An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands

S. G. Howell; A. D. Clarke; S. Freitag; C. S. McNaughton; V. Kapustin; V. Brekovskikh; J.-L. Jimenez; M. J. Cubison

doi:10.5194/acp-14-5073-2014

Howell, S. G.;Clarke, A. D.;Freitag, S.;McNaughton, C. S.;Kapustin, V.;Brekovskikh, V.;Jimenez, J.-L.;Cubison, M. J.;

2014-05-23 00:00:00

Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ doi:10.5194/acp-14-5073-2014 © Author(s) 2014. CC Attribution 3.0 License. An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands 1 1,2 2 1,* 1 1 3 S. G. Howell , A. D. Clarke , S. Freitag , C. S. McNaughton , V. Kapustin , V. Brekovskikh , J.-L. Jimenez , and 3,** M. J. Cubison Department of Oceanography, University of Hawaii, Honolulu, Hawaii, USA Department of Oceanography, University of Hawaii, Honolulu, Hawaii, USA Cooperative Institute for Research in the Environmental Sciences and Department of Chemistry and Biochemistry, University of Colorado, Boulder, Colorado, USA now at: Golder Associates Ltd., Saskatoon, Saskatchewan, Canada ** now at: Tofwerk AG, Thun, Switzerland Correspondence to: S. G. Howell ([email protected]) Received: 31 July 2013 – Published in Atmos. Chem. Phys. Discuss.: 15 August 2013 Revised: 1 February 2014 – Accepted: 5 March 2014 – Published: 23 May 2014 Abstract. During the Arctic Research of the Composition be active. Altogether, organic aerosol and black carbon emis- of the Troposphere from Aircraft and Satellites (ARCTAS) sions from the oil sands operations are small compared with campaign, two NASA research aircraft, a DC-8 and a P- annual forest ﬁre emissions in Canada. The oil sands do con- 3B, were outﬁtted with extensive trace gas (the DC-8) and tribute signiﬁcant sulfate and exceed ﬁre production of SO aerosol (both aircraft) instrumentation. Each aircraft spent by an order of magnitude. about a half hour sampling air around the oil sands mining and upgrading facilities near Ft. McMurray, Alberta, Canada. The DC-8 circled the area, while the P-3B ﬂew directly over 1 Introduction the upgrading plants, sampling close to the exhaust stacks, then headed downwind to monitor the aerosol as it aged. Canada’s oil sand deposits represent 30 % of total world At short range, the plume from the oil sands is a complex oil reserves (Alboudwarej et al., 2006) and are estimated at mosaic of freshly nucleated ultraﬁne particles from a SO - about 180 billion barrels. Most of these resources are in Al- and NO -rich plume, soot and possibly ﬂy ash from indus- berta near the Athabasca River. However, the bitumen con- trial processes, and dust from dirt roads and mining opera- tained within the sand is extremely viscous, requiring heat or tions. Shortly downwind, organic aerosol appears in quanti- solvents to extract from the sand. The surface mining oper- ties that rival SO , either as volatile organic vapors condense ators (e.g., Syncrude Canada, Suncor Energy, Albian Sands or as they react with the H SO . The DC-8 pattern allowed 2 4 Energy) extract the bitumen using a hot water process. About us to integrate total ﬂux from the oil sands facilities within 80 % of the deposits are not recoverable by surface mining about a factor of 2 uncertainty that spanned values consistent and require in situ recovery using steam injection. Before the with 2008 estimates from reported SO and NO emissions, 2 2 bitumen can be sent through pipelines and reﬁned, it must be though there is no reason to expect one ﬂyby to represent av- upgraded, a combination of processes that consume natural erage conditions. In contrast, CO ﬂuxes exceeded reported gas and produce synthetic crude oil and CO . regional emissions, due either to variability in production or There are many sources of aerosols and aerosol precur- sources missing from the emissions inventory. The conver- sors in the oil sand extraction and upgrading processes. The sion rate of SO to aerosol SO of ∼ 6 % per hour is con- 2 4 bitumen itself releases SO , H S and light hydrocarbons as 2 2 sistent with earlier reports, though OH concentrations are in- well as CO and CO on heating (Strausz et al., 1977). Sur- sufﬁcient to accomplish this. Other oxidation pathways must face mining releases dust directly and road dust and soot are Published by Copernicus Publications on behalf of the European Geosciences Union. 5074 S. G. Howell et al.: Particulate emissions from Athabasca oil sands produced by trucks. Since the Athabasca bitumen contains approximately 5 % sulfur and 0.5 % nitrogen, the upgrad- ing process can release large quantities of these as SO and NO as well as soot. The SO forms sulfate aerosol at up to 2 2 6 % per hour (Cheng et al., 1987). Fly ash has been docu- mented from a coke-burning power plant associated with the upgraders (Barrie, 1980). Development of the oil sands began with the Great Cana- dian Oil Sands Company in 1967 but started expanding rapidly around 2000. As production has increased, the in- dustry has invested heavily in emissions abatement equip- ment on existing facilities and state-of-the-art low-emission technologies for new facilities in order to ensure regional air quality stays within regulated limits. Oil sands opera- tors are also required to report emissions to the Environ- miles 10 km 20 Figure 1. Imagery of the oil sands region from 29 June and ﬂight ment Canada’s National Pollutant Release Inventory (http: tracks from 10 July. The DC-8 track is blue; the P3B red. The two //www.ec.gc.ca/inrp-npri/) and fund air quality monitoring planes entered the area from the north along the parallel tracks to the in the region. As part of their response to environmental is- right. The DC-8 then looped counterclockwise around the facilities sues, the oil sands operators have provided capital and on- at about 350 m a.g.l; 90 min later the P-3B approached from the going support for the Wood Buffalo Environmental Asso- south and ﬂew the ﬁgure 8 shown, penetrating the Syncrude facility ciation (WBEA), which is a collaboration of communities, plume three times, twice at about 250 m a.g.l. and once at 600 m. environmental groups, industry, government and aboriginal The Suncor upgrader plume was sampled once, at 250 m. stakeholders. WBEA operates an environmental monitoring program that measures the ambient air quality at about 15 sta- tions throughout the area and continuously monitors environ- primarily of light hydrocarbons, presumed to be either from mental effects of air emissions through the Terrestrial Envi- the bitumen or the solvents used to mobilize it. The other was ronmental Effects Monitoring Program, which addresses is- a more typically industrial plume with high concentrations sues such as soil acidiﬁcation, trace metals in foods harvested of NO and SO together with CO, CO , and a variety of 2 2 2 by aboriginal communities, and vegetation stress. Data from alkanes, solvents, and halocarbons. The work reported here many of these sites can be accessed directly through the extends that analysis to include aerosol emissions. WBEA website (http://www.wbea.org/). One of the unusual features of the oil sands facilities is their location in an area lacking large populations and other industrial facilities. This makes the emissions more obvious 2 Experimental against a relatively clean background. Hence, the oil sands emissions can be discerned in global satellite maps of NO The ﬂights around the Ft. McMurray oil sands upgrading and SO (McLinden et al., 2012) even though total emissions facilities took place on 10 July 2008. The DC-8 and P-3B are smaller than many urban regions. This input of pollutants took the opportunity to do a joint ﬂight over the oil sands into a nearly pristine area has been of concern since develop- to perform an instrument intercomparison and a brief eval- ment began in the area. Studies of aerosol formation (Cheng uation of the aerosol and gas emissions from the facilities et al., 1987), deposition (Barrie, 1980; Proemse et al., 2012b) there (Fig. 1). The DC-8 ﬂew a loop around the facilities at and composition (Proemse et al., 2012a) have been ongoing about 350 m above ground level (a.g.l.), passing through the as have air quality modeling efforts (Davies, 2012). entire plume and providing upwind samples for contrast. The Despite this emphasis on measuring oil sands emissions, P-3B ﬂew a ﬁgure-8-shaped pattern, passing through visible there had been no recent evaluations of them using an exten- plumes from the tallest stacks in the facilities (Fig. 2). The sive airborne instrumentation suite until the summer of 2008, Syncrude plume was sampled twice at about 240 m a.g.l. and when the NASA DC-8 and P-3B research aircraft were de- once at 625 m, while the Suncor plume was sample once, at ployed at the Canadian Forces Base Cold Lake in Alberta, about 250 m. After the third Syncrude plume penetration, the Canada. The main focus of the project was to study smoke P-3B turned downwind and followed the evolving plume. plumes from the forest ﬁres that occur every year in north- In addition to the July 10 ﬂights, there were 3 plume pen- ern Canada (Jacob et al., 2010), but the two planes detoured etrations on other ARCTAS (Arctic Research of the Compo- one day to measure aerosol and trace gas emissions from the sition of the Troposphere from Aircraft and Satellites) ﬂights facilities near Ft. McMurray. (Fig. 3). The P3-B descended into the mixed layer during The gas-phase data revealed two different types of plumes transits from ﬁre plume studies on 28 and 29 June, while (Simpson et al., 2010). One was a broad plume that consisted the DC-8 ﬂew a low pass early on 28 June looking for a ﬁre Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5075 58.5 P-3 6/28 P-3 7/10 58.0 DC-8 6/29 P-3 aging run 7/10 57.5 DC-8 lidar pass 7/10 DC-8 7/10 57.0 P-3 6/29 -111 -110 -109 -108 Longitude Figure 2. The Suncor facility about 12 s before the P-3B encoun- tered the white/gray plume. Stacks emitting darker plumes are visi- Figure 3. Flight tracks during ARCTAS that passed near the oil ble to the left. The Athabasca River, settling ponds, and mined areas sands facilities. Red traces are the P-3B; DC-8 tracks are blue. Em- are visible in the background. The image is a still from a windshield- phasized sections are where the aircraft intercepted the plume. mounted video camera. so high that multiple particles were in the sensing volume si- plume from the previous day and coincidentally met the oil multaneously and were not counted properly. The counters sands plume. used for CNcold and CNhot do not compensate for coinci- 3 −3 dent particles, which becomes a problem at > 10×10 cm . 