Document XOGRBzKye67d0gRbrk1b4LbRx

r Afszb - \h-l*h 000161 DRAFT REPORT DRY RUN CREEK WASHINGTON, WOOD COUNTY, WEST VIRGINIA NOVEMBER 1997 PREPARED BY: Mark D. Sprenger, Ph.D. Environmental Response Team AND Michael T. Horne, Ph.D. U.S. Fish & Wildlife Service/Environraental Response Team IN CONJUNCTION WITH: Mark Huston REAC/ERT Environmental Response Team Center Office of Emergency & Remedial Response 000162 USEPA 6861 0 0 0 - it < .\ USFW 0579 TABLE OF CONTENTS LIST OF T A B L E S .............................................................................................................................................................. .... LIST OF FIGURES .........................................................................................................................................................viii SECTION I. 1.0 2.0 3.0 TECHNICAL APPROACH. SUMMARY OF FIELD EFFORT RESULTS. AND PRELIMINARY RISK S C R E E N ..................................................................................................................................... 1 IN TRO D U CTIO N ................................................................................................................................ 1 1.1 Objective ........................................................................................ ......................................... 1 1.2 Site B ackground...................................................................................................................... I METHODOLOGY ................................................................................................................... 1 2.1 Soil Sampling ........................................................................................................................ 1 2.2 Sediment S am pling................................................................................................................ 2 2.3 Surface Water Sampling ..........................................................................................................2 2.4 Drinking Water Well S a m p lin g ....................................................................... '. ................ 2 2.5 Biological S a m p lin g .............................................................................................................. 3 2.5.1 Small Mammal S tu d y .............................................................................................. 3 2.5.2 Vegetation S a m p lin g .............................................................................................. 3 2.5.3 Aquatic Macroinvenebrate S am pling.................................................................. 4` 2.5.4 Fish C ollection......................................................................................................... 4 2.6 Toxicity T estin g ...................................................................................................................... 4 2.6.1 Eisenia foerida (Earthworm) Toxicity T e s ts ........................................................ 4 2.6.2 Hyalella azseca (Amphipod) Toxicity T e s ts ........................................................ 5 2.6.3 Pimephalespromelas (Minnow) Toxicity Tests ............................................... 5 2.7 Sampling Equipment Decontamination................................................................................ 5 2.8 Standard Operating P ro c e d u re s........................................................................................... 5 2.8.1 Docum entation......................................................................................'.................. 5 2.8.2 Sample Packaging, Shipment, Storage, Preservation, and Handling ........... 5 2.8.3 Field Sampling and Analytical T echniques........................................................ 5 2.8.4 Health and S a f e ty .................................................................................................... 6 R E S U L T S ..................................................................... ............................................................... '. . . 1 6 3.1 Water, Soil, and SedimentAnalysis ..................................................................................... 6 3.1.1 BNAs ............................................................................................................. 6 C 0 k-J A -- 000163 USEPA 6862 USFW 0580 3.1.2 TAL M etals............................................................................................................ 7 3.1.3 Pesticides/PC B s.................................................................................................... 8 3.1.4 3.1.5 3.1.6 VOCs ..................................................................................................................... Total Fluoride ....................................................................................................... O rganofluorides..................................................................................................... 8 9 9 3.1.7 Total Organic Carbon and Grain Size of Soil and Sedim ent..............................10 3.1.8 Water Quality P aram eters.......................................................................................10 3.1.9 Bovine Fecal Samples ............................................................................................ 10 3.2 Biotic Sampling and Tissue A nalysis.................................................................................... 10 3.2.1 Benthic M acroinvenebrates....................................................................................11 3.2.2 M am m al.................................................................................................................... 12 3.2.3 F is h .............................................................................................................................13 3.2.4 Earthworm ............................................................................................................... 13 3.2.5 V egetation..................................................................................................................14 3.3 Histological assay of small mammal liver and k id n e y .................................................... 14 . 3.4 Toxiciry T esting..................................................................................................................... 14 4.0 SUMMARY OF PRELIMINARY ECOLOGICAL RISK ASSESSMENT S C R E E N .................15 5.0 DISCUSSION ........................................................................................................................................15 SECTION II 1.0 2.0 ECOLOGICAL RISK ASSESSMENT ............................................................................... 16 INTRODUCTION.................................................................................... : .........................................16 1.1 O b je c tiv e .................................................................................................................................. 16 1.2 Site Background........................................................................................................................16 PROBLEM FORMULATION ...........................................................................................................16 2.1 Ecological Risk Assessment ..................................................................................................16 2.2 Identification of the Contaminants of C o n cern ................................................................. .1 7 2.3 Exposure Characterization .....................................................................................................17 2.4 Hazard Characterization/Toxicity A ssessm ent.................................................................... 17 2.4.1 F lu o rid e ..................................................................................................................... 17 2.4.2 O rganofluorides........................................................................................................18 2.4.3 A lum inum .................................................................................................................. 18 2.4.4 Arsenic ..................................................................................................................... 18 2.4.5 B ery lliu m ...................................................................................................................19 n USEPA 6863 0 0 0 - ;. : 000164 USFW 0581 2.4.6 C hrom ium ...................................................................................................................19 2.4.7 C o p p e r........................................................................................................................20 2.4.8 I r o n ................................................ 21 2.4.9 Lead ...........................................................................................................................21 2.4.10 Manganese ................................................................................................................21 2.4.11 Nickel ........................................................................................................................22 2.4.12 V anadium ...................................................................................................................22 2.4.13 Zinc .......................................................................................................................... 23 2.5 Selection of Assessment E n d p o in ts....................................................................................... 23 2.6 Production of Testable H ypotheses....................................................................................... 24 2.7 Conceptual Model ...................................................................................................................25 2.8 Selection of Measurement E ndpoints..................................................................................... 26 2.9 Life History/Exposure Profile Inform ation.......................................................................... 29 2.9.1 The amphipod {HyalleLa azteca) as Representative o f Benthic Invertebrates ................................................................................................................................... 29 2.9.2 Earthworm (Eisenia foetida) as Representative o f Terrestrial Invertebrates ................................................................................................................................... 30 2.9.3 Fathead Minnow (Pimephalespromelas) as Representative of FishCommunity 31 2.9.4 American Robin (Turdus migratorius) as Representative of Worm-eating Birds ................................................................................................................................... 32 2.9.5 Red-tailed Hawk (Buteo jamaciertsis) as Representative of Carnivorous Birds. ..................................................................................................................................33 2.9.6 Red Fox ( Vulpes vulpes) as Representative of Carnivorous Mammals . . . . 34 2.9.7 Mink (Aiusiela vison) as Representative of Carnivorous M a m m a ls ................ 35 2.9.8 Raccoon (Procyon lotor) as Representative of Omnivorous Mammals . . . . 37 2.9.9 Short-tailed Shrew (Biarina brevicauda) as Representative of Insectivorous M a m m a ls................................................................................................................. 38 2.9.10 Meadow Vole (Microtus penrtsylvanicus) as Representative of Herbivorous M a m m a ls..................................................................................................................40 A SSU M PTIO N S.....................................................................................................................................42 EFFECTS P R O F IL E .............................................................................................................................43 4.1 F lu o rid e ......................................................................................................................................43 m eo o 4 ^ o o o l6 5 4.2 O rganofluorides................................................................... 4.3 Aluminum ............................................................................... 4.4 Arsenic ................................................................................. 4.5 Beryllium '............................................................................... 4.6 Chromium ............................................................................ 4.7 Copper ................................................................................. 4.8 Iron ....................................................................................... 4.9 Lead ....................................................................................... 4.10 M an g an ese............................................................................ 4.11 N ic k e l.................................................................................... 4.12 V anadium ............................................................................... 4.13 Zinc ....................................................................................... 5.0 RISK CHARACTERIZATION......................................................... 5.1 Benthic Invertebrate Community Structureand Function 5.2 Soil Invertebrate Community Structure and Function . . 5.3 Fish Communities ............................................................ 5.4 Worm-eating B ird s .............................................................. 5.5 Carnivorous Birds .............................................................. 5.6 Carnivorous Mammals (Terrestrially fe ed in g )................ 5.7 Piscivorous Mammals ......................................................... 5.8 O mnivorous Mammals ...................................................... 5.9 Insectivorous M am m als..................................................... 5.10 Herbivorous Mammals ...................................................... 6.0 UNCERTAINTY ANALYSIS ......................................................... 7.0 CONCLUSION S................................................................................. 7.1 Benthic Invertebrate Community Structureand Function 7.2 Soil Invertebrate Community Structure and Function . 7.3 Fish Communities ............................................................ 7.4 Worm-eating B ird s ............................................................ 7.5 Carnivorous Birds .............................................. .. 7.6 Carnivorous Mammals .................................................... 7.7 Piscivorous M am m als....................................................... 7.8 Omnivorous Mammals .................................................... o 000166 43 44 44 44 45 45 45 46 46 46 47 47 47 48 48 48 48 48 48 49 49 49 49 50 50 50 , 51 . 51 . 51 . 51 7.9 Insectivorous M am m als.......................................................................................................... 52 7.10 Herbivorous Mammals .................................................................................. 52 8.0 SU M M A R Y ........................................................................................................ ................................... ;2 LITERATURE C I T E D .........................................................................................................................................................53 APPENDIX A Small Mammal Data S h eets................................................................................................................................... 60 APPENDIX B Analytical Reports ................................................................................................................................................. 61 APPENDIX C Toxicity Testing R ep o rts......................................................................................................................................... 62 APPENDIX D Field N o t e s ............................................................................................................................................................... 63 APPENDIX E Statistical Analysis ..................................................................................................................................................64 000167 000 V USEPA 6866 LIST OF TABLES NUMBER 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 TEILE Concentration of BNA's in Water Concentration of BNA's in Soil Concentration of BNA's in Sediment Concentrations of Metals in Water Results of Concentrations of Metals in Soil Results of Concentrations of Metals in Sediment Results of the Analysis for Pesticide/PCB in Water Results of the Analysis for Pesticide/PCB in Soil Results of the Analysis for Pesticide/PCB in Sediment VOA Concentrations in Water VOA Concentrations in Soil VOA Concentrations in Sediment Concentrations of Fluoride in Water Concentrations of Fluoride in Soil Concentrations of Fluoride in Sediment Results of the Organo-fluoride Analysis in Sediment Concentrations of TOC in Soil Results of the Analysis for Grain Size in Soil Concentrations of TOC in Sediment Results of the Analysis for Grain Size in Sediment In Situ Water Quality Parameters -Concentrations of Bromide, Chloride, Nitrate, Phosphorus, and Sulfate in Water VI O G O -.O v 000168 USEPA 6867 USFW 0585 23 Concentrations of BNA's in Fecal Samples 24 Concentrations of Metals in Fecal Samples 25 Concentrations of Fluoride in Fecal Samples 25 Frequency and Abundance of Benthic Macroinvertebrates 27 Concentrations of Metals in Small Mammals 28 Concentrations of Fluoride in Small Mammals 29 Lipid Concentrations in Mammal Tissue 30 Concentrations of Metals in Fish Tissue 31 Concentrations of Puoride in Fish Tissue 32 Lipid Concentrations in Fish Tissue 33 Results of the Analysis for TAL Metals in Earthworm Tissue 34 Results of the Analysis for Fluoride in Earthworm Tissue 35 Lipid Concentrations in Earthworm Tissue 16 Concentrations of Metals in Vegetation 7 Concentrations of Fluoride in Vegetation Lipid Concentrations in Plant Tissue Results of Histopathology for the (Meadow Vole, Short-tail shrew, Meadow jumping mouse, and White-footed mouse) Summary of Toxicity Test Results Summary of Initial Risk Screen Risk Calculations Based on Wet Weight * u s e y ^ 6S6S ooS1v**i'>. ` vii 0 * 6 9 NUMBER 1 LIST OF FIGURES IITLE Sampling Site Map vm USEPA 6869 0 0 9 -U i', 000170 USFW 0587 SECTION I. TECHNICAL APPROACH, SUMMARY OF FIELD EFFORT RESULTS. AND PRELIMINARY RISK SCREEN 1.0 INTRODUCTION 1.1 Objective The objective of this project was to provide technical support to the U.S. Environmental Protection Agency Region III Removal Program in conducting an evaluation of ecological risks from alleged contamination of soil, sediment, and water at a working beef production farm located down gradient of a landfill The effort resulted in the collection of soil, sediment, surface water, and biota samples for contaminant analyses and soil, sediment, and surface water for laboratory toxicity testing. The primary goals of the project were to: 1) identify contaminan ts present, 2) determine the extent of contamination, and 3) produce an ecological risk assessment based on the collected data. 1.2 Site Background The site is a working beef production farm located in Washington, Wood County, WV. The owner of the farm has filed numerous complaints with the West Virginia Department of Natural Resources and the U. S. EPA alleging that contaminants are being discharged from an industrial landfill owned by the DuPont corporation, into Dry Run. Dry Run flows through the farmer's property and is a primary source of water for his cattle. The farmer maintains that numerous deaths, blindness, and other unusual illnesses observed in his herd are directly attributable to the contaminants that are discharged into Dry Run from the DuPont landfill. It has also been reported that fish and wildlife Vills have also occurred in the area, which may be associated with the abnormalities observed in the cattle. 2.0 METHODOLOGY The approach used in this document followed current U.S. EPA guidance for designing and conducting ecological risk assessments (U.S. EPA 1997). Based on the problem formulation phase of the risk assessment design, the following field study was conducted to provide data needed to complete the assessment. A screening-level ERA was conducted after the field investigation, as little data on site contamination was available prior to the effort. Numerous fish and wildlife kills, in addition to problems in the cattle, had been reported prior to this activation. 2.1 Soil Sampling Surface soil samples were collected at 4 sample areas along Dry Run and in one reference sample area (Figure 1). Sample areas were selected based on distance from the landfill outfall in an attempt to identify1a contaminant concentration gradient. Sampling was concentrated in the meadows along the stream bed. Three replicate samples were taken in each sampling area. Replicate sampling locations were determined by gridding the sampling area and randomly choosing three grid nodes for sampling through the use of a random numbers table. Sampling grid nodes were determined by usinz a random numbers table. Surface soil samples were collected using a decontaminated stainless steel trowel or spoon from the top 6 inches of the soil according to ERTC/REAC Standard Operating Procedure (SOP) *2012, Soil Sampling. All soil samples were analyzed for total organic carbon (TOC); grain size; target analyte GOO- USEPA 6870 ; 000171 USFW 058S list (TAJL) metals: TCL pesticides/PCBs; TCL Base, Neutral, and Acid Extractable (BNAs) compounds: TCL volatile organic compounds (VOCs), total fluoride, and organofluoride compounds. Additional soil was collected from the sample node closest to the stream bed for use in an earthworm toxicity test. A vegetation sample was also taken at each of the soil sampling nodes. 2.2 Sediment Sampling Sediment samples were collected at 5 sample areas on site in Dry Run, one reference sample area, and one area in Lee Creek. Sample areas were selected based on distance from the landfill outfall in an attempt to identify a contaminant concentration gradient. Sampling was concentrated in the depositional areas along the stream bed. All sediment sampling was conducted according to ERTC/REAC SOP #2016, Sediment Sampling. At each sample station, sediment was collected from the top 6 inches using a decontaminated trowel. The sample was composited into a decontaminated 5-gallon stainless steel bucket, homogenized, and divided into the appropriate sample containers for chemical analyses. Additional sediment was collected in the reference area. Tributary A, Tributary B, Area II, and Area IV, for use in a Hyalella azieca whole sediment bioassay. 