https://cpw.cvlcollections.org/files/original/19a3e76af26a00bce6715021e821ff7b.pdf 22ca6296b70667cea2d93febb21136d1 PDF Text Text Effects of the 2013 West Creek Petroleum Spill on Stream Ecosystem Structure and Function: Responses of Periphyton, Macroinvertebrates and Fish Sam B. Duggan, Dr. Paula Schaffer1, Pete Cadmus2, and Dr. William H. Clements Department of Fish, Wildlife and Conservation Biology Colorado State University Fort Collins, CO 80521 1 Department of Microbiology, Immunology and Pathology Veterinary Diagnostic Laboratory Colorado State University Fort Collins, CO 80521 2 Colorado Parks and Wildlife Aquatic Research Fort Collins, Colorado 80526 �Acknowledgements This research was funded by Colorado Parks and Wildlife (16-IAA-82320) and the Colorado State University Energy Institute. Colorado State University’s Intuitional Animal Care and Use Committee approved this research under protocols 15-6086A and 16-6367. We thank the numerous individuals and laboratories that participated – aquatic toxicologist Abbie Jefferson as well as aquatic biologists Lori Martin and Eric Gardunio of Colorado Parks and Wildlife; toxicologists and biologists Dr. Don Tillitt and Diane Nicks of the USGS Columbia Environmental Research Center; clinical pathologists Dr. Linda Vap and Lynne Shannon of Colorado State University’s Veterinary Teaching Hospital Diagnostic Laboratory; analytic chemist Dr. Terry Wade of Texas A&M University’s Geochemical and Environmental Research Group; analytic chemists from ALS Global; field or laboratory assistance from Colorado State University’s Aquatic Ecotoxicology Laboratory including Chris Kotalik, Brian Wolff, Britney Dabney, Hannah Riedl, Richard Salas, Thomas Shields, Connor Nikkola, Ben Kinne, Eric Tokuyama, Julianne Nikirk, Siena Schroeder, Jessica Ruhlman, Courtney Larson, Valerie Doebley, Ashley Hagen, Spencer Elliot; additional laboratory assistance from Jordan Anderson of Colorado State University; taxonomic assistance from Chris Kotalik. ii �List of Acronyms AR1 – Arkansas River Site 1 (Reference Community) AR5 – Arkansas River Site 5 (Tolerant Community) ATL – Aquatic Toxicology Laboratory at Colorado Parks and Wildlife BDL – Below Detection Limit BLM – Bureau of Land Management BQL – Below Quantification Limit BTEX – Benzene, Toluene, Ethylbenzene and Xylene CPW – Colorado Parks and Wildlife DRO – Diesel Range Organics GC-FID – Gas Chromatography Coupled with Flame Ionization Detection GC-MS – Gas Chromatography Coupled with Mass Spectrometry HPF – High-Power-Field MMI – Multimetric Index PAH – Polycyclic Aromatic Hydrocarbon PCA – Proximate Component Analysis PCB – Polychlorinated Biphenyl PPB – Parts-Per-Billion PPM – Parts-Per-Million PTFE - Polytetrafluoroethylene RBT – Rainbow Trout SEM – Standard Error of The Mean SMA – Spleenic Melanomacrophage Aggregate SRL – Stream Research Laboratory at Colorado State University U.S.EPA – United States Environmental Protection Agency VOC – Volatile Organic Compound WC1 – West Creek Site 1 (Upstream Reference Site) WC2 – West Creek Site 2 (Immediately Downstream of the 2013 Petroleum Spill) WC3 – West Creek Site 3 (Downstream from WC2) WC4 – West Creek Site 4 (Downstream from WC3) iii �Table of Contents Acknowledgements ......................................................................................................................... ii List of Acronyms ........................................................................................................................... iii Table of Contents ........................................................................................................................... iv List of Figures .............................................................................................................................. viii List of Tables ............................................................................................................................... viii Abstract .......................................................................................................................................... ix Report Structure .............................................................................................................................. x Section 1: Literature Review ....................................................................................................... 1 The Petroleum Hydrocarbon Mixture, its Chemistry and Dynamics ...................................... 1 Comparing Crude Oil, Gasoline and Diesel ............................................................................ 3 Petroleum Toxicity .................................................................................................................. 4 Section 2: The 2013 West Creek Petroleum Spill ....................................................................... 8 Initial West Creek Spill Remediation and Chemical Quantification ....................................... 9 Colorado Parks and Wildlife’s Fish Population Surveys and Management of West Creek.. 11 Bureau of Land Management’s Benthic Macroinvertebrate Surveys ................................... 13 Section 3: Field Observations, Bioassays and Mesocosm Experiments ................................... 14 Introduction ........................................................................................................................ 14 Petroleum Spill Loading Rates (Spill Size) ....................................................................... 14 Overall Goals and Objectives ............................................................................................ 16 2015 West Creek Biomonitoring ........................................................................................... 16 Goals and Objectives ......................................................................................................... 16 Methods.............................................................................................................................. 16 West Creek ..................................................................................................................... 16 Site Selection:................................................................................................................. 17 Routine Physicochemical Analysis ................................................................................ 18 Analytic Chemistry ........................................................................................................ 18 Benthic Macroinvertebrate Community Structure ......................................................... 19 Mottled Sculpin Collection and Processing ................................................................... 19 Differential Leukocyte Counts ....................................................................................... 20 Histology ........................................................................................................................ 20 iv �Statistical Analyses ........................................................................................................ 21 Results and Discussion ...................................................................................................... 22 Analytic Chemistry ........................................................................................................ 22 Aquatic Macroinvertebrate Communities ...................................................................... 24 Mottled Sculpin Blood Differentials .............................................................................. 27 Mottled Sculpin Histopathology .................................................................................... 28 Conclusions ........................................................................................................................ 33 Rainbow Trout Bioassay ....................................................................................................... 33 Goals and Objectives ......................................................................................................... 33 Methods.............................................................................................................................. 33 Rainbow trout, Experiment Set-up and Acclimation ..................................................... 33 Exposure Regime, Physical Chemistry and Analytic Samples ...................................... 34 Analytic Chemistry ........................................................................................................ 35 Fish Processing............................................................................................................... 35 EROD ............................................................................................................................. 36 Differential Leukocyte Counts ....................................................................................... 37 Histology ........................................................................................................................ 37 Statistical Analyses ........................................................................................................ 38 Results and Discussion ...................................................................................................... 38 Analytic Chemistry ........................................................................................................ 38 Behavioral Effects .......................................................................................................... 38 Total Ammonia .............................................................................................................. 40 Mortality ......................................................................................................................... 40 EROD Rate..................................................................................................................... 41 Histology ........................................................................................................................ 43 Hematology .................................................................................................................... 43 Conclusions ........................................................................................................................ 47 Periphyton Bioassay .............................................................................................................. 48 Goals and Objectives ......................................................................................................... 48 Methods.............................................................................................................................. 48 Exposure Scenario .......................................................................................................... 49 Biomass Measurements .................................................................................................. 49 v �Analytic Chemistry ........................................................................................................ 49 Statistical Analyses ........................................................................................................ 50 Results and Discussion ...................................................................................................... 50 Analytic Chemistry ........................................................................................................ 50 Periphyton Biomass........................................................................................................ 51 Conclusions ........................................................................................................................ 52 Aquatic Insect Mesocosm Experiments ................................................................................ 52 Goals and Objectives ......................................................................................................... 52 Methods.............................................................................................................................. 53 Insect Community Collection......................................................................................... 53 Mesocosm Setup ............................................................................................................ 54 Diesel Spill Exposure Scenario ...................................................................................... 54 Analytic Chemistry ........................................................................................................ 55 Statistical Analysis ......................................................................................................... 56 Results and Discussion ...................................................................................................... 56 Analytic Chemistry ........................................................................................................ 56 Aquatic Insect Drift ........................................................................................................ 58 Benthic Abundance ........................................................................................................ 62 Conclusions ........................................................................................................................ 65 Section 4: Final Remarks and Recommendations ..................................................................... 66 Inference of Causation ........................................................................................................... 66 Empirical Weight-of-Evidence Approach ......................................................................... 67 Bradford Hill’s Criteria for Causation ............................................................................... 69 Lessons Learned: Final Recommendations for Managers and Stakeholders ........................ 71 References ..................................................................................................................................... 77 vi �List of Figures 1) Petroleum Hydrocarbon Structures ......................................................................................1 2) CPW West Creek Fish Surveys .........................................................................................12 3) Sampling Locations on West Creek ...................................................................................17 4) West Creek Sediment PAH Concentrations ......................................................................23 5) West Creek Sediment Aliphatic Concentrations................................................................23 6) West Creek Aquatic Macroinvertebrate PCA ....................................................................24 7) West Creek Aquatic Macroinvertebrate Abundances .......................................................27 8) West Creek Mottled Sculpin Spleenic Melanomacrophages (Quantitative) .....................30 9) West Creek Mottled Sculpin Histological Panel ...............................................................32 10) Rainbow Trout Bioassay Mortality....................................................................................41 11) Rainbow Trout Bioassay Mortality Time-Course .............................................................41 12) Rainbow Trout Bioassay EROD Rates .............................................................................43 13) Rainbow Trout Bioassay White Blood Cell Differentials ................................................44 14) Rainbow Trout Bioassay Plasma Panel 1 ..........................................................................46 15) Rainbow Trout Bioassay Plasma Panel 2 ..........................................................................47 16) Periphyton Bioassay Biomass ............................................................................................52 17) Mesocosm Aquatic Insect Drift (Exp1) ............................................................................59 18) Mesocosm Aquatic Insect Proportion Drifting (Exp1) ......................................................60 19) Mesocosm Aquatic Insect Drift (Exp2) ............................................................................61 20) Mesocosm Aquatic Insect Proportion Drifting (Exp2) ......................................................62 21) Mesocosm Benthic Abundance (Exp1) .............................................................................64 22) Mesocosm Benthic Abundance (Exp2) .............................................................................65 23) Empirical Weight of Evidence Approach ..........................................................................67 vii �List of Tables 1) General Properties of Petroleum .........................................................................................4 2) BLM West Creek Aquatic Macroinvertebrate Data ..........................................................13 3) Hypothetical Diesel Loading Rates (Tanker Truck Spills) ................................................15 4) Hypothetical Diesel Loading Rates (Drum Spills) ............................................................15 5) West Creek White Blood Cell Differentials for Old Sculpin ............................................28 6) West Creek White Blood Cell Differentials for Young Sculpin........................................28 7) West Creek Mottled Sculpin Melanomacrophages (Qualitative) ......................................30 8) Rainbow Trout Bioassay Analytic Chemistry ..................................................................39 9) Periphyton Bioassay Analytic Chemistry ..........................................................................51 10) Aquatic Insect Mesocosm Experiments Analytic Chemistry (Exp1) ...............................57 11) Aquatic Insect Mesocosm Experiments Analytic Chemistry (Exp2) ................................57 viii �Abstract Oil development has expanded dramatically in Colorado over the last decade. Associated with this rapid expansion has been a significant increase in the number of accidental releases into the environment. On January 2013, West Creek which flows along a scenic byway in Unaweep Canyon, Colorado, was impacted by a petroleum spill from an overturned tanker truck. 22,700 liters of gasoline and 7,300 liters of diesel discharged into the stream killing an estimated 1,206 Brown Trout, Salmo trutta, and 8,172 Mottled Sculpin, Cottus bairdii. Subsequent electrofishing surveys indicated that the fishery was not quickly recovering particularly with regard to Mottled Sculpin populations, but also Brown Trout. In June and October 2015, as part of ongoing efforts to determine long term effects of this spill, we explored health indicators across multiple levels of biological organization. Histopathological abnormalities (e.g., ectopic neural tissue, cystic kidney, increased melanomacrophage aggregates) were observed in Mottled Sculpin collected from the spill site and nearby downstream sites. Altered benthic macroinvertebrate community structure was observed at the spill site compared with a reference site one kilometer upstream. Interestingly, a GC-MS finger-printing analysis of polycyclic aromatic hydrocarbons (PAHs) in stream sediment revealed that PAH concentrations were typical of minimally impacted streams flowing adjacent to roads. These results suggest that effects of the spill were persisting after contaminant concentrations had returned to ‘normal’ by Fall 2015. Subsequently, we conducted two mesocosm experiments, using naturally colonized benthic macroinvertebrate communities. Exposure to simulated spill conditions caused concentration-dependent macroinvertebrate drift and substantial mortality that occurred rapidly after the spills were initiated and at lower concentrations than expected. In addition, concentration-dependent lethal and sub-lethal effects were observed in Rainbow Trout, Oncorhynchus mykiss, during simulated spill bioassays. Periphyton biofilms were also adversely affected. We conclude that petroleum spills in coldwater streams risk adverse acute, chronic, lethal and sub-lethal effects to aquatic communities across numerous levels of biological organization. And these effects were evident after the 2013 West Creek petroleum spill. Moreover, by utilizing field observations, mesocosms and bioassays we gained insights into consequences of petroleum spills using an ecotoxicological weight-ofevidence approach. Importantly, the methods used in this project can be employed at future spill events as field useful bioassessment techniques to aid in the process of holding responsible parties appropriately accountable for damages to stream communities. ix �Report Structure The intention of this document is to assemble facts surrounding the 2013 West Creek petroleum spill and appraise the consequences of that spill using a collaborative, interdisciplinary weight-of-evidence approach. This approach was utilized because causal inferences of chronic, latent and indirect effects associated with pollution events are notoriously challenging to demonstrate. Our goal was to use the West Creek spill investigation as a platform to inform and improve bioassessment as well as forecasting the potential consequences of petroleum spills in coldwater streams of Colorado. In the first section, our objective is to briefly review the science of inland petroleum spills as well as their chemistry and toxicity relevant to Colorado Parks and Wildlife (CPW) management and contingency planning. In section 2, we focus on the 2013 West Creek Spill. We discuss the spill itself and the resulting fish-kill. We describe the remediation efforts of Groendyke contractors as well as the initial chemical analysis conducted by contractors hired by the U.S.EPA. Then we discuss historic aquatic insect sampling conducted by the Bureau of Land Management (BLM) which conveniently bracketed the date of the spill. We also briefly discuss CPW electrofishing surveys conducted to monitor potential recovery of the trout fishery. In section 3, we report our observations of suspected chronic impacts, in fish and aquatic macroinvertebrates, stemming from the 2013 West Creek spill. We discuss in detail our methodology, results and implications regarding aquatic macroinvertebrate community structure, sculpin hematology and histopathology as well as an analysis of persistent petroleum hydrocarbon compounds in water and sediment. Additionally, we detail the laboratory portion of this project in which we simulated a series of diesel spills under controlled conditions. We describe how simulated spill bioassays conducted on naturally colonized periphyton biofilms affected biomass. We report on a simulated spill bioassay conducted on Rainbow Trout where mortality, hematology and biochemical endpoints were explored. Then, we discuss two mesocosm experiments that subjected naturally colonized communities of insects to simulated spill conditions. We conclude this report in section 4, by discussing inference of causality and by providing recommendations for improving preparedness, response and biomonitoring methods that may help strengthen future spill related bioassessments in Colorado streams. x �Section 1: Literature Review The Petroleum Hydrocarbon Mixture, its Chemistry and Dynamics: Petroleum in its raw form, commonly referred to as crude oil, as well as its refined products such as gasoline and diesel are complex mixtures composed of hundreds to tens-ofthousands of discrete compounds (Gough and Rowland 1990; Hoffman et al. 2002). In general, petroleum is composed of carbon and hydrogen chains of various configurations and sizes (Figure 1). The dominant hydrocarbons classes include aliphatic compounds (i.e., straight-chain alkanes, branched alkanes and cycloalkanes) and cyclic aromatics containing conjugated double bonded carbon atoms. Additionally, nonhydrocarbon compounds are present including those with oxygen, sulfur, nitrogen or metals bonded to a carbon and hydrogen skeleton. Each discrete compound behaves physically and chemically different, although differences are less apparent among similarly sized compounds within similar classes. Additionally, synthetic and natural additives are commonly added to petroleum products as stability or performance enhancers and as lubricants (Bleyl 1990). For the sake of simplicity, this document will refer to the two most common classes of hydrocarbons: aliphatic and aromatic hydrocarbons, which include polycyclic aromatic hydrocarbons (PAHs). Nonhydrocarbons and petroleum additives will not be discussed in detail. However, their existence and potential relevance to petroleum spill risk assessment should be recognized. Figure 1: Examples of common petroleum hydrocarbon structures (Saadoun 2015). �The composition of petroleum is dynamic and changes rapidly over time, particularly in the event of an uncontained spill. Low-molecular weight compounds (e.g., BTEXs and naphthalene), which are predominant in gasoline, but also present in diesel and crude oil, quickly volatilize into the atmosphere (Posthuma 1977; Owens et al. 1993). Typically, 10%-75% of compounds volatilize from crude oil and gasoline, respectively (Hoffman et al. 2002). The speed of volatilization varies based on water temperature, aeration and turbulence, thickness of the petroleum surface film and the present of other aqueous solutes (Hoffman et al. 2002; API 2016). As a result, the composition of spilled petroleum shifts towards less volatile and