2.1 Instrumentation A correction algorithm was applied (Baron and Willeke, 3 −3 The two aircraft had complementary payloads. The DC-8 2001), but the algorithm fails above about 35× 10 cm , was outﬁtted with an extensive suite of gas-phase and aerosol when particles are in the sampling volume ≥ 50 % of the instruments. The P-3B had similar aerosol measurements time. The UCN counter automatically compensates for co- 3 −3 and several radiation propagation measurements, but no gas- incident particles, but saturates at 100× 10 cm . phase capability beyond CO and O . Both aircraft used the Aerosol size distribution measurements differed modestly aerosol inlet characterized by McNaughton et al. (2007), between the platforms. Both used scanning mobility parti- who found that it efﬁciently conveyed dust particles of up cle sizers (SMPSs, using TSI long DMA (differential mo- to a few microns in diameter. Additional information can be bility analyzer) model 3081s) for particles from ∼ 10 to found at http://www.espo.nasa.gov/arctas/airborne_inst.php. 400 nm, optical particle counters (OPCs) for rapid measure- Rather than describing the entire set of measurements, a short ment of the accumulation mode (0.2 to 1 μm), and aerody- description of the most relevant ones is included here. namic particle sizers (APS, TSI model 3321) to measure Number concentration measurements were almost identi- aerodynamic particle diameters between ∼ 0.8 μm < D < ae cal on both aircraft. The concentration of particles >3 nm 10 μm. The DC-8 used a DMT UHSAS (ultra-high sensitivity were measured with TSI model 3025A ultraﬁne condensa- aerosol spectrometer) OPC, which resolves smaller particles tion nuclei (UCN) counters. Particles&10 nm were measured (0.055 to 1 μm) at a higher rate (1 Hz) than the modiﬁed LAS- with a pair of TSI model 3010 condensation nuclei (CN) X OPC on the P3-B (0.2 to 3 μm, 0.33 Hz), though the latter counters operated with a temperature difference of 22 C be- had a set of heated inlets that could explore volatility at 150, tween saturator and condenser. One (called CNcold hence- 300, and 400 C (residence times > 1 s). The P-3B also had forth) had no additional heat applied to the inlet, while the an SMPS with a TSI radial DMA with unheated and 300 C, other had an inlet tube heated to 350 C for about 0.1 s to 0.1 s inlets that spanned 10 to 200 nm. The two SMPSs on remove most volatile material (CNhot). CNhot is typically the P-3B used grab samplers (Clarke et al., 1998) to ensure associated with combustion, as ﬂames produce nonvolatile that size distributions were from a single sample of air, not black carbon (BC) and organic material, though dust, sea affected by changes in airmass over the scanning period. salt, and volatile particles too large to evaporate entirely can Aerosol composition measurements were similar on the also contribute (Clarke, 1991). Differences between UCN two aircraft. Nonrefractory submicron composition was mea- and CNcold show particles in the 3 to 10 nm range indicative sured with Aerodyne high-resolution time-of-ﬂight aerosol of recent new particle formation (nucleation). UCN and CN- mass spectrometers (AMSs) with vaporizers at 600 C (De- cold suffered from saturation of the counters–concentrations Carlo et al., 2006). The P-3B AMS ran on an approximately www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014 Latitude 5076 S. G. Howell et al.: Particulate emissions from Athabasca oil sands 40 % duty cycle, blanking 2 s, sampling for 2 s, and taking the aircraft at 591 and 1064 nm. Data are presented here as about a second to prepare for the next cycle. The DC-8 ran in attenuated aerosol backscatter ratio, deﬁned as one less than so-called “fast mode” (Kimmel et al., 2011) with a duty cy- the ratio of lidar return L to that calculated for an atmosphere cle of about 75 %. Both instruments used pressure-controlled lacking aerosol L . inlets (Bahreini et al., 2008) to stabilize conditions at the L [β + β ] β p m p aerodynamic lens. The P-3B instrument was controlled at 2 R ≡ −1 = (1−ε ) −1 ≈ if ε 1, (1) ba p p L β β 600 hPa during the periods reported here, while the DC-8 was m m m at 200 hPa. At 780 hPa, Liu et al. (2007) found that the AMS where β and β are backscattering due to particles and p m effectively sampled particles from roughly 60 to 700 nm vac- molecules (air), respectively, and ε is extinction along the uum aerodynamic diameter (50 % sampling efﬁciency cut- beam due to particles. For the qualitative analysis here, this offs). Note that vacuum aerodynamic diameter (D ) is in- VA approximation is sufﬁcient, but note that R is an underes- ba versely proportional to density, so cuts are smaller for typical timate of the actual backscatter ratio when ε is signiﬁcant, atmospheric particles, probably 30 or 45 nm at the low end to which often occurs when the DIAL penetrates thin clouds. roughly 400 for the P3-B (the DC-8 50 % size cut for large particles was measured at about 1 μm D , or about 700 nm VA actual diameter.) 3 Results and discussion The measurements used here are bulk (not size-resolved) SO , NO , NH and nonrefractory organic aerosol (OA) 4 3 4 The 10 July 2008 ﬂights provide the bulk of our data. It mass concentrations extracted using the publicly available was a clear day with scattered low clouds and some higher Squirrel software (Allan et al., 2003, 2004; DeCarlo et al., clouds in the distance (Fig. 2). Surface winds below 1 km −1 2006) and the collection efﬁciencies described in Middle- were from the SSW at 3 to 7 m s (near the climatological brook et al. (2012). The AMS has difﬁculty separating −1 mean) but increased with altitude to 15 m s at 1 km and organic from inorganic sulfates, nitrates, and ammonium −1 about 20 m s at 3.5 km. Above 1 km winds transitioned to (Farmer et al., 2010); we refer to those radicals without ◦ about 215± 15 . Convective activity and clouds were present charges (e.g., SO as opposed to SO ) as a reminder that we 4 in the area, but no precipitation was evident. Some clouds in- may not be measuring only inorganic ions. teracted with emission plumes, apparently mixing them to BC was measured with DMT single-particle soot pho- higher altitudes. tometers (SP2s; Schwarz et al., 2006; Stephens et al., 2003). Unfortunately, the DC-8 SP2 was not operational on 10 July. 3.1 Near-ﬁeld characterization Particle sizes that can be detected by SP2s depend on details of instrument setup, but generally particles below 100 nm are The closest plume penetrations by the P-3B were too quick not always detected and 80 nm particles are missed entirely, for the PSAPs and size distributions, but the rapid-response as their incandescence is too faint to be detected. For typical measurements showed the plumes clearly (Fig. 4). SO and soot plumes, SP2s detect nearly the entire mass of BC, but CO were found almost exclusively in the visible plumes from can miss a signiﬁcant fraction of the BC number concentra- the tallest stacks. In contrast BC and OA appear to be pro- tion. The DC-8 had additional aerosol composition measure- duced in other locations as well, probably including the ﬂar- ments, but at rates too slow to capture the oil sands plume. ing stacks visible at the lower left in Fig. 2. Scattering and The aerosol optics packages on both aircraft included TSI CN show both inﬂuences, dominated by the tall stacks but model 3563 nephelometers (Anderson et al., 1996) measur- with other sources evident. This shows that the plumes from ing light scattering B at 450, 550, and 700 nm and Radiance the upgrading facilities are spatially heterogeneous, with a sp Research particle soot absorption photometers (PSAPs) ob- variety of plume compositions. taining light absorption B at 470, 530, and 660 nm. Neph- The behavior of organic aerosol after the last penetration ap elometer data were corrected for truncation errors using the is striking. Little OA was emitted directly – there were only procedures from Anderson and Ogren (1998). PSAP correc- small ﬂuctuations during the close passes – but OA concen- tions were performed according to Virkkula et al. (2005) and trations jumped to almost equal SO as soon as the plane Virkkula (2010). turned downwind at 20:15 (this and all times mentioned As mentioned above, the DC-8 had an extensive set of are UTC, 6 h later than local daylight savings time). The trace gas measurements. Of particular interest here are CO, roughly 1:1 ratio between OA and SO persisted for the rest SO , NO , NO , and a suite of hydrocarbons and halocar- of the leg.