Characterizing the Organic Composition of Snow and Surface Water Across the Athabasca Region: Phase 2

Jean Birks
Yi Yi
Sunny Cho
Emily Taylor
John Gibson
Resource Date:
Page Length

This study was conducted to characterize the composition of polar dissolved organic compounds present in snow and surface waters in the Athabasca Oil Sands Region (AOSR) with the goal of identifying whether atmospherically-derived organics present in snow are an important contributor to the dissolved organics detected in surface waters in the AOSR. The Phase 1 OSRIN study (2013) was a pilot scale project conducted in 2011-2012 to evaluate whether Electrospray Ionization (ESI) coupled with Fourier Transform Ion Cyclotron Mass Spectrometry (FTICR MS) would be a useful analytical technique to characterize the dissolved organics in snow. Although a limited number of samples (i.e., 7 snow samples) were used in the Phase 1 study, the results indicated differences in organic signatures between the snow samples closest to oil sands activities and the more far-field samples. The Phase 2 project includes a similar comparison of the composition of organics present in snow and surface water as was conducted in Phase 1, but is based on a more spatially and temporally comprehensive set of samples which allows a more extensive investigation of the spatial, temporal and species variations in snow and river water. Phase 2 also combines hydrometric data with the stable isotopic composition of snow and river water to identify when snowmelt appears in river discharge. The dissolved organic composition results identified three snow groups. Group 1 snow tended to have O2 as the dominant compound class, followed by O4 compound classes. The snow samples from locations farthest from industrial activities had Group 1 organic profiles. The organic profiles for Group 2 had O4 as the most abundant compound class and a pattern of decreasing relative contributions from the O4 to O12 classes. There were only six Group 2 snow samples, but they were collected from either the geographical centre (GC) or near mining activities. The remaining snow samples that did not have similar dissolved organic compositions as Group 1 or Group 2 were categorized as Group 3 and were obtained from various locations. The organic profiles obtained for the 110 river samples (84 tributary samples and 26 main stem Athabasca River samples) showed large differences between the composition of dissolved organics present in river water and those present in snow. River samples tended to have a greater relative contribution of O6 to O8 and S2On (n = 4 to 9) compound classes than snow samples. More subtle differences in organic profiles were also evident between the individual river samples related to sampling location and season. Comparing the organic profile results between the river and snow samples show the different types of relationships that exist between river and snow dissolved organic compositions. The monthly river samples collected from the main stem Athabasca River and from one tributary sampling location (i.e., Muskeg 8) tend to have organic compositions that become more similar to Group 1 snow samples over the open water season. The other tributary sampling locations tended to have dissolved organic compositions that become more similar to Group 2 or Group3 snow compositions over the open water season. The river samples differed from snow in that the dissolved organics present in river water are dominated by O6 to O8 classes in oxygen containing compounds, and contain a greater relative contribution S2On (n = 4 to 9). Also, the Athabasca River samples had slightly different organic compositions than the tributaries, with higher relative contributions of O2 class compounds than in the tributaries. The main stem Athabasca River samples also contained some SO3 compounds that were not detected in the tributary samples. All of the river samples showed seasonal variations in dissolved organics, with larger variations in the Athabasca River than in tributaries. The distribution of compound classes in the river samples did not change significantly between May and September, but the dominance of O2 classes becomes more pronounced in September, particularly in the Athabasca main stem sample. The river discharge and stable water isotope data indicate that snowmelt was a major component of the May river samples, but the dissolved organics present in the May river samples did not resemble those present in snow. The months with the greatest similarity between snow and river organic compositions were low flow periods in March, April, and September, which could indicate significant delays between when atmospheric organics are released from the snowpack and when they reach the rivers, or that some of the organics present in snow are similar to organics that characterize baseflow. In summary, the results of this comprehensive profiling of organics in snow and river water across the AOSR suggest that nitrogen and sulphur containing compounds may be the most useful in improving our understanding of the sources and fate of atmospherically derived organics in the oil sands region. There are still some endmembers that need improved organic characterization, including baseflow (groundwater inputs and soil water in disturbed and undisturbed watersheds) to the Athabasca River and its tributaries. Direct sampling of dissolved organics that can be attributed to natural and anthropogenic atmospheric sources of organics (e.g., forest fire, stack emissions, fugitive emissions) are also needed.