TRWP were collected at a road simulator laboratory located within the Bundesanstalt für Straßenwesen (BASt), the German Federal Highway Research Institute, as previously described by Kreider et al. (2010). Briefly, the laboratory used an interior drum testing system containing actual asphalt pavement in cassettes. This system was electronically programmable to mimic a variety of driving conditions by varying speed, temperature, acceleration, braking, and steering. For the TRWP collection, the pavement consisted of a standardized asphalt concrete with 6.1 % proportion of bitumen (B50/70) according to ISO 10844. To prevent overheating from friction, the road surface temperature was maintained at approximately 20 °C for summer tires and 14 °C for the winter tire using a climate control system in the drum during the tests. The TRWP was collected using a vacuum system mounted behind one of the simulator wheels. Both summer and winter silica based tires (Michelin Pilot Primacy 225/55 R16 95W and Pirelli Sottozero 225/55 R16 95W M + S) and a carbon-black based summer tire (Bridgestone Potenza RE 88 205/65 R15 94W) were used to generate the TRWP. Particles from each tire were combined to form a single composite (2:1:1 Bridgestone: Michelin: Pirelli) that was sieved at 150 μm to remove any large pavement pieces. The TRWP was shipped on ice in amber glass jars to the toxicity testing laboratory under full chain of custody and were thereafter stored in the dark at 4 °C.
Sediment and elutriate preparation
Freshwater reference sediment was collected from a drinking water reservoir near Dillon Beach, CA. This reservoir has previously been used by the toxicity testing laboratory as a source of reference sediment due to the isolation and absence of any known sources of contamination and no major drainage inputs. Analytical testing of the sediment indicated metals concentrations were generally below sediment screening criteria. To prepare the mixtures for use in toxicity testing, the reference sediment was spiked with 10 g/kg (10,000 ppm by weight) of TRWP and homogenized using mechanical mixing for 15 min. The sediment was allowed to equilibrate in cold storage (4 °C) for 48 h. Unspiked reference sediment was also mechanically mixed to control for any effects associated with the mixing process. Both mixed and unmixed reference sediment were used as controls in whole sediment toxicity tests. The spiked sediment was used for whole sediment toxicity tests in Chironomus dilutus and Hyalella azteca, and was also used to prepare the sediment elutriate used for treatment of Ceriodaphnia dubia and Pimephales promelas. All chronic toxicity tests were performed by Nautilus Environmental, LLC (San Diego, CA).
To test chronic toxicity in water-dwelling organisms, a sediment elutriate design was used. Elutriate designs for preparing test solutions that replicate sediment mobilization and storm water phenomena are recommended and have been successfully employed by other investigators for exposing water column species such as algae and daphnids (ASTM 2011; Bosch et al. 2009; Novelli et al. 2006). To prepare the sediment elutriate, one part sediment was mixed with four parts of moderately hard water for a period of 24 h at 15 °C using stainless steel mechanical mixers. After mixing, the sediment was allowed to settle for 24 h, at which time the elutriate was siphoned off and centrifuged at 1,100×g for 15 min. The supernatant was removed and used as the test and renewal material for the Ceriodaphnia dubia and Pimephales promelas exposures.
Whole sediment exposures
Chironomus egg cases were obtained from Aquatic BioSystems (ABS) in Fort Collins, CO and acclimated to test conditions prior to test initiation. During acclimation, the test organisms were observed for any indications of stress, as indicated by deterioration of the egg cases. Water quality parameters including conductivity, temperature, dissolved oxygen, and pH were monitored daily. The egg cases were kept at test temperature and monitored daily until larvae hatched and vacated the egg case, at which time the test was initiated (<8-h post-hatch).