(Fig. 4). It is conceivable but unlikely that OA was 2 2 x bons collected in stainless steel chambers and measured with present at diameters too small for the AMS to detect; that gas chromatography at UC Irvine. More extensive gas-phase would require a huge population of particles growing more data are reported in Simpson et al. (2010). slowly than the freshly nucleated sulfates. The origin of the The DC-8 was equipped with a differential absorption li- organic matter is not clear; the high correlation with SO sug- dar (DIAL; Browell et al., 1998; Dupont et al., 2012) which gests a common source, but it is conceivable that organic ma- provides a 2-D curtain of backscatter data above and below terial was mixing up into the SO plume from lower-altitude Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ GPS alt B CO SO ap 2 2 -1 ppmv pptv km sMm Pressure altitude, km S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5077 A B C D NCAR 5000 UCB 0 3 Total SO SO OA NH OA 0 0 1000 400 1 0 375 5 3 Caltech 10 100x10 GA Tech 0 0 10 UCN CNcold CNhot 18:30 18:40 18:50 20:05 20:10 20:15 20:20 20:25 10 July 2008 UTC 10 July 2008 UTC Figure 5. The oil sands plume as seen by in situ instruments on Figure 4. Data along the P-3B ﬂight track shown in Fig. 1. Syn- the DC-8. The light-blue region was when the plane was upwind crude upgrader plume penetrations occurred at points A, B, and D, of the facilities, while light brown indicates the plume. The plane while the Suncor plume was encountered at point C. Light-blue re- ﬂew parallel to the wind at 18:34, so no ﬂux could be calculated. gions indicate periods when the plane was among clouds; darker The gap at 18:43 is there because of the NO and B spikes, which 2 ap blue shows cloud penetrations. Note that the UCN and CN saturate appear consistent with diesel truck trafﬁc. The period from 18:25 5 −3 4 −3 at 1× 10 cm and 3× 10 cm , respectively; actual values are to 18:28 was downwind of both mining operations and the town of higher than shown. The AMS was on a 5 s cycle, blanking half the Ft. McMurray. Scattering, B , is at 550 nm, while absorption, B , sp ap time, so it may have missed peak concentrations. CN, CO, and scat- is at 532 nm. tering (B ) are on log scales; CO has the minimum concentration sp of 93.6 subtracted. The light-blue background shows periods among clouds; darker blue indicates cloud penetrations. Light scattering, sociated with close approaches. Similar pulses were present B , is at 550 nm. sp in and between clouds during the descent (blue regions in Fig. 4). It is not clear how much of the sulfate in and around sources. Perhaps the extensive vapor-phase measurements of clouds was due to the well-known aqueous-phase reaction of the DC-8 could distinguish between those possibilities, but SO with H O (Penkett et al., 1979). In this case the reac- 2 2 2 the P-3B payload lacked that capability. It is also not im- tion is limited by the roughly 500 pptv concentration of H O 2 2 −3 mediately apparent whether the organic aerosol was simply seen by the DC-8, which could produce as much as 2 μg m condensation of semivolatile vapors as the plume cooled with SO . Reaction of SO with NO in cloud droplets could be 4 2 2 altitude (and is thus primary OA) or whether photochemi- a signiﬁcant source of SO (Littlejohn et al., 1993; Sarwar cal reactions could have produced secondary organic mate- et al., 2013), though this is a pH-dependent reaction that may rial in the short period since emission. Elemental analysis be limited by available NH (which was not measured). of the AMS mass spectra revealed median O : C= 0.26 and Figure 5 shows a similar data set from the DC-8. At the H : C= 1.5, characteristic of fresh mixed primary and sec- 10 km distance of this sampling, the plume is more mixed, ondary OA (Aiken et al., 2008), which is also consistent with but some of the same features are visible. There is a double- the mass spectral pattern observed by Ng et al. (2011). hump structure to many of the species. Given the winds and It is likely that reactions with H SO are responsible for ﬂight direction, the ﬁrst peak is from Suncor and the second 2 4 much of the OA. The high acidity of the plume could catalyze from Syncrude. As with the P-3B, SO was found in a nar- polymerization of aldehydes (Jang et al., 2002), react with rower region, indicating a couple of discrete sources. SO be- alkenes such as isoprene (Surratt et al., 2007) or react with haved similarly, while other species were more widely spread alcohols such as 2-methyl-3-buten-2-ol (Zhang et al., 2012). out. NH was also higher in the plume, but not sufﬁciently to While much of the organic vapor in the oil sands plume was neutralize the SO . NH : SO dropped from ∼1.5 in back- 4 4 4 aliphatic and thus unreactive, alcohols and alkenes such as ground air to 0.5 in the heart of the plume, indicating that isoprene, α-pinene, and β -pinene were enhanced (Simpson NH was limited and the plume was acidic. Before the main et al., 2010). Biogenic vapors from the surrounding forest plumes, from 18:25 to 18:28, scattering, absorption, OA, CO, may also have contributed. CO , and CNhot were all moderately enhanced, due either to As part of the downwind leg, the P-3B explored the ver- mining or the city of Ft. McMurray. In contrast, NO was tical structure of the plume by ascending through the mixed quite low during that period. layer and the cloud layer, then proﬁling back down to the Size distributions in the plume seen by the DC-8 are com- mixed layer. Cloud base was about 2.2 km. During the ascent, plex (Fig. 6). The APS (top panel) shows modes at 3 μm at clouds were avoided, but pulses of aerosol were clearly as- 18:30 and 18:32. That is typical for dust made mechanically www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014 CN ΔCO B AMS BC sp -3 -3 -3 -1 ppbv scm µg sm ng sm sMm CNhot ∆CO B AMS NO sp 2 -3 -3 -1 pptv ppbv µg sm scm sMm Scattering CN -1 -3 sMm scm dV / d log d 5078 S. G. Howell et al.: Particulate emissions from Athabasca oil sands 3.2 Vertical structure 10 60 Total <1µm Although aircraft have the ability to survey in 3 dimensions, in situ measurements are limited to a 1-D line of observations along the ﬂight track. A lidar curtain is an effective way to add a dimension, and thus context, to aerosol data. However, because aerosol size depends upon water uptake in response 100x10 UCN to relative humidity that generally varies with altitude (Shi- CNcold 20 nozuka et al., 2011), the backscatter cannot directly quantify CNhot 50 emitted dry aerosol mass. The DC-8 lidar had difﬁculty pen- etrating clouds in some spots, but did provide an overview of 0.1 the plume (Fig. 7). Figure 7b relates the lidar curtain and location of the plume 18:29 18:30 18:31 18:32 18:33 18:34 to winds and geography. The wind vectors do not line up 10 July 2008 UTC precisely with the 18:03 start of the plume or maximum at 18:06 as seen by the lidar, but small changes in wind direc- Figure 6. Aerosol volume superimposed on some of the data from Fig. 5. The upper panel shows total and submicron light scattering at tion are likely in the few hours it took the plume to reach the 550 nm and the coarse aerosol volume distributions measured by the lidar track. The backscatter peak between 18:10 and 18:11 APS. The lower panel shows UCN, CNcold and CNhot over sub-μm is probably due to mining operations (see Fig. 1), but its volume distribution from the UHSAS. Pale traces show UCN and magnitude is deceptive – the cloud at 4 km between 18:05 CNcold saturation; true values are higher. and 18:09 suppresses the lower-altitude backscatter. The con- tinued plume to 18:13:30 is nearly downwind of the city of Ft. McMurray. Aerosol in the main plume appears to be and is likely to be road dust, from mining, or from crush- mixed fairly uniformly to about 1450 m, but is drawn higher ing or mixing gravel. The mode peaking at 1 μm around by convection around 18:05, presumably in a manner similar 18:31:40 is more unusual – too small for road dust (e.g., to that seen by the P-3B among clouds in Fig. 4. Singh et al., 2002) or natural dust, even after long transport (e.g., Mahowald et al., 2014; Maring et al., 2003), and much 3.3 Fluxes larger than the typical accumulation mode of 0.1 to 0.3 μm from gas-to-particle conversion. The latter is also seen in the One of the advantages of sampling from a mobile platform is UHSAS data at that time (bottom panel) at about 0.6 μm. that one can cross the entire plume and integrate a total ﬂux Since aerodynamic diameter as measured by the APS is re- if the vertical structure is known. The ﬂux in the direction lated to geometric diameter by approximately the square of the wind through the vertical projection of the ﬂight track root of density, this implies a particle density of at least between times t and t is 0 1 −3 2.8 g cm (or higher if the refractive index caused the t z 1 m Z Z UHSAS to oversize the particles). This could be ﬂy ash Q = (S − S )v v sin(φ − φ )dz dt, (2) 0 w a a w from petroleum coke combustion, as was reported by Bar- 0 0 rie (1980). Total and submicrometer light scattering is also plotted in the top panel. It is no surprise that coarse-mode where z is the top of the boundary layer, S and S are the m 0 scattering is high during the dust events. However, maxi- plume and background concentrations, v and φ are aircraft a a mum scattering occurred with the suspected ﬂy ash, since speed and heading in Earth coordinates, and v and φ are w w 1 μm particles are near the peak of mass scattering efﬁciency. the wind speed and direction (White et al., 1976; Ryerson Figure 6 also reveals that near 18:30:30 the 0.1 to 0.2 μm et al., 1998). If we assume winds and concentration are con- accumulation mode caused 80 % of the scattering. This is co- stant with altitude throughout the mixed layer and there is no incident with the SO and OA peaks from the AMS (Fig. 5). vertical ﬂux above the mixed layer, then the discrete version These sizes are very effective as cloud condensation nuclei of Eq. (2) becomes (CCN) in boundary layer (BL) clouds. CNhot, which typi- cally consists mainly of BC and refractory organics (Clarke Q = z (S − S )v v sin(φ − φ )1t. (3) m 0 w a a w et al., 2007) is enhanced during the same period but also shows narrow peaks at other times (e.g., near 18:30). This The ﬂux from a