2.3 Surface Water Sampling Surface water samples were collected at locations which corresponded to each of the seven sediment sample stations. Surface water samples were collected directly into two 1-liter polypropylene bottles for metals analyses and into 1-liter glass bottles for organic (i.e.,' BNAs, Pesticide/PCBs, VOCs) analyses as per ERTC/REAC SOP #2013, Surface Water Sampling. Water samples were collected prior to collecting sediment samples and upstream of any stream disturbances caused by the sampler. One sample at each location was filtered through a 0.45 micron (pm) filter in the field prior to TAL metals analysis; all the remaining TAL metals samples and all the organic samples were analyzed unfiltered. All samples analyzed for metals were preserved by adding 40 percent nitric acid until a pH of less than 2 in the sample was obtained. The filtered sample submined for TAL metal analysis was preserved after the sample was filtered. All surface water samples were submitted for TAL metals, TCL BNAs, TCL Pesticide/PCBs, TCL VOCs, chloride, fluoride, bromide, nitrate, sulfate, and phosphate analyses. Additional sample was taken in the reference area. Tributary A, Tributary B, Area II, and Area IV, for use in a Pimephaies promelas aquaeous phase toxicity test. Water quality parameters were measured using a Horiba"' water quality meter. The meter was used to measure temperature in degrees Celsius (*C), pH, dissolved oxygen [milligrams per liter (mg/L)], conductivity (millimhos per centimeter (mmhos/cm)]. oxidation reduction potential [volts (V)]. The meter was calibrated prior to and after data collection. In-situ water quality data was transcribed from the digital display of the HoribaTM into a field logbook at the time of collection. The Horiba" was used in accordance the manufacturer's operating manual. 2.4 Drinking Water Well Sampling Water was sampled from a drinking water well on the Tennant farm. Parameters were analyzed as outlined above. Samples to be analyzed were taken from a tap that was located directly on the pump head after the well had been purged for a period of approximately five minutes. 2 USEPA 6871 O O Q o.;: 000172 USFW 058! 2.5 Biological Sampling 2.5.1 Small Mammal Study Small mammals were collected from the site to determine body burden levels of TAL metals and total fluoride and to evaluate histopathological effects of exposure to site contaminants. Tissue burdens of small mammals trapped on site were compared to animal collected from the reference area. All field trapping activities were conducted in accordance with ERTC/REAC Draft Standard Operating Procedure SOP #2029, Small Mammal Sampling and Processing. Four trapping areas were established on site in areas corresponding to the soil sampling locations. A fifth grid was established on a reference area located just to the north of Dry Run in similar meadow habitat as that observed along the stream corridor (Figure 1). The reference area was chosen because the habitat present was sim ilar to that in the meadows near Dry Run, and because it was outside the area that could be directly influenced by surface water from Dry Run. The length of the trapping period and the trapping effort varied among each of the trap areas and was based on the length of time and effort required to capture a sufficient number of mammals for statistical evaluation. Sampling was performed using Museum Special snap traps set in grids. All traps were spaced 10 feet apart and baited with a rolled oats and peanut butter mixture. The traps were checked: twice daily, once in the morning and once in the evening. During trap checks, traps were rebaited and reset as necessary. Recovered animals were labeled with the trap area, trap number, species, and date of capture while in the field and then were transferred in coolers to the staging area for processing. For each animal, prior to performing the necropsy, data from the specimen label was transferred to a small mammal data sheet (Appendix A). Body metrics including total body weight, body length, tail length, ear length, liver weight, and kidney weight were measured and recorded on the data sheet. During the necropsy any abnormaliz e were noted and the contents of the gastrointestinal tract were removed from each specimen. Sections of the liver and kidney (approximately 0.5 g each) were removed for histopathological analyses. The sections were placed in a labeled 40-mL glass vial and preserved with 10 percent neutral buffered formalin. Preserved liver and kidney sections were submitted to Animal Reference Pathology (ARP) for histopathological evaluation. The remaining tissue was submitted for homogenization and TAL metal, total fluoride, percent moisture, and percent lipid analysis. 2.5.2 Vegetation Sampling Vegetation was collected by hand for residue analysis per ERTC/REAC SOP #2038 Vegetation Assessment Field Protocol. The most abundant grass taxa observed at all sampling locations was targeted for residue analysis. Grass samples were taken in each area at the same grid nodes as the soil samples were taken. The above ground portion of plants from the immediate vicinity of the soil sampling node were collected by cutting the steins at the soil surface with a decontaminated knife. All samples were analyzed for TAL metals, total fluoride, percent moisture, and percent lipids. 2.5.3 Aquatic Macroinvenebrate Sampling l GO0 USEPA 6872 000173 USFW n.FQO The infaunal macroiavenebrate community was sampled per Draft ERTC/REAC SOP "2032 Bemhic Macroinvertebrate Sampling and U.S. EPA (1983, 1989, and 1990). Macroinvertebrate samples were collected for evaluation of community structure. In this investigation, macroinvertebrates were defined as organisms that impinged on a 0.5 millimeter (mm) sieve. A total of three replicates were collected from each of five sediment sampling locations (Figure 1). A long-handled, D-frame kick net, measuring approximately 45 centimeters wide and 20 centimeters tall, with 0.5 mm mesh was used. The net was used to disturb submerged vegetation and debris and collect dislodged invertebrates. Each replicate collection was performed over a uniform area at each sampling location. Benthic invertebrate samples were transferred to 500 ml polyethylene jars and preserved with a 70 percent 2-propanol solution. In the laboratory, the sample was rinsed in clean water and placed in a white 12 x 18-inch polyethylene pan with just enough water added to allow complete dispersion of the material within the pan. Large debris, stones, and other extraneous materials were removed from the tray and inspected for attached or clinging organisms. All organisms picked from the pan were identified to the lowest positively identified taxonomic level, enumerated, and recorded on a laboratory bench sheet. The size and life history stage of the organisms and state of taxonomic knowledge of the group determined the level of identification. The organisms were identified using appropriate taxonomic references and a representative subsample were identified by a second individual to meet the Quality Assurance,'Quality Control (QA/QC) requirements of the taxonomic analysis. 2:5.4 Fish Collection Fish were collected from Dry Run to determine body burden levels of TAL metals and total fluoride. A CoffeltTM battery powered backpack electroshocker was used and operated as per the manufacturer's instructions. The sampling team consisted of one individual operating the electroshocker and one individual collecting stunned fish with a dip net. Stunned fish were placed in a 5-gallon bucket filled with site water. Following collection, fish were identified to the lowest taxonomic level possible in the field and live specimens were released. Voucher and dead specimens were preserved with a dilute formaldehyde solution and returned to the ERT/REAC biological laboratory for confirmation of field taxonomic analyses. Fish tissue was homogenized and submined to the laboratory for TAL metal, total fluoride, % lipid and % moisture analysis. Toxicity Testing 2.6.1 Eisenia foetida (Earthworm) Toxicity Tests Five soil samples were taken for evaluation in an earthworm toxicity test. Four of the samples were taken in the meadow sampling areas along Dry Run and one in the reference meadow area as outlined above. The test was run for a period of 28 days, at which time mortality and growth in each of the test soils was enumerated. Earthworm tissue resulting from each of the treatments was submined for TAL metals, total fluoride, % lipid, and % moisture analysis. Figure 1 details the earthworm toxicity test soil sampling locations. 2.6.2 Hyalella azteca (Amphipod) Toxicity Tests USEPA 6873 4 000174 USFW 0591 Five sediment samples were taken for evaluation in an amphipod toxicity test. Four of the samples were taken in Dry Run and one in a reference area stream. The test was run for a period of 10 days, at which time mortality and growth in each of the test sediments was enumerated. Figure 1 details the amphipod toxicity test sediment sampling locations. 2.6.3 Pimepholes promelas (Minnow) Toxicity Tests Five surface water samples were taken for evaluation in a fathead minnow toxicity test. Four of the samples were taken in Dry Run and one in a reference area stream. The test was run for a period of 7 days, at which time mortality and growth in each of the test waters was enumerated. Figure 1 details the fathead minnow toxicity test water sampling locations. 2.7 Sampling Equipment Decontamination The following sampling equipment decontamination procedure was employed prior to and subsequent to sampling in the following numerical sequence: 1. physical removal 2. nonphosphate detergent wash 3. potable water rinse 4. 10 percent nitric acid rinse 5. distilled water rinse 6. solvent rinse [acetone] 7. air dry 2.8 Standard Operating Procedures 2.8.1 Documentation Documentation was conducted in accordance with the following SOPs: -ERTC/REAC SOP #2002, Sample Documentation -ERTC/REAC SOP #4001, Logbook Documentation -ERTC/REAC SOP #4005, Chain o f Custody Procedures 2.8.2 Sample Packaging. Shipment, Storage, Preservation, and Handling Sample packaging, shipment, storage, preservation and handling were conducted in accordance with the following SOPs: -ERTC/REAC SOP #2003, Sample Storage, Preservation and Handling -ERTC/REAC SOP #2004, Sample Packaging and Shipment 2.8.3 Field Sampling and Analytical Techniques Field sampling activities and field analytics were conducted in accordance with the following SOPs: -ERTC/REAC SOP #2001, Cenerai Field Sampling Guidelines 5 coo USEPA 6874 000175 USFW 0592 -ERTC/REAC SOP #2005, Quality Assurance/Quality Control Sample: -ERTC/REAC SOP #2006, Sampling Equipment Decontamination -ERTC/REAC SOP #2012, Soil Sampling -ERTC/REAC SOP #2013, Surface Water Sampling -ERTC/REAC SOP #2016, Sediment Sampling -REAC SOP #2029, Small Mammal Trapping and Processing -REAC SOP #2032, Benthic Sampling 2.8.4 Health and Safety Health and Safety was conducted in accordance with the following SOPs: -ERTC/REAC SOP #3001, REAC Health and Safety Program Policy and Implementation -ERTC/REAC SOP #3012, REAC Health and Safety Guidelines at Hazardous Waste Sites -ERTC/REAC SOP #3020, Inclement Weather, Heat Stress and Cold Stress RESULTS 3.1 Water. Soil, and Sediment Analysis 3.1.1 BNAs Surface Water Analysis of the surface water samples from Dry Run. the reference stream, and Lee Creek produced only one detection on the standard BNA scan. A sample taken in the Upper Tributary B location contained an estimated concentration of 2 ug/L of Bis(2EthylhexyOphthalate. In addition, numerous Tentatively Identified Compounds (TICs) including unknown alkane and alkene compounds were found in the surface water samples. Results for the BNA analysis of surface water samples taken in in Dry Run are presented in Table 1 and in Appendix B. Well Water The sample taken at the Tennant Farm well produced no detections from the standard BNA list. Several TICs were identified, however only one of the detected compounds could be tentatively characterized and identified as an alkene. Results for the BNA analysis of the well water sample taken at the Tennant farm is presented in Table 1 and in Appendix B. Sail Analysis of the surface soil samples from the meadows adjacent to the streambed and the reference meadow area produced a few isolated hits from the standard BNA list. Fluoranthene was detected at an estimated concentration of 23 ug/Kg in one of the three reference samples. Carbazole was detected at an estimated concentration of 41 ug/Kg in one of the three Area I soils. Di-n-burylphthalate was detected at concentrations of 22, 27, 26. and 30 ug/kg in one sample from area II, one sample from area three, and two of the three samples from area IV, respectively. Bis(2-Ethylhexyl)phthalate was detected at estimated concentrations of 27 and 62 ug/Kg in one sample from Area III and one sample 6 USEPA 6875 0 0 0 USFW 059: 'i := 0 0 0 1 7 6 from Area IV. Di-n-octylphthalate was detected at an estimated concentration of 180 ug/kg in one of the samples from area IV. In addition, numerous TICs including unknown alkane, cycloalkane, alkene, aldehyde, sterols, alcohols, PAH, acid, and other organic compounds were found in the surface soil samples. Results for the BNA analysis of soil samples is presented in Table 2 and in Appendix B. S.cdimcm Analysis of the sediment samples from the five site and two off-site stream locations produced only a few isolated detections of BNA compounds. Di-n-butylphthalate was detected in the Area IV sediment sample at an estimated concentration of 30 ug/Kg. Bis(2Ethylhexyl)phthalate was detected in the Area III sediment sample at an estimated concentration of 52 ug/Kg. No other standard list compounds were found in any of the Lee Creek, reference stream, or Dry Run sediment samples. Numerous TICs including unknown alkane, cycloalkane, alkene, aldehyde, sterols, alcohols, PAH, acid, and other organic compounds were found in the Dry Run, Lee Creek, and reference sediment samples. Results for the BNA analysis of sediment samples is presented in Table 3 and in Appendix B. 3.1.2 TAL Metals Surface Water Analysis of the surface water samples from Dry Run, the reference stream, and Lee Creek included both filtered and unfiltered samples for TAL metals analysis. Antimony, arsenic, beryllium, cadmium, cobalt, mercury, nickel, selenium, silver, thallium, and vanadium were not detected in any of the filtered or unfiltered water samples. Aluminum, barium, calcium, copper, iron, magnesium, manganese, potassium, sodium, and zinc were detected in the filtered water samples. Of the list of detected metals in the filtered samples, it appears that aluminum, copper, and zinc are found in higher concentrations in the Dry Run Creek drainage, including the reference stream, than in Lee Creek. Aluminum, barium, calcium, copper, iron, magnesium, manganese, potassium, sodium, and zinc were detected in the unfiltered samples. O f the list of detected metals in the unfiltered samples, concentrations of aluminum, iron, and zinc appear to be higher in Dry Run than those measured in Lee Creek. Detailed results of the TAL metals analysis in filtered and unfiltered water samples are presented in Table 4 and in Appendix B. W siL water Well water sampled from the well on the Tennant farm was analyzed as an unfiltered sample. Antimony, arsenic, beryllium, cadmium, chromium, cobalt, mercury, nickel, selenium, silver, thallium, and vanadium were not detected in the well sample. Concentrations measured for the remaining list of TAL metals are presented in Table 4 and in Appendix B. 7 USEPA 6876 0 0 0 ,, -J 000177 USFW 0594 Soil Three replicate surface soil samples from the four Dry Run meadow areas and the reference meadow area were analyzed for TAL metals. Antimony, cadmium mercury, selenium, silver, and thallium were not detected in any of the soil samples. One-way analysis of variance determined that soil manganese concentrations were significantly higher in Area II compared to the reference area, but the same as those noted in areas I, ID. and IV (p=0.094). Area II had the highest mean manganese concentration with mean concentrations from the other areas ranging from 680 to 1310 mg/kg. Further results of the TAL metals analysis of site and reference soil samples are presented in Table 5 and Appendix B. Sediment Seven sediment samples were submined for TAL metals analysis. Five of the samples were taken in the streambed of Dry Run, one was taken in Lee Creek, and one was taken in the reference stream. Antimony, cadmium, mercury, selenium, silver, and thallium were not detected in any of the sediment samples. In comparison to the levels measured in the Lee Creek sample, it appears the Dry Run Creek reach may be enriched in aluminum, arsenic, barium, calcium, chromium, cobalt, copper, iron, lead, magnesium, manganese, nickel, potassium, sodium, vanadium, and zinc. Based on the results of the aluminum, barium, chromium, cobalt, copper, iron, lead, manganese, nickel, vanadium, and zinc analysis, there also appears to be a general trend that metal concentrations" decrease with increasing distance from the landfill. Further results of the TAL metals analysis of site and reference sediment samples are presented in Table 6 and Appendix B. 3.1.3 Pesticides/PCBs Water No pesticides or PCBs were detected in the Dry Run samples, the Lee Creek sample, or in the reference stream sample (Table 7; Appendix B). Soil No pesticides or PCBs were detected in the Dry Run meadow samples, or in the reference meadow samples (Table 8; Appendix B). Sediment No pesticides or PCBs were detected in the Dry Run samples, the Lee Creek sample, or in the reference stream sample (Table 9, Appendix B). 3.1.4 VOCs Wati No volatile organic carbon compounds were detected in the Dry Run samples, the Lee Creek sample, or in the reference stream sample (Table 10; Appendix B). 8 USEPA 6877 0 9 9 .i ; v 000178 USFW 059 sail Trichlorofluoromethane was detected in every soil sample taken in the meadows adjacent to Dry Run at concentrations ranging from 0.9 to 3.6 ug/Kg. In addition, tetrachloroethene was detected in one replicate Area III soil sample at a concentration of 4.4 ug/Kg. No other volatile organic carbon compounds were detected in the Dry Run samples, the Lee Creek sample, or in the reference stream sample. Results of the VOC analysis of site and reference soil samples are presented in Table 11 and Appendix B. Sediment Acetone was detected in the Area IV sample at a concentration of 7.2 ug/Kg. Chloroform was detected at a concentration of 0.5 ug/kg in the Area III sample. No other volatile organic carbon compounds were detected in the Dry Run samples, the Lee Creek sample, or in the reference stream sample. Results of the VOC analysis of site and reference sediment samples are presented in Table 12 and Appendix B. 1.5 Total Fluoride Water / Well Water Fluoride was not detected in the Dry Run samples, the Lee Creek sample, the reference stream sample, or in the well sample taken on the Tennant farm (Table 13; Appendix B). Soil Ftuoride was detected in the Dry Run meadow and in the reference meadow samples. Soil fluoride concentrations ranged from a low of 180 mg/kg in Area IV to a high of 370 mg/kg in Area III. There appear to be no statistically significant differences in total soil fluoride concentration. Results of the soil fluoride analysis are presented in Table 14 and Appendix B. Sediment Fluoride was detected in the Dry Run creekbcd and in the reference creekbed samples, but not in Lee Creek. Fluoride concentrations ranged from a low of 290 mg/kg in the Area IV sampling area to a high of 450 mg/kg in the Upper Tributary A sampling area. Fluoride was not detected in Lee Creek. Overall, fluoride concentrations tend to decrease with increasing distance from the landfill. Sediments sampled in the Dry Run Creek reach appear to be enriched with fluoride, which is not found in Lee Creek. Results of the sediment fluoride analysis are presented in Table 15 and in Appendix B. 1.6 Oreanofluorides Sediment USEPA 6878 Because of methodology problems, specifically in obtaining appropriate standards, and the high volatility of some standards, only a limited suite of organofluoride compounds could be scanned for in the sediment samples. These compounds are presented in Table 16. Of the list that was analyzed for (Tetrafluoroethylene, hexafluoropropylene, 9 000- 000179 USFW 0596 chlorodifluoromethane, perfluorocyclobutane, l-Chloro-l, 1.2.2, tetrafluoroethane, 2Chloro-l,1.1.2,3.3.-hexafluoropropane, and Perfluoroisobutylene), none of die organofluoride compounds were detected in site sediments. Results of the organofluoride analysis is reported in Table 16 and in Appendix B. 3.1.7 Total Organic Carbon and Grain Size of Soil and Sediment Summaries of total organic carbon and grain size analysis are presented in Tables 17-20 and in Appendix B. TOC in the soil ranged from an average low of 5.655 in Area III soils to an average high of 9.2 in Area IV soils. Soil grain size determinations are summarized in Table 18. TOC in the sediment ranged from a low of 1.955 in Lee Creek to a high of 4.5 55 in Area IV. Sediment grain size determination is presented in Table 20. 3.1.8 Water Quality Parameters Water quality parameters including pH. conductivity, turbidity, dissolved oxygen, temperature, bromide, chloride, nitrate, phosphate, and sulfate was measured in the Dry Run Creek reach, the reference stream, and in Lee Creek. The most notable observations were that conductivity and sulfate concentration decreased with increasing distance from the landfill. Other parameters appeared to be in the expected range. Results of these measurements and analyses are presented in Tables 21 and 22. 3.1.9 Bovine Fecal Samples ; Six fecal samples were taken to determine if environmental contaminants were showing up in the digestive products of the affected cattle. Phenol was detected in all six fecal samples at concentrations ranging from 2.6 to 8.0 mg/kg. Additionally, 4-meihylphenol was detected in all six samples in concentrations ranging from 45 to 110 mg/kg. Benzoic acid was detected in two of the samples at a concentration of 30 mg/kg. Additional information is presented in Table 23 and in Appendix B. TAL Metals Aluminum, barium, calcium, copper, iron, lead, magnesium, manganese, potassium, sodium, vanadium, and zinc were detected in the fecal samples. Additional information presented in Table 24 and in Appendix B. Fluoride Fluoride was not detected in any of the fecal samples (Table 25; Appendix B). Biotic Sampling and Tissue Analysis 3.2.1 Benthic Macroinvenebrates 10 USEPA 6879 000180 O O O .llii USFW 0597 A tool of 27 taxonomic groups were collected from the 5 locations sampled (Table 26). Of ihese, there was 1 Oligochaete, 1 Mollusc, 1 Turbellarian. 