Long-Range Transported North American Wildfire Aerosols

Environment International 139 (2020) 105680

Contents lists available at ScienceDirect

Environment International

journal homepage: www.elsevier.com/locate/envint

Long-range transported North American wildﬁre aerosols observed in marine boundary layer of eastern North Atlantic T

Guangjie Zhenga,b, Arthur J. Sedlacekb, Allison C. Aikenc, Yan Fengd, Thomas B. Watsonb, ⁎ Shira Raveh-Rubine, Janek Uinb, Ernie R. Lewisb, Jian Wanga,b, a Center for Aerosol Science and Engineering, Department of Energy, Environmental and Chemical Engineering, Washington University in St. Louis, MO, USA b Environmental and Climate Science Department, Brookhaven National Laboratory, Upton, NY, USA c Earth System Observations, Los Alamos National Laboratory, Los Alamos, NM, USA d Environmental Science Division, Argonne National Laboratory, Lemont, IL, USA e Earth and Planetary Sciences, Weizmann Institute of Science, Rehovot, Israel

ARTICLE INFO ABSTRACT

Handling Editor: Xavier Querol Wildfire is a major source of biomass burning aerosols, which greatly impact Earth climate. Tree species in North fi Keywords: America (NA) boreal forests can support high-intensity crown res, resulting in elevated injection height and North American wildfires longer lifetime (on the order of months) of the wildfire aerosols. Given the long lifetime, the properties of aged Dry intrusion NA wildfire aerosols are required to understand and quantify their effects on radiation and climate. Here we Long-range transport present comprehensive characterization of climatically relevant properties, including optical properties and Optical properties cloud condensation nuclei (CCN) activities of aged NA wildfire aerosols, emitted from the record-breaking Cloud condensation nuclei activity Canadian wildfires in August 2017. Despite the extreme injection height of ~12 km, some of the wildfire plumes Marine low clouds descended into the marine boundary layer in the eastern North Atlantic over a period of ~2 weeks, owing to the dry intrusions behind mid-latitude cyclones. The aged wildfire aerosols have high single scattering albedos at

529 nm (ω529; 0.92–0.95) while low absorption Ångström exponents (Åabs) at 464 nm/648 nm (0.7–0.9). In

comparison, Åabs of fresh/slightly aged ones are typically 1.4–3.5. This low Åabs indicates a nearly complete loss of brown carbon, likely due to bleaching and/or evaporation, during the long-range transport. The nearly complete loss suggests that on global average, direct radiative forcing of BrC may be minor. Combining Mie calculations and the measured aerosol hygroscopicity, volatility and size distributions, we show that the high

ω529 and low Åabs values are best explained by an external mixture of non-absorbing organic particles and absorbing particles of large BC cores (> ~110 nm diameter) with thick non-absorbing coatings. The accelerated descent of the wildﬁre plume also led to strong increase of CCN concentration at the supersaturation levels

representative of marine low clouds. The hygroscopicity parameter, κCCN, of the aged wildfire aerosols varies from 0.2 to 0.4, substantially lower than that of background marine boundary layer aerosols. However, the high fraction of particles with large diameter (i.e., within accumulation size ranges, ~100–250 nm) compensates for the low values of κ, and as a result, the aged NA wildfire aerosols contribute more efficiently to CCN population. These results provide direct evidence that the long-range transported NA wildfires can strongly influence CCN concentration in remote marine boundary layer, therefore the radiative properties of marine low clouds. Given the expected increases of NA wildfire intensity and frequency and regular occurrence of dry intrusion following mid-latitude cyclones, the influence of NA wildfire aerosols on CCN and clouds in remote marine environment need to be further examined.

1. Introduction Jacobson, 2004; Reid et al., 2005a; Reid et al., 2005b; Bond et al., 2013) and indirectly by modifying the properties of clouds (Johnson Biomass burning aerosols can greatly impact climate both directly et al., 2004; Sakaeda et al., 2011; Wilcox, 2012; Lu et al., 2018). The by interacting with solar radiation (Crutzen and Andreae, 1990; climatic eﬀects of biomass burning aerosols depend on a number of

⁎ Corresponding author at: Center for Aerosol Science and Engineering, Department of Energy, Environmental and Chemical Engineering, Washington University in St. Louis, Missouri, USA. E-mail address: [email protected] (J. Wang). https://doi.org/10.1016/j.envint.2020.105680 Received 7 August 2019; Received in revised form 18 March 2020; Accepted 20 March 2020 Available online 06 April 2020 0160-4120/ © 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/). G. Zheng, et al. Environment International 139 (2020) 105680 factors, including the injection height (Zhu et al., 2018), dispersion (Ansmann et al., 2018; Khaykin et al., 2018; Peterson et al., 2018; Kloss conditions (Adebiyi and Zuidema, 2016; Lelieveld et al., 2018), optical et al., 2019). Despite the high injection height (up to ~13 km) properties, and Cloud Condensation Nuclei (CCN) and ice nucleation (Ansmann et al., 2018; Kloss et al., 2019), some of the Canadian activity. The optical properties and CCN activities of biomass burning wildfire aerosol aloft travelled across the Atlantic ocean and descended aerosols are strongly influenced by biomass fuel type, fuel load, and into the remote marine boundary layer (MBL) over the eastern North combustion conditions (e.g., flaming or smoldering, etc.) (Andreae and Atlantic (ENA), > 7000 km away from the fires. The well-aged wildfire Merlet, 2001; Petters et al., 2009a; Bond et al., 2013; Saleh et al., 2014; aerosol dominated the properties of MBL aerosols in ENA from Aug. Liu et al., 2014), and these climate-relevant properties will likely 24th to 29th, 2017. Here we present comprehensive characterization of change as the particle size distribution, chemical composition, and climatically relevant properties of the aged wildfire aerosol, including mixing state evolve during the transport in the atmosphere (Jacobson, size distribution, chemical composition, optical properties, hygro- 2001; Janhäll et al., 2010; Levin et al., 2010; Cubison et al., 2011; scopicity, and CCN concentrations. The potential impact of the NA Ortega et al., 2013; Vakkari et al., 2014; Zhao et al., 2015; Sumlin et al., wildfires on the CCN spectrum and therefore cloud properties in remote 2017). Presently, biomass burning aerosols and their evolution remain marine boundary layer is also discussed. not well understood and poorly represented in climate models (Stocker, 2014), and these topics have been the focus of several recent field 2. Method campaigns, for example, the ObseRvations of Aerosols above CLouds and their intEractionS (ORACLES) campaign (Lang, 2015), the Layered This study utilizes comprehensive in-situ measurements conducted Atlantic Smoke Interactions with Clouds (LASIC) campaign (Zuidema at the Atmospheric Radiation Measurement Climate Research Facility et al., 2016; Zuidema et al., 2018), the Green Ocean Amazon Experi- ENA site (https://www.arm.gov/capabilities/observatories/ena), ment (GoAmazon2014/5) (Martin et al., 2017), and the Biomass which is located on Graciosa Island in the Azores, Portugal (39° 5′ 30″ Burning Observation Project (BBOP) (Collier et al., 2016). N, 28° 1′ 32″ W, 30.48 m above mean sea level). The site straddles the Wildfires, as well as controlled agricultural burning and cooking, is boundary between the subtropics and mid-latitudes in the eastern North an important source of biomass burning aerosols (Andreae and Merlet, Atlantic, and receives a diverse range of air masses from North America, 2001; van der Werf et al., 2010; Williams and Abatzoglou, 2016). the Arctic, and northern Europe (Wood et al., 2015; Wang et al., 2016;