source equals that through the ﬂight path is similar to those seen in the P3B data for stack emissions curtain if there are no signiﬁcant sources or sinks along the with high BC (Fig. 4). Consequently, the size distributions in- way. dicate different sources or processes embedded in the broad The DC-8 loop around the two major upgraders is well oil sands plume contribute differently to both coarse and ﬁne suited to such a calculation. The upwind part of the loop pro- aerosol. vides the background concentrations (blue areas of Fig. 5) Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ UHSAS diameter APS diameter µm µm SO pptv Backscatter S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5079 (A) (B) 58.0 40x10 4km 18:00 1 57.5 18:05 0.1 18:10 z ,1200m 57.0 Flux leg,700 m 18:15 56.5 Ft McMurray 17:50 17:55 18:00 18:05 18:10 18:15 10 July 2008 UTC 112 111ºW Figure 7. (A) Near-infrared (1064 nm) lidar curtain as the DC-8 approached the oil sands area. The plane was at 5.5 km until 18:10, when it started descending. Gray areas are the blanking interval around the plane, below ground level, or obscured by cloud. The plume is obvious starting at about 18:03 with a maximum at 18:06. Note that cloud partly obscures the peak, so the aerosol scattering ratio is an underestimate. Around 18:04 some of the plume is clearly exiting the mixed layer, perhaps due to convection associated with the cloud remnant at 3.2 km. The dashed line is the mixed-layer top (from Fig. 8) and the magenta line is at the altitude of the in situ plume penetration. (B) Location of the curtain with respect to sources and the plume as observed in situ. The track from 17:50 to 18:15 is colored with the lidar curtain data from the in situ altitude, while the low-altitude loop is colored by SO concentration. Magenta arrows show 2 h of wind advection as measured by the DC-8. −1 to subtract from the plume concentrations (brown). The pri- consuming 310 pptv SO h , or about 1 % per hour. If SO 2 2 mary difﬁculty and major source of uncertainty is in estab- from the stacks is the only source of SO , reaction with OH is lishing the mixed-layer height z . The DC-8 did no vertical less than 20 % of the total SO →SO conversion. The rapid m 2 4 proﬁles in the plume, so we have no direct measurements reaction of SO with H O in cloud droplets is not relevant 2 2 2 of mixed-layer depth then and there. Figure 8 shows vertical here, as the few clouds present were well above the mixed proﬁle data from the lidar and in situ instruments. The lidar layer (Fig. 7). Aqueous chemistry can occur immediately as proﬁles are near the in situ leg, while the proﬁles were up- the efﬂuent leaves the stacks (Fig. 2 shows apparent liquid wind but closer in time to the plume penetration. The lidar water droplets), but we were not equipped to determine how shows a fairly uniform mixed layer topping out at 1000 to much H O could diffuse in before the water vapor evapo- 2 2 1200 m. In contrast, the potential temperature suggests a very rates. stable boundary layer with very little vertical mixing. Water Since the SP2 was not functioning on the DC-8, we used vapor and CN concentrations are not uniform, but have min- absorption data from the sub-μm PSAP. Particulate mass ab- ima near 1200 m. That, together with the lidar, makes 1200 m sorption efﬁciency (MAE) is usually reported as between 5 2 −1 a reasonable choice for mixed-layer depth. Since the ground and 20 m g BC, depending on the geometry of the soot level is about 300 m, we use a mixed-layer depth of 900 m, and its coatings (Fuller et al., 1999). The P-3B had func- but it could be as small as 700 or as large as 1500 m (the P- tioning PSAPs and SP2 that day, so MAE immediately ad- 3B proﬁle about 100 min later showed CO capped at about jacent to the upgraders could be determined, as shown in 1800 m). gray in Figure 9. Water vapor ﬂuctuations in the biggest Table 1 shows the results of the ﬂux calculations for plumes caused large positive and negative artifacts in the both mixed-layer depths. We note that the AMS SO ﬂux PSAP signals that had to be edited out, but BC was pri- is roughly 5 % of the SO ﬂux roughly an hour downwind marily found outside those plumes, so the average should of the upgraders, consistent with Cheng et al. (1987), who be valid. The data are entirely consistent with the MAE 2 −1 found that SO from the upgraders reacted at about 6 % of 7.5± 1.2 m g for uncoated soot found by Bond and per hour. This conversion rate cannot be explained by re- Bergstrom (2006). However, rapid changes in particle com- action with OH alone. The OH concentration measured on position as the plume ages are likely to affect MAE, so the the DC-8 was slightly elevated, averaging about 0.11 pptv P-3B ﬂyby on 29 June is probably more representative of across the plume except at the SO peak, when it dropped the 10 July DC-8 data, as both were about 10 km down- to 0.05 pptv. Background concentrations averaged 0.07 pptv. wind. As seen in black in Fig. 9, the data, while scattered At 0.1 pptv OH, 286 K, and 92 500 Pa the reaction rate coefﬁ- and a bit sparse, indicate MAE is roughly double that of −12 3 −1 −1 cient is 1.06× 10 cm molecule s (Blitz et al., 2003), the 10 July close pass. This ampliﬁcation is as expected for www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014 Syncrude Suncor Altitude ºN 5080 S. G. Howell et al.: Particulate emissions from Athabasca oil sands CNcold and CNhot 1.2 0 1000 2000 3000 4000 5000 2 -1 2.5km 15 m g 1.0 Cloud base 2 -1 0.8 2.0 8 m g 0.6 1.5 0.4 z =1.2 0.2 1.0 0.0 10 km away, 6/29 0 5 10 15 20 0.5 close pass, 7/10 Aerosol backscatter ratio -0.2 Ground 292 294 296 298 0.00 0.04 0.08 0.12 0.16 Potential temperature, K -3 BC Mass, µg m 0 20 40 60 80 100 RH, % Figure 9. Relationship between BC mass and aerosol light ab- Figure 8. Vertical proﬁle data from the DC-8 near the oil sands sorption from the P-3B from directly over the upgrader facilities plume. Aerosol backscatter ratios are from 18:06:20 (near the peak (10 July) and from about 10 km downwind (29 June). One-second of the plume downwind of the sampling loop) and 18:10:20 (near data have been smoothed with a 21-point box ﬁlter, and periods on where the lidar curtain crossed the sampling loop). Other data are 10 July where the PSAP ﬂuctuated wildly due to humidity transi- from the descent into (thin lines) and ascent out of (thick lines) the tions in the plumes have been eliminated. The dotted lines are not sampling leg, which occurred upwind of the emissions sources. The ﬁts to the data; they are shown to illustrate that the data obtained choice of mixed layer depth, z , for the ﬂux calculation was based 2 −1 were similar to literature values for uncoated (8 m g ) and coated on the lidar data. Potential temperature indicates a surprisingly sta- 2 −1 (15 m g ) soot. ble boundary layer. In contrast, water vapor and RH suggest that the mixed layer extends to near cloud base at 2.0 to 2.2 km. particles with soot cores surrounded by nonabsorbing coat- Table 1. Fluxes calculated from the DC-8 loop around the oil sands ings (Bond et al., 2006) and is similar to laboratory studies facilities on 10 July and the P3-B plume penetration on 28 June. of coated soot (Schnaiter et al., 2005). Note that if soot par- Estimated errors are roughly a factor of 2 for the DC-8 ﬂyby (Ta- ticles were initially too small for the SP2 to detect but grew ble 2) and higher for the P3-B. The latter column includes a range of due to coagulation, the opposite pattern would be seen, as the values since the plume width is ambiguous. See text and Fig. 10b. PSAP would detect absorption from BC missed by the SP2, “Reported” values are from Environment Canada (2008a) and the artiﬁcially raising the MAE in the near ﬁeld. Another poten- Greenhouse Gas Inventory (for CO ), reporting year 2008. All sources within the DC-8 loop are included, though the Syncrude tial artifact could be caused by formation of “brown carbon” and Suncor upgraders are by far the largest. (BrC), nonsoot organic material capable of absorbing light (Andreae and Gelencser, 2006; Sun et al., 2007). This possi- −1 Flux, g s bility can be examined by analyzing the wavelength depen- Species DC-8, 10 km P3-B, 180 km Reported dence of absorption: BC absorption is inversely proportional to the wavelength, while BrC has a stronger relationship, ab- NO (NCAR) 440 sorbing little at long wavelengths and much more strongly at NO (as NO ) 910 730 short wavelengths. During the 29 June ﬂyby, the PSAP sam- SO (GA Tech) 4500 pling sub-μm particles showed the pattern expected for BC. AMS SO 280 600–650 SO + SO 4800 3100 Meanwhile, the PSAP sampling all particle sizes did detect 2 4 AMS OA 260 550–750 enhanced absorption at short wavelengths, which is charac- AMS total 600 teristic of dust as well as BrC (Yang et al., 2009). BC (from PSAP) 9 0.5–5 Table 2 enumerates most of the errors. “Instrument gaps” CO 2100 1800–4000 290 refer to how well the instrument can sample plume extremes. CO 1 700 000 640 000 The AMS, for example, is blanking 25 % of the time and so might miss peak concentrations. Convection out of the mixed layer is clearly visible in the lidar curtain (Fig. 7) between 18:04 and 18:06 and by the P-3B when it is over Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ GPS altitude -1 sub-µm B Mm ap, 530 nm S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5081 2.3 km (Fig. 4) but is sporadic and thus hard to quantify. The plume about 3 times more, and SO : CO is higher by a factor maximum error would be if such an updraft were lofting part of 50. However, the average CO enhancement in the oil sands of the plume for the entire hour between emission and the plume averaged 10.3 ppbv and peaked at 98 ppbv, while for- ﬂyby. The clouds were weak and so are likely to have up- est ﬁre peaked an order