3 Crustaceans, and 21 Insect taxa. Of the latter, the dominant group, in terms of taxonomic diversity, were the Coleopteria which were represented by 7 taxa. The Dipteria were represented by 4 taxa and the Ephemeropteria were represented by 3 taxa. The Plecopteria, Hemipteria Tricopteria, and Megaloptera were the least diverse groups and were represented by one or two taxa. The greatest taxonomic diversity was observed at the reference location, where 19 taxa were collected. Fewer taxa were observed at locations I through IV, and the lowest diversity was observed at location IV where 11 taxa were collected. The difference in taxonomic diversity between the reference location and locations I, II. HI, and IV was primarily due to the presence of a greater number of rare taxa at the former location. The number of individuals collected per replicate ranged from 203 at the reference location (replicate A) to 24 at location IV (replicate C). The observed density of individuals throughout the study area is primarily the result of the numerical abundance of only several taxa (Table 26). The numerically dominant taxa collected from the study area includes Leucrocuxa and Asellidae. When present, these taxa were typically the most numerically abundant organisms and were represented by 221 and 391 individuals, respectively. Other taxa. including Perlista, Chironomidae, Hyalella. the Turbellaria, and to a lesser extent, Lepiophtebia, Baeris, and Pseudolimnophilia, were present at most locations in consistently significant proportions. In general, most taxa collected were relatively rare and were represented by five or fewer individuals at most locations. For example, of the 129 total taxonomic observances, 54 were represented by one individual. 36 were represented by two to five individuals, 11 by six to 10 individuals, 15 by 11 to 20 individuals, and 13 by zreater than 21 individuals. Several taxa were collected from all locations sampled including Leucotricia, Perlesta, Chironomidae, and Asellidae (Table 26). Several taxa were not collected from all locations but were broadly represented throughout the drainage including Lepsophlebia, Agabus, Hyalella, and Turbellaria. Of the nine taxa observed at only one location, four, including Elmidae. Scinidae, Pseudolimnophila, and Stratiomvidae were collected only from the reference location. The most common distribution observed was one where a taxa was collected in relatively low numbers, and at few locations. For example. Elmidae, Hydropsyche, Limnophilidae, Nigronia, and Ceratoponidae were collected infrequently and in low numbers. Similarly, Lipogomphus, Dytiscidae, Curculidae, Elmidae, Scinidae. Histeridae, Pseudolimnophila, Stratiomyidae, and Physa were collected in low numbers at only one location. Five functional feeding groups were collected from the Dry Run drainage (Table 26). Resulting from the presence of Asellidae and Hyalella, omnivores were the dominant functional group at most locations. Although less dominant, collector-gatherers and scrapers were consistently collected from all locations in the study area and included the mayflies Leucrocuia and Leptophlebia. The dominant scraper was the mayfly Leucrocuta. Predators were dominant at locations II and III and were represented primarily by the stonefly Perlesta. Tne overall assessment of ecological condition first focuses on the evaluation of habitat quality, and secondly on the analysis of biological components in light of habitat. Habitat, 11 USEPA 6880 000- 0 000181 USFW 0598 as the principal determinant of biological potential, sets the stage for interpreting biosurvev data and can be used as a general predictor of biological condition. High quality habitat will support high quality biological communities and responses to minor alterations will be subtle and of little consequence. However, as a habitat declines in quality, discemable biological, impairment results. When habitat and biological data are systematically collected together, empirical relationships can be quantified and subsequently used for screening impact and discriminating water quality effects from habitat degradation. The watershed that drains the Dry Run study area has been modified as a result of past and present land use, particularly with respect to cattle grazing and other agricultural practices, as well as the siting of commercial and industrial facilities. The loss of riparian vegetation, through replacement by species resistant or adapted to grazing, or elimination by grazing has several consequences that should be considered when evaluating the distribution of benthic raacroinvertebrates and macroinvertebrates in the current study. The attainable biological potential of a stream or river is primarily determined by the quality of the habitat at a particular location. The Dry Run study area is situated in a rural area utilized primarily for grazing cattle and, although historic indications of grazing are evident, significant portions of the riparian area remain vegetated, and there are few areas with a completely open canopy and exposed soil. Portions of the Dry Run drainage, though somewhat degraded, support a surprisingly diverse and apparently robust aquatic community. The taxonomic diversity and numerical abundance of the macroinvertebrate was relatively high at the reference area. In contrast, the diversity and abundance at locations I. II. Ill, and IV was reduced substantially. Since habitat considerations at all locations in Dry Run are similar, the presence of contamination at the latter locations may be significant. 3.2.2 Mammal Four species of small mammals (meadow voles, short-tailed shrews, white-footed mice, and meadow jumping mice) were caught during the trapping effort. Whole bodies were submitted for lipid, TAL metal and total fluoride analysis. The trapping effort revealed at least one important field observation, which was there was extremely low trapping success in Area I, the area nearest the landfill outfall, as compared to the other areas. This is highly irregular given the similar habitats present site-wide, and may indicate an ecological threat. Field necropsies identified several significant problems with the small mammals collected in the meadow areas adjacent to Dry Run. Short-tail shrews sampled from all areas showed blackened and degenerating teeth. Shrews commonly have what is known as chestnut tipped teeth, where the extreme points of the dentitia are a light brown color. The black, mottled, and degenerating teeth observed in this study are not normally observed in shrews. One meadow vole sampled from the area was missing the left kidney, and another appeared to have and extra kidney or an extra lobe on the right kidney, independent of the adrenals. Sufficient numbers of meadow voles were caught from the Reference Area, Area II. Area III, and Area IV for statistical comparisons of tissue concentrations. Lipid concentration of meadow voles was significantly depressed in the Reference Area, Area II, and Area III, compared to that observed in Area IV (p<0.001). Barium concentration was significantly lower in the Reference Area, Area II, and Area III. compared to that observed in Area IV 12 USEPA 6881 i.T 000182 USFW 0599 ( p =0.067). Sufficient numbers of short-tail shrews were caught from the Reference Area and Area III for statistical comparisons. There were no differences in body concentration of lipid. TAL metals, and fluoride in shrews taken from these two areas. Results of the trapping success, TAL metals, and fluoride analysis are presented in Tables 27-29, and in Appendix B. Results are presented by species and trapping location. 3.2.3 Fish Four species of fish were collected from Dry Run in Areas II, III, and IV. No fish were observed in Area I or the Reference Area. Creek chubs and fantail darters were collected in Area IV. Creek chubs and river chubs were collected in area III. Creek chubs, river chubs, fantail darters, central stone rollers, and black-nose dace were collected in Area II. Fish sampled during the electrofishing effort in Dry Run were submitted for whole body lipid, TAL metal and fluoride analysis. A composite sample that was taken during a historical fish kill in Dry Run was also analyzed. Since creek chubs were the only species common to all three sampling locations, statistical analysis concentrated on differentiating between tissue concentrations in this species. Aluminum, arsenic, barium, beryllium, cobalt, iron, lead, manganese, thallium, and mvanadium were significantly higher in creek chubs from area II than those sample in Areas and IV. Likewise, concentrations of these metals in Area III creek chubs were higher than those in Area IV. Conversely, cadmium and silver concentrations showed the reverse trend, with tissue concentrations in Area IV significantly higher than those measure in Areas II or III. In spite of this result, it is clear that fish inhabiting upper reaches of Dry Run. nearer to the landfill are being dosed with a significantly higher amount of metals than those in the lower reaches. There were no difference in lipid or fluoride concentration. Results of the chemical analyses are presented in Tables 30-32 and in Appendix B. 3.2.4 Earthworm Tnree replicate earthworm samples were produced from each of the earthworm toxicity test soil samples. One-way analysis of variance was used to look for significant differences in tissue concentrations between earthworms exposed to each of the five soil treatments. Cadmium and thallium concentrations in the earthworm tissue was significantly higher in the Area I exposures than those in Areas II, III, IV, or the Reference soils. Cobalt levels were lower in worms taken from the reference area, but tended to increase with increasing distance from the landfill. Copper and nickel concentrations were higher in worms taken from the reference soils than those observed in worm taken from the other soil area exposures. Further results of the earthworm tissue analyses for lipid, TAL metals and fluoride concentration are presented in Tables 33-35 and in Appendix B. 3.2.5 Vegetation 13 000 - USEPA 6882 000183 USFW 06C fluoride, aluminum, calcium, magnesium, nickel, potassium, and sodium. Further, there were strong negative associations between the growth endpoint and chromium, copper, lead, and Tine concentrations, although the relationships were not significant at the 0.10 level. Further results of the amphipod toxicity test are presented in Table 40 and in Appendix C. SUMMARY OF PRELIMINARY ECOLOGICAL RISK ASSESSMENT SCREEN Sediment, soil and water concentrations were compared against Region III BTAG screening values (U.S. EPA 1995). Hazard quotients were generated by dividing the maximum site concentration measured in each matrix by the corresponding Region III benchmark values. All contaminants for which maiimum concentrations exceeded benchmarks for sediment, soil, and water in the initial screening-level risk assessment are listed in the following sections. Contaminants that failed the initial screening process will be further evaluated in a final risk assessment for the site. Sediment Table 1 lists maximum concentrations, screening criteria, and quality criteria factors for sediment contaminants. The maximum concentration recorded at the site exceeded the benchmark values for the following compounds: arsenic, chromium, copper, manganese, and nickel. Because of the lack of a screening benchmark, the following compounds are still considered as risk factors as well: fluoride, aluminum, barium, beryllium, cobalt, iron, and vanadium. W aur Table I lists marimum concentrations, benchmarks, and quality criteria factors for water contaminants. The maximum concentration recorded at the site exceeded the benchmark values for the following compounds: aluminum, copper, and iron. Because of the lack of a screening benchmark, fluoride is considered to be a potential risk factor. Soil Table 1 lists maximum concentrations, screening criteria, and quality criteria factors for soil contaminants. The maximum concentration recorded at the site exceeded the benchmark values for the following compounds: aluminum, beryllium, chromium, copper, iron, lead, manganese, vanadium, and zinc. Because of the lack of a screening benchmark, the following compounds are still considered as risk factors as well: fluoride and trichlorofluoromethane. DISCUSSION The data generated during the field effort suggests that there are potential problems associated with conditions at the Dry Run Creek site. Minimally, the results of the sediment and water toxicity tests suggest potential problems in the stream. Some metals appear in higher tissue concentrations in biota sampled nearest the outfall of the landfill, with those levels progressively dropping with increasing distance from the landfill area. Likewise, sediments sampled in Dry Run near the landfill outfall appear to have higher fluoride and metals concentrations than those sampled further downstream. A preliminary screen of potential risk factors suggests that other problems, specifically elevated levels of fluorides, organofluorides, and some metals, may be present as well. Data gathered during the field effort will be further analyzed through a base-line ecological risk assessment for the Dry Run Creek site USEPA 6884 15 000184 000- USFW 0602 Three replicate samples of meadow grass were taken in each of the five soil sampling locations. One way analysis of variance was used to look for differences in plant tissue concentrations across the five sampling areas. Barium concentration was significantly hieher in the Area I vegetation than in the Area III and IV vegetation, but similar to that observed in plants take in the Reference and in Area II. Manganese concentration was sienificantly higher in the Area I and reference vegetation than in the Area II. Ill and IV vegetation. Cobalt was significantly higher in Area IV vegetation than that taken in any of the other areas. There was no difference in the fluoride or lipid concentration. Results of the TAL metals and fluoride analysis are presented in Tables 36-38 and in Appendix B. 3.3 Histological assay of small mammal liver and kidney Histological analysis of liver and kidney sections of meadow voles, short-tail shrews, meadow jumping mice, and white-footed mice trapped in each of the five trapping areas concluded that there were no substantive changes in liver or kidney morphology. The absence of a kidney in one animal, and the presence of an extra lobe on the right kidney of another provide anecdotal evidence of an effect, however we are unable to ascertain the importance of these observations. Full summaries of the histopathological work are presented in Table 39 and in Appendix B. 3.4 Toxicity Testing Earthworm Based on the toxicity evaluation of soils, there is no evidence for growth or survival effects on earthworms tested in any of the soil samples collected at the Dry Run Creek site. Survival was 100% in all treatment replicates and growth ranged from 32.4 to 54.3%. Further results of the earthworm toxicity test are presented in Table 40 and in Appendix C. Fathead minnow Based on the toxicity evaluation of surface water samples to the fathead minnow, it appears that surface water taken in the Upper Tributary A location induced significant mortality. Survival in the Upper Tributary A sample was 58% while survival in all other samples, including the reference location, ranged from 87 to 100%. There appear to be no growth related effects water, on the minnows in any of the water samples taken from Dry Run Creek. Survival was negatively correlated potassium concentrations, however these correlations are not statistically significant at the 0.10 level. There was a significant positive correlation between fathead survival and iron concentrations in the filtered water samples. Further results of the fathead minnow toxicity test are presented in Table 40 and in Appendix C. Amphiood Based on the results of the 10 day solid phase whole sediment toxicity test with the amphipod, Hyalella azteca, growth of the organisms was inhibited in the sediment samples taken at the Upper Tributary A, and Area II locations. There were no effects observed on survival in any of the samples taken. The observed negative growth effect was significantly negatively correlated with 14 USEPA 6883 0 0 0 - 3 000185 USFW 0601 SECTION II ECOLOGICAL RISK ASSESSMENT 1.0 INTRODUCTION 1.1 Objective The objective of this risk assessment was to provide technical support to the U.S. Environmental Protection Agency Region III in conducting an evaluation of potential ecological threat due to existing contaminant levels in soil, sediment, and water at a working beef production farm located down gradient of a landfill. Soil, sediment, surface water, and biota samples were collected for contaminant analyses and soil, sediment, and surface water were collected for laboratory toxicity testing. The information gathered during this field effort was incorporated into in an ecological risk assessment for the Dry Run Creek site. 1.2 Site Background The site is a working beef production farm located in Washington, Wood County, WV. The owner of the farm has filed numerous complaints with the West Virginia Department of Natural Resources and the U. S. EPA alleging that contaminants are being discharged from an industrial landfill owned by the Dupont corporation, into Dry Run. Dry Run flows through the farmer's property and is a primary source of water for his cattle. The fanner maintains that numerous deaths, blindness, and other unusual illnesses observed in his herd are directly attributable to the contaminants that are discharged into Dry Run from the Dupont landfill. In addition to these abnormalities, numerous fish and wildlife kills have also been reported in Dry Run since the construction of the landfill. ; 2.0 PROBLEM FORMULATION This risk assessment was designed to evaluate the potential threats to ecological receptors from exposure to site contaminants. During the preliminary risk assessment, the problem formulation process included the identification of COCs and a comparison of the maximum concentration o f COCs with accepted benchmarks. This information was then used to identify complete exposure pathways of compounds exceeding benchmarks to ecological receptors and their appropriate measurement endpoints. The first step of the preliminary risk assessment process compared all chemicals analyzed in soil, sediment, and water during the field with established toxicological benchmarks. Benchmarks for sediment and soil were used to identify potential contaminants of concern for the protection of aquatic biota (U.S. EPA 1995, Long and Morgan 1990, Long et al. 1995, Persuad et al. 1992, U.S. EPA 1992, Suter and Mabrey 1994). Compounds exceeding benchmarks were retained for further evaluation using ingestion-based exposure models for higher vertebrates, and direct exposure assays for fish, benthic and terrestrial invertebrates. 2.1 Ecological Risk Assessment This ecological risk assessment was written to determine the risk associated with the exposure of biota to site-related contaminants. The following steps were completed for this preliminary risk assessment: (1) A literature search was conducted to locate life history information for selected indicator species, to determine ecotoxicological effects of site contaminants, and to locate bioconcentration factors for site contaminants. (2) An evaluation of ecological receptors was prepared. This consisted of the following: Exposure scenarios were determined based on site contaminant levels, the extent and magnitude of contamination, and the toxicological mechanisms of the 16 USEPA 6885 000 18 6 e00.;;;5 USFW 06( contaminants. Indicator species were selected based on species present and/or potentially present on site, the availability o f toxicity information from the literature, and the potential for exposure to site contaminants based on habitat use or behavior. Exposure pathway(s) were determined for each indicator species. Exposure and effect profiles were written for each indicator species and each site contaminant. A risk characterization was conducted which involved the calculation of hazard quotients (HQs) for each species for a range of exposure scenarios. Based on the results of this evaluation, the COCs identified in the initial screen were further evaluated through the use o f conservative risk models. Identification of the Contaminants o f Concern The contaminants o f potential concern were identified using the initial contaminant screen. The COCs for this site that were retained through the preliminary screen include metals, fluoride, and organofluoride compounds. ; Exposure Characterization The objective of the exposure assessment is to determine the pathways and media through which receptors may be exposed to site contaminants. Potential exposure pathways are dependant on habitats and receptors present on site, extent and magnitude o f contamination, and environmental fate and transport o f COCs. In this base-line ecological risk assessment, it will be concluded that "a potential risk" exists if the HQ calculated from the maximum site concentration and the No Observed Apparent Effect Level (NOAEL) equals or exceeds l. Exposure to COCs present in forage and prey species via ingestion could cause toxicity in higher trophic level organisms. In addition to exposure via consumption of contaminated forage, ecological receptors may also be exposed through ingestion of water and incidental ingestion of soil/sediment. The exposure of benthic invertebrates, terrestrial invertebrates, and fish was determined by examining results of the toxicity tests. Hazard CharacterizatioaToxicity Assessment To determine the effects of contaminants on biota, it is necessary to understand the mechanisms of toxicity of the chemicals and the systems that they affect. Knowledge o f the fate, effects, and mode of action of the COCs allows for the selection of appropriate assessment endpoints. Next is a review of the general toxicological information for the COCs identified in Section 2.4. 2.4.1 Fluoride Inorganic fluoride compounds are ubiquitous in nature. However, industrial processes such as manufacturing and mining have contributed to the environmental load of fluoride, primarily through atmospheric deposition.^Jn low doses, it is accepted that fluoride is 17 USEPA 6886 0 0 0 ..-2 G 000187 USFW 0604 protective of teeth in humans as well as other animals. Hcm-.ver in higher levels it is generally accepted that fluoride can be toxic to both plant and animal life. Dental and skeletal lesions. lameness, stiffness o f gait, appetite impairment, decreased weight gain, decreased milk production, posture abnormalities, tremors, stillbirths, overgrowth of hooves, severe diarrhea, and death have been associated with mammalian fluoride toxicity (Suttie, 1977; Shupe et al, 1992). In addition to the efFects known in mammals, birds are also susceptible to fluoride toxicity. Mortality, decreased growth rates, depressed appetite, and decreased eggshell quality have been reported as toxicological endpoints of fluoride exposure in birds (Fleming and Schuler 1988; Fleming et al. 1987; Guenter and Hahn 1986). 2.4.2 Organofluorides Organofluorides are used in a variety of industrial processes including the production of Teflon"*, propellants, and refrigerants. Available toxicological data generally concentrates on inhalation exposure and dermal absorption. Acute (10 day) exposure of rats to chlorodifluormethane produced decreased maternal and fetal weights, as well as an increased frequency of anopthalmia and subsequent blindness in newborn fetuses (IARC 1986). Hexafluoropropene exposure induced an increased incidence of hamster ovary cell aberrations and increased frequency grossly abnormal cells (HSDB 1997). 