more persistent high-molecular weight compounds over time. The composition is also altered when aromatic compounds are photoxidized by UV-light in the presence of oxygen (Hoffman et al. 2002; Lee 2013). Because photoxidation is proportional to UV-light intensity this process is influenced by season, ozone thickness, cloud cover, shade, aspect, water depth, turbidity and thickness of the petroleum surface film (Lee 2013; Morris et al. 2015). Additionally, petroleum compounds are commonly oxidized and biotransformed through metabolism by microorganisms, fungi, animals and plants (Whyte et al. 2000). Most petroleum related compounds, particularly those with high-molecular weights, are extremely hydrophobic, thus they do not readily solubilize in water. However, they may disperse throughout the water column as fine droplets. Rather than dissolving, petroleum compounds generally float or adsorb to surfaces (e.g., sediments and organic matter) (Hoffman et al. 2002). It should be noted that some crude oils and heavily weathered petroleum products will sink over time as low-molecular weight compounds volatilize and the specific gravity of the mixture increases. Low-molecular weight compounds that tend to volatize are also relatively soluble in water, especially those with non-hydrocarbon atoms and double bonds. Therefore, low-molecular weight compounds can dissolve, incorporate and readily distribute throughout the water column (API 2016). In general, smaller, more oxidized compounds containing double bonds and/or nonhydrocarbon petroleum compounds are more water soluble than large single bonded hydrocarbons. Importantly, as the composition of petroleum hydrocarbons changes by physical, chemical and biological processes, their toxic effects and properties also change. This dynamic process should be accounted for during analytic chemical analysis to appropriately quantify the 2 �presence of petroleum compounds in water and sediment after a spill. Dynamics of the chemical composition, physical state (i.e., aqueous or gaseous), extent of dispersion in the water column, absorption by organisms and adsorption in sediments or other surfaces influence hazard and exposure to aquatic life. Comparing Crude Oil, Gasoline and Diesel: Regarding crude oil and its refined products, this document refers to their characteristics and effects in generalities only. Specific characteristics are variable and influenced by their source, extraction method, subsequent transport, processing, and blending (Table 1). In addition, this document does not discuss all refined products, but focuses on diesel, gasoline and crude oil because they are most commonly transported and spilled into inland waters. Characteristics of other commonly transported petroleum products (e.g., kerosene and jet fuel) have properties intermediate to gasoline and diesel, although there are exceptions to this principle. Moreover, chemical composition of gasoline and diesel varies based on the producer, anticipated applications and season. For example, diesel comes in summer and winter blends and gasoline varies in its octane rating. Refined products are also commonly co-transported and spilled simultaneously during an accident, as was the case during the 2013 gasoline and diesel spill in West Creek. Therefore, it should be noted that the characteristics and effects of spilled petroleum do not come in a “one-size-fits-all” package. Refined petroleum products are typically distilled from crude oil, although they can be produced synthetically. Gasoline contains the lightest and most volatile fraction of crude oil. As a result, gasoline compounds are small, readily evaporate into the atmosphere and are extremely flammable due to high vapor pressure and low flashpoint. Although all petroleum is technically considered insoluble in water, gasoline is more soluble than diesel or crude oil (Parker et al. 1976). Diesel contains some of the larger molecules that exist in gasoline, but it also contains compounds much heavier than those in the gasoline range. The fractions of diesel compounds that overlap with gasoline share certain characteristics. However, the heavier compounds, which make up the majority of diesel by volume, are less volatile and more environmentally persistent than those found in gasoline (Song 2000). Thus, diesel is less flammable, more viscous, has a 3 �greater capacity to smother surfaces and potential for causing chronic effects on organisms than gasoline (Owens et al. 2003). Crude oil contains compounds within the range of both gasoline and diesel, but it also contains smaller trapped gases and much larger wax-like compounds. Therefore, the characteristics of crude oil are very diverse. However, as a whole, crude is less volatile and flammable, more viscous, more environmentally persistent and has a greater smothering capacity than diesel or gasoline. Additionally, as opposed to gasoline and diesel, some crude oils will sink in water. Table 1: General properties and comparisons of characteristics among gasoline, diesel and crude oil ( 1Hyne and Norman 2012; 2Bleyl 1990; 3Owens et al 1993; 4U.S DOE, 2014; 5API 2016; 6Curl and O'Donnell, 7 Hoffman et al. 2002). Characteristic Average Carbon Chain Lengths Flash Point Specific Gravity Relative Volatility Relative Viscosity Relative Acute Toxicity Relative Physical Smothering Potential Relative Environmental Persistence Relative Solubility in Water % Aromatic Compounds (by Volume) Forms Stable Emulsions with Water Mass Lost to Evaporation Gasoline C5 – C10 6 100°F - 40°F 3,4 0.65 – 0.8 3,4 High 2,3 Low 2,3 High 2,3 Low 3 Low 2,3 High 2 10% - 50% 2 Typically No 3 75%-100% 5, 7 Diesel Fuel C12 - C20 6 100°F - 165°F 3,4 0.81 – 0.87 3,4 Medium 2,3 Medium 2,3 Medium 2,3 Medium 3 Medium 2,3 Medium 2 5% - 35% 2 Yes 3 ~25% 5 Crude Oil < C4 - > C40 6 20°F - > 150°F 3,4 0.85 – > 1.00 3,4,5 Low – High 2 High 2,3 Low 2,3 High 3 High 2,3 Low 2 3% - 30% 1 Yes 3 Extremely Variable 5 Petroleum Toxicity: Potential environmental consequences of a petroleum spill were first reported by Howard (1892) who demonstrated that a ‘thin’ layer of kerosene was sufficient to kill all insects in a small pond. Today, there are numerous examples of petroleum’s adverse effects to aquatic life (Hoffman et al. 2002). However, much of that research focuses on marine ecosystems impacted by large, high profile spills such as the Exxon Valdez and the Deepwater Horizon incidents. In contrast, relatively little work has been conducted on freshwater ecosystems affected by smaller, low profile spills. Below, we briefly review available research on the adverse impacts of petroleum spills to aquatic life with special attention to freshwater organisms and relevant marine science which likely translate to freshwater ecosystems. 4 �In general, petroleum compounds with low molecular weight and relatively high water solubility are considered the most toxic compounds within the petroleum mixture because of their high bioavailability (Bruce et al. 1976). Refined products are thought to be more acutely toxic than crude oil; gasoline more acutely toxic than diesel (Bury 1972; Anderson 1977; Connell et al. 1981). There also is evidence that weathered petroleum, especially crude oil, can be more toxic than fresh petroleum (Morris et al. 2015). Moreover, toxicity is increased by any mechanism that disperses petroleum throughout the water column such as wave action, wind, rapids, chemical dispersants or physical agitation (Southward and Southward 1978; Wu et al. 2012). Petroleum spills harm aquatic life through several distinct actions. Petroleum physically acts on an aquatic organism by smothering gills leading to asphyxiation, or coating fur and feathers diminishing insulating properties and causing hypothermia (Engelhardt 1983; Waldichuk 1990). Physical smothering can also fill the interstitial space in gravel or sediment, leading to habitat loss for benthic organisms and fish (Connell et al. 1981; Crunkilton et al. 1990). Petroleum can act through chemical action by disrupting cellular membranes and influencing biochemical reactions (Neff 1985). Additionally, petroleum can elicit indirect effects by altering food availability, predator-prey relationships, community dynamics and ecosystem services (Peterson et al. 2003). Petroleum can directly kill or chronically poison organisms so that mortality is delayed or fecundity, fitness and recruitment are affected. Research suggests that aliphatic compounds are most responsible for physical effects, whereas aromatics compounds induce chemical effects. Low-molecular weight compounds are most responsible for acute mortality and larger compounds for chronic toxicity (Hoffman et al. 2002). However, some large 3-5 ring PAHs are also extremely acutely toxicity (Adams et al. 2014). Chemical effects are thought to be primarily mediated by interactions within the liver in vertebrates, as monocyclic and polycyclic aromatic hydrocarbons (PAHs) are biotransformed by metabolizing enzymes of the cytochrome P450 family into toxic or carcinogenetic intermediates (Whyte et al. 2000). Alternatively, similar toxic intermediates are created by other oxidation pathways such as photoxidation by UV-light. The ways in which PAHs chemically harm organisms are diverse and often non-specific. They act by disrupting cellular membrane function, binding to proteins altering their structure and function, and binding to DNA, thereby disrupting 5 �gene regulation, repair and replication (Neff 1985; Hoffman et al. 2002). This manifests as a breadth of adverse effects. For example, PAH toxicity leads to cancers, genetic mutations, developmental defects and biochemical disruptions in exposed organisms (Santodonato et al. 1981; Varanasi et al 1989; Eisler 2000). However, these effects are not specific to PAHs as other unrelated chemical exposures, genetic predisposition or even the ageing process itself can cause similar effects. Therefore, establishing a causal link between PAH exposure from a petroleum spill to sub-lethal, latent or chronic effects in aquatic organism is challenging and requires an integrated, weight of evidence approach. Reduced densities and biodiversity in aquatic macroinvertebrates has been reported after petroleum spills (Harrel 1985; Crunkilton et al 1990; Lytle et al. 2001; Smith et al. 2010). However, the magnitude of effects vary widely and reduced biodiversity as measured by species richness does not always occur after a spill. Species replacement of sensitive taxa by tolerant taxa has been reported (Harrel 1985; Lytle et al. 2001). Following a petroleum spill recovery of invertebrates occurs on the scale of weeks to a decade (Hoffman et al. 2002). Sublethal effects have also been observed. An increased rate of aquatic insect drift, a general escape response to perturbations, has been reported after a petroleum spill (Bury 1972). Additionally, the chemosensory food homing system in crawfish was disrupted after exposure to petroleum (Jurcak et al. 2015). Organisms, such as mollusks, that lack effective cytochrome p450 enzymes, will bioaccumulate petroleum compounds and transfer these compounds to predators even when primary sources of exposure have been eliminated (Peterson et al. 2003). In marine systems observations of acute fish kills following a petroleum spill are uncommon because fish can easily migrate away from areas where petroleum concentrations are highest, and also because of high dilution potential. In contrast, acute fish kills in freshwater streams are much more common for the opposite reasons (Hoffman et al. 2002). In the laboratory, mortality due petroleum toxicity has been demonstrated in water concentrations as low as 1 part-per-million (ppm) after long term exposures (Woodward et al. 1983). In acute exposures mortality generally occurs when concentrations exceed 100 ppm (Barnett and Toews 1978; Hedtke et al 1982; Anderson et al. 1987; Little et al 2000). However, sublethal effects have been reported at concentrations less than 0.5 ppm. Reported sublethal effects include altered respiration and heart rate, narcosis, diminished growth rates, gill and liver damage, 6 �immunosuppression, reproductive impairment, biochemical and hematological changes and altered behavior (Chambers et al. 1979a; Chambers et al. 1979b; Thomas et al. 1980; Malins et al. 1981; Moles and Norcross 1998; Carls et al. 1998; Khan 1999; Ali et al. 2014). Recently there has been considerable attention paid to cardiotoxicity and genotoxicity of spilled petroleum. Although, this research was conducted on marine fish, it has important implications for freshwater species. Concentrations of petroleum as low as 1 part-per-billion (ppb) crude oil was sufficient to cause developmental heart defects in fish embryos (Incardona et al. 2014). Other experiments demonstrated increased prevalence of heart defects in fish embryos that were exposed to 1 ppb petroleum for 48 hours, and swim performance of fish were compromised compared with control fish (Mager et al. 2014; Beyer et al. 2016). Heart cells exposed to petroleum displayed disrupted potassium and calcium ion regulation, which manifested as heart arrhythmias and poor swim performance (Brette et al. 2014). Downregulation of genes associated with heart muscle fibers and contractility were also detected (Edmunds et al. 2015). Exposure to low concentrations of petroleum altered regulations of genes involved with control of the cell cycle, transcription, apoptosis, DNA repair and the immune system (Pilcher et al. 2014). Beyer et al. (2016) concluded that “gene expression in fish was a robust indicator of oil exposure.” Petroleum spills have been implicated as a factor leading to increased incidents of cancers and birth defects in fish on numerous occasions (Roberts et al 1989; Hoffman et al. 2002; Peterson et al. 2003). These observations are, at least partly, caused by the fact that PAHs in petroleum bond with DNA and proteins in forming adducts that disrupt molecular and biochemical processes such as DNA repair and replication (Aas et al. 2000). Long-term fish population level effects are often observed following petroleum contamination in marine and in freshwater systems (Mankii and Vauras 1974; Teal and Howarth 1984; Squire 1992; Kuehn et al 1995; Hoffman et al. 2002; Peterson et al. 2003). Petroleum exposure can also depreciate the value of a fishery without reducing fish densities by tainting the flavor of fish meat. Typically, fouling occurs at petroleum concentrations of 0.45 ppm to 300 ppm, depending on the exposure durations and petroleum type (Connell, 1981; Heras et al. 1992; Davis et al. 1992). However, one study reported that exposure 7 �to 10 ppb petroleum for 24 hours caused tainted fish flesh (Nitta, 1972). Fouled flesh flavor can linger for months after exposures have ended (Höfer 1998). Although fish and insects are often the focus of biomonitoring studies after petroleum spills, other animals can be harmed. Bull frogs have been reported dead after petroleum exposure and surviving frogs had increased incidences of fatty liver syndrome, abnormally inflated lungs and behavioral changes (McGrath and Alexander 1979). Green tree frog tadpoles exposed to motor oil were prevented from metamorphosing into adult frogs (Mahaney 1994; Lefcort et al. 1997). Bury (1972) reported large numbers of dead tadpoles and garter snakes after a gasoline spill in a creek. Birds commonly die in marine oil spills (Hoffman et al. 2002) and mortality has also been observed after spills in steams (Bury, 1972). Additionally, aquatic reliant mammals such as beaver, muskrat and otter have died after petroleum spills (Albers and Gay 1982; Monson et al. 2000). Petroleum spills can also injure aquatic vegetation and alter community assemblages of phytoplankton and periphyton (Snow and Scott 1975; Burk 1977; Bott and Rogenmuser 1978; Federle 1979; Baca and Getter 1985). Microbial community changes, even proliferation of microbes, have been reported because petroleum can be toxic to some microbes and a food source to others (Heitkamp and Johnson 1984; Braddock et al. 1995). Section 2: The 2013 West Creek Petroleum Spill On the morning of January 25th 2013 a tanker truck, owned by Groendyke Inc., overturned and ruptured into West Creek adjacent to Highway 141 (mile marker 120.2) near Gateway, Colorado. As a result, approximately 6,000 gallons of gasoline and 2,000 gallons of diesel fuel were discharged directly into the creek. Subsequently, released petroleum caught fire and burned nearly one mile of downstream riparian habitat. Responding emergency crews and the local fire department allowed the fire to burn off naturally without use of flame retardants or chemical petroleum dispersants. To assess acute adverse ecological effects resulting from the petroleum spill responding CPW staff followed Administrative Directive W-5 recommendations for determining injury to aquatic life (State of Colorado 2007). Although winter weather conditions, ice on the creek and banks, fire and dangerous volatilizing petroleum compounds complicated and challenged 8 �assessments, CPW staff determined the extent and magnitude of the resulting acute fish-kill. Between January 25th and February 15th 2013, CPW made visual observations, collected dead organisms and conducted electrofishing surveys at six West Creek locations on several occasions. CPW determined that the fish-kill zone extended 2.6 miles downstream from the spill; no live fish were collected or observed within that zone. Additionally, numerous dead crawfish (Astacidae) and cranefly larvae (Tipulidae) were discovered. Petroleum sheen and odor, and dead Brown Trout (Salmo trutta) and Mottled Sculpin (Cottus bairdii) were observed as late as February 15th 2013. CPW estimated that 1,206 Brown Trout, 16 Rainbow Trout (Oncorhynchus mykiss), and 8,172 Mottled Sculpin were killed by the petroleum spill. Initial West Creek Spill Remediation and Chemical Quantification: Groendyke Inc., the responsible party, hired Environmental Management Inc. to conduct and manage spill cleanup activities. The United States Environmental Protection Agency (U.S.EPA) contracted Weston Solutions Inc., Superfund Technical Assessment and Response Team, to document and determine the “efficacy and completeness” of cleanup activities immediately following the spill (Weston Solutions Inc., 2013). Active remediation occurred during the subsequent three-week period starting at the spill site and extending for approximately one mile downstream. This included removal of tanker truck debris as well as petroleum saturated plant material. Floating petroleum was also physically removed by sorbent booms (additional booms were deployed farther downstream by CPW staff) and pumping by vacuum truck. Supplemental aeration and power washing enhanced volatilization and flushing of petroleum from sediment and rock surfaces. Additionally, sticks and poles were used to agitate rocks and sediment and to break up ice, further facilitating volatilization and flushing. “Remediation effectiveness and completeness” were determined by chemical sampling and quantification of highly toxic, low-molecular weight benzene, ethylbenzene, toluene, and xylene (BTEX) and gasoline-range volatile organic compounds (VOCs) by gas chromatography coupled with mass spectrometry by standard U.S.EPA Method 8260B (Weston Solutions Inc. 2013). It should be noted that persistent, semi-volatile compounds associated with diesel fuel were apparently not quantified during remediation efforts. Water and sediment samples were 9 �collected on numerous occasions between January 26th and February 11th, 2013 from eleven locations bracketing the spill site and West Creek’s confluence with the Dolores River. On January 26th, at the spill site, water BTEX concentrations were quantified at low concentrations of 2 ppb benzene, 24 ppb toluene, 8 ppb ethylbenzene, and 49 ppb xylenes (Weston Solutions Inc. 2013). This suggested that within one day of the spill, the majority of BTEX compounds had degraded, transformed, adsorbed to surfaces, volatized or were transported downstream. At the Bureau of Land Management picnic area and at sites farther downstream, BTEX water concentrations were barely detectable. By January 27th concentrations at the spill site were further reduced, and concentrations at the picnic area and downstream had risen to quantifiable but low ppb ranges. Over the next 3 days water BTEX concentrations at all sites diminished rapidly. By early February 2013 water BTEX and VOC concentrations were no longer detectable (Weston Solutions Inc. 2013). However, the fact that a distinct petroleum odor and a detectable sheen were observed at the spill site on February 15th 2013 by CPW staff calls into question the usefulness of quantifying only volatile gasoline-range compounds. In general, sediment samples had higher concentrations of BTEX compounds than in water. At the spill site on January 26th 2013, sediment contained 184 ppb ethylbenzene and 929 ppb xylenes. Approximately 700 feet downstream, at the highway 141 bridge-crossing, sediment contained 410 ppb toluene, 752 ppb ethylbenzene, and 4,330 ppb xylenes (Weston Solutions Inc. 2013). No BTEX compounds were detected in sediment samples at locations downstream of the highway 141 bridge crossing. Results from the subsequent sediment sampling event on February 11th 2013 failed to detect BTEX compounds at any location (Weston Solutions Inc. 2013). Approximately six months after the spill occurred, water and sediment samples were collected on July 10th, 2013. Although, BTEX and VOCs were detected at some sampling locations, concentrations were in the low ppb range, rarely above detection limits and sporadically located (Weston Solutions Inc. 2013). Such low, intermittent concentrations indicated that these detections of BTEX and VOC compounds were not attributable to the West Creek petroleum spill. However, persistent diesel-range compounds may have been present, but were not apparently quantified. 10 �Colorado Parks and Wildlife’s Fish Population Surveys and Management of West Creek: Historically, West Creek was a popular angling destination that provided fishing opportunities for wild Brown Trout and occasional wild Rainbow Trout. This was particularly true at the spill site and downstream due to the presence of public land and ease of access. Because of its popularity and to maintain the sustainability of this rarely stocked fishery, special regulations were in place - bag and possession limits of two trout rather than the statewide limit of four trout. West Creek also contained self-sustaining populations of native Speckled Dace (Rhinichthys osculus) and Mottled Sculpin. Following the 2013 petroleum spill, CPW began a biannual two to three-pass removal electrofishing survey program to monitor potential recovery of the fishery. The initial survey in June 2013 documented that Brown Trout and Mottled Sculpin populations were markedly reduced immediately downstream of the spill when compared to an upstream reference, as well as sites further downstream (Figure 2). In general, the following six surveys at sites downstream of the spill site documented continuing reductions in fish populations and biomass. Interestingly, at the upstream reference site the Mottled Sculpin population increased over time as the Brown Trout population diminished (CPW 2016). It is unclear why Brown Trout populations diminished. However, reduced Brown Trout predation on Mottled Sculpin at the reference site is a possible explanation for Mottled Sculpin population growth. CPW electrofishing surveys clearly indicated that West Creek Brown Trout fishery as well as populations of native Mottled Sculpin had not recovered by the most recent survey date of October 2016. Due to the absence of natural recovery at sites downstream of the 2013 petroleum spill, CPW began a Brown Trout stocking program in the summer of 2016 on public lands. Five thousand 2-3 inch adipose fin clipped Brown Trout were stocked at sites downstream of the spill (CPW 2016). Importantly, these fish were not included in the subsequent October 2016 Brown Trout population estimates. Additionally, 700 catchable adipose fin clipped Rainbow Trout were stocked by a private landowner at the upstream reference site in the summer of 2016 (CPW 2016). 11 �Figure 2: Figures on fish population surveys gathered from the 2016 West Creek fishery report prepared by Eric Gardunio (CPW 2016). 12 �Bureau of Land Management’s Benthic Macroinvertebrate Surveys: The Bureau of Land Management (BLM) collected aquatic macroinvertebrate samples at sites relevant to this investigation before and after the 2013 Spill. The most pertinent results are briefly discussed here. Although data on aquatic insects in West Creek from 1997-2000 are also available from Trout Unlimited, these data are of limited value to this investigation because it is unclear which sampling and sorting methods were utilized. Additionally, data on non-insect macroinvertebrates were not reported and data on aquatic insect abundance were reported in proportions, rather than densities. Using both quantitative (Surber sampler) and qualitative (kick net) approaches, the BLM collected aquatic macroinvertebrate samples in 2009, 2010 and 2011, and in 2013 approximately six months after the petroleum spill. Samples were collected at three sites downstream of the spill location, but were not collected upstream of the 2013 spill. Notably, highly tolerant Annelida (i.e., Oligochaeta) and Turbellaria (i.e., Planariidea) were detected after the spill but not before (BLM 2011; BLM 2013). Conversely, highly sensitive stoneflies (order, Plecoptera) were present at the spill site in 2011 before the spill, but were not detected in 2013 after the spill (BLM 2011; BLM 2013). A multimetric index (MMI) was also calculated to quantify the overall quality and health of the spill site. The MMI was determined to be 63.4 in 2011 and 48.5 in 2013, indicating a reduction in quality and health of the aquatic macroinvertebrate community at the spill site bracketing the date of the spill (BLM 2013) Table 2: Densities of Annelda, Tubellaria and Plecoptera as well as multimetric index scores of the aquatic macroinvertebrate community (BLM 2011; BLM 2013). Samples were collected before and after the January 2013 West Creek petroleum spill at a site immediately downstream of the spill. Date August 2011 August 2013 Annelida 2 0 per m 405 per m2 Turbellaria Plecoptera 2 0 per m 228 per m2 130 per m 0 per m2 2 MMI 63.4 48.5 13 �Section 3: Field Observations, Bioassays and Mesocosm Experiments Introduction Associated with the rapid expansion of oil and natural gas development in the U.S. has been a significant increase in the number of accidental releases and spills of petroleum into the environment. Petroleum hydrocarbons, especially mono and polycyclic aromatic hydrocarbons are highly toxic, carcinogenic and mutagenic (Hoffman et al 2002; Beyer et al 2016). The primary anthropogenic sources include urban runoff, industrial discharges and accidental spills. Because of several high-profile spills (e.g., Exxon Valdez, Deepwater Horizon), much is known about the impacts of petroleum hydrocarbons on marine ecosystems. In contrast, our understanding of the effects of petroleum releases in streams, particularly in the western U.S., is quite limited. Ironically, petroleum spills are among the leading causes of fish kills in North America, and after a spill, residual oil can continue to impact ecosystems because the most toxic fractions are often the most persistent (Green and Trett 1989; Hoffman et al. 2002). Quantifying effects of petroleum spills in aquatic ecosystems is challenging, in part because of the diverse range of constituents (e.g., aliphatics, aromatics) that comprise the petroleum hydrocarbon mixture. Additionally, the persistence and toxicity of these compounds are highly variable. Given the challenges associated with planning for and responding to petroleum spills as well as the increased likelihood of spills occurring due to increased production, research on their effects in cold-water streams is necessary. Petroleum Spill Loading Rates (Spill Size) To provide context on diesel concentrations used in the following laboratory and mesocosm experiments, below are two tables describing loading rates for hypothetical diesel spills (Tables 3 and 4). It is important to note that concentrations used in this experiment were relevant to realistic spill scenarios. It is also important to realize that spill size (loading rate) is highly dependent on the speed at which petroleum is discharged into the waterway and the diluting capacity of the receiving water. Even a spilled 55-gallon drum can lead to toxic petroleum concentrations in many of Colorado’s lotic waterways. 14 �Table 3: Approximate diesel loading rates (mg/L) after hypothetical tanker spills (8,500 gallons) across various stream discharges and durations of actively spilling diesel. Bold numbers indicate concentrations that are bracketed by treatments used in this report’s mesocosm and rainbow trout bioassays (75 mg/L – 1,200 mg/L). Red numbers are concentrations higher than those used in this report (> 1,200 mg/L) and blue numbers are lower (< 75 mg/L). All values should be considered approximations to provide context and would change based on the distance from the point source, mixing, the specific gravity of the spilled petroleum product, temperature, volatilization, adsorption and absorption as well as other factors. Here, diesel was given a specific gravity of (0.83g/mL). Stream Discharge (CFS) 1 Minute Tanker Spill (mg/L) 10 Minute Tanker Spill (mg/L) 1 Hour Tanker Spill (mg/L) 6 Hour Tanker Spill (mg/L) 2 887,000 441,000 116,000 21,400 10 612,000 136,000 25,600 4,360 200 73,000 7,820 1,310 219 1,000 15,500 1,570 263 43.8 20,000 787 78.8 13.1 2.19 Table 4: Approximate diesel loading rates (mg/L) after hypothetical 55-gallon drum spills across various stream discharges and durations of actively spilling diesel. Bold numbers indicate concentrations that are bracketed by treatments used in this report’s mesocosm and rainbow trout bioassays (75 mg/L – 1,200 mg/L). Red numbers are concentrations higher than those used in this report (> 1,200 mg/L) and blue numbers are lower (< 75 mg/L). All values should be considered approximations to provide context and would change based on the distance from the point source, mixing, specific gravity of the spilled petroleum product, temperature, volatilization, adsorption and absorption as well as other factors. Here, diesel was given a specific gravity of (0.83g/mL). Stream Discharge (CFS) 2 10 200 1,000 20,000 1 Minute Drum Spill (mg/L) 48,500 10,100 510 102 5.10 10 Minute Drum Spill (mg/L) 5,070 1,020 50.9 10.2 0.509 1 Hour Drum Spill (mg/L) 848 170 8.49 1.70 0.085 6 Hour Drum Spill (mg/L) 141 28.3 1.41 0.283 0.014 15 �Overall Goals and Objectives The primary objective of this research is to identify and evaluate a suite of biochemical, physiological and ecological endpoints that can be used to quantify effects of petroleum hydrocarbons on stream ecosystems. Below we report results of field biomonitoring studies conducted at West Creek, laboratory toxicity tests conducted with fish and periphyton, and stream mesocosm experiments conducted with aquatic macroinvertebrate communities. 2015 West Creek Biomonitoring Goals and Objectives Our specific objectives were to assess ecological characteristics of West Creek two to three years after the 2013 petroleum spill by: 1) examining water and sediment chemistry in West Creek to determine if persistent petroleum hydrocarbons were present or if chemical recovery had occurred; 2) determine whether aquatic macroinvertebrate community structure differed among sites upstream and downstream of the spill location and to make inferences on ecosystem impairment or recovery; 3) examine Mottled Sculpin (Cottus bairdii) health characteristics; and 4) explore the utility of bioassessment techniques that can be used to quantify biological effects of petroleum hydrocarbons on coldwater streams. Our goal was to use the experience and knowledge gained at West Creek to inform future studies and to develop a weight of evidence approach to quantify impacts on long-term stream health. Methods West Creek: West Creek is a cobble bottom, 3rd to 4th order cold-water tributary to the Dolores River in Western Colorado. West Creek is moderately-hard (170 mg/L CaCO3, +/- 2.8) and alkaline (174 mg/L CaCO3, +/- 2.8), with a pH of 8.3 (+/- 0.3) and relatively high specific conductivity of 321µS (+/- 9.9). Mean stream widths in sampled locations ranged from 3.8 – 7.7 meters. Historically, the creek supported a self-sustaining fishery of naturalized Rainbow Trout, Oncorhynchus mykiss, and Brown Trout, Salmo trutta, as well as non-game native species 16 �including Speckled Dace, Rhinichthys osculus, Fathead Minnow, Pimephales promelas, and Mottled Sculpin, Cottus bairdii. West Creek runs directly adjacent to Highway 141, a scenic byway in Unaweep Canyon near Gateway, Colorado. Water quality is influenced by ground water, run-off from the Uncompahgre Plateau, run-off from the Highway 141, ranching, agriculture and off-stream gravel mining. Additionally, a crude oil spill occurred in 2011 on land adjacent to the creek, but was not thought to have affected water quality or biota. Site Selection: Sampling locations (n=4) were selected to bracket the site of the 2013 petroleum spill as well as closely correspond with historic sampling locations (discussed in Section 2). WC1 was a reference site located approximately 1 km upstream from the spill. WC2 was immediately downstream of the spill point source. WC3 was approximately 2 km further downstream and within the documented fish-kill reach. WC4 was approximately 6.5 km downstream from WC1 and was below the documented fish-kill reach. Each site was sampled in early summer (June 24, 2015) and fall (October 19, 2015). Figure 3: Sampling locations on West Creek. West Creek runs directly adjacent to the yellow line representing highway 141. Information contained within parentheses refers to Colorado Parks and Wildlife site identifiers. 17 �Routine Physicochemical Analysis: During each sampling event, routine physicochemical characteristics were measured. Water temperature and dissolved oxygen were determined by YSI Pro ODO meter; pH and specific conductance by YSI 63 meter (YSI Incorporated, Yellow Springs, OH). Flow rate and depth were measured at aquatic macroinvertebrate sampling locations by Flow-Mate Model 2000 (Marsh-McBirney Inc., Frederick, Md). Water samples (0.5 L) were collected, placed on ice and returned to the laboratory for determination of water hardness and alkalinity using standard titration procedures. Analytic Chemistry: All sediment samples were collected in 8 oz certified, TraceClean®, wide-mouth, amber glass jars with PTFE lined lids. Water samples were collected in 32 oz certified, TraceClean®, narrow-mouth, amber glass jars with PTFE lined lids. HCl added as a preservative (VWR, Randor, PA). In June, 2015 water and sediment samples were collected from each sampling location, in duplicate. Water was collected by complete submersion without headspace and sediment was collected by scooping the top layer of sediment with a solvent rinsed jar. Samples were stored on ice until analysis was conducted by ALS Global Laboratory in Fort Collins, Colorado within 1 week of collection. Extraction and analysis procedures were based on Texas Natural Resource Conservation Commissions standard protocol TX1005 for quantification of total petroleum hydrocarbons (TPH). Briefly, samples were extracted in solvent with 2,3,4triflurotoluene and O-terphenyl added as surrogate. TPH were detected by Gas Chromatography with a Flame Ionization Detector (GC-FID). Concentrations were quantified by summing all C6C28 compounds and comparing that summation to a minimum of a 5-point calibration curve of TPH standards. Quality controls included a field blank, method blank, laboratory control samples, matrix spikes and duplicate samples. Although not discussed in detail in this report, additional West Creek sediment samples were screened for potential organic contaminants unrelated to petroleum hydrocarbons by GC-MS set to full scan mode. No significant nonhydrocarbon organic compounds were detected. In October 2015 sediment samples were collected from each site, in duplicate, for identification and quantification of semi-volatile polycyclic aromatic hydrocarbons (PAHs) and aliphatic hydrocarbons. Sediment was collected by pressing a jar (discussed above) with a 18 �solvent cleaned exterior into the sediment, open end first, and then scoping upward filling the jar approximately half full. Water was carefully decanted from the sediment then jars were capped and stored on ice until samples were frozen at -4°C. Samples were shipped frozen on dry ice to the Geochemical and Environmental research Group at Texas A&M University for quantitative analysis of polycyclic aromatic hydrocarbons and aliphatic hydrocarbons using methods outlined by Short et al. (1996). Briefly, samples were spiked with deuterated PAH and aliphatic surrogates and extracted by dichloromethane. PAHs were quantified and identified by gas chromatography with a mass selective detector (GC-MS). Aliphatics were quantified and identified by GC-FID. Benthic Macroinvertebrate Community Structure: Benthic macroinvertebrates were collected from each site in June and October 2015 using a 0.1-m2 Hess Sampler. Samples (n=5 per site per sampling event) were collected from riffle habitats with similar physical features (e.g., depth and velocity). Substrate within each sampling area was thoroughly agitated and individual rocks scrubbed by hand to dislodge benthic organisms. Samples were washed through a 350-µm sieve and organisms preserved in 70% ethanol in the field. Macroinvertebrate samples were sorted in the laboratory using a standard 300-count protocol (Moulton et al. 2000). Macroinvertebrates were generally identified to genus. However, non-insects were identified to order, diptera were identified to family and caddisflies (Trichoptera) in the family Hydroptilidae were identified to family. Mottled Sculpin Collection and Processing: Mottled Sculpin were selected for a thorough health assessment because they are benthic predators and are strongly associated with sediment where petroleum contaminants concentrate. They are also less mobile than other local species (e.g., trout) and would be less likely to migrate among sites. Therefore, sculpin collected at a particular site should represent environmental conditions at that site. Old (> 2+ and > 20g) and young (< 1+ and < 4g) Mottled Sculpin (n=5 and 5, respectively) were collected opportunistically by Colorado Parks and Wildlife during routine electrofishing surveys at each sampling site. Sculpin were weighted, measured and anesthetized in 1 g/L neutral buffered of tricaine methanesulfonate (MS-222). Blood was collected from the caudal vein with a heparinized 28-gauge syringe. To prevent clotting, collected blood was thoroughly mixed by gently moving the plunger up and down 5 times 19 �without discharging it from the syringe. A small drop of blood was placed on a clean microscope slide and a blood film was made after the syringe tip was removed to limit cell lysis during blood expulsion from the tip (Campbell, 2015). Blood films were air dried then fixed in the field by soaking in ACS grade methanol for 15 seconds to preserve cellular architecture and air dried until they were stained by submersion in a modified Wright-Giemsa stain by professional clinicians at Colorado State University’s Clinical Pathology Laboratory within the Veterinary Teaching Hospital. Lateral body walls and caudal fins were removed and individuals were preserved in Bouin’s fixative solution. After 96 hours Bouin’s solution was replaced with 80% ethanol and stored until processed for histology. Otoliths were extracted from sculpin heads, none of which were preserved in Bouin’s solution, and fish were aged using light microscopy. Differential Leukocyte Counts: Stained blood films were analyzed by professional clinicians at Colorado State University’s Clinical Pathology Laboratory within the Veterinary Teaching Hospital. Leukocyte differentials were determined by manually identifying 100 leukocytes as lymphocytes, neutrophils, monocytes, basophils or eosinophil’s by light microscopy. Histology: Fixed tissues were trimmed in the transverse plane, routinely processed, and embedded in paraffin blocks. Processed tissues were sectioned at 5 microns, collected on charged slides, and stained with hematoxylin and eosin for evaluation by light microscopy. Sculpin tissues were routinely surveyed and evaluated for abnormalities by a Board Certified Veterinary Anatomical Pathologist. Tissues included gill, oral cavity mucosa, skeletal muscle, bone, heart, liver, kidney/inter-renal tissue, adrenocortical cells, exocrine and endocrine pancreas, adipose, testis/ovary, spinal cord, esophagus, stomach, intestine, cecae, spleen, lateral line, and swim bladder. The mucosal surface of gill rakers was used as a proxy for the mucosa of the oral cavity, when available. In some cases, gall bladder, Brockman’s body, the Corpuscle of Stannius, and thymus were also available for evaluation. Liver Histology - Individual necrotic hepatocytes and melanomacrophages in sections of liver were counted in ten high-power-fields (HPFs) (400x magnification). Hepatocyte 20 �anisocytosis and anisokaryosis were scored subjectively, as 1 = mild, 2 = moderate, and 3 = severe. Splenic Melanomacrophages - The pattern of melanomacrophages in splenic tissue were scored based on distribution and aggregation, as 0 = no recognizable melanomacrophages present, 1 = melanomacrophages evenly dispersed throughout the parenchyma in low or moderate numbers, and 2 = melanomacrophages organized into discrete nodular melanomacrophage centers as well as more loosely dispersed in the parenchyma. Individual melanomacrophages were also counted in a HPF (400x) at the location on the spleen with their highest density. Renal Melanomacrophages - Renal melanomacrophages were counted in a single HPF (400x) representing their highest density. The number of discrete melanomacrophage centers was also counted per HPF (400x), and a renal melanomacrophage index was calculated (number of melanomacrophages/HPF and number of melanomacrophage centers/HPF). Statistical Analyses: Differences in macroinvertebrate abundance and richness among stations in West Creek were analyzed using 1-way analysis of variance in SAS 9.3 (PROC ANOVA) (SAS Institute Inc. Cary, North Carolina). Data were log-transformed to meet assumptions of parametric statistics. If the overall F-statistic was statistically significant (p<0.05), Duncan’s multiple range test was used to determine differences among individual stations. We used canonical discriminant analysis (PROC CANDISC) to examine separation and overlap of macroinvertebrate communities among stations based on abundance of the 11 dominant taxa. For analytic chemistry, differential leukocyte counts and histology, descriptive statistics were calculated on Microsoft Excel (Microsoft Corp., Redmound, WA). Statistical differences (p<0.05) were calculated by one-way ANOVA with Tukey’s post-hoc analysis on Sigma Plot 11.0 (Systat Software Inc., Chicago, IL). Log10 transformations were conducted were necessary to meet the assumptions of parametric statistics. For DRO and EROD analysis, data below quantification limits (BQL) were given a value of half the quantification limit, and data below detection limits (BDL) were given a value of zero. 21 �Results and Discussion Analytic Chemistry: When aliphatic hydrocarbons and PAHs were quantified in West Creek sediments (Fall 2015) neither were present in high concentrations. However, numerous congeners were detected. Individual PAHs were quantified at low concentrations (0.4 – 28.3 µg/kg), resulting in total sediment PAH concentrations ranging from 71.7 to 144.5 µg/kg (Figure 4). When analyzed for significant differences among sites, differences were not detected for either total PAHs or total aliphatic hydrocarbons in sediments. It is important to note that these sediment concentrations were considered typical for a 3rd to 4th order stream running adjacent to a highway. For example, sediment PAH concentrations in West Creek were several orders of magnitude lower than urban stream sediment and similar to unimpacted streams (Beasley and Kneale 2002; Brown and Peake 2006). Although PAH concentrations at the spill site and downstream were not significantly elevated, more than 2.5 years had elapsed since the spill occurred, providing sufficient time for downstream sediment transport and degradation. PAHs resulting from the 2013 spill would have likely be deeply buried or relocated further downstream by the time samples were collected in 2015. Of the PAHs detected in West Creek sediments, the majority (68.9%-82.3%) were alkylated homologs rather than parent compounds (Figure 4) indicating that petroleum was their probable source (Saha et al. 2009). However, it is unlikely that the 2013 spill was the source of contamination because a concentration gradient was not detected among sites and concentrations were within a range typically observed in minimally impacted freshwater bodies (Beasley and Kneale 2002; Cox and Clements 2013). Sediment PAHs detected in West Creek were presumably from upstream sources, runoff from the adjacent highway and from naturally occurring sources or atmospheric deposition. Furthermore, a concentration gradient was not observed among sites for aliphatic hydrocarbons. Unlike PAHs, most aliphatic hydrocarbons were not derived from petroleum. Instead, 67.7% to 87.4% (Figure 5) were most likely decomposition products of plants and their epicuticular waxes (Baker 1982). West Creek water was also measured for petroleum hydrocarbons, but none were detected. Additionally, other potential organic contaminants were screened in West Creek water 22 �and sediment. Organic chemicals unrelated to petroleum, such as pesticides and flame retardants, were not detected. Sediment PAH Concentration (ng/g) 160 140 Total PAHs Alkylated PAHs 120 100 80 60 40 20 0 WC1 WC2 WC3 WC4 Figure 4: PAH concentrations for West Creek sediments from samples collected in Fall 2015 analyzed by GCMS. Alkylated PAHs refer to PAHs with carbon side chains that are more often of derived from petroleum sources rather than combustion. Data are means with SEM of two replicate samples. Sediment Aliphatic Concentration (ng/g) 60000 50000 Total Aliphatics (<C36) Epicuticular Wax Contribution 40000 30000 20000 10000 0 WC1 WC2 WC3 WC4 Figure 5: Aliphatic hydrocarbon concentrations (C12-C36) for West Creek sediments from samples collected in Fall 2015 analyzed by GC-FID. Epicuticular wax contribution refer to aliphatics with sizes C 27, C29, C31 and C33 which are typically derived from plant decomposition (Baker 1982). Data are means with SEM of two replicate samples. 23 �Aquatic Macroinvertebrate Communities: Aquatic macroinvertebrate communities were sampled in early summer and fall 2015. Despite similar physical and chemical characteristics among sites and their proximity (6.5 km between WC1 and WC4), aquatic macroinvertebrate communities varied greatly among sites. A principle component analysis (PCA) of the most dominant taxa at each site demonstrated that the reference community (WC1) was strikingly different from the community immediately downstream of the spill location (WC2) and also from communities further downstream (WC3 and WC4) (Figure 6). Moreover, at each site, Hess samples (n=10 per site) were similar to each other. Therefore, observed dissimilarity between the reference community and other communities was not a function of intra-site patchiness; it was the result true differences in overall community structure among the sites. WC1 WC2 WC3 WC4 Canonical Variable 2 (37%) 6 4 2 0 -2 -4 -6 -8 -6 -4 -2 0 2 4 6 Canonical Variable 1 (57%) Figure 6: Canonical variable analysis of community structure (early Summer and Fall 2015) at sites on West Creek. Data points within a given symbol are replicate Hess samples at a particular site. Overlap and proximity of data points indicate similarities in community structure. Differences among communities are proportional to increasing distances between points. Differences in community structure among sites were predominantly driven by the insect order Plecoptera (stoneflies), families within the order Diptera (Stratiomyidae, Simulidae and 24 �Chironomidae), and Oligochaeta. Petroleum hydrocarbon sensitive stoneflies were abundant at the reference site (WC1) compared with the site immediately downstream of the spill (WC2), while the opposite was true for tolerant Oligochaeta and Stratiomyidae (Figure 6). Chironomidae were also present in high densities at WC1, but density decreased at downstream sites and the reverse occurred for Simulidae. According to historic BLM sampling, stoneflies were detected at the spill site in 2011 before the spill occurred. However, in August 2013, 8 months after the spill, stoneflies were not detected (BLM 2011; BLM 2013). In 2015, Isoperla stoneflies were detected by our sampling at the impacted sites (WC2-WC4), but the density was significantly lower (P < 0.001) compared to the reference site. Isoperla (order, Plecoptera) is a highly sensitive, long lived, predatory stonefly which is particularly susceptible to petroleum toxicity. Crunkilton and Duchrow (1990) identified Isoperla as one of the most adversely affected taxa after a large crude oil spill in a river and similar results were obtained by Poulton et al. (1998). Additional studies have also classified stoneflies as highly sensitive to petroleum contamination (Simpson 1980; Woodward and Riley 1983). Moreover, they are slow to recolonize streams after a petroleum spill compared with other taxa (Michaelis 1983; Crunkilton and Duchrow 1990). Results of our study at West Creek show that stoneflies were extirpated from WC2, immediately downstream from the spill, and have not recovered by fall 2015. During our 2015 sampling, Oligochaeta densities were significantly higher at WC2, immediately downstream from the spill, compared to WC1 (P < 0.001; Figure 7). According to historic BLM sampling Oligochaeta were not present at WC2 or downstream in 2011 but were detected at WC2 in August 2013, 8 months after the spill (BLM 2011; BLM 2013). Proliferation of non-insects, particularly Oligochaeta, after petroleum spills has been documented previously (Harrel 1985; Guiney 1987). Other studies determined that Oligochaeta were completely unaffected or more tolerant to petroleum contamination than other taxa (McCauley 1966; Smith et al. 2010). Crunkilton and Duchrow (1990) concluded that Oligochaeta, Simulidae and Chironomidae were the earliest colonizers after near complete extirpation of all aquatic macroinvertebrates following a large crude oil spill. Moreover, in laboratory aquaria, Vorob’ev et al. (2010) demonstrated that Oligochaetes (Limnodrilus hoffmeisteri) not only survived 25 �sediment crude oil concentrations (16.72 g/Kg) that would be lethal to nearly all taxa, but they also reduced the sediment oil load 17 – 42% faster than aquaria without oligochaetes. When standard metrics of aquatic macroinvertebrate and stream health were quantified differences among sites were less dramatic. For example, total EPT abundance (sum of Ephemeroptera, Trichoptera and Plecoptera) was significantly reduced at WC2 compared with the reference site, but recovered at other downstream sites (Figure 7). Furthermore, when total abundance of organisms and species richness were analyzed, no differences were detected among sites. These results contrast with observational studies which reported significant changes in total and EPT abundances as well as species richness immediately after a spill (Harrel 1985; Guiney 1987; Crunkilton and Duchrow 1990; Lytle and Peckarsky 2001; Smith et al. 2010). Had our sampling at West Creek occurred immediately after the 2013 spill, we would have likely seen more pronounced effects in these metrics. Importantly, the standard metrics of species richness and total abundance do not appear to appropriately characterize the 2015 West Creek aquatic macroinvertebrate communities. Abundances of Plecoptera, Oligochaeta and PCA clearly demonstrate that the reference community was substantial different from sites downstream from the spill. Moreover, it is worth considering how community differences likely manifest as altered ecosystem services. For example, non-insects like Oligochaeta do not emerge from the water as adults, severing critical aquatic-terrestrial linkages at WC2 and WC3 which are necessary for healthy riparian zones (Baxter et al. 2005). Despite striking difference in community structure among sites, sophisticated analytical chemical procedures failed to detect significantly elevated concentrations of petroleum hydrocarbons at downstream sites. We hypothesize that observed community differences are not the result of persistent, chronic petroleum exposure but rather a legacy of the 2013 spill. The West Creek spill altered the equilibrium state of the community, which is now in the process of recovery. The spill was highly lethal to dominant, long-lived taxa (e.g., stoneflies) at the spill site and to a lesser extent downstream. The spill opened niches for colonization by tolerant taxa (e.g., Oligochaeta) and prevented recolonization by more sensitive pre-spill taxa. While the original pre-spill community is expected to eventually recolonize the downstream sites, the recovery timeframe is unclear. 