Wildfires has increased in frequency, duration, areas burned, and in- Zheng et al., 2018). Measurement of the trace gases (e.g., CO and O3), tensity in the past decades, partially due to the human-induced am- meteorological parameters, and aerosol optical properties are detailed plification (Field et al., 2009). Proposed amplification mechanisms in- in Zheng et al. (2018). Aerosol size distribution with particle diameter clude the agricultural management processes like less forest thinning ranging from 0.07 to 1 μm was measured by an Ultra-High Sensitivity (Pollet and Omi, 2002), drier conditions resulting from anthropogenic- Aerosol Spectrometer (UHSAS, Droplet Measurement Technologies, induced climate warming (Marlon et al., 2008; Kelly et al., 2013; Jolly Boulder, CO, USA). Non-refractory submicron aerosol composition et al., 2015; Williams and Abatzoglou, 2016; Dennison et al., 2014), etc. (organics, sulfate, nitrate, ammonium, and chloride) was characterized In turn, the increasing wildfire activities can greatly impact climate by by an Aerosol Chemical Speciation Monitor (ACSM; Aerodyne Research, modifying landscapes, altering surface albedo, and injecting substantial Inc., Billerica, MA, USA). Aerosol hygroscopic growth factor at 90% RH amounts of greenhouse gases and aerosols into the atmosphere was obtained by a Humidified Tandem Differential Mobility Analyzer (Randerson et al., 2006). (HTDMA, Brechtel Manufacturing Inc., CA, USA) (Uin, 2016). These The amplification of wildfire activities is especially evident in North above measurements are part of the routine measurements at the ENA America (NA) temperate and higher-latitude boreal forests, including site since October 2013 (Mather and Voyles, 2013). During the Aerosol western U.S. and Canadian forests (Westerling et al., 2006; Gavin et al., and Cloud Experiments in Eastern North Atlantic (ACE-ENA) campaign 2007; Marlon et al., 2009; Turco et al., 2014; Abatzoglou and Williams, from June 2017 to August 2018, additional measurements were carried 2016). North America boreal forest wildfires may also influence climate out at the ENA site, as detailed below. Only the data collected during and ecosystems differently than the wildfires of tropical grassland or the 14-month ACE-ENA campaign are used for further analysis. Eurasia boreal forests, as NA tree species can support high-intensity crown fires (Wooster and Zhang, 2004; Randerson et al., 2006; Rogers 2.1. Aerosol size distributions et al., 2013; Rogers et al., 2015). The stronger pyrocumulonimbus activities of NA wildfires result in higher aerosol injection heights (typi- Aerosol size distribution from 10 to 470 nm was measured by a cally 4~7 km) than fires in other regions (mostly below 3 km) (Labonne scanning mobility particle analyzer (SMPS, Model 3938, TSI et al., 2007; Zhu et al., 2018). These higher injection heights lead to Incorporated, Shoreview, MN, USA). Concurrently, total aerosol longer aerosol lifetime of up to over one month (Liang et al., 2009; number concentrations (CN) were measured by a standalone con- Bond et al., 2013; Laing et al., 2016), during which the aerosol prop- densation particle counter (CPC, Model 3772, TSI Incorporated, erties may change substantially due to various aging process such as Shoreview, MN, USA). The standalone CPC and the SMPS were oper- photochemical transformation (Forrister et al., 2015; Lin et al., 2015; ated side-by-side, and the measurements alternated every 4 min be- Liu et al., 2016a; Liu et al., 2015; Schnitzler and Abbatt, 2018; Sumlin tween ambient samples and samples processed by a custom-made et al., 2017), coagulation, cloud processing, and gas-particle parti- thermal denuder operated at 300 °C. Components that survived the tioning (Saleh et al., 2013; Vakkari et al., 2014; Taylor et al., 2014; thermal denuder are referred to as non-volatile components hereinafter. Haviland et al., 2015; Vakkari et al., 2018; Janhäll et al., 2010). Relative humidity of the aerosol sample was reduced to below 25% Therefore, characterization of well-aged (i.e, for more than several using a Nafion dryer before being introduced into the system. The flow days) wildfire aerosol properties is required to quantify the climate rates and classifying voltage of the SMPS were calibrated every effects. 3 months and exhibited negligible (< 1%) variation throughout the Despite the importance, there have been only a few field observa- ACE-ENA campaign. tions (Fiebig et al., 2003; Petzold et al., 2007) focusing on the prop- The measured number size distributions were fitted as a sum of up erties of well-aged NA wildfire aerosols after long-range transport. That to three lognormal modes, which were subsequently classified ac- is partially due to the elevated injection heights and therefore long cording to the number mode geometric mean diameters (Dpn). For the distances from the sources (Ditas et al., 2018), making it difficult to ambient aerosol size distribution, the modes are classified as: nuclea- predict plume locations and plan measurements accordingly. Here we tion mode (Dpn < 30 nm), Aitken mode (30 < Dpn < 110 nm) and report direct measurements of well-aged (> 10 days) aerosol properties accumulation mode (Ac, 110 < Dpn < 300 nm). For the non-volatile from the extreme wildfires in British Columbia, Canada in August 2017 component, the modes are classified into non-volatile Aitken

2 G. Zheng, et al. Environment International 139 (2020) 105680

(14 < Dpn < 100 nm) and non-volatile Accumulation (100 < Dpn < 250 nm). For non-volatile aerosol component, the thresholds of mode diameters are generally smaller than those of ambient aerosols, based on the typical non-volatile volume for each mode at ENA. The non-volatile nucleation mode (Dpn < 14 nm) was rarely observed (< 4% annually) and is not considered here. All modes with

Dpn above 300 nm for ambient aerosols, or above 250 nm for non-volatile aerosols, are considered as sea spray aerosol modes (Quinn et al., 2017; Zheng et al., 2018).

2.2. Aerosol hygroscopicity from size-resolved CCN activation fraction

The CCN activation spectrum (i.e., activation fraction as a function of super-saturation) of size-selected particles was measured using a Differential Mobility Analyzer (DMA, TSI Inc Model 3081) coupled to a CPC (TSI Inc., Model 3010) and a cloud condensation nuclei counter (CCNC, Droplet Measurement Technologies, Boulder, CO) (Frank et al., 2006; Moore et al., 2010; Petters et al., 2007; Mei et al., 2013b). The operation of the same system is detailed in Thalman et al. (2017), and is only briefly described here. During the ACE-ENA campaign, the DMA stepped through 6 aerosol dry diameters of 40, 50, 75, 100, 125, and 150 nm. At each diameter, the super-saturation level inside the CCNC was changed by changing the flow rates (QCCN, ranging from 0.3 to 1 L −1 △ min ), temperature gradient ( T, changing among 4, 6.5, and 10 K), Fig. 1. Aerosol properties and air mass origins at the ENA site from August 24th or a combination of both. At each super-saturation, sampling time was a to 29th, 2017. (a) Air mass origins classified according to 10-days back-tra- minimum of 30 s and until 1500 particles were detected by the CPC, or jectory analysis. The bar graph shows the number of trajectories (out of 12 a maximum time of 120 s. A measurement cycle through the 6 particle during a 6-h period) arriving at the site at the respective time, with their origins diameters took between 1 and 2 h. The super-saturation inside the classified as North America (orange), the Arctic (blue), subtropics (brown) or CCNC was calibrated using ammonium sulfate particles following es- others (grey). See section 2.3 and Fig. S1 for more details on the trajectory tablished procedures (e.g., Mei et al., 2013a; Thalman et al., 2017) for calculation and source definitions. (b) Time series of size distributions for ambient and thermally denuded samples, gas-phase species including CO, O3 each operation condition (QCCN and △T), and ranged from 0.07% to 1.34% at 298 K. Dependence of super-saturation on the temperature at and water vapor, and non-refractory chemical compositions during this episode. the top of CCNC column (instrument temperature T1) (Rose et al., 2008; Thalman et al., 2017) was also derived from the calibrations with different T1 values, and applied to retrieve the CCNC super-saturation origin; (iv) Trajectories not meeting any of the above criteria are clas- levels. For the measurements of size-resolved CCN activation spectrum, sified as “others”. This class corresponds to 13% of all trajectories. influence from multiple charged particles is taken into consideration using the same approach described in Thalman et al. (2017). The par- 3. Strong impact of Canadian wildfire on boundary layer aerosol ticle hygroscopicity parameter under supersaturated conditions, κ properties (Petters and Kreidenweis, 2007), was derived from the activation spectrum at each particle size using approaches detailed in the litera- Abnormally high CO mixing ratio and number concentration of non- ture (Lance et al., 2013; Mei et al., 2013a; Thalman et al., 2017). volatile accumulation mode aerosol, NAc, TD (Fig. 1) were observed at the ENA site from 00 UTC 24th August to 00 UTC 29th August 2017 2.3. Classification of air mass origin (referred to as the wildfire episode hereinafter). CO mixing ratio averaged 144 ppb, which is 1.4 times of the average and presenting To understand the origin of the air masses arriving at the ENA site, 99.9th percentile of the measurements during the one-year ACE-ENA we performed backward trajectory calculations using reanalysis data. campaign. While non-volatile accumulation mode was present nearly Wind data were obtained from the European Centre for Medium-range all the times during this episode, it was observed for only 12% of the

Weather Forecasts (ECMWF) interim reanalysis (ERA-Interim) (Dee observational time during the ACE-ENA campaign. Average NAc, TD − et al., 2011). The 6-hourly ERA-Interim data were interpolated hor- during the wildfire episode (227 cm 3) is 3.3 times higher than the izontally to 1° by 1° grid, at 60 vertical hybrid levels. The interpolated average of those observed outside of the episode. If we consider the data were then used in the Lagrangian analysis tool (LAGRANTO) concentration of NAc, TD as zero in the absence of a non-volatile Ac version 2.0 (Sprenger and Wernli, 2015) to calculate 10-day backward mode, then the average NAc, TD during this episode is 28 times higher air mass trajectories arriving at the ENA site, for the 12 available model than the annual average. Back trajectory analysis (section 2.3) shows levels between the surface and 850-hPa. Calculated every 6 h from 00 that 54% of the air masses arriving at lower troposphere (< 850 hPa) UTC 24th August to 00 UTC 29th August 2017, a total of 252 backward over the ENA site during this episode originated from North America, trajectories were subsequently classified into 4 origin groups (Fig. S1) with the rest from the Arctic (18%), subtropical latitudes (14%) or with the following criteria. (i) North American origin if a trajectory is others (Fig. 1a; Fig. S1). Elevated aerosol number concentration and CO found at least once within 120° − 60° W longitude, while at latitude mixing ratio were observed when air masses mostly originated from lower than 62° N and at times 96–240 h prior to arrival in the ENA; (ii) North America (orange bars in Fig. 1a), while lower values were gen- Arctic origin if a trajectory is found at least once within 60° − 0°W erally associated with the air masses dominated by those with Arctic longitude, while at latitude higher than 62° N and at times 96–240 h origin (blue bars in Fig. 1a). Active Canadian wildfires were observed prior to arrival in the ENA (and was not recognized as having a North (Fig. S1) and reported (https://www.bbc.com/news/world-us-canada- American origin); (iii) Subtropical origin if a trajectory is found at least 41040625) just prior to this episode. The strong impact of the Canadian once at latitude lower than 35° N and at times 6–150 h prior to arrival wildfire on the aerosol in the ENA MBL is further supported by the in the ENA, and is neither recognized as a North American nor Arctic in mixing ratios of trace gases and characteristics of aerosol properties

3 G. Zheng, et al. Environment International 139 (2020) 105680

Fig. 2. Distinct characteristics of trace gases and aerosol properties during the wildﬁre episode. Orange dots indicate data observed during August 24th to 29th, 2017 when air masses were not from the Arctic, and grey dots indicate data observed at the ENA site during the 14-month ACE-ENA campaign but outside of this wildﬁre episode. Variables shown include: (i) volume

mixing ratios of CO and O3 in part per billion (ppb), and the water vapor mixing ratios in g − kg 1; (ii) the volume concentrations of accumulation mode aerosols under ambient condi-

tions (Ambient VAc) and those remaining after

heated to 300 ℃ (VAc, TD); (iii) total number concentrations of ambient aerosols (CN) and

those remaining after heated to 300 ℃ (CNTD); and (iv) mass concentration of non-refractory organics in sub-micron (diameter < 1 μm) aerosols.

(section 3.1), and agreed with the synoptic conditions that facilitated Pacific(Clarke et al., 2013). Although the non-volatile volume fraction the downward transport of the wildfire plumes into the MBL (section is only ~30% (Fig. 2d), non-volatile number fraction (CNTD / CN) 3.2). during non-Arctic periods of this episode is around 90% (not shown here), indicating that nearly all particles had a non-volatile component. 3.1. Trace gas mixing ratios, aerosol number and volume concentrations This is in agreement with results from previous studies of biomass burning aerosols (Clarke et al., 2004; Petzold et al., 2007). The good During this episode, one prominent feature is the elevated CO correlation between VAc and CN, and that between VAc, TD and CNTD mixing ratio, which is especially evident from the correlations with O3 (Fig. 2g, h) suggest relatively uniform spectral shape of size distribu- and water vapor (Fig. 2a,b). Within the ENA MBL, CO usually correlates tions (section 5.1). Both ambient and non-volatile aerosols show higher fi positively with O3 while negatively with water vapor (Zheng et al., volume-to-number ratios during the wild re episode than other periods 2018). During the non-Arctic periods (Fig. 1) of this episode, the CO (Fig. 2g, h), consistent with the larger mode diameter of the accumu- mixing ratio is much higher than the other periods with similar H2Oor lation mode aerosols (see section 5.1). O3 levels. This is attributed to the higher CO mixing ratio in wildfire plumes than in other types of anthropogenic emissions (Clarke and 3.2. Vertical transport and dry intrusions Kapustin, 2010). Aerosol properties during non-Arctic periods (Fig. 1) of this episode On 12 August 2017, an extremely intense wildfire occurred in also exhibit distinct features, including strong linear correlations among western Canadian. The corresponding strong pyrocumulonimbus ac- the volume concentration of accumulation mode aerosol (VAc), volume tivity injected aerosols into the upper troposphere and lower strato- concentration of non-volatile accumulation mode at 300 ℃ (VAc, TD), sphere at ~12 km (Ansmann et al., 2018; Peterson et al., 2018; Khaykin mass concentration of non-refractory organics measured by ACSM, and et al., 2018; Kloss et al., 2019). At these high altitudes, it can take over CO mixing ratios (Fig. 2c-e). Similar linear correlations were also ob- one month for aerosols to descend into the boundary layer (Liang et al., served in other fresh and slightly aged NA wildfire plumes (Clarke et al., 2009; Bond et al., 2013; Laing et al., 2016). Therefore, one would ex- 2007; Kondo et al., 2011). Both ambient CN (Fig. 2c,g) and non-volatile pect little influence in the boundary layer from this event. However, as

CN at 300 ℃ (CNTD)(Fig. 2f) show strong correlations with CO during shown in section 3.1, the MBL aerosols were strongly inﬂuenced by this the episode, similar to those observed in the long-range transported wildﬁre after only ~12 days, suggesting an accelerated descent of the biomass burning plumes in the free troposphere of central equatorial plumes.

4 G. Zheng, et al. Environment International 139 (2020) 105680

Fig. 3. Descent of Canadian wildﬁre air masses by dry intrusions behind a cold front. (a) 10-day backward trajectories of the air masses arriving at the ENA site (red star) at 12:00 UTC, Aug. 27th. The trajectories are colored according to their pressure level (hPa). The blue dots mark the location of the air masses at 00:00 UTC, Aug. 21st, when the front is positioned oﬀ the east coast (blue line). (b) Meteorological conditions at 00:00 UTC, Aug. 21st. The 850-hPa equivalent potential tem-

perature θe (in K, shaded), mean sea-level pressure (in hPa, black contours, contours are dashed and unlabeled for 1016 hPa and above) and potential vorticity on 330-K isentropic surface (2-PVU, red contour) are shown. The curved blue line marks the

location of the front along the strong θe gradients (veriﬁed against NOAA surface analysis, not shown). The schematic orange arrow illustrates the descending airﬂow at this time behind the upper- tropospheric trough and the cold front.

This accelerated descent is attributed to dry intrusions. Dry intru- synoptic-scale pathway that facilitates the relatively fast transport of sion (Browning, 1997; Wernli, 1997), also termed as “dry conveyor NA wildfire aerosols into the MBL. The fast transport contrast with the belt” (Danielsen, 1968), refers to the strong descent of upper-tropo- conventional perception of an extended period of time being required spheric dry air in post-cold-front regions, and is often observed over the for wildfire aerosol aloft to descend into MBL. As dry intrusions fre- ENA region (e.g., Naud et al., 2018). Backward trajectories shows that quently occur behind cold fronts in mid-latitudes, (e.g., Naud et al., the majority of NA air masses arriving at the ENA site indeed descended 2018), we expect substantial impact of NA wildfires on the properties of slantwise from middle-upper troposphere towards the ENA lower tro- MBL aerosol and therefore low clouds over the North Atlantic. posphere within 10 days prior to reaching the site (Fig. S1). One example of dry-intrusion air masses is shown in Fig. 3. Fig. 3a 4. Optical properties of well-aged wildfire aerosol shows the 10-day backward trajectory starting from 12:00 UTC, 27 August. This period was characterized by a high CO mixing ratio The extreme Canadian wildfire and the accelerated descent of the (> 140 ppb), while all MBL air masses arriving at that time originated wildfire plume into the ENA boundary layer provided an excellent from NA (Fig. 1). The backward trajectories suggest that the air masses opportunity to characterize the properties of well-aged wildfire aero- descended from 500 to 700 hPa to the MBL during 3–4 days as they sols. As strong pyrocumulonimbus activities were observed on August travelled from Northern Canada to the eastern coast of North America 12th and 13th (Peterson et al., 2018), we estimate the wildfire aerosols by Aug. 21st (Fig. 3a). At this stage, the air masses were located in the observed from August 24th to 29th had an aging time of around two cyclone cold sector, behind the cold front off the east coast. Following weeks. Sea spray aerosol can contribute substantially to the aerosol this descent, i.e., after Aug. 21st, the air masses were advected at a scattering coefficient in the ENA MBL (Zheng et al., 2018)(Fig. 4). As a rather constant altitude towards the ENA site, while turning anti-cy- result, the aerosol scattering coefficient (Bsca) of the NA wildfire aerosol clonically. Similarly, during Aug. 14th – 29th, a series of five extra- cannot be measured directly. In this study, the size distribution mea- tropical cyclones occurred over eastern North America, each having a sured by UHSAS was fitted as sum of lognormal modes, and Bsca of the cold trailing front with descending air behind it, which later partially NA wildfire aerosol at 529 nm was calculated from the fitted accumu- reached the ENA site (not shown). During the same time period, cy- lation mode using Mie theory (see detailed calculations and corrections clones over the North Atlantic governed the descent of Arctic air as well in SI S1). This approach assumes that the accumulation modes during towards ENA. The latter are similar to the cold sector of the cyclone this episode were dominated by the aged wildfire aerosols, as indicated north of ENA in Fig. 3b. by the much higher VAc, VAc, TD and CO than the background values Raveh-Rubin (2017) suggested a Lagrangian descent criterion of (Fig. 2). Measured aerosol absorption coefficient (Babs) is all attributed 400 hPa descent in 48 h for the identification of dry intrusions. Such a to the aged wildfire aerosols because the contribution from sea spray defining criterion resulted in events predominantly in the extended aerosols is expected to be negligible. winter season, a third of which indeed occur in association with cold trailing fronts (Catto and Raveh-Rubin, 2019). For the summer case here, however, the descent does not qualify for the criterion in Raveh- 4.1. Single scattering albedo Rubin (2017), as air mostly descends < 400 hPa, over a time period fi longer than 48 h. Yet, the dry northwesterly descending flow is situated During this wild re episode, measured Babs is linearly correlated 2 between a cyclone-anticyclone dipole, behind a cold front (Fig. 3)as with calculated accumulation mode Bsca (r = 0.91) (Fig. 5a), in- ω observed in winter dry intrusions (Raveh-Rubin and Catto, 2019), dicating a relatively constant single scattering albedo at 529 nm ( 529). suggesting that the dry intrusions here are qualitatively similar, al- Here the analysis is focused on the periods with CO > 140 ppb (99% though shallower and weaker compared to strong winter dry intrusions. percentile of the 14-month data during the ACE-ENA campaign), when The evidence shown here indicates that dry intrusions are the aerosol properties were dominated by wildfire plumes (Figs. 1, 2). During these periods, Bsca and Babs were also above 99% percentile of

5 G. Zheng, et al. Environment International 139 (2020) 105680

Fig. 4. Influence of sea spray aerosols on optical properties of ENA MBL aerosols during the wildfire episode. (a) Average volume size distribution with fitted sea spray aerosols mode shown for reference, and (b) calculated size-de-

pendent contribution for Bsca at 529 nm. The blue dash line indicates the approximate threshold of sea spray aerosols dominating size

ranges (Dp > 500 nm).

−1 −1 the 14-month measurements, and were > 22 Mm and 1.6 Mm , The low Åabs observed here indicates negligible contribution of respectively (Fig. 5, the range of data used in the following calculations brown carbon (BrC) (Clarke et al., 2007; Shinozuka et al., 2007; indicated by the red dash lines and arrows). Average ω529 of the aged Gyawali et al., 2009; Forrister et al., 2015; Laing et al., 2016; Liu et al., wildfire aerosol is derived as 0.92. This value may represent a lower 2017; Selimovic et al., 2018; Sedlacek III et al., 2018). In comparison, fi limit, as the filter-based PSAP can overestimate Babs by up to 70% at Åabs of fresh wild re aerosols and those aged for less than a few days high organic loadings (Lack et al., 2008; Cappa et al., 2008). With this typically ranges from ~1.4 to ~3.5 (Table 1), suggesting the presence taken into consideration, the upper limit of ω529 is estimated as 0.95. of substantial BrC upon emission. This apparent reduction of the BrC The ω529 value of 0.92 – 0.95 is near the upper end of the range re- contribution to absorption is in agreement with Forrister et al. (2015), ported by previous field studies of NA wildfire aerosols (Table 1). Note which examined the BrC concentration and Åabs of western U.S. forest that in some studies (e.g., Gyawali et al. (2009), Forrister et al. (2015); fire as a function of aging time up to 50 h. Their result shows that most fi Liu et al., (2017); Selimovic et al. (2018)) the ω were measured (~94%) BrC emitted from NA wild res was lost within a day (Åabs at ~400 nm and 870 nm, and exhibit negligible difference at the two reduced from ~3.6 to ~1.4), while the remaining fraction was persistent and not affected by further aging up to 50 h (Forrister et al., 2015). wavelengths. Therefore, we expect ω529 to be at same level. While the results listed in Table 1 (including this study) are from different NA The even lower Åabs value observed here essentially suggested a com- wildfires, they suggest a generally increasing ω with the plume age of plete loss of BrC following the extended aging time. The reduction of the NA wildfire aerosol. Such trend can be explained by the collapse of the BrC contribution to absorption could be due to a combination of chain-aggregates and condensation of secondary species (Reid et al., bleaching (Zhao et al., 2015; Liu et al., 2016a; Sumlin et al., 2017; 2005a,b; Ahern et al., 2019), bleaching and/or evaporation of ab- Schnitzler and Abbatt, 2018) and/or evaporation of volatile BrC sorbing particles, etc (section 4.2). (Forrister et al., 2015). The derived Åabs value (i.e., 0.7–0.9) is also near the lower end of the typical range (0.6–1.3) observed for particles consisting of black Å 4.2. Absorption ngström exponents carbon (BC) without contribution from brown carbon (Bergstrom et al., 2007; Clarke et al., 2007; Gyawali et al., 2009; Bond et al., 2013; Liu For periods when aerosol was dominated by the aged wildfire par- et al., 2018). This low Åabs value is typically associated with large ticles (i.e., when CO > 140 ppb), values of absorption Ångström ex- compact BC particles (diameter > ~ 110 nm), or large BC core with ponents (Åabs) derived from the PSAP measurements at 464 and 648 nm thick non-absorbing coating (Schnaiter et al., 2005; Gyawali et al., are mostly between 0.6 and 0.8 (Fig. 5c). Early studies show that Å abs 2009; Lack and Cappa, 2010; Liu et al., 2018; Bergstrom et al., 2007; Li derived from filter-based PSAP may be underestimated by ~ 14% when et al., 2016), as also reproduced in our calculations (SI S2). The HTDMA compared to that derived from photoacoustic based measurements measurement shows that the vast majority of the particles has a hy- (Chow et al., 2009). A 14% increase of the derived Å leads to a range abs groscopic growth factor > 1.23 at 80% RH, excluding the possibility of of 0.7–0.9.

Fig. 5. Optical properties of long-range transported wildfire aerosols. (a) Dependence of Accumulation mode Bsca on CO, and identification of wildfire-dominated periods (i.e., CO > 140 ppb). (b) Babs and accumulation mode Bsca at 529 nm, and (c) aerosol absorption Ångström exponent between 464/648 nm.

6 .Zeg tal. et Zheng, G.

Table 1 Summary of optical properties from previous ﬁeld observations of NA wildﬁre aerosols.

Source Region Observation Site Plume type Year Campaign Single Scattering Albedo Absorption Ångström Exponent Reference

wave-length valuea wave-length valuea (nm) pairs (nm)

California Forest fire Surface, Reno, NV fresh (several hours) 2008 / 405 0.88–0.93b 405/870 1.42–2.07 Gyawali et al. (2009) 870 0.88–0.96b Southwestern U.S. forest (Las Surface, NM fresh 2012 / 532 0.908 (0.83–0.95) 405/781 2.1 ± 0.5 Liu et al., (2014) Conchas, NM) Northwestern U.S. forest Pacific Northwest, Richland, WA fresh (< 5h) 2013 BBOP 532 ~0.94 / / Sedlacek III et al. (2018) Southwestern U.S. forest Surface, NM slightly aged (~9h) 2012 / 532 0.911 (0.90–0.93) 405/781 3.3 ± 0.4 Liu et al., (2014) (Whitewater Baldy, NM) 7 Northern California Mt. Bachelor Observatory, FT in slightly aged 2016 / / / 470/660 1.8 or 2.4c Laing et al. (2019) central Oregon, U.S. (10 ~ 15 h) western U.S. forest fire western, central, and southeastern fresh to slightly aged 2013 NASA SEAC4RS 401 ~0.90b, d 401/870 ~3.5d Forrister et al. (2015); Liu et al., regions of continental U.S. (< 3 days) 870 ~0.91b, d (2017); Selimovic et al. (2018) Western US wildfires Mt. Bachelor Observatory, FT in fresh to slightly aged 2015 / 528 0.978 ± 0.007 467/660 3.45 ± 1.04 Laing et al. (2016) central Oregon, U.S. (3–35 h) Canadian Forest fire FT of NA east coast aged (3–5 days)e 2004 INTEX / 550 0.96 (0.95–0.97) 470/660 2.1 (1–3) Clarke et al., (2007) ICARTT Canadian Forest fire ENA well-aged (> 10 days) 2017 ACE-ENA 529 0.93–0.96 464/648 ~0.8 This study

a Read a ± b as mean ± 1 σ, and read a (c-d) as mean (min - max). b Corresponding single scattering albedo at 529 nm are expected to be in-between that measured at 405 and 870 nm c Depending on the correction method applied. d Estimated through combining the lab studies and field observations of the Rim fires; see Selimovic et al. (2018) for more detail. e Estimated from typical air mass transport time from western to eastern U.S. (Colarco et al., 2004). Environment International139(2020)105680 G. Zheng, et al. Environment International 139 (2020) 105680 a substantial contribution from compact BC particles, which is hydro- average. phobic. Large BC core sizes are consistent with previous observations of The optical properties also influence the lifetime and geographic NA wildfire plumes, which generally show the geometric mean dia- spread of aged wildfire aerosols, which further affect their direct ra- meter or BC cores of 120 to 150 nm (Kondo et al., 2011; Taylor et al., diative effects. Yu et al (2019) shows that the Canadian wildfire plumes 2014; Ditas et al., 2018). The thick coating is supported by earlier rose from 12 to 23 km within 2 months owing to solar heating of the studies, which show that the shell-core ratio based on SP2 measure- wildfire aerosols, extending their lifetime and latitudinal spread. In the ments is 1.3 to 1.6 for fresh NA wildfire plumes (Kondo et al., 2011), absence of measurements, model simulations suggest the observed and ~2.25 after days of aging (Taylor et al., 2014; Ditas et al., 2018). In plume rising is best explained by aged wildfire aerosol particles of a addition, BC particles with thick coating also help explain the hygro- coated fractal structure with 2% BC mass and corresponding ω532 scopic growth shown by the HTDMA. We note that thickly coated BC of ~0.82, or a coated sphere structure with 3% BC mass and corre- particles alone cannot explain the relatively high ω529 (0.92~0.95) (see sponding ω532 of ~0.92. Our measurements of aerosol optical proper- SI S2). Earlier studies (Lack et al., 2012; Ditas et al., 2018) show that ties, however, are best explained by an external mixture of non-ab- large amounts of particulate organics can be externally mixed with BC- sorbing particles and coated sphere BC particles, which differs from the containing particles in biomass burning plumes. The aged wildfire scenarios above. This provides observational constraints for further aerosols observed are therefore likely external mixtures of non-ab- evaluations of the transport and direct radiative effects of the wildfire sorbing organic particles and thickly coated particles with a large BC aerosols. core. The averaged shell-core ratio and the number fraction of externally mixed non-absorbing particles are derived from measured op- 5. Evolutions in CCN concentrations tical properties and aerosol size distributions (see SI S2). The results suggest that on average, BC particles have a shell-core ratio above ~2 5.1. Ambient number size distributions and non-absorbing particles represent a majority of the aged wildfire aerosols. In regional or global models, size distributions of fresh biomass burning aerosols are often assumed to be unimodal, with default

4.3. Implications on aerosol direct effects number mode diameter (Dpn) of 80 to 150 nm and mode width (σn)of 1.6 to 2 (Stier et al., 2005; Dentener et al., 2006; Ramnarine et al., Given the elevated injection height, wildfire plumes can stay aloft 2018). The default mode parameters are based on the observed number for extended periods of time and spread over large geographic areas, size distributions of fresh biomass burning plumes (Janhäll et al., 2010; including the tropics (Yu et al., 2019; Khaykin et al., 2018; Kloss et al., Levin et al., 2010). These simulation (Sakamoto et al., 2016; Ramnarine 2019). In addition, the radiative forcing efficiency at higher latitudes et al., 2018) and field observations (Janhäll et al., 2010; Levin et al., are much stronger (up to ~10 times) than at the lower altitudes 2010) show that the size distribution of biomass burning aerosols re-

(Zarzycki and Bond, 2010; Samset and Myhre, 2011; Ban-Weiss et al., mains unimodal, but with increasing Dpn (170 ~ 300 nm) and de- 2012; Samset et al., 2013). The optical properties presented here help creasing σn (1.7 to 1.3) during the aging in the atmosphere. With suf- quantify the radiative effects of wildfire aerosols and BrC. When the ficient time, σn is expected to converge towards the coagulation limit optical properties of BrC are parameterized using measurements of of ~1.2 (Sakamoto et al., 2016; Ramnarine et al., 2018). To capture the freshly emitted biomass burning aerosols (Saleh et al., 2014) (i.e., loss variation trend, some recent models are assuming two size modes for of BrC is not taken into consideration), simulated global mean direct biomass burning aerosols, one fresh mode and one aged mode (Liu radiative effects (DRE) of BrC assuming internal mixtures generally et al., 2016b; Rasch et al., 2019). After heating to 300 to 400 ℃, most − ranged from 0.08 to 0.13 W m 2 (Saleh et al., 2015; Jo et al., 2016; observational studies show an unimodal distribution of biomass Brown et al., 2018; Zhang et al., 2019). This is comparable to that of burning aerosols (Clarke et al., 2004; Clarke et al., 2007; Rose et al., − BC. The DRE of BrC decreases by over 50% to ~0.05 W m 2, when 25% 2011; Ueda et al., 2016). BrC absorption is assumed to remain following photo-bleaching (Wang Unlike the unimodal size distribution represented in some global et al., 2018; Brown et al., 2018). The DRE further decreases models, aerosols observed during this wildfire episode show a bimodal − to ~0.013 W m 2 if a smaller fraction (i.e., 6%) of primary BrC ab- number size distribution (Fig. 6), containing one minor Aitken mode sorption remains after photo-bleaching (Zhang et al., 2019). Our results (At: Dpn = 88 nm, σn = 2.11) and a dominant accumulation mode (Ac: here indicate negligible contribution of BrC to absorption in the aged Dpn = 230 nm, σn = 1.48). Note that the Aitken mode evident in the wildfire aerosols. If this finding holds for wildfires in other regions as number size distribution has a negligible contribution to the volume well, it will suggest that the overall DRE of BrC may be minor on global size distribution (Fig. 4) due to the relatively small particle diameters.

Fig. 6. Number size distributions observed during the long-range transported Canadian wildﬁre episode when CO > 140 ppb. The black circles indicate the observed size distribution, based on which the lognormal ﬁttings are conducted to derive modal parameters. For both ambient and non-volatile aerosols, the size distributions are bimodal, including one Aitken (At) mode and one accumulation (Ac) mode.

8 G. Zheng, et al. Environment International 139 (2020) 105680

Table 2 Summary of ambient number size distributions from ﬁeld observations for NA wildﬁre aerosols.

Source Region Observation Site Plume Type Year Campaign Modes observed in Aitken - Accumulation Reference mode ranges (30~400 nm)

Mode number Dp,n (nm) σ

Canadian forest fire FT of eastern Canada fresh (< a few hours) 2008 NASA ARCTAS 1 224 1.33 Kondo et al., 2011 Western US wildfires Mt. Bachelor Observatory, FT in fresh to slightly aged 2015 / 1 170 1.67 Laing et al. central Oregon, U.S. (3–35 h) (2016) Forest fires in northwest FT of eastern Canada aged for 1~2 days 2011 BORTAS-B 1 ~220a n/a Taylor et al. Ontario (2014) NA wildfire FT in Central Europe aged for 4–6 days 2004 ICARTT-ITOP 2 85 1.9 Petzold et al. 230 1.39 (2007) Northern Canada lower-middle FT (3.8–4.3 km), aged for 6–7 days 1998 LACE 98 2 57 2 (Fiebig et al., Lindenberg, Germany 340 1.35 2003) middle FT (4.7–5.2 km), 2502 Lindenberg, Germany 230 1.45 NA wildfire FT in Central Europe aged for 6–9 days 2004 ICARTT-ITOP 3 65 1.6 Petzold et al. 260 1.3 (2007) 350 1.8 aged for 6–9 days 2 80 1.6 270 1.31 aged for 6–9 days 2 78 1.56 260 1.31 aged for 7–10 days 2 90 1.4 300 1.3 aged for 10–13 days 2 65 1.64 270 1.32 Canadian forest wildfire MBL in eastern North Atlantic aged for over 10 days 2017 ACE-ENA 2 88 2.11 This study 230 1.48

a Estimated from Fig. 4 in Taylor et al., (2014)

The non-volatile component also exhibited a bimodal number size values over 0.6. Such an increase of κCCN is consistent with condensa- distribution during this wildfire episode (Fig. 6). The bimodal dis- tion of sulfate and in-cloud production of sulfate, which can grow the tributions of non-volatile components are not an artifact due to time- Aitken mode particles into the accumulation mode size range (Zheng averaging, but was observed continuously during periods when aerosol et al., 2018). was dominated by the aged wildfire particles (Fig. 1). While bimodal As shown in Fig. 7a, both κCCN and κGF show strong negative cor- distribution is rarely observed in fresh or slightly-aged NA wildfire relation with CO mixing ratio. κCCN decreased from ~0.55 for back- plumes (Kondo et al., 2011; Taylor et al., 2014; Laing et al., 2016), it ground particles to ~0.3 for well-aged wildfire particles. The decrease has been reported for well-aged NA plumes (plumes aged over 6 days) is especially evident for the accumulation mode (Dp > 75 nm) parti- (Fiebig et al., 2003; Petzold et al., 2007)(Table 2). cles, which is the dominate mode of wildfire aerosols (Fig. 2). The κCCN value of wildfire aerosols (~0.3) is in good agreement with previous lab 5.2. Aerosol hygroscopicity and the influencing factors studies of aerosols produced using typical biomass fuels consumed in NA wildfires. These studies (Petters et al., 2009a; Engelhart et al., 2012; κ fi Aerosol hygroscopicity can be quantified by hygroscopicity para- Lathem et al., 2013) show that, although CCN of fresh wild re aerosols meter, κ (Petters and Kreidenweis, 2007), which ranges from 0 to 1.3 can vary substantially (0.06 to 0.8), the value quickly converges to a κ for representative atmospheric aerosol particles. Atmospheric particles range of 0.08 to 0.3 after several hours of aging. The CCN value for the fi consisting mostly of inorganic species typically have higher κ values wild re aerosols is also consistent with the range of aged biomass than those enriched in organic compounds. Particle hygroscopicity burning aerosols observed in central Amazonia (Thalman et al., 2017; parameter κ can be derived under super-saturated conditions from size- Palm et al., 2018). The value of κGF derived from HTDMA measurement shows similar resolved CCN activation spectrum (κCCN, section 2.2), or under sub-saturation conditions from particle hygroscopic growth factor (κ ) negative correlation with CO, decreasing from ~ 0.35 under back- GF fi κ measured by HTDMA (section 2). Aerosols observed during the wildfire ground conditions to ~0.15 for wild re aerosols. The value of GF is κ ff episode were classified into three groups based on the concurrent consistently lower than that of CCN for all three groups. Such di erence measurement of CO mixing ratio: background (CO < 110 ppb), in- has been observed in a number of earlier studies (e.g., Wex et al., 2009; termediate (110 < CO < 140 ppb), and well-aged wildfire aerosols Ovadnevaite et al., 2011) when aerosols consist of substantial amounts (CO > 140 ppb). Derived values of κ are shown for each group (Fig. 7). of organic compounds with higher molecular weights, and was attributed to high non-ideality of compounds in aerosol water under sub- In background air masses, κCCN derived at all six diameters (i.e., 40, saturated conditions (Petters et al., 2009b), the conversion of hydrogels 50, 75, 100, 125, and 150 nm) were around 0.55. This κCCN values suggest a mixture dominated by sulfate (including sulphuric acid under sub-saturated conditions into multiple smaller molecules under (κ ~ 0.9), ammonium sulfate (κ ~ 0.6) and ammonium bisulfate super-saturated conditions (Ovadnevaite et al., 2011), and/or the dif- (κ ~ 0.7)) and organic compounds (κ generally below 0.4) (Petters and ference in solubility and phase states under sub- and super-saturated Kreidenweis, 2007; Schmale et al., 2018). This is further supported by conditions (Pajunoja et al., 2015; Rastak et al., 2017). κ the lower κ than κ under background conditions, which suggested The variation in can be further explained by the non-volatile vo- GF CCN fi the presence of organics (see discussions below). Moreover, under the lume fractions (fv, TD), de ned as: background conditions, while the majority (> 90%) of the 40 and fv, TD = (corresponding Dp after heated to 300 ℃ / selected Dp in 50 nm particles have κCCN below 0.6, a substantial fraction (25% to size-resolved CCN measurements)3 50%) of particles with diameters of 100, 125, and 150 nm exhibit κCCN

9 G. Zheng, et al. Environment International 139 (2020) 105680

Fig. 7. Dependence of hygroscopicity parameter, κ, with CO and aerosol sizes. (a) Super-saturated κ by size-resolved CCN measurement, or κCCN. (b) Sub-saturated κ by HTDMA measurement, or κGF. The boxes and whiskers indicate 5th, 25th, 50th, 75th and 95th percentiles, respectively. The blue circles indicate group means.

Fig. 8. Dependence of wildﬁre aerosol hygro-

scopicity with the non-volatile volume fraction, fv,

TD. Only data during wildﬁre-inﬂuenced periods (i.e., when CO > 110 ppb) are shown. (a) Super- saturated κ from size-resolved CCN measurement

(κCCN), and (b) sub-saturated κ by HTDMA mea-

surement (κGF). The black dash lines are used to

indicate the general trend. Both κCCN and κGF decrease with increasing non-volatile volume frac-

tion. See derivations of the fv, TD in SI S3.

which can be estimated from the paired size distributions of am- lower κ (Wang et al., 2019). The variations of κCCN and κGF at given fv, bient and thermally denuded aerosols (SI S2; Fig S2). Here we derived TD levels can be caused by differences in the oxygen content of aerosols fv, TD only during the wildfire-influenced periods (i.e., when CO > (e.g., O:C ratio) (Jimenez et al., 2009; Duplissy et al., 2011)orby 110 ppb). Both κCCN and κGF are negatively correlated with fv, TD thermal volatilities (Massoli et al., 2010; Rickards et al., 2013), as well (Fig. 8), as non-volatile components are likely water insoluble BC or as uncertainties in fv, TD estimations (SI S2). organics with higher molecular weight, which is shown to result in

10 G. Zheng, et al. Environment International 139 (2020) 105680

Fig. 9. Contribution of wildfire aerosols to CCN concentrations at different super-saturation