2.4.3 Aluminum Because of its strong reactivity, aluminum (Al) is not found as a free metal in nature. Aluminum has only one oxidation sate (+3), thus its behavior in the environment depends on its ordination chemistry and the surrounding conditions. In soils, a low pH generally results in an increase in aluminum mobility. In water, an equilibrium with a solid phase is established that controls the extent of aluminum dissolution (ATSDR 1990). Plants vary in their ability to remove aluminum from soils, although bioconcentration factors for plants are generally less than one. Biomagnification o f aluminum in terrestrial food chains does not appear to occur. There is no data on the biomagnincation of aluminum in aquatic food chains (ATSDR 1990). The nervous system may be a target area for aluminum. Aluminum accumulates in neurofibrillary tangles in humans with Alzheimer's disease. Aluminum may also interact with neuronal DNA to alter gene expression and protein formation. Mammalian studies do not indicate that aluminum affects reproduction although some developmental effects have been reported in mammals (ATSDR 1990). Aluminum is known to interfere with gill transport of oxygen and carbon dioxide in fish, and has also been identified in ionoregulatory disruption. 2.4.4 Arsenic Several review articles are available which discuss the toxic effects o f As (Eisler 1988a, Nriagu 1994). Arsenic tends to be widespread in the environment (Woolson 1975) and is constantly being oxidized, reduced, or mobilized (Eisler 1938a). Physical processes are important in determining As bioavailability in aquatic environments. For example, arsenates are readily adsorbed onto sediments with high organic matter, and arsenates are more strongly adsorbed onto sediments than other As forms. However, absorption depends on the As concentration, sediment characteristics, pH, and ionic concentration of other compounds USEPA 6887 O l*-J 000188 < i t USFW 060 (Eisler 1988a; U.2. U4. 19811. The U.S. EPA (1981) noted that arsenate (pentavalent) is the predominant As form in oxygeu_;= rnd that arsenite (trivalent) is the predominant As form in anaerobic conditions. Arsenic is not significantly concentrated in aquatic invertebrates; whole body concentration factors for invertebrates range from 3 to 17 for exposure to arsenic trioxide (trivalent) and from 0 to 7 for arsenic pentoxide (pentavalent). Arsenic may be bioconcentrated by organisms at the bottom o f the food chain; however, data do not indicate that significant biomagnification occurs (U.S. EPA 1985). 2.4.5 Beryllium The majority o f the beryllium (Be) in the environment is the result o f coal and oil combustion. Beryllium naturally enters waterways through the weathering o f rock and soil, and through deposition of atmospheric beryllium. Upon reaching water and soil, beryllium is most likely retained as an insoluble form that is generally immobile. However, beryllium chloride, fluoride, nitrate, phosphate, and sulfate (tetrahydrate) are all water-soluble forms (ATSDR 1993a). Due to its geochemical similarity to aluminum, beryllium may be expected to adsorb onto clay surfaces at low pHs, and it may remain precipitated as insoluble complexes at higher pHs. Therefore, beryllium is expected to have limited mobility in soil (ATSDR 1993a). ; Beryllium is not expected to bioconcentrate in aquatic animals and no evidence for significant biomagnificarion within food chains has been found. Beryllium is extremely toxic to warmwater fish in soft water. The degree of toxicity decreases with increasing hardness (ATSDR 1993a). Major exposure routes for aquatic ecological receptors include ingestion of contaminated soil and sediment. Although several studies point out the negative effects o f beryllium in mammalian systems, no studies that evaluated the relationship between sediment beryllium concentration and observed toxicity to benthic organisms could be found (ATSDR 1993a). 2.4.6 Chromium Chromium (Cr) can exist in oxidation states ranging from -2 to +6, but is most frequently converted to the relatively stable trivalent (+3) and hexavalent (--6) oxidation states (Eisler 1986a). In both freshwater and marine systems, hydrolysis and precipitation are the most important processes that determine the fate and effects o f Cr, whereas adsorption and bioaccumulation are relatively minor. Precipitated Cr*1 hydroxides remain in sediments under aerobic conditions. However, under anoxic and low pH conditions, Cr*1hydroxides may solubilize and remain as ionic Cr*' unless oxidized to Cr*4 through mixing and aeration (Eisler 1986a). In soils, the solubility and bioavailability o f Cr are governed by soil pH and organic complexing substances, although organic complexes play a more significant role (James and Bartlett 1983a; James and Bartlett 1983b). Th* trivalent state is the form usually found in biological materials. This form functions as an essential element in mammals by maintaining efficient glucose, lipid, and protein metabolism (Stevens et al. 1976). Chromium is beneficial but not essential to higher plants (Eisler 1986a). The biomagnification and toxicity of Cr*1 is low relative to Cr*4 because of 19 USEPA 6888 000189 /> -r. ' 1 O USFW 0606 its low membrane permeability and its noncorrosivity. However, a large degree of accumulation by aquatic and terrestrial plants and animals in the lower trophic levels has been documented (Eisler 1986a), although, the mechanism of accumulation remains largely unknown. Chromium is mutagenic, carcinogenic, and teratogenic, with Cr"4 exhibiting the greatest toxicity; relatively less is known about the toxicity of Cr*1 . At high concentrations" Cr*4 is associated with abnormal enzyme activity, altered blood chemistry, lowered resistance to pathogenic organisms, behavioral modifications, disrupted feeding, histopathology, osmoregulatory upset, alterations in population structure, and inhibition of photosynthesis. Rabbits fed dietary Cr accumulated hyaluronates, chondroitin sulfates, and neutral mucopolysaccharides in the soft tissues, causing pericapillary sclerosis (Kucher and Shabanov 1967). This accumulation blocked blood tissue barriers, which are permeable under normal conditions, preventing the normal transport o f metabolites. One manifestation of this condition was the inhibition of insulin production in the pancreatic islets due to damage to the beta-cells contained therein. Chromium also leads to nephron damage via swelling and loss of microvilli, the formation of intracellular vacuoles, mitochondrial swelling, and cytoplasmic liquefication and loss of cells lining the nephron surface (Evan and Dail 1974). The preliminary step in Cr-induced respiratory cancer is speculated to be the scarring of alveolar tissue, followed by the elicitation o f inflammatory reactions in lung tissue leading to bronchopneumonia, alveolar epithelial changes, atrophy, and benign tumor formation. Direct skin contact with highly corrosive chromic acid and its anhydride produces skin ulcers and necrosis by a mechanism independent o f any allergic response (Steven et al. 1976). 2.4.7 Copper Copper (Cu) does not appear to have mutagenic properties (IRIS 1990), but it is a teratogen (RTECS 1991) and a possible carcinogen (Venugopal and Luckey 1978). Copper is caustic, and acute toxicity is primarily related to this property (Hatch 1978). Copper is an essential element for animals and is a component of many metalloenzymes and respiratory pigments (Demayo et al. 1982). It is also essential to iron (Fe) utilization and functions in enzymes for energy production, connective tissue formation, and pigmentation (Venugopal and Luckey 1978). Excess Cu ingestion leads to accumulation in tissues, especially in the liver. High levels of Cu modify hepatic metabolism (Brooks 1988), which may lead to inability of the liver to store and excrete additional Cu. When liver concentration exceeds a certain level, the metal is released into the blood, causing hemolysis and jaundice. High Cu levels also inhibit essential metabolic enzymes (Demayo et al. 1982). Toxic symptoms appear when the liver accumulates 3 to 15 times the normal level of Cu (Demayo et al. 1982). Although the exact mechanism of toxicity is not known, the following mechanisms have been proposed: formation of stable inhibitory complexes with cytochrome P-450 (Wiebel et al. 1971); impairment of function ofNADPH-cytochrome c reductase and alteration of mixed function oxidations (Reiners etal. 1986); and inhibition of heme biosynthesis (Martell 1981). Intranuclear inclusions may act as a detoxifying mechanism where Cu is compiexed by 20 USEPA 6889 000190 0 0 0 - i '0 USFW 060-, protein ligands, protecting cytoplasmic organelles (Demayo et al. 1982). Ruminants are the most sensitive mammal species to Cu toxicosis. Young animals retain more dietary Cu than older animals and are more sensitive to Cu toxicity (Venugopal and Luckey 1978). 2.4.8 Iron Iron (Fe) is commonly detected in concentrations o f 5 percent or more in soil. It is used primarily in the production o f steel and other alloys as well as a major source o f hydrogen. Iron is a constituent of hemoglobin and is essential to plant and animal life as well as being an important component in cellular oxidative processes. The disposition of ingested iron is regulated by a complex mechanism to maintain homeostasis. Therefore, bioconcentration in biota is not expected to be a significant process for iron. Generally, about 2 to 15 percent o f ingested iron is absorbed from the gastrointestinal tract, and elimination is approximately 0.01 percent of the body burden per day. Adverse effects of iron toxicity may include renal failure and hepatic cirrhosis. The mechanism of toxicity begins with acute mucosal cell damage and absorption of ferrous ions directly into circulation, resulting in capillary endothelial cell damage to the liver (Shacklette and Boemgen 1984). 2.4.9 Lead Lead does not biomagnify to a great extent in food chains, although accumulation by plants and animals has been extensively documented (Wixson and Davis 1993, Eisler 1988b). Older organisms typically contain the highest tissue Pb concentrations, with the majority of the accumulation in the bony tissue of vertebrates (Eisler 1988b). Predicting the accumulation and toxicity of Pb is difficult since its effects are influenced to a very large degree, relative to other metals, by interactions among physical, chemical, and biological variables. In general, organolead compounds are more toxic than inorganic Pb compounds, and young, immature organisms are most susceptible to its effects (Eisler 1988b). In plants, Pb inhibits growth by reducing photosynthetic activity, mitosis, and water absorption. The mechanism by which photosynthetic activity is reduced is attributed to the blocking of sulfhydryl groups, inhibiting the conversion o f coproporphyrinogen to prcporphyrinosen (Holl and Hampp 1975). The toxic effects o f Pb on aquatic and terrestrial organisms are extremely varied and include mortality, reduced growth and reproductive output, blood chemistry alterations, lesions, and behavioral changes. However, many effects exhibit general trends in their toxic mechanism. Generally, Pb inhibits the formation of heme, adversely affects blood chemistry, and accumulates at hematopoietic organs (Eisler 1988b). At high concentrations near levels causing mortality, marked changes to the central nervous system occur prior to death (Eisler 1988b). Plants can uptake Pb through surface deposition in rain, dust, and soil, or by uptake through the roots. The ability of a plant to uptake Pb from soils is inversely related to soil pH and organic matter content. Lead can inhibit photosynthesis, plant growth, water absorption. 2.4.10 Manganese USEPA 6890 000191 USFW 06( soluble pentavalent form. The mobility o f vanadium in soil is affected by pH, redox potential, and the presence o f particulates. Relative to other minerals, vanadium is mobile in neutral or alkaline soils and its mobility decreases in acidic soils (ATSDR 1991; Van Zinderen Bakker and Jaworski 1980). In the terrestrial systems, bioconcentration is more common in lower plant species. In addition, vanadium concentrations in plants are dependent on the amount of water-soluble vanadium, pH, and growing conditions. Vanadium appears to be present in all terrestrial mammals but the concentrations are usually below the detection limits. The highest concentration o f vanadium is usually found in the liver and skeletal tissues (ATSDR 1991). Vanadium is very poorly absorbed into the gastrointestinal tract and the toxic mechanism of vanadium on the respiratory system is similar to other metals (Castronova et al. 1984). Vanadium damages the alveolar macrophages by decreasing the macrophage membrane integrity. Damaged macrophages inhibit the ability o f the respiratory system to clear itself o f other particles. In vitro experiments indicate that the mechanism o f toxicity of vandium is by inhibiting sodium-potassium ATPase activity, which inhibits the sodium-potassium pump. This pump is necessary for the transport o f material across cell membranes (Nechay and Saunders 1978). 2.4.13 Zinc Zinc (Zn) is essential for normal growth and reproduction in plants and animals and is regulated by metallothioneins. Metallothioneins act as temporary Zn storage sites and aid in reducing the toxicity ofZn to both vertebrates and invertebrates (Olsson et al. 1989). Zinc is not known to bioaccumulate in food chains, because it is regulated by the body and excess Zn is eliminated. Zinc has its primary metabolic effect on Zn-dependant enzymes that regulate the biosynthesis and catabolic rate of ribonucleic (RNA) acid and deoxyribonucleic acid (DNA). High levels o f Zn induce Cu deficiency and interfere with metabolism o f calcium (Ca) and Fe (Goyer 1986). The pancreas and bone seem to be the primary targets of Zn toxicity in birds and mammals. Pancreatic effects include cytoplasmic vacuolation, cellular atrophy, and cell death (Lu and Combs 1988, Kazacos and Van Vleet 1989). Zinc preferentially accumulates in bone, and induces osteomalacia (a softening o f bone caused by a deficiency o f Ca, phosphorus and other minerals) (Kaji et al. 1988). Gill epithelium is the primary target site in fish. Zinc toxicosis results in destruction of gill epithelium and tissue hypoxia (Spear 1981). Selection of Assessment Endpoints The information gathered during a site reconnaissance and during the field work, and subsequent discussions with the U.S. EPA on-scene coordinator and the Region III Biological Technical Assistance Group, allowed for the selection of assessment endpoints that corresponded to the habitat types present at the Dry Run Creek site. The site is composed o f a variety of habitats including forested and old-field uplands, grassy meadows, the creek, and associated riparian areas. A variety of birds, mammals, and fish may use the site for feeding and nesting. Likewise, terrestrial and benthic invertebrates are key elements in the functions of these systems. Therefore, the assessment endpoints focused toward these faunal groups. Viability of terrestrial, avian, and aquatic populations and organism survivability were selected as assessment endpoints for this risk assessment. Listed next are 0 0 0 -':,? 2 000192 USEPA 6892 USFW 0610 Manganese (Mn) does not occur as a free metal in the environment but is a component of numerous minerals. Elemental manganese and inorganic manganese compounds have negligible vapor pressures, but may exist in air as suspended particulate matter derived from industrial emissions or the erosion of soil. Removal from the atmosphere is mostly through aravitational settling. The transport and partitioning o f manganese in water is controlled by the solubility of the specific chemical form present. The metal may exist in water in any of four oxidation states (2+, 3+, 4+, or 7+). Divalent manganese (Mn-*-2) predominates in most waters (pH 4 to 7), but may become oxidized at a pH greater than 8 or 9. Manganese is often transported in moving water as suspended sediments. The tendency o f soluble manganese compounds to adsorb to soils and sediments depends mainly on the cation exchange capacity and the organic composition o f the soil. Manganese in water may be significantly bioconcentrated at lower trophic levels. However, biomagnification in the food chain may not be significant (ATSDR 1990). The amount of manganese absorbed across the gastrointestinal tract is variable. There does not appear to be a marked difference between manganese ingested in food or in water. One of the key determinants o f absorption appears to be dietary iron intake, with low iron levels leading to increased manganese absorption. This is probably because both iron and manganese are absorbed by the same transport system in the gut (ATSDR 1990). 2.4.11 Nickel Pure nickel (Ni) is a hard, white metal that is usually used in the formation of alloys (such as stainless steel) and Ni combined with other elements is found in all soils. Nickel is the 24'*' most abundant element and is found in the environment as oxides or sulfides. Nickel may be released into the environment through mining, oil-burning power plants, coal-burning power plants, and incinerators. Nickel will attach to soil or sediment panicles, especially those containing Fe or manganese (Mn). Under acidic conditions, Ni may become more mobile and seep into the groundwater. The typical Ni concentration reported in soils is from 4 - 8 0 milligrams per kilograms (mg/kg). The speciation and physicochemical state of Ni is important in considering its behavior in the environment and its availability to biota. The most probable exposure routes of Ni are through dermal contact, inhalation of dust, and ingestion o f Ni-contaminated soil. The respiratory system is the primary target of Ni exposure following inhalation. Manifestations such as inflammation of the lungs, fibrosis, macrophage hyperplasia, and increased lung weight have been noted in animals exposed to Ni. Animals exposed to Ni through oral exposure were noted to have lethargy, ataxia, irregular breathing, salivation, and squinting (ATSDR 1996). 2.4.12 Vanadium Elemental vanadium does not occur naturally but it can exist in 50 different ores and fossil fuels. Other anthropogenic sources include acid-mine leachate, sewage sludge, and fertilizers. The principal use of vanadium is as an alloy constituent, especially in steel. The addition of vanadium to steel removes oxygen and nitrogen, which improves the strength. The average concentration of vanadium in the earths crust is 150 mg/kg and in the U.S. soils are 200 mg/kg (Byemum et al. 1974). The release of vanadium to water and soil occurs as a result of the weathering of rocks and from soil erosion. This process usually converts the less-soluble trivalent form to the more- 22 USEPA 6891 /i 000- 000193 USFW 060! on the site? Are levels of site contaminants sufficient to cause negative impacts on growth, survival, and reproductive success of carnivorous birds due to the ingestion of contaminated forage, soil, and water on the site? Are levels of site contaminants sufficient to cause negative impacts on growth, survival, and reproductive success of carnivorous mammals due to the ingestion o f contaminated forage, soil, and water on the site? Are levels o f site contaminants sufficient to cause negative impacts on growth, survival, and reproductive success of piscivorous mammals due to the ingestion o f contaminated forage, soil, and water on the site? Are levels of. site contaminants sufficient to cause negative impacts on growth, survival, and reproductive success of omnivorous mammals due to the ingestion o f contaminated forage, soil, and water on the site? Are levels of site contaminants sufficient to cause negative impacts on growth, survival, and reproductive success of insectivorous mammals due to the ingestion of contaminated forage, soil, and water on the site? Are levels of site contaminants sufficient to cause negative impacts on growth, survival, and reproductive success o f herbivorous mammals due to the ingestion o f contaminated forage, soil, and water on the site? Conceptual Model The conceptual model relies on contaminant and habitat characteristics to identify critical exposure pathways to the selected measurement endpoints. At the Dry Run Creek site, contaminants in the soil may come in contact with subsurface (earthworms) and above-ground terrestrial receptors (small mammals) inhabiting the wooded, wetland, and open field areas of the site. Subsurface terrestrial receptors in these areas may be exposed to site contaminants through direct contact with the soil, and in some cases, the intentional ingestion of soil. Above-ground terrestrial receptors may be exposed to contaminants through direct contact with the soil, the ingestion of subsurface terrestrial organisms, the ingestion of other above-ground terrestrial receptors, the incidental ingestion of soil adhered to food items, and the intentional ingestion of surface water from any o f the on-site surface drainages. The wooded areas, riparian area, and meadow areas provide distinct habitat types that may support a variety o f terrestrial and avian receptors. For example, a small omnivorous mammal may occupy one or all the habitat types, whereas, an individual carnivorous mammal may regularly traverse all three habitats daily in search of food items. Avian piscivores and carnivores may be exposed to site contaminants in much the same way as an above-ground terrestrial receptor. The consumption of contaminated prey, the incidental ingestion of soil/sediment, and the consumption of surface water may transfer contaminants to these receptors. The conceptual model relies on contaminant and habitat characteristics to identify critical exposure pathways to the selected measurement endpoints. The preliminary risk screen identified metals, fluoride, and trichlorofiuoromethane as the primary contaminants exceeding benchmarks in site sediment, soil, and water. Benthic macroinvertebrates, fish, and terrestrial invertebrates may be 25 USEPA 6894 0 0 0 34 000194 USFW 0612 the specific assessment endpoints selected for this ecological risk assessment. Ten assessment endpoints were chosen to evaluate the risk of contaminants at the Dry Run Creek site: 1) protection o f benthic invertebrate community structure and function 2) protection of soil invertebrate community structure and function 3) protection of fish communities to ensure that direct exposure to contaminants does not have a potential negative impact on growth, survival, or reproductive success. 4) protection of worm-eating birds to ensure that ingestion of contaminants in forage does not have a potential negative impact on growth, survival, or reproductive success. 5) protection of carnivorous birds to ensure that ingestion of contaminants in forage does not have a potential negarive impact on growth, survival, or reproductive success. 6) protecrion of carnivorous mammals to ensure that ingestion o f contaminants in forage does not have a potential negative impact on growth, survival, or reproductive success. 7) protection of piscivorous mammals to ensure that ingestion of contaminants in forage does not have a potential negative impact on growth, survival, and reproductive success. 8) protection of omnivorous mammals to ensure that ingestion of contaminants in forage does not have a potential negative impact on growth, survival, and reproductive success. 9) protection of insectivorous mammals to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success. 10) protection of herbivorous mammals to ensure that ingestion of contaminants in forage does not have a negative impact on growth, survival, and reproductive success. Production of Testable Hypotheses The testable hypotheses are specific risk questions that are based on the assessment endpoints. Based on the mechanism of contaminant toxicity, the number of exposure pathways that may exist for an assessment endpoint, or other factors, there may be more than one question for each assessment endpoint. Are levels of site contaminants sufficient to have negative effects on benthic invertebrate community structure and function? Are levels o f site contaminants sufficient to have negative effects on soil invertebrate community structure and function? Are levels of site contaminants sufficient to cause direct toxicity to fish growth, survival, and reproductive success? Are levels of site contaminants sufficient to cause negative impacts on growth, survival, or reproductive success of worm-eating birds due to the ingestion of contaminated forage, soil, and water 24 USEPA 6893 000 <-3 <7J 0 0 0 1 9 5 USFW 061 a) Ingestion of vegetation b) Incidental ingestion of sediment c) Incidental ingestion of water 8 Selection of Measurement Endpoints Measurement endpoints are measurable ecological characteristics that are related to the valued characteristics selected as assessment endpoints. Measurement endpoints should be linked to the assessment endpoints by the mechanism of toxicity and the route o f exposure. Measurement endpoints are used to derive a quantitative estimate of potential effects, and form a basis for extrapolation to the assessment endpoints. Measurement endpoints were selected on the basis o f potential presence o f receptors on site, and the potential for exposure to contaminants o f concern. The availability o f the appropriate toxicity information on which risk calculations could be based was also an important consideration. Endpoints selected were determined to be representative o f exposure pathways and assessment endpoints identified for the site. Next is a list of specific measurement endpoints that correspond to the assessment endpoints identified in Section 2.5. M easurem ent endpoints for assessment endpoint: protection of benthic invertebrate communities structure and function To evaluate the structure and function of the benthic community, benthic macroinvertebrates were collected from five locations in Dry Run. Existing community structure was evaluated at each of the five locations by determining taxonomic diversity and through an evaluation of functional feeding groups. Sediment was also collected in each o f the five areas for toxicity testing using the amphipod, Hyallela azteca. Tne endpoints o f these tests were survival and growth. Collocated sediment samples were also collected and analyzed for target analyte list (TAL) metals. BNA's, Pest/PCBs, VOCs, fluoride, grain size, and total organic carbon (TOC). The chemistry results were then correlated with observed adverse biotic responses in the toxicity tests in order to determine risk potential. M easurem ent endpoints for assessment endpoint: protection of soil invertebrate community structure and function To evaluate the structure and function of the benthic community, soil was collected from each o f the meadow locations and tested using the earthworm, Eiseniafoetida in toxicity tests. The endpoints of these tests were survival and growth. Collocated soil samples were also collected and analyzed for target analyte list (TAL) metals, BNA's, Pest/PCBs, VOCs, fluoride, grain size, and total organic carbon (TOC). M easurem ent endpoints for assessment endpoint: protection of fish communities to ensure that direct exposure to contaminants does not have a negative impact on growth, survival, and reproductive success. 27 USEPA 6896 0G0A3S 000196 I |Qcr\A/ n e -i a exposed to contaminated sediment, water, or soil through direct toxicity. For the purposes of this risk assessment, the concentration of the contaminants of concern found in the sediment, water, or soil were correlated with toxicity levels identified in the corresponding toxicity tests to determine if benthic invertebrates fish, or terrestrial invertebrates may be at risk. Terrestrial receptors may be exposed to contaminants by feeding on organisms which have accumulated COCs in their tissues. Higher trophic level receptors may also be exposed to contaminants from food ingestion and via incidental ingestion o f soil/sediment and water. The pathway to the reference area meadow is unknown, however the pathway to the reference area stream may involve groundwater transport. The following pathways were evaluated in this risk assessment: I. Benthic invertebrates a) Direct exposure to sediment II. Soil invertebrates a) Direct exposure to soil III. Fish a) Direct exposure to water IV. Worm'eating bird a) Ingestion of earthworms b) Incidental ingestion o f soil 0 Incidental ingestion of water V. Carnivorous bird a) Ingestion of small mammals b) Incidental ingestion o f soil c) Incidental ingestion o f water VI. Carnivorous mammal a) Ingestion of small mammals b) Incidental ingestion o f soil c) Incidental ingestion o f water VII. Piscivorous mammal a) Ingestion of forage fish b) Incidental ingestion o f sediment c) Incidental ingestion o f water VIII. Omnivorous mammal a) Ingestion of forage fish b) Incidental ingestion o f sediment c) Incidental ingestion o f water IX. Insectivorous mammal a) Ingestion of earthworms b) Incidental ingestion o f sediment c) Incidental ingestion o f water X. Herbivorous mammal 26 0 r, n t USEPA 6895 000197 USFW 0613  Protection of omnivorous mammals to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success. Food chain accumulation studies were selected to evaluate risk to mammalian species which utilize the site and adjacent areas. The selected measurement endpoint receptor species is the raccoon. Procyan lotor, as a model for omnivorous mammalian species. Appropriate forage species (fish) were identified for the above receptors and the dietary exposure of receptors to contaminants was quantified. M easurement endpoints for assessment endpoint: Protection o f insectivorous mammals to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success Food chain accumulation studies were selected to evaluate risk to mammalian species which utilize the site and adjacent areas. The selected measurement endpoint receptor species is the short-tail shrew, Blarina brevicauda, as a model for insectivorous mammalian species. Appropriate forage species (earthworms) were identified for the above receptors and the dietary exposure o f receptors to contaminants was quantified. Measurement endpoints for assessment endpoint: Protection o f herbivorous mammals to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success Food chain accumulation studies were selected to evaluate risk to mammalian species which utilize the site and adjacent areas. The selected measurement endpoint receptor species is the meadow vole, Microtuspenrtsylvanicus, as a model for herbivorous mammalian species. Appropriate forase species (vegetation) was identified for the above receptors and the dietary exposure of receptors to contaminants was quantified. Life History/Exposure Profile Information Receptor species were selected from several trophic levels. Organisms which were likely to be exposed to contaminants because o f specific behaviors, patterns o f habitat use. or feeding habits were selected for evaluation in this risk assessment. The availability of appropriate toxicity information on which risk calculations could be based was also an important consideration. The terrestrial invertebrate receptor selected for this assessment is the earthworm. The terrestrial vertebrate receptor species selected for this risk assessment are: meadow vole, short-tail shrew, raccoon, mink, and red fox. The avian receptor species selected for this risk assessment are: American robin and red-tailed hawk. The aquatic vertebrate receptor species for this risk assessment is the fathead minnow. The aquatic invertebrate receptor is H. azteca. 2.9.1 The amphipod (Hyallela azteca) as Representative o f Benthic Invertebrates Justification Hyallela azteca was selected as representative o f benthic invertebrates due to their direct contact with sediment for a significant portion o f their life cycle, ubiquitous distribution in aquatic systems, importance as a food item for aquatic-invertebrate consumers, and ease of 29 USEPA 6898 000 '^000198 Fathead minnow, Pimephales promelas. toxicity tests were used to determine the toxicity of the water in Dry Run. The endpoints of these tests were survival and growth. Collocated water samples were also collected and analyzed for target analyte list (TAL) metals, BNA's, Pest/PCBs, VOCs, fluoride, grain size, and total organic carbon (TOC). The chemistry results were then correlated with observed advene biotic responses in the toxicity tests in order to determine risk potential. M easurement endpoints for assessment endpoint: Protection of worm*eating birds to ensure that ingestion of contaminants in forage does not have a negative impact on growth, survival, and reproductive success. Food chain accumulation studies were selected to evaluate risk to avian species which utilize the site as a feeding area. The selected measurement endpoint receptor species is the American robin, Turdus migratorius. Appropriate forage species (earthworms) were identified for the above receptor, and the dietary exposure o f receptors to contaminants was quantified and compared to existing toxicity data for these, or other closely related species. M easurement endpoints for assessment endpoint: Protection o f carnivorous birds to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success. Food chain accumulation studies were selected to evaluate risk to avian species which utilize the site as a feeding area. The selected measurement endpoint receptor species is the red-tailed hawk, Buteo jamaciensis. Appropriate forage species (small mammals) were identified for the above receptor, and the dietary exposure of receptors to contaminants was quantified and compared to existing toxicity data for these, or other closely related species. M easurement endpoints for assessment endpoint: Protection o f carnivorous mammals to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success. Food chain accumulation studies were selected to evaluate risk to mammalian species which utilize the site and adjacent areas. The selected measurement endpoint receptor species is the red fox, Vulpes vulpes. Appropriate forage species (small mammals) were identified for the above receptors and the dietary exposure of receptors to contaminants was quantified. Measurement endpoints for assessment endpoint: Protection of piscivorous mammals to ensure that ingestion o f contaminants in forage does not have a negative impact on growth, survival, and reproductive success. Food chain accumulation studies were selected to evaluate risk to mammalian species which utilize the site and adjacent areas. The selected measurement endpoint receptor species are the mink. Mustela vison. Appropriate forage species (fish) were identified for the above receptors and the dietary exposure of receptors to contaminants was quantified. M easurement endpoints for assessment endpoint: USEPA 6897 000199 0 0 0 --..: V USFW 061 earthworms were observed in both the wooded and open field areas o f the Dry Run Creek site. Lift Histcn Earthworms feed on dead and decaying plant and animal remains and on free-living soil microflora and microfauna. Their primary source o f food is dead plant material, especially plant liner. Next to food, their most important requirement is adequate moisture. Water conservation mechanisms are poorly developed; respiration depends on diffusion of gases through the body wall which must be kept moist. Earthworms are generally absent or rare in soiis with very coarse texture, in soils with high clay content in regions o f high rainfall, and in soils with a pH o f less than 4 (Lee 1985). Earthworms are hermaphroditic and most species reproduce by cross-fertilization, although many species can also produce cocoons parthenogenetically. Sexual reproduction cannot occur without a ditellum, ovaries, oviducts, and possibly the ovisacs, but male organs are not essential. The population of an earthworm species at any one time consists of young immature, well-grown immature (adolescent), mature, and senescent individuals (Edwards and Lofty 1977). Earthworms have several ways o f surviving adverse environmental conditions such as soil desiccation and ambient cold and h eat In terms o f population survival, the cocoons can resist desiccation and temperature extremes much more easily than mature individuals. Worms may also migrate to deeper soil or undergo states o f inactivity until environmental conditions become favorable once again (Edwards and Lofty 1977). Some species of worms grow throughout their lives by continually adding segments proliferated from a growing zone located just in front o f the anus. Other species, such as . foetida, possess the adult number o f segments upon hatching and increase in size without increasing the number o f segments. The life span o f Eisenta foetida was reported to be approximately 4.5 years under laboratory conditions (Edwards and Lofty 1977). Exposure Profile Direct contact with contaminated soil is the primary route o f exposure for earthworms in this risk assessment. Survival and growth endpoints following exposure to site soils will be used to evaluate risk to these organisms. Tissue residue analysis will also be conducted on the worms to determine exposure to higher trophic level organisms. .9.3 Fathead Minnow (Pimephales promelas) as Representative o f Fish Community Justification The fathead minnow was selected as representative o f omnivorous fish due to its dietary composition, direct contact with water throughout the life cycle, ubiquitous distribution in aquatic systems, importance as a food item for fish-eating consumers, and ease o f use in laboratory toxicity evaluations. Lift Histotv USEPA 6900 31 0 0 0 - 0 0 00200 I to m * / use in laboratory toxicity evaluations. These species are also likely to occur in the surface sediment at the Dry Run Creek site. 1 ife History (Hvalleta azteca) The amphipod, Hyallela azteca, is commonly found in freshwater lakes, streams, ponds, and rivers throughout North and South America. In preferred habitats, they are known to reach densities in excess of 10,000 per square meter. They may also be found in sloughs, marshes, and ditches, but generally in lower numbers (U.S. EPA 1994). Hyallela azteca are epibenthic detritivores that feed on coarse paniculate organic material. They typically burrow into surface sediment, and avoid bright light. Because of their feeding and behavioral characteristics, they are ideal test organisms for toxicological evaluation of freshwater sediments. Avoidance of light by movement into the sediment keeps these organisms almost constantly in contact with sediment contaminants (U.S. EPA 1994). Reproduction in this crustacean is sexual. Males are larger than females and have larger front snathopods that are presumably used for holding the female during amplexus and copulation. During amplexus, the male and female feed together for a period of up to one week. The pair separates temporarily while the female goes through a molting period. Immediately after the molt, the two rejoin and copulation begins. During copulation, the male releases sperm near the female's marsupium. The female sweeps the sperm into her marsupium. and simultaneously releases eggs from her oviducts, into the marsupium, where fertilization take:s place. The "average brood size for female Hyallela azteca is 18 eggs per brood, but this number can vary with environmental conditions and physiological stress (U.S. EPA 1994). Developing embryos and hatched young are kept inside the female's marsupium until she undergoes a second molt. At that time, the juvenile Hyallela azteca are released into the surrounding environment. Under favorable conditions, each female produces approximately one brood during every ten day time period (U.S. EPA 1994). Hyallela azteca have a minimum of 9 instars, with 5 to 8 pre-reproductive stages. Tne first five stages are juvenile stages; instars 6 and 7 form the adolescent stages; and stages 8 and higher are considered adult (fully reproductive) stages (U.S. EPA 1994). Exposure Profile for Hvallela azteca Since direct contact with contaminated sediment in the toxicity evaluation is the primary route of exposure for Hyallela azteca in this risk assessment, the results of the test will be used to indicate exposure. .9.2 Earthworm (Eisenia foetida) as Representative o f Terrestrial Invertebrates Justification Earthworms were selected as representative of terrestrial invertebrates due to their feeding habits, ubiquitous distribution throughout many habitats and soil conditions, and importance in providing a food base for many small- to medium-sized predators. A diet of detritus, microflora, and microfauna, combined with direct contact with the surrounding soil, presents a potential link between soil contaminants and soil-invertebrate consumers. In addition. 30 USEPA 6899 000201 000 l !RFW 06` Breeding territories are established by male robins. Most foraging occurs close to these territories during the breeding season; however, if densities of robins are high in a given area or if food resources are limited, adult robins will leave to temporarily forage elsewhere. Outside of the breeding period, robins typically return to the same foraging sites and roost within 1 to 3 kilometers (km) of these areas (U.S. EPA 1993). Exposure Profile Adult American robins are reported to weigh from 77.3 to 133.8 g (U.S. EPA 1993). Territory sizes vary from 0.3 to 1 acre, with foraging home ranges reported up to 2 acres (U.S. EPA 1993). The lowest reported body weight (77.3 g) and the smallest reported home range of (0.3 acres) were assumed for this risk assessment. A food ingestion rate of 0.89 to 1.52 g/g BW/day and a water ingestion rate of 0.14 g/g BW/day are reported for this species (U.S. EPA 1993). Assuming a 77.3 g body weight, an American robin can be expected to consume 117.5 g/day of food and 10.8 g/day of water. Tne diet of the American robin consists of seasonally variable proportions of invertebrates (e.g., earthworms, snails, beetles, caterpillars, spiders) and fruit (e.g., dogwood, cherry, sumac, hackberries, raspberries) (U.S. EPA 1993, Ehrlich et al. 1988). During spring, summer, and fall, the dietary composition is reported to change from 93 percent invertebrates and 7 percent fruit in the spring (nesting season) to 92 percent fruit and 8 percent invertebrates in fall (migratory season). The summer dietary proportion is reported as 68 percent fruit and 32 percent invertebrates (U.S. EPA 1993). For the purposes of this risk assessment, a diet of 100% earthworms will be assumed. An incidental soil ingestion rate for the American robin could not be found in the literature. However, a soil ingestion rate of 10.4 percent of the diet reported for the American woodcock will be used as a substitute ingestion rate for the American robin (Beyer et al. 1994). Assuming a food ingestion rate of 117.5 g/day, the soil ingestion rate for the American robin is 12.2 g/day. .9.5 Red-tailed Hawk (Buieo jamaciensis) as Representative of Carnivorous Birds. Justification. The red-tailed hawk was selected as representative of a carnivorous bird due to its dietary composition, relative abundant distribution, and likelihood of occurrence at the Dry Run Creek site. Its diet allows for the evaluation of contamination in site soils. In addition, the concentration of contaminants found in small mammal tissue will also provide an accurate dose to the red-tailed hawk which allows for the evaluation of contaminants in the food source. Life History Red-tailed hawks are the most common and widespread American Buteo (Bull and Farrand 1977). Their habitat is highly variable, but they are commonly found in wooded areas near open land. They also inhabit plains, prairie groves, and deserts in the western United States (N'GS 1987). This species is absent, however, from tundra, and rare in extensive unbroken forest. An opportunistic feeder, the red-tailed hawk hunts from a perch or.on the wing for 33 USEPA 6902 009 000202 I IQ C\A/ The fathead minnow, P. promelas, is widely distributed in North America and is found in a variety of habitats such as small streams, ponds, and small lakes. It is uncommon or absent in streams of moderate and high gradients. It is tolerant of high temperature' high turbidity, and low oxygen concenmations (U.S. EPA 1985). The fathead minnow is primarily omnivorous. Young typically feed on detritus, algae, and zooplankton. Adults feed on aquatic insects, worms, small crustaceans, and other animals. This species is considered an important food source for other fish and birds (U.S. EPA 1985). Adult fathead minnows spawn in the spring and continue to spawn throughout most of the summer. The minimum spawning temperature appears to be approximately I6aC. The ovaries of the females contain eggs in all stages of development, and they spawn repeatedly as the eggs mature. The average number of eggs per spawn per female is 100 to 150. Larger females may lay 400 to 500 eggs per spawn. Hatching times depend on temperature and average about six days. In warm water with an ample food supply, spawning may occur as early as the first year. In cooler water with a moderate food supply, spawning usually occurs during the second year. Survival to the third year is relatively uncommon (U.S. EPA 1985). Exposure Profile Since direct contact with contaminated water in the toxicity evaluation is the primary route of exposure for fathead minnows in this risk assessment, the results of the test will be used to indicate exposure. 9.4 American Robin (Turdus migratorius) as Representative of Worm-eating Birds Justification The American robin was selected as representative of omnivorous and carnivorous birds because of its ubiquitous distribution and dietary composition. The preference for soil invertebrates in its omnivorous diet allows this species to be used as both an omnivorous and carnivorous receptor in this risk assessment. This species is also likely to occur at the Dry Run Creek site. Life History The American robin (Turdus migratorius) occurs throughout most of the continental U.S. and Canada, wintering in the southern half of the U.S., Mexico, and Central America. Given the increase in open habitat and lawns, the robin's breeding range has expanded in the recent times. Habitat requirements for breeding robins include access to fresh water, protected nesting sites, and productive foraging areas. These requirements are commonly met in moist forests, swamps, open woodlands, and other open areas. Non-breeding robins occupy similar habitats although proximity to fruit bearing trees is of more importance. The primary foraging technique for robins is to hop along the ground in search of grounddwelling invertebrates, although they commonly search for insects and fruit in tree branches as well. The robin's diet during the breeding season consists mainly of invertebrates and some fruit, but fruit is the primary food consumed outside of the breeding season. As robins exhibit a low digestive efficiency for fruit, they often consume more than their own body weight in fruit to meet their metabolic needs (U.S. EPA 1993). 