26 �Number per Sample 140 120 100 80 60 40 20 0 A 1000 Plecoptera (P < 0.0001) 800 Number per Sample AB AB B 400 200 B B B 0 1000 Non-insects WC1 WC2 WC3 WC4 2000 (P < 0.0001) Total Abundance (P = 0.7655) 1600 A 600 1200 400 200 A 600 WC1 WC2 WC3 WC4 800 EPT (P = 0.0649) 800 A B B 0 400 0 WC1 WC2 WC3 WC4 WC1 WC2 WC3 WC4 Figure 7: Number of individual organism collected by Hess sampler at sites on West Creek. WC1 is an upstream reference site, WC2 is immediately downstream of the spill location, WC3 and WC4 are further downstream. Data are means plus SEM of 10 replicate samples taken over two sampling periods (early Summer and Fall 2015). P-values indicate significance of an ANOVA calculation. Letters above histograms indicate significant differences among treatments determined by post-hoc analysis. Mottled Sculpin Blood Differentials: Inferences on stress type or the nature of disease can be made based on the proportions of specific types of white blood cells in peripheral blood (Campbell 2015). Peripheral blood differential counts are commonly used in human and veterinary medicine to elucidate disease when impairment is suspected (Shapiro and Greenfield 1987; Campbell 2015). Blood differentials change in relation to stress, sex, species, nutritional status, age and season (Blaxhall and Daisley 1973; Ellsaesser et al. 1985). But, when properly applied as part of a weight of evidence approach, they aid in detecting suspected infection, inflammation and acute or chronic toxic exposure. For example, an increase in the proportion of lymphocytes indicates viral or bacterial infection, whereas increases in monocytes can indicate chronic inflammation. However, blood differential as well as all other sublethal diagnostic tools should not be evaluated in a vacuum; they should be included as a part in a holistic analysis. 27 �We measured white blood cell differential counts from young and old Mottled Sculpin at each of the four sites on West Creek (Tables 5 and 6). Across all sites young sculpin had much higher proportions of lymphocytes than older sculpin which is typical of fish (Campbell 2015). In young sculpin, neutrophils and monocytes were significantly elevated at WC4. For older sculpin, monocytes were significantly elevated at sites WC1 and WC2 compared with WC3 and WC4. Additionally, at WC2 there was a sculpin with aberrantly elevated monocytes. This sculpin was discovered to have inflamed ovarian tissue of unknown etiology. As a whole, differences among sites were slight which was expected given current low petroleum hydrocarbon concentrations in sediments. However, blood differentials at WC4 suggest a modest but significant immune response in young sculpin. Table 5: White blood cell differentials for old sculpin at each of four sites on West Creek. Data are means with SEM in parentheses. Letters following data indicate significate differences among treatments. At WC2 an outlier was excluded from statistics and is presented below. Site WC1 WC2 WC3 WC4 WC2 (Outlier) % Neutrophils 30.0 (2.09) 30.75 (4.23) 31.20 (4.58) 19.60 (5.62) 2.0 (NA) % Lymphocytes 63.0 (3.34) 63.25 (4.50) 66.80 (4.92) 79.40 (5.62) 17.0 (NA) % Monocytes 7.0(1.29) A 6.0 (0.71) A 2.0 (1.03) B 1.0 (0.77) B 81.0 (NA) % Eosinophils 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (NA) % Basophils 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (NA) Table 6: White blood cell differentials for young sculpin at each of four sites on West Creek. Data are means with SEM in parentheses. Letters following data indicate significate differences among treatments. Site WC1 WC2 WC3 WC4 % Neutrophils 8.0 (1.0) A 6.80 (1.69) A 6.25 (1.38) A 16.2 (1.11) B % Lymphocytes 90.50 (0.50) A 92.40 (1.62) A 91.0 (1.37) A 76.40 (1.91) B % Monocytes 1.50 (0.50) A 0.60 (0.24) A 2.50 (0.87) A 7.40 (1.9) B % Eosinophils 0.0 (0.0) 0.0 (0.0) 0.25 (0.25) 0.0 (0.0) % Basophils 0.0 (0.0) 0.20 (0.20) 0.0 (0.0) 0.0 (0.0) Mottled Sculpin Histopathology: Melanomacrophage aggregates are discrete centers that increase in size and/or frequency with senescence, disease and with intensifying environmental stress (Agius and Roberts 2003; Campbell 2015). They contain immune cells, pigments and antioxidants and act as depositories or recycling centers for cellular breakdown products, infectious agents and toxic material (Agius and Roberts 2003). Melanomacrophage centers, particularly those in the spleen, are reliable 28 �indicators of water quality (Blazer et al 1997; Schmitt and Dethloff 2000; Agius and Roberts 2003). Importantly, increases in splenic melanomacrophages aggregate (SMA) sizes and frequency have been shown to correlate with exposure to petroleum in the laboratory and field (Khan et al. 1984; Ali et al 2014) We observed a significant (P < 0.01) increase in the number of melanomacrophage cells associated with SMAs at WC3 compared with WC1 and WC2 (Figure 8). Because of the difficultly determining what constitutes clusters of lone melanomacrophage cells and what constitutes an aggregate of melanomacrophages, as well as the fact that these counts only represented the location on the spleen with the highest SMA density, a qualitative scoring system was employed to characterize SMA patterns. For qualitative scoring, all available spleen tissue was survey and scored as 0 (no recognizable melanomacrophages present), 1 (melanomacrophages evenly dispersed throughout the parenchyma in low or moderate numbers) and 2 (melanomacrophages organized into discrete nodular melanomacrophage centers as well as more loosely dispersed in the parenchyma). No discernible SMAs were present in young sculpin except in one fish from WC4 (pattern 1). In contrast, older sculpin from sites WC1 and WC2 had SMAs corresponding to patterns 0, 1 and 2. Older sculpin from WC3 and WC 4 showed SMA pattern 2, except one pattern 1 fish from WC4 (Table 7). Older sculpin at WC3 and WC4 had markedly increased SMAs compared with WC1 and WC2, suggesting increased environmental stress at WC3 and WC4. Interestingly, the most disturbed site during the 2013 petroleum spill (WC2), had fewer SMAs than sites further downstream. This would be an unexpected result if sculpin at WC2 had lived through the spill or quickly recolonized the site after the spill and were chronically exposed to petroleum contamination. However, all fish at sites WC2 and WC3 were killed during the spill. Sculpin quickly recolonized WC3 within six months after the spill and likely emigrated from the nearest colonization source (WC4), which was downstream of the reported fish-kill, but was exposed to petroleum during and after the spill (see section 2). Conversely, sculpin had only marginally begun recolonization at WC2 within six months and the closest recolonization source were unexposed sculpin upstream of the spill. Therefore, older sculpin analyzed in this study were likely exposed at higher concentrations for longer durations at WC3 and WC4 than at WC2. 29 �Macrophage Cells contained within SMAs per HPF 500 400 b (P < 0.01) ab 300 200 a 100 a 0 WC1 WC2 WC3 WC4 Figure 8: The number of melanomacrophage cells contained within splenic melanomacrophage aggregates (SMA) per 400x high-power-field (HPF) in old Mottled Sculpin at each site on West Creek. Counts were conducted at the location on the spleen with the highest density of SMAs. Data are means with SEM of four replicates (WC1, WC3, WC4) or five replicates (WC2). P-values indicate significance of an ANOVA calculation. Letters above histograms indicate significant differences among treatments determined by posthoc analysis. Table 7: Splenic melanomacrophage aggregate (SMA) patterns from old Mottled Sculpin in West Creek. SMA patterns were scored 0 (no recognizable melanomacrophages present), 1 (SMA were evenly dispersed throughout the parenchyma in low or moderate numbers), and 2 (SMA were organized into discrete nodular melanomacrophage centers as well as more loosely dispersed in the parenchyma). Data are presented as the number of fish at each site with a particular SMA pattern as well as the number a sculpin with spleens available for histological analysis from that site. Site WC1 WC2 WC3 WC4 Pattern 0 (mild) 1 of 4 1 of 5 0 of 4 0 of 4 Pattern 1 (moderate) 2 of 4 3 of 5 0 of 4 1 of 4 Pattern 2 (severe) 1 of 4 1 of 5 4 of 4 3 of 4 Congenital abnormalities: Congenital abnormalities were identified in two young sculpin from WC2 and WC4. In contrast, no congenital abnormalities were identified in older fish or fish from the reference site (WC1). Increased prevalence of congenital abnormalities has been observed in fish populations after petroleum spills (Peterson et al. 2003; Dubansky et al. 2013) and has been experimentally induced in laboratory experiments (de Soysa et al 2012; Incardona et al., 2014). Because moderate or severe congenital effects would be directly or indirectly lethal due to inhibited performance and reduced competitive ability, it was striking that two deformed fish were discovered in a small sample size of 20 young sculpin. 30 �The abnormal young sculpin at WC2 had roughly 1/3 of renal tubules ectatic to markedly dilated up to 800 µm in diameter (10-20x normal diameter) (Figure 9). No evidence of ureteral obstruction was identified on macroscopic or histologic exam. The ectatic tubules were predominantly lined by well differentiated columnar epithelial cells with modest amounts of eosinophilic cytoplasm and oval nuclei that had stippled cytoplasm and rare punctate nucleoli. Rarely, the epithelium was non-uniformly attenuated to low cuboidal or squamous morphology. The dilated tubules contained sparse amounts of stippled to finely fibrillar eosinophilic debris. The abnormal young sculpin at WC4 had an anomalous tract of well differentiated but disorganized neural tissue located in the dorsolateral musculature (Figure 9). This tract was evident in the transverse section at the level of the distal esophagus. It was ~60 µm at its widest point and extended from the dorsal spinal cord dorsally into skeletal muscle for approximately 2 mm. The tract joined the spinal cord through a small perforation in the bone and cartilage of the vertebral dorsal lamina. Overall it was well demarcated, though in some fields it entrapped isolated skeletal muscle fibers. It was lined by delicate meninges and comprised of homogenous neuropil with low numbers of glial cells scattered loosely within. Low numbers of cells bearing melanin pigment were present along the meningeal surfaces and in the perivascular interstitium. Hepatic injury: Individual hepatocyte necrosis was identified in all fish. The average count of necrotic hepatocytes per 10 HPFs were significantly higher in old sculpin than in the young age class (7.6/10 HPF +/- 1.52 vs 2.3/10 HPF +/- 0.41, P < 0.001). Renal melanomacrophages: Renal melanomacrophages were more common in the older age class when counted as melanomacrophage cells per single HPF (34.2/HPF +/- 9.23 vs 0.2/HPF +/- 0.12, P < 0.001) or when counted as number of melanomacrophage centers per HPF (2.5/HPF +/- 0.32 vs 0.2/HPF +/- 0.12, P < 0.001). Additional histologic findings: Many fish had evidence of mild or moderate stomatitis of uncertain etiology. Branchitis was common and ranged from mild to focally severe. When present, branchitis was proliferative with partial to complete filling of interlamellar troughs by epithelial cells and few inflammatory cells. An etiology was not evident histologically. All five adult fish from WC3 had multifocal branchial vascular telangiectasia. One fish from WC2 and one from WC1 also had focal telangiectasia. No distinct pattern was identified between site, age, gender, and the presence of stomatitis and branchitis. 31 �A B C D E F Figure 9: Mottled Sculpin histological panel. A) Normal kidney in a young sculpin. B) Pathological kidney with ectatic renal tubules in a young sculpin from site WC2 – notice dilated tubules. C) Normal spinal/neural tissue in a young sculpin. D) Pathological spinal/neural tissue in a young sculpin from site WC4 – notice light pink tissue extending dorsally from the spinal cord. E) Old sculpin with pattern 0 splenic melanomacrophages (not observable) from site WC1. F) Old sculpin with pattern 2 splenic melanomacrophages (organized into discrete nodular centers) from site WC3 – notice lightly pigmented areas in bottom portion of the spleen. 32 �Conclusions Sediment chemistry in West Creek was similar among stations by Fall 2015 with regard to persistent petroleum hydrocarbon concentrations. However, aquatic macroinvertebrate recovery, particularly at WC2 lagged and this was most evident in loss of stoneflies as well as proliferation of Oligochaeta at that site. Both patterns are classical effects of petroleum spills in streams. This was likely a result of the legacy of petroleum contamination, not current toxic exposure. Mottled sculpin was also adversely affected by the legacy of contamination, but more so at WC3 and WC4 than WC2. We hypothesize that this was due to complete mortality of all fish at WC2 and WC3 during the 2013 spill. However, WC3 was probably recolonized by downstream petroleum exposed fish and WC2 by unexposed upstream fish due their disparate proximity to source populations. Rainbow Trout Bioassay Goals and Objectives Our objective was to examine the effects of simulated diesel spills on Rainbow Trout, (Oncorhynchus mykiss) mortality and to establish concentration and time response relationships. We also wanted to explore field methods for gathering data on indicators of sublethal exposure and injury to spilled petroleum. Our goal was to provide information on the critical timing of field procedures employed immediately after a diesel spill and to validate the utility of promising bioassessment tools. Methods Rainbow trout, Experiment Set-up and Acclimation: Six-month old (18.1 g +/- 0.44; 122.8 mm +/- 1.1) Mt. Shasta strain Rainbow Trout (RBT) reared outdoors under sunlight were obtain from the Colorado Parks and Wildlife Bellevue Hatchery. RBT were transported to the Colorado Parks and Wildlife Aquatic 33 �Toxicology Laboratory (ATL) and stored indoors in large holding tanks at 13°C in dechlorinated tap water. Water quality characteristics were typical of montane Rocky Mountain streams, including low hardness (49-57 mg/L CaCO3), alkalinity (37−42 mg/L CaCO3), moderate pH (7.4) and specific conductivity (124-128 μS). Wide spectrum visible, UV-A and UV-B light provided a semi-natural lighting regime with AgroMax® T5 UV-A Plus and Exo Terra® ReptiGlo 10.0 fluorescent bulbs on 14hr:10hr (light:dark) cycle (AgroMax, Summerdale, AL and Hagen, Mansfield, MA). Fish were acclimated for 7 days until they were transplanted to experimental tanks. RBT (n=6 fish per aquaria) were placed into 24 9.5-liter glass aquaria and acclimated to experimental conditions for 48 hours. Each tank contained ATL water, two food-grade stainlesssteel refugia structures and 750 g of a sand/gravel mixture collected from the Cache La Poudre River. Tanks resided in a chilled water bath holding temperatures between 11°C-14°C and were vigorously aerated by air pumps. Each tank was provided the same light regime as described above resulting in 6.2 µW/cm2 UV light (+/- 0.12) during lite hours. RBT were feed daily and 33% water changes were conducted daily. Exposure Regime, Physical Chemistry and Analytic Samples: After 48-hours of acclimation, each tank was randomly assigning a nominal diesel loading rate (0, 75, 150, 300, 600, and 1200 mg/L diesel in water), creating 4 replicates of 6 treatments. Simulated spills were initiated as a single pulse of diesel fuel poured directly into tanks to achieve diesel loading rates. Diesel fuel was ultra-low sulfur, summer blend purchased from a local gas station. After six-hours water samples were collected from each tank for diesel range organic chemical analysis (DRO) in a 8 oz certified, TraceClean®, narrow-mouth, amber glass jars with PTFE lined lids and frozen at -4°C (VWR, Randor, PA). Immediately after DRO samples were collected, a 33% water change was conducted and thereafter water changes returned to a once every 24 hour schedule. Because no additional diesel was added after the first treatment, diesel exposure concentrations were expected to decrease over time. Fish were feed daily and dead fish were removed when mortality or moribundity was observed. Temperature, dissolved oxygen and pH were measured by YSI Pro ODO meter; pH and specific conductance by YSI 63 meter (YSI Incorporated, Yellow Springs, OH). Total UV-A and UV-B intensity were measured one inch above the water’s surface with a certified calibrated UVA/B light meter (Sper 34 �Scientific, Scottsdale, AZ). Ammonia was measured by Hach Ammonia (Nitrogen) test strips (Hach Co, Loveland CO). At the conclusion of the experiment (96 h), water and sediment samples were collected for DRO analysis as previously described. Analytic Chemistry: Water samples were analyzed for DRO by ALS Global Laboratory in Fort Collins, Colorado. Extraction and analysis procedures were based on U.S.EPA methods SW-846, 8000C and 8015D. Briefly, samples were extracted in solvents (i.e. hexane and methylene chloride) with O-terphenyl added as surrogate. Total DRO was detected by GC-FID. Concentrations were quantified by summing all C10-C28 compounds and comparing that summation to a minimum of a 5-point calibration curve of DRO standards. Quality controls included a method blank, laboratory control samples, matrix spikes and duplicate samples. Fish Processing: After 96 hours, exposed fish were anesthetized in 1 g/L neutral buffered tricane methanesulfanate (MS-222). Lengths and weights were measured. Blood was collected from the caudal vein with a heparized 28-gauge syringe. To prevent clotting, collected blood was thoroughly mixed by gently moving the plunger up and down 5 times without discharging it from the syringe. After the syringe tip was removed to limit cell lysis during blood expulsion from the tip, a small drop of blood was placed on a clean microscope slide and a blood film made by gently spreading the blood drop to create a film with a feathered edge (Campbell, 2015). Blood films were air dried then fixed by soaking in methanol for 15 seconds and air dried to preserve cellular architecture until they were stained by submersion in a modified WrightGiemsa stain. A drop of blood from each fish was also tested for hemoglobin concentration by HemoCue 201 hb+ (Hemocue America, Brea, CA). Hemoglobin, readings were then corrected for salmonids with the equation [Hb] = 0.815 (HemoCue [Hb]) − 2.198 (Andrewartha et al. 2016). Fresh RBT blood (100-300 µL) of was added to a 1.3 mL heparin tube and pooled from each surviving fish within a replicate tank to create a pooled blood sample (Nümbrecht, Germany). Pooled blood was held on ice until processed for hematocrit and blood plasma 35 �collection by centrifugation (2,500 g for 5 min in microhematocrit capillary tubes and 2,500 g for 15 min in the heparin tubes, respectively). Blood plasma (500 µL) was collected from the blood pellet and frozen at -4°C until a blood plasma panel was conducted by the Colorado State University’s Clinical Pathology Laboratory within the Veterinary Teaching Hospital using a Roche Cobas c501 Chemical Analyzer (Roche Diagnostics, Indianapolis, IN). Fish livers were excised from each fish, placed in 2 mL Eppendorf Safe-lock 2 mL cryotubes (Hamburg, Germany) then flash frozen in powdered dry ice. Livers were sent frozen on dry ice to the USGS Colombia Environmental Research Center for 7-Ethoxyresorufin-ODeethylase (EROD) activity assays. Remaining carcasses were prepared for histology by removing lateral body walls and caudal fins then preserving the body and organs in Bouin’s fixative solution at a ratio of 4:1 (Bouin’s:fish, by volume). After 96 h, Bouin’s solution was replaced with 80% ethanol and stored until processed for histology. EROD: Microsomes were prepared by homogenizing liver tissue, then centrifugation at 9,000 x g to produce an S9 fraction followed by centrifugation at 100,000 g to pellet microsomes. EROD assays were performed on the same day as microsome preparation. A resorufin standard curve, concentration range 0 to 80 pmol resorufin per assay well, was prepared each assay date. The standard curve was obtained by preparing dilutions from a resorufin super stock solution (200 M). The resorufin super stock was prepared by dissolving 5.0 mg of resorufin in 25 mL of methanol and stored at –20oC. The concentration of the resorufin super stock was checked spectrophotometrically at 571 nm on each assay date and appropriate dilutions were made to achieve a consistent standard curve on each assay date. In sample plates, the EROD reaction was initiated and then fluorescence responses resulting from resorufin production were monitored once a minute in each well for a period of 20 minutes. The analysis was performed on the first 10 minutes of the reaction separately from the second 10 minutes of the reaction. This allowed selection of the optimal reaction time. For each of the 10 minute intervals, the relative fluorescence intensities for the sample plates were compared to the corresponding linear fit of the seven point resorufin standard curve (6 replicates /concentration) and relative intensity values converted into pmol of resorufin. The resorufin content in each well was plotted against time to evaluate any deviations from linearity in the rate of formation of resorufin. A linear regression 36 �analysis was performed on each sample well to obtain the slope and estimate the rate of reaction (pmol/min). The reaction rate observed in each well was normalized according to the measured protein content, generating a value of specific activity in units of pmols/(min*mg) of protein. Reported results are the average of at least three replicates from the best linear fit 10 minute reaction time interval. Protein content in each well was determined using a fluorescamine based protein assay. Fluorescence measurements were made and the background corrected values compared to a BSA protein standard curve. The BSA protein standard curve was generated by making appropriate dilutions of a standard 4 mg/mL BSA stock solution and covered the range from 0 to 90 g of BSA/well. Fluorescence values were measured for each sample well, compared to those obtained for the standard curve, and the protein content was then calculated. The positive control (PC) used in the EROD assay was S9 fractions (Catalog #0850412, Lot # Q2101) obtained from rat hepatoma purchased from MP Biomedicals, LLC, Solon, OH. The material was prepared from Spague Dawley male rats induced with one intraperitoneal injection of Aroclor 1254 (500 mg/kg) suspended in corn oil. The S9 was prepared as a KCl homogenate (mg protein/ml) by the Lowery Method. All EROD assay reagents, as well as the kinetic reactions, were maintained at approximately 25oC. Samples for kinetic analysis were maintained at approximately 25oC for ten minutes prior to data collection. Differential Leukocyte Counts: Blood films were stained by modified Wright-Giemsa at Colorado State University’s Clinical Pathology Laboratory within the Veterinary Teaching Hospital. Differentials were determined by identifying and enumerating 300 consecutive leukocytes with light microcopy at 1000x using an oil immersion objective. Leukocytes were classified as lymphocytes or granulocytes (i.e., neutrophils, monocytes, eosinophils and basophils) according to Campbell (2015) identification procedures. Histology: Preserved RBT tissues were trimmed in the transverse plane, routinely processed, and embedded in paraffin blocks. Processed tissues were sectioned at 5 microns, collected on charged slides, and stained with hematoxylin and eosin for evaluation by light microscopy. 