The physical and chemical fate of oil accidentally spilled into the natural environment is challenging to study in a controlled manner. The environmental behaviour of diluted bitumen (dilbit), consisting of viscous bitumen plus a volatile diluent fraction added to aid pipeline flow, is especially poorly known. A recent controlled study at the International Institute for Sustainable Development Experimental Lakes Area (IISD-ELA) in northwestern Ontario, Canada, used contained cylindrical units called limnocorrals to mimic a series of accidental spills. The researchers then examined the physical and chemical fate of diluted bitumen after release.
In a paper published in Environmental Science and Technology, a group of eleven scientists led by Sawyer Stoyanovich, now an environmental scientist with Kilgour and Associates in Ottawa, then a University of Ottawa doctoral student supervised by environmental toxicologist Professor Jules Blais, closely monitored weathering of a bitumen mixture called the Cold Lake Winter Blend (CLWB). Bitumen was released into seven limnocorrals installed in IISD-ELA Lake 260 in summer 2018. Two untreated limnocorrals were controls. The seven limnocorrals received from 1.5 to 180 litres of CLWB, with volumes chosen to emulate previous accidental oil spills. Limnocorrals were monitored for 70 days post-spill.
Studying chemical composition over time, the team collected oil samples using paper-thin Teflon fabric strips passed through the oil slicks. Dipped strips, stored in sterile sampling jars, were shipped for Ottawa lab analysis. Using a gas chromatograph, they quantified petroleum hydrocarbons like n-alkanes, polycyclic aromatic hydrocarbons (PAHs), and petroleum biomarkers.
Once a spill occurs, weathering ensues via biotic and abiotic processes. The most dominant processes are evaporation, photooxidation – breakdown by sunlight – and biodegradation, breakdown by microbes. Some compounds also dissolve or emulsify in the water column. “All of those processes are happening at the same time once the oil spill occurs,” says Stoyanovich, transforming the slick’s colour and consistency.
They predicted that limnocorrals treated with higher oil volumes might see slower depletion but did not find such a relationship.
The team used diagnostic ratios, employing biomarker ratios from samples taken over time compared to a reference sample. “We use diagnostic ratios to give us an idea of which process is dominating” at various points in time, adds Blais, referring to the processes of evaporation, photooxidation, dissolution and biodegradation.
Diagnostic ratios for concentrations of n-alkanes, isoprenoids, and PAHs indicated that evaporation and photooxidation were predominant processes contributing to dilbit weathering. Low molecular weight hydrocarbons – short-chained alkanes including benzene and toluene – were prone to evaporation, consistent with previous studies. Evaporation was dominant early on. Once low molecular weight compounds were depleted, other processes, like photooxidation, broke down larger aromatic ring compounds like PAHs, which are more prone to light energy absorption.
Dissolution and biodegradation were less prevalent. Although oil-degrading microbes are ubiquitous, especially prevalent in marine waters, “in these pristine Boreal Lake conditions, we would expect to see less of a background community,” says Stoyanovich.
Heather Dettman, Senior Science Advisor at Natural Resources Canada’s CanmetENERGY research center in Devon, Alberta, suggests their methodology may underestimate biodegradation, noting the same microbes that break down vegetation falling into a lake can break down oil. Nevertheless, with diluted bitumen, she says, “the easy stuff to be biodegraded has already been biodegraded before it was even produced.” Diluted bitumen “is the crust left after the microbes in a reservoir have been chewing on it for millions of years.” Dettman would love to see this same study using conventional crude. Still, as someone that extensively studied bitumen behaviour in the lab, she commends the study as important because of its natural setting with “real sunshine, real weather, real wind, and real water.”
This study is part of the BOREAL project (Boreal Lake Oil Release Experiment by Additions to Limnocorrals) investigating ecosystem-level impacts of dilbit following freshwater spills. A critical finding of the larger study is that after a spill, bitumen sinks to the bottom “under conditions that were otherwise not thought to cause submergence,” says Blais, even without high wind, wave and suspended sediment loads.
“We’ve shown that there’s a certain amount of time to recover the oil before it sinks and then becomes very difficult to recover,” says Blais. Referencing the spill of diluted bitumen into Michigan’s Kalamazoo River, originating from Canada’s Athabasca oil sands, “there’s still an enormous amount of diluted bitumen at the bottom of that river because it was deemed too difficult and too potentially harmful to remove,” says Blais. “So it’s there forever.”
Nevertheless, Dettman notes that this experiment’s first sinking was after 15 days, “well beyond a reasonable response window for responders to get out there and collect it,” she says.
“We learned a lot in terms of what oil does in a real, natural environment,” says Blais, who underscores the value of the Experimental Lakes Area. “You need to be able to do these experiments in natural systems.”