32 USEPA 6901 000203 USFW 0619 dose to the red fox which allows for the evaluation of contaminants in the food source. Life History Red fox inhabit open meadows, ditch banks, field and wood edges, fencerows, stream and lake borders, and farmlands (Hoffineister 1989; Jones and Bimey 1988; Merritt 1987). With the exception of the breeding season, red fox have no permanent home but sleep on the ground (Schwartz and Schwartz 1981). A den, usually modified from an existing woodchuck or fox den, is dug during the breeding season and exceptionally cold winters (Barbour and Davis 1974). These scent-marked dens have multiple rooms, entrances, and trails leading to and from hunting areas (Schwartz and Schwartz 1981). In addition to their dens, both males and females will defend their scent-marked hunting territory from intruders (Jones and Bimey 1988). The red fox is primarily an opportunistic carnivore, consuming food items such as rabbits, opossums, muskrats, skunks, rodents, birds, eggs, carrion, invertebrates, snakes, and frogs (Barbour and Davis 1974; Merritt 1987). Some vegetable matter such as fruits and nuts are also consumed when in season (Jones and Bimey 1988). During times of abundant food supply, the red fox will bury surplus food to return to for consumption at a later time (Schwartz and Schwartz 1981). Male and female foxes pair for life, remaining together from midwinter to summer. Females bear one litter per year usually between March and April (Merritt 1987). Gestation periods last from about 49 to 56 days, with most averaging 53 days (Schwartz and Schwartz 1981). The pups are weaned at about 60 days, leave the den in the autumn, and are sexually mature by their first winter (Merritt 1987). Natural predators of the red fox are few but include large hawks and owls, and possibly coyotes (Merritt 1987; Schwartz and Schwartz 1981). Red fox may live from six to ten years in the wild (Schwartz and Schwartz 1981). Exposure Profile Adult red fox weigh from 2.7 to 7 kg (Barbour and Davis 1974; Jones and Bimey 1988). Home ranges vary from 245 to 1,235 acres (Merritt 1987). The food ingestion rates of the red fox range from 0.069 g/g BW/day for a nonbreeding adult, to 0.16 g/g BW/day for a juvenile (U.S. EPA 1993). The water ingestion rate for an adult red fox is estimated to be approximately 0.086 g'g BW/day (U.S. EPA 1993). To express these values in units of g/day, the highest reported food ingestion rate of0.16 g/g BW/day and the water ingestion rate of 0.086 g/g BW/day were multiplied by the lowest reported body weight of 2.7 kg (2,700 g) to yield a food ingestion rate of 432 g/day and a water ingestion rate of 232.2 g'day (232.2 mL/day). For the purposes of this risk assessment, a diet of 100% small mammals will be assumed. A soil ingestion rate of 2.8 percent of the total diet has been reported (Beyer et al. 1994) for the red fox. To express this value in units of g/day, the soil ingestion rate of 2.8 percent was multiplied by the food ingestion rate of 432 g/day to yield a soil ingestion rate of 12.1 g/day. 2.9.7 Mink (Mustela vison) as Representative of Carnivorous Mammals Justification USEPA 6904 35 0 0 0 A 0 00204 USFW 0622 food items such as small mammals (e.g., mice, chipmunks, rabbits), birds (usually grounddwelling species), reptiles, insects, and occasionally, prey species that are too heavy to lift off the ground (Burton 1989). Tne breeding season starts with aerial courtship displays, commonly followed by mating on a perch and nest-building by both sexes. Nests are placed in tall trees, high rock ledges, or tall cacti and are often refurbished annually for use in consecutive years. Incubation of two to three eggs is carried out by both sexes and lasts for approximately 30 days. The young are able to feed themselves at 4 to 5 weeks and fledge in about 45 days (Bull and Farrand 1977; Burton 1989). Exposure Profile Adult male and female red-tailed hawks are reported to weigh 960 g and 1,235 g, respectively (DeGraaf and Rudis 1983; U.S. EPA 1993). Home ranges vary from 148.26 to 395.36 acres (Kirkwood 1980). The lowest reported body weight of 0.960 kg and the smallest reported home range of 148.26 acres were assumed for this risk assessment. The diet of a red-tailed hawk consists of mammals, birds, reptiles, and insects which vary in importance with season and availability (U.S. EPA 1993). For the purposes of this risk assessment, the hawk will be assumed to consume 100% small mammals. Food ingestion rates are reported to range from 136 to 400 g/day (Kirkwood 1980). The highest reported food ingestion rate o f400 g/day was assumed for this risk assessment. A water ingestion rat of approximately 0.059 g/g BW/day has been estimated for this species (U.S. EPA 1993). To express this value in units of g/day, the water ingestion rate was multiplied by the lowest reported body weight of 960 g to yield a water ingestion rate o f56.64 g/day (56.64 mL/day). A soil ingestion rate for the red-tailed hawk could not be found in the literature; therefore, the amount of soil predicted to be entrained in the digestive tract of a white-footed mouse was used to calculate this value. A soil ingestion rate of less than 2 percent of the total diet has been reported (Beyer et al. 1994) for the white-footed'mouse. From this value, a conservative soil ingestion rate of 1.9 percent of the total diet was assumed for the whi.tefooted mouse. To express this value in units of g/day, the soil ingestion rate of 1.9 percent was multiplied by the food ingestion rate of the white-footed mouse (4.50 gMay) (U.S. EPA 1993) to yield a soil ingestion rate of 0.09 g/day. This value was assumed to represent the amount of soil entrained in the digestive tract of the white-footed mouse that remains constant over time. To express 0.09 g in units of grams of soil per gram of mouse body weight, this value was divided by the lowest reported body weight (13 g) of the white-footed mouse (Merritt 1987) to yield a value of 0.007 g/g BW. This value was then multiplied by the food ingestion rate of the red-tailed hawk (400 g'day) to yield a soil ingestion rate of 2.8 g'day. 2.9.6 Red Fox ( Vu!pes vuipes) as Representative of Carnivorous Mammals Justification The red fox was selected as representative of a carnivorous mammal due to its dietary composition, relative abundant distribution, and likelihood of occurrence at the Dry Run Creek site. Its diet allows for the evaluation of contamination in site soils. In addition, the concentration of contaminants found in small mammal tissue will also provide an accurate 34 USEPA 6903 goo. 000205 USFW 062- The mink was selected as representative of a carnivorous mammal due to its dietary composition, relative abundant distribution, and likelihood of occurrence at the Dry Run Creek site. Its diet allows for the evaluation of contamination in site soils. In addition, the concentration of contaminants found in clams and fish tissue will also provide an accurate dose to the mink which allows for the evaluation of contaminants in the food source. Lift History Mink are distributed over much of boreal North America, southward throughout the eastern United States and in the west to California, New Mexico, and Texas (Jones and Bimey 1988). They can be found in virtually any habitat containing permanent water thus, they are not commonly found in upland areas (Jones and Bimey 1988). Although primarily nocturnal, their activity often extends into midday (Hoffmeister 1989). Dens are always near water, and they are usually an old muskrat burrow or constructed by the mink itself (Jones and Bimey 1988). Males tend to live in their own burrows which are less elaborate than ones occupied by females (Barbour and Davis 1974). Home ranges tend to be linear since mink often follow a shoreline (Jones and Bimey 1988). Mink are solitary and mark their territories by spraying (Merritt 1987). Seasonal food availability governs the dietary composition (Barbour and Davis 1974). Their diets may consist of crayfish, frogs, fish, snakes, rodents, rabbits, and plants among other items (Jones and Bimey 1988; Schwartz and Schwartz 1981). Crayfish are a major portion of the summer diet in many regions of North America (Barbour and Davis 1981; Jones and Bimey 1988; Merritt 1987). Breeding occurs from January to early April with highly variable gestation periods ranging from 40 to 75 days (Merritt 1987; Schwartz and Schwartz 1981). A highly variable single liner of 1to 17 young may be produced (Schwartz and Schwartz 1981). Average liner sizes vary among regions (Barbour and Davis 1974; Hoffmeister 1989; Jones and Bimey 1988; Merritt 1987; Schwartz and Schwartz 1981). Young are weaned at about five to six weeks of age and are sexually mature by ten months (Merrin 1987; Schwartz and Schwartz 1981). Occasionally great homed owls, foxes, coyotes, bobcats, and dogs will prey on mink (Merrin 1987; Schwartz and Schwartz 1981). Although some individuals have lived up to six years, mink seldom exceed two years of age in the wild (Schwartz and Schwartz 1981). Effects Profile Adult mink weigh from 520 to 1,730 g (Merrin 1987; U.S. EPA 1993). Home ranges'vary from 19 to 1,900 acres (U.S. EPA 1993). A year-round food ingestion rate of 0.22 g/g BW/day has been estimated for both male and female mink (U.S. EPA 1993). To express this value in units of g/day, the food ingestion rate was multiplied by the lowest reported body weight (520 g) to yield a food ingestion rate of 114 g/day. An estimated water ingestion rate of 0 .11 g/g BW/day was reported for farmraised females (U.S. EPA 1993). To express this value in units of g/day, this water ingestion rate was multiplied by the lowest reported body weight of 520 g to yield a water ingestion rate of 57.2 g/day (57.2 mL/day). For the purposes of this risk assessment, a diet of 100% fish will be assumed. USEPA 6905 36 009 15 000206 USFW 0623 An incidental sediment ingestion rate was not available from the literature; therefore, a predicted incidental ingestion rate for sediment that may be entrained in the digestive system of the prey item (fish) was used for this risk assessment. Consumption of this prey item was assumed to be the primary mechanism by which mink may incidentally ingest sediment. The derivation of the predicted level of incidental sediment ingestion via consumption of fish is described next. Life history information for the bluegill (Lepomis machrochirus) was used to predict the amount of sediment that may be ingested by mink via consumption of fish. Adult bluegills range in size from 100 to 230 mm (Pflieger 1975; Smith 1985). In keeping with the conservative approach of this risk assessment, the amount of sediment entrained in the lowest body size of 100 mm in length was predicted. The weight of a 100 mm bluegill was calculated to be 18.11 g based on the following algorithm relating length to weight (Hillman 1982); log Weight (g) - -5.374 +3.316 log Length (mm) A daily food ingestion rate of 1.75 percent BW/day has been reported for the bluegill (Kolehmainen 1974). This provides a predicted intake rate of 0.32 g of food per day for a 18.11 g fish. A study evaluating the stomach contents of 153 bluegills reported an average content of detritus and sediment to be 9.6 percent of the total diet (Kolehmainen 1974). If a conservative assumption is made that 9.6 percent of the food ingested is entirely sediment,' it can be predicted that a fish o f this size may contain 0.03 g of sediment in its digestive system. For the purpose of this model, it was assumed that the level of sediment contained in the digestive system of a fish remains constant over time. This value (0.03 g) was divided by the predicted fish body weight (18.11 g) to express sediment entrained in fish digestive systems in units of grams of sediment per gram of fish body weight. This provided a value of 0.0017 g sediment/g body weight. When this value is multiplied by the food ingestion rate of the mink (114 g/day), the predicted sediment ingestion rate for the mink through consumption of fish is 0.2 g/day. 2.9.8 Raccoon (Procyon lotor) as Representative of Omnivorous Mammals Justification The raccoon was selected as representative of a omnivorous mammal due to its dietary composition, relative abundant distribution, and likelihood of occurrence at the Dry Run Creek site. Its diet allows for the evaluation of contamination in site sediment. In addition, the concentration of contaminants found in forage fish tissue and clams will also provide an accurate dose to the raccoon which allows for the evaluation of contaminants in the food source. Life History Raccoons are medium-sized omnivores and are abundant throughout North America. Raccoons prefer aquatic habitats, particularly hardwood swamps, flood plains, freshwater wetlands, and salt marshes (Kaufrnann 1982). Raccoons have also adapted well to residential areas and farmlands. Raccoons rely heavily on surface waters for foraging and as a source 37 USEPA 6906 000207 I tOCAA/ n c o / i of drinking water (Srnewer 1943). Raccoons are active primarily from dusk to dawn (Stuewer 1943) but will alter their activities to opportunistically feed on whatever is available (Sanderson 1987). For example, raccoons living near a salt marsh may become active during the day to take advantage of feeding opportunities during low tide (Ivey 1948). Raccoons feed primarily on fruits, nuts, acoms, grains, insects, frogs, crayfish, and eggs (Palmer and Fowler 1975). Raccoons in the southern regions of the United States are active year round (Goldman 1950). Adult raccoons are normally solitary but will come together for short periods of time during mating (Kaufman 1982). Mating occurs from March to June in southern areas and each male may mate with several females during each season (Sanderson 1987; Kaufman 1982). Young males are normally not sexually mature in the first breeding season but mature later in the summer, while females mature in the first year (Sanderson 1951). The home range of a raccoon depends on the animal's age, habitat, food resources, and season (Sanderson 1987). Home ranges are typically a few hundred hectares (ha) but ranges as large as a few thousand ha have been reported (Sanderson 1987). Population densities also depend strongly on the amount of resources in the area. Numbers of 0.1 to 0.2 animals per ha are common (Hoffman and Gottschang 1977). Raccoons are found near every aquatic habitat. During the last 50 years raccoon populations have increased greatly (Sanderson 1987). In Alabama, adult male raccoons weighed up to 8.8 kilograms (kg) (mean 43 1 kg) while adult females can weigh up to 5.9 kg (mean 3.67 kg) (Johnson 1970). Adult raccoons weigh between 2 and 12 kg (Nowak 1991), and consume 0.5 kg of food per day (Newell et al. 1987). Raccoons feed primarily on fruits, nuts, acoms, grains, insects, frogs, crayfish, eggs (Palmer and Fowler 1975). In a Maryland forested bottom land, the dietary composition of raccoons during the summer was principly made up of insects (39 percent), wild cherry (17 percent), blackberries (16 percent), crayfish (8 percent), snails (5 percent), herptiles (5 percent), fish (2 percent), rodents (2 percent), com (1 percent), and trace amounts of Smilax, acoms and pokeberry (Llewellyn and Uhier 1952). At Washington state tidewater area raccoons displayed the following dietary composition: molluscs, mussels and oyster (44 percent), Crustacea, shrimp and crabs (25 percent), fish (9 percent), marine worms (20 percent), and Echiurida worms (1 percent) (Tyson 1950). The home range of a raccoon depends on the animal's age, habitat, food resources, and season (Sanderson 1987). Home ranges are typically a few hundred hectares but ranges as large as a few thousand hectares have been reported (Sanderson 1987). The home range for adult male raccoon found in coastal Georgia raccoons is approximately 65 ha ( 18 SE) while the home range for adult females in the same area is approximately 39 ha ( 16 SE) (Lotze 1979). Population densities also depend strongly on the amount of resources in the area. Numbers of 0.1 to 0 2 animals per hectare is common (Hoffman and Gottschang 1977). EiSPOSUr? Profile For the purposes of this risk assessment, a body weight of 2 kg, an ingestion rate of 0.5 kg'day, and a diet of 80 percent forage fish and 20 percent clams were assumed. A soil ingestion rate of 9.4 percent of the diet has been reported for raccoons (Beyer et al. 1991). Multiplying the ingestion rate by 9.4 percent yields a sediment ingestion rate of 0.047 kg'day. 38 USEPA 6907 o o o : 000208 USFW 0625 A daily water ingestion rate of 0.18 Liters per day (IVday) was calculated using an allometric equation derived by Calder and Braun (1983). A diet of 100% fish will be assumed. ^rr.raiied Shrew (Blarina brevicauda) as Representative of Insectivorous Mammals Justification The short-tailed shrew was selected as representative of insectivorous mammals because of its dietary composition, relative abundant distribution in both moist and dry habitats, and likelihood of occurrence at the Dry Run Creek site. Although their diets may consist of plants as well as insects, they tend to favor soil invertebrates when they are in abundance. Hence, by assuming that their dietary composition comprises solely invertebrates in this risk assessment, this species may represent an insectivorous mammal. Life History The short-tailed shrew is an extremely active, large, and heavy-bodied shrew common within its range (Jones and Bimey 1988). It occupies a variety of moist and dry habitats such as marshes, bogs, moist forest floors with ample decaying matter, brushland, fencerows, weedfields, and pastures (Barbour and Davis 1974; Jones and Bimey 1988). Short-tailed shrews are active both day and night throughout the year, although most of this activity is subnivean (Merritt 1987). During harsh winters, this species may undergo a period of torpor' (Hoffmeister 1989). The home range of this species varies with their dramatic population cycles. In peak years, animal density may be greater than 25 individuals per acre (Schwartz and Schwartz 1981). In other years, this species may have an animal density of one individual per acre (Merritt 1987). Although short-tailed shrews strongly prefer animal matter, they'are opportunistic omnivores and will voraciously consume whatever food items are in ample supply (Barbour and Davis 1974). These food items include earthworms, slugs, snails, insects, arthropods, fungi, vegetable matter, seeds, snakes, salamanders, small mammals, and young birds CBarbour and Davis 1974; Jones and Bimey 1988; Schwartz and Schwartz 1981). Plant matter is generally consumed to a greater extent in winter (Schwartz and Schwartz 1981). In some regions, plant matter may constitute up to 20 percent of the shrew's diet (Barbour and Davis 1974). Submaxillary glands ptoduce a venom that quickly immobilizes their prey (Merritt 1987). Prey items that are not consumed immediately are stored in a cache (Merritt 1987). Using echolocation and scent-marking, short-tailed shrew rely heavily on their hearing and sense of smell to locate food and to move about (Hoffmeister 1989). An elaborate system of runways and tunnels are constructed usually just a few inches below the ground surface (Schwartz and Schwartz 1981). Two types of nests are built by this species, a breeding nest and a resting nest. Both nests are built underground beneath a log. rock, or other cover, and have multiple entrances. The breeding nest is typically larger than the resting nest (Merritt 1987). Breeding appears to commence in early spring and extend into the fall, although in some regions, breeding may subside in early and midsummer but peak again in early fall (Hofimeister 1989; Jones and Bimey 1988). Gestation periods are approximately 21 to 22 3 9 USEPA 6908 000:* 8 000209 U SFW ncoo days with litter sizes of approximately four to ten young (Jones and Bimey 1988; Schwartz and Schwartz 1981). The young are fully mature from one to three months of age (Barbour and Davis 1974; Schwartz and Schwartz 1981). Both sexes may breed their first spring (Schwartz and Schwartz 1981). Natural predators of the short-tailed shrew include fish, snakes, owls, hawks, shrikes, opossums, raccoons, foxes, weasels, bobcats, skunks, and feral cats, although most of these predators do not consume the shrew (or at least all of the shrew) because of their distasteful musk glands (Barbour and Davis 1974; Jones and Bimey 1988; Merritt 1987; Schwartz and Schwartz 1981). The life expectancy of a short-tailed shrew in the wild is approximately one year (Schwartz and Schwartz 1981). Exposure Profile Adult short-tailed shrews weigh from 12 to 30 grams (g) (Jones and Bimey 1988; Merrin 1987). Home ranges vary from 0.5 to 1 acre (Memtt 1987). Therefore, it was assumed that a short-tailed shrew could obtain 100 percent of its diet from the contaminated area (area use factor of 1), since the area comprising the on-site sampling locations was approximately 20 acres. Food ingestion rates ranging from 0.49 to 0.62 gram per gram of body weight per day (g/g BW/dav ) have been reported (U.S. EPA 1993). An average food ingestion rate of 7.95 g'dav has also been reported (U.S. EPA 1993). To express the former food ingestion rates in units of g/day for comparison to the latter ingestion rate, the former ingestion rates were multiplied by the lowest reported body weight of 12 grams to yield food ingestion rates of 5.SS to 7.44 g/day. Of these values, the highest food ingestion rate of 7.95 g'dav will be used for the purposes of this risk assessment. A water ingestion rate of 0.223 g/g BW/day has been reported (U.S. EPA 1993). To express this value in units of g/day, the water ingestion rate was multiplied by the lowest reported body weight of 12 g to yield a water ingestion rate of 2.7 g'day (2.7 milliliters per day [mL/day]). A soil ingestion rate for the short-tailed shrew was not available from the literature, therefore, the soil ingestion rate of the opossum was used. The opossum's diet is similar to that of the short-tailed shrew since they are both opportunistic omnivores with a strong preference for animal matter (Schwartz and Schwartz 1981). A soil ingestion rate of 9.4 percent of the diet was reported for the opossum (Beyer et al. 1994). This value was multiplied by the highest food ingestion rate of the short-tailed shrew (7.95 g/day) to yield a soil ingestion rate of 0.74 g'dav. For the purposes of the food chain model in this risk assessment, it was assumed that 100 percent of the diet of the short-tailed shrew was comprised of earthworms. .9.10 Meadow Vole (Microtus pennsylvanicus) as Representative of Herbivorous Mammals Justification The meadow vole was selected as representative of herbivorous mammals because of its dietary composition, abundance in North America, preference for moist areas, and likelihood of occurrence at the Dry Run Creek site. USEPA 6909 000210 OOO-JS USFW 0627 t.ife History The meadow vole is one of the largest and most abundant voles in North America (Jones and Bimey 1988; Merritt 1987). Although they are more commonly found in habitats such as moist meadows, bogs, swamps, stream banks, and lakeshores, they have also been known to inhabit cultivated fields, roadside ditches, and fencerows (Barbour and Davis 1974; Jones and Bimey 1988; Merritt 1987; Schwartz and Schwartz 1981). Dense vegetative cover appears to be one of the major prerequisites for habitation (Hoffineister 1989; Jones and Bimey 1988). The home range of the meadow vole varies in size with season, habitat, and population size (Jones and Bimey 1988). Populations tend to fluctuate drastically every two to five years, with peak population density levels exceeding 100 voles per acre (Barbour and Davis 1974; Jones and Bimey 1988). Activity occurs during both day and night, and throughout the year, although it is greatest at dawn and dusk (Barbour and Davis 1974). Well-worn intersecting runways under vegetative cover are distinctive of meadow vole inhabitation (Jones and Bimey 1988). Elaborate spherical nests are commonly built aboveground in the center of a tussock of grass, although underground nests are also built in drier areas (Barbour and Davis 1974; Jones and Bimey 1988). The meadow vole is herbivorous, feeding primarily on grasses, sedges, legumes, tubers, and roots (Merritt 1987); however, insectivory and cannibalism have been reported in some individuals (Barbour and Davis 1974; Hoffineister 1989). Bluegrass (Poa sp.) is a major component of the diet in some regions (Jones and Bimey 1988; Hoffineister 1989). This species hoards food for the winter in above- and below-ground caches (Merritt 1987). The meadow vole is one of the most prolific mammals, producing litter after litter in rapid succession (Barbour and Davis 1974). Breeding occurs during the warmer months of the year (Jones and Bimey 1988). The gestation period is about 21 days with litter sizes ranging from 1 to 11 young (averaging four to seven) (Barbour and Davis 1974; Jones and Bimey 1988). The helpless young mature rapidly and may breed by 25 days of age (Barbour and Davis 1974). Meadow voles are preyed upon by nearly all species of predatory birds and mammals (Barbour and Davis 1974). These predators include owls, hawks, shrikes, bluejays, crows, foxes, weasels, mink, cats, raccoons, skunks, opossums, shrews, and snakes (Barbour and Davis 1974; Merritt 1987). Due to heavy predation, only a small proportion of the population exceeds sixty days of age (Schwartz and Schwartz 1981). Exposure Profile Adult meadow voles weigh from 20 to 65 grams (Merritt 1987; U.S. EPA 1993). The home range of this species varies from less than one acre to 3.2 acres (Merritt 1987). Therefore, it was assumed that a meadow vole could obtain 100 percent of its diet from the contaminated area (area use factor of 1), since the area comprising the on-site sampling locations was approximately 20 acres. A food ingestion rate ranging from 0.30 to 0.35 g/g BW/day, and a mean water ingestion rate of 0.21 g;g BW/day is reported for this species (U.S. EPA 1993). To express these values in units of g'day, the highest reported food ingestion rate of 0.35 g'g BW/day and the water 41 USEPA 6910 G GQ -.C.O 0 00 2 11 USFW 0628 ingestion rate of 0.21 g/g BW/day were multiplied fay the lowest reported body weight of 20 g to yield a food ingestion rate of 7.0 g/day and a water ineestion rate of 4 2 g/dav (4 > ml/day). " 3v A soil ineestion rate the meadow voie. o1f02e.4xppreerscsentht ios f*thieuteoiunl di* units 'of *--n re^-- g/day, the u '.tSe>'er et al- 1994) soil ingestion rate of for 2.4 percent was multiplied by the food ingestion rate of 7.0 g/day to vi*<>* - .u uigcsuon rate of 0.17 g/day. For the purposes of the food chain model in this risk assessment, it was assumed that 100 percent of the diet was comprised of plants. .0 ASSUMPTIONS This risk assessment evaluates exposure to contaminants through food and incidental sediment/soil ingestion. The following conservative assumptions were made to conduct this risk assessment in the absence of site- specific data: The maximum of the contaminant levels measured in sediment, soil, or water collected on site was used in risk calculations. The maximum concentrations of COCs reported in sediment, soil, water, and biota were assumed to be present site-wide. .; An area use factor (AUF) of 1 was assumed for all species using the site for feeding. Contaminants were assumed to be 100 percent bioavailable. Dietary composition information was obtained from the literature for the receptor species. However, simplifications of complex diets were performed for the receptors. A literature search was conducted to determine the chronic toxicity of the contaminants of concern when ingested by the indicator species. If no toxicity values could be located for the receptor species, values reported for a closely related species were used. All studies were critically reviewed to determine whether study design and methods were appropriate. When values for chronic toxicity were not available, LD,a (median lethal dose) values were used. For purposes of this risk assessment, a factor of 100 was used to convert the reported LDJ0 to a No Observed Apparent Effect Level . (NOAEL). A factor of 10 was used to convert a reported Lowest Observed Adverse Effect Level (LOAEL) to a NOAEL. and a factor of 10 was used to convert a reported LD,, to a LOAEL. If several toxicity values were reported for a receptor species, the most conservative value was used in the risk calculations regardless of toxic mechanism. Toxicity values obtained from long-term feeding studies were used in preference to those obtained from single dose oral studies. No other safety factors were incorporated into this risk assessment. In some cases, contaminant doses were reported as part per million contaminant in diet. These were convened to daily intake (in milligrams per kilogram body weight per day; mg/kg-day) by using the formula: Intake (mg'kg/day)*Contaminant Dose (mg/kg diet) x Ingestion Rate (kg/day) x 1/Bodyweight (kg) 42 0 0 0-: n USEPA 6911 000212 USFW 0629 This conversion allows dietarytoxicity levels cited for one species to be convened to a daily dose for a different species based on body weight. For the purposes of this risk assessment, incidental nii/sediment ingestion was also included in the calculation to determine the total daily intake for the recepto. . r -n-;- j,;ivdose may then be used to evaluate the risk to other species if no specific toxicity data art available for a (ccepiu>. Some containin of cericer" (e.g. aluminum) are not food chain accumulators, but instead are direct toxins when ingested at the prescn-.! levels. EFFECTS PROFILE Many contaminants detected at the Dry Run Creek site do not have benchmarks. This excluded them from further consideration in this risk assessment, but does not exclude them as potential contaminants of concern. Based on the results of the preliminary risk assessment, the following compounds were considered COCs and their toxic effects are presented next: fluoride, trichlorofluoromethane, aluminum, arsenic, beryllium, chromium, copper, iron, lead, manganese, nickel, vanadium, and zinc. Based on the chemistry results, these compounds will be further evaluated using food chain accumulation models. Contaminants exceeding their respective benchmarks are assumed to be affecting receptor species and negatively impacting species, populations, and communities in the aquatic and terrestrial ecosystems at the Dry Run Creek site. 4.1 Fluoride Maurer et al (1990) identified skeletal and dental abnormalities in rats that were exposed to sodium fluoride for a period o f 99 weeks. The LOAEL identified in this study was 4 mg Fl/kg BW/day. A NOAEL was calculated from the LOAEL using an accepted conversion factor of 10. Based on these results, a LOAEL of 4mg/kg BW/day and an estimated NOAEL of 0.4/kg BW/dav will be used to evaluate the risk posed by fluoride mammalian receptors Fleming et al. (1987) found significant growth rate reduction in European starling fed a diet containing as low as 13 mg Fl/kg BW/day. No effects were observed at 10 mg Fl/kg BW/day. As such, this risk assessment will estimate fluoride related risk using a LOAEL of 13 mg'kg BW/day and a NOAEL of 10 mg'kg BW/day. 4.2 Organofluorides No studies pertaining to the dietary toxicity of trichlorofluoromethane or any other fluorinated oreanic compound was found in the literature. 4.3 Aluminum Dixon et al. (1979) conducted a study that evaluated the reproductive success of rats exposed to aluminum in drinking water for 90 days prior to breeding. The highest dose administered was 77.5 milligrams per kilogram body weight per day (mg/kg BW/day) and did not result in reproductive abnormalities. Lai et al. (1993) conducted a 180-day drinking water study in which rats were exposed to 55 mg'kg BW/day of aluminum. At this dose, behavioral effects were observed, including a significant reduction in spontaneous locomotor activity and significant deficits in acquisition and retention of learned responses. Based on these results, a LOAEL of 55 mg/kg BW/day and an estimated NOAEL of 5.5 mg'kg BW/day will be used to evaluate the risk posed by aluminum to mammalian receptors (Table 1). USEPA 6912 43 000213 0 0 0 . ; ' '' IOC1M n o n , No effects were observed when Japanese quail were fed a diet containing 0.05 percent (84 mgkg BW/day) aluminum for four weeks (Hussein et al. 1988). When quail were fed a diet containing 0.1 percent (165 mg/kg BW/day) aluminum, a decrease in egg shell breaking strength was observed. Finally, when quail were fed a diet containing 0.15 percent (257 mg/kg BW/day) aluminum, a decrease in body weight, egg shell strength, and egg shell production was observed. A 48-day feeding study using chickens concluded that dietary levels of 28.4 mg/kg BW/day aluminum resulted in a decrease in weight gain, feed intake, and plasma inorganic phosphorus, as well as an increase in plasma calcium (Hussein 1990). However, only the altered metabolism of calcium and phosphorus could be attributed to the direct effects of aluminum. The associated NOAEL for this effect is 22.8 mgkg BW/day. Because a range of concentrations were used and the endpoints were ecologically significant and related to the dose, the study by Hussein et al. (1988) was used to the develop the NOAEL and LOAEL values. A NOAEL of 84 mg/kg BW/day and a LOAEL of 165 mg/kg BW/day will be used to evaluate the risk posed by aluminum to avian receptors (Table 1). 4.4 Arsenic Several studies were located which determined the effects of As to mammals. A study conducted on cats indicated that a chronic oral toxicity dose was 1.5 mg/kg BW/day (Pershagen and Vahter 1979). In addition. National Resources Council of Canada (1978) states that mammals in general have oral LDwS that range from 10 to 50 mg/kg of lead arsenate. A study conducted on mice indicated an oral dose LDj0of 39.4 mg/kg BW/day and an oral dose LD#of 10.4 mg/kg BW/day after 96 hours (NAS 1977). For the purposes of this risk assessment, the chronic value for the cat was used to calculate HQs for mammals (1.5 mg/kg BW/day). This value was convened to a NOAEL by dividing by a factor of 10. Eisler (1988a) reviewed several studies in which the toxicity of inorganic arsenicals were measured. Inorganic As is more mobile than organic As and may pose greater risk by leaching into surface water. Studies were also described in which organoarsenical compounds were measured. Studies indicate that sensitive species include the California quail (single oral dose LDJ0 of 47.6 mgkg BW/day) (Hudson etal. 1984) and chicken (single oral dose LD* of 33 mg/kg BW/day) (NAS 1977). For the purposes of this risk assessment, a value of 3.3 mg/kg BW/day was used to determine the HQ to birds. This value was convened to a NOAEL by dividing by a factor of 10. 4.5 Beryllium Two separate chronic dietary exposure studies using rats reponed similar musculoskeletal effects. Guyatt et al. (1933) fed large amounts of beryllium carbonate to rats at concentrations of 10, 20,40, 80, 160, and 240 mgkg BW/day. Rats from all exposure levels developed rickets, with the fragility of the bones varying directly with the exposure concentration. Similar results were reponed by Jacobson (1933) who reponed severely weakened bones in rats fed beryllium carbonate at dietary levels of 121 and 242 mgkg BW/day. For this risk assessment, a dietary exposure level of 10 mgkg BW/day was used to estimate risk of beryllium to the short-tailed shrew. A NOAEL of 0.10 mgkg BW/day was derived from this LOAEL using an accepted conversion factor of 10. No studies pertaining to the dietary toxicity of beryllium to avian receptors were found in the literature. 4.6 Chromium USEPA 6913 44 G00`:'-'53 0 0 0 2 1 4 USFW 0631 Heinz and Haseltine (1981) exposed 2- to 3-year old breeding pairs of black ducks (Anas mbripes) to a diet containing 0, 20, or 200 mg/kg, wet weight, (0.2.77, or 27.77 mg/kg BW/day) of Cr*1as chromium potassium sulfate [CrK (S O ^'PHjO] for a period of approximately five months, until the onset of egg-laying by the females. Hatched ducklings were then fed a mash diet containing the same Cr concentrations that the parents were fed. Seven-day old chicks were tested for avoidance behavior in response to a fright stimulus. None of the Cr concentrations resulted in alteration of avoidance behavior. However, Haseltine et al. (1985), in an unpublished study reported by Eisler (1986a) notes that black duck ducklings suffered reduced survival and altered growth patterns when exposed to 10 mg/kg and 50 mg/kg of an unspecified Cr'5compound in their diets. The percent reduction in survival and a detailed explanation of the altered growth patterns were not available in this unpublished study. For the purposes of this risk assessment, dietary levels of 10 mg/kg (1 mg/kg BW/day) of Cr in prey was used as a LOAEL for the avian species. However, due to the conflicting results, a NOAEL was derived from the same study in which the LOAEL was selected to maintain a degree of consistency regarding the Cr species evaluated. A NOAEL of 0.1 mg/kg BV//day was derived from the LOAEL using a conversion factor of 10. A study conducted with dogs indicated that 2.5 mg/kg/day of Cr** ingested in the diet caused death (Steven et al. 1976). For the purposes of this risk assessment, a LOAEL of 0.25 and a NOAEL of 0.025 were used for the red fox, raccoon, and mink. 4.7 Copper One study was located which determined the effects of ingestion of Cu to mammalia species. An oral dose of 100 mg/kg/day to a dog caused death (OHMD 1987). For the purposes of this risk assessment, a LOAEL of 10 mg/kg/day was used and a NOAEL of I mg/kg/day were used for the exposure of mammals. Several studies were located which determined the effects of Cu on chickens. A dose of 350 mg/kg (61.3 mgkg/day) caused a significant decrease in growth and food consumption (Smith 1969). Another study found that a dose of 325 mg/kg (23.5 mg/kg/day) caused respiratory problems (Hatch 1978). Assuming that respiratory problems are an acute effect, a LOAEL of 2.35 mg/kg/day and a NOAEL of 0.235 mg/kg/day were used to determine risk to avian species. 4.8 Iron No studies pertaining to the dietary toxicity of iron to mammalian or avian receptors were found in the literature. 4.9 Lead The gastric motility of adult male and female red-tailed hawks fed 0.82 and 1.64 mg Pb/kg BW/day in a single oral dose was evaluated through the use of surgically implanted transducers for a period of three weeks following the dose. Neither concentration had any effect on gastric contractions or egestion of undigested material pellets (Lawler et al. 1991). A study conducted on red-tailed hawk found that 3 mg/kg/dav of Pb caused the clinical symptoms of Pb poisoning (Reiser and Temple 1981). A similar study found that 3 mg/kg'day fed to starlings caused a reduction in muscle condition and altered their feeding activity (Osbome et al. 1983). For USEPA 6914 45 O 0O -\c 000215 I ICCHA/ n c n r the purposes of this risk assessment, a LOAEL of 3 mg/kg/day was used to determine risk to avian species and a NOAEL of 0.3 was used. Several studies were located which determined the effects of Pb ingestion to mammals. A study conduced on mice indicated that 1.5 mg/kg/day of Pb caused a reduction in success of implanted ova (Clark 1979). Another study found that 2 2 mg/kg/day caused a reduction in the frequency of pregnancy when the dose was administered 3 to 5 days following mating (Clark 1979). For the purposes of this risk assessment, a NOAEL of 0.15 mg/kg/day and a LOAEL of 1.5 mg/kg/day were used to determine risk to mammals. 4.10 Manganese The effects ievels for manganese toxicity vary widely, most likely attributable to the form of manganese tested. Rats exposed to 13 mg/kg BW/day of manganese as Mn304 in their diet for 224 days exhibited reduced testosterone levels (Laskey et al. 1982). In mice, a dietary level of 140 mg/kg BW/day, also of Mn304 for 90 days resulted in decreased activity (Gray and Laskey 1980). A much higher exposure concentration of 2,300 mg/kg BW/day of manganese as MnC12 resulted in reduced dopamine levels (Gianutsos and Murray 1982). In contrast, levels as high as 930 mg/kg BW/day of manganese as MnS04 for 103 weeks had no efFec on the respiratory, cardiovascular, gastrointestinal, hematological, musculoskeletal, hepatic, renal, dermal, and ocular systems of mice (Hejtmancik et al. 1987). For this risk assessment, a dietary exposure level of 13 mg/kg BW/day will be used as a LOAEL to estimate risk of manganese to the selected mammalian receptor. A NOAEL of 1.3 mg-kg BW/day was derived from this LOAEL using an accepted conversion factor of 10. No studies pertaining to the dietary toxicity of manganese to an avian receptor were found in the literature. 4.11 Nickel Several studies were available which determined the effects of Ni ingestion to mammals. Wistar rats fed Ni sulfate indicated a NOAEL of 187.5 mg/kg/day to most systems except for body weight. This level of Ni sulfate caused a 27 to 29 percent decreased body weight (Ambrose et al. 1976). In a similar study with a beagle, a NOAEL of 62.5 mg/kg/day was noted (Ambrose et al. 1976). For the purposes of this risk assessment a NOAEL of 62.5 mg/kg/day was used to determine risk to mammals. This value was converted to a LOAEL of 625.0 mg/kg'day by multiplying the NOAEL by a factor of 10. No studies were available that determined the dose of Ni to avian species. Tnerefore, the risk to avian species from ingested Ni will not be determined. 4.12 Vanadium Gavage studies in mice have found an LC50 of 31 mg Vn/Kg diet (Schroeder and Balassa 1967). This dose was converted to a LOAEL of 3.1 mg/kg and a NOAEL of 0.31 using an accepted factor of 10 conversion. This food dose was convened to a daily dose by multiplying the LOAEL or NOAEL concentration by an ingestion rate commonly observed in mice (0.003 kg of food/day) and then by the inverse of the body weight (0.025 kg)(RTECS 1985). This calculation resulted in a LOAEL of 0.372 46 USEPA 6915 000'io.[) 00023.6 USFW 0632 mg V/kg BW/day and a NOAEL of 0.0372 mg V/kg BW/day. These values will be used to estimate risk to mammalian receptors in this risk assessment. Rosomer (1960) exposed chickens to varying concentrations of vanadium. The study involved feeding 4 replicates of 13 chickens varying concentrations of vanadium for a period of 21 days. The study found that a dietary level of 40 mg/kg in the diet resulted in a marked depression in weight gain and efficiency of food utilization. At levels of 200 mg/kg, mortality was noted in all test chickens. The authors reported that a dietary level of 20 mg/kg could be tolerated with no resultant toxic effects. This dietary level was converted to a daily dose as above by multiplying the dietary concentration by a representative chicken ingestion rate (0.140 kg/day) and then by the inverse of the body weight (0.800 kg)(RTECS 1985). This calculation resulted in LOAEL of 7 mg V/kg BW/day and a NOAEL of 3.5 mg V/kg BW/ day. These values will be used to estimate risk to avian receptors. 4.13 Zinc Several studies were available which determined the effects of ingested Zn to birds. A concentration of 144.5 mg/kg/day caused a decrease in growth and anemia in chickens (Stahl et al. 1989). In a similar study conducted on chickens, a concentration of 361 mg/kg/day caused a reduction in body weight (Dean et al. 1991). In a study conducted on Japanese quail, a concentration of 139 mg/kg/day caused 7 percent mortality in chicks and reduced food intake (Hill and Camardese 1986). For the purposes of this risk assessment, a LOAEL of 139 mg/kg/day was used to determine the effects to avian species. This value was convened to aNOAEL of 13.9 mg/kg/day by dividing the LOAEL by a factor of 10. A study conducted on dogs, indicated that 1,000 mg/kg (25 mg/kg/day) caused no effects after one year (NAS 1979). For the purposes of this risk assessment, a LOAEL of 250 and aNOAEL of 25 were used to determine risk to the fox and the mouse. In a study conducted on fenets, a dose of 370 mg'kg day caused a decrease in food intake and weight loss. Because the ferret is similar to the mink, a LOAEL of 370 mg/kg/day was used and a NOAEL of 37 was used to determine risk to the mink. RISK CHARACTERIZATION The following method was used to calculate risk. To estimate the risk to wildlife in the model systems utilizing the Dry Run Creek site, implications of the exposure concentrations need to be determined. The HQ method (U.S. EPA 1989, Bamthouse et al. 1986) compares exposure concentrations to ecological endpoints such as reproductive failure or reduced growth. The comparisons are expressed as ratios of potential intake values to population effect levels, or: Hazard Quotient (HQ) - ______ Mean Exposure Concentration No Observed Adverse Effect Level (NOAEL) A HQ greater than one indicates that exposure to the contaminant has the potential to cause adverse effects in the organism. A HQ less than one does not indicate a lack of risk. The HQ should be interpreted based on the severity of the effect reported. The results of the risk characterization are presented next. 5.1 Benthic Invertebrate Community Structure and Function The benthic invertebrate community in Dry Run appears to be at risk for two reasons. The benthic community survey showed a decrease in community taxonomic diversity and abundance in Dry Run as compared to the Reference stream. Since land use and available habitat are the same along both 47 USEPA 6916 0 0 0 -lc S 00021V streams, the decrease in diversity and abundance in Dry Run may be attributed to contamination present in sediments in Dry Run. In addition, the amphipod toxicity test dearly demonstrates that acute exposure sub-lethal effects can be produced in the benthic community, especially under conditions present in Tributary A and in Area II. The observed negative growth effect was significantly negatively correlated with fluoride, aluminum, calcium, magnesium, nickel, potassium, and sodium. Further, there were strong negative associations between the growth endpoint and chromium, copper, lead, and zinc concentrations, although the relationships were not significant at the 0.10 level. Since the sediments closer the landfill along the whole Dry Run reach appear to be enriched with metals, the observed toxicity may represent a significant threat. 