of magnitude higher, so OA in ﬁre −1 −3 drafts < 0.1 m s (anonymous reviewer #1, personal com- plumes often reached hundreds of μg m (Cubison et al., munication, 2013). Over an hour, that would exhaust 360 m 2011; Hecobian et al., 2011). Similarly, BC in the oil sands of the 900 m mixed layer, for a 40 % maximal loss. Dry de- plume is small compared with even a single ﬁre plume. It position losses are not included because they are relatively could be argued that using CO as the standard here is not −1 small; even a deposition velocity of 1 cm s would remove ideal, since the large discrepancy between measured and re- < 4 % of the column amount in the hour between emission ported CO ﬂuxes reveals that we do not really understand the and the DC-8 sampling. NO and SO also have photochem- relationship between CO and other emissions from oil sands 2 2 ical sinks, but ≤ 6 % per hour for SO (Cheng et al., 1987). processing. Unfortunately, no good alternatives are available: Environment Canada maintains a public website with his- the ideal denominator would be measured on both aircraft, toric emissions information. A search for SO emissions in present in both sources, and stable on the timescales of the 2008 reveals that the Syncrude and Suncor plants at Fort sampling. CO does have the advantage that it is commonly McMurray together released about 1× 10 kg in 2008, or used as a standard in biomass burning plumes. −1 3100 g s . Given our estimated errors and the questionable Comparisons between the oil sands plume and other assumption that SO production on 10 July were representa- aerosol sources suffer from some ambiguity, as results are tive of annual emissions, the combined SO and SO ﬂuxes strongly dependent on the temporal and spatial scales cho- 2 4 −1 of 4800 g s constitute excellent agreement. NO is even sen. As an arbitrary example, Table 4 shows estimated annual closer. The dramatic differences between reported and mea- emissions from forest ﬁres throughout Canada from Amiro sured ﬂuxes for CO are thus something of a surprise. CO et al. (2009) in comparison to the oil sands ﬂuxes from Ta- ﬂuxes were also higher than expected, but by a factor of 2.5. ble 1. On this scale the oil sands are an insigniﬁcant source Even though this was a single snapshot and cannot be of BC, OA, and CO (even though we found much higher CO assumed to be representative, the agreement of measured than expected), but they still dwarf ﬁre production of SO . SO + SO and NO ﬂuxes with the Environment Canada Two caveats should be noted here: BC and OA have no pre- 2 4 x database is probably not fortuitous; the upgraders run essen- cise chemical deﬁnition, so there are likely to be components tially continuously, are by far the dominant reported sources of each that are primarily from the oil sands; and the ﬁre sea- in the area, and are likely to dwarf undocumented emissions son lasts only a couple of months, so aerosol from oil sands (only large stationary sources are in the database). Those processing is certainly a more signiﬁcant fraction of the total conditions may be less true for OM, CO, CO , and BC. burden in other seasons. Fluxes of gases and aerosols have often been measured from aircraft using calculations similar to Eq. (2), (e.g., 3.4 Aerosol evolution White et al., 1976; Ryerson et al., 1998, 2011). Since those ﬂight plans were designed for the purpose, with multiple The P-3B and DC-8 ﬂights near the sources on 10 July were plume penetrations and more extensive vertical structure the primary measurements for the oil sands emissions. How- measurements, they often reported much smaller uncertain- ever, we also encountered the plume downwind at other times ties. Such improvements would be entirely practical in this as shown in Fig. 3. These “encounters of opportunity” pro- region. Perhaps most valuable would be a lidar curtain di- vide data to examine plume development over time and some rectly over the ﬂux-measurement legs, which could address evolution in gas and aerosol properties. three of the four largest uncertainties: mixed-layer depth, Figure 10 shows P-3B data for the 29 June and 28 June mixed-layer uniformity, and losses from the mixed layer. ﬂight segments, about 10 and 180 km downwind of the up- In fact, given some knowledge about aerosol optical prop- graders. Unfortunately, on 29 June, the CO went through a erties and wind speed, a lidar alone can be used to mea- calibration cycle at exactly the peak of the plume. The SO sure aerosol ﬂuxes (or anything a lidar can measure well) peak that day was much smaller than on 10 July, indicating even with poorly mixed plumes (e.g., Porter et al., 2002). At that the plane may have missed the main part of the plume best, ﬂux calculation accuracy would probably be limited by or that SO production was lower that day. As with the DC-8 aerosol or gas measurement errors and knowledge of wind data discussed above and collected at roughly the same dis- velocity proﬁles. tance, the plume in Fig. 10a is complex, indicating a variety Because they are unaffected by dilution, emission ratios of sources. OA appears to have a much higher background, can be determined with better precision than ﬂuxes. Table 3 perhaps of biogenic origin as BC was low. A broad plume shows emission ratios of the aerosol species to CO from this with dust and organic aerosol from 22:38 to 22:40 may cor- work and from fresh forest ﬁre plumes during ARCTAS. At respond to mining operations and the evaporative plume from the 10 km DC-8 distance from the source, emission ratios Simpson et al. (2010). The enhanced SO and very large were similar for OA, but BC is enhanced in the oil sands UCN around 22:42 indicate the upgrader plume. As with the www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014 2 -3 dA/dlogD, µm cm -1 Abs, Mm D , µm aero 5082 S. G. Howell et al.: Particulate emissions from Athabasca oil sands Table 2. Flux error estimates. Total errors are calculated by taking the square root of the sum of the squares of the individual errors. Totals are asymmetrical because losses from the mixed layer are not included in the ﬂux calculation but could have had an effect. Error AMS BC NO SO CO CO 2 2 2 Instrument accuracy 35 % 50 % 10 % 10 % 2 % 0.1 % Clean/plume difference 5 % 5 % 5 % 2 % 20 % 5 % Instrument gaps 5 % 0 % 5 % 0 % 0 % 5 % Mixed-layer height 50 % Mixed-layer uniformity 30 % (from lidar) Wind speed Assume 30 % Loss from mixed layer +40 %,−0 % Total +85 % +90 % +80 % +75 % +80 % +75 % −75 % −85 % −70 % −65 % −70 % −65 % Industrial plume (A) (B) Road (dust,BC) Organic+dust 2 10 10 10 1 1 1 3 40 0.4 0 0 0.0 4 4 OA 2 2 SO 450 nm 550 nm 5 5 700 nm 10 10 UCN 3 3 CNhot 10 10 22:40 22:44 19:35 19:45 28 June 2008 UTC 29 June 2008 UTC Figure 10. P-3B crossings of the oil sands plume at (A) about 10 km (on 29 June) and (B) about 180 km (28 June) from the upgraders. Vertical scales are the same on both plots except for CO, BC, and absorption. Absorption is for 530 nm light. The gap in CO data around 22:42 on 29 June was due to an automatic calibration cycle. Altitude on 28 June was constant at 1200 m, so it is not shown. Low BC and absorption rendered the data sufﬁciently noisy that a 21 s box ﬁlter was applied to make changes visible. The gray region shows the minimal plume period used to calculate ﬂuxes in Table 1, deﬁned by the enhanced SO . The vertical dashed line marks the end of the maximum plume period, determined from BC and the 1 μm mode visible in the APS data. Table 3. Aerosol enhancement ratios relative to CO. Units are ter the plane ﬂew over a road where trucks could be seen −1 mg aerosol (g CO) . Forest ﬁre data are from fresh plumes ob- raising clouds of dust. Dust was responsible for much of the served during ARCTAS (Singh et al., 2010). light scattering seen throughout the plume crossing; the very similar patterns at all wavelengths can only occur with par- Species Oil sands Forest ﬁres ticles large compared to the wavelength. While SO was far smaller than that seen by the DC-8 a similar distance away, OA 120± 60 120± 50 the dust seen by the APS was similar, though the 1 μm mode SO 130± 60 2.4± 1.4 was absent, further evidence that the plane either missed part BC 4.3± 2.3 1.6± 0.9 of the plume or that the industrial facilities were operating differently. In contrast to the detail visible at 10 km, the plume at 10 July data, most of the particles are volatile and thus not 180 km downwind (Fig. 10b) is broad and smooth. Wind −1 detected as CNhot. A review of the forward video reveals speeds were about 10 m s , so the plume was about 5 h that the very narrow peak at 22:43:39 occurred shortly af- old. The dust is no longer obvious, presumably diluted rather Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ CN Scat AMS BC CO GPS alt -3 -1 -3 -3 cm Mm µg m ng m ppbv km 3 -3 dV / d log D, µm cm S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5083 Table 4. Emissions from the oil sands in comparison to estimated 3 10 km 182 km 40x10 forest ﬁre production. BC, NO , and CO are directly from Amiro et al. (2009), while the others use ratios to CO during ﬁre plume 182 km penetrations in ARCTAS (as identiﬁed by acetonitrile > 0.5 ppbv). −1 Units are Gg yr . The SO numbers are a bit misleading, as some of the SO will react to create more sulfate. 