ARTICLE pubs.acs.org/est
Blinded Taste Panel Evaluations To Determine if Fish from near the Oil Sands Are Preferred Less than Fish from Other Locations in Alberta, Canada Brenda Barona,† Rozlyn F. Young,‡ Phillip M. Fedorak,*,‡ and Wendy V. Wismer† † ‡
Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, Alberta, Canada T6G 2P5 Department of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E9 ABSTRACT: The oil sands industry is rapidly expanding surface mining and bitumen extraction operations near the Athabasca River in northeastern Alberta, Canada. There are anecdotal comments that the fish from the Athabasca River have an off-taste, implying that the oil sands operations are the cause. This study was done to determine if the taste of wild fishes caught near the Athabasca oil sands was less preferred than the taste of fishes collected from two other river basins in Alberta. In blinded experiments, consumer sensory panels, of 40 to 44 participants, tasted steamed samples of each of three fish species (walleye (Sander vitreus), northern pike (Esox lucius), and lake whitefish (Coregonus clupeaformis)) from three different sources in Alberta (the Athabasca River, Buck Lake, and McGregor Lake). Data analyses showed that there was no evidence from the consumer preference rankings that the taste of the fish from the Athabasca River was preferred less than the taste of fish from two other water bodies in Alberta.
’ INTRODUCTION The Athabasca oil sands deposit in northeastern Alberta, Canada is one of the largest petroleum deposits in the world.1 Recently, the Alberta oil sands have drawn considerable international investment and attention. Many countries realize the importance of this vast petroleum reserve, and large international companies are developing these resources. Much attention has also been focused on environmental issues that accompany petroleum extraction from the oil sands. Some of the concerns are largescale land disruption, air quality, and water use with the potential for seepage of the extraction waters into nearby rivers. The Athabasca deposit contains highly biodegraded, viscous, tar-like petroleum, known as bitumen. Erosion by the Athabasca River and its tributaries has led to the exposure of bitumen-containing outcrops in some river valleys. Conly et al.2 provide a geological cross-section that illustrates how this erosion cuts into the oil sands bearing stratum. These outcrops were the first indication of the presence of oil sands. Full-scale surface mining and bitumen extraction activities began in 1967, and for about 35 years, two major oil sands companies operated just north of Fort McMurray, Alberta. Initially, both operations were on the west bank of the Athabasca River, but now mining has expanded to the east side of the river, directly across from the original site. Over the past decade, many international petroleum companies have invested in the oil sands industry, and new mining and extraction operations have recently started or are being planned. Many of the leases are adjacent to rivers in this area, and their operations have the potential to alter water quality. In general, the extraction of bitumen from the mined Athabasca oil sands is done using a dilute caustic solution in water which produces tailings consisting of water, sand, clay, and residual bitumen. These tailings are held in vast settling ponds, and the clarified cap water is recycled into the extraction process. Under r 2011 American Chemical Society
current regulations, none of these tailings waters can be released into the Athabasca River system because of their toxicity to fishes and other aquatic organisms. The Athabasca River originates from glaciers in the Rocky Mountains of Alberta and it flows northeast into Lake Athabasca (Figure 1), a distance of about 1230 km.3 Lake Athabasca supports a commercial fishery, and Jardine and Hrudey4 reported that this fishery was closed in 1982 after an upset at one of the oil sands plants during the winter of 1981 to 1982 caused the release of hydrocarbons into the Athabasca River. At that time, fish were reported to have “petroleum-like off-flavors”.4 Since then, fish raised or held in oil sands tailings waters have been used in several studies to assess the potential of these waters to taint fish.5-8 Results of these studies, summarized by Rogers et al.,9 have shown that dilutions of oil sands process-affected waters cause fish tainting that can be detected by a sensory panel. Upstream of the oil sands, there are five pulp and paper mills on the Athabasca River and its tributaries.10 As summarized in a review by Kenefick et al.,11 fish tainting has been also attributed to pulp and paper mills effluents. Thus, there are multiple potential sources of inputs into this river system that could cause fish taint. Fishes used in this study were collected close to the oil sands operations to focus potential tainting on this source. According to GESAMP12 the principal components of petroleum causing taint include phenols, mercaptans, methylated naphthalenes, dibenzothiophenes, tetradecanes and naphthenic acids. These can all be found in oil sands tailings waters, and Kelly et al.13 found that airborne particles in the vicinity of the oil sands operations contribute many polycyclic aromatic compounds to the surrounding watershed. Jardine and Hrudey4 examined selected Received: October 4, 2010 Accepted: December 21, 2010 Published: January 19, 2011 1730
dx.doi.org/10.1021/es103359f | Environ. Sci. Technol. 2011, 45, 1730–1736
Environmental Science & Technology
ARTICLE
Figure 1. Map of Alberta with river basins and sampling locations: X = Athabasca River near oil sands, Buck Lake, McGregor Lake.
compounds found in oil sands process-affected waters (i.e., pxylene, benzothiophene, dibenzothiophene, naphthalene, 2,6dimethylnaphthalene, 2,3,5-trimethylnaphthalene, and 2,3-dimethylphenol) and demonstrated that these aromatic compounds could cause fish tainting. In addition, Koning and Hrudey5 found phenol, cresols, and dimethylphenols in oil sands tailing waters and identified these as potential fish tainting compounds. Recently, Edge et al.14 demonstrated that some commercial naphthenic acids and naphthenic acids extracted from an oil sands experimental reclamation pond can cause a detectable odor (taint) in steamed walleye fillets (Sander vitreus). There are anecdotal comments in the media and on the Internet that report off-flavor or off-odor in fish caught near the Athabasca oil sands. For example, K€ohler15 reported “fish pulled from the Athabasca downstream of the oil sands taste of gasoline and smell of burning galoshes in the fry pan”. However, there is no clear documentation that wild fishes taken from natural waters near the Athabasca oil sands operations are tainted, either by seepage from various nonpoint sources related to mining and extraction or by natural erosion of bitumen, or by deposition of airborne polycyclic aromatic compounds. Thus, we sought to determine if a sensory panel of consumers could distinguish between wild fishes collected from near the Athabasca oil sands and wild fishes collected
from two other Alberta locations in different river basins distant from the Athabasca oil sands. The use of sensory panels was imperative because at this time there is no information about which organic chemicals might be causing an off-flavor. We hypothesized that if tainted fish were common near the Athabasca oil sands, a sensory panel would judge fish from the Athabasca oil sands region to be the “least liked” of the fish samples from the three source locations. In these blinded experiments, consumer sensory panelists were presented with samples of the three major fish species eaten by people in the Athabasca region. These species are walleye (Sander vitreus, formerly Stizostedion vitreum), northern pike (Esox lucius), and lake whitefish (Coregonus clupeaformis) (Brenda Miskimmin, Summit Environmental Consultants Ltd., personal communication, 2009). These three species are also important to the commercial fishery in Alberta, and thus, these three species of fish were collected from three different Alberta river basins for presentation to a sensory panel.
’ EXPERIMENTAL SECTION Sampling Locations. All of the wild fishes used in this study were collected in September 2008. Figure 1 shows three of the river basins in Alberta, Canada and the locations from which the fish were gathered. Fish were obtained from the Athabasca River 1731
dx.doi.org/10.1021/es103359f |Environ. Sci. Technol. 2011, 45, 1730–1736
Environmental Science & Technology
ARTICLE
at a few different locations near the oil sands deposits and mining operations as part of the Regional Aquatics Monitoring Program (RAMP) in Alberta. Fish were also collected from McGregor Lake and Buck Lake during Fall Index fish netting by the Province of Alberta, as part of the Sustainable Resource Development Ministry’s Fish and Wildlife Fisheries Management program. McGregor Lake is in the South Saskatchewan River Basin, and Buck Lake is in the North Saskatchewan River Basin.16 Based on total phosphorus concentrations, McGregor Lake is classified as mesotrophic and Buck Lake is classified as eutrophic.17 Based on total chlorophyll a concentrations, McGregor Lake is classified as eutrophic and Buck Lake is classified as hypereutrophic.17 These two lakes are more than 450 km from Fort McMurray. Fish Collection and Storage. Wild walleye, northern pike, and lake whitefish were collected from each location. To mimic the manner in which anglers take fish from Alberta waters, fish of the appropriate lengths were kept regardless of the age or gender of the fish. Personnel from RAMP provided gutted, whole fish collected from the Athabasca River. These fishes were stored on ice in coolers and transported to the University of Alberta. Whole fish from McGregor Lake and Buck Lake were gutted immediately in the field before placing them in individual plastic bags. The bagged fish were stored on ice in coolers for transportation to the University of Alberta. At the University, the fish were kept on ice in coolers that were stored in a walk-in refrigerator at 4 °C. Rodríguez et al.18 reported that acceptable freshness is maintained for up to 6 days when gutted fish are stored on ice. All of Table 1. Summary of the Numbers of Each Fish Species and Mass of Fillets from the Three Different Sources Athabasca
Buck
McGregor
River
Lake
Lake
6 3.3
3 2.8
7 2.9
species lake whitefish number of fish caught total mass of fillets (kg) walleye
number of fish caught
3
8
8
total mass of fillets (kg)
1.9
2.0
3.3
northern pike number of fish caught total mass of fillets (kg)
3
5
7
2.2
2.7
3.7
our fish were filleted in a food-grade facility within 6 days of being caught. Two fillets were obtained from each fish. Each fillet was rinsed with cold tap water, placed in a separate plastic bag, and frozen for use in sensory studies 4 months later. Previous investigations19-21 have shown that frozen fish stored for 8 to 12 months remains acceptable to consumers. Table 1 summarizes the numbers and total masses of fillets of each fish species collected at each sample location. Consumer Sensory Panel Evaluation. Three consumer sensory panels were conducted in the food sensory testing facilities at the University of Alberta; one for each of the fish species (walleye, northern pike, or lake whitefish). Steamed fish samples were presented to sensory panelists to determine if there were any preference differences among fish from the three geographic sources. Panel sizes ranged from 40 to 44 individuals, a sufficient number of participants for the simple and intuitive nature of the ranked preference test.22 Panelists were informed that they would receive fish samples from each of northern, central, and southern Alberta but were not told the identity of the three river basins. The consumer sensory panelists were untrained, over 18 years of age and recruited through posters, in classes and from outdoor recreation clubs to meet the criterion of having consumed all or one of the fish species within the past year (Table 2). Often participants consumed these fish species more frequently than once per year. Most of the consumed fish were caught by the participants or obtained from friends or relatives. The consumer panel participants were nearly evenly divided between the genders, and as expected at a university, tended to be younger than 35 years of age (Table 2). For the sample preparation, one fillet from each fish was thawed overnight in a refrigerator at
