Search

Oil and gas wells (OGWs) can lead to soil and well emissions of methane (CH 4 ), a potent greenhouse gas, and hydrogen sulfide (H 2 S), a highly toxic gas, both of which reduce air quality and can cause explosions when emitted into confined spaces. Developments have been occurring over OGWs, posing health and safety risks. However, to our knowledge, previous studies have not conjunctively analyzed well and soil emissions while considering development on or near OGWs. In this paper, we characterize 343 CH 4 and H 2 S emission rate measurements from 67 non-producing (abandoned) and 35 producing (active) OGWs, including 205 measurements from soils surrounding 81 OGWs in Ontario and Quebec. We also provide the first emission rate estimates from an abandoned water and OGW-linked explosion and map OGWs in urban and built-up areas in Ontario and Quebec. We estimate the explosion-linked emissions to be 3,000 g CH 4 /hour and 7 g H 2 S/hour. Moreover, we find that 7,264 and 161 OGWs in Ontario and Quebec, respectively, are in urban and built-up areas, with 94% of these wells being abandoned. For the 102 wells we measured, of which 9.7% had H 2 S detections, we find OGW emission rate ranges of -16 to 47,000 mg CH 4 /hour and 0.001 to 3,300 mg H 2 S/hour. Although soil CH 4 emissions at a 1-m distance from the wells are most correlated with well emissions, the highest soil emission rate was observed at a 3-m distance, indicating the potential for OGW-related emissions into buildings to occur away from the well. Overall, our multi-faceted measurement dataset provides a basis for conjunctive analysis of the broad range of environmental risks of OGWs to climate, indoor and outdoor air quality, and explosions., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 The Authors. Published by Elsevier Inc. All rights reserved.)

Tailings ponds in the oil sands (OS) region in Alberta, Canada, have been associated with fugitive emissions of volatile organic compounds (VOCs) and other pollutants to the atmosphere. However, the contribution of tailings ponds to the total fugitive emissions of VOCs from OS operations remains uncertain. To address this knowledge gap, a field study was conducted in the summer of 2017 at Suncor's Pond 2/3 to estimate emissions of a suite of pollutants including 68 VOCs using a combination of micrometeorological methods and measurements from a flux tower. The results indicate that in 2017, Pond 2/3 was an emission source of 3322 ± 727 tons of VOCs including alkanes, aromatics, and oxygenated and sulfur-containing organics. While the total VOC emissions were approximately a factor of 2 higher than those reported by Suncor, the individual VOC species emissions varied by up to a factor of 12. A chemical mass balance (CMB) receptor model was used to estimate the contribution of the tailings pond to VOC pollution events in a nearby First Nations and Metis community in Fort McKay. CMB results indicate that Suncor Pond 2/3 contributed up to 57% to the total mass of VOCs measured at Fort McKay, reinforcing the importance of accurate VOC emission estimation methods for tailings ponds.

An "event-based" approach to characterize complex air pollutant mixtures was applied in the Oil Sands region of northern Alberta, Canada. This approach was developed to better-inform source characterization and attribution of the air pollution in the Indigenous community of Fort McKay, within the context of the lived experience of residents. Principal component analysis was used to identify the characteristics of primary pollutant mixtures, which were related to hydrocarbon emissions, fossil fuel combustion, dust, and oxidized and reduced sulfur compounds. Concentration distributions of indicator compounds were used to isolate sustained air pollution "events". Diesel-powered vehicles operating in the mines were found to be an important source during NO x events. Industry-specific volatile organic compound (VOC) profiles were used in a chemical mass balance model for source apportionment, which revealed that nearby oil sands operations contribute to 86% of the total mass of nine VOC species (2-methylpentane, hexane, heptane, octane, benzene, toluene, m , p -xylene, o -xylene, and ethylbenzene) during VOC events. Analyses of the frequency distribution of air pollution events indicate that Fort McKay is regularly impacted by multiple mixtures simultaneously, underscoring the limitations of an exceedance-based approach relying on a small number of air quality standards as the only tool to assess risk.

The air-sea gas exchange of alpha-hexachlorocyclohexane (α-HCH) in the Canadian Arctic was estimated using a micrometeorological approach and the commonly used Whitman two-film model. Concurrent shipboard measurements of α-HCH in air at two heights (1 and 15 m) and in surface seawater were conducted during the Circumpolar Flaw Lead study in 2008. Sampling was carried out during eight events in the early summer time when open water was encountered. The micrometeorological technique employed the vertical gradient in air concentration and the wind speed to estimate the flux; results were corrected for atmospheric stability using the Monin-Obukhov stability parameter. The Whitman two-film model used the concentrations of α-HCH in surface seawater, in bulk air at 1 and 15 m above the surface, and the Henry's law constant adjusted for temperature and salinity to derive the flux. Both approaches showed that the overall net flux of α-HCH was from water to air. Mean fluxes calculated using the micrometeorological technique ranged from -3.5 to 18 ng m(-2) day(-1) (mean 7.4), compared to 3.5 to 14 ng m(-2) day(-1) (mean 7.5) using the Whitman two-film model. Flux estimates for individual events agreed in direction and within a factor of two in magnitude for six of eight events. For two events, fluxes estimated by micrometeorology were zero or negative, while fluxes estimated with the two-film model were positive, and the reasons for these discrepancies are unclear. Improvements are needed to shorten air sampling times to ensure that stationarity of meteorological conditions is not compromised over the measurement periods. The micrometeorological technique could be particularly useful to estimate fluxes of organic chemicals over water in situations where no water samples are available.

Shipboard measurements of organohalogen compounds in air and surface seawater were conducted in the Canadian Arctic in 2007-2008. Study areas included the Labrador Sea, Hudson Bay, and the southern Beaufort Sea. High volume air samples were collected at deck level (6 m), while low volume samples were taken at 1 and 15 m above the water or ice surface. Water samples were taken within 7 m. Water concentration ranges (pg L(-1)) were as follows: α-hexachlorocyclohexane (α-HCH) 465-1013, γ-HCH 150-254, hexachlorobenzene (HCB) 4.0-6.4, 2,4-dibromoanisole (DBA) 8.5-38, and 2,4,6-tribromoanisole (TBA) 4.7-163. Air concentration ranges (pg m(-3)) were as follows: α-HCH 7.5-48, γ-HCH 2.1-7.7, HCB 48-71, DBA 4.8-25, and TBA 6.4 - 39. Fugacity gradients predicted net deposition of HCB in all areas, while exchange directions varied for the other chemicals by season and locations. Net evasion of α-HCH from Hudson Bay and the Beaufort Sea during open water conditions was shown by air concentrations that averaged 14% higher at 1 m than 15 m. No significant difference between the two heights was found over ice cover. The α-HCH in air over the Beaufort Sea was racemic in winter (mean enantiomer fraction, EF = 0.504 ± 0.008) and nonracemic in late spring-early summer (mean EF = 0.476 ± 0.010). This decrease in EF was accompanied by a rise in air concentrations due to volatilization of nonracemic α-HCH from surface water (EF = 0.457 ± 0.019). Fluxes of chemicals during the southern Beaufort Sea open water season (i.e., Leg 9) were estimated using the Whitman two-film model, where volatilization fluxes are positive and deposition fluxes are negative. The means ± SD (and ranges) of net fluxes (ng m(-2) d(-1)) were as follows: α-HCH 6.8 ± 3.2 (2.7-13), γ-HCH 0.76 ± 0.40 (0.26-1.4), HCB -9.6 ± 2.7 (-6.1 to -15), DBA 1.2 ± 0.69 (0.04-2.0), and TBA 0.46 ± 1.1 ng m(-2) d(-1) (-1.6 to 2.0).

Organocohlorine pesticide (OCP) residues in agricultural soils are of concern due to the uptake of these compounds by crops, accumulation in the foodchain, and reemission from soils to the atmosphere. Although it has been about three decades since DDT was banned for agricultural uses in Canada, residues persist in soils of some agricultural areas. Emission of DDT compounds to the atmosphere from a historically treated field in southern Ontario was determined in fall 2004 and spring 2005. The sigmaDDTs concentration in the high organic matter (71%) soil was 19 +/- 4 microg g(-1) dry weight. Concentration gradients in the air were measured at 5, 20, 72, and 200 cm above soil using glass fiber filter-polyurethane foam cartridges. Air concentrations of sigmaDDTs averaged 5.7 +/- 5.1 ng m(-3) at 5 cm and decreased to 1.3 +/- 0.8 ng m(-3) at 200 cm and were 60-300 times higher than levels measured at a background site 30 km away. Soil-air fugacity fractions, fs/(fs + fa), of p,p'-DDE, p,p'-DDD, and p,p'-DDT ranged from 0.42 to 0.91 using air concentrations measured above the soil and > or = 0.99 using background air concentrations, indicating that the soil was a net source to the background air. Fractionation of DDT compounds during volatilization was predicted using either liquid-phase vapor pressures (PL) or octanol-air partition coefficients (KOA). Relative emissions of p,p'-DDE and p,p'-DDT were better described by PL than KOA, whereas either PL or KOA successfully accounted for the fractionation of p,p'-DDT and o,p'-DDT. Soil-to-air fluxes were calculated from air concentration gradients and turbulent exchange coefficients determined from micrometeorological measurements. Average fluxes of sigmaDDTs were 90 +/- 24 ng m(-2) h(-1) in fall and 660 +/- 370 ng m(-2) h(-1) in spring. Higher soil temperatures in spring accounted for the higher fluxes. A volatilization half-life of approximately 200 y was estimated for sigmaDDT in the upper 5 cm of the soil column, assuming the average flux rate for 12 h d-(1) over 8 months of the year. Thus, in the absence of other dissipation processes, the soil will continue to be a source of atmospheric contamination for a very long time.

Plume rise parameterizations calculate the rise of pollutant plumes due to effluent buoyancy and exit momentum. Some form of these parameterizations is used by most air quality models. In this paper, the performance of the commonly used Briggs plume rise algorithm was extensively evaluated, through a comparison of the algorithm's results when driven by meteorological observations with direct observations of plume heights in the Athabasca oil sands region. The observations were carried out as part of the Canada-Alberta Joint Oil Sands Monitoring Plan in August and September of 2013. Wind and temperature data used to drive the algorithm were measured in the region of emissions from various platforms, including two meteorological towers, a radio-acoustic profiler, and a research aircraft. Other meteorological variables used to drive the algorithm include friction velocity, boundary-layer height, and the Obukhov length. Stack emissions and flow parameter information reported by continuous emissions monitoring systems (CEMSs) were used to drive the plume rise algorithm. The calculated plume heights were then compared to interpolated aircraft SO2 measurements, in order to evaluate the algorithm's prediction for plume rise. We demonstrate that the Briggs algorithm, when driven by ambient observations, significantly underestimated plume rise for these sources, with more than 50% of the predicted plume heights falling below half the observed values from this analysis. With the inclusion of the effects of effluent momentum, the choice of different forms of parameterizations, and the use of different stability classification systems, this essential finding remains unchanged. In all cases, approximately 50% or more of the predicted plume heights fall below half the observed values. These results are in contrast to numerous plume rise measurement studies published between 1968 and 1993. We note that the observations used to drive the algorithms imply the potential presence of significant spatial heterogeneity in meteorological conditions; we examine the potential impact of this heterogeneity in our companion paper (Akingunola et al., 2018). It is suggested that further study using long-term in situ measurements with currently available technologies is warranted to investigate this discrepancy, and that wherever possible, meteorological input variables are observed in the immediate vicinity of the emitting stacks. [ABSTRACT FROM AUTHOR]