10 2 10 km Species Oil sands Forest ﬁres OS:FF 0 0 BC 0.3 34 1 % 2 3 4 5 6 7 2 3 4 5 NO 29 181 16 % y 0.01 0.1 CO 66 6450 1 % Particle mobility diameter, µm SO 9 13 70 % Figure 11. SMPS number and volume size distributions obtained SO 140 13 1000 % 10 and 180 km from the source. OA 8 900 1 % ther acid-catalyzed polymerization or organosulfate produc- than sedimented out, as dry deposition timescales are about tion. The relative isolation of the oil sands plume and its high 12 h for 5 μm particles, which have a deposition velocity of acidity may provide an excellent opportunity to study these −1 roughly 2 cm s in coniferous forests with that wind speed reactions outside of a laboratory environment. (Zhang et al., 2001). However, the roughly 1 μm possible ﬂy The increased small particle scattering in Fig. 10b is due to ash mode is evident. The disparity between the red and blue growth of particles initially too small to affect visible light. channels reveals that small particles now dominate the scat- This is revealed by comparing SMPS distributions obtained tering. While overall particle concentrations are lower (note in each plume. In Fig. 11 we compare number distributions the log scales on the CN data), there are now very few volatile measured near the source (10 km downwind, black) with sev- particles, and higher CNhot suggests that at least some of the eral distributions measured at 172 to 192 km downwind (red). secondary organic material in the aerosol has low volatility. The mean of the latter (heavy dashed red) has a peak near Once again, SO and OA are well correlated. CO, which is 4 0.06 μm compared to 0.015 μm near the source. This indi- stable at these timescales, rises about 6 ppbv. cates that some sizes near the source were too small to be Flux calculations like those in Sect. 3.3 can be be ap- active as CCN in BL clouds but have grown to be effective plied here, though errors are even more poorly quantiﬁed: CCN (i.e., larger than about 0.06 μm) for low clouds. Corre- the mixed-layer top of 2900 m is solely from a descent prior sponding growth in the larger mode near 0.1 μm is also evi- to the plume crossing; the P-3B wind speeds were compro- dent, but, as most in this mode were also larger than 0.06 μm mised by the radiometer on the nose of the aircraft; and the before aging, this growth strongly affects light scattering but end time of the plume is rather ambiguous: if the plume is de- has only a second-order effect on their effectiveness as CCN. ﬁned by elevated SO , the penetration took 5 min, while the As shown in Fig. 3, the DC-8 also intercepted the plume BC suggests that the crossing took 15 min. It is likely that on 29 June. It was inadvertent; the plan was to sample a the SO plume is from the industrial facilities while the BC ﬁre plume seen the previous day. The low-altitude sampling is from a broader range of sources, some of which may not leg was cut short when no smoke was found. However, ele- be associated with the oil sands. Thus, these ﬂuxes are just vated SO and NO suggest the oil sands plume was present 2 y rough estimates. Table 1 shows the range of ﬂuxes calculated (Fig. 12). A variety of hydrocarbons were also detected, con- for reasonable deﬁnitions of the end of the plume. Values for sistent with those identiﬁed in the oil sands plume by Simp- SO and OA do not change much in this calculation, while son et al. (2010). This was not a forest ﬁre plume – CO was BC and CO are sensitive to the plume duration and are par- only slightly enriched and other biomass indicators like HCN ticularly uncertain. were not elevated at all. The gas ratios did not resemble those While very approximate, the ﬂuxes calculated in this ﬂyby identiﬁed in the Simpson et al. (2011) analysis of boreal for- do conﬁrm the high CO emissions from the DC-8 loop. The est ﬁre plumes. increased SO : CO is as expected, since sulfate production 4 Low-altitude winds and back trajectories are consistent from SO continues while CO is inert at these timescales. If 2 with transport from the Ft. McMurray area (Fig. 13). Tra- SO emissions were the same those two days, then the addi- 2 jectories from the southern end of the leg did pass directly −1 tional 320 to 370 g s represents about 8 % of the SO in the 2 over the upgraders about 5 h earlier, but at an altitude 1400 m roughly 5 h since emission, still more than can be attributed above the facilities. Low-altitude winds evidently carried the to reaction with 0.1 pptv OH. plume north for a few hours before lofting it to 900 m, where The correlation between OA and SO at 10 and 180 km 4 the back trajectories led to the northern end of the ﬂight leg. and during the P3-B downwind leg reinforce the idea that SO production governs secondary OA formation through ei- www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014 -3 dN / d log D, cm Pressure altitude, km 5084 S. G. Howell et al.: Particulate emissions from Athabasca oil sands 2.00 Caltech NCAR 1.95 UCN CNcold CNhot 1.5 2.0 Traj. alt, km 112 110 108ºW 1.90 Trajectories (1 mark/hr) DC-8 (width by NO ) Wind vanes (~300m ASL) Surface wind (1 hr travel) Org SO Upgrader locations Figure 13. Low-altitude winds and back trajectories to the DC-8 ﬂight track on 29 June. Kinematic back trajectories are from the 1.85 Florida State University group and are driven by hourly FSU/WRF 15:36 15:40 15:44 winds on a 45 km grid initialized from GFS. Surface winds are from 29 June 2008 UTC ground stations at about 5 h before the DC-8 ﬂight. Figure 12. DC-8 plume interception on 29 June. NO , NO , and 2 y SO are consistent with the oil sands plume; low HCN and mini- mally enriched CO indicate that this is not a ﬁre plume. SO and NO emission inventories lie within error esti- 2 2 mates of our calculated ﬂux, though our measurements of CO ﬂuxes substantially exceeded reported emissions during two However, the aerosol signature usually present in the oil plume penetrations. CO emissions were also above reported sands plume was weak. CN and scattering rose only slightly. emissions, but not by a large factor. Black carbon ﬂuxes are −1 More surprisingly, while AMS OA correlated well with SO , highly uncertain, but appear to be roughly 10 g s . NO , and organic vapors, the OA : SO ratio was 8 : 1 rather y 4 While neither the mining operations nor the industrial fa- than 1 : 1. This suggests some removal mechanism, rather cilities produce much particulate organic matter directly, or- than simply dilution, was active. Rain was reported at Ft. Mc- ganic aerosol appears rapidly within the industrial plume in a Murray at 2 and 3 a.m local time (09:00 and 10:00 UTC) on short time, creating particulate mass approximately equal to 29 June (Environment Canada, 2008b), about 7 h before the the sulfate. Some of this may be primary OA, due to conden- DC-8 ﬂight. One likely explanation is that much of the SO sation of vapors as the plume cools, but reactions between reacted in cloud to form SO , which was effectively removed H SO and either biogenic or plume-derived organic vapors 2 4 by rain, while leaving insoluble organic vapors unaffected. may be important. At least over short ranges, we conﬁrm Photochemical reactions after sunrise could then produce the the SO photochemical loss rates from Cheng et al. (1987), organic aerosol seen in Fig. 12. though reaction with OH appears insufﬁcient to account for the conversion rate to SO . The industrial plume has tremendously high aerosol num- 4 Conclusions ber concentrations, saturating our counting instruments, but those particles are too small to scatter light or serve as effec- While 2 ﬂybys and 3 incidental plume penetrations can only tive CCN. They do coagulate and grow, so by the time the provide short-term estimates of the aerosol output of the Al- plume is a few hours old CCN concentrations and scattering berta oil sands mining and upgrading operations, these ﬂights are affected. Major light scattering components of the aerosol have established that emissions ﬂuxes can be measured, at are the accumulation mode in the industrial plume, dust, and least under favorable weather conditions, but with uncertain- what appears to be ﬂy ash. ties of about a factor of 2. The primary sources of error are Compared to estimates of annual forest ﬁre emissions in mixed-layer depth and homogeneity and exchange with over- Canada, the oil sands facilities are a minor source of aerosol lying air. Careful ﬂight planning, extended sampling includ- number, aerosol mass, particulate organic matter, and black ing multiple altitudes, and lidar data along the ﬂux cross sec- carbon. They produce roughly comparable sulfate aerosol tion could greatly enhance the accuracy of the ﬂux calcula- and far more sulfur dioxide. tions, probably to within 50 %. Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ HCN AMS Scat. CN CO NO SO y 2 -3 -1 -3 µg m Mm cm pptv ppbv 5 x NO pptv pptv 57 58ºN S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5085 Acknowledgements. We thank H. Maring (NASA) for his support J. Atmos. Ocean. Tech., 13, 967–986, doi:10.1175/1520- of this objective and some of the analysis undertaken here as 0426(1996)013<0967:PCOAHS>2.0.CO;2, 1996. part of our NASA grant NNX08AD39G. M. J. Cubison and J.