TOXICITY OF PHYSICALLY AND CHEMICALLY DISPERSED OILS UNDER CONTINUOUS AND ENVIRONMENTALLY REALISTIC EXPOSURE CONDITIONS: APPLICABILITY TO DISPERSANT USE DECISIONS IN SPILL RESPONSE PLANNING

摘要

As part of efforts to develop standardized testing protocols under the Chemical Response to Oil Spills Environ-mental Research Forum (CROSERF) and apply the results to real-world scenarios, three types of oils and two dispersants were tested in both continuous and short-term spiked exposures using the early life-stages of several marine organisms. Test species included embryo-larval stages of Pacific oyster (Crassostrea gigas), two marine mysids (Holmesimysis costata and Mysidopsis bahia), and two marine fishes (turbot, Scophthalmus maximus and inland silverside, Menidia beryllina). Oils were physically dispersed in seawater by vortex mixing in a flask and chemically dispersed using the same approach with COREXIT? 9527 or COREXIT? 9500 applied in a 10:1 oil-to-dispersant ratio to generate maximum exposure concentrations. Continuous exposure tests followed standard testing protocols for 96-hour or 48-hour duration, according to demands of the test species. Spiked exposures reflect continuous dilution of water column concentrations (half-life ～107 minutes), as observed in the field when oil is dispersed into open waters. Results are reported as the acute LC50s. Tests oils included fresh and weathered Kuwait crude, fresh Forties crude, and a Medium Fuel Oil (MFO) mix. Exposure concentrations for oil tests were quantified using gas chromatography and expressed as the sum of the C10 to C36 components, or TPH(resolved). Dispersant exposure concentrations were verified by UV spectrophotometric analysis. Not all species were tested with each oil and dispersant. For dispersants tested individually, constant exposure LC50s ranged from 3 to 75 mg/L, with oyster the most sensitive and turbot the least sensitive species. Spiked exposure LC50s ranged from 14 to >1055 mg/L among all test species. Dispersants were up to 36 times less toxic under spiked exposure conditions compared to similar treatments under constant exposure conditions. For oils, LC50s based on TPH(resolved) are similar for both the physically and chemically dispersed oil, demonstrating that dispersant did not increase the toxicity of oils based on measured exposures. Under constant exposure conditions, test species are very similar in sensitivity to the oils, with most LC50s around 0.5 ppm TPH(resolved). Spiked exposures were 4 to 100 fold less toxic to these test organisms. The more environmentally realistic spiked exposures demonstrate that standard, continuous exposure test data overestimate the potential toxicity of dispersed oil. When laboratory toxicity data are used as part of a dispersant approval process for spill response, the decision should take into account whether exposure durations and sensitivity of test species are representative of conditions in the spill area.