential for indirect effects to the SMHM via affects to small mammals and birds that
may help to provide suitable habitat was evaluated.

RQ values representing exposures of endosulfan to small mammals that may serve as prey for
the terrestrial-phase CRLF, are also used to represent exposures of endosulfan to small
mammals that may help to provide suitable habitat (nests) for the SMHM. As described in
Section 5.2.4.1 for direct effects on the SMHM, the Agency determined that a potential for
direct effects to the SMHM exits as result of labeled endosulfan use in California, although
this potential is likely to be limited to small areas spatially. The Agency further determined
that a potential for direct effects of endosulfan use exists on birds, as described for the
terrestrial phase CRLF (Section 5.2.1.1).  Therefore, the same potential exists for habitat
modification due to loss of mammalian and avian-created  nesting sites.

       5.2.5 San Joaquin Kit Fox

              5.2.5.1 Direct Effects

Endosulfan is currently registered for numerous diverse agricultural uses that  span a large
variety of geographical regions throughout the entire  state of California. Therefore, there is
the potential for endosulfan use across the state to spatially and temporally coincide with the
SJKF breeding season.

SJKF occupies a variety of habitats, including grasslands,  scrublands (e.g., chenopod scrub
and sub-shrub scrub), vernal pool areas, oak woodland, alkali meadows and playas, and an
agricultural matrix of row crops, irrigated pastures, orchards, vineyards, and grazed annual
grasslands. Kit foxes dig their own dens, modify and use those already constructed by other
animals (ground squirrels, badgers, and coyotes), or use human-made structures (culverts,
abandoned pipelines, or banks in sumps or roadbeds). They move to new dens within their
home range often (likely to avoid predation by coyotes). The SJKF forages in California
prairie and Sonoran grasslands in the vicinity of freshwater marshes and alkali sinks, where
there is a dense ground cover of tall grasses and San Joaquin saltbush. Seasonal flooding in
such habitats is normal. It feeds on small animals including blacktailed hares,  desert
cottontails, mice, kangaroo rats, squirrels, birds, lizards, insects and grass. The San Joaquin
                                        204

-------
kit fox satisfies its moisture requirements from prey and does not depend on freshwater
sources.

Temporal and Spatial Overlap

The breeding season of the SJKF is from late December - March, with a gestation period of
48 to 52 days. Litters are born February - late March. Pups emerge from their dens at about 1-
month of age and may begin to disperse after 4-5 months usually in Aug. or Sept. Pup
dispersal coincides with the peak endosulfan application in August through September
(Figure 5.1).

The SJKF inhabits grasslands in the San Joaquin Valley and eastern Bay Area counties of
California. Since pesticide use information is only available at the county level, the degree
that the SJKF and endosulfan actually coincide cannot be precisely determined. The spatial
overlap of endosulfan use and SJKF distribution is summarized at the county level in Table
5.7. Approximately 80% of endosulfan use (71,104 Ibs) is applied in counties in which the
SJKF occurs. These counties represent approximately 73% of the SJKF range by county.
Examination of the overlap of SJKF occurrence sections and distribution records (Figure E.4
of Appendix E) with cultivated crop, orchard and vineyard areas suggests a moderate amount
of spatial overlap with potential endosulfan use areas.

Direct Exposure

Direct exposure of the SJKF and the resulting risks in terrestrial environments were evaluated
based on dose- and dietary-based EECs estimated using two approaches (i.e., T-REX for
foliar spray applications and using fugacity-based modeling for insectivorous wildlife). All
estimated EECs (i.e., EECs for short grass, tall grass, broadleaf plants/small insects,
fruits/pods/seeds/large insects) were considered relevant for evaluation of direct effects
because the SJKF has been known to feed on insects and grasses. A mammalian body weight
of lOOOg was modeled based on the adult size of the kit fox.

Based on surrogate toxicity data (LD50=10 mg/kg-bw and NOAEL = 15 mg/kg-diet), the
maximum allowable application rate (1 application, 2.5 Ibs a.i./acre/application), the foliar
dissipation half-life of 4 days for endosulfan from Willis and McDowell (1987), and upper
bound Kenaga values from T-REX, there is a potential for direct adverse effects on SJKF
individuals from foliar spray applications of endosulfan in CA (Table 5.4).

For spray applications of endosulfan, the dose-based acute RQs range from 2.38 to 11.92,
from 1.09 to 5.46, from 1.34 to 6.70, and from 0.15 to 0.74 for the short grass, tall grass,
broadleaf plants/small insects, and fruits/pods/seeds/large insect dietary categories,
respectively (Table 5.4). For acute dose-based RQs, all modeled scenarios (N=20) resulted in
an exceedance of the Agency's acute listed species LOG (RQ>0.1) for mammals that feed on
short grass, tall grass, broadleaf plants/small insects, and fruits/pods/seeds/large insects.

The dose-based chronic RQs ranged from 31.78 to 158.90, from 14.57 to 72.83, from 17.88 to
89.38, and from 1.99 to 9.93, for the short grass, tall grass, broadleaf plants/small insects, and
                                        205

-------
fruits/pods/seeds/large insect dietary categories, respectively (Table 5.4). For chronic dose-
based RQs, all modeled scenarios (N=20) resulted in an exceedance of the Agency's acute
listed species LOG (RQ>1) for mammals that feed on short grass, tall grass, broadleaf
plants/small insects, and fruits/pods/seeds/large insects.

The dietary-based chronic RQs range from 8 to 40.00, from 3.67 to 18.33, from 4.50 to 22.50,
and from 0.50 to 2.50, for the short grass, tall grass, broadleaf plants/small insects, and
fruits/pods/seeds/large insects dietary categories, respectively (Table 5.4). For chronic
dietary-based RQs, all modeled scenarios (N=20) resulted in an exceedance of the Agency's
acute listed species LOG (RQ>1) for mammals that feed on short grass, tall grass, and
broadleaf plants/small insects. Thirteen out of 20 modeled scenarios resulted in an exceedance
of the Agency's chronic risk LOG (RQ>1) for mammals that feed on fruits/pods/seeds/large
insects. Therefore, RQs for exceed the Agency's acute listed species LOG (RQ>0.1) and
chronic risk LOG (RQ>1) for mammals (see Table 5.4) for foliar spray applications of
endosulfan.

As described in Section 5.2.4.1, the potential exists for direct effects to insectivorous wildlife
(including SJKF) via soil contaminated with endosulfan. In order to explore the potential
exposures of SJKF to total residues of endosulfan that have accumulated in earthworms
inhabiting endosulfan-treatment sites, a simple fugacity approach was employed to estimate
endosulfan concentrations in  earthworms and subsequent exposures to the SJKF consuming
earthworms. Additional description of this approach is provided in Appendix N.  Results
presented in Table 5.23 indicate dose-based RQs exceed the Agency's listed acute a LOG for
a lOOOg mammal even when  using the exposure scenario with the lowest soil EECs
(eggplant). Furthermore, chronic dose- and dietary-based RQs for a lOOOg mammal exceed
the chronic LOG for the eggplant exposure scenario.

In conclusion, the Agency determines that based on:
1. Exceedances of the Agency's listed species acute risk LOG (RQ>0.1) and chronic risk
   LOG (RQ>1) for large mammals feeding on a variety of terrestrial food items for the
   various endosulfan application scenarios considered,
2. The high probability of an individual mortality occurrence based on the highest acute RQ
   for mammals (~1 in 1; Table 5.12),
3. LOG exceedences via consumption of earthworms via soil contamination (Table 5.23),
   and
4. A moderate degree of spatial and temporal overlap of SJKF occurrence and potential
   endosulfan use areas,

It appears there is a potential  for direct effects to the SJKF as result of labeled endosulfan use
in California.

              5.2.5.2 Indirect Effects (via Reductions in Prey Base)

Potential forage items of the SJKF include small mammals, grasses, and insects.
                                       206

-------
Terrestrial Invertebrates

RQ values representing exposures of endosulfan to terrestrial invertebrates that may serve as
prey for the terrestrial-phase CRLF, CIS, SFGS SMHM are also used to represent exposures
of endosulfan to terrestrial invertebrates that may serve as prey for the SFKF.  As described in
Section 5.2.1.2, the Agency determined that based on the fact that endosulfan is a highly
efficacious broad spectrum insecticide, exceedances of the Agency's interim LOG for listed
terrestrial invertebrates for all of the modeled scenarios assessed, the high percentage of tested
terrestrial invertebrate species that would be affected at the modeled EECs, and the high
probability of an individual mortality occurrence based on the RQs for terrestrial
invertebrates, there appears to be a reasonable potential for direct effects on terrestrial
invertebrates from the labeled endosulfan use in California.  Furthermore, there appears to be
a moderate amount of spatial overlap between SJKF occurrence and both current and potential
future endosulfan use areas, as described in Section 5.2.5.2. Therefore, the Agency
determines that there are sufficient lines of evidence for indirect effect on the SJKF through
reductions in terrestrial invertebrate prey, assuming the sensitivity of tested terrestrial
invertebrates is similar to the terrestrial invertebrate prey being consumed by the SLKF.

Small Mammals

RQ values representing exposures of endosulfan to small mammals that may serve as prey for
the terrestrial-phase CRLF, are also used to represent exposures of endosulfan to small
mammals that may help to provide prey for the  SJKF.  As described in Section 5.2.1.2 for the
terrestrial-phase CRLF, the Agency determined that based on the exceedances of the
Agency's listed species acute risk LOG and chronic risk LOG for small mammals feeding on
a variety of terrestrial food items for the various endosulfan seed, spray, and granular
application scenarios considered, the high probability of an individual mortality occurrence
based on the  highest acute RQ for mammals (~1 in 1), and the  relative insensitivity of risk
conclusions to selection of less conservative acute and chronic endpoints, there are adequate
lines of evidence for potential direct effects on mammals as a result of labeled endosulfan use
in California. Furthermore, there appears to be  a moderate amount of spatial overlap between
SJKF occurrence and both current and potential future endosulfan use areas, as described in
Section 5.2.5.2. Therefore, the Agency determines that there are sufficient lines of evidence
for indirect effect on the SJKF through reductions in small mammalian prey, assuming the
sensitivity of tested mammals is similar to the mammalian prey being consumed by the SLKF.

Aquatic (Vascular and Non-vascular) and Terrestrial Plants

As noted in Section 5.2.1.3, no data are available to reliably, quantitatively evaluate the
effects and the potential risks of endosulfan to vascular aquatic plants or terrestrial plants. No
incidents have been reported to the Agency that involve any plants (all terrestrial plants), and
none have reliably linked endosulfan to the observed effects with a certainty index of
"probable" or higher, despite that it is regularly  directly applied on or near a very wide variety
of agricultural and home garden plants. Therefore, although effects to vascular aquatic and
terrestrial plants cannot be quantified due to the lack of data, the available lines of evidence
                                        207

-------
provide no compelling reason to believe that there is a potential for indirect effects to the
SJKF from loss of plant food items as result of labeled endosulfan use in California.

              5.2.5.3 Indirect Effects (via Habitat Effects)

Aauatic (Vascular and Non-vascular) and Terrestrial Plants
Aquatic plants serve several important functions in aquatic ecosystems. Non-vascular aquatic
plants are primary producers and provide the autochthonous energy base for aquatic
ecosystems. Vascular plants provide structure as attachment sites and refugia for many
aquatic invertebrates, fish, and juvenile  organisms, such as fish and frogs. In addition,
vascular plants also provide primary productivity and oxygen to the aquatic ecosystem.
Rooted plants help reduce sediment loading and provide stability to nearshore areas and lower
streambanks. In addition, vascular aquatic plants are important as attachment sites for egg
masses of aquatic species.

Terrestrial plants serve several important habitat-related functions for the SJKF. In addition
to providing habitat and cover for invertebrate prey  terrestrial plants also provide energy to
the terrestrial ecosystem through primary production. Upland vegetation including grassland
and woodlands provides cover during dispersal. Riparian vegetation helps to maintain the
integrity of aquatic systems by providing bank and thermal stability, serving as a buffer to
filter out sediment, nutrients, and contaminants before they reach the watershed, and serving
as an energy source.

As stated in Section 5.2.1.3 covering the evaluation of potential indirect effects to the CRLF
via habitat effects, although effects to vascular aquatic and terrestrial plants cannot be
quantified due to the lack of data, since  the RQs do  not exceed the Agency's LOG (1) for
aquatic non-vascular plants based on the most sensitive data available to the Agency, and
because available lines of evidence provide no compelling reason to believe that endosulfan
will  affect any type of plants to the extent that it would affect the habitat integrity of the
SJKF, labeled endosulfan use in California appears  not to indirectly affect the SJKF via
impacts to habitat and/or primary production.