Chironomus larvae were exposed to reference sediment (both mixed and unmixed) or sediment containing 10,000 mg/kg TRWP for 35 days and evaluated for survival, growth, and emergence in accordance with the Organization for Economic Cooperation and Development (OECD) Technical Guidance Document (TGD) 218 (OECD 2004). In addition, laboratory control sediment consisting of a combination of Scripps sand and peat was used as a control. Each control and test sediment sample was evaluated in 10 replicates. An additional replicate was included for each control and test sample as a surrogate for measurement of water quality parameters, which occurred daily. The test chamber contained 2 cm of sediment and 400 mL of moderately hard water. A Zumwalt water renewal system was used to facilitate water renewals without disturbing the sediment (U.S. EPA 2000). Each chamber was supplied with a continuous aeration rate of three to four bubbles per second with a glass pipette. The test was conducted in an environmental chamber maintained at 23 ± 1 °C under a 16 h (light):8 h (dark) cycle. Each test chamber contained 12 larvae.
On day 10, sediments in five test chambers were removed and sieved through a 0.5 mm screen, and the number of larvae surviving was recorded. Surviving organisms were dried at 60 °C for 24 h and the dry weight of each replicate was measured to the nearest 0.01 mg. Samples were then placed in a furnace at 550 °C for 24 h to ash organic material and reweighed to provide an ash weight. The difference between the two measurements (dry weight minus ash weight) represents the organic mass of the larvae (ash-free dry weight [AFDW]).
The remaining five replicates were used to evaluate Chironomus emergence. Beginning on day 20 and until the end of the test period (day 35), the number and sex of emerged adult Chironomus were recorded daily. The test series was terminated after five consecutive days of non-emergence from the laboratory control.
In addition to the above described control and test samples, a positive control reference toxicant test using copper chloride (375, 750, 1500, 3000, and 6000 μg/L) was conducted concurrently to estimate the sensitivity of the test organisms in relation to those historically tested at the laboratory.
Hyalella azteca were obtained from Aquatic Indicators (St. Augustine, FL); the amphipods were 8 days old upon arrival. The Hyalella were acclimated to test conditions prior to test initiation in order to promote and confirm animal health. During the acclimation phase, the amphipods were observed for mortality and any indicators of stress, such as abnormal swimming behavior or discoloration. Mortality was considered significant if it was greater than 10 % during the holding and acclimation periods.
Hyalella were exposed to reference sediment (mixed or unmixed) or sediment spiked with 10,000 mg/kg TRWP for 42 days, in accordance with methods recommended by the U.S. Environmental Protection Agency (USEPA) (U.S. EPA 2002). The exposure chamber and conditions were maintained and monitored as described for treatment of Chironomus, with 12 replicates per treatment group. Four of the 12 replicates were used for intermediate endpoint evaluation (day 28) including survival and growth. On day 28, the amphipods from the remaining eight test chambers were placed into clean jars without sediment; a nitex screen was added as a clean substrate. On day 35, each test chamber was evaluated for reproduction; the number of young produced in each chamber was recorded and the young removed. At day 42, the additional number of young produced was recorded for each test chamber, as well as the number of adult amphipods and their gender. Growth was determined by dry weight after oven-drying of the amphipods.
In addition to the above described control and test samples, a positive control reference toxicant test using copper chloride (100, 200, 400, 800, and 1600 μg/L) was conducted concurrently to estimate the sensitivity of the test organisms in relation to those historically tested at the laboratory.
The Ceriodaphnia dubia used for this study were obtained from internal laboratory culture. Prior to test initiation, neonatal water fleas (<24 h old) were isolated from brood stock cultures and placed in individual holding cups containing clean culture water and food. Cultures were maintained in a temperature controlled room at 25 ± 1 °C. Isolated females were transferred daily to new cups containing fresh water and food. Neonates produced within the 24 h prior to testing were selected for chronic toxicity if produced by individuals that had at least three broods averaging eight or more neonates each.
Ceriodaphnia were tested for chronic toxicity over a period of 7 days following methods recommended by the USEPA (U.S. EPA 2002). The organisms were treated using sediment elutriate, prepared as described above, from 10,000 mg/kg spiked sediment. Measurements of pH, dissolved oxygen, temperature and conductivity were measured and recorded for each treatment and control. The test consisted of ten replicates exposed to either TRWP sediment elutriate, control sediment elutriate, or control laboratory water (moderately hard water). Test solutions were renewed and organisms fed once daily. Water quality was monitored daily in both freshly prepared test renewal solution and test solution collected from the test chambers. Survival status and reproductive output were recorded once per day. At test termination (7 days), final observations were made and organisms discarded.
In addition to the above described control and test samples, a positive control reference toxicant test using copper chloride (12.5, 25, 50, 100, 200 μg/L) was conducted concurrently to estimate the sensitivity of the test organisms in relation to those historically tested at the laboratory.
Fathead minnow embryos and larval fish were purchased from Aquatic Biosystems (Fort Collins, CO). Upon receipt, each batch of animals was acclimated to the proper test temperature of 25 ± 1 °C prior to test initiation. Water quality, including conductivity, temperature, dissolved oxygen, and pH, in the larval fish and embryo holding chambers was monitored daily during holding.
To evaluate chronic toxicity in fathead minnows, the hatching success, survival and growth (mass and length) of fathead minnows exposed to the test material (TRWP sediment elutriate) were assessed over a period of 32 days, according to OECD TGD 210 methods and criteria (OECD 1992). The test was initiated with unhatched embryos ranging from 52 to 56 h post-fertilization. Controls included a reference sediment elutriate and a laboratory water (moderately hard water) control. Testing was conducted using six replicate test chambers containing 200 mL of test solution maintained at 25 ± 1 °C with a 16:8 h light:dark cycle. Each test chamber contained ten fish embryos of healthy appearance at test initiation. All test chambers were initiated with continuous, light aeration to maintain dissolved oxygen content in an acceptable range. Water renewals were performed every 8 days due to limited test material.
To assess toxicity, hatching status and survival of post-hatch larvae was recorded daily. At test termination, final observations were made and test animals were prepared for weight and length determination. Fork length of each fish, measured from the tip of the snout to the posterior end of the middle caudal fin rays to the nearest 0.5 mm, was recorded at test termination. Fish growth was determined by measuring the pooled dry weight of all fish in each test chamber.
In addition to the above described control and test samples, a positive control reference toxicant test using copper chloride (15, 30, 60, 120, and 240 μg/L) was conducted concurrently to estimate the sensitivity of the test organisms in relation to those historically tested at the laboratory.
Sediment and elutriate test comparisons
Survival (proportional) data were arcsin square-root transformed prior to analysis. Growth data was log-transformed as needed to satisfy statistical assumptions. Reproduction data was not transformed for any species prior to analysis. Prior to and following transformation, all datasets were examined for normality using either the Shapiro–Wilk, or the D’Agostino & Pearson omnibus normality test. Homogeneity of variance was determined using either the Bartlett’s test or F-ratio test.
Ceriodaphnia dubia, Chironomus dilutus, Hyalella azteca, and Pimephales promelas test data were analyzed with GraphPad Prism® Statistical Software, Version 4.02. Statistical comparisons to determine whether the TRWP impacted Chironomus dilutus and Hyalella azteca performance were conducted between the spiked sediment and mixed reference sediment. Differences between other treatments were evaluated for comparative purposes only. Likewise, statistical comparisons to determine the effects of TRWP on fathead minnows were conducted between the elutriate control and the TRWP-spiked elutriate.
A one-way analysis of variance (ANOVA) was performed to determine if there were any significant differences in treatments means for the toxicity tests that were performed (Neter et al. 1990). If the ANOVA indicated that there was at least one significant difference at 95 % confidence level (p < 0.05), the Tukey pairwise multiple comparison tests were performed to identify the specific significant differences. The assumption of a normal distribution was verified for each ANOVA using probability plots of the data for each treatment group.
Reference toxicant tests
Statistical analyses for standard reference toxicant tests were conducted using CETIS Version 1.6.3revE. Normality of the data was established by Shapiro–Wilk’s Normality Test and equality of variance by either Levene’s or Bartlett’s test. A medial lethal concentration (LC50) value was then calculated using either the Trimmed Spearman-Karber or Linear Interpolation methods. Statistical methods were chosen based on methodology for data analysis as outlined by the USEPA (U.S. EPA 2002).