37 �Tissues were surveyed and evaluated for abnormalities a Board Certified Veterinary Anatomical Pathologist. Statistical Analyses: Descriptive statistics were calculated on Microsoft Excel (Microsoft Corp., Redmound, WA). Statistical differences (α = 0.05) were calculated by one-way ANOVA with Tukey’s posthoc analysis on Sigma Plot 11.0 (Systat Software Inc., Chicago, IL). Log10 transformations were conducted where necessary to meet the assumptions of parametric statistics. For DRO and EROD analysis, data below quantification limits (BQL) were given a value of half the quantification limit, and data below detection limits (BDL) were given a value of zero. Results and Discussion Analytic Chemistry: Measured diesel concentrations immediately prior to the first water change were 1-2 orders of magnitude lower than nominal concentrations (Table 8). This was likely due to adsorption of diesel onto sediment and others surfaces, volatilization, absorption into tissue, as well as low solubility of diesel in water. After 96 hours of exposure, aqueous diesel concentrations had only marginally decreased. This suggested that despite loss of aqueous diesel from water changes, there was sufficient diesel sources (e.g., contaminated sediment and other surfaces) within each tank to partially replace lost aqueous diesel. Sediment DRO analysis reveal that concentrations were substantially higher than aqueous concentrations (Table 8). Therefore, diesel partitioned to sediment with greater affinity than to water and contributed to the persistence of diesel within tanks as is typical for petroleum contamination (Hoffman et al. 2002). Behavioral Effects: Diesel exposure resulted in obvious, but unquantified behavioral effects during the experiment. These effects were most prominent at 300 mg/L diesel loading rates and above. Most notably was the observation of gulping behavior at the surface of the water which is traditionally associated with oxygen deprivation and has been previously described in dieselexposed rainbow trout (Khan et al. 2007). Although there was sufficient dissolved oxygen (7.3 38 �mg/L +/- 0.06) in each treatment, diesel may have physically smothered gills impeding oxygen transport over gill membranes. Another explanation could be biochemical asphyxiation, where oxygen was not processed or perfused into tissues adequately. In the higher diesel loading rates (1,200 mg/L) most trout began gulping 6 to 24 hours post-exposure. Because a disproportionate amount of diesel floated at the water’s surface, gulping behavior likely increased diesel exposure. Table 8: Nominal diesel treatment concentrations compared with measured aqueous and sediment DRO concentrations measured by GC-FID. Aqueous concentrations were measured 6 hours post exposure (immediately preceding the first post-exposure water change), at 96 hours (after four 33% water changes) and in sediment 96 hours post exposure. Data are means of 4 replicates with SEM in parentheses. BQL = “below quantification limits” indicating DRO was detected, but detection occurred at such low concentrations that quantification was unreliable. For 96 hours post-exposure aqueous samples only 2 replicates tanks were analyzed in treatments 75 and 300 mg/L and 3 replicates tanks for the control treatment samples due to container brakeage. Nominal 6 hours Post-Exposure 96 hours Post-Exposure 96 hours Post Exposure Diesel Measured Aqueous Measured Aqueous Measured Sediment Loading Rate Diesel Concentration Diesel Concentration Diesel Concentration (SEM) (SEM) (SEM) 0 mg/L 0.07mg/L (0.07) BQL 0.59 mg/L (0.18) BQL 2.40 mg/kg (0.00) BQL 75 mg/L 6.75 mg/L (1.55) 1.90 mg/L (0.00) 21.00 mg/kg (1.83) 150 mg/L 7.45 mg/L (1.44) 2.55 mg/L (0.64) 45.00 mg/kg (3.49) 300 mg/L 11.97 mg/L (1.03) 6.80 mg/L (1.00) 99.75 mg/kg (11.03) 600 mg/L 12.47 mg/L (4.53) 10.08 mg/L (3.76) 102.00 mg/kg (16.13) 1,200 mg/L 20.25 mg/L (4.87) 11.65 mg/L (3.55) 167.5 mg/kg (37.50) Lethargy or narcosis was also evident in all treatments greater than 75 mg/L. For example, during water changes fish from control treatments were startled and swam erratically. However, fish exposed to diesel responded slower or not at all during water changes. Moreover, some individuals within the 600 mg/L and 1,200 mg/L treatments did not respond to being gently prodded by a glass rod, whereas control fish avoided being prodded. Feeding in control treatments and the 75 mg/L treatments was vigorous, but vigor diminished as exposure increased and targeted feeding was not observed in the 1,200 mg/L treatment. 39 �Total Ammonia: Ammonia (3 ppm total ammonia) was detected in all treatments on day-3 post-exposure. However, there were no differences in concentration among treatments. Lethal concentrations of free unionized ammonia was not likely present as no fish died in control or 75 mg/L treatments during the experiment. Moreover, due to circumneutral pH (7.4), and low temperatures (11°C13°C), the vast majority of free ammonia was likely in a non-toxic, ionized NH4+ form rather than its more toxic unionized NH3 form. Therefore, ammonia was not suspected to have directly influenced mortality. However, it may have influenced gill histology results (discussed below). Mortality: Rainbow trout mortality occurred in a concentration and time-dependent manner (Figures 10 and 11). Mortality did not occur in the control and 75 mg/L treatments. However, in the 1,200 mg/L treatments approximately 90% of the fish died within 96 hours. Although ANOVA failed to distinguish significant differences among the lower treatments, a clear concentration response relationship was observed. Additionally, across all treatments we observed relatively little mortality within the first 6-24 hours post-exposure as most fish died only after a full day of exposure (Figure 11). 40 �Total Mortality 100 b (p < 0.01) % Mortality 80 60 a 40 a 20 a a a 0 75 0 150 300 600 1200 Diesel Loading Rate (mg/L) Figure 10: Effects of diesel exposure on rainbow trout mortality. Data are means +/- SEM of four replicates. P-value indicated significance of an ANOVA test and letters above histograms indicate a significant difference among treatments. 0 mg/L 75 mg/L 150 mg/L 300 mg/L 600 mg/L 1200 mg/L 100 % Mortality 75 50 Mortality Time-course 25 0 0 20 40 Time (hrs) 60 80 Figure 11: Effects of diesel exposure on the time-course of rainbow trout mortality over the duration of the experiment. Data are means with SEM of four replicates. EROD Rate: One of the primary mechanisms of injury associated with aromatic hydrocarbon exposure is the formation of free radicals, which causes a response in antioxidant defense mechanisms 41 �(Livingstone 2001; Mos et al. 2008). Because free radicals are difficult to measure, researchers have attempted using upregulation of antioxidants (e.g., catalase, superoxide dismutase, and glutathione-S-tranferase) to approximate aromatic hydrocarbon linked injury. However, results from these studies have been variable, difficult to interpret and expensive. Sample collection also requires highly controlled conditions which are more suitable to laboratory procedures than field settings. Moreover, recent work has suggested that antioxidant biomarkers are more likely to be “induced by frequent small exposures compared to infrequent large ones” (Sandrini-Neto et al. 2016). Therefore, an antioxidant response could be disproportionately affected by petroleum runoff from standard road usage. Because of these concerns, ethoxyresorufin-O-deethylase (EROD) rate, which is indirectly related to free radical production from aromatic hydrocarbon exposure, was measured in this study. EROD rate is a highly sensitive, well-researched indicator of exposure to aromatic hydrocarbons as well as other organic contaminants such polychlorinated biphenyls (PCBs) (Schmitt and Dethloff 2000). Briefly, EROD is an analytic procedure which measures the activity of the cytochrome p450 biotransformation system, which is critical for metabolism and clearance of a breath of toxic compounds including PAH’s (Whyte et al. 2000). Numerous studies have established strong relationships between EROD and fish health (Schein et al. 2009; Wu et al. 2012; Beyer et al. 2016). Increased EROD rates signals an organism is mounting a biochemical response to a toxin, which is bioavailable and interacting with tissues. In this study, EROD activity in control fish was significantly lower than all diesel treatments and increased in a strong concentration-dependent manner (Figure 12). This result provided clear evidence that aromatic hydrocarbon compounds from diesel exposure were entering fish liver tissue and were interacting with its biochemical systems. This fact is important because it suggest that toxicity observed in these experiments was not solely caused by physical smothering of gills. Although oxygen deprivation by physical smothering may have played an important role in trout mortality, biochemical intoxication was also important. Interestingly, there is an apparent, although not significant, decrease in EROD rate at the highest exposure concentration (1,200 mg/L), which was likely caused by diminished liver function and reduced ability to mount a biochemical defense. This hypothesis is supported by a characterization of other parameters utilized in this study (discussed below), which examined 42 �liver function and implicates liver deterioration (i.e., AST, ALT and blood urea nitrogen). Because cytochrome p450s are primarily liver enzymes, it is reasonable that as liver function decreased, its ability to metabolize toxins by the cytochrome p450 system would be diminished and thus EROD rate also decreased. EROD Rate 30 Mean EROD Rate (pmol/min/mg) c 25 (p < 0.001) bc 20 15 bc 10 b 5 b a 0 0 75 150 300 600 1200 Diesel Loading Rate (mg/L) Figure 12: Effects of diesel exposure on ethoxyresorufin-O-deethylase (EROD) rate from rainbow trout livers after 96 h exposure to diesel. Data are means with SEM of four replicates. P-value indicated significance of an ANOVA calculation and letters indicate significant differences among treatments. Histology: All rainbow trout were preserved for histopathological analysis and a subset was surveyed for abnormalities. We focused on liver and gill tissue, the expected targets of toxic action. However, no treatment effects were discovered histologically. Liver tissue appeared normal with no apparent differences among treatments. Additionally, no treatment effect was observed in gill tissue. However, moderate clubbing of gill lamella was observed across treatments and controls. This effect was attributed to incidental ammonia exposure during the experiment. Hematology: In response to environmental stress, immunological changes occur; often the number of lymphocytes in peripheral blood decreases while the number of granulocytes increase (Agius and 43 �Roberts 2003). Because petroleum exposure has been implicated as a causative factor in immune dysfunction, RBT blood film were analyzed for differential counts (Mos et al. 2008; Beyer et al. 2016). Here, we observed a significant increase in the proportion of granulocytes (defined as the sum of monocytes, neutrophils, basophils and eosinophils) at the 1,200 mg/L diesel treatment (Figure 13). Similar results were observed in Pearl Dace (Margariscus margarita) collected from a diesel contaminated pond (Khan 1999). Large increases in the proportion of granulocytes compared with lymphocytes generally indicate either infection or inflammation of tissue (Ellsaesser et al 1985; Campbell 2015). Because petroleum is a known inflammatory agent, inflammation seems a likely cause over an infectious disease (Khan 1999; Rodrigues et al. 2010). However, infection is possible given petroleum exposure has been linked to immune suppression and increased susceptibility to disease (Ali et al. 2016). Following the Deepwater Horizon oil spill, blood differentials were used to support effects of petroleum on the immune system (Ali et al. 2014; Beyer et al. 2016). Given that blood films are easy to collect, non-lethal, archivable and a routine procedure in human and veterinary medicine, their use in aquatic field studies will likely increase and become a helpful addition to bioassessment protocols. White Blood Cell Differentials 60 % Granulocytes 50 (p < 0.01) B 40 30 20 10 A A A A 300 600 A 0 0 75 150 1200 Diesel Loading Rate (mg/L) Figure 13: Effects of diesel exposure on percent granulocyte immune cells in rainbow trout white blood cell differentials. Data are means with SEM of four replicates. However, only 2 replicates of 1,200 mg/L treatment were available due to mortality. P-value indicated significance of an ANOVA calculation and letters above histograms indicate significant differences among treatments. 44 �Blood plasma was also analyzed as part of routine diagnostic methods commonly used in human and veterinary medicine. Interestingly, across numerous parameters, analyte levels were shifted when comparing the 1,200 mg/L treatment to all other treatments (Figures 14 and 15). Although not statistical significant due to low sample size because of high mortality in the 1,200 mg/L treatment, when examined holistically a trend was evident. Numerous indicators of liver and kidney function as well as muscle damage were elevated in the 1,200 mg/L treatment compared with other treatments (e.g., ALT, AST, blood urea nitrogen and creatine kinase). Glucose was also elevated, potentially indicating a stress response associated with increased short-term energy availability which would be necessary for fueling biochemical defense mechanisms to mitigate toxic injury. This effect has been reported by studies examining plasma glucose after petroleum exposure (Al-Kindi et. al 2000). However, triglyceride lipids, were concurrently reduced potentially signaling that trout were using energy in the short term at the expense of the long term. Simultaneously, albumin, a blood protein often used as a normalization parameter, remained constant across treatments, suggesting that the previously discussed effects were not anomalies associated with a small sample size. When examined holistically, the blood plasma panel for the 1,200 mg/L treatment appears abnormal and implies toxic injury. Total protein, phosphate, magnesium, calcium, sodium, potassium, chloride and cholesterol were also measured in blood plasma. However, no effects were apparent among treatments. Treatment effects were not detected in hemoglobin concentration or hematocrit measurements of whole blood either. 45 �10 Blood Urea Nitrogen (mg/dL) 180 Glucose (mg/dL) 160 140 120 100 80 60 40 20 0 75 150 300 4 2 0 600 1200 350 75 150 300 600 1200 1200 6000 300 Creatine Kinase IU/L Triglycerides (mg/dL) 6 0 0 250 200 150 100 5000 4000 3000 2000 1000 50 0 0 0 75 150 300 600 1200 60 0 75 150 300 600 0 75 150 300 600 1200 1.8 1.6 50 1.4 Albumin (g/dL) Iron (ug/dL) 8 40 30 20 1.2 1.0 0.8 0.6 0.4 10 0.2 0 0.0 0 75 150 300 600 1200 Diesel Loading Rate (mg/L) Figure 14: Effects of diesel exposure on blood panel parameters analyzed in rainbow trout blood plasma. Data are means with SEM of four replicates. Open histograms indicate that only one replicate was available for analysis due to mortality in that treatment. 46 �Aspartate Aminotransferase (IU/L) Alanine Aminotransferase (IU/L) 18 15 12 9 6 3 0 0 75 150 300 1000 800 600 400 200 0 0 600 1200 75 150 300 600 1200 Diesel Loading Rate (mg/L) Figure 15: Effects of diesel exposure on alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels in rainbow trout blood plasma. Data are means with SEM of four replicates. Open histograms indicate that only one replicate was available for analysis due to mortality in that treatment. Conclusions Rainbow Trout were adversely affected by simulated diesel spills with regard to spill size and duration of exposure. However, even when exposures were lethal, mortality occurred over the course of several days and would have likely continued after the experiment ended. Importantly, this has consequences for the timing of fish-kill assessments after a petroleum spill occurs. Delayed mortality observed in our study suggests that several fish-kill surveys should be conducted over time to more completely quantify fish mortality. Hematological changes were only evident in fish exposed to the highest diesel loading rate, but they were profoundly affected in treatments where most other fish died. Therefore, fish that survived 4 days of exposure were presumably injured and their recovery prognosis was unclear. At low concentrations, ecologically important fish behaviors (e.g., swimming position in the water column, feeding vigor, lethargy, reaction) were adversely impacted, although these responses were not quantified. Because proper behavior is critical for survival in the wild, fish that did not die from acute exposures would be at greater risk of indirect mortality. Moreover, EROD analysis indicated that even when diesel exposures were not sufficient to cause mortality, petroleum compounds were 47 �bioavailable and entered fish tissue. Therefore, fouling of flesh likely occurred and trophic transfer of contaminants was possible. Histological changes were not an especially useful response in this experiment seemingly because of the short exposure duration and moderate clubbing of gill lamella due to low ammonia concentrations. As a result, the use of histological abnormalities as an assessment tool for short-term petroleum exposure requires further investigation. In summary, by using a breadth of field transferable sampling methods, exposure and hazard of a simulated diesel spill were characterized in a valuable fish species and in a manner than can be mirrored during actual petroleum spill scenarios. Periphyton Bioassay Goals and Objectives The objective of this experiment was to explore effects of simulated diesel spills on the biomass of a fundamental aquatic food source, periphyton communities. Additionally, we sought to examine the utility of the Bethotorch (BBE Moldaenke, Germany) as a potential field tool to examine changes in periphyton biomass over time. Methods Natural periphyton communities were colonized on 2’’x 2’’ porcelain tiles for two weeks in the Stream Research Laboratory (SRL), a stream mesocosm facility located in a greenhouse at the Colorado State University Foothills Campus (see Aquatic Insect Mesocosm Experiment for details). Following the colonization period, periphyton tiles were transplanted to CPW’s ATL (see Rainbow Trout Bioassay for details). Colonized tiles (n=10 per tank) were placed into each of 15 food-grade, stainless steel tanks (32.4 x 18 x 15 cm), filled with 4 L of dechlorinated ATL water that was continuously aerated by an air pump. Tanks resided in a chilled water bath holding water at 13°C-14°C. Light was provided by visible and UV-A/B bulbs set to a 16hr:8hr light:dark cycle (see Rainbow Trout Bioassay for details). Water changes (33%) were conducted daily and tiles were acclimated to laboratory conditions for 48 hours prior to the initiation of diesel exposures. 48 �Exposure Scenario: Each tank was randomly assigning a diesel loading rate to serve as a treatment concentration (0, 20, 200, 2000, and 20000 mg/L), creating 3 replicates of each treatment. Simulated spills were initiated as a single pulse of diesel fuel poured directly into the water to achieve appropriate diesel loading rates. Diesel fuel was ultra-low sulfur, summer blend purchased from a local gas station. After six-hours, water samples were collected from each tank for diesel range organic chemical analysis (DRO) in a 8 oz certified, TraceClean®, narrowmouth, amber glass jars with PTFE lined lids and frozen at -4°C (VWR, Randor, PA). Immediately after DRO samples were collected a 33% water change was conducted. Thereafter water changes occurred every 24 hour. Because no additional diesel was added after the first addition, diesel exposure concentrations were expected to decrease over time. At the conclusion of the experiment, water samples were collected for DRO analysis. Biomass Measurements: Using a portable, hand-held, fluorometric, Benthotorch (bbe Moldaenke, Schwentinental, Germany), periphyton biomass was measured from tiles (n=4 per tank) on 3 occasions: T1) immediately preceding diesel exposure T2) 12 hours post-exposure and T3) 84 hours postexposure. A stainless steel and PTFE support structure, developed by the ATL’s director (Pete Cadmus), permitted Benthotorch biomass readings without disrupting periphyton on the surface of tiles, thereby making multiple readings from the same tile possible. At each timepoint, Benthotorch readings were collected on the same tiles (n=4 per tank) allowing positive or negative growth to be observed over time. Analytic Chemistry: Water samples were analyzed for DRO by ALS Global Laboratory in Fort Collins, Colorado. Extraction and analysis procedures were based on U.S.EPA methods SW-846, 8000C and 8015D. Briefly, samples were extracted in solvents (i.e. hexane and methylene chloride) with O-terphenyl added as a surrogate. Total DRO was detected by GC-FID. Concentrations were quantified by summing all C10-C28 compounds and comparing that value to a minimum of a 5-point calibration curve of DRO standards. Quality controls included a method blanks, laboratory control samples, matrix spikes and duplicate samples. 49 �Statistical Analyses: Descriptive statistics were calculated on Microsoft Excel (Microsoft Corp., Redmound, WA). Percent of day-0 biomass (12-hour) was calculated by dividing the T2 biomass by the T1 biomass for each respective measured tile and then multiplying by 100. Percent of day-0 biomass (84-hour) was calculated by dividing the T3 biomass by the T1 biomass for each respective measured tile and then multiplying by 100. Statistical differences (α = 0.05) were calculated by one-way ANOVA with Tukey’s post-hoc analysis on Sigma Plot 11.0 (Systat Software Inc., Chicago, IL). For DRO analysis, data below quantification limits (BQL) were given a value of half the quantification limit, and data below detection limits (BDL) were given a value of zero. All biomass data are presented as summations of values for diatoms, blue-green algae and cyanobacteria. Results and Discussion Analytic Chemistry: As with other experiments presented in this report, measured diesel concentrations measured prior to the water changes were one to two orders of magnitude lower than nominal loading rates (Table 9). This was likely caused by absorption of diesel into periphyton, adsorption on surfaces, volatilization into the air and the low water solubility of diesel in water. Moreover, on day-3 of the experiment, after three water changes had occurred, measured aqueous diesel concentrations declined substantially. 50 �Table 9: Nominal diesel loading rates compared with measured aqueous DRO concentrations measured by GC-FID. Aqueous concentrations were measured 6 hours post exposure (immediately preceding the first water change) and at 84 hours (after four 33% water changes). Data are means of 3 replicates with SEM in parentheses. BQL stands for “below quantification limits” indicating DRO was detected, but detection occurred at such low concentrations that quantification was unreliable. Nominal Diesel Loading Rate 6 hours Post-Exposure Measured Diesel Concentration (SEM) 84 hours Post-Exposure Measured Diesel Concentration (SEM) 0 mg/L 20 mg/L 200 mg/L 2,000 mg/L 20,000 mg/L 0.15 mg/L (0.08) BQL 0.72 mg/L (0.32) 13.00 mg/L (5.70) 131.67 mg/L (24.92) 640 mg/L (245.44) Not measured Not measured Not measured 1.01 mg/L (0.10) 2.2 mg/L (0.46) Periphyton Biomass: The BenthoTorch is a real time, in situ, fluorimeter that accurately measures thin films of periphyton using cholorophyll-a concentration as a proxy for biomass (Kahlert and McKie 2014; Echenique-Subiabre et al 2016). Because of its field portability and ease of use, the BenthoTorch is a convenient bioassessment tool for quantifying periphyton density in streams. In this experiment, periphyton biomass was significantly reduced (P < 0.01) as diesel loading rate increased at both 12 hours (T2) and 84 hours (T3) post-exposure (Figure 16). These results are similar to other short-term studies examining acute toxicity of petroleum to periphyton (Nayar et al. 2004; Baek et al. 2013). In contrast, a longer-term study that included macroinvertebrates demonstrated growth in periphyton after petroleum exposure (Lock et al. 1981). When petroleum stimulated growth occurs, it is generally thought to be caused by reducing herbivory by grazing as well as tolerant periphyton taxa replacing extirpated sensitive taxa (McCauley 1966; Singh and Gaur 1989; Carmen et al. 2000; Baek et al. 2013). Regardless of how periphyton is affected there is consensus that changes to periphyton and algal communities occur after petroleum exposure. Given that autotrophs are fundamental components at the base of stream food webs, monitoring changes in periphyton is an important consideration when determining petroleum spill impacts and assessing recovery. 51 �T2 Periphyton Biomass (12 hours post-exposure) T3 Periphyton Biomass (84 hours post-exposure) % of Day-0 Biomass 180 180 p < 0.001 150 120 a a a 150 120 a 90 p < 0.01 ab ab 90 b 60 b 30 60 bc c 30 0 0 0 20 200 2000 20000 0 20 200 2000 20000 Diesel Loading Rate (mg/L) Figure 16: Effects of diesel exposure on periphyton biomass as measured by chlorophyll-a concentration at T2 (12 hours) and T3 (84 hours) after simulated spills were initiated. All data are means +/- SEM of three replicates. P-values indicate significance of an ANOVA calculation. Letters above histograms indicate significant differences among treatments. Data greater than 100% of day-0 biomass represent positive growth while data less than 100% represent negative growth when compared biomass measurement immediately preceding the initiation of the spill (T1). Conclusions Periphyton biomass, in the absence of herbivores, was profoundly and quickly affected by diesel in a concentration and time dependent manner. Because periphyton is a fundamental component of aquatic food webs, and provides critical ecosystem services such as oxygen production, contaminant removal and habitat for organisms, altered periphyton biomass would cause meaningful impacts to aquatic stream ecosystem services. Moreover, with the appropriate use of a Benthotorch, altered periphyton biomass can be easily monitored after a petroleum spill. Aquatic Insect Mesocosm Experiments Goals and Objectives We conducted two mesocosm experiments to quantify effects of a simulate diesel spill on macroinvertebrate communities. The specific objectives of these experiments were different. In 52 �Exp1 we examined how diesel spills of various sizes affected a healthy community of aquatic insects collected from the Arkansas River, CO. In Exp2 we investigated the context dependency of a diesel spill and asked the question: “Do communities collected from different locations respond similarly to a diesel spill?” Our primary goal was to determine if spilled diesel was acutely toxic to aquatic insects. Our secondary goal was to determine whether drift assessments would potentially be a useful biomonitoring tool during a spill event. Methods Insect Community Collection: To quantify short-term effects of a diesel spill on aquatic macroinvertebrate communities, two mesocosm experiments (Exp1 and Exp2) were conducted using naturally colonized insect communities collected from field sites. Briefly, 10 x 10 x 6 cm plastic trays with three holes (2.5 cm diameter) in each side were filled with clean cobble and pebbles. Trays were secured to stream bed in riffle habitat at two locations on the Arkansas River near Leadville, Colorado (stations AR1 and AR5) each with distinctive communities of aquatic insects (Clements et al. 2010). AR1 was a reference community characterized by higher densities and diversity of taxa generally thought to be sensitive to metals (orders: Plecoptera and Ephemeroptera), while AR5 had higher densities of metal-tolerant taxa (order: Trichoptera) as a result of long-term exposure to low concentrations of heavy metal pollution due to historic mining activity (Clements et al. 2010). Trays were colonized instream for 30 and 36 days for Exp1 and Exp2, respectively. Previous research demonstrated that resulting insect community colonizing these trays closely resembled the actual assemblage residing in the stream benthos (Courtney and Clements 2002). After the assigned colonization period, trays were haphazardly collected from the stream bed and placed in 4L coolers (4 trays per cooler) filled with stream water and transported to Colorado State University’s Stream Research Laboratory (SRL) housed in a temperature controlled greenhouse. The contents of each cooler were randomly assigned to one of 18 or 16 stream mesocosms for Exp1 or Exp2, respectively. 53 �Mesocosm Setup: Mesocosms were constructed from food-grade, stainless steel pans (53 x 32 x 15cm) in a flow-through design. Flows were maintained at 0.25 m/s and water depths at 11 cm by a foodgrade, stainless steel paddle wheels and stand pipes. Diluent water was delivered to each mesocosm from a deep mesotrophic reservoir at a rate of 0.3 L/min, resulting in a water residence time of approximately 40 minutes within each mesocosm. Diluent water was characteristic of montane Rocky Mountain streams with low hardness (30−38 mg/L CaCO3), alkalinity (25−29 mg/L CaCO3) and dissolved organic carbon (2.5−3.0 mg/L) (Clements et al. 2013). Wide spectrum UV-A and UV-B light were supplemented to each mesocosm by AgroMax® T5 UV-A Plus and Exo Terra® Repti-Glo 10.0 fluorescent bulbs (AgroMax, Summerdale, Al and Hagen, Mansfield, MA). Diesel Spill Exposure Scenario: Exp1: This experiment explored how a gradient of diesel concentrations, simulating various petroleum spill sizes, affected the structure and function of an aquatic insect community (AR5). All trays (n=4 trays) from within a given cooler (n=18 coolers) were randomly assigned to a mesocosm and each mesocosm was randomly assigned to one of six treatments, in triplicate (control, 75, 150, 300, 600 and 1200 µg/mL diesel). Diesel was ultra-low sulfur, summer blend purchased from a gas station in Fort Collins, Colorado. After 48 hours of acclimation to mesocosm conditions, diluent water was turned off temporarily, shifting mesocosms from a flow-through to a static system. Paddlewheels remained on during this period to maintain water current, turbulence and oxygenation. A single pulse of diesel was poured into each mesocosm to achieve an initial diesel loading rate reflecting the six experimental treatments. After 2 hours of exposure and prior to initiating flow-through conditions, a 40 mL water sample was collected in a certified clean, Teflon lid, head space vial containing HCl as a preservative. Water samples were placed on ice in darkness until diesel range organic (DRO) compounds were quantified by gas-chromatography coupled with a flame ionization detector (GC-FID). Subsequently, a 3’’ x 4’’ fine mesh net (< 350 µm) was placed into each stream for 120 seconds to capture organisms drifting in the water column. Collected organisms were preserved in 70% ethanol. Streams were returned to flow-through conditions after 6 hours. Simultaneously, and a fine mesh (< 350 µm ) net was place at the outlet of each standpipe to collect organisms emigrating from the 54 �mesocosms. Captured organisms were preserved in 70% ethanol after a 24 hour collection period. Fine mesh (< 350 µm) net was then placed over each standpipe to prevent further emigration. An additional water sample was collected as previously described three days after the initial dosing. Temperature (10.5 – 14.4°C) and dissolved oxygen (7.88 - 8.56mg/L) were measured on three occasions using a YSI Pro ODO meter; pH (7.3 – 7.4) and specific conductance (67.2 – 80.2 µS) were measured on 1 occasion using a YSI 63 meter (YSI Incorporated, Yellow Springs, OH). Total UV-A and UV-B intensity were measured on 2 occasions (7.0 µW/cm2 +/- 0.28) with a certified calibrated UVA/B light meter (Sper Scientific, Scottsdale, AZ). Four days after dosing was initiated all organisms remaining within each mesocosm were filtered through a 350 µm sieve. All organisms on the sieve were preserved in 70% ethanol then identified to genus, or family for order: diptera. Exp2: This experiment explored the context dependency of two different aquatic insect communities affected by simulated spill conditions. All trays (n=4) from a given cooler collected from station AR1 or AR5 were randomly assigned to an individual mesocosm and each mesocosm was randomly assigned a diesel loading rate (0 or 600 mg/L). The resulting experimental design was a 2 x 2 factorial (AR1 control, AR1 diesel, AR5 control and AR5 diesel) with 4 replicates. All other factors of this experiment were similar to those described above with the exception of timing of sample collections. Here, the initial water samples for DRO analysis were collected at 4 hours post-exposure rather than 2 hours post-exposure. The second water sample collection for DRO analysis occurred on 6 days post-exposure, rather than 3 days post-exposure. Additionally, all remaining organism within each mesocosm were collected and preserved on 7 days post-exposure rather than 4 days post-exposure. Analytic Chemistry: Water samples were analyzed for DRO by ALS Global Laboratory in Fort Collins, Colorado. Extraction and analysis procedures were based on U.S.EPA methods SW-846, 8000C and 8015D. Briefly, samples were extracted in solvents (i.e. hexane and methylene chloride) with O-terphenyl added as surrogate and total DRO was detected by GC-FID. Concentrations were quantified by summing all C10-C28 hydrocarbons and comparing this value to a minimum of a 5-point calibration curve of DRO standards. Quality controls included a method blank, laboratory control samples, matrix spikes and duplicate samples. 55 �Statistical Analysis: Differences in macroinvertebrate drift, abundance and richness among treatments in mesocosm Exp1 were analyzed using 1-way analysis of variance in SAS 9.3 (PROC ANOVA) (SAS Institute Inc. Cary, North Carolina). If the overall F-statistic was statistically significant (p<0.05), Duncan’s multiple range test was used to determine differences among individual treatments. To compare responses of macroinvertebrate communities collected from Arkansas River stations AR1 and AR5 to diesel in mesocosm Exp2 we used 2-way ANOVA. To determine if effects of diesel varied between the 2 stations, we tested for diesel x station interaction effects. All data were log-transformed to meet assumptions of parametric statistics. Results and Discussion Analytic Chemistry: Diesel concentrations measured prior to the resumption of flow through conditions in the two experiments were 1-2 orders of magnitude lower than nominal concentrations (Tables 10 and 11). This was likely caused by absorption of diesel into biota (e.g., periphyton, microbes and insects), adsorption on surfaces (e.g., mesocosm walls, paddlewheels and substrate), volatilization as well as diesel’s low solubility in water. Interestingly, when comparing measured DRO concentrations between the two mesocosm experiments at the loading rate of 600mg/L diesel, Exp1 had more variability and higher concentrations than Exp2. This disparity was likely caused by the timing of when samples were collected. During Exp1, samples were collected approximately two hours after the spills were initiated, while samples in Exp2 were collected approximately four hours after the spills were initiated. As a result, Exp2 had lower measured DRO concentrations and less variability than Exp1, even though their loading rates were the same (600 mg/L diesel). For samples taken after 72 and 144 h during Exp1 and Exp2, DRO concentrations were uniformly below detection limits. However, diesel remained adsorbed to surfaces throughout each experiment. Although sediment concentrations were not measured, diesel odor was present, requiring the use of a VOC rated respirators during processing of aquatic insect communities at the end of each experiment. Therefore, the exposure scenario likely proceeded through three phases: 1) exposure from predominantly aqueous diesel, 56 �including pure diesel and the solubilized water accommodating fraction; 2) exposure from aqueous diesel and adsorbed diesel as well as from dietary uptake from contaminated biota; and 3) exposure from adsorbed diesel and from contaminated biota in the absence of aqueous exposures. Importantly, this replicated the expected phases of an actual petroleum spill in a river. Table 10: Diesel loading rates in Exp1 compared with measured aqueous DRO concentrations quantified by GC-FID. Aqueous concentrations were measured 2 hours post-exposure (preceding the resumption of flow through conditions) and at 72 hours post-exposure. Data are means of 3 replicates with SEM in parentheses. BDL stands for “below detection limits”. BQL stands for “below quantification limits” indicating DRO was detected, but detection occurred at such low concentrations that quantification was unreliable. Diesel Loading Rate 2-hours Post Exposure Measured Aqueous Diesel Concentration (SEM) 72-hours Post Exposure Measured Aqueous Diesel Concentration (SEM) 0 mg/L 75 mg/L 0.16 mg/L (0.08) BQL 0.72 mg/L (0.87) Not Measured Not Measured 150 mg/L 1.17 mg/L (0.14) Not Measured 300 mg/L 4.03 mg/L (0.44) BDL 600 mg/L 11.67 mg/L (4.67) BDL 1,200 mg/L 22.53 mg/L (7.24) BDL Table 11: Diesel loading rate in Exp2 compared with measured aqueous DRO concentrations quantified by GC-FID. Aqueous concentrations were measured 4 hours post exposure (preceding the resumption of flow through conditions) and at 144 hours post-exposure. Data are means of 4 replicates with SEM in parentheses. BDL stands for “below detection limits”. Diesel Loading Rate 4-hours Post Exposure Measured Aqueous Diesel Concentration (SEM) 144-hours Post Exposure Measured Aqueous Diesel Concentration (SEM) AR1 0 mg/L BDL Not Measured AR5 0 mg/L BDL Not Measured AR1 600 mg/L 2.95 mg/L (0.35) BDL AR5 600 mg/L 2.90 mg/L (0.62) BDL 57 �Aquatic Insect Drift: When aquatic insects migrate downstream to colonize new habitat or escape unfavorable conditions often their most efficient and quickest method is drifting in the water column (Brittain and Eikeland 1988). Although drift is a normal behavior in many species, numerous studies have demonstrated exposure to contaminants increases drift rates (Miller et al. 1986; Allan 1987; Wallace et al. 1989; Clements 2004; Araújo et al. 2014). Here, we explored how simulated diesel spills affected drift responses and we observed catastrophic aquatic insect drift shortly after spills were initiated. Exp1: Drift results were highly significant (p < 0.01) for Ephemeroptera, Trichoptera, Plecoptera and Diptera (Figure 15). The most common drifting insects were Baetis (order, Ephemeroptera) and Chironomidae (order, Diptera). Baetis have a high drift propensity even under normal conditions (Brittain and Eikeland 1988). However, their drift rates increased substantially as diesel loading increased. Chironomidae, do not typically drift to the same extent as Baetis under normal conditions, but do when exposed to organic pollutants like pesticides and petroleum (Miller et al 1986; Wallace et al. 1989). Plecopterans have also been observed readily drifting from diesel contamination (Miller et al 1986). Not only were drift results significant at higher diesel loading rates representing larger spill sizes, results were also significant at much lower diesel loading. For each taxonomic order discussed above, there were significant differences between controls and the lowest diesel loading rate (75 mg/L) (Figure 17). This is a striking result given that measured aqueous diesel concentrations for that treatment were in the parts per billion range at 0.72 mg/L (SEM 0.87). Clear concentration response relationships for Ephemeroptera and Trichoptera as well as for the proportion of drifting organisms were also apparent (Figure 18). 58 �Number Drifting per Mesocosm Number Drifting per Mesocosm 250 200 Ephemeroptera p < 0.0001 a ab 150 100 b 50 d c c 0 0 30 25 75 150 300 600 1200 a Trichoptera p < 0.001 a Plecoptera p < 0.01 a 300 a a a a b 0 250 20 75 150 300 600 1200 Diptera p < 0.001 a a 200 b 15 10 18 16 14 12 10 8 6 4 2 0 b ab 150 5 50 0 0 75 150 300 600 1200 a a 100 c 0 a b 0 75 150 300 600 1200 Diesel Loading Rate (mg/L) Figure 17: Effects of diesel exposure on number of drifting aquatic macroinvertebrates. Data are means of 3 replicates and error bars are SEMs. P-values indicate significance of an ANOVA calculation. Letters above histograms indicate significant differences among treatments. To estimate the proportion of aquatic insects drifting within a given treatment, drift data was normalized to the final benthic abundance from control treatments (discussed below). A proportion was calculated by dividing the number of drifting organisms by the total abundance of organisms from controls at the end of the experiment. These data provided an estimate of the fraction of organisms that drifted from the entire community. This procedure normalized the drift data to give context to the magnitude of the drift response in relation to the size of the aquatic insect community. Accordingly, less than 2% of insects drifted in control treatments. However, 59 �more than 30% drifted at the highest loading rate (1,200 mg/L diesel), and a strong concentration Esitmated Total Proportion Drifting per Mesocosm response relationship was observed across treatments (Figure 18). 0.4 p < 0.001 b bc 600 1200 0.3 0.2 ab a ab 75 150 0.1 a 0.0 0 300 Diesel Loading Rate (mg/L) Figure 18: Effects of diesel on the estimated proportion of total aquatic macroinvertebrates drifting. Data are means of 3 replicates and error bars are SEMs. P-values indicate significance of an ANOVA calculation. Letters above histograms indicate significant differences among treatments. Exp2: As in Exp1, diesel exposure significantly increased aquatic insect drift in Exp2. For Ephemeroptera, Trichoptera, Plecoptera and Diptera, exposure to a loading rate of 600 mg/L diesel caused a massive drift response, particularly within the genus Baetis (order, Ephemeroptera) and family Chironomidae (order, Diptera). Interestingly, the magnitude of effects for Plecopterans and Dipterans were dependent on whether they were collected from a reference community (AR1) or from a historically metal-polluted community (AR5). Significantly fewer Plecopterans and Dipterans drifted from AR5 than from AR1. For Plecoptera specifically, a significant site by treatment interaction was also detected, indicating that the effects of diesel on drift was greater for individuals collected from the reference site (AR1) compared with the historically metal-polluted site (AR5). 60 �Number Drifting per mesocosm Number Drifting per mesocosm 200 Ephemeroptera 150 T*** 80 Plecoptera 60 S* T*** SxT* 100 40 50 20 0 0 AR1 40 30 AR1 AR5 Trichoptera T*** 800 Diptera 600 S** T*** 20 400 10 200 0 0 AR1 AR5 Control Diesel AR1 AR5 AR5 Figure 19: Effects of diesel exposure on total number of drifting aquatic macroinvertebrates per mesocosm. One community (AR1) was collected from reference site dominated by taxa sensitive to metal pollution. The other (AR5) was collected from a historically metal polluted site that was dominated by more tolerant taxa. Data are means of 4 replicates and error bars are SEMs. P-values indicate significance of the main effects (site and diesel treatment) and the interaction effect. S =site effect, T = treatment effect and S x T = site by treatment interaction. Asterisks indicate the level of significance (* = P < 0.05; ** = P < 0.01: *** = P < 0.001). Using the same technique as described for Exp1, an estimate of the proportion of drifting aquatic insects was calculated to normalize drift results to the size of the communities. A site by treatment interaction effect was detected, indicating that the drift response from the community as a whole was more sensitive to diesel exposure at AR1 than at AR5 (Figure 20). Therefore, it appears that organisms in the AR1 community were more likely to drift to escape diesel exposure than organisms from a community previously stressed by metals (AR5). 61 �Figure 20: Effects of diesel exposure (600 mg/L) on the estimated proportion of total aquatic macroinvertebrates drifting per mesocosm. Responses of two distinct communities were compared. One community (AR1) was collected from reference site dominated by taxa sensitive to metal pollution. The other (AR5) was collected from a metal polluted site dominated by more tolerant taxa. Data are means and error bars are SEM of 4 replicates. Asterisks indicate the level of significance (* = P < 0.05; ** = P < 0.01: *** = P < 0.001). S =site effect, T = treatment effect and S x T = site by treatment interaction. Benthic Abundance: Abundance of Ephemeroptera, Plecoptera, Diptera, as well as total abundance were significantly reduced when exposed to diesel compared with controls, regardless of loading rate or experiment (Figures 21 and 22). However, Trichoptera (caddisflies) were not significantly affected in Exp1 as they were in Exp2. Catastrophe drift was a major factor in reduced abundances after diesel exposure, but drift did not fully explain reduced benthic abundances indicting that diesel exposure caused acute mortality in aquatic insects. These effects were observed across numerous taxa, including those generally considered tolerant to contaminants (e.g., Diptera). Exp1: Benthic abundances were significantly reduced for orders Ephemeroptera, Plecoptera and Diptera after 4 days across all diesel exposures (Figure 21). Interestingly, for most orders there was not an obvious concentration-response relationship. Instead diesel appeared to affect most groups in an all–or–none manner. When diesel exposure occurred, even 62 �at low loading rates, benthic abundances were adversely impacted similarly as observed at much higher loading rates. It was not clear why clear concentration response relationships were not clearly observed, especially when drift was strongly related to concentration. However, this may simply be the result of short-term exposure duration (4 days). It was possible that a concentration response relationship would have been observed over a longer time frame. Regardless, these results showed that low diesel loading rates were highly toxic to aquatic insects across several taxonomic orders. For example, the 75 mg/L treatment caused an approximate 50% reduction in total abundance of aquatic insects after 4 days (Figure 21). Trichoptera were the only dominant order that was not significantly affected by diesel; interestingly, caddisfly drift was significant. Exp2: After 7 days of exposure to simulated diesel spill conditions abundances of Ephemeroptera, Trichoptera, Plecoptera and Diptera were significantly reduced compared with control exposures (Figure 22). Ephemeroptera abundance also had a significant site by treatment interaction, indicating that mayflies from the metal pollution impacted site (AR5) were more sensitive to diesel exposure than individuals from the reference site (AR1). This apparent contradiction can be explained by the theory of Cost of Tolerance where individuals and communities of organisms that develop tolerance to a particular stressor (i.e., metals) become more susceptible to other, novel stressors (Kashian et al. 2007). For example, the benthic community at AR5, which is dominated by taxa that are tolerant to metals, has been shown to more susceptible to stonefly predation, UV-B light and low pH than the AR1 community (Clements 1999; Courtney and Clements 2000; Kashian et al. 2007). 63 �Abundance per Mesocosm 600 Ephemeroptera p < 0.0001 500 a ab b Abundance per Mesocosm b 300 30 b bc bc bc b c 20 200 c 10 100 0 80 Plecoptera p < 0.05 a 40 400 0 0 75 150 300 600 1200 Trichoptera 60 0 75 1000 Diptera p < 0.01 800 a 150 300 600 1200 600 b 40 400 20 b b b b 200 0 0 0 Abundance per Mesocosm 50 75 1600 1400 150 300 600 1200 25 Total abundance p = 0.0005 a 75 150 300 600 1200 Number of taxa 20 1200 1000 b 800 0 b 600 15 b b c 400 10 5 200 0 0 0 75 150 300 600 1200 0 75 150 300 600 1200 Diesel Loading Rate (mg/L) Figure 21: Effects of diesel exposure on benthic abundance of aquatic macroinvertebrates after conclusion of the experiment on Day-4. Data are means of 3 replicates and error bars are SEMs. P-values indicate significance of an ANOVA test; results were insignificant where no P-value was provided. Letters above histograms indicate significant differences among treatments. 64 �Number per mesocosm 300 Ephemeroptera 250 S*** 100 200 SxT* 80 150 60 100 40 50 20 0 0 T*** AR1 Number per mesocosm 120 Plecoptera 150 125 AR5 1200 Trichoptera S*** T*** 1000 100 800 75 600 50 400 25 200 0 Control Diesel S*** T*** AR1 AR5 AR1 AR5 Diptera S** T*** 0 AR1 AR5 Figure 22: Effects of diesel exposure (600 mg/L) on insect abundance after seven days of exposure. Responses of two distinct communities were compared. One community (AR1) was collected from reference site dominated by taxa sensitive to metal pollution. The other (AR5) was collected from a metal polluted site dominated by more tolerant taxa. Data are means and error bars are SEM of four replicates. Asterisks indicate the level of significance (* = P < 0.05; ** = P < 0.01: *** = P < 0.001). S =site effect, T = treatment effect and S x T = site by treatment interaction. Conclusions Experimental diesel spills had highly significant impacts on aquatic insect communities across all simulated spill sizes. Catastrophic drift occurred in a concentration-dependent and a consistent manner immediately after exposure. Interestingly, generally tolerant taxa such as Diptera drifted similarly to sensitive taxa like Baetis, suggesting that the traditional model of tolerance to contaminants was not valid for petroleum hydrocarbon exposure. Moreover, profound effects were detected in aquatic insect abundance after diesel exposure, regardless of 65 �whether exposure duration was four days or seven. Surprisingly, a low concentration of diesel (75 mg/L loading rate) caused an approximate 50% reduction in abundance after four days of exposure. Although bioassays with Rainbow Trout showed no mortality at 75 mg/L after four days, Rainbow Trout heavily rely on aquatic insects for a food source. Therefore, even if no trout mortality occurred due to direct petroleum exposure, the long-term survival and fitness of fish populations may be affected by loss of prey resources. Importantly, this response is rarely considered when negotiating settlements because typically only direct fish mortality is used to hold responsible parties accountable for damaging a fishery. However, the indirect effects of even small spills may be much greater than the initial fish-kill. Aquatic insects provide a diversity of critical ecosystem services, connect aquatic and riparian habitats, and link microscopic organisms with higher trophic levels in aquatic food chains. Section 4: Final Remarks and Recommendations Inference of Causation By integrating field observations, mesocosms and bioassays we gained insights into consequences of petroleum spills using a holistic, interdisciplinary approach. Moreover, by combining our results with evidence gathered from other sources, such as the Bureau of Land Management’s pre/post-spill aquatic macroinvertebrate community data, Colorado Parks and Wildlife’s acute fish-kill assessments and their biannual, post-spill fish population surveys, the insights become more powerful (BLM 2011; BLM 2013; CPW 2016). An aggregated, holistic, synthesis of evidence was required to discern petroleum impacts. For example, by Fall 2015, nearly three years after the spill, chemical analysis demonstrated that West Creek water and sediments were similar to what we would expect before the spill. In other words, sediment and water chemistry had returned to ‘normal’. However, fish populations had not recovery from the spill, aquatic insect communities were severely degraded at the spill site, and pathological abnormalities were observed in Mottled Sculpin at the spill site and downstream. We conclude that the West Creek petroleum spill caused adverse acute, chronic, lethal and sub-lethal effects to 66 �aquatic communities across numerous levels of biological organization. Below we discuss two methods for establishing causal inference to support our conclusion. Empirical Weight-of-Evidence Approach: The empirical weight-of-evidence approach employed in our study integrates analytic chemistry with biological field assessments and laboratory toxicity tests (Figure 23). Its intention is to 1) determine what compounds are in the environment and at what concentrations; 2) identify environmental effects based on field assessments at reference and contaminated sites; and 3) employ controlled experiments to establish a coherent relationship between the contaminants and the adverse effects observed in the field. Figure 23: A diagram representing the intersection of the procedures involved in establishing causal inference by the empirical weight of evidence approach. 1 – Analytic Chemistry: In this report, we discussed shortcomings of the analytic chemical data collected at West Creek. Immediately following the spill, only volatile compounds associated with gasoline were analyzed in water and sediment. Persistent diesel range compounds were apparently omitted from analysis by federal contractors. Our analysis in 2015 failed to detect elevated concentrations of petroleum hydrocarbons in West Creek more than two 67 �years after the spill. However, we know that petroleum sheen and odor were observed by Colorado Parks and Wildlife nearly a month after the spill, suggesting persistence of petroleum at the spill site. 2 – Biological Assessments (Field): Approximately 10,000 fish died during and immediately after the spill occurred and fish populations have not since recovered by Fall 2016 (CPW 2016). Also in Fall 2015, two young mottled sculpin, born after the spill, were found with suspected birth defects. Older scuplin, which were thought to be exposed to chronic petroleum contamination, had indications of chronic environmental stress (e.g., increased SMAs) at sites downstream from the spill. Additionally, in 2015 aquatic macroinvertebrate communities were very different at the spill site and downstream compared with the upstream reference site. The spill site was dominated by tolerant taxa (e.g., Oligochaeta), and the reference site was the only site with high densities of Plecoptera which are known to be particularly sensitive to petroleum contamination and are slow to recolonize after environmental conditions improve. 3 – Toxicology (Laboratory): Although it was impossible for short-term laboratory experiments to verify all of the chronic effects observed in West Creek (i.e., birth defects in sculpin, proliferation in Oligochaeta), associations between field and lab responses were evident. Across all experiments, measured petroleum hydrocarbon concentrations were orders of magnitude lower than the spill loading rates. Also, simulated diesel spills caused acute mortality to fish at concentrations that were substantially lower than what would have been expected during the 2013 West Creek spill. Bioassays also confirmed that, even at low concentrations where mortality did not occur, diesel was bioavailable and biochemically interacted with liver tissue, leading to altered behavior and apparent sublethal injury. Periphyton, a fundamental food source for stream communities, was adversely impacted by exposure at environmentally relevant petroleum concentrations. For aquatic insect communities, exposure to simulated spills in mesocosms caused catastrophic drift and massive mortality. This would have opened niches for colonization by tolerant taxa like Oligochaeta. Plecopterans, which were mostly absent from sites impacted by the West Creek spill, were found to be among the most sensitive insects to petroleum exposure during mesocosm experiments. 68 �Bradford Hill’s Criteria for Causation: Another method for supporting causal inference is Bradford Hill’s Criteria for Causation (Hill 1965). Below, we follow our conclusion (“the West Creek petroleum spill caused adverse acute, chronic, lethal and sub-lethal effects to aquatic communities across numerous levels of biological organization”) through Hill’s Criteria. According to this method, a causal relationship can be presumed depending on how the following principles are satisfied: 1) Strength of Association – Is a statistical or obvious correlation apparent between a suspected cause and its effects?  Yes. Strong statistical effects were observed in insect communities immediately downstream from the West Creek spill site compared with other sites. Mottled Sculpin collected from downstream sites had significantly increased SMA’s, indicating increased environmental stress at locations downstream from the spill. Moreover, the trout fishery in West Creek had not recovered by Fall 2016, nearly four years after the petroleum spill. 2) Constancy of Effects – Have similar effects been observed on multiple, separate occasions when a similar cause was suspected?  Yes. Numerous researchers have document reduced insect abundance and fish-kills after petroleum spills (Bury 1972; Harrel 1985; Crunkilton et al, 1990; Lytle et al. 2001; Hoffman et al. 2002; Smith et al. 2010). Chronically depressed fish populations have been observed after spills (Hoffman et al. 2002). Oligochaeta dominated benthic macroinvertebrate communities have been documented after petroleum spills (McCauley 1966; Harrel 1985; Guiney 1987; Smith et al. 2010). Increases in SMAs have been observed in fish exposed to petroleum (Khan et al. 1984; Ali et al 2014). Increased incidence birth defects have been observed after petroleum exposure (Roberts et al. 1989; Hoffman et al. 2002; Peterson et al. 2003). 3) Specificity of Effects – Are observed effects particular to the suspected cause?  No. Effects observed in West Creek cannot be specifically and singularly linked to a particular contaminant or mixture of contaminants. The complex and dynamic chemical composition of petroleum hydrocarbons and the non- 69 �specific toxicity of PAHs do not satisfy this principle. However, effects did specifically manifest at locations impacted by the spill. 4) Temporarily – Did the suspected cause precede its effects?  Yes. We know that the fish in West Creek were alive before the spill occurred and died immediately after. We know that fish populations have not recovered in West Creek. We know that in 2011, Plecoptera were present at the spill site and Oligochaeta absent. However, by August 2013, the opposite was true. 5) Biological Gradient – Is there a concentration response relationship between the suspected cause and its effects?  Yes. Effects on aquatic insect communities were strongest at the spill site compared with other downstream sites. Sublethal effects in Mottled Sculpin were strongest at downstream sites where surviving fish were exposed to higher concentrations of petroleum for longer durations than fish at upstream sites. Laboratory and mesocosm experiments also demonstrated strong concentration-response relationships. 6) Plausibility – Do reasonable mechanisms between a suspected cause and its effects exist?  Yes. PAHs have been mechanistically linked to free radical production, and DNA or protein adduct formation (Aas et al. 2000; Livingstone 2001). Each of these have been linked to numerous diseases involving cancers, liver and kidney dysfunction, increases in developmental or birth-defects and death (Livingstone 2001; Luschak 2007). Aliphatic hydrocarbons have also been mechanistically linked to irritation, death and habitat destruction by physical smothering (Connell et al., 1981; Crunkilton et al. 1990; Peterson et al. 2003). 7) Coherence – Does a suspected cause and its effects conflict with science and facts?  No. Effects do not conflict with current knowledge of biology, chemistry or physics. 8) Experimentation – Do controlled experiments support the conclusion of causality?  Yes. Controlled experiments, including those undertaken in the project support the conclusion that petroleum spills in coldwater streams cause acute, chronic, lethal and sub-lethal effects to aquatic communities across numerous 70 �levels of biological organization (Khan 1999; Hoffman et al. 2002; Peterson et al. 2003; Beyer et al 2016). 9) Analogy – Have similar relationships been established as causal?  Yes. A recent surge of excellent research has developed out of examining effects of the Deepwater Horizon oil spill in the Gulf of Mexico. Researchers have linked exposure from petroleum contamination to a breadth of acute, chronic, lethal and sublethal effects. See Beyer et al. (2016) for a review. Additionally, similar effects were observed after the Exxon Valdez Oil Spill of the Coast of Alaska. See Peterson et al. (2003) for a review. Lessons Learned: Final Recommendations for Managers and Stakeholders In the following section, we supply a list of recommendations for improving petroleum spill related bioassessments of stream communities as well as preparedness for spill events. Here, our goal is to provide natural resource managers with suggestions for improving their existing preparedness and response procedures. We also seek to help natural resource managers, as well as stakeholders, establish causal inferences for petroleum spill related lethal, sublethal, chronic and acute effects on stream communities. However, it is important to keep in mind that we are not lawyers, this is not an exhaustive list and items discussed below are not in order of importance. This is merely a jumping off point for future planning. 1. Prioritize non-game fish during population surveys the same way game fish are. Often, small, non-game fish (i.e., Mottled Sculpin and Speckled Dace) are counted in a qualitative (present/absent or rare/plentiful) manner during routine population surveys rather than as quantitative population estimates. However, when recovering funds from a responsible party in Colorado, non-game fish ($100 per dead fish) are worth more than game fish ($35 per dead fish). Establishing quantitative data on non-game fish can be a powerful tool when making fish-kill mortality estimates, especially because small bodied, non-game fish can be difficult to detect when dead as they easily get wedged between rocks and are obscured from view. 71 �2. After a spill, it is typical to conduct multiple fish-kill assessments over several days to more completely characterize the magnitude of a fish-kill event. This is important, because fish will not necessarily die immediately or in unison. Moreover, the quantity of dead fish will likely be the most important factor in recovering funds from the responsible party. 3. Aquatic insects are a critical food source for economically important fish and riparian fauna. They are also critical to instream ecosystem services (i.e., organic decomposition, nutrient cycling, mitigation of potentially harmful proliferations of microbes or algae). Many aquatic insects are also very sensitive to petroleum contamination. However, they are not routinely monitored after spills. a. Deploy drift nets upstream and downstream soon after a spill to quantify how a petroleum spill may impact insect emigration from the affected area. b. Use a quantitative aquatic macroinvertebrate sampler (e.g., Hess or Surber), not a qualitative sampler (e.g., seine or kick net), to monitor changes in community structure over time. 4. The Benthotorch is a convenient way to monitor how a petroleum spill affects the biomass of periphyton, a critical stream food source. Monitoring periphyton biomass over time may provide useful bioassessment data after a spill. Note that the Benthotorch does not efficiently measure thick mats of periphyton. 5. When collecting biological samples to assess sublethal and chronic effects in fish after a petroleum spill, collect samples that are indicators of exposure and others that are indicators of injury. Rarely, will a single assay indicate both injury and exposure. a. Indicators of exposure to petroleum hydrocarbons include: i. Cytochrome p450 gene upregulation ii. Increased EROD activity in liver iii. PAH metabolites in bile iv. PAHs accumulation in bivalve mollusks tissue b. Indicators of injury include: 72 �i. Histological abnormalities 1. Gill damage 2. Skin irritation 3. Liver/Kidney damage 4. Spleenic melanomacrophage aggregates (medium to long term) 5. Cancers and congenital effects (long term to mutigenerational) ii. Hematological abnormalities: 1. Altered blood plasma panel 2. Proliferation of white blood cell granulocytes iii. Condition factor and organo-somatic indices iv. Pathologies found by necropsy 6. Less mobile and more resident species of fish (i.e., Mottled Sculpin) are good choices to sample for sublethal effects because their health would likely reflect environmental conditions at the sampling location rather than another location. 7. Samples that can be archived (e.g., blood films and histology biopsy’s, aquatic insect communities, preserved fish tissue) are especially useful because they can be stored after analysis. Therefore, if the validity of the results were ever at issue the samples could be reexamined by a third-party. 8. Non-lethal sampling methods can include some biopsies for histology as well as blood draws for blood films and plasma panels. This may be important for characterizing health in threatened or endangered species. 9. Determine bodies of water at highest risk of a spill and develop baseline data in these watersheds. Baseline data should include sampling methods that would be employed in a petroleum spill response and bioassessment scenario so that ‘before spill’ and ‘after spill’ results can be compared. Areas at high risk include winding roads adjacent to water bodies with heavy traffic and high speed limits, areas where trains run adjacent to water 73 �bodies, locations where pipelines cross water, areas where flooding might compromise hydraulic fracturing sites and locations where spills have occurred in the past. 10. Stay safe. Petroleum fumes are toxic to humans and are extremely flammable. Do not risk injury or death by collecting samples when the scene is dangerous. If odor is detected, use a volatile organic hydrocarbon rated respirator. Responders also risk exposure through skin contact. Wear personal protective equipment and use common sense when responding to a spill. 11. Relevant and useable data collection is often time sensitive. Data must be collected properly and quickly after a spill occurs. 12. Develop a communications pipeline to facilitate timely, accurate reporting of spills from government agencies, witnesses, stakeholders and responsible parties to responders Consider an incentives program. 13. Set money aside in an emergency fund which would be available to pursue bioassessment and analytical procedures after a spill occurs. Settling with a responsible party requires time, but the data needed to hold that party fully responsible are often costly and time sensitive. Therefore, funds for data/sample collection and analysis needs to be available prior to negotiating with the responsible party. 14. Develop a spill response and sample collection training program. This will greatly increase the speed, accuracy and confidence with which samples are collected. 15. Follow chain-of-custody protocols when collecting samples. 16. Develop and distribute spill preparedness and sample collection kits. See Schmitt et al. (1999) for an example. 74 �17. Develop relationships with professionals who would process collected samples such veterinarians or pathologists for histology and hematology, analytic chemists for contaminant quantification and identification, as well as biochemists for enzyme analysis. 18. Develop relationships with attorneys who would potentially represent you during future petroleum spill related claims. Discuss your weaknesses in bioassessments, your past shortcoming or success in negotiations with responsible parties, your goals for future settlements, and how to determine best practices for applying field assessments to legal proceedings. Discuss ways to settle disputes without providing the responsible party immunity if chronic or more subtle impacts are discovered after settlement. 19. Use lessons learned from West Creek and from other spills to inform your settlement objectives. For example, in West Creek, the responsible party was fined for an acute fishkill (approximately 10,000 dead fish). They were not fined for 4+ years of a lost fishery or the dramatic loss of ecosystem services that resulted from the impacts to benthic communities. 20. During a spill response take reasonable action to mitigate damages (i.e., deploy numerous sorbent booms and frequently replace them, remove petroleum saturated debris, etc.). However, take care not to exacerbate damages to natural resources. For example, chemical dispersants often increase the acute toxicity of petroleum. The use of chemical flame retardants as fire suppressants may contribute persistent contaminants to the aquatic environment. Widespread agitation of the benthos (e.g., raking or dredging) used as a cleanup procedure often disproportionately harms the stream. 21. Do not necessarily rely on Federal analytic chemistry guidelines, or chemical analysis conducted by Federal contractors without adequate consideration and context. For example, after the 2013 West Creek spill, only volatile, highly ephemeral petroleum hydrocarbon compounds associated with gasoline were quantified in water and sediment, and those results were made publicly available after the spill. However, much more persistent compounds associated with spilled diesel were not analyzed or made publicly 75 �available. Therefore, according to data released by federal contractors there was an appearance of “remediation effectiveness and completeness” in West Creek as determined by an absence of volatile, gasoline range petroleum hydrocarbons shortly after the spill. This conclusion was reached despite reports from CPW staff who observed petroleum sheen and odor at the spill site. Residual petroleum was likely in the diesel range and would not have been detected by a gasoline range analysis of VOCs. In addition, PAHs are generally considered to be the most toxic constituents within diesel and crude oil’s petroleum hydrocarbon mixture. However, only 16 PAH compounds, all of which are parent PAHs, are listed as U.S.EPA priority pollutants. Thus, these 16 parent PAHs are most likely to be quantified at sites contaminated by petroleum. Unfortunately, alkylated PAHs are far more common in petroleum than parent PAHs. These alkylated PAHs would not be identified in contaminated water or sediment unless sophisticated (and expensive) analytic procedures were utilized such as U.S.EPA method 8270D. The total PAH concentrations (parent PAHs plus alkylated PAHs) would be substantial higher than the concentrations revealed by testing for only the 16 parent PAHs on the U.S.EPA’s priority list. For the reasons outlined above it is worth closely examining any analytic chemistry data from contaminated sites. It is also worth considering personally collecting chemistry samples and having them analyzed by a third-party. Additionally, it is important to keep in mind that petroleum measured in water and sediment after a spill does not necessarily reflect the concentration organisms were exposed to during the spill. 22. After a spill, consider the use of active biomonitoring to assess stream health and recovery over time. This could include a mark and recapture program of surviving, or stocked, fish which employs collecting non-lethal samples and data (e.g., blood draws for blood film and plasma panels as well as condition factor data, etc.) to monitor fish health and habitat occupancy. A more intensive program could include physically transplanting marked fish from a location upstream of the spill to a location immediately downstream of the spill and vice versa. 76 �References Aas, E.; Baussant, T.; Balk, L.; Liewenborg, B.; Andersen, O. K., PAH METABOLITES IN BILE, CYTOCHROME P4501A AND DNA ADDUCTS AS ENVIRONMENTAL RISK PARAMETERS FOR CHRONIC OIL EXPOSURE: A LABORATORY EXPERIMENT WITH ATLANTIC COD. Aquatic Toxicology 2000, 51 (2), 241-258. Agius, C.; Roberts, R. J., MELANO-MACROPHAGE CENTRES AND THEIR ROLE IN FISH PATHOLOGY. Journal of Fish Diseases 2003, 26(9), 499-509. Albers, P. H.; Gay M. L., UNWEATHERED AND WEATHERED AVIATION KEROSINE: CHEMICAL CHARACTERIZATION AND EFFECTS ON HATCHING SUCCESS OF DUCK EGGS." Bulletin of environmental contamination and toxicology 1982, 28 (4), 430-434. Ali, A. O.; Hohn, C.; Allen, P. J.; Ford, L.; Dail, M. B.; Pruett, S.; Petrie-Hanson, L., THE EFFECTS OF OIL EXPOSURE ON PERIPHERAL BLOOD LEUKOCYTES AND SPLENIC MELANO-MACROPHAGE CENTERS OF GULF OF MEXICO FISHES. Marine Pollution Bulletin 2014, 79 (1-2), 87-93. Al-Kindi, A. Y. A.; Brown J. A.; Waring C. P. ENDOCRINE, PHYSIOLOGICAL AND HISTOPATHOLOGICAL RESPONSES OF FISH AND THEIR LARVAE TO STRESS WITH EMPHASIS ON EXPOSURE TO CRUDE OIL AND VARIOUS PETROLEUM HYDROCARBONS. Sultan Qaboos University Journal for Science [SQUJS] 2000, 5, 130. Allan, J. D., MACROINVERTEBRATE DRIFT IN A ROCKY-MOUNTAIN STREAM. Hydrobiologia 1987, 144 (3), 261-268. Anderson, J. W. "RESPONSES TO SUBLETHAL LEVELS OF PETROLEUM HYDROCARBONS: ARE THEY SENSITIVE INDICATORS AND DO THEY CORRELATE WITH TISSUE CONCENTRATION." Fate and Effects of Petroleum Hydrocarbons in Marine Organisms and Ecosystems. D. A. Wolfe, Ed., Pergammon Press 1977, 95-114. Anderson, J. W.; Riley, R.; Kiesser, S.; Gurtisen, J., "TOXICITY OF DISPERSED AND UNDISPERSED PRUDHOE BAY CRUDE OIL FRACTIONS TO SHRIMP AND FISH." American Petroleum Institute International Oil Spill Conference 1987, 1. Andrewartha, S. J.; Munns, S. L.; Edwards, A., CALIBRATION OF THE HEMOCUE POINTOF-CARE ANALYSER FOR DETERMINING HAEMOGLOBIN CONCENTRATION IN A LIZARD AND A FISH. Conservation Physiology 2016, 4, 6. API, OPTIONS FOR MINIMIZING ENVIRONMENTAL IMPACTS OF INLAND SPILL RESPONSE. 2016 http://www.oilspillprevention.org/~/media/Oil-SpillPrevention/spillprevention/r-and-d/inland/options-for-minimizing-e20161228t134857.pdf (Technical Report) Araujo, C. V. M.; Moreira-Santos, M.; Sousa, J. P.; Ochoa-Herrera, V.; Encalada, A. C.; Ribeiro, R., CONTAMINANTS AS HABITAT DISTURBERS: PAH-DRIVEN DRIFT BY ANDEAN PARAMO STREAM INSECTS. Ecotoxicology and Environmental Safety 2014, 108, 89-94. Baca, B. J.; Getter C. D., FRESHWATER OIL SPILL CONSIDERATIONS: PROTECTION AND CLEANUP. American Petroleum Institute International Oil Spill Conference 1985 1. 77 �Baker, E. A., CHEMISTRY AND MORPHOLOGY OF PLANT EPICUTICULAR WAXES." Linnean Society symposium series. 1982. Barnett, J.; Toews, D., EFFECTS OF CRUDE-OIL AND DISPERSANT, OILSPERSE 43, ON RESPIRATION AND COUGHING RATES IN ATLANTIC SALMON (SALMOSALAR). Canadian Journal of Zoology-Revue Canadienne De Zoologie 1978, 56 (2), 307-310. Baxter, C. V.; Fausch, K. D.; Saunders, W. C., TANGLED WEBS: RECIPROCAL FLOWS OF INVERTEBRATE PREY LINK STREAMS AND RIPARIAN ZONES. Freshwater Biology 2005, 50 (2), 201-220. Beasley, G.; Kneale, P., REVIEWING THE IMPACT OF METALS AND PAHS ON MACRO INVERTEBRATES IN URBAN WATERCOURSES. Progress in Physical Geography 2002, 26 (2), 236-270. Beyer, J.; Trannum, H. C.; Bakke, T.; Hodson, P. V.; Collier, T. K., ENVIRONMENTAL EFFECTS OF THE DEEPWATER HORIZON OIL SPILL: A REVIEW. Marine Pollution Bulletin 2016, 110 (1), 28-51. Blaxhall, P. C.; Daisley, K. W., ROUTINE HEMATOLOGICAL METHODS FOR USE WITH FISH BLOOD. Journal of Fish Biology 1973, 5 (6), 771-781. Blazer, V. S.; Fournie, J. W.; WeeksPerkins, B. A., MACROPHAGE AGGREGATES: BIOMARKER FOR IMMUNE FUNCTION IN FISHES? Environmental Toxicology and Risk Assessment: Modeling and Risk Assessment (Sixth Volume) 1997, 131, 360375. Bott, T. L.; Rogenmuser, K., EFFECTS OF NO 2 FUEL OIL, NIGERIAN CRUDE-OIL, AND USED CRANKCASE OIL ON ATTACHED ALGAL COMMUNITIES - ACUTE AND CHRONIC TOXICITY OF WATER-SOLUBLE CONSTITUENTS. Applied and Environmental Microbiology 1978, 36 (5), 673-682. Braddock, J. F.; Lindstrom, J. E.; Brown, E. J., DISTRIBUTION OF HYDROCARBONDEGRADING MICROORGANISMS IN SEDIMENTS FROM PRINCE-WILLIAMSOUND, ALASKA, FOLLOWING THE EXXON-VALDEZ OIL-SPILL. Marine Pollution Bulletin 1995, 30 (2), 125-132. Brette, F.; Machado, B.; Cros, C.; Incardona, J. P.; Scholz, N. L.; Block, B. A., CRUDE OIL IMPAIRS CARDIAC EXCITATION-CONTRACTION COUPLING IN FISH. Science 2014, 343(6172), 772-776. Brittain, J. E.; Eikeland, T. J., INVERTEBRATE DRIFT - A REVIEW. Hydrobiologia 1988, 166 (1), 77-93. Brown, J. N.; Peake, B. M., SOURCES OF HEAVY METALS AND POLYCYCLIC AROMATIC HYDROCARBONS IN URBAN STORMWATER RUNOFF. Science of the Total Environment 2006, 359 (1-3), 145-155. Burk, C. John. A FOUR YEAR ANALYSIS OF VEGETATION FOLLOWING AN OIL SPILL IN A FRESHWATER MARSH. Journal of Applied Ecology 1977, 515-522. Bury, R. B., EFFECTS OF DIESEL FUEL ON A STREAM FAUNA. California Fish and Game 1972, 58 (4), 291-295. Bleyl, D. W. R., IARC MONOGRAPHS ON THE EVALUATION OF CARCINOGENIC RISKS TO HUMANS. VOL. 47. SOME ORGANIC SOLVENTS, RESIN MONOMERS AND RELATED COMPOUNDS, PIGMENTS AND OCCUPATIONAL EXPOSURES IN PAINT MANUFACTURE AND PAINTING. Herausgegeben von WHO und IARC. 535 Seiten IARC, Lyon 1989. Preis: 85,Sw.fr. Food/Nahrung 1990, 34 (10), 952-952. 