5.2 Soil Invertebrate Community Structure and Function The soil invertebrate community does not appear to be at risk based on current soil conditions at Dry Run. The earthworm toxicity test identified no problems with survival or growth. 5.3 Fish Communities The fish community at Dry Run may be at risk. Results of the fathead minnow toxicity bioassay show that water conditions in Upper Tributary A induce mortality to larval fish. This mortality could not directly be associated with a suite of contaminants as in the amphipod test, but survival was negatively correlated with potassium concentrations, however this correlation was not statistically significant at the 0.10 level. There was a significant positive correlation between fathead survival and iron concentrations in the filtered water samples. Low species diversity and abundance observed during the eiectroshocking effort may be reflected by the results of the toxicity test. 5.4 Worm-eating Birds A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that worm-eating birds may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that aluminum, chromium, copper, lead, vanadium, zinc, and fluoride are risk factors based on conservative inputs. By default, beryllium, iron, manganese, nickel, and trichlorofluoromethane are risk factors due to lack of toxicological benchmarks for these compounds. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 5.5 Carnivorous Birds A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that carnivorous birds may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that aluminum, chromium, copper, lead, zinc, and fluoride are risk factors based on conservative inputs. By default, beryllium, iron, manganese, nickel, vanadium, and trichlorofluoromethane are risk factors due to lack of toxicological benchmarks for these compounds. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 5.6 Carnivorous Mammals (Terrestrially feeding) A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that carnivorous mammals may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that aluminum, chromium, copper, lead. 48 009 USEPA 6917 000218 USFW 0635 manganese, vanadium, and fluoride are risk factors based on conservative inputs. By default, iron and trichlorofluoromethane are risk factors due to lack of toxicological benchmarks for these compounds. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 5.7 Piscivorous Mammals A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that piscivorous mammals may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that chromium, manganese, and fluoride are risk factors based on conservative inputs. Trichlorofluoromethane is not considered a risk factor because it was not detected in site sediments. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 5.8 Omnivorous Mammals A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that omnivorous mammals may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that arsenic, chromium, copper, manganese, vanadium, and fluoride are risk factors based on conservative inputs. Trichlorofluoromethane is not considered a risk factor because it was not detected in site sediments. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 5.9 Insectivorous Mammals A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that insectivorous mammals may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that aluminum, chromium, copper, lead, manganese, vanadium and fluoride are risk factors based on conservative inputs. By default, iron, and trichlorofluoromethane are considered risk factors due to lack of toxicological benchmarks for these compounds. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 5.10 Herbivorous Mammals A conservative risk assessment model based on wet-weight concentrations of contaminants for the Dry Run Creek site has determined that herbivorous mammals may be at risk due to ingestion of contaminated forage, soil, and water. The model predicts that aluminum, chromium, lead, manganese, and fluoride are risk factors based on conservative inputs. By default, iron, vanadium, and trichlorofluoromethane are considered risk factors due to lack of toxicological benchmarks for this compound. Food chain risk calculations and resultant hazard quotients are presented in Table 42. 6.0 UNCERTAINTY ANALYSIS There are factors inherent in the risk assessment process which contribute to uncertainty and need to be considered when interpreting results. Major sources of uncertainty include natural variability, error, and insufficient knowledge. Error can be introduced by use of invalid assumptions in the conceptual model. Conservative assumptions were made in light of the uncertainty associated with the risk assessment process. This was done to minimize the possibility of concluding that no risk is present when a threat actually does exist (e.g., elimination of false negatives). Whenever possible, risk calculations were based on conservative values. For example, NOAELs 49 USEPA6918 000219 i io n r r n c used to calculate HQs were the lowest values found in the literature, regardless of toxic mechanism. An important contributor to uncertainty is the incompleteness of the data or information upon which the risk assessment is based. Risk calculations are based on maximum COC levels in sediment, water, and soil samples. Literature values for the toxicity of COCs were not available for all receptor species. An attempt was made to identify studies using closely related species to make risk estimates for the selected receptors. Species respond differently to exposure to toxins; responses to COCs by the indicator species may be different from species for which the toxicity data are reported. Methodological problems were also apparent in several of the studies from which NOAELs were obtained. Unfortunately, studies which were more suitable for this assessment were not found for some of the selected receptors. A literature search was conducted to identify appropriate NOAELs and LOAELs for this risk assessment. The values used to calculate HQs were the lowest values found in the literature. In many of the studies reviewed, adverse effects were observed at the lowest exposure concentration. This made it impossible to identify appropriate NOAELs for some receptors. In these cases, a factor of 10 was used to convert the LOAEL to a NOAEL, which adds uncertainty to the NOAEL-based calculations. Doses in toxicological studies can be reported in units of mg contaminant/kg diet, or in units of mg contaminant/kg body weight/day. All doses reported as mg/kg in diet were convened to units of mg/kg BW/day. Ifbody weights were reponed for the test animals in a given study, these values were used for making this conversion. Otherwise, the body weight and ingestion rate for the species reponed in other literature sources were used. Another source of uncertainty arises from the use of toxicity values reponed in the literature which are derived from single-species, single-contaminant laboratory studies. Prediction of ecosystem effects from laboratory studies is difficult. Laboratory studies cannot take into account the effects of environmental factors which may add to the effects of contaminant stress. NOAELs were generally selected from studies using single contaminant exposure scenarios. Species utilizing the Dry Run Creek site are exposed to a variety of contaminants. There is very little information available in the literature regarding the rates of incidental soil/sediment ingestion for wildlife species. In this risk assessment, most of these values were based on estimates reponed for species similar to the indicator species. Exposure concentrations were calculated for each target receptor species based on levels of contaminants detected in site media, daily food ingestion rates, incidental soil/sediment ingestion rates, and body weight reported in the literature. Tnis ecological risk assessment was conducted with the intent of completing a baseline risk assessment. In this risk evaluation it is concluded that a "potential ecological risk" exists if the HQ calculated from the maximum area concentration and the NOAEL equals or exceeds one. Within the calculation spreadsheets, alternate calculations were made using LOAEL toxicity benchmarks. CONCLUSIONS 7.1 Benthic Invertebrate Community Structure and Function Data from both the benthic survey and the toxicity tests indicate that fluoride and metal contamination may be a significant problem in Dry Run. Numerous fish kills historically reported in Dry Run also 50 USEPA 6919 OOOAoO 0 0 0 2 2 0 USFW 0637 provide evidence for potential effects on the benthic community. 7.2 Soil Invertebrate Community Structure and Function The structure and function of the soil invertebrate community does not appear to be at risk under current conditions found at the Dry Run Creek site. However, since earthworms comprise a significant amount of the forage base of some organisms (e.g. American robins, short-tail shrews, etc.), food chain problems may result from contaminants being tied up in the earthworm tissue. Based on our food chain models, it appears that this may be the case. 7.3 Fish Communities It was shown through the results of the fathead minnow bioassay, that larval fish were susceptible to contamination currently present near the landfill outfall at Dry Run. This finding is further supported by the results of the benthic invertebrate toxicity tests, where toxicity was observed at the same location. Negative effects of contaminants on the benthic community may directly affect fish communities, in that a portion of the fish food base in Dry Run (i.e. benthic invertebrates) may also be removed from the system. Reports of historical fish kills are also an important piece of evidence that suggests an ecological risk. In addition, high levels of metals were noted in the fish, which may present problems to upper level consumers due to dietary toxicity. 7.4 Worm-eating Birds Results of the food chain model for worm-eating birds such as the American robin indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. This risk is associated with these contaminants in the soil and/or in earthworm tissue. Reports of historical wildlife kills also suggest ecological risk to avian receptors. 7.5 Carnivorous Birds Results of the food chain model for carnivorous birds such as the red-tail hawk indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. This risk is associated with these contaminants in the soil and'or in small mammal tissue. Reports of historical wildlife kills also suggest ecological risk to avian receptors. 7.6 Carnivorous Mammals Results of the food chain model for terrestrially feeding carnivorous mammals such as the red fox indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. "his risk is associated with these contaminants in the soil and/or in small mammal tissue. Reports of historical wildlife kills also suggest ecological risk to mammalian receptors. 7.7 Piscivorous Mammals Results of the food chain mode! for piscivorous mammals such as the mink indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. This risk is associated with these contaminants in the soil and'or in fish tissue. Reports of historical wildlife kills also suggest ecological risk to mammalian receptors. 7.8 Omnivorous Mammals 51 USEPA 6920 000 , so 000221 o 5. i io c\M n fi8 8 Results of the food chain model for omnivorous mammals such as the raccoon indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. This risk is associated with these contaminants in the soil and/or in fish tissue. Reports of historical wildlife kills also suggest ecological risk to mammalian receptors. 7.9 Insectivorous Mammals Results of the food chain model for insectivorous mammals such as the short-tail shrew indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. This risk is associated with these contaminants in the soil and/or in earthworm tissue. Physiological abnormalities, specifically the tooth structure of the shrews taken on site, further suggest ecological risk. In addition to the direct potential risk for the shrews, some of these animals had high concentrations of metals and fluoride in their tissues. This could present problems to organisms that feed on shrews and other small mammals on the site due to dietary toxicity. Reports of historical wildlife kills also suggest ecological risk to mammalian receptors. 7.10 Herbivorous Mammals Results of the food chain model for herbivorous mammals such as the meadow vole indicate a potential risk due to metals, fluoride, and trichlorofluoromethane. This risk is associated with these contaminants in the soil and/or in plant tissue. In addition to the direct potential risk for the voles; some of these animals had high concentrations of metals and fluoride in their tissues. This could present problems to organisms that feed on voles and other small mammals on the site due to dietary toxicity. Reports of historical wildlife kills also suggest ecological risk to mammalian receptors. SUMMARY During the past several years, a fanner who grazes his cattle along the reach of Dry Run Creek, has reported severe abnormalities in his herd. These abnormalities have included an increased incidence of stillborn calves, blindness in newborn and adult cattle, erratic behavior, stiffness of gait in adult cattle, abnormal posture, mottled teeth, and a high mortality rate across all age classes of his herd. In addition to problems with his herd, the farmer and others have also reported numerous fish kills in Dry Run, and wildlife kills (e.g. deer) for animals drinking from Dry Run. The results of this risk assessment support his assertion that effluent from the Dry Run Creek landfill may be having adverse effects on the ecological communities that inhabit the old field, deciduous forest, meadow, stream, and riparian habitats that are present on the site. These effects may be related to enriched levels of metals, fluoride, and trichlorofluoromethane that appear to be resultant of the landfill drainage. At a minimum, the symptoms manifest by the herd are characteristic of fluoride toxicity, and consistent with the conclusions of the risk assessment. In addition to the compounds that were studied in this risk assessment, numerous other compounds were present in Dry Run (specifically those identified as TICs or Tentatively Identified Compounds in the BNA scan) that could not be accurately identified. These compounds may also present a threat to the system, an certainly merit further investigation. The DuPont landfill that drains into Dry Run is the only apparent source of trichlorofluoromethane in soils adjacent to the stream. USEPA 6921 000222 1ISFW 0639 LITERATURE CITED Agency for Toxic Substances and Disease Registry (ATSDR). 1990. Toxicological Profilefor Aluminum. Prepared bv Sciences International Inc. Under U.S. Department of Health and Human Services Contract No. 205-93-0606. Research Triangle Park, NC. Agency for Toxic Substances and Disease Registry (ATSDR). 1990. Toxicological Profilefor Manganese. Prepared bv Sciences International Inc. Under U.S. Department of Health and Human Services Contract No. 205-93-0606. Research Triangle Park, NC. Azency for Toxic Substances and Disease Registry (ATSDR). 1991. Toxicological Profilefar Vanadium. Prepared by Sciences International Inc. Under U.S. Department of Health and Human Services Contract No. 205-93-0606. Research Triangle Park, NC. Agency for Toxic Substances and Disease Registry (ATSDR). 1993. Toxicological Profilefo r Beryllium. Prepared by Sciences International Inc. Under U.S. Department of Health and Human Services Contract No. 205-93-0606. Research Triangle Park, NC. Agency for Toxic Substances and Disease Registry (ATSDR). 1996. Toxicological Profilefo r Nickel. Prepared by Sciences International Inc. Under U.S. Department of Health and Human Services Contract No. 205-93-0606. Research Triangle Park, NC. Ambrose, A.M., P.S. Larson, and J.F. Borzelleca. 1976. "Long-term toxicological assessment of nickel in rats and dogs." J. Food. Sci. Technol.. 13:181-187. Barbour, R.W. and W.H. Davis. 1974. Mammals o f Kentucky. Lexington, KY: University of Kentucky Press. 322p. Bamthouse, L.W., G.W. Suter, S.M. Bartell, J.J. Beauchamp, R.H. Gardner, E. Linder, R.V. O'Neill and A.E. Rosen. 1986. Users Manualfor Ecological Risk A tsessment. Publication Number 2679. ORNL-6251. Environmental Services Division, Oak Ridge National Laboratory, Oak Ridge, TN. Beyer, W.N., E.E. Conner, and S. Gerould. 1994. "Estimates of Soil Ingestion by Wildlife." J. W'tldl. Manage., 58(2):375-382. Brooks, L. 1988. "Inhibition of NADPH-cytochrome c reductase and attenuation of acute diethylnitrosamine hepatotoxicity by copper." Ph.D. Dissertation, Rutgers University, New Brunswick, N.J Bull, J. And J. Farrand, Jr. 1977. The Audubon Society Field Guide to North American Birds. Eastern Region. New York, NY: Alfred A. Knopf, Inc. Burton, P. 1989. Birds o f Prey o f the World. New York, NY: W.H. Smith Publishers. Byerrum, R.U., R.E. Eckardt, L.L. Hopkins. 1974. Vanadium. Washington, D.C. National Academy of Sciences. USEPA 6922 53 * 0 0 v- 000223 Effects on Aquatic Biota: 994 Revision. ES/ER/TM-96/R1, Martin Marietta Energy Systems, Inc. Surtie, J.W. 1977. "Effects of Fluoride on Livestock." Journal o f Occupational Medicine. 19(I):40-48. Tyson, E.L. 1950. "Summer Food Habits o f the Raccoon in Southwest Washington." J. Mammal., 31:448-449. U .S . Environmental Protection Agency (U.S. EPA). 1981. A n exposure a n d risk assessm entfo r arsenic. Office of Water Regulations and Standards, Criteria and Standards Division, Washington, D.C. EPA-440/4-85-005. U.S. Environmental Protection Agency (U.S. EPA). 1985. Ambient water quality criteriafo r arsenic. Office of Water Regulations and Standards, Criteria and Standards Division, Washington, D.C. U.S. Environmental Protection Agency (U.S. EPA). 1985. Methodsfo r Measuring the Acute Toxicity o f Effluents to Freshwater and Marine Organisms. United States Environmental Protection Agency. EPAj'600/4-85/013. ' U.S. Environmental Protection Agency (U.S. EPA). 1989. Risk Assessment Guidance fo r Superfund. Volume I. Washington, D.C. EPA/540/1-89/002. U.S. Environmental Protection Agency (U.S. EPA). 1992. Ambient Water Quality Criteriafo r the Protection o fAquatic Life. Federal Register. Volume 57. No. 246. December 22. U.S. Environmental Protection Agency (U.S. EPA). 1993. Wildlife Exposure Factors Handbook Volume l o fII. United States Environmental Protection Agency, Office of Research and Development, Washington, D.C. EPA/600/R-93/187a. U.S. Environmental Protection Agency (U.S. EPA). 1994. Methodsfo r Measuring the Toxicity and Bioaccumulation o fSediment-Associated Contaminants with Freshwater Invertebrates. United States Environmental Protection Agency. EPA/600/R-94/024. U.S. Environmental Protection Agency (U.S. EPA). 1995. Revised Region III BTAG Benchmark Values. U.S. EPA Region III BTAG. Technical Support Section. Philadelphia, PA. Van Zinderen Bakker and J.F. Jaworski. 1980. Effects o f Vanadium in the Canadian Environment. Ottawa, Canada: National Research Council of Canada, Associate Committee Scientific Criteria for Environmental Quality. Venugopal, B. and T.D. Luckey. 1978. Metal Toxicity in Mammals: 2. Chemical Toxicity o f Metals andMetalloids. Plenum Press, New York, NY. Wiebel. F.J., J.C. Leutz. L.Diamond and H.V. Gelboin. 1971. "Aryl Hydrocarbon (Benzo(a)pyrene) Hydroxylase in Microsomes from Rat Tissues: Differential Inhibition and Stimulation by Benzoflavones and Organic Solvents." Arch. Biochem. Biophys., 144:78-86. Wixson, B.G. and B.E. Davis. 1993. "Lead in Soil." Lead in Soil Task Force, Science Reviews. Northwood. 132 pp. Woolson, E.A. 1975. Arsenical pesticides. ACS Ser7:l -176 (as cited in Eisler. 994). USEPA 6924 59 000224 USFW Calder, W.A., and E.J. Braun. 1983. "Scaling of osmotic regulation in mammals and birds." American Journal o f Physiology, 244: R601-R606. Castranova, V., L. Bowman, and J.R. Wright. 1984. "Toxicity of Metallic Ions in the Lung: Effects on Alveolar Macrophages and Alveolar Type II Cells." J. Toxicol. Environ. Health 13:845-856. Clark, D.R. Jr. 1979. "Lead Concentrations: Bats vs. Terrestrial Mammals Collected near a Major Highway." Environ. Sci. Tech., 13:338-341. Dean, C.E., B.M. Hargis and P.S. Hargis. 1991. "Effects of Zinc Toxicity on Thyroid Function and Histology in Broiler Chicks." Toxicol. Letters, 57:309-318. DeGraaf, R.M. and D.D. Rudis. 1983. New England Wildlife: Habitat, Natural History, and Distribution. Amherst, MA. The University of Massachusetts Press. DeMayo, A., M.C. Taylor and K.W. Taylor. 1982. "Effects of Copper on Humans, Laboratory and Farm Animals, Terrestrial Plants and Aquatic Life." CRC Critical Reviews in Environmental Control, 12(3):183-255. Dixon, R.L., R.J. Sherins, and I.P. Lee. 1979. "Assessment of Environmental Factors Affecting Male Fertility." Environmental Health Perspectives. 30:53-68. Edwards, C.A. and J.R. Lofty. 1977. The Biology o f Earthworms. John Wiley and Sons, New York, NY. Ehrlich, P.R., D.S. Dobkin, and D. Wheye. 1988. The Birders Handbook. Simon and Schuster, Fireside. New York. 785 PP- Eisler, R. 1986. "Chromium Hazards to Fish, Wildlife, and Invertebrates: a Synoptic Review." U.S. Fish and Wildlife Service Biological Report, 85(1.86). 60p. Eisler, R. 1988a. "Arsenic Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review." U.S. Fish and Wildlife Service Biological Report 85(1.12). Eisler, R. 1988b. "Lead Hazards to Fish, Wildlife, and Invertebrates: A Synoptic Review." United States Fish and Wildlife Biological Report. 85(1.14). Evan, A.P. and W.G. Dail. 1974. "The Effects of Sodium Chromate on the Proximal Tubules of the Rat Kidney." Lab. Invest., 30:704-715 In: Steven, J.D., L.J. Davies, E.K. Stanley, R.A. Abbott, M. Inhat, L. Bidstrup, and J.F. Jaworski. 1976. "Effects of Chromium in the Canadian Environment." Nat. Res. Counc. Can., NRCC No. 15017. 168p. Fleming, W.J. and C.A. Schuler. 1988. "Influence of the Method of Fluoride Administration on Toxicity and Fluoride Concentrations in Japanese Quail." Environmental Toxicology and Chemistr. 7(10):841-846. Fleming, W.J., C.E. Grue, C.A. Schuler, and C.M. Bunck. 1987. "Effects of Oral Doses of Fluoride on Nestling European Starlings." Arch. Env. Com. Toxicol. 16(4): 483-490. Gianutsos, G. and M.T. Murray. 1982. "Alterations in Brain Dopamine and GABA Following Inorganic or Organic Manganese Administration." Neurotoxicol. 3:75-81. Goldman, E.A. 1950. Raccoons o f North and Middle America. Washington, DC: U.S. Fish and Wildl. Service. 54 USEPA 6923 000225 USFW 0641 90 Mdsn Z Z O r 9