5.2.6 Bay Checkerspot Butterfly
       5.2.6.1 Direct Effects
The BCB's life cycle is closely tied with the biology of its host plants. The host plants
germinate anytime from early October to late December and senesce from early April to mid
May, and most of the active parts of the BCB life cycle also occur during this time (see
Figure 5.9). The BCB reproduces once and dies within a single year.  Adults emerge from
pupae, feed on nectar, mate and lay eggs during a flight season that lasts 4 to 6 weeks from
late February to early May. While females normally (although not always) only mate once,
males emerge up to 10 days prior to the emergence of females and may mate several times
before dying. Adults of both sexes live on average for 10 days (with a maximum adult life
span of over 3 weeks reported).  Females lay up to 5 egg masses (250  eggs/mass) typically in
                                        208

-------
March and April. Eggs are deposited primarily near the base of dwarf plantain plants, and less
commonly on purple owl's-clover and exserted paintbrush.
                                                       /ith
Larvae hatch from eggs in roughly 10 days and grow to the 4  instar in about two weeks.
Once reaching the 4l instar, the larvae then spend a period of dormancy (diapause) under
rocks or in soil cracks that lasts through the summer. The larvae resume activity with the start
of the rainy season and the germination of dwarf plantain plants. The post-diapause larvae are
more mobile than the pre-diapause larvae and may travel tens of meters in search of food
and/or warm microclimates to bask or pupate in. Larvae pupate, with the pupae suspended
few meters above the ground  on vegetation, once they reach a weight of 300 - 500 milligrams.
Adults emerge within 15 to 30 days depending on thermal conditions, although there  is some
evidence that a few larvae in very dry years may enter into a second diapause and complete
their development the second spring after hatching.
Jan,
m
March
April
May
June
July
&ttfc,
§SBfc.
Q&
Nov;
Dec.
         Pupa
Emergence, Adult
Flight, Lav Eggs
Prediapause
  Larvae
Diapause (Larvae Dormancy)
Postdiapause
  Larvae
Figure 5.9 General Annual Life-History Parameters for the BCB
As previously discussed in Section 5.2.1.1, endosulfan is currently registered for a wide
variety of agricultural in crops, orchards, and nursery (only one non-agricultural use: ear tags
for cattle), which span a large variety of use sites and geographical regions throughout the
entire state of California and allow for the potential for almost year-round use (see Figure
5.1). Therefore, there is the potential for endosulfan use in any given area across the state to
spatially and temporally coincide with all of the critical  life-stages of the BCB, and disrupt its
life-cycle  at various points. However, during the months of peak usage August and September
(Figure 5.1), larvae are in their dormancy stage (see Figure 5.9).
For the purposes of evaluating direct exposure and effects to the BCB, larvae for the BCB
were considered 'small insects' in this assessment, while the adults of this species were
considered 'large insects'. Therefore, the potential for direct exposure and effects specifically
to the BCB resulting from endosulfan spray applications was evaluated by considering the
lowest available acute contact toxicity endpoint for terrestrial invertebrates along with the T-
REX estimated EECs for small (broadleaf plants/small insects dietary category) and large
(fruits/pods/seeds/large insects dietary category) insects. Based on the beet webworm
(Pyrausta sticticalis) toxicity data (LD50 = 0.15 ug a.i./g) the RQs for small insects ranged
from 547 to 2250, and those for large insects from 50 to 250).
RQ values representing exposures of endosulfan to terrestrial invertebrates that may serve as
prey for the terrestrial-phase CRLF, are also used to represent direct exposures of endosulfan
to the BCB. As described in Section 5.2.1.2, the Agency determined that based on the fact
that endosulfan is a highly efficacious broad spectrum insecticide, the exceedances of the
Agency's  interim LOG for listed terrestrial invertebrates for all of the modeled scenarios
assessed (for both large (adult BCB) and small (BCB larvae) insects), and the high probability
of an individual mortality occurrence based on the highest acute RQ for terrestrial
                                        209

-------
invertebrates (~1 in 1),  it appears that there are adequate lines of evidence to conclude that
there is a potential for direct effects to the BCB as result of labeled endosulfan use in
California.
However, it should be noted that the spatial distribution of current average endosulfan use and
BCB distribution do not coincide (see Table 5.7; Figure 5.7). If cultivated crop, orchard and
vineyard use areas are considered as regions of potential endosulfan use in the future, then a
substantial degree of overlap between BCB occurrence, its critical habitat and where
endosulfan use could occur (Figure E.5 of Appendix E).  Furthermore, there is potential for
endosulfan to be transported long distances from the application sites (Section 3.2.4.4).

       5.2.6.2 Indirect Effects (via Reduction  in Prey Base & Habitat Effects)
The primary diet for the BCB larvae are dwarf plantain plants (although they may also feed
on purple owl's-clover or exserted paintbrush if the dwarf plantains senesce before the larvae
pupate). Adults  feed on the nectar of a variety of plants found in association with serpentine
grasslands [e.g., California goldfields, tidy-tips,  desert parsley, scytheleaf (Allium
falcifolium), sea muilla (muilla maritime), false babystars  (Linanthus androsaceus), and
intermediate fiddleneck (Amsinckia intermedia)}.
In addition to serving as the primary dietary item of the BCB, terrestrial plants serve several
important habitat-related functions that are described below in Section 5.2.6.3 in detail with
regards to critical habitat. Therefore, the potential for indirect effects to the BCB via loss of
terrestrial plant  food items and impacts to habitat and/or primary production was considered.
Terrestrial Plants
For the purposes of this assessment, the potential for indirect effects to the BCB via loss of
terrestrial plant  food items and impacts to habitat and/or primary production was assessed by
considering effects to terrestrial plants. As noted in Section 5.2.1.3, there were no data to
reliably quantitatively evaluate the effects and the potential risks of endosulfan to terrestrial
plants. However, aquatic non-vascular plants are not particularly sensitive to endosulfan,
endosulfan has a neural toxic mode of action, and no studies demonstrating significant
adverse effects of endosulfan to any vascular aquatic or terrestrial plant have been identified
in the open literature. The only terrestrial plant information comes from studies of single
application to food crops to determine the efficacy of endosulfan for the control of insect
pests. By design, the exposure concentrations for the control of insect pests are intended not to
harm the crop plants. However, these studies are only available for herbaceous plants and
none for woody plants. In addition, since 1970 no ecological incidents have been reported to
the Agency that involved any documented harm to either aquatic or terrestrial plants, despite
that it is regularly directly applied on or near a very wide variety of agricultural and home
garden plants. Therefore, although effects to terrestrial plants cannot be quantified due to the
lack of data, the available lines of evidence provide no compelling reason to believe that there
is a potential for indirect effects to the BCB via loss of terrestrial plant food items and impacts
to habitat and/or primary production as result of labeled endosulfan use in California.
       5.2.6.3 Modification to Designated Critical Habitat
The primary constituent elements (PCEs) of the  BCB include:
                                         210

-------
   •   The presence of annual or perennial grasslands with little to no overstory that provide
       north/south and east/west slopes with a tilt of more than 7 degrees for larval host plant
       survival during periods of atypical weather (e.g., drought).
   •   The presence of the primary larval host plant, dwarf plantain (Plantago erecta) (a
       dicot) and at least one of the secondary host plants, purple owl's-clover or exserted
       paintbrush, are required for reproduction, feeding, and larval development.
   •   The presence of adult nectar sources for feeding.
   •   Aquatic features such as wetlands, springs, seeps, streams, lakes, and ponds and their
       associated banks, that provide moisture during periods of spring drought; these
       features can be ephemeral, seasonal, or permanent.
   •   Soils derived from serpentinite ultramafic rock (Montara, Climara, Henneke, Hentine,
       and Obispo soil series) or similar soils (Inks, Candlestick, Los Gatos, Pagan, and
       Barnabe soil series) that provide areas with fewer aggressive, nonnative plant species
       for larval host plant and adult nectar plant survival and reproduction.
   •   The presence of stable holes and cracks in the soil, and surface rock outcrops that
       provide shelter for the larval stage of the bay checkerspot butterfly during summer
       diapause.
For the purposes of this assessment, the potential for indirect effects to the BCB as result of
effects to the PCEs of its designated critical habitat is assessed by considering effects to
terrestrial  plants. Similar to what was noted in Section 5.2.6.2 above, in which the potential
for indirect effects to the BCB via loss of terrestrial plant food items and impacts to habitat
and/or primary production was assessed, although effects to terrestrial plants cannot be
quantified due to the lack of data, the Agency concludes that the weight-of-evidence suggests
that effects to terrestrial plants as a result of labeled endosulfan use in California are not
expected to the extent that there will be modification of BCB designated critical habitat.

5.2.7 Valley Elderberry Longhorn Beetle
              5.2.7.1 Direct Effects
The VELB "feeds on at least one species of elderberry (Sambucus) and perhaps as many as
three elderberry taxa" including S. glauca, S.  caemlea, and S. mexicana (USFWS, 1984).
VELB adults consume elderberry foliage and possibly also the flowers whereas the "larvae
are borers and feed on the soft pith in stems and roots of the elderberry" (USFWS, 1984 and
2007 b; and LSA, 2004).  The adults eat from when they emerge in the spring until about June
when they begin to mate (California's Endangered Insects).
The life cycle of the VELB is divided into four stages: egg, larva, pupa, and adult, and has
been assumed to encompass two years, but recent information from rearing experiments
suggests that a one year cycle is possible,  if not probable (Barr, 1991). Female VELB lay their
eggs singly or in small groups on live elderberry leaves, in crevices in the bark, at the
stem/trunk junctions, or at the stem/petiole junctions of the elderberry (Barr, 1991;
California's Endangered Insects; and Talley,  no date). The first instar larvae are exposed on
the surface of the shrub anywhere from a few minutes to a day before they bore to the center
of elderberry stems where they create a characteristic feeding gallery in the pith at the center
of the stem (Talley et al.,  2006 and Barr, 1991). The larvae develop for 1 or 2 years feeding
                                        211

-------
on pith (Talley et al., 2006), eventually creating a pupal chamber. In the pupal chamber, the
larvae metamorphose into their pupae between December and April (Talley et al., 2006).
Pupation is thought to take about one month and the adult may remain in the chamber for up
to several weeks (Talley et al., 2006). The VELB adults typically emerge at about the same
time the elderberry flowers, between mid-March and mid-June" (USFWS,  1984 and Talley et
al., 2006). The adults live for a few days to a few weeks between mid-March and early-June
(Talley, no date and 2006; and USFWS, 1984). However, most records are for late-April to
mid-May (Talley et al., 2006 and USFWS, 1984). During this period, the adults feed on
elderberry leaves and possibly flowers, and reproduce within the canopy (Talley, no date and
2006). The females also lay their eggs during this period (USFWS, 2006). The lifespan of the
adult VELB is unknown; however, it is suspected that they die after reproducing (LSA, 2004).
As with the BCB, there is the potential for endosulfan use in any given area across the state to
temporally coincide with all of the critical life-stages of the BCB, and disrupt its life-cycle at
various points. For the purposes of evaluating direct exposure and effects to the VELB, larvae
were considered 'small insects' in this assessment, while the adults of this species were
considered 'large insects'. Therefore, the potential for direct exposure and effects specifically
to the VELB resulting from endosulfan spray applications was evaluated by considering the
lowest available acute contact toxicity endpoint for terrestrial invertebrates along with the T-
REX estimated EECs for small (broadleaf plants/small insects dietary category) and large
(fruits/pods/seeds/large insects dietary category) insects. The RQ analysis for VELB  is the
same as evaluated previously for BCB (Section 5.2.5). Therefore, as with BCB, it appears that
there are adequate lines of evidence to conclude that there is a potential for direct effects to
the VELB as result of labeled endosulfan use in California. However, the VELB or its critical
habitat occurs in counties with less than 0.01% of the average endosulfan use in California
from 2005-2006 (Table 5.7 and Figure 5.7). When the cultivated crop, orchard, and vineyard
agricultural uses are considered as potential area of future endosulfan use, then a moderate
amount of spatial overlap with VELB occurrence can be observed (Figure  E.6 of Appendix
E).  Furthermore, there is potential for endosulfan to be transported long distances from the
application sites (Section 3.2.4.4).

              5.2.7.2 Indirect Effects (via Reduction in Prey Base & Habitat Effects)
The VELB feeds on at least one species of elderberry (Sambucus) and perhaps as many as
three elderberry taxa including S. glauca, S. caemlea, and S. mexicana (USFWS, 1984). In
addition to serving as the primary dietary item of the VELB, terrestrial plants serve several
important habitat-related functions that are described below in  Section 5.2.5.2 in detail with
regards to critical habitat. Therefore, the potential for indirect effects to the VELB via loss of
terrestrial plant food items and impacts to habitat and/or primary production was considered.
Terrestrial Plants
For the purposes of this assessment, the potential for indirect effects  to the VELB via loss of
terrestrial plant food items and impacts to habitat and/or primary production was assessed by
considering effects to terrestrial plants. As noted in Section 5.2.1.3, there were no data to
reliably quantitatively evaluate the effects and the potential risks of endosulfan to terrestrial
plants. However, aquatic non-vascular plants are not particularly sensitive to endosulfan,
endosulfan has a neural toxic mode of action, and no studies demonstrating significant
adverse effects of endosulfan to any vascular aquatic or terrestrial plant have been identified
                                        212

-------
in the open literature. The only terrestrial plant information comes from studies of single
application to food crops to determine the efficacy of endosulfan for the control of insect
pests. By design, the exposure concentrations for the control of insect pests are intended not to
harm the crop plants. However, these studies are only available for herbaceous plants and
none for woody plants. In addition, since 1970 no ecological incidents have been reported to
the Agency that involved any documented harm to either aquatic or terrestrial plants, despite
that it is regularly directly applied on or near a very wide variety of agricultural and home
garden plants. Therefore, although effects to terrestrial plants cannot be quantified due to the
lack of data, the available lines of evidence provide no compelling reason to believe that there
is a potential for indirect effects to the VELB  via loss of terrestrial plant food items and
impacts to habitat and/or primary production as result of labeled endosulfan use in California.
              5.2.7.3 Modification to Designated Critical Habitat
The primary constituent elements (PCEs) of the VELB include the presence of riparian
elderberry trees during its entire life cycle. For the purposes of this assessment, the potential
for indirect effects to the VELB as result of effects to the PCEs of its designated critical
habitat is assessed by considering effects to terrestrial plants. Similar to what was noted in
Section 5.2.6.2 above,  in which the potential for indirect effects to the VELB via loss of
terrestrial plant food items and impacts to habitat and/or primary production was assessed,
although effects to terrestrial plants cannot be quantified due to the lack of data, the Agency
concludes that the weight-of-evidence suggests that effects to terrestrial plants as a result of
labeled endosulfan use in California are not expected to the extent that there will be
modification of VELB designated critical habitat.
                                         213

-------
  6  Uncertainties

6.1  Exposure Assessment Uncertainties

6.1.1 Maximum Use Scenario

The screening-level risk assessment focuses on characterizing potential ecological risks
resulting from a maximum use scenario, which is determined from labeled statements of
maximum application rate and number of applications with the shortest time interval between
applications. The frequency at which actual uses approach this maximum use scenario may
be dependant on pest resistance, timing of applications, cultural practices, and market forces.