78 �Bureau of Land Management. WEST CREEK BENTHIC MACROINVERTEBRATE DATA. 2011. (unpublished). Bureau of Land Management. WEST CREEK BENTHIC MACROINVERTEBRATE REPORT. Prepared by Miller. S., and Judson. S., 2013. (unpublished) Cadmus, P.; Clements, W. H.; Williamson, J. L.; Ranville, J. F.; Meyer, J. S.; Gines, M. J. G., THE USE OF FIELD AND MESOCOSM EXPERIMENTS TO QUANTIFY EFFECTS OF PHYSICAL AND CHEMICAL STRESSORS IN MINING-CONTAMINATED STREAMS. Environmental Science & Technology 2016, 50 (14), 7825-7833. Campbell, T. W., EXOTIC ANIMAL HEMATOLOGY AND CYTOLOGY, 4th Edition. Exotic Animal Hematology and Cytology, 4th Edition 2015, 1-402. Carls, M. G.; Marty, G. D.; Meyers, T. R.; Thomas, R. E.; Rice, S. D., EXPRESSION OF VIRAL HEMORRHAGIC SEPTICEMIA VIRUS IN PRESPAWNING PACIFIC HERRING (CLUPEA PALLASI) EXPOSED TO WEATHERED CRUDE OIL. Canadian Journal of Fisheries and Aquatic Sciences 1998, 55 (10), 2300-2309. Chambers, J. E.; Heitz, J. R.; McCorkle, F. M.; Yarbrough, J. D., ENZYME-ACTIVITIES FOLLOWING CHRONIC EXPOSURE TO CRUDE-OIL IN A SIMULATED ECOSYSTEM .1. AMERICAN OYSTERS AND BROWN SHRIMP. Environmental Research 1979, 20 (1), 133-139. Chambers, J. E.; Heitz, J. R.; McCorkle, F. M.; Yarbrough, J. D., ENZYME-ACTIVITIES FOLLOWING CHRONIC EXPOSURE TO CRUDE-OIL IN A SIMULATED ECOSYSTEM .2. STRIPED MULLET. Environmental Research 1979, 20 (1), 140-147. Clements, W. H., METAL TOLERANCE AND PREDATOR-PREY INTERACTIONS IN BENTHIC MACROINVERTEBRATE STREAM COMMUNITIES. Ecological Applications 1999, 9 (3), 1073-1084. Clements, W. H., SMALL-SCALE EXPERIMENTS SUPPORT CAUSAL RELATIONSHIPS BETWEEN METAL CONTAMINATION AND MACROINVERTEBRATE COMMUNITY RESPONSES. Ecological Applications 2004, 14 (3), 954-967. Clements, W. H.; Cadmus, P.; Brinkman, S. F., RESPONSES OF AQUATIC INSECTS TO CU AND ZN IN STREAM MICROCOSMS: UNDERSTANDING DIFFERENCES BETWEEN SINGLE SPECIES TESTS AND FIELD RESPONSES. Environmental Science & Technology 2013, 47 (13), 7506-7513. Clements, W. H.; Vieira, N. K. M.; Church, S. E., QUANTIFYING RESTORATION SUCCESS AND RECOVERY IN A METAL-POLLUTED STREAM: A 17-YEAR ASSESSMENT OF PHYSICOCHEMICAL AND BIOLOGICAL RESPONSES. Journal of Applied Ecology 2010, 47 (4), 899-910. Colorado Parks and Wildlife, WEST CREEK. Prepared by Gardunio, E., 2016. (Internal Report) Connell, D. W.; Miller, G. J., PETROLEUM-HYDROCARBONS IN AQUATIC ECOSYSTEMS - BEHAVIOR AND EFFECTS OF SUBLETHAL CONCENTRATIONS .2. Crc Critical Reviews in Environmental Control 1981, 11 (2), 105-162. Courtney, L. A.; Clements, W. H., SENSITIVITY TO ACIDIC PH IN BENTHIC INVERTEBRATE ASSEMBLAGES WITH DIFFERENT HISTORIES OF EXPOSURE TO METALS. Journal of the North American Benthological Society 2000, 19 (1), 112-127. Courtney, L. A.; Clements, W. H., ASSESSING THE INFLUENCE OF WATER AND SUBSTRATUM QUALITY ON BENTHIC MACROINVERTEBRATE 79 �COMMUNITIES IN A METAL-POLLUTED STREAM: AN EXPERIMENTAL APPROACH. Freshwater Biology 2002, 47 (9), 1766-1778. Cox, O. N.; Clements, W. H., AN INTEGRATED ASSESSMENT OF POLYCYCLIC AROMATIC HYDROCARBONS (PAHS) AND BENTHIC MACROINVERTEBRATE COMMUNITIES IN ISLE ROYALE NATIONAL PARK. Journal of Great Lakes Research 2013, 39 (1), 74-82. Crunkilton, R. L.; Duchrow, R. M., IMPACT OF A MASSIVE CRUDE-OIL SPILL ON THE INVERTEBRATE FAUNA OF A MISSOURI OZARK STREAM. Environmental Pollution 1990, 63 (1), 13-31. Curl, H. C.; O'Donnell, K., CHEMICAL AND PHYSICAL PROPERTIES OF REFINED PETROLEUM PRODUCTS. US Department of Commerce, National Oceanic and Atmospheric Administration, Environmental Research Laboratories No. ERL MESA-17, 1977. de Soysa, T. Y.; Ulrich, A.; Friedrich, T.; Pite, D.; Compton, S. L.; Ok, D.; Bernardos, R. L.; Downes, G. B.; Hsieh, S.; Stein, R.; Lagdameo, M. C.; Halvorsen, K.; Kesich, L. R.; Barresi, M. J. F., MACONDO CRUDE OIL FROM THE DEEPWATER HORIZON OIL SPILL DISRUPTS SPECIFIC DEVELOPMENTAL PROCESSES DURING ZEBRAFISH EMBRYOGENESIS. Bmc Biology 2012, 10, 24. Dubansky, B.; Whitehead, A.; Miller, J. T.; Rice, C. D.; Galvez, F., MULTITISSUE MOLECULAR, GENOMIC, AND DEVELOPMENTAL EFFECTS OF THE DEEPWATER HORIZON OIL SPILL ON RESIDENT GULF KILLIFISH (FUNDULUS GRANDIS). Environmental Science & Technology 2013, 47 (10), 50745082. Edmunds, R. C.; Gill, J. A.; Baldwin, D. H.; Linbo, T. L.; French, B. L.; Brown, T. L.; Esbaugh, A. J.; Mager, E. M.; Stieglitz, J.; Hoenig, R.; Benetti, D.; Grosell, M.; Scholz, N. L.; Incardona, J. P., CORRESPONDING MORPHOLOGICAL AND MOLECULAR INDICATORS OF CRUDE OIL TOXICITY TO THE DEVELOPING HEARTS OF MAHI MAHI. Scientific Reports 2015, 5, 18. Eisler, R., HANDBOOK OF CHEMICAL RISK ASSESSMENT: HEALTH HAZARDS TO HUMANS, PLANTS, AND ANIMALS. CRC Press, 2000. Ellsaesser, C. F.; Miller, N. W.; Cuchens, M. A.; Lobb, C. J.; Clem, L. W., ANALYSIS OF CHANNEL CATFISH PERIPHERAL-BLOOD LEUKOCYTES BY BRIGHT-FIELD MICROSCOPY AND FLOW-CYTOMETRY. Transactions of the American Fisheries Society 1985, 114 (2), 279-285. Engelhardt, F. R., PETROLEUM EFFECTS ON MARINE MAMMALS. Aquatic Toxicology 1983, 4 (3), 199-217. Federle, T. W.; Vestal, J. R.; Hater, G. R.; Miller, M. C., EFFECTS OF PRUDHOE BAY CRUDE-OIL ON PRIMARY PRODUCTION AND ZOOPLANKTON IN ARCTIC TUNDRA THAW PONDS. Marine Environmental Research 1979, 2 (1), 3-18. Gough, M. A.; Rowland, S. J., CHARACTERIZATION OF UNRESOLVED COMPLEXMIXTURES OF HYDROCARBONS IN PETROLEUM. Nature 1990, 344 (6267), 648650. Guiney, P. D.; Sykora, J. L.; Keleti, G., ENVIRONMENTAL-IMPACT OF AN AVIATION KEROSENE SPILL ON STREAM WATER-QUALITY IN CAMBRIA COUNTY, PENNSYLVANIA. Environmental Toxicology and Chemistry 1987, 6 (12), 977-988. 80 �Harrel, R. C., EFFECTS OF A CRUDE-OIL SPILL ON WATER-QUALITY AND MACROBENTHOS OF A SOUTHEAST TEXAS STREAM. Hydrobiologia 1985, 124 (3), 223-228. Hauschildtlillge, D., LONG-TERM EFFECTS OF PETROLEUM-HYDROCARBONS ON THE LIFE-CYCLE AND PRODUCTIVITY OF THE LITTORAL OLIGOCHAETE LUMBRICILLUS-LINEATUS. Netherlands Journal of Sea Research 1982, 16 (DEC), 502-510. Hedtke, S. F.; Puglisi, F. A., SHORT-TERM TOXICITY OF FIVE OILS TO FOUR FRESHWATER SPECIES. Archives of environmental contamination and toxicology 1982, 11 (4), 425-430. Heras, H.; Ackman, R. G.; Macpherson, E. J., TAINTING OF ATLANTIC SALMON (SALMO-SALAR) BY PETROLEUM-HYDROCARBONS DURING A SHORTTERM EXPOSURE. Marine Pollution Bulletin 1992, 24 (6), 310-315. Heitkamp, M.A.; Johnson B. T., IMPACT OF AN OIL FIELD EFFLUENT ON MICROBIAL ACTIVITIES IN A WYOMING RIVER." Canadian Journal Of Microbiology 30.6 1984, 30 (6), 786-792. Hill, A. B., THE ENVIRONMENT AND DISEASE: ASSOCIATION OR CAUSATION?,” Proceedings of the Royal Society of Medicine 1965, 58, 295-300. Hofer, T., TAINTING OF SEAFOOD AND MARINE POLLUTION. Water Research 1998, 32 (12), 3505-3512. Hoffman, D.J.; Rattner, B.A.; Burton Jr, G.A.; and Cairns Jr, J., HANDBOOK OF ECOTOXICOLOGY. CRC press 2002. Hyne, Norman J. NONTECHNICAL GUIDE TO PETROLEUM GEOLOGY, EXPLORATION, DRILLING, AND PRODUCTION. PennWell Books 2012. Incardona, J. P.; Gardner, L. D.; Linbo, T. L.; Brown, T. L.; Esbaugh, A. J.; Mager, E. M.; Stieglitz, J. D.; French, B. L.; Labenia, J. S.; Laetz, C. A.; Tagal, M.; Sloan, C. A.; Elizur, A.; Benetti, D. D.; Grosell, M.; Block, B. A.; Scholz, N. L., DEEPWATER HORIZON CRUDE OIL IMPACTS THE DEVELOPING HEARTS OF LARGE PREDATORY PELAGIC FISH. Proceedings of the National Academy of Sciences of the United States of America 2014, 111 (15), E1510-E1518. Jurcak, A. M.; Gauthier, S. J.; Moore, P. A., THE EFFECTS OF BIODIESEL AND CRUDE OIL ON THE FORAGING BEHAVIOR OF RUSTY CRAYFISH, ORCONECTES RUSTICUS. Archives of Environmental Contamination and Toxicology 2015, 69 (4), 557-565. Kashian, D. R.; Zuellig, R. E.; Mitchell, K. A.; Clements, W. H., THE COST OF TOLERANCE: SENSITIVITY OF STREAM BENTHIC COMMUNITIES TO UV-B AND METALS. Ecological Applications 2007, 17 (2), 365-375. Khan, R. A., STUDY OF PEARL DACE (MARGARISCUS MARGARITA) INHABITING A STILLWATER POND CONTAMINATED WITH DIESEL FUEL. Bulletin of Environmental Contamination and Toxicology 1999, 62 (5), 638-645. Khan, R. A., STUDY OF PEARL DACE (MARGARISCUS MARGARITA) INHABITING A STILLWATER POND CONTAMINATED WITH DIESEL FUEL. Bulletin of Environmental Contamination and Toxicology 1999, 62 (5), 638-645. Khan, R. A.; Kiceniuk, J., HISTOPATHOLOGICAL EFFECTS OF CRUDE-OIL ON ATLANTIC COD FOLLOWING CHRONIC EXPOSURE. Canadian Journal of Zoology-Revue Canadienne De Zoologie 1984, 62 (10), 2038-2043. 81 �Kuehn, R. L.; Berlin, K. D.; Hawkins, W. E.; Ostrander, G. K., RELATIONSHIPS AMONG PETROLEUM REFINING, WATER AND SEDIMENT CONTAMINATION, AND FISH HEALTH. Journal of Toxicology and Environmental Health 1995, 46(1), 101-116. Lefcort, H.; Hancock, K. A.; Maur, K. M.; Rostal, D. C., THE EFFECTS OF USED MOTOR OIL, SILT, AND THE WATER MOLD SAPROLEGNIA PARASITICA ON THE GROWTH AND SURVIVAL OF MOLE SALAMANDERS (GENUS AMBYSTOMA). Archives of Environmental Contamination and Toxicology 1997, 32 (4), 383-388. Little, E. E.; Cleveland, L.; Calfee, R.; Barron, M. G., ASSESSMENT OF THE PHOTOENHANCED TOXICITY OF A WEATHERED OIL TO THE TIDEWATER SILVERSIDE. Environmental Toxicology and Chemistry 2000, 19 (4), 926-932. Livingstone, D. R., CONTAMINANT-STIMULATED REACTIVE OXYGEN SPECIES PRODUCTION AND OXIDATIVE DAMAGE IN AQUATIC ORGANISMS. Marine Pollution Bulletin 2001, 42 (8), 656-666. Lushchak, V. I., FREE RADICAL OXIDATION OF PROTEINS AND ITS RELATIONSHIP WITH FUNCTIONAL STATE OF ORGANISMS. Biochemistry-Moscow 2007, 72 (8), 809-827. Lytle, D. A.; Peckarsky, B. L., SPATIAL AND TEMPORAL IMPACTS OF A DIESEL FUEL SPILL ON STREAM INVERTEBRATES. Freshwater Biology 2001, 46 (5), 693-704. Mager, E. M.; Esbaugh, A. J.; Stieglitz, J. D.; Hoenig, R.; Bodinier, C.; Incardona, J. P.; Scholz, N. L.; Benetti, D. D.; Grosell, M., ACUTE EMBRYONIC OR JUVENILE EXPOSURE TO DEEPWATER HORIZON CRUDE OIL IMPAIRS THE SWIMMING PERFORMANCE OF MAHI-MAHI (CORYPHAENA HIPPURUS). Environmental Science & Technology 2014, 48 (12), 7053-7061. Mahaney, P. A., EFFECTS OF FRESH-WATER PETROLEUM CONTAMINATION ON AMPHIBIAN HATCHING AND METAMORPHOSIS. Environmental Toxicology and Chemistry 1994, 13 (2), 259-265. Malins, D. C.; Hodgins, H. O., PETROLEUM AND MARINE FISHES - A REVIEW OF UPTAKE, DISPOSITION, AND EFFECTS. Environmental Science & Technology 1981, 15(11), 1272-1280. Mankki, j; Vauras, J., LITTORAL FISH POPULATIONS AFTER AN OIL TANKER DISASTER IN THE FINNISH SW ARCHIPELAGO. Annales Zoologici Fennici. Societas Biologica Fennica Vanamo, 1974 120-126. McCauley, R. N., BIOLOGICAL EFFECTS OF OIL POLLUTION IN A RIVER. Limnology and Oceanography 1966, 11 (4), 475-486. McGrath, E. A.; Alexander, M. M., OBSERVATIONS ON THE EXPOSURE OF LARVAL BULLFROGS TO FUEL OIL. Transactions of the Northeast Section of the Wildlife Society 1979 36, 45-51. Michaelis, F. B., EFFECT OF TUROA OIL-SPILL ON AQUATIC INSECTS IN THE MANGAWHERO RIVER SYSTEM. New Zealand Entomologist 1983, 7 (4), 447-455. Miller, M. C.; Stout, J. R.; Alexander, V., EFFECTS OF A CONTROLLED UNDER-ICE OILSPILL ON INVERTEBRATES OF AN ARCTIC AND A SUB-ARCTIC STREAM. Environmental Pollution Series a-Ecological and Biological 1986, 42 (2), 99-132. 82 �Moles, A.; Norcross, B. L., EFFECTS OF OIL-LADEN SEDIMENTS ON GROWTH AND HEALTH OF JUVENILE FLATFISHES. Canadian Journal of Fisheries and Aquatic Sciences 1998, 55 (3), 605-610. Monson, D. H.; Doak, D. F.; Ballachey, B. E.; Johnson, A.; Bodkin, J. L., LONG-TERM IMPACTS OF THE EXXON VALDEZ OIL SPILL ON SEA OTTERS, ASSESSED THROUGH AGE-DEPENDENT MORTALITY PATTERNS. Proceedings of the National Academy of Sciences of the United States of America 2000, 97 (12), 6562-6567. Morris, J.M.; Lay, C.R.; Forth, H. P., EFFECTS OF WEATHERING ON THE TOXICITY OF OIL TO EARLY-LIFESTAGE FISH AND INVERTEBRATES. DWH NRDA Toxicity Technical Working Group Report. Prepared for National Oceanic and Atmospheric Administration by Abt Associates, Boulder, CO. 2015 (Technical Report) Mos, L.; Cooper, G. A.; Serben, K.; Cameron, M.; Koop, B. F., EFFECTS OF DIESEL AN SURVIVAL GROWTH, AND GENE EXPRESSION IN RAINBOW TROUT (ONCORHYNCHUS MYKISS) FRY. Environmental Science & Technology 2008, 42 (7), 2656-2662. Moulton S.R. II; Carter, J.L.; Grotheer, S.A.; Cuffney, T.F; Short, T.M., METHODS OF ANALYSIS BY THE U. S. GEOLOGICAL SURVEY NATIONAL WATER QUALITY LABORATORY: PROCESSING, TAXONOMY, AND QUALITY CONTROL OF BENTHIC MACROINVERTEBRATE SAMPLES. 2000. Open-File Report 00-212, U. S. Geological Survey. Muller, K., STREAM DRIFT AS A CHRONOBIOLOGICAL PHENOMENON IN RUNNING WATER ECOSYSTEMS. Annual Review of Ecology and Systematics 1974, 5(1), 309323. Neff, J. M. POLYCYCLIC AROMATIC HYDROCARBONS. FUNDAMENTALS OF AQUATIC TOXICOLOGY: METHODS AND APPLICATIONS. Hemisphere Publishing Corporation 1985, 416-454. Nitta, T. MARINE POLLUTION IN JAPAN. Marine Pollution and Sea Life. FAO 1972, 77-81. Owens, E. H.; E. Taylor, Marty, R.; Little, D. I., AN INLAND OIL SPILL RESPONSE MANUAL TO MINIMIZE ADVERSE ENVIRONMENTAL IMPACTS." American Petroleum Institute International Oil Spill Conference 1993, 105-109. Parker, B. L.; Brammer, J. D.; Whalon, M. E.; Berry, W. O., CHRONIC OIL CONTAMINATION AND AQUATIC ORGANISMS WITH EMPHASIS ON DIPTERA: STATUS AND BIBLIOGRAPHY JAWRA Journal of the American Water Resources Association 1976 12(2), 291-305. Peterson, C. H.; Rice, S. D.; Short, J. W.; Esler, D.; Bodkin, J. L.; Ballachey, B. E.; Irons, D. B., LONG-TERM ECOSYSTEM RESPONSE TO THE EXXON VALDEZ OIL SPILL. Science 2003, 302 (5653), 2082-2086. Posthuma, J., THE COMPOSITION OF PETROLEUM. Rapports et Procès-Verbaux des Réunions du Conseil Permanent International pour l'Exploration de la Mer 1977. Pilcher, W.; Miles, S.; Tang, S.; Mayer, G.; Whitehead, A., GENOMIC AND GENOTOXIC RESPONSES TO CONTROLLED WEATHERED-OIL EXPOSURES CONFIRM AND EXTEND FIELD STUDIES ON IMPACTS OF THE DEEPWATER HORIZON OIL SPILL ON NATIVE KILLIFISH. Plos One 2014, 9 (9), 11. Poulton, B. C.; Callahan, E. V.; Hurtubise, R. D.; Mueller, B. G., EFFECTS OF AN OIL SPILL ON LEAFPACK-INHABITING MACROINVERTEBRATES IN THE CHARITON RIVER, MISSOURI. Environmental Pollution 1998, 99 (1), 115-122. 83 �Roberts, M. H.; Hargis, W. J.; Strobel, C. J.; Delisle, P. F., ACUTE TOXICITY OF PAH CONTAMINATED SEDIMENTS TO THE ESTUARINE FISH, LEIOSTOMUSXANTHURUS. Bulletin of Environmental Contamination and Toxicology 1989, 42 (1), 142-149. Rodrigues, R. V.; Miranda-Filho, K. C.; Gusmao, E. P.; Moreira, C. B.; Romano, L. A.; Sampaio, L. A., DELETERIOUS EFFECTS OF WATER-SOLUBLE FRACTION OF PETROLEUM, DIESEL AND GASOLINE ON MARINE PEJERREY ODONTESTHES ARGENTINENSIS LARVAE. Science of the Total Environment 2010, 408 (9), 2054-2059. Saha, M.; Togo, A.; Mizukawa, K.; Murakami, M.; Takada, H.; Zakaria, M. P.; Chiem, N. H.; Tuyen, B. C.; Prudente, M.; Boonyatumanond, R.; Sarkar, S. K.; Bhattacharya, B.; Mishra, P.; Tana, T. S., SOURCES OF SEDIMENTARY PAHS IN TROPICAL ASIAN WATERS: DIFFERENTIATION BETWEEN PYROGENIC AND PETROGENIC SOURCES BY ALKYL HOMOLOG ABUNDANCE. Marine Pollution Bulletin 2009, 58 (2), 189-200. Sandrini-Neto, L.; Pereira, L.; Martins, C. C.; de Assis, H. C. S.; Camus, L.; Lana, P. C., ANTIOXIDANT RESPONSES IN ESTUARINE INVERTEBRATES EXPOSED TO REPEATED OIL SPILLS: EFFECTS OF FREQUENCY AND DOSAGE IN A FIELD MANIPULATIVE EXPERIMENT. Aquatic Toxicology 2016, 177, 237-249. Santodonato, J.; Howard, P.; Basu, D., HEALTH AND ECOLOGICAL ASSESSMENT OF POLYNUCLEAR AROMATIC-HYDROCARBONS. Journal of Environmental Pathology and Toxicology 1981, 5 (1), 1-364. Schein, A.; Scott, J. A.; Mos, L.; Hodson, P. V., OIL DISPERSION INCREASES THE APPARENT BIOAVAILABILITY AND TOXICITY OF DIESEL TO RAINBOW TROUT (ONCORHYNCHUS MYKISS). Environmental Toxicology and Chemistry 2009, 28 (3), 595-602. Saadoun, Ismail MK. "IMPACT OF OIL SPILLS ON MARINE LIFE." Emerging Pollutants In The Environment-Current And Further Implications, 2015. Schmitt, C.J.; Blazer, V.S.; Dethloff, G.M.; Tillitt, D.E.; Gross, T.S., BIOMONITORING OF ENVIRONMENTAL STATUS AND TRENDS (BEST) PROGRAM: FIELD PROCEDURES FOR ASSESSING THE EXPOSURE OF FISH TO ENVIRONMENTAL CONTAMINANTS. No. USGS/BRD/ITR--1999-0007. Geological Survey Columbia Mo Biological Resources Div, 1999. Schmitt, C. J.; Dethloff. G. M., BIOMONITORING OF ENVIRONMENTAL STATUS AND TRENDS (BEST) PROGRAM: SELECTED METHODS FOR MONITORING CHEMICAL CONTAMINANTS AND THEIR EFFECTS IN AQUATIC ECOSYSTEMS. No. USGS/BRD/ITR-2000-0005. Geological Survey Columbia Mo Biological Resources Div, 2000. Shapiro, M. F.; Greenfield, S., THE COMPLETE BLOOD-COUNT AND LEUKOCYTE DIFFERENTIAL COUNT - AN APPROACH TO THEIR RATIONAL APPLICATION. Annals of Internal Medicine 1987, 106 (1), 65-74. Shapiro, M. F.; Greenfield, S., THE COMPLETE BLOOD-COUNT AND LEUKOCYTE DIFFERENTIAL COUNT - AN APPROACH TO THEIR RATIONAL APPLICATION. Annals of Internal Medicine 1987, 106 (1), 65-74. Short, J. W.; Jackson, T. J.; Larsen, M. L.; Wade, T. L., ANALYTICAL METHODS USED FOR THE ANALYSIS OF HYDROCARBONS IN CRUDE OIL, TISSUES, 84 �SEDIMENTS, AND SEAWATER COLLECTED FOR THE NATURAL RESOURCES DAMAGE ASSESSMENT OF THE EXXON VALDEZ OIL SPILL. Proceedings of the Exxon Valdez Oil Spill Symposium 1996, 18, 140-148. Simpson, K. W., ABNORMALITIES IN THE TRACHEAL GILLS OF AQUATIC INSECTS COLLECTED FROM STREAMS RECEIVING CHLORINATED OR CRUDE-OIL WASTES. Freshwater Biology 1980, 10 (6), 581-&. Smith, P.; Snook, D.; Muscutt, A.; Smith, A., EFFECTS OF A DIESEL SPILL ON FRESHWATER MACROINVERTEBRATES IN TWO URBAN WATERCOURSES, WILTSHIRE, UK. Water and Environment Journal 2010, 24 (4), 249-260. Snow, N. B.; Scott B. F., THE EFFECT AND FATE OF CRUDE OIL SPILT ON TWO ARCTIC LAKES. American Petroleum Institute International Oil Spill Conference. 1975, 1, 527-534. Song, C., CHEMISTRY OF DIESEL FUELS. CRC Press, 2000. Southward, A. J.; Southward, E. C., RECOLONIZATION OF ROCKY SHORES IN CORNWALL AFTER USE OF TOXIC DISPERSANTS TO CLEAN UP TORREYCANYON SPILL. Journal of the Fisheries Research Board of Canada 1978, 35 (5), 682-706. Squire Jr, James L., EFFECTS OF THE SANTA BARBARA, CALIF., OIL SPILL ON THE APPARENT ABUNDANCE OF PELAGIC FISHERY RESOURCES. Marine Fisheries Review 1992 54 (1), 7-14. State of Colorado. Department of Natural Resources. Division of Wildlife. ADMINISTRATIVE DIRECTIVE W-5. INVESTIGATING AND REPORTING NATURAL RESOURCE INJURY DUE TO KNOWN OR SUSPECTED POLLUTION OF WATER IN COLORADO 2007. Teal, J. M.; Howarth, R. W., OIL-SPILL STUDIES - A REVIEW OF ECOLOGICAL EFFECTS. Environmental Management 1984, 8 (1), 27-43. Thomas, P.; Woodin, B. R.; Neff, J. M., BIOCHEMICAL RESPONSES OF THE STRIPED MULLET MUGIL-CEPHALUS TO OIL EXPOSURE .1. ACUTE RESPONSES INTERRENAL ACTIVATIONS AND SECONDARY STRESS RESPONSES. Marine Biology 1980, 59 (3), 141-149. U.S. Department of Energy. ALTERNATIVE FUELS DATA CENTER – FUEL PROPERTIES COMPARISON 2014 https://www.afdc.energy.gov/fuels/fuel_comparison_chart.pdf Varanasi, U., Nishimoto, M., Baird, W. M., & Smolarek, T. A.; METABOLIC ACTIVATION OF PAH IN SUBCELLULAR FRACTIONS AND CELL CULTURES FROM AQUATIC AND TERRESTRIAL SPECIES." Metabolism of polycyclic aromatic hydrocarbons in the aquatic environment. CRC Press 1989, 203-251. Vorob'ev, D. S.; Frank, Y. A.; Lushnikov, S. V.; Zaloznyi, N. A.; Noskov, Y. A., OIL DECONTAMINATION OF BOTTOM SEDIMENTS USING LIMNODRILUS HOFFMEISTERI (OLIGOCHAETA: TUBIFICIDAE). Contemporary Problems of Ecology 2010, 3 (1), 15-18. Waldichuk, M., SEA OTTERS AND OIL POLLUTION. Marine Pollution Bulletin 1990, 21 (1), 10-15. Wallace, J. B.; Lugthart, G. J.; Cuffney, T. F.; Schurr, G. A., THE IMPACT OF REPEATED INSECTICIDAL TREATMENTS ON DRIFT AND BENTHOS OF A HEADWATER STREAM. Hydrobiologia 1989, 179 (2), 135-147. 85 �Weston Solutions Inc., WEST CREEK TRIP REPORT 2013, https://response.epa.gov/site/doc_list.aspx?site_id=8394 Whyte, J. J.; Jung, R. E.; Schmitt, C. J.; Tillitt, D. E., ETHOXYRESORUFIN-ODEETHYLASE (EROD) ACTIVITY IN FISH AS A BIOMARKER OF CHEMICAL EXPOSURE. Critical Reviews in Toxicology 2000, 30 (4), 347-570. Woodward, D. F.; Riley, R. G., PETROLEUM HYDROCARBON CONCENTRATIONS IN A SALMONID STREAM CONTAMINATED BY OIL-FIELD DISCHARGE WATER AND EFFECTS ON MACROBENTHOS. Archives of Environmental Contamination and Toxicology 1983, 12 (3), 327-334. Woodward, D. F.; Riley, R. G.; Smith, C. E., ACCUMULATION, SUBLETHAL EFFECTS, AND SAFE CONCENTRATION OF A REFINED OIL AS EVALUATED WITH CUTTHROAT TROUT. Archives of Environmental Contamination and Toxicology 1983, 12 (4), 455-464. Wu, D. M.; Wang, Z. D.; Hollebone, B.; McIntosh, S.; King, T.; Hodson, P. V., COMPARATIVE TOXICITY OF FOUR CHEMICALLY DISPERSED AND UNDISPERSED CRUDE OILS TO RAINBOW TROUT EMBRYOS. Environmental Toxicology and Chemistry 2012, 31 (4), 754-765. 86 � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Reports Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Effects of the 2013 West Creek Petroleum Spill on Stream Ecosystem Structure and Function: Responses of Periphyton, Macroinvertebrates and Fish Description An account of the resource <p><strong>Funding: </strong>This research was funded by Colorado Parks and Wildlife (16-IAA-82320) and the Colorado State University Energy Institute. Colorado State University’s Intuitional Animal Care and Use Committee approved this research under protocols 15-6086A and 16-6367.<br /><br /><strong>Abstract: </strong>Oil development has expanded dramatically in Colorado over the last decade. Associated with this rapid expansion has been a significant increase in the number of accidental releases into the environment. On January 2013, West Creek which flows along a scenic byway in Unaweep Canyon, Colorado, was impacted by a petroleum spill from an overturned tanker truck. 22,700 liters of gasoline and 7,300 liters of diesel discharged into the stream killing an estimated 1,206 Brown Trout, Salmo trutta, and 8,172 Mottled Sculpin, Cottus bairdii. Subsequent electrofishing surveys indicated that the fishery was not quickly recovering particularly with regard to Mottled Sculpin populations, but also Brown Trout. In June and October 2015, as part of ongoing efforts to determine long term effects of this spill, we explored health indicators across multiple levels of biological organization. Histopathological abnormalities (e.g., ectopic neural tissue, cystic kidney, increased melanomacrophage aggregates) were observed in Mottled Sculpin collected from the spill site and nearby downstream sites. Altered benthic macroinvertebrate community structure was observed at the spill site compared with a reference site one kilometer upstream. Interestingly, a GC-MS finger-printing analysis of polycyclic aromatic hydrocarbons (PAHs) in stream sediment revealed that PAH concentrations were typical of minimally impacted streams flowing adjacent to roads. These results suggest that effects of the spill were persisting after contaminant concentrations had returned to ‘normal’ by Fall 2015. Subsequently, we conducted two mesocosm experiments, using naturally colonized benthic macroinvertebrate communities. Exposure to simulated spill conditions caused concentration-dependent macroinvertebrate drift and substantial mortality that occurred rapidly after the spills were initiated and at lower concentrations than expected. In addition, concentration-dependent lethal and sub-lethal effects were observed in Rainbow Trout, Oncorhynchus mykiss, during simulated spill bioassays. Periphyton biofilms were also adversely affected. We conclude that petroleum spills in coldwater streams risk adverse acute, chronic, lethal and sub-lethal effects to aquatic communities across numerous levels of biological organization. And these effects were evident after the 2013 West Creek petroleum spill. Moreover, by utilizing field observations, mesocosms and bioassays we gained insights into consequences of petroleum spills using an ecotoxicological weight-of evidence approach. Importantly, the methods used in this project can be employed at future spill events as field useful bioassessment techniques to aid in the process of holding responsible parties appropriately accountable for damages to stream communities.</p> Bibliographic Citation A bibliographic reference for the resource. Recommended practice is to include sufficient bibliographic detail to identify the resource as unambiguously as possible. Duggan, S. B., P. Schaffer, P. Cadmus, and W. H. Clements. 2017. Dept. of Fish, Wildlife and Conservation Biology, Colorado State University, Fort Collins, Colorado. Creator An entity primarily responsible for making the resource Sam B. Duggan, Department of Fish, Wildlife and Conservation Biology. Colorado State University. Dr. Paula Schaffer, Department of Microbiology, Immunology and Pathology, Veterinary Diagnostic Laboratory, Colorado State University. <a href="https://cpw.cvlcollections.org/items/show/532">Pete Cadmus</a>, Aquatic Research, Colorado Parks and Wildlife. Dr. William H. Clements, Department of Fish, Wildlife and Conservation Biology. Colorado State University. Extent The size or duration of the resource. 96 pages Date Created Date of creation of the resource. 2017 Rights Information about rights held in and over the resource <a href="https://rightsstatements.org/page/InC-NC/1.0/?language=en">In Copyright - Non-Commercial Use Permitted</a>