- Andreae, M. O. and Gelencsér, A.: Black carbon or brown car- L. Jimenez were supported by NASA grants NNX08AD39G and bon? The nature of light-absorbing carbonaceous aerosols, At- NNX12AC03G. We also appreciate the data collected and assem- mos. Chem. Phys., 6, 3131–3148, doi:10.5194/acp-6-3131-2006, bled by other P3B and DC-8 researchers during ARCTAS that ap- 2006. pear or underlie some of the data/information referenced here. The Bahreini, R., Dunlea, E. J., Matthew, B. M., Simons, C., Docherty, authors also thank the NASA P3B and DC-8 support staff and their K. S., DeCarlo, P. F., Jimenez, J. L., Brock, C. A., and Middle- assistance in collecting this data. We particularly appreciate the in- brook, A. M.: Design and Operation of a Pressure Controlled terest and support of the Clean Air Task Force and Ellen Baum in Inlet for Airborne Sampling with an Aerodynamic Aerosol Lens, helping purchase our SP2 used in ARCTAS and for the partial sup- Aerosol Sci. Tech., 42, 465–471, 2008. port provided for preparation of this report. Baron, P. A. and Willeke, K.: Aerosol measurement: principles, This is SOEST contribution 8982. techniques, and applications, Wiley New York, 2001. Barrie, L. A.: The Fate Of Particulate Emissions From An Isolated Edited by: J. Liggio Power Plant In The Oil Sands Area Of Western Canada, Ann. NY Acad. Sci., 338, 434–452, 1980. Blitz, M. A., Hughes, K. J., and Pilling, M. J.: Determination of References the High-Pressure Limiting Rate Coefﬁcient and the Enthalpy of Reaction for OH + SO , J. Phys. Chem. A, 107, 1971–1978, Aiken, A. C., DeCarlo, P. F., Kroll, J. H., Worsnop, D. R., Huff- doi:10.1021/jp026524y, 2003. man, J. A., Docherty, K. S., Ulbrich, I. M., Mohr, C., Kimmel, Bond, T. C. and Bergstrom, R. W.: Light Absorption by Carbona- J. R., Sueper, D., Sun, Y., Zhang, Q., Trimborn, A., Northway, ceous Particles: An Investigative Review, Aerosol Sci. Tech., 40, M., Ziemann, P. J., Canagaratna, M. R., Onasch, T. B., Alfarra, 27–67, doi:10.1080/02786820500421521, 2006. M. R., Prevot, A. S. H., Dommen, J., Duplissy, J., Metzger, A., Bond, T. C., Habib, G., and Bergstrom, R., W.: Limitations in the Baltensperger, U., and Jimenez, J. L.: O/C and OM/OC ratios enhancement of visible light absorption due to mixing state, J. of primary, secondary, and ambient organic aerosols with high- Geophys. Res., 111, D20211, doi:10.1029/2006JD007315, 2006. resolution time-of-ﬂight aerosol mass spectrometry, Environ. Browell, E. V., Ismail, S., and Grant, W. B.: Differential absorption Sci. Technol., 42, 4478–4485, doi:10.1021/es703009q, 2008. lidar (DIAL) measurements from air and space, Appl. Phys. B- Alboudwarej, H., Felix, J., Taylor, S., Badry, R., Bremner, C., Lasers O., 67, 399–410, doi:10.1007/s003400050523, 1998. Brough, B., Skeates, C., Baker, A., Palmer, D., Anchorage, K. P., Cheng, L., Peake, E., and Davis, A.: The Rate of SO to Sul- Beshry, M., Krawchuk, P., Brown, G., Calvo, R., Triana, J. A. C., fate Particle Formation in an Air Parcel from an Oil Sands Ex- Hathcock, R., Koerner, K., Hughes, T., Kundu, D., de Cár- traction Plant Plume, JAPCA J. Air Waste Ma., 37, 163–167, denas, J. L., and West, C.: Highlighting Heavy Oil, Schlum- doi:10.1080/08940630.1987.10466211, 1987. berger Oilﬁeld Review, 18, 34–53, http://www.slb.com/~/media/ Clarke, A., McNaughton, C., Kapustin, V., Shinozuka, Y., Howell, Files/resources/oilﬁeld_review/ors06/sum06/heavy_oil.pdf (last S., Dibb, J., Zhou, J., Anderson, B., Brekhovskikh, V., Turner, H., access: 9 August 2013), 2006. and Pinkerton, M.: Biomass burning and pollution aerosol over Allan, J. D., Jimenez, J. L., Williams, P. I., Alfarra, M. R., Bower, North America: Organic components and their inﬂuence on spec- K. N., Jayne, J. T., Coe, H., and Worsnop, D. R.: Quantitative tral optical properties and humidiﬁcation response, J. Geophys. sampling using an Aerodyne aerosol mass spectrometer: 1. Tech- Res., 112, D12S18, doi:10.1029/2006JD007777, 2007. niques of data interpretation and error analysis, J. Geophys. Res, Clarke, A. D.: A Thermo-optic Technique for in situ Analysis of 108, 4090, doi:10.1029/2002JD002358, 2003. Size-resolved Aerosol Physicochemistry, Atmos. Environ., 25A, Allan, J. D., Delia, A. E., Coe, H., Bower, K. N., Alfarra, M., 635–644, 1991. Jimenez, J. L., Middlebrook, A. M., Drewnick, F., Onasch, Clarke, A. D., Varner, J. L., Eisele, F., Mauldin, R. L., Tanner, D., T. B., Canagaratna, M. R., Jayne, J. T., and Worsnop, D. R.: and Litchy, M.: Particle production in the remote marine atmo- A generalised method for the extraction of chemically resolved sphere: Cloud outﬂow and subsidence during ACE 1, J. Geophys. mass spectra from Aerodyne aerosol mass spectrometer data, J. Res., 103, 16397–16409, doi:10.1029/97JD02987, 1998. Aerosol. Sci., 35, 909–922, doi:10.1016/j.jaerosci.2004.02.007, Cubison, M. J., Ortega, A. M., Hayes, P. L., Farmer, D. K., Day, D., Lechner, M. J., Brune, W. H., Apel, E., Diskin, G. S., Fisher, Amiro, B. D., Cantin, A., Flannigan, M. D., and de Groot, W. J.: J. A., Fuelberg, H. E., Hecobian, A., Knapp, D. J., Mikoviny, Future emissions from Canadian boreal forest ﬁres, Can. J. Forest T., Riemer, D., Sachse, G. W., Sessions, W., Weber, R. J., Wein- Res., 39, 383–395, doi:10.1139/X08-154, 2009. heimer, A. J., Wisthaler, A., and Jimenez, J. L.: Effects of aging Anderson, T. L. and Ogren, J. A.: Determining Aerosol on organic aerosol from open biomass burning smoke in aircraft Radiative Properties Using the TSI 3563 Integrat- and laboratory studies, Atmos. Chem. Phys., 11, 12049–12064, ing Nephelometer, Aerosol Sci. Tech., 29, 57–69, doi:10.5194/acp-11-12049-2011, 2011. doi:10.1080/02786829808965551, 1998. Davies, M.: Air Quality Modeling in the Athabasca Oil Sands Re- Anderson, T. L., Covert, D. S., Marshall, S. F., Laucks, M. L., Charl- gion, in: Alberta Oil Sands, edited by: Percy, K. E., vol. 11 of son, R. J., Waggoner, A. P., Ogren, J. A., Caldow, R., Holm, R. L., Developments in Environmental Science, chap. 12, 267–309, El- Quant, F. R., Sem, G. J., Wiedensohler, A., Ahlquist, N. A., and sevier, doi:10.1016/B978-0-08-097760-7.00012-3, 2012. Bates, T. S.: Performance Characteristics of a High-Sensitivity, Three-Wavelength, Total Scatter/Backscatter Nephelometer, www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014 5086 S. G. Howell et al.: Particulate emissions from Athabasca oil sands DeCarlo, P. F., Kimmel, J. R., Trimborn, A., Northway, M. J., Jayne, D. R., and Deshler, T.: Transmission Efﬁciency of an Aero- J. T., Aiken, A. C., Gonin, M., Fuhrer, K., Horvath, T., Docherty, dynamic Focusing Lens System: Comparison of Model Cal- K. S., Worsnop, D. R., and Jimenez, J. L.: Field-deployable, culations and Laboratory Measurements for the Aerodyne high-resolution, time-of-ﬂight aerosol mass spectrometer, Anal. Aerosol Mass Spectrometer, Aerosol Sci. Tech., 41, 721–733, Chem, 78, 8281–8289, doi:10.1021/ac061249n, 2006. doi:10.1080/02786820701422278, 2007. Dupont, R., Pierce, B., Worden, J., Hair, J., Fenn, M., Hamer, Mahowald, N., Albani, S., Kok, J. F., Engelstaeder, S., Scanza, R., P., Natarajan, M., Schaack, T., Lenzen, A., Apel, E., Dibb, J., Ward, D. S., and Flanner, M. G.: The size distribution of desert Diskin, G., Huey, G., Weinheimer, A., Kondo, Y., and Knapp, dust aerosols and its impact on the Earth system, Aeolian Re- D.: Attribution and evolution of ozone from Asian wild ﬁres us- search, in press, doi:10.1016/j.aeolia.2013.09.002, 2014. ing satellite and aircraft measurements during the ARCTAS cam- Maring, H., Savoie, D. L., Izaguirre, M. A., Custals, L., and paign, Atmos. Chem. Phys., 12, 169–188, doi:10.5194/acp-12- Reid, J. S.: Mineral dust aerosol size distribution change 169-2012, 2012. during atmospheric transport, J. Geophys. Res., 108, 8592, Environment Canada: National Pollutant Release Inventory Facility doi:10.1029/2002JD002536, 2003. Data Search-Alberta Oilsands and Heavy Oil, http://www.ec.gc. McLinden, C. A., Fioletov, V., Boersma, K. F., Krotkov, N., Sioris, ca/inrp-npri/donnees-data/index.cfm?lang=En (last access: May C. E., Veefkind, J. P., and Yang, K.: Air quality over the Cana- 2014), 2008a. dian oil sands: A ﬁrst assessment using satellite observations, Environment Canada: National Climate Data and Informa- Geophys. Res. Lett., 39, L04804, doi:10.1029/2011GL050273, tion Archive, http://www.climate.weather.gc.ca/climateData/ 2012. hourlydata_e.html?timeframe=1&Prov=ALTA&StationID= McNaughton, C. S., Clarke, A. D., Howell, S. G., Pinkerton, M., 2519&hlyRange=1953-01-01|2008-09-25&Year= Anderson, B., Thornhill, L., Winstead, E., Hudgins, C., Dibb, 2008&Month=6&Day=29 (last access: May 2014), 2008b. J. E., Scheuer, E., and Maring, H.: Results from the DC-8 in- Farmer, D. K., Matsunaga, A., Docherty, K. S., Surratt, J. D., Se- let characterization experiment (DICE): Airborne versus surface infeld, J. H., Ziemann, P. J., and Jimenez, J. L.: Response of an sampling of mineral dust and sea salt aerosols, Aerosol Sci. aerosol mass spectrometer to organonitrates and organosulfates Tech., 41, 136–159, doi:10.1080/02786820601118406, 2007. and implications for atmospheric chemistry, P. Natl. Acad. Sci. Middlebrook, A. M., Bahreini, R., Jimenez, J. L., and Cana- USA, 107, 6670–6675, doi:10.1073/pnas.0912340107, 2010. garatna, M. R.: Evaluation of Composition-Dependent Col- Fuller, K. A., Malm, W. C., and Kreidenweis, S. M.: Effects of mix- lection Efﬁciencies for the Aerodyne Aerosol Mass Spec- ing on extinction by carbonaceous particles, J. Geophys. Res., trometer using Field Data, Aerosol Sci. Tech., 46, 258–271, 104, 15941–15954, doi:10.1029/1998JD100069, 1999. doi:10.1080/02786826.2011.620041, 2012. Hecobian, A., Liu, Z., Hennigan, C. J., Huey, L. G., Jimenez, Ng, N. L., Canagaratna, M. R., Jimenez, J. L., Zhang, Q., Ul- J. L., Cubison, M. J., Vay, S., Diskin, G. S., Sachse, G. W., brich, I. M., and Worsnop, D. R.: Real-Time Methods for Esti- Wisthaler, A., Mikoviny, T., Weinheimer, A. J., Liao, J., Knapp, mating Organic Component Mass Concentrations from Aerosol D. J., Wennberg, P. O., Kürten, A., Crounse, J. D., Clair, J. St., Mass Spectrometer Data, Environ. Sci. Technol., 45, 910–916, Wang, Y., and Weber, R. J.: Comparison of chemical character- doi:10.1021/es102951k, 2011. istics of 495 biomass burning plumes intercepted by the NASA Penkett, S. A., Jones, B. M. R., Brich, K. A., and Eggleton, A. E. J.: DC-8 aircraft during the ARCTAS/CARB-2008 ﬁeld campaign, The importance of atmospheric ozone and hydrogen peroxide in Atmos. Chem. Phys., 11, 13325–13337, doi:10.5194/acp-11- oxidising sulphur dioxide in cloud and rainwater, Atmos. Envi- 13325-2011, 2011. ron., 13, 123–137, doi:10.1016/0004-6981(79)90251-8, 1979. Jacob, D. J., Crawford, J. H., Maring, H., Clarke, A. D., Dibb, J. E., Porter, J. N., Horton, K. A., Mouginis-Mark, P. J., Lienert, B., Emmons, L. K., Ferrare, R. A., Hostetler, C. A., Russell, P. B., Sharma, S. K., Lau, E., Sutton, A. J., Elias, T., and Oppen- Singh, H. B., Thompson, A. M., Shaw, G. E., McCauley, E., Ped- heimer, C.: Sun photometer and lidar measurements of the erson, J. R., and Fisher, J. A.: The Arctic Research of the Compo- plume from the Hawaii Kilauea Volcano Pu‘u O‘o vent: Aerosol sition of the Troposphere from Aircraft and Satellites (ARCTAS) ﬂux and SO lifetime, Geophys. Res. Lett., 29, 30-1–30-4, mission: design, execution, and ﬁrst results, Atmos. Chem. Phys., doi:10.1029/2002GL014744, 2002. 