6.1.2 Aquatic Exposure Modeling of Endosulfan

The standard ecological water body scenario (EXAMS pond) used to calculate potential
aquatic exposure to pesticides is intended to represent conservative estimates, and to avoid
underestimations of the actual exposure. The standard scenario consists of application to a
10-hectare field bordering a 1-hectare, 2-meter deep (20,000 m3) pond with no outlet.
Exposure estimates generated using the EXAMS pond are intended to represent a wide variety
of vulnerable water bodies that occur at the top of watersheds including prairie pot holes,
playa lakes, wetlands, vernal pools, man-made and natural ponds, and intermittent and lower
order streams.  As a group, there are factors that make these water bodies more or less
vulnerable than the EXAMS pond.  Static water bodies that have larger ratios of pesticide-
treated drainage area to water body volume would be expected to have higher peak EECs than
the  EXAMS pond. These water bodies will be either smaller in size  or have larger drainage
areas. Smaller water bodies have limited storage capacity and thus may overflow and carry
pesticide in the discharge, whereas the EXAMS pond has no discharge. As watershed size
increases beyond 10-hectares, it becomes increasingly unlikely that the entire watershed is
planted with a single crop that is all treated simultaneously with the pesticide. Headwater
streams can also have peak concentrations higher than the EXAMS pond, but they likely
persist for only short periods of time and are then carried and dissipated downstream.

The Agency acknowledges that there are some unique aquatic habitats that are not accurately
captured by this modeling scenario and modeling results may, therefore, under- or over-
estimate exposure, depending on a number of variables.  For example,  some organisms may
inhabit water bodies of different size and depth and/or are located adjacent to larger or smaller
drainage areas than the EXAMS pond.  In addition, the Services agree that the existing
EXAMS pond represents the best currently available approach for estimating aquatic
exposure to pesticides (USFWS/NMFS 2004).

PRZM-EXAMS modeled EECs are intended to represent exposure of aquatic organisms in
relatively small ponds and low-order streams.  Therefore it is likely that EECs generated from
the  PRZM-EXAMS model will over-estimate potential concentrations in larger receiving
water bodies such as estuaries, embayments, and coastal marine areas because chemicals in
runoff water (or spray drift, etc.) should be diluted by a much larger volume of water than
                                       214

-------
would be found in the 'typical' EXAMS pond. However, as chemical constituents in water
draining from freshwater streams encounter brackish or other near-marine-associated
conditions, there is potential for important chemical transformations to occur. Many chemical
compounds can undergo changes in mobility, toxicity, or persistence when changes in pH, Eh
(redox potential), salinity, dissolved oxygen (DO) content, or temperature are encountered.
For example, desorption and re-mobilization of some chemicals from sediments can occur
with changes in salinity (e.g., Means 1995; Swarzenski et al. 2003; Jordan et al. 2008),
changes in pH (e.g., Wood and Baptista 1993; Parikh et al. 2004; Fernandez et al. 2005), Eh
changes (Wood and Baptista 1993; Velde and Church  1999), and other factors.  Thus,
although chemicals in discharging rivers may be diluted by large volumes of water within
receiving estuaries and embayments, the hydrochemistry of the marine-influenced water may
negate some of the attenuating impact of the greater water volume; for example, the effect of
dilution may be confounded by changes in chemical mobility (and/or bioavailability) in
brackish water. In addition, freshwater contributions from discharging streams and rivers do
not instantaneously mix with more saline water bodies. In these settings, water will
commonly remain highly stratified, with fresh water lying atop denser, heavier saline water -
meaning that exposure to concentrations found in discharging stream water  may propagate
some distance beyond the outflow point of the stream (especially near the water surface).
Therefore, it is not assumed that discharging water will be rapidly diluted by the entire water
volume within an estuary, embayment, or other coastal aquatic environment. PRZM-EXAMS
model results should be considered consistent with concentrations that might be found near
the head of an estuary unless there is specific information - such as monitoring data - to
indicate otherwise. Conditions nearer to the mouth of a bay or estuary, however, may be
closer to a marine-type system, and thus more subject to the notable buffering,  mixing, and
diluting capacities of an open marine environment. Conversely, tidal effects (pressure waves)
can propagate much further upstream than the actual estuarine water, so  discharging river
water may become temporarily partially impounded near the mouth (discharge point) of a
channel, and resistant to mixing until tidal forces are reversed.

The Agency does not currently have sufficient information regarding the hydrology and
hydrochemistry of estuarine aquatic habitats to develop alternate scenarios for assessed listed
species that inhabit these types of ecosystems. The Agency acknowledges that there are
unique brackish and estuarine  habitats that may not be accurately captured by PRZM-
EXAMS modeling results, and may, therefore, under-  or over-estimate exposure, depending
on the aforementioned variables.

In general, the linked PRZM/EXAMS model  produces estimated aquatic concentrations that
are expected to be exceeded once within a ten-year period. The Pesticide Root Zone Model is
a process or "simulation" model that calculates what happens to a pesticide  in an agricultural
field on a day-to-day basis.  It considers factors such as rainfall and plant transpiration of
water, as well as how and when the pesticide is applied. It has two major components:
hydrology and chemical transport. Water movement is simulated by the use of generalized
soil parameters, including field capacity, wilting point, and saturation water content. The
chemical transport component can simulate pesticide application on the soil or on the plant
foliage.  Dissolved, adsorbed,  and vapor-phase concentrations in the soil are estimated by
                                        215

-------
simultaneously considering the processes of pesticide uptake by plants, surface runoff,
erosion, decay, volatilization, foliar wash-off, advection, dispersion, and retardation.

Uncertainties associated with each of these individual components add to the overall
uncertainty of the modeled concentrations. Additionally, model inputs from the
environmental fate degradation studies are chosen to represent the upper confidence bound on
the mean values that are not expected to be exceeded in the environment approximately 90
percent of the time. Mobility input values are chosen to be representative of conditions in the
environment. The natural variation in soils adds to the uncertainty of modeled values.
Factors such as application date, crop emergence date, and canopy cover can also affect
estimated concentrations, adding to the uncertainty of modeled values. Factors within the
ambient environment such as soil temperatures, sunlight intensity, antecedent soil moisture,
and surface water temperatures can cause actual  aquatic concentrations to differ for the
modeled values.

Unlike spray  drift, tools are currently not available to evaluate the effectiveness of a
vegetative setback on runoff and loadings.  The effectiveness of vegetative setbacks is highly
dependent on the condition of the vegetative strip.  For example, a well-established, healthy
vegetative setback can be a very effective means of reducing runoff and erosion from
agricultural fields. Alternatively, a setback of poor vegetative quality or a setback that is
channelized can be ineffective at reducing loadings. Until such time as a quantitative method
to estimate the effect of vegetative setbacks on various conditions on pesticide loadings
becomes available, the aquatic exposure predictions are likely to overestimate exposure where
healthy vegetative setbacks exist and underestimate exposure where poorly developed,
channelized, or bare setbacks exist.

6.1.2.1 Potential Groundwater Contributions to Surface Water Chemical
Concentrations

Although the potential impact of discharging groundwater on CRLF populations is not
explicitly delineated, it should be noted that groundwater could provide a source of pesticide
to surface water bodies - especially low-order streams, headwaters, and groundwater-fed
pools.  This is particularly likely if the chemical is persistent and mobile.  Soluble chemicals
that are primarily  subject to photolytic degradation will be very likely to persist in
groundwater, and can be transportable over long distances.  Similarly, many chemicals
degrade slowly under anaerobic conditions (common in aquifers) and are thus more persistent
in groundwater. Much of this groundwater will eventually be discharged to the surface -
often supporting stream flow in the absence of rainfall.  Continuously flowing low-order
streams in particular are sustained by groundwater discharge, which can constitute 100% of
stream flow during baseflow (no runoff) conditions. Thus, it is important to keep in mind that
pesticides in groundwater may have  a major (detrimental) impact on surface water quality,
and on CRLF habitats.

SciGrow may be used to determine likely  'high-end' groundwater vulnerability, with the
assumption (based upon persistence  in sub- and anoxic conditions, and mobility) that much of
the compound entering the groundwater will be transported some distance and eventually
                                        216

-------
discharged into surface water. Although concentrations in a receiving water body resulting
from groundwater discharge cannot be explicitly quantified, it should be assumed that
significant attenuation and retardation of the chemical will have occurred prior to discharge.
Nevertheless, groundwater could still be a significant consistent source of chronic background
concentrations in surface water,  and may also add to surface runoff during storm events (as a
result of enhanced groundwater  discharge typically characterized by  the 'tailing limb' of a
storm hydrograph).

6.1.3 Multiple Growing Seasons per Year

Most endosulfan product labels  specify application rates on a per crop cycle basis (not on a
per year basis).  Information from BEAD indicates that several crops  can be grown more than
one time/year in California, as indicated in Section 3. Since standard PRZM scenarios only
consist of one crop per year, applications to only one crop per year were modeled. The crops
that may be grown multiple times in a calendar year that can be treated by endosulfan include
Broccoli & Lettuce (two crops); Cabbage, Chinese cabbage, Cauliflower, Collards, Mustard
Green & Sweet Corn (three crops ); and Kale (four crops). If endosulfan is applied for
multiple cropping cycles within  a year, EECs presented in this assessment may underpredict
exposures. For all other labeled uses, it was assumed that a maximum seasonal application
specified on the label was equivalent to a maximum annual application.

6.1.4 Usage  Uncertainties

County-level usage data were obtained from California's Department of Pesticide Regulation
Pesticide Use Reporting (CDPR PUR) database.  Thirteen years of data (1994- 2006) were
included in this analysis although error identification was not available for the older data from
1994 tO  1998. Statistical methodology for identifying outliers, in terms of area treated and
pounds  applied, was provided by CDPR for later years only (1999 to 2006).  All  data was
included in order to analyze historical usage of endosulfan and no data was used as input
parameter for modeling. CDPR PUR documentation indicates that errors in  the data may
include  the following: a misplaced decimal; incorrect measures, area treated, or units; and
reports of diluted pesticide concentrations.  In addition, it is possible that the data may contain
reports for pesticide uses that have been cancelled. The CPDR PUR data does not include
home owner applied pesticides;  therefore, residential uses are not likely to be reported.  As
with all  pesticide usage data, there may be instances of misuse and misreporting. The Agency
made use of the most current, verifiable information; in cases where  there were discrepancies,
the most conservative information was used.

6.1.5 Terrestrial Exposure Modeling of Endosulfan

The Agency relies on the work of Fletcher et al. (1994) for setting the assumed pesticide
residues in wildlife dietary items. These residue assumptions are believed to reflect a realistic
upper-bound residue estimate, although the degree to which this assumption reflects a specific
percentile estimate is difficult to quantify. It is important to note that the field measurement
efforts used to develop the Fletcher estimates of exposure involve highly varied sampling
techniques.  It is entirely possible that much of these data reflect residues averaged over entire
above ground plants in the case  of grass and forage sampling.
                                        217

-------
It was assumed that ingestion of food items in the field occurs at rates commensurate with
those in the laboratory. Although the screening assessment process adjusts dry-weight
estimates of food intake to reflect the increased mass in fresh-weight wildlife food intake
estimates, it does not allow for gross energy differences. Direct comparison of a laboratory
dietary concentration- based effects threshold to a fresh-weight pesticide residue estimate
would result in an underestimation of field exposure by food consumption by a factor of 1.25
- 2.5 for most food items.

Differences in assimilative efficiency between laboratory and wild diets suggest that current
screening assessment methods do not account for a potentially important aspect of food
requirements.  Depending upon species and dietary matrix, bird assimilation of wild diet
energy ranges from 23 - 80%, and mammal's assimilation ranges from 41 - 85% (U.S.
Environmental Protection Agency, 1993). If it is assumed that laboratory chow is formulated
to maximize assimilative efficiency (e.g., a value of 85%), a potential for underestimation of
exposure may exist by assuming that consumption of food in the wild is comparable with
consumption during laboratory testing. In the screening process, exposure may be
underestimated because metabolic rates are not related to food consumption.

For the terrestrial exposure analysis of this risk assessment, a generic bird or mammal was
assumed to occupy either the treated field or adjacent areas receiving a treatment rate on the
field.  Actual habitat requirements of any particular terrestrial species were not considered,
and it was assumed that species occupy, exclusively and permanently, the modeled treatment
area. Spray drift model predictions suggest that this assumption leads to an overestimation of
exposure to species that do not occupy the treated field exclusively and permanently.

Exposure assessment for the SJKF did  not include small mammal, avian, and reptilian dietary
items. Diet items of SJKF include small animals (e.g. blacktailed hares, desert  cottontails,
mice, kangaroo rats, squirrels), birds, lizards,  insects and grass. Although grass and insects are
included in the dietary assessment of the fox,  the current assessment is not inclusive of direct
effects to the fox through all of its diet items.  Exclusion of vertebrate  dietary items is a
limitation of the T-Rex model, but is not expected to affect the risk characterization outcome
for the SJKF. For acute dose-based RQs, all modeled scenarios (N=20) resulted in an
exceedance of the Agency's acute listed species LOG (RQ>0.1) for mammals and it was
determined that there is a potential for  direct adverse effects on SJKF  individuals from foliar
spray applications of endosulfan in CA. Inclusion of mammalian and avian dietary items
would increase the potential risk to SJKF. The current assessment, while conservative,
represents the potential risk of SJKF exposure to endosulfan as highly probable for foliar
spray applications.

The potential for endosulfan biomagnification in terrestrial food webs (i.e., increasing
concentration with increasing trophic level) was not evaluated due to limitations in current
risk assessment tools. However,  a number of recent studies have been conducted for
evaluating and predicting the bioaccumulation of organic chemicals in food webs that involve
terrestrial (air-respiring) organisms (e.g., Armitage and Gobas, 2007; Kelly et al., 2007; Czub
and McLachlan,  2004; Kelly and Gobas, 2003; Sharp and Mackay, 2000; McLachlan, 1996).
                                        218

-------
A key component from these studies has been the relationship between observed or predicted
bioaccumulation of poorly metabolized chemicals in terrestrial animals and the octanol-air
partition coefficient (KOA). For terrestrial organisms, the relationship between the KOA and
bioaccumulation is somewhat analogous to the use of the octanol-water partition coefficient
(Kow) for predicting the bioaccumulation of nonionic organic chemicals by aquatic (water-
respiring) organisms. Based on model simulations and supporting observations of chemical
accumulation in Arctic terrestrial, purely aquatic and marine mammalian food webs, Kelly et
al. (2007) indicate that slowly-metabolized organic chemicals with relatively low to moderate
KOW values (i.e., log K0w between 2-5) and high KOA values (i.e., log KOA > 6) have the
potential to biomagnify in terrestrial and marine mammal food webs but not purely aquatic
food webs.  In their model, the  conceptual basis for biomagnification of this group of
compounds by terrestrial organisms relates largely to the greater ability of terrestrial
organisms to assimilate food from their diet and their slower ability to eliminate these
chemicals through respiratory processes relative to aquatic organisms.

Based on a log K0w of 3.7 and  log KOA of 7.9 for p-endosulfan, Kelly et al. (2007) calculate
that biomagnification factors (BMFs) range from 2.5 to 28 for various herbivorous and
carnivorous terrestrial organisms, but are all less than 1 for aquatic organisms. Importantly,
these calculated BMP values assume that the chemical is not metabolized in tissues of biota.
Model predictions were evaluated against measured concentrations in tissues of organisms
occupying terrestrial, piscivorous, and marine arctic food webs (e.g., lichen, caribou, wolf
food chain; macroalgae, zooplankton, bivalve, fish, whale/seal, polar bear food web) for a
variety of chlorinated biphenyls, chlorinated benzenes, hexachlorocyclohexanes, cyclodienes
(including endosulfan), and DDT-related compounds.  Agreement between the mean model-
predicted and observed concentrations in biota (lipid normalized) was generally good (i.e.,
many fell within a factor of three and all fell within a factor of 10 of observed data). Limited
data were available for endosulfan to directly verify model predictions and thus, the potential
for biomagnifications in terrestrial food webs is largely based  on indirect evidence (e.g.,
model predictions).

Regarding the bioaccumulation assessment in aquatic ecosystems (e.g., using the KABAM
model as described in Section 5), a significant sources of uncertainty pertains to the
variability associated with measured values of Kow for endosulfan isomers (alpha, beta).  The
reported log Kow values for the d isomer range from 3.55 to 4.74 (McConnell et al., 1998 and
MRID 41421501, respectively) and those for the P isomer range from 3.62 to 4.79 from the
same sources.  For parameterizing the KABAM model, the higher value of Kow was selected

6.1.6 Spray Drift Modeling

Although there may be multiple endosulfan applications at a single site, it is unlikely that the
same organism would be exposed to the maximum amount of spray drift from every
application made. In order for  an organism to receive the maximum concentration of
endosulfan from multiple applications, each application of endosulfan would have to occur
under identical atmospheric conditions (e.g., same wind speed and - for plants - same wind
direction) and (if it is an animal) the animal being exposed would have to be present directly
downwind at the same distance after each application.  Although there may be sites where the
                                        219

-------
dominant wind direction is fairly consistent (at least during the relatively quiescent conditions
that are most favorable for aerial spray applications), it is nevertheless highly unlikely that
plants in any specific area would receive the maximum amount of spray drift repeatedly. It
appears that in most areas (based upon available meteorological data) wind direction is
temporally very changeable, even within the same day.  Additionally, other factors, including
variations in topography, cover, and meteorological conditions over the transport distance are
not accounted for by the AgDRIFT/AGDISP model (i.e., it models spray drift from aerial and
ground applications in a flat area with little to no ground cover and a steady, constant wind
speed and direction). Therefore, in most cases, the drift estimates from AgDRIFT/AGDISP
may overestimate exposure even from single applications, especially as the distance increases
from the site of application, since the model does not account for potential obstructions (e.g.,
large hills, berms, buildings, trees, etc.).  Furthermore, conservative assumptions are often
made regarding the droplet size distributions being modeled ('ASAE Very Fine to Fine' for
orchard uses and 'ASAE Very Fine' for agricultural uses), the application method (e.g.,
aerial), release heights and wind speeds.  Alterations in any of these inputs would change the
area of potential effect.

6.2 Effects Assessment Uncertainties

6.2.1 Age Class and Sensitivity of Effects Thresholds

It is generally recognized that test organism age may have a significant impact on the
observed sensitivity to a toxicant. The acute toxicity data for fish are collected on juvenile
fish between 0.1 and 5 grams.  Aquatic invertebrate acute testing is performed on
recommended immature age classes (e.g., first instar for daphnids, second instar for
amphipods,  stoneflies, mayflies, and third instar for midges).

Testing of juveniles may overestimate toxicity at older age classes for pesticide active
ingredients that act directly without metabolic transformation because younger age classes
may not have the enzymatic  systems associated with detoxifying xenobiotics. In so far as the
available toxicity data may provide ranges of sensitivity information with respect to age class,
this assessment uses the most sensitive life-stage information as measures of effect for
surrogate aquatic animals, and is therefore, considered as protective.

6.2.2 Use of Surrogate Species Effects Data

Guideline toxicity tests and open literature data on endosulfan considered suitable for
quantitative use are not available for frogs or any other aquatic-phase amphibian; therefore,
freshwater fish are used as surrogate species for aquatic-phase amphibians.  Although no data
are available for endosulfan , the available open literature information on endosulfan toxicity
to aquatic-phase amphibians shows that acute and chronic ecotoxicity endpoints for aquatic-
phase amphibians range from approximately 2 to > 4000 times less sensitive than the most
sensitive freshwater fish (LC50 0.01 ug a.i./L) used in this assessment.  Therefore, endpoints
based on freshwater fish ecotoxicity data are assumed to be protective of potential direct
effects to aquatic-phase amphibians  including the CRLF, and extrapolation of the risk
conclusions from the most sensitive  tested species to the aquatic-phase  CRLF and CTS  and
                                        220

-------
may overestimate the potential risks to those species. For reptiles (SFGS) and terrestrial
phase amphibians (CRLF, CTS), no data on their sensitivity to endosulfan was identified.
Therefore, the uncertainty in using avian effects data as a surrogate for reptiles and terrestrial-
phase amphibians could not be characterized. Efforts are made to select the organisms most
likely to be affected by the type of compound and usage pattern; however, there is an inherent
uncertainty in extrapolating across phyla. In addition, the Agency's LOCs are intentionally
set very low,  and conservative estimates are made in the screening level risk assessment to
account for these uncertainties.

6.2.3 Sublethal Effects

When assessing acute risk, the screening risk assessment relies on the acute mortality
endpoint as well as a suite of sublethal responses to the pesticide, as determined by the testing
of species response to chronic exposure conditions and subsequent chronic risk assessment.
Consideration of additional sublethal data in the effects determination t is exercised on a case-
by-case basis and only after careful consideration of the nature of the sublethal effect
measured and the extent and quality of available data to support establishing a plausible
relationship between the measure of effect (sublethal endpoint) and the assessment endpoints.
However, the full suite of sublethal effects from valid open literature studies is considered for
the purposes of defining the action area.

6.2.4 Location of Wildlife Species

For the terrestrial  exposure analysis of this risk assessment, a generic bird or mammal was
assumed to occupy either the treated field or adjacent areas receiving a treatment rate on the
field. Actual habitat requirements of any particular terrestrial species were not considered,
and it was assumed that species occupy, exclusively and permanently, the modeled treatment
area.  Spray drift model predictions suggest that this assumption leads to an overestimation of
exposure to species that do not occupy the treated field exclusively  and permanently.

6.2.5 Endocrine Disruption

EPA is required under the FFDCA, as amended by FQPA, to develop a screening program to
determine whether certain substances (including all pesticide active and other ingredients)
"may have an effect in humans that is similar to an effect produced by a naturally occurring
estrogen, or other such endocrine effects as the Administrator may designate. " Following the
recommendations of its Endocrine Disrupter Screening and Testing Advisory Committee
(EDSTAC), EPA determined that there were scientific bases for including, as part of the
program, androgen and thyroid hormone systems, in addition to the estrogen hormone system.
EPA also adopted EDSTAC's recommendation that the Program include evaluations of
potential  effects in wildlife.  When the appropriate screening and/or testing protocols being
considered under the Agency's Endocrine Disrupter Screening Program (EDSP) have  been
developed and vetted, endosulfan may be subjected to additional screening and/or testing to
better characterize effects related to endocrine disruption.
                                        221

-------
     7   Summary and Conclusions

  In fulfilling its obligations under Section 7(a)(2) of the Endangered Species Act, the
  information presented in this endangered species risk assessment represents the best data
  currently available to assess the potential risks of endosulfan to the CRLF, CTS, SFGS,
  SMHM, SJKF, BCB, and VELB and the designated critical habitat for CRLF, CTS, BCB and
  VELB.

  Based on the best available information, the Agency makes a Likely to Adversely Affect
  determination for the CRLF,  CTS, SFGS, SMHM, SJKF, BCB, and VELB from the use of
  endosulfan. Additionally, the Agency has determined that there is the potential for effects to
  the designated critical habitat for the CRLF, CTS, BCB and VELB from the use of the
  chemical.  Given the LAA determination for the CRLF, CTS, SFGS, SMHM, SJKF, BCB,
  and VELB and potential effects to designated critical habitat for CRLF, CTS, BCB and
  VELB, a description of the baseline status and cumulative effects for the CRLF is provided
  in Attachment 2 and the baseline status and cumulative effects for the CRLF, CTS, SFGS,
  BCB, VELB, SMHM, and SJKF in Attachment 4.

  A summary of the risk conclusions and effects determinations for the CRLF, CTS, SFGS,
  SMHM, SJKF, BCB, and VELB and their critical habitat, given the uncertainties discussed in
  Section 6, is presented inError! Reference source not found, and Table 7.2.
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
    Effects
Determination 1
                        Basis for Determination
California red-
 legged frog
 (Rana aurora
  draytonii)
   (CRLF)
LAA1
                                              POTENTIAL FOR DIRECT EFFECTS
Aquatic-phase CRLF (Eggs, Larvae, and Adults):
Freshwater Fish RQ
- Acute RQs for freshwater fish (used as a surrogate for aquatic-phase CRLFs) exceed
the listed species acute risk LOG for all 20 modeled crop scenarios.
- Chronic RQs for freshwater fish (used as a surrogate for aquatic-phase CRLFs)
exceed the chronic risk LOG for all 20 modeled crop scenarios.

Likelihood of Individual Mortality
-The chance of individual effects (i.e., mortality) for freshwater fish (surrogate for
aquatic-phase CRLFs) is as high as ~1 in 1.

Ecological Incident Reports
- 67 of the 83 endosulfan-associated ecological incidents reported to the Agency
involve fish and 53 of these are classified as 'probable' or 'highly probable;' 18 of the
20 incident reports associated with 'registered uses' involved mortality to fish.

Species Sensitivity Differences
- Model-based EECs exceed the LOG for approximately 80 to 97% of the tested
freshwater fish species

Surface Water Monitoring Data
- The highest reported values of total endosulfan in California equal or exceed the
acute LOG for freshwater fish.
                                           222

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan  on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
    Effects
Determination 1
Basis for Determination
                                 Bioaccumulation in Aquatic Prey
                                 - Based on consumption of aquatic prey that is predicted to bioaccumulate
                                 endosulfan, acute dose-based RQs exceed the acute listed LOG in all 20 crop
                                 exposure scenarios modeled for small, medium and large CRLF.

                                 Temporal and Spatial Overlap
                                 - Based on current endosulfan use data and potential endosulfan use on agricultural
                                 crops/orchards/vineyard areas, there appears to be a potential for both temporal and
                                 spatial overlap between aquatic-phase CRLF distribution and endosulfan agricultural
                                 use (Appendix E).	
                                  Terrestrial-phase CRLF (Juveniles and Adults):
                                  Direct Deposition on Forage Items: Avian RQ
                                  - Acute-dose based RQs for 20g birds (surrogate for terrestrial-phase amphibians)
                                  consuming large and small insects contaminated with endosulfan from direct
                                  deposition exceed LOG for all 20 crop scenarios modeled.
                                  - Chronic RQs for the small and large insect categories exceed the Agency LOG of 1
                                  for 20 and 3 crop exposure scenarios, respectively.

                                  Direct Deposition on Forage Items: Refined Herpetofauna Modeling
                                  - Using refined modeling, acute listed LOG was exceeded in 39% or more of the
                                  acute dose-based; and acute dietary-based species/diet combinations modeled. The
                                  chronic LOG was exceeded in 46% of the chronic dietary-based species/diet
                                  combinations modeled.

                                  Likelihood of Individual Mortality
                                  The chance of individual effects (i.e., mortality) for birds (surrogate for terrestrial -
                                  phase CRLFs) and herpetofauna based on direct deposition onto food items is as high
                                  as~l in 1.

                                  Temporal and Spatial Overlap
                                  - Based on current endosulfan use data and potential endosulfan use on agricultural
                                  crops/orchards/vineyard areas, there appears to be a potential for both temporal and
                                  spatial overlap between terrestrial -phase CRLF distribution and endosulfan
                                  agricultural use (Appendix E).	
                                                   POTENTIAL FOR INDIRECT EFFECTS
                                  CRLF Aquatic Prey Items, Aquatic Habitat, Cover and/or Primary Productivity
                                  Freshwater fish and aquatic-phase amphibians:
                                  - Acute and chronic RQs exceed the listed species acute and chronic LOCs; high
                                  likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
                                  tested fish species and a large number of aquatic incidents involving fish, as
                                  described above for CRLF (see "Potential Direct Effects;  Aquatic Phase CRLF
                                  [eggs, larvae, adults])"

                                  .Freshwater Invertebrates:
                                  -Acute and chronic RQs for freshwater invertebrates exceed the listed species acute
                                  and chronic risk LOG for all 20 crop exposure scenarios modeled.
                                  -The chance of individual effects (i.e., mortality) for freshwater invertebrates is as
                                  high as ~1 in 1.
                                  - Model-based EECs exceed the LOG for approximately 45% to 60% of the tested
                                  freshwater invertebrate species
                                  - The highest reported values of total endosulfan in California equal or exceed the
                                  listed acute LOG and chronic LOG for freshwater invertebrates.
                                                 223

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
   Species
     Effects
Determination 1
        Basis for Determination
                                  CRLF Terrestrial Prey Items, Riparian Habitat
                                  Terrestrial-phase Amphibians'.
                                  - Exceedence of acute and chronic LOCs for terrestrial-phase amphibians as
                                  described above for CRLF (see "Potential Direct Effects;  Terrestrial Phase CRLF
                                  [juveniles and adults])"

                                  Terrestrial Invertebrates:
                                  - Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
                                  all 20 crop exposure scenarios modeled.
                                  - EECs exceed for large and small insects exceed the LD50 values for 80% and 99%
                                  of the tested terrestrial invertebrate species, respectively.
                                  -The chance of individual effects (i.e., mortality) for terrestrial invertebrates resulting
                                  from direct deposition in application sites is as high as ~1 in 1.
                                  - When indirect effects are considered via exposure of terrestrial insects to spray drift,
                                  spatial overlap between potential endosulfan use on agricultural crops, orchards, and
                                  vineyard areas can extend up to 2 miles from the source (Appendix E).

                                  Small Mammals: Direct Deposition on Forage Items
                                  - Acute and chronic dose-based RQs for small mammals foraging on food items
                                  receiving direct deposition of applied endosulfan exceed the acute listed LOG for
                                  mammals in all 20 crop scenarios modeled
                                  - Chronic diet-based RQs for small mammals foraging on food items receiving direct
                                  deposition of applied endosulfan exceed the chronic LOG for mammals in 13 of the
                                  20 crop scenarios modeled
                                  - The chance of individual effects (i.e., mortality) for small mammals is as high as ~1
                                  inl.

                                  Small Mammals: Bio accumulation in Terrestrial Prey
                                  - RQs exceed of acute listed and chronic LOCs in all 20 exposure scenarios modeled
                                  based on small mammals eating earthworms that are predicted to bioaccumulate
                                  endosulfan from soil.
California
tiger
salamander
 (Ambystoma
 californiense
LAA1
POTENTIAL FOR DIRECT EFFECTS
                  Direct Deposition on Forage Items: Avian RQ (Terrestrial-Phase CTS)
                  - Acute-dose based RQs for 20g birds (surrogate for terrestrial-phase amphibians)
                  consuming large and small insects contaminated with endosulfan from direct
                  deposition exceed LOG for all 20 crop scenarios modeled.
                  - Chronic RQs for the small and large insect categories exceed the Agency LOG of 1
                  for 20 and 3 crop exposure scenarios, respectively.

                  Direct Deposition on Forage Items: Refined Herpetofauna Modeling
                  - Using refined modeling, acute listed LOG was exceeded in 39% or more of the
                  acute dose-based; acute dietary-based species/diet combinations modeled. The
                  chronic LOG was exceeded in 46% of the chronic dietary-based species/diet
                  combinations modeled.

                  Likelihood of Individual Mortality
                  The chance of individual effects (/'. e., mortality) for birds (surrogate for reptiles and
                  terrestrial phase amphibians) and herpetofauna based on direct deposition onto food
                  items is as high as ~1 in 1.

                  Bioaccumulation in Aquatic Prey
                                                 224

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species

San Francisco
garter snake
(Thamnophis
sirtalis
tetrataenia)
Effects
Determination 1

LAA1
Basis for Determination
- Based on consumption of aquatic prey that is predicted to bioaccumulate
endosulfan, acute dose-based RQs exceed the acute listed LOG in all 20 crop
exposure scenarios modeled for small CRLF (used as surrogate for CTS exposure
potential).
Bioaccumulation in Terrestrial Prey
- Based on consumption of terrestrial prey (earthworm) that is predicted to
bioaccumulate endosulfan, acute dose-based RQs and chronic diet-based RQs for
small birds (surrogate for terrestrial-phase amphibians) exceed the acute listed and
chronic LOCs in all 20 crop exposure scenarios modeled.
Freshwater Fish (Aquatic-phase CTS) :
- Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
high likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
tested fish species and a large number of aquatic incidents involve fish, as described
above for CRLF (see "Potential Direct Effects; Aquatic Phase CRLF [eggs, larvae,
adults])"
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of CTS was
identified.
- A large amount of spatial overlap between CTS occurrence and current (and
potential future use) of endosulfan was identified (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
Freshwater Fish'.
- Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
high likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
tested fish species and a large number of aquatic incidents involve fish, as described
above for CRLF (see "Potential Direct Effects; Aquatic Phase CRLF [eggs, larvae,
adults])"
Freshwater Invertebrates:
- Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
high likelihood of individual mortality, and exceedence of LOCs for 45-60% of
freshwater invertebrate species tested, as described above for CRLF (see "Potential
Indirect Effects; CRLF Aquatic Prey Items, Aquatic Habitat, Cover and/or Primary
Productivity)"
Terrestrial Invertebrates:
- Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled, there is a high likelihood of individual
mortality, and EECs exceed LD50 values for vast majority of tested species, as
described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
Items, Riparian Habitat)
POTENTIAL FOR DIRECT EFFECTS
Direct Deposition on Forage Items: Avian RO
- Acute-dose based RQs for 20g birds (surrogate for terrestrial-phase amphibians)
consuming large and small insects contaminated with endosulfan from direct
deposition exceed LOG for all 20 crop scenarios modeled.
- Chronic RQs for the small and large insect categories exceed the Agency LOG of 1
for 20 and 3 crop exposure scenarios, respectively.
225

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species

Salt marsh
harvest mouse
Effects
Determination 1

LAA1
Basis for Determination
Direct Deposition on Forase Items: Refined Herpetofauna Modeling
- Using refined modeling, acute listed LOG was exceeded in 39% or more of the
acute dose-based; and acute dietary -based species/diet combinations modeled. The
chronic LOG was exceeded in 46% of the chronic dietary -based species/diet
combinations modeled.
Likelihood of Individual Mortality
The chance of individual effects (/'. e., mortality) for birds (surrogate for reptiles and
terrestrial phase amphibians) and herpetofauna based on direct deposition onto food
items is as high as ~1 in 1.
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of SFGS was
identified.
- No spatial overlap identified at the County level based on current endosulfan use
reported for California. If cultivated crop/orchard/vineyard uses are considered areas
of potential endosulfan use in the future, then a substantial amount of spatial overlap
can be observed relative to the total area of SFGS occurrence sections (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
Prey Items, Habitat, Cover And/Or Primary Productivity
Freshwater Fish:
- Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
high likelihood of individual mortality, EECs exceed the LOG for 80% or more of the
tested fish species and a large number of aquatic incidents involve fish, as described
above for CRLF (see "Potential Direct Effects; Aquatic Phase CRLF [eggs, larvae,
adults])"
Freshwater Invertebrates:
- Acute and chronic RQs exceed the listed species acute and chronic LOCs, there is a
high likelihood of individual mortality, and exceedence of LOCs for 45-60% of
freshwater invertebrate species tested, as described above for CRLF (see "Potential
Indirect Effects; CRLF Aquatic Prey Items, Aquatic Habitat, Cover and/or Primary
Productivity)"
Terrestrial Invertebrates:
- Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled, there is a high likelihood of individual
mortality, and EECs exceed LD50 values for vast majority of tested species, as
described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
Items, Riparian Habitat)
Small Terrestrial Vertebrates
- Acute and chronic LOG exceedence for birds (surrogate for reptiles and terrestrial-
phase amphibians) and herpetofauna for direct deposition on food items, a high
likelihood of individual mortality, and RQ exceedence of acute listed LOG and
chronic LOG based on consumption of terrestrial and aquatic prey by birds and
mammals as described above for the CRLF (see "Potential Direct Effects;
Terrestrial Phase CRLF [juveniles and adults] and "Potential for Indirect Effects:
CRLF Terrestrial Prey Items, Riparian Habitat) "
POTENTIAL FOR DIRECT EFFECTS
Direct Deposition on Forase Items: Mammalian RO
226

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species
(Reithrodonto
mys
raviventris)
(SMHM)
San Joaquin
kit fox
(Vulpes
macrotis
muticd)
Effects
Determination 1

LAA1
Basis for Determination
- Acute and chronic dose-based RQs for small mammals foraging on food items
receiving direct deposition of applied endosulfan exceed the acute listed LOG for
mammals in all 20 crop scenarios modeled
- Chronic diet-based RQs for small mammals foraging on food items receiving direct
deposition of applied endosulfan exceed the chronic LOG for mammals in 13 of the
20 crop scenarios modeled
Likelihood of Individual Mortality
- The chance of individual effects (i.e., mortality) for small mammals resulting from
direct deposition onto forage items is as high as ~1 in 1.
Bioaccumulation in Terrestrial Prey
- RQs exceed of acute listed and chronic LOCs in all 20 exposure scenarios modeled
based on small mammals eating earthworms that are predicted to bioaccumulate
endosulfan from soil.
Temporal and Spatial
- Temporal overlap between endosulfan application and all life stages of the SMHM
was identified.
- A small amount of spatial overlap identified at the County level based on current
endosulfan use reported for California and based on areas of potential endosulfan use
in the future (i.e., cultivated crop, orchard and vineyard use areas).
- The potential for endosulfan to be transported long distances in the atmosphere may
increase SMHM exposure, particularly in areas of potential endosulfan use west of
SMHM occurrence sections in Solano County (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
Prey items, habitat, cover and/or primary productivity
Terrestrial invertebrates'.
— Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled, there is a high likelihood of individual
mortality, and EECs exceed LD50 values for vast majority of tested species, as
described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
Items, Riparian Habitat)
Small Birds and Mammals (rearing sites):
- Acute and chronic RQ for birds and mammals (whose rearing sites are potentially
used by SMHM) exceed listed acute and chronic LOCs as a result of direct deposition
onto forage items and bioaccumulation in terrestrial prey as described above for the
CRLF (Potential for Direct Effects: Terrestrial-phase CRLF [Juveniles and Adults])
and for the SMHM (see "Potential for Direct Effects ")
POTENTIAL FOR DIRECT EFFECTS
Direct Deposition on Forase Items: Mammalian RO
- Acute and chronic dose-based RQs exceed the Agency acute listed and chronic LOG
in all modeled scenarios (N-20) for large mammals. Chronic dietary -based RQs
exceed the chronic LOG in all 20 modeled scenarios for large mammals that feed on
short grass, tall grass, and broadleaf plants/small insects (chronic dietary-based RQs
for 13/20 scenarios exceed the LOG for large mammals consuming
fruits/pods/seeds/large insects).
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of SJKF was
227

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species

Bay
checkerspot
butterfly
(Euphydryas
editha
bayensis)
(BCB)
Effects
Determination 1

LAA1
Basis for Determination
identified.
- A moderate amount of spatial overlap between SJKF occurrence and current (and
potential future use) of endosulfan was identified (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
Terrestrial Invertebrates:
- Acute RQs exceed the Agency's interim LOC for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled, there is a high likelihood of individual
mortality, and EECs exceed LD50 values for vast majority of tested species, as
described above for CRLF (see "Potential for Indirect Effects: CRLF Terrestrial Prey
Items, Riparian Habitat)
Small Birds and Mammals'.
- Acute and chronic RQ for birds and mammals serving as prey to the SJKF exceed
listed acute and chronic LOCs as a result of direct deposition onto forage items and
bioaccumulation in terrestrial prey as described above for the CRLF (Potential for
Direct Effects: Terrestrial-phase CRLF [Juveniles and Adults]) and for the SMHM
(see "Potential for Direct Effects ")
POTENTIAL FOR DIRECT EFFECTS
Terrestrial Invertebrate ROs:
- Acute RQs exceed the Agency's interim LOC for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled.
- EECs exceed for large and small insects exceed the LD50 values for 80% and 99%
of the tested terrestrial invertebrate species, respectively.
Likelihood of Individual Mortality
-The chance of individual effects (i.e., mortality) for terrestrial invertebrates resulting
from direct deposition in application sites is as high as ~1 in 1.
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of BCB was
identified.
- No spatial overlap identified at the County level based on current endosulfan use
reported for California. If cultivated crop/orchard/vineyard uses are considered areas
of potential endosulfan use in the future, then a substantial amount of spatial overlap
can be observed relative to the total area of BCB occurrence sections.
- When exposure of terrestrial insects via spray drift is considered, spatial overlap
between potential endosulfan use on agricultural crops, orchards, and vineyard areas
can extend up to 2 miles from the source.
- The potential for endosulfan to be transported long distances in the atmosphere may
increase BCB exposure, particularly in areas of potential endosulfan use west of BCB
occurrence sections in Santa Clara County (Appendix E).
POTENTIAL FOR INDIRECT EFFECTS
-Although effects to terrestrial plants cannot be quantified due to the lack of data,
aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
- No studies demonstrating significant adverse effects of endosulfan to any vascular
aquatic or terrestrial plant have been identified in the open literature. The one toxicity
study found to be acceptable for quantitative use suggests that aquatic nonvascular
228

-------
Table 7.1 Effects Determination Summary for Effects of Endosulfan on the CRLF, CTS, SFGS, SMHM, SJKF,
BCB, AND VELB
Species

Valley
elderberry
longhorn
beetle
(Desmocems
califomicus
dimorphus)
Effects
Determination 1

LAA1
Basis for Determination
plants are insensitive relative to aquatic animals.
- No ecological incidents have been reported to the Agency that involve any plants
and this was classified with a certainty index of "probable" or higher linking
endosulfan as a cause, despite that it is regularly directly applied on or near a very
wide variety of agricultural plants.
POTENTIAL FOR DIRECT EFFECTS
Terrestrial Invertebrate RQs:
- Acute RQs exceed the Agency's interim LOG for listed terrestrial invertebrates in
all 20 crop exposure scenarios modeled.
- EECs exceed for large and small insects exceed the LD50 values for 80% and 99%
of the tested terrestrial invertebrate species, respectively.
Likelihood of Individual Mortality
-The chance of individual effects (i.e., mortality) for terrestrial invertebrates resulting
from direct deposition in application sites is as high as ~1 in 1.
Temporal and Spatial Overlap
- Temporal overlap between endosulfan application and all life stages of VELB was
identified.
- A small amount of spatial overlap identified at the County level based on current
endosulfan use reported for California. If cultivated crop/orchard/vineyard uses are
considered areas of potential endosulfan use in the future, then a moderate amount of
spatial overlap can be observed relative to the total area of VELB occurrence sections
(Appendix E).
- When exposure of terrestrial insects via spray drift is considered, spatial overlap
between potential endosulfan use on agricultural crops, orchards, and vineyard areas
can extend up to 2 miles from the source
- The potential for endosulfan to be transported long distances in the atmosphere may
increase VELB exposure, particularly in areas of potential endosulfan use west of
VELB occurrence sections.
POTENTIAL FOR INDIRECT EFFECTS
-Although effects to terrestrial plants cannot be quantified due to the lack of data,
aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
- No studies demonstrating significant adverse effects of endosulfan to any vascular
aquatic or terrestrial plant have been identified in the open literature. The one toxicity
study found to be acceptable for quantitative use suggests that aquatic nonvascular
plants are insensitive relative to aquatic animals.
- No ecological incidents have been reported to the Agency that involve any plants
and this was classified with a certainty index of "probable" or higher linking
endosulfan as a cause, despite that it is regularly directly applied on or near a very
wide variety of agricultural plants.
Table 7.2 Effects Determination Summary
Designated
Critical Habitat
for:
Effects
Determination 1
for the Critical Habitat Impact Analysis
Basis for Determination
229

-------
    CRLF
May Affect
-There is a potential for direct effects to the aquatic-phase CRLF and
indirect effects via reduction of aquatic-phase prey items (aquatic
invertebrates, fish, and aquatic-phase amphibians) as described in Table 7.1
above.
- There is a potential for direct effects to the terrestrial-phase CRLF and
indirect effects via reduction of terrestrial-phase prey items (mammals,
amphibians, and terrestrial invertebrates) as described in Table 7.1 above.
     CTS
May Affect
There is a potential for direct effects to the CTF and indirect effects via
reduction of aquatic-phase prey items (aquatic invertebrates and fish) as
described in Table 7.1 above.
- There is a potential for direct effects to the CTF and indirect effects via
reduction of terrestrial-phase prey items (terrestrial invertebrates) as
described in Table 7.1 above.
     BCB
   NE1
-Although effects to terrestrial plants cannot be quantified due to the lack of
data, aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
-No studies demonstrating significant adverse effects of endosulfan to any
vascular aquatic or terrestrial plant have been identified in the open
literature. The one toxicity study found to be acceptable for quantitative use
suggests that aquatic nonvascular plants are insensitive relative to aquatic
animals.
- No ecological incidents have been reported to the Agency that involve any
plants and this was classified with a certainty index of "probable" or higher
linking endosulfan as a cause, despite that it is  regularly directly applied on
or near a very wide variety of agricultural plants.	
    VELB
   NE1
-Although effects to terrestrial plants cannot be quantified due to the lack of
data, aquatic non-vascular plants are not particularly sensitive to endosulfan
- Endosulfan has a neural toxic mode of action.
-No studies demonstrating significant adverse effects of endosulfan to any
vascular aquatic or terrestrial plant have been identified in the open
literature. The one toxicity study found to be acceptable for quantitative use
suggests that aquatic nonvascular plants are insensitive relative to aquatic
animals.
- No ecological incidents have been reported to the Agency that involve any
plants and this was classified with a certainty index of "probable" or higher
linking endosulfan as a cause, despite that it is  regularly directly applied on
or near a very wide variety of agricultural plants.	
No effect (NE)
 Based on the conclusions of this assessment, a formal consultation with the U. S. Fish and
 Wildlife Service under Section 7 of the Endangered Species Act should be initiated.
 When evaluating the significance of this risk assessment's direct/indirect and adverse habitat
 modification effects determinations, it is important to note that pesticide exposures and
 predicted risks to the listed species and its resources (i.e., food and habitat) are not expected to
 be uniform across the action area.  In fact, given the assumptions of drift and downstream
 transport (i.e.., attenuation with distance), pesticide exposure and associated risks to the
 species and its resources are expected to decrease with increasing distance away from the
 treated field or site of application.  Evaluation of the implication of this non-uniform
 distribution of risk to the species would require  information and assessment techniques that
 are not currently available.

 When evaluating the significance of this risk assessment's direct/indirect and adverse habitat
 modification effects determinations, it is important to note that pesticide exposures and
                                              230

-------
predicted risks to the species and its resources (i.e., food and habitat) are not expected to be
uniform across the action area. In fact, given the assumptions of drift and downstream
transport (i.e., attenuation with distance), pesticide exposure and associated risks to the
species and its resources are expected to decrease with increasing distance away from the
treated field or site of application. Evaluation of the implication of this non-uniform
distribution of risk to the species would require information and assessment techniques that
are not currently available.  Examples of such information and methodology required for this
type of analysis would include the following:

          •   Enhanced information on the density and distribution of CRLF, CTS, SFGS,
              BCB, VELB, SMHM, and SJKF life stages within the action area and/or
              applicable designated critical habitat.  This information would allow for
              quantitative extrapolation of the present risk assessment's  predictions of
              individual effects to the proportion of the population extant within
              geographical areas where those effects are predicted. Furthermore, such
              population information would allow for a more comprehensive evaluation of
              the significance of potential resource impairment to individuals of the  assessed
              species.
          •   Quantitative information  on prey base requirements for the assessed species.
              While existing information provides a preliminary picture  of the types of food
              sources utilized by the assessed species, it does not establish minimal
              requirements to sustain healthy individuals at varying life  stages.  Such
              information could be used to establish biologically relevant thresholds of
              effects on the prey base, and ultimately establish geographical limits to those
              effects.  This information could be used together with the density data
              discussed above to characterize the likelihood of adverse effects to individuals.
          •   Information on population responses of prey base organisms to the pesticide.
              Currently, methodologies are limited to predicting exposures and likely levels
              of direct mortality, growth or reproductive impairment immediately following
              exposure to the pesticide. The degree to which repeated exposure events  and
              the inherent demographic characteristics of the prey population play into the
              extent to which prey resources may recover is not predictable. An enhanced
              understanding of long-term prey responses to pesticide exposure would allow
              for a more refined determination of the magnitude and duration of resource
              impairment,  and together with the information described above, a more
              complete prediction of effects to individual species and potential modification
              to critical habitat.
                                        231

-------
  8  References
Registrant Submitted Studies
MRID 000369-35    Atkins. E.L., E.A. Greywood, and R.L. MacDonald.  1975. Toxicity of
      pesticides and other agricultural chemicals to honey bees.  Laboratory studies. Univ.
      of Calif., Div. Agric. Sci. Leaflet 2287.  38 pp.

MRID 45421402     Gries, T.; van der Kolk, J. (2000) (Carbon 14) Endosulfan sulfate
      (metabolite of (Carbon 14-Endosulfan) Acute Toxicity Test with Carp (Cyprinus
      carpio) under Static Conditions: Lab Project Number: 1049.014.170: C010869.
      Unpublished study prepared by Springborn Laboratory (Europe) AG. 52 p.

MRID 45421403     Gries, T.; van der Kolk, J. (2000) (Carbon 14) Endosulfan sulfate
      (metabolite of (Carbon 14-endosulfan) Acute Immobilisation Test with Daphnids
      (Daphnia magnd) under Static Conditions: Lab Project Number: 1049.014.110:
      C012413. Unpublished study prepared by Springborn Laboratory (Europe) AG. 64 p.

MRID 22923        Hill, E.F.; Heath, R.G; Spann, J.W.; et al. (1975) Lethal Dietary
      Toxicities of Environmental Pollutants to Birds: Special Scientific Report—Wildlife
      No. 191. (U.S. Dept. of the Interior, Fish and Wildlife Service, Patuxent Wildlife
      Research Center; unpublished report).

MRID 5005824      Schimmel, S.C.; Patrick,  J.M., Jr.; Wilson, A.J., Jr. (1977) Acute
toxicity to and bioconcentration of endosulfan by estuarine animals. Pages 241-252, in
Aquatic Toxicology and Hazard Evaluation. Edited by F.L. Mayer and J.L. Hamelink.
Philadelphia, Pa.: American Society for Testing and Materials.

MRID 38307        Reno, F.E. (1975) Final Report: Acute Oral Toxicity Study in Rats:
      Project No. 915-108. (Unpublished study received Jul 28, 1980 under 2749-487;
      prepared by Hazleton Laboratories America, Inc., submitted by Aceto Chemical Co.,
      Inc., Flushing, N.Y.; CDL: 243082-A)

MRID 1328         Earnest, R. (1970) Annual progress report: 1970 effects of pesticides on
      aquatic animals in the estuarine and marine environment. (Unpublished  study received
      Nov 26, 1976 under 241-132; prepared by U.S. Fish and Wildlife Service, fish-
      Pesticide Research Laboratory, submitted by American Cyanamid Co., Princeton,
      N.J.; CDL: 129448-W).

MRID 1999         Atkins, E.L.,Jr; Anderson, J.D. (1967). Toxicity of pesticides and other
      agricultural chemicals to honey bees: laboratory studies. (Unpublished study received
      Jan. 30, 1969 under 9G0802; prepared by Univ. of California-Riverside, Dept. of
      Entomology, submitted by Hercules, Inc., Agricultural Chemicals, Wilmington, Del.;
      CDI: 093111-D).
                                       232

-------
MRID 38806        Buccafusco, RJ. (1976) Acute Toxicity of Endosulfan Technico to
      Bluegill (Lepomismacrochirus). (Unpublished study received Jul 28, 1980 under
      2749-487; prepared by EG&G, Bionomics, sub- mitted by Aceto Chemical Co., Inc.,
      Flushing, N.Y.; CDL: 243084-A)

MRID 136998       Roberts, N.; Phillips, C. (1983) The Acute Oral Toxicity (LD50) of
      Endosulfan-Technical (Code: HOE 002671 OIZD97 0003) to the Mallard Duck:
      HST 226/83493. (Unpublished study received Dec 27,  1983 under 8340-13; prepared
      by Huntingdon Research Centre, Eng., submitted by American Hoeschst Corp.,
      Somerville, NJ; CDL: 252043-F).

MRID 137189       Roberts, N.; Phillips, C. (1983) The Acute Oral Toxicity (LD50) of
      Endosulfan Technical (Code: Hoe 002671 OI ZD97 0003) to the Bobwhite Quail:
      HST 224/83566; A27035. (Unpublished study received Jan 16, 1984 under 11678-5;
      prepared by Huntingdon Research Centre, Eng., submitted by Makhteshim Beer Sheva
      Chemical Works, Ltd., New York, NY; CDL:252229-A).

MRID 146843       Roberts, N.; Phillips, C.; Dawe, I; et al. (1985) The Effects of Dietary
      Inclusion of Endosulfan—Technical on Reproduction in the Mallard Duck. HRC
      Report No. HST 228b/841012. Unpublished study prepared by Huntington Research
      Centre pic. 211 p.

MRID 148264       Edwards, J.; Reid, Y.; Offer, J.; et al. (1984) Effect of Endosulfan-
      Technical on Reproductive Function of Multiple Generations of the Rat: Report No.
      HST 204/83768. Unpublished study prepared by Huntingdon Research Centre pic. 422
      P-

MRID 160000       Hudson, R.; Tucker, R.; Haegele, M. (1984) Handbook of toxicity of
      pesticides to wildlife: Second edition. US Fish and Wildlife Service: Resource
      Publication 153.  91 p.

MRID 4640601       Cafarella, M. (2003) Endosulfan Sulfate - Acute Toxicity to Mysids
      (Americamysis bahia) Under Flow-Through Conditions. Project Number: 13798/6137,
      EBENX015, B004570. Unpublished study prepared by Springborn Smithers
      Laboratories.  53  p.

MRID 5004151       Stevenson, J.H. (1968) Laboratory studies on the acute contact and oral
      toxicities of insecticides to honeybees. Annals of Applied Biology 61 (3): 467-472.

MRID 5008271       Macek, K.J.; Lindberg, M.A.; Sauter, S.; Buxton, K.S.; Costa, P.A.
      (1976) Toxicity of Four Pesticides to Water Fleas and Fathead Minnows. Duluth,
      Minn.: U.S. Environmental Protection Agency, Environmental Research Laboratory.
      (EPA report no. EPA-600/3-76-099; available from: NTIS, Springfield, VA; PB-262
      912)
                                      233

-------
MRID 5008936      Clinch, P.O. (1967) The residual contact toxicity to honey bees of
      insecticides sprayed on white clover (Trifolium repens L) in the laboratory. New
      Zealand Journal of Agricultural Research 10(2): 289-300.

MRID 5012881      Gorecki, K. (1973) Zatrucia Pszczoly Miodnej—Apis mellifica L.
      (Hym. Apidae) Insektycydami Stosowanymi W Kraju. [Harmful effects of insecticides
      used in Poland on the honey bee—Apis Mellifica L. (Hym. Apidae).] Polskie Pismo
      Entomologiczne. [Polish Journal of Entomology] 4(1):201-210.

MRID 40335001     Beavers, J.; Frank, P.; Jaber, M. (1987) Endosulfan Technical Sub-
      stance (Code: HOE 002671 OIZD95 0005): A One-generation Reproduction Study
      with the Mallard (Anasplatyrhynchos): Lab Project No. 125-137. Unpublished study
      prepared by Wildlife Inter- national Ltd. 146 p.

MRID 40335002     Beavers, J.; Frank, P.; Jaber, M. (1987) Endosulfan Technical Sub-
      stance (Code: HOE 002671 OI ZD95 0005): A One-generation Reproduction Study
      with the Bobwhite (Colinus virginianus): Lab Project No.  125-134. Unpublished study
      prepared by Wildlife International Ltd. 144 p.

MRID 45868601     Dionne, E. (2002) Endosulfan: The Chronic Toxicity to the Fathead
      Minnow (Pimephalespromelas) During Full Life-Cycle Exposure: Lab Project
      Number: 13726.6140: B004189. Unpublished study prepared by Springborn Smithers
      Laboratories. 142 p.

MRID 46382604     Cafarella, M. (2003) Endosulfan Sulfate - Acute Toxicity to Bluegill
      Sunfish (Lepomis macrochirus) Under Flow-Through Conditions. Project Number:
      13798/6127, 33278/EBENX, B004613. Unpublished study prepared by Springborn
      Smithers Laboratories. 53 p.

MRID 46382603     Christ

MRID 46382601     Christ, M.; Larn, C. (2004) Technical Endosulfan-Sulfate (A
      Metabolite of Endosulfan): A Subacute Dietary LC50 with Mallards. Project Number:
      ES720801, 201058, B004669. Unpublished study prepared by Bayer Corp. 29 p.

MRID 46382605     Putt, A.E. (2004) Endosulfan Sulfate toxicity to midge (Chironomus
      tentans) during a 10-day sediment exposure. Springborn Smithers laboratories,
      Wareham, MA. Laboratory Report ID 13798.6142.

MRID 46430502     Sabbert, T. (2004) Endosulfan Sulfate Technical: A Subactute Dietary
      LC50 with Northwhite Bobwhite. Project Number: ES/721701, 200619. Unpublished
      study prepared by Bayer Cropscience LP. 36 p.

MRID 46781601     Cafarella, M. (2006) Endosulfan Sulfate - Life-Cycle Toxicity Test
      withMysids (Americamysis bahia). Project Number: EBENM019, 13798/6178.
                                      234

-------
       Unpublished study prepared by Springborn Smithers Laboratories. 71 p. Relates to
       L0001048.

MRID 46929001     Putt, A. (2006) Endosulfan Sulfate - Toxicity to Estuarine Amphipods
       (Leptocheirus plumulosus) During a 28-Day Sediment Exposure. Project Number:
       13798/6186, EBENX023. Unpublished study prepared by Springborn Smithers
       Laboratories. 60 p. Relates to LOGO 1083.

MRID 47318101 Weltje, L; Pupp, A. (2008) Endosulfan sulfate - life cycle toxicity testing
       exposing midges (Chironomus tentans) to a test substance applied to sediment under
       static-renewal conditions following EPA test methods. Springborn Smithers
       Laboratories, Wareham, MA. Laboratory Report ID. 13798.6185.

Other References

Armitage, J.M. and F.A.P.C. Gobas. 2007. A Terrestrial Food-Chain Bioaccumulation Model
       for POPs. Environ. Sci. Technol., 41(11):4019 -4025.

Arnot, J.A. and F.A.P.C.  Gobas. 2004.  A food web bioaccumulation model for organic
       chemicals in aquatic ecosystems. Environ. Toxicol. Chem. 23(10): 2343-2355.

Barr, C. B., 1991. The distribution, habitat, and status of the Valley Elderberry Longhorn
       Beetle Desmocerus californicus dimorphus Fisher (Coleoptera: Cerambycidae). U.S.
       Fish and Wildlife Service. Sacramento, CA. 134 pp. Available online at:
       www.fws.gov/sacramento/es/documents/VELB_Report/velb_report.htm

Berrill M, Coulson D, McGillivray L, Pauli B. 1998. Toxicity of Endosulfan to Aquatic
       Stages of Anuran Amphibians. Environ Toxicol Chem. 17: 1739-1744. ECOTOX#
       19467.

Bernabo I; Brunelli E; Berg C; Bonacci A; Tripepi S;.2008.Endosulfan Acute Toxicity in
       Bufo bufo Gills: Ultrastructural  Changes and Nitric Oxide Synthase Localization.
       Aquat Toxicol 86(3): 447-456. ECOTOX # 103223.

Broomhall S. 2002.The Effects of Endosulfan and Variable Water Temperature on
       Survivorship and  Subsequent Vulnerability to Predation in Litoria citropa Tadpoles.
       Aquat Toxicol 61(3/4): 243-250. ECOTOX # 65732.

Broomhall SD. 2004.Egg Temperature Modifies Predator Avoidance and the Effects of the
       Insecticide Endosulfan on Tadpoles of an Australian Frog.J Appl Ecol 41(1): 105-113.
       ECOTOX # 73400.

Broomhall S; Shine R;. 2003 .Effects of the Insecticide Endosulfan and Presence of Congeneric
       Tadpoles on Australian Treefrog (Litoria freycinetf) Tadpoles. Arch Environ Contam
       Toxicol 45(2): 221-226. ECOTOX #  71866.
                                      235

-------
Brunelli E, Bernabo H, Berg C, Lundstedt_Enkel K, Bonacci A, Tripepi S. 2009.
       Enviornmentally Relevant Concentrations of Endosulfan Impair Development,
       Metamorphosis and Behaviour in Bufo bufo Tadpoles. Aquatic Toxicology. 91: 135-
       142. Not in ECOTOX.

Burns, L.A. 1997. Exposure Analysis Modeling System (EXAMSII) Users Guide for Version
       2.97.5, Environmental Research Laboratory, Office of Research and Development,
       U.S. Environmental Protection Agency, Athens, GA.

Carsel, R.F. , J.C. Imhoff, P.R. Hummel, J.M. Cheplick and J.S. Donigian, Jr. 1997. PRZM-3,
       A Model for Predicting Pesticide and Nitrogen Fate in Crop Root and Unsaturated Soil
       Zones: Users Manual for Release 3.0; Environmental Research Laboratory, Office of
       Research  and Development, U.S. Environmental Protection Agency, Athens, GA.

Czub, G. and M.S. McLachlan. 2004. Bioaccumulation Potential of Persistent Organic
       Chemicals in Humans. Environ. Sci. Technol.  38:2406-2412.

DeLorenzo ME;Taylor LA;Lund SA;Pennington PL;Strozier ED;Fulton MH;.2002.Toxicity
       and Bioconcentration Potential of the Agricultural Pesticide Endosulfan in
       Phytoplankton and Zooplankton.Arch Environ Contam Toxicol 42(2): 173-181.
       ECOTOX #65915.

Dey M, Gupta A. 2002. Acute Toxicity of Endosulfan on Three Anuran Tadpoles. J
       Ecotoxicol Environ Monit. 12: 61-65. Not in ECOTOX.

Ecobichon, D. J.  1991. Toxic effects of pesticides. In M. O. Amdur, J. Doull and C. Kaassen
       eds., Toxicology The Basic Science of Poisons. Pergamon Press., New York.  Pages
       65 - 622.

Fellers, G.M, L.L. McConnell, D. Pratt, S. Datta. 2004. Pesticides in Mountain Yellow-
       Legged Frogs (Rana Mucosa) from the Sierra Nevada Mountains of California, USA.
       Environmental Toxicology & Chemistry  23 (9):2170-2177.

Fernandez, S., C. Santin, J. Marquinez, and M.A. Alvarez. 2005. Changes in soils  due to
       polderization in coastal plain estuaries. Geophysical Research Abstracts 7,  3pp.

Fisk AT, Norstron RJ, Cymbalisty CC, Muir DCB.  1998.  Dietary accumulation and
       depuration of hydrophobic organochlorines: Bioaccumulation parameters and their
       relationship with the octanol/water partition coefficient. Environ Toxicol Chem
       17:951-961.

Fletcher, J.S., I.E. Nellessen, and T.G. Pfleeger.  1994.  Literature review and evaluation of
       the EPA food-chain (Kenaga) nomogram, and instrument for estimating pesticide
       residues on plants. Environmental Toxicology and Chemistry 13 (9): 1383-1391.
                                      236

-------
Geng, B.R., D. Yao and Q.Q. Xue. 2005. Acute toxicity of the pesticide dichlorvos and the
       herbicide butachlor to tadpoles of four anuran species. Bull. Environ. Contam.
       Toxicol. 75:343-349.

Gupta PK.1976. Endosulfan-Induced Neurotoxicity in Rats and Mice. Bull Environ Contam
       Toxicol 15(6): 708-713. ECOTOX # 103467.

Gupta, P. K. and R. C. Gupta.  1979. Pharmacology, toxicology and degradation of
       endosulfan. A review.  Toxicology 13: 115 - 130.

Hayes, M.P. and M.R. Jennings.  1988.  Habitat correlates of distribution of the California
       red-legged frog (Rana aurora draytonii) and the foothill yellow-legged frog (Rana
       boylii): Implications for management.  Pp. 144-158. In Proceedings of the
       symposium on the management of amphibians, reptiles, and small mammals in North
       America. R. Sarzo, K.E. Severson, and D.R. Patton (technical coordinators). USDA
       Forest Service General  Technical Report RM-166.

Hack R;Ebert E;Leist KH;. 1995.Chronic Toxicity and Carcinogenicity Studies with the
       Insecticide Endosulfan  in Rats and Mice. Food Chem Toxicol 33(11): 941-950.
       ECOTOX # 103384.

Hageman K.J., Simonich S. L., Campbell D.H. Wilson G.R., and D.H. Landers. 2006.
       Atmospheric deposition of current-use and historic use pesticides in snow at National
       Parks in the Western  United States. Environ. Sci. and Tech. 40: 3174-3180.

Hansen DJ;Cripe GM;.1991. Interlab oratory Comparison of the Early Life-Stage Toxicity
       Test Using Sheepshead Minnows (Cyprinodon variegatus). In: M A Mayes and M G
       Barron (Eds ), Aquatic  Toxicology and Risk Assessment, Vol 14, ASTM STP 1124,
       Philadelphia, PA14Q: 354-375. ECOTOX # 14143.

Harris ML;Bishop CA;Struger J;Ripley B;Bogart JP;. 1998.The Functional Integrity of
       Northern Leopard Frog (Ranapipiens) and Green Frog (Rana clamitans) Populations
       in Orchard Wetlands. II. Effects of Pesticides and Eutrophic Conditions on Early Life
       Stage Development. Environ Toxicol Chem 17(7): 1351-1363. ECOTOX # 19300.

Hoerger, F.,  and E.E. Kenaga.  1972.  Pesticide residues on plants: Correlation of
       representative data as a basis for estimation of their magnitude in the environment.  In
       F. Coulston and F. Korte, eds., Environmental Quality and Safety: Chemistry,
       Toxicology, and Technology, Georg Thieme Publ, Stuttgart, West Germany, pp. 9-28.

Hose GC;Lim RP;Hyne RV;Pablo F;.2002.A Pulse of Endosulfan-Contaminated Sediment
       Affects Macroinvertebrates in Artificial Streams. Ecotoxicol Environ Saf 51(1): 44-52.
       ECOTOX # 62267.
                                       237

-------
Hose GC;Lim RP;Hyne RV;Pablo F;.2003.Short-Term Exposure to Aqueous Endosulfan
       Affects Macroinvertebrate Assemblages. Ecotoxicol Environ Saf 56(2): 282-294.
       ECOTOX# 72555.

Jordan, T.E., J.C. Cornwell, R.B. Walter, and J.T. Anderson. 2008. Changes in phosphorus
       biogeochemistry along an estuarine salinity gradient. Limnology and Oceanography
       53(1): 172-184.

Kelly, B.C., Ikonomou, M.G., Blair, J.D., Morin, A.E., and F.A.P.C. Gobas. 2007. Food Web-
       Specific Biomagnification of Persistent Organic Pollutants. Science. 317:236-239.

Kelly, B.C. and F.A.P.C. Gobas. 2003. An Arctic Terrestrial Food-Chain Bioaccumulation
       Model for Persistent Organic Pollutants. Environ. Sci. Technol. 37:2966-2974.

Klaassen CD, & Watkins JB (1999) Casarett & Doull's toxicology: the basic science of
       poisons. 5th ed. New York, MacGraw-Hill Publishing Company 542-547.

Landers, D.H., S.L. Simonich, D.A. Jaffe, L.H. Geiser, D.H. Campbell, A.R. Schwindt, C.B.
       Schreck, M.L> Kent, W.D. Hafner, H.E. Taylor, KJ. Hageman, S. Usenko, L.K.
       Ackerman, Schrlau J. E, N.L. Rose, T.G. Blett, and M.M. Erway. 2008. The Fate,
       Transport, and Ecological Impacts of Airborne Containants in Western National
       Parks (USA). EPA/600/R-07/138. U.S. Environmental Protection Agency, Office of
       Research and Development, NHEERL, Western Ecology Division, Corvallis, Oregon.

LeNoir, J.S., L.L. McConnell, G.M. Fellers, T.M. Cahill, J.N. Seiber. 1999. Summertime
       Transport of Current-use pesticides from California's Central Valley to the  Sierra
       Nevada Mountain Range,USA. Environmental Toxicology & Chemistry 18(12): 2715-
       2722.

Leonard AW;Hyne RV;Lim RP;Chapman JC;.1999.Effect of Endosulfan Runoff from Cotton
       Fields on Macroinvertebrates in the Namoi River. Ecotoxicol Environ Saf 42(2): 125-
       134. ECOTOX# 20012.

Leonova IN;SlynkoNM;. 1996.Comparative Study of Insecticide Susceptibility and Activities
       of Detoxification Enzymes in Larvae and Adults of Cotton Bollworm, Heliothis
       armigera. Arch Insect Biochem Physiol 32(2): 157-172. ECOTOX # 103049.

Leonova IN;Slynko NM;.2004.Life  Stage Variations in Insecticidal Susceptibility and
       Detoxification Capacity of the  Beet Webworm, Pyrausta sticticalis L. (Lep.,
       Pyralidae). J Appl Entomol 128(6): 419-425. ECOTOX #  100430.

LSA Associates, Inc. 2004. Valley Elderberry Longhorn Beetle. Solano HCP/NCCP Solano
       County Water Agency. Species Descriptions. Available online at:
       http ://www. scwa2. com/hcp/Final%20Draft%202.0/ Appendices/ APPX%20B -
       Species%20Descriptions/Valley%20Elderberry%20Longhorn%20Beetle.pdf
                                      238

-------
Madeira OJD;Hebling-Beraldo MJA;.1989.Laboratory Toxicity of Insecticides to Workers of
       Trigona spinipes (F., 1793) (Hymenoptera, Apidae). J Apic Res 28(1): 3-6. ECOTOX
       #51755.

McConnell, L.L., J.S. LeNoir, S. Datta, J.N. Seiber.  1998. Wet deposition of current-use
       pesticides in the Sierra Nevada mountain range, California, USA. Environmental
       Toxicology & Chemistry 17(10):1908-1916.

McKenney C Jr.;Matthews E;Lawrence D;.1981.Mysid Life-Cycle Tests.In: Prog Rep , FY81,
       Exp Environ Branch, U S EPA, Gulf Breeze, FL(): 62-73. ECOTOX # 3736.

McLachlan, M.S. 1996. Bioaccumulation of Hydrophobic Chemicals in Agricultural Food
       Chains. Environ. Sci. Technol. 30:252-259.

Means, J.C. 1995. Influence of salinity upon sediment-water partitioning of aromatic
       hydrocarbons. Marine Chemistry 51(1): 3-16.

Mosleh YY;Paris-Palacios S;Couderchet M;Vernet G;.2002.Biological Effects of Two
       Insecticides on Earthworms (Lumbricus terrestris L.) Under Laboratory
       Conditions.Med Fac Landbouww Univ Gent67(2): 59-68. ECOTOX # 87129.

Mosleh YY;Paris-Palacios S;Couderchet M;Vernet G;.2003.Acute and Sublethal Effects of
       Two Insecticides on Earthworms (Lumbricus terrestris L.) Under Laboratory
       Conditions. Environ Toxicol 18(1): 1-8. ECOTOX # 86741.

Parikh S.J., J. Chorover, and W.D Burgos. 2004. Interaction of phenanthrene and its primary
       metabolite (l-hydroxy-2-naphthoic acid) with estuarine sediments and humic
       fractions. Journal of Contaminant Hydrology 72(1-4): 1-22.

Rauh J. J., E. Benner, M. E. Schnee, D. C.ordova, C. W. Holyoke, M. H. Howard, D. Bai, S.
       D. Buckingham, M. L. Hutton, A. Hamon, R. T. Roush, and D.B. Sattelle. 1997.
       Effects of [3H]-BIDN, a novel bicyclic dinitrile radioligand for GABA-gated chloride
       channels of insects and vertebrates. British Journal of Pharmacology 121 (7):  1496 -
       1505.

Schimmel SC;. 1981.Results: Interlaboratory Comparison  - Acute Toxicity Tests Using
       Estuarine Animals.Final Draft, EPA  600/4-81-003, U S EPA, Gulf Breeze, FL: 13 p.
       ECOTOX #3740

Sharpe, S. and D. Mackay. 2000. A Framework for Evaluating Bioaccumulation in Food
       Webs. Environ. Sci. Technol. 34:2373-2379.

Shenoy K, Cunningham BT, Renfroe JW, Crowley PH. in press. Growth and Survival of
       Northern Leopard Frog (Ranapipiens) Tadpoles Exposed to Two Common Pesticides.
       Environ Toxicol Chem. not in ECOTOX.
                                      239

-------
Siddique, Okeke, Arshad and Frankenberger. 2003. Biodegradation kinetics of endosulfan by
      Fusarium ventricosum and aPandoraea species. J. Agric. FoodChem. 51, 8015-8019

Singh ND;Sharma AK;Dwivedi P;Patil RD;Kumar M;.2007.Citrinin and Endosulfan Induced
      Teratogenic Effects in Wistar Rats. J Appl Toxicol 27(2): 143-151. ECOTOX #
      103238.

SinhaN;Narayan R;Shanker R;Saxena DK;. 1995.Endosulfan-Induced Biochemical Changes
      in the Testis of Rats. Vet Hum Toxicol 37(6): 547-549. ECOTOX # 103592.

Sparling, D.W.,  G.M. Fellers, L.L. McConnell. 2001.  Pesticides and amphibian population
      declines in California, USA. Environmental Toxicology & Chemistry 20(7): 1591-
      1595.

Sunderam RIM;Cheng DMH;Thompson GB;.1992.Toxicity of Endosulfan to Native and
      Introduced Fish in Australia.Environ Toxicol Chem 11(10): 1469-1476 (OECDG Data
      File). ECOTOX # 5850.

Swarzenski, P.W., D. Porcelli, P.S. Andersson, and J.M Smoak.  2003. The behavior of U- and
      Th-series nuclides in the estuarine environment. Reviews in Mineralogy and
      Geochemistry 52(1):  577-606.

Talley, T.S., D. Wright and M. Holyoak. 2006. Assistance with the 5-year review of the falley
      elderberry longhorn beetle (Desmocerus californicus dimorphus).  US FWS,
      Sacramento. Available online at:
      www.fws.gov/sacramento/es/documents/VELB  5yr review Talleyetal.pdf

Talley, T.S. No date. VELB Info. Available online at:
      http://web.mac.com/tsinicrope/Stie/VELB_info/Entries/2006/ll/4_Life-cycle.html

Teske, Milton E., and Thomas B. Curbishley. 2003.AgDisp ver 8.07 Users Manual. USDA
      Forest Service, Morgantown, WV.

Urban, DJ. and NJ. Cook. 1986. Hazard Evaluation Division Standard Evaluation Procedure:
      Ecological Risk Assessment. Office of Pesticide Programs, US. EPA. Report # 540/9-
      85-001. 96 pp.

U.S. EPA. 2008. White Paper on Methods for Assessing Ecological Risks of Pesticides with
      Persistent, Bioaccumulative and Toxic Characteristics Submitted to the FIFRA
      Scientific Advisory Panel For Review and Comment.  October 28-31, 2008. Office of
      Prevention, Pesticides and Toxic Substances, Office of Pesticide Programs,
      Environmental Fate and Effects Division, Washington, D.C. (available at:
      http://epa.gov/oscpmont/sap/meetings/2008/october/sapjbt whitepaper final Oct  7
        08d.pdf)
                                      240

-------
U.S. EPA. 2004. Overview of the Ecological Risk Assessment Process in the Office of
       Pesticide Programs. Office of Prevention, Pesticides, and Toxic Substances. Office of
       Pesticide Programs. Washington, D.C. January 23, 2004.

U.S. EPA. 2003. Methodology for Deriving Ambient Water Quality Criteria for the Protection
       of Human Health (2000), Technical Support Document Volume 2: Development of
       National Bioaccumulation Factors.  Office of Water. EPA-822-R-03-030. Washington,
       D.C

U.S. EPA. 2000. Methodology for Deriving Ambient Water Quality Criteria for the Protection
       of Human Health (2000). Office of Water, EPA-822-B-00-004. Washington, DC.
       (available at:
       http://www.epa.gov/waterscience/criteria/humanhealth/method/index.html)

U.S. Environmental Protection Agency (U.S.  EPA). 1998. Guidance for Ecological Risk
       Assessment. Risk Assessment Forum. EPA/630/R-95/002F, April 1998.

U.S. EPA. 1997. Incidence and Severity of Sediment Contamination in Surface Waters of the
       United States (Volume 1 EPA 823-R-97-006; Volume 2 EPA 823-R-97-007; Volume
       3 EPA 823-R-97-008). Office of Water, Washington, DC.
       http://www.epa.gov/waterscience/cs/report/1997/index.htmltfreport

U.S. Fish & Wildlife Service (USFWS). 2006. 5-Year Review: Summary and Evaluation for
       the Valley Elderberry Longhorn Beetle (Desmocerus californicus dimorphus).
       USFWS, Sacramento. 28 pp. Available online at:
       http://ecos.fws.gov/docs/five year  reviewAioc779.pdf

U.S. Fish & Wildlife Service (USFWS). 2007a. Species Account: Valley Elderberry
       Longhorn Beetle (Desmocerus californicus dimorphus). USFWS, Sacramento.
       Available online at: www.fsw.gov/sacramento/es/animal spp  acct/velb.pdf

U.S. Fish & Wildlife Service (USFWS). 2007b. Species Account—Kid-Friendly: Valley
       Elderberry Longhorn Beetle. USFWS, Sacramento. Available online at:
       www.fsw.gov/sacramento/es/animal  spp acct/valey elderberry  longhorn beetle_kf.
       htm

U.S. Fish & Wildlife Service (USFWS). 2006. Endangered and threatened wildlife and plants:
       determination of critical habitat for the California red-legged frog. 71 FR 19244-
       19346.

U.S. Fish & Wildlife Service (USFWS). Website accessed: 30 December 2006.
       http://www.fws.gov/endangered/features/rl  frog/rlfrog.html#where

U.S. Fish and Wildlife Service (USFWS). 1996. Endangered and threatened wildlife and
       plants: determination of threatened status for the California red-legged frog. Federal
       Register 61(101):25813-25833.
                                       241

-------
U.S. Fish & Wildlife Service (USFWS).  1984. Valley elderberry longhorn beetle recovery
       plan. Portland, Oregon. 62 pp. Sacramento, CA. 134 pp. Available online at:
       www.iws.gov/sacramento/es/recoveiYjlans/velb recovery_plan.pdf

U.S. Fish and Wildlife Service (USFWS) and National Marine Fisheries Service (NMFS).
       1998.  Endangered Species Consultation Handbook: Procedures for Conducting
       Consultation and Conference Activities Under Section 7 of the Endangered Species
       Act. Final Draft. March 1998.

USFWS/NMFS.  2004.  50 CFR Part 402. Joint Counterpart Endangered Species Act Section
       7 Consultation Regulations; Final Rule. FR 47732-47762.

Velde B. and T. Church. 1999. Rapid clay transformations in Delaware salt marshes. Applied
       Geochemistry 14(5): 559-568

Verma, Agrawal, Faroog, Misra and Hans. 2006. Endosulfan degradation by a Rhodococcus
       strain isolated from earthworm gut. Ecotoxicology and Environmental Safety 64:377-
       381.

Wan MT;Kuo JN;Buday QSchroeder G;Van Aggelen G;Pasternak J;.2005.Toxicity of
       [alpha]-, [beta]-, ([alpha]+[beta])-Endosulfan and Their Formulated and Degradation
       Products to Daphnia magna, Hyalella azteca, Oncorhynchus mykiss,  Oncorhynchus
       kisutch, and Biological Implications in Streams. Environ Toxicol Chem 24(5): 1146-
       1154. ECOTOX# 87973.

Willis, G.H. and L.L. McDowell. 1987.  Pesticide Persistence on Foilage in Reviews of
       Environmental Contamination and Toxicology.  100:23-73.

Weston D.P. J. You., and MJ. Lydy. 2004. Distribution and toxicity of sediment-associated
       pesticides in agriculture-dominated water bodies of California's Central Valley.
       Environ. Sci. and Tech. 38: 2752-2759

Wood, T.M. and A.M. Baptista. 1993.  A model for diagnostic analysis of estuarine
       geochemistry. Water Resources Research 29(1): 51-71.
                                       242

-------