Year of Publication: 2011
Abstract:
A pilot study was conducted by Alberta Innovates – Technology Futures during 2009 and 2010 to assess potential for labelling process-affected water from oil sands operations using a suite of isotopic and geochemical tracers, including inorganic and organic compounds in water. The study was initiated in response to a request from Alberta Environment and grant funds for the project were obtained from the Oil Sands Research and Information Network, University of Alberta. Three oil sands operators participated in the study, providing logistical support and/or personnel to assist with on-lease water sampling. Alberta Environment and its consultants also provided support for sampling of groundwater. At the outset of the study, Worley Parsons was subcontracted to carry out a detailed electromagnetic survey of the Athabasca River from Fort McMurray to the confluence of the Firebag River, to map high conductivity seeps as potential targets for water sampling. While the priority of this first phase of the study was fingerprinting of water sources (i.e., tailings ponds vs. natural groundwater, lakes, and river water), the survey also sampled a selection of river bed seeps to test application of the methods to identify the origin of these waters near the point of discharge to the Athabasca River. In total 39 samples were collected for this study. These included 8 process-affected water samples, 6 groundwater samples, 8 river bed seepage samples, and 15 river samples. A variety of isotope tracers were measured including oxygen-18 (18OH2O) and deuterium (2HH2O) in water, enriched tritium (e3H) in water, carbon-13 in dissolved organic carbon (13CDOC), carbon-13 and carbon-14 in dissolved inorganic carbon (13CDIC, 14C), sulfur-34 in dissolved sulfate (34SSO4), chlorine-37 in dissolved chloride (37Cl), and strontium-87 versus strontium-86 (87Sr/86Sr) and boron-11 (11B) in dissolved solids. Geochemical analyses included major-, minor- and trace elements, a range of metals, nutrients and total organic carbon, as well as 113 priority pollutants and naphthenic acids. Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) was also used to scan for thousands of organic compounds in the water samples. Overall, while selected isotopic and geochemical tracers were found to be definitive for labelling water sources in some locations, it is unreliable to attempt any universal labelling of water sources based solely on individual tracers or simple combinations of tracers. Understanding of the regional hydrogeological system, and interpretation of tracer variations in the context of a biogeochemical systems approach on a case by case basis offers the greatest potential for comprehensive understanding and labelling of water source and pathways. While limited in number of samples, the survey demonstrates the complimentary use of various fingerprinting techniques. Preliminary evaluation of statistical approaches for differentiating various water types using inorganic, organic and combined datasets yielded promising results. These methods potentially offer multiple lines of evidence for fingerprinting and should be further evaluated, refined and applied as part of more comprehensive future investigations. While organic and inorganic tracers were capable of fingerprinting process-affected water sources from different operators, identification of seep sources along the Athabasca River was much more challenging due to presence of complex water mixtures including groundwater and significant river water. The presence or absence of process-affected water in seeps along developed portions of the river remains to be verified and will require further baseline surveys. FT-ICR MS offers capability to resolve thousands of organic compounds, and may be the simplest, most cost-effective approach to build a baseline dataset for use in identification of process-affected waters in the natural aquatic environment. A wide range of organic compounds are observed in process-affected water and these are not limited to naphthenic acids and hydrocarbons. Further work to constrain sources, pathways and receptors of process-affected water needs to be undertaken. From a riverine perspective, synoptic surveys offer an integrative method for better understanding of evolution of the Athabasca River and tributaries as it may be affected by addition of both natural and potentially process-affected water. We find no evidence of robust connections between tailings ponds and the river seeps that were sampled over the 125-km reach traversing the oil sands development area, although many seeps were not sampled. Although the seeps we did sample appear to be directly related to occurrence of natural groundwater seepage, we do not have enough evidence at this point to rule out the possibility that minor or trace amounts of process-affected water may be present in some of these seeps.