10, 5191–5212, doi:10.5194/acp-10-5191-2010, 2010. Proemse, B. C., Mayer, B., Chow, J. C., and Watson, J. G.: Jang, M., Czoschke, N. M., Lee, S., and Kamens, R. M.: Isotopic characterization of nitrate, ammonium and sulfate Heterogeneous Atmospheric Aerosol Production by Acid- in stack PM2.5 emissions in the Athabasca Oil Sands Catalyzed Particle-Phase Reactions, Science, 298, 814–817, Region, Alberta, Canada, Atmos. Environ., 60, 555–563, doi:10.1126/science.1075798, 2002. doi:10.1016/j.atmosenv.2012.06.046, 2012a. Kimmel, J. R., Farmer, D. K., Cubison, M. J., Sueper, D., Tanner, C., Proemse, B. C., Mayer, B., and Fenn, M. E.: Tracing industrial Nemitz, E., Worsnop, D. R., Gonin, M., and Jimenez, J. L.: Real- sulfur contributions to atmospheric sulfate deposition in the time aerosol mass spectrometry with millisecond resolution, Int. Athabasca Oil Sands Region, Alberta, Canada, Appl. Geochem., J. Mass. Spectrom., 303, 15–26, doi:10.1016/j.ijms.2010.12.004, 27, 2425–2434, doi:10.1016/j.apgeochem.2012.08.006, 2012b. 2011. Ryerson, T. B., Buhr, M. P., Frost, G. J., Goldan, P. D., Hol- Littlejohn, D., Wang, Y., and Chang, S. G.: Oxidation of aqueous loway, J. S., Hübler, G., Jobson, B. T., Kuster, W. C., McK- sulﬁte ion by nitrogen dioxide, Environ. Sci. Technol., 27, 2162– een, S. A., Parrish, D. D., Roberts, J. M., Sueper, D. T., Trainer, 2167, doi:10.1021/es00047a024, 1993. M., Williams, J., and Fehsenfeld, F. C.: Emissions lifetimes and Liu, P. S. K., Deng, R., Smith, K. A., Williams, L. R., Jayne, ozone formation in power plant plumes, J. Geophys. Res., 103, J. T., Canagaratna, M. R., Moore, K., Onasch, T. B., Worsnop, 22569–22583, doi:10.1029/98JD01620, 1998. Atmos. Chem. Phys., 14, 5073–5087, 2014 www.atmos-chem-phys.net/14/5073/2014/ S. G. Howell et al.: Particulate emissions from Athabasca oil sands 5087 Ryerson, T. B., Aikin, K. C., Angevine, W. M., Atlas, E. L., Blake, ern latitudes: Boreal and California forest ﬁre emissions, Atmos. D. R., Brock, C. A., Fehsenfeld, F. C., Gao, R.-S., de Gouw, Environ., 44, 4553–4564, doi:10.1016/j.atmosenv.2010.08.026, J. A., Fahey, D. W., Holloway, J. S., Lack, D. A., Lueb, R. A., 2010. Meinardi, S., Middlebrook, A. M., Murphy, D. M., Neuman, Singh, M., Jaques, P. A., and Sioutas, C.: Size distribution and diur- J. A., Nowak, J. B., Parrish, D. D., Peischl, J., Perring, A. E., Pol- nal characteristics of particle-bound metals in source and recep- lack, I. B., Ravishankara, A. R., Roberts, J. M., Schwarz, J. P., tor sites of the Los Angeles Basin, Atmos. Environ., 36, 1675– Spackman, J. R., Stark, H., Warneke, C., and Watts, L. A.: At- 1689, doi:10.1016/S1352-2310(02)00166-8, 2002. mospheric emissions from the Deepwater Horizon spill constrain Stephens, M., Turner, N., and Sandberg, J.: Particle Identiﬁcation air-water partitioning, hydrocarbon fate, and leak rate, Geophys. by Laser-Induced Incandescence in a Solid-State Laser Cavity, Res. Lett., 38, L07803, doi:10.1029/2011GL046726, 2011. Appl. Optics, 42, 3726–3736, doi:10.1364/AO.42.003726, 2003. Sarwar, G., Fahey, K., Kwok, R., Gilliam, R. C., Roselle, S. J., Strausz, O., Jha, K. N., and Montgomery, D. S.: Chemical composi- Mathur, R., Xue, J., Yu, J., and Carter, W. P.: Potential impacts of tion of gases in Athabasca bitumen and in low-temperature ther- two SO oxidation pathways on regional sulfate concentrations: molysis of oil sand, asphaltene and maltene, Fuel, 56, 114–120, Aqueous-phase oxidation by NO and gas-phase oxidation by doi:10.1016/0016-2361(77)90128-4, 1977. Stabilized Criegee Intermediates, Atmos. Environ., 68, 186–197, Sun, H., Biedermann, L., and Bond, T. C.: Color of brown doi:10.1016/j.atmosenv.2012.11.036, 2013. carbon: A model for ultraviolet and visible light absorption Schnaiter, M., Linke, C., Möhler, O., Naumann, K. H., Saathoff, by organic carbon aerosol, Geophys. Res. Lett., 34, L17813, H., Wagner, R., Schurath, U., and Wehner, B.: Absorp- doi:10.1029/2007GL029797, 2007. tion ampliﬁcation of black carbon internally mixed with Surratt, J. D., Lewandowski, M., Offenberg, J. H., Jaoui, M., Klein- secondary organic aerosol, J. Geophys. Res., 110, D19204, dienst, T. E., Edney, E. O., and Seinfeld, J. H.: Effect of Acid- doi:10.1029/2005JD006046, 2005. ity on Secondary Organic Aerosol Formation from Isoprene, Schwarz, J. P., Gao, R. S., Fahey, D. W., Thomson, D. S., Watts, Environ. Sci. Technol., 41, 5363–5369, doi:10.1021/es0704176, L. A., Wilson, J. C., Reeves, J. M., Darbeheshti, M., Baum- 2007. gardner, D. G., Kok, G. L., Chung, S. H., Schulz, M., Hen- Virkkula, A.: Correction of the Calibration of the 3-wavelength Par- dricks, J., Lauer, A., Kärcher, B., Slowik, J. G., Rosenlof, ticle Soot Absorption Photometer (3λ PSAP), Aerosol Sci. Tech., K. H., Thompson, T. L., Langford, A. O., Loewenstein, M., 44, 706–712, doi:10.1080/02786826.2010.482110, 2010. and Aikin, K. C.: Single-particle measurements of midlatitude Virkkula, A., Ahlquist, N. C., Covert, D. S., Arnott, W. P., Sheridan, black carbon and light-scattering aerosols from the boundary P. J., Quinn, P. K., and Coffman, D. J.: Modiﬁcation, calibration layer to the lower stratosphere, J. Geophys. Res., 111, D16207, and a ﬁeld test of an instrument for measuring light absorption doi:10.1029/2006JD007076, 2006. by particles, Aerosol Sci. Tech., 39, 68–83, 2005. Shinozuka, Y., Redemann, J., Livingston, J. M., Russell, P. B., White, W. H., Anderson, J. A., Blumenthal, D. L., Husar, R. B., Clarke, A. D., Howell, S. G., Freitag, S., O’Neill, N. T., Reid, Gillani, N. V., Husar, J. D., and Wilson, W. E., J.: Formation and E. A., Johnson, R., Ramachandran, S., McNaughton, C. S., Ka- Transport of Secondary Air Pollutants: Ozone and Aerosols in pustin, V. N., Brekhovskikh, V., Holben, B. N., and McArthur, L. the St. Louis Urban Plume, Science, 194, 187–189, 1976. J. B.: Airborne observation of aerosol optical depth during ARC- Yang, M., Howell, S. G., Zhuang, J., and Huebert, B. J.: Attri- TAS: vertical proﬁles, inter-comparison and ﬁne-mode fraction, bution of aerosol light absorption to black carbon, brown car- Atmos. Chem. Phys., 11, 3673–3688, doi:10.5194/acp-11-3673- bon, and dust in China – interpretations of atmospheric measure- 2011, 2011. ments during EAST-AIRE, Atmos. Chem. Phys., 9, 2035–2050, Simpson, I. J., Blake, N. J., Barletta, B., Diskin, G. S., Fuelberg, H. doi:10.5194/acp-9-2035-2009, 2009. E., Gorham, K., Huey, L. G., Meinardi, S., Rowland, F. S., Vay, S. Zhang, H., Worton, D. R., Lewandowski, M., Ortega, J., Ru- A., Weinheimer, A. J., Yang, M., and Blake, D. R.: Characteriza- bitschun, C. L., Park, J.-H., Kristensen, K., Campuzano-Jost, P., tion of trace gases measured over Alberta oil sands mining opera- Day, D. A., Jimenez, J. L., Jaoui, M., Offenberg, J. H., Klein- tions: 76 speciated C –C volatile organic compounds (VOCs), dienst, T. E., Gilman, J., Kuster, W. C., de Gouw, J., Park, C., 2 10 CO , CH , CO, NO, NO , NO , O and SO , Atmos. Chem. Schade, G. W., Frossard, A. A., Russell, L., Kaser, L., Jud, W., 2 4 2 y 3 2 Phys., 10, 11931–11954, doi:10.5194/acp-10-11931-2010, 2010. Hansel, A., Cappellin, L., Karl, T., Glasius, M., Guenther, A., Simpson, I. J., Akagi, S. K., Barletta, B., Blake, N. J., Choi, Y., Goldstein, A. H., Seinfeld, J. H., Gold, A., Kamens, R. M., Diskin, G. S., Fried, A., Fuelberg, H. E., Meinardi, S., Rowland, and Surratt, J. D.: Organosulfates as Tracers for Secondary Or- F. S., Vay, S. A., Weinheimer, A. J., Wennberg, P. O., Wiebring, ganic Aerosol (SOA) Formation from 2-Methyl-3-Buten-2-ol P., Wisthaler, A., Yang, M., Yokelson, R. J., and Blake, D. R.: (MBO) in the Atmosphere, Environ. Sci. Technol., 46, 9437– Boreal forest ﬁre emissions in fresh Canadian smoke plumes: 9446, doi:10.1021/es301648z, 2012. C –C volatile organic compounds (VOCs), CO , CO, NO , Zhang, L., Gong, S., Padro, J., and Barrie, L.: A size-segregated 1 10 2 2 NO, HCN and CH CN, Atmos. Chem. Phys., 11, 6445–6463, particle dry deposition scheme for an atmospheric aerosol doi:10.5194/acp-11-6445-2011, 2011. module, Atmos. Environ., 35, 549–560, doi:10.1016/S1352- Singh, H. B., Anderson, B. E., Brune, W. H., Cai, C., Cohen, R. C., 2310(00)00326-5, 2001. Crawford, J. H., Cubison, M. J., Czech, E. P., Emmons, L., Fuel- berg, H. E., Huey, G., Jacob, D. J., Jimenez, J. L., Kaduwela, A., Kondo, Y., Mao, J., Olson, J. R., Sachse, G. W., Vay, S. A., Wein- heimer, A., Wennberg, P. O., and Wisthaler, A.: Pollution inﬂu- ences on atmospheric composition and chemistry at high north- www.atmos-chem-phys.net/14/5073/2014/ Atmos. Chem. Phys., 14, 5073–5087, 2014

http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

Atmospheric Chemistry and Physics Unpaywall

http://www.deepdyve.com/lp/unpaywall/an-airborne-assessment-of-atmospheric-particulate-emissions-from-the-Yhyr3Yy2Ji

An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands

Loading next page...

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher: Unpaywall
ISSN: 1680-7316
DOI: 10.5194/acp-14-5073-2014
Publisher site: See Article on Publisher Site

Abstract

Journal

Atmospheric Chemistry and Physics – Unpaywall

Published: May 23, 2014

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands

An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands

An airborne assessment of atmospheric particulate emissions from the processing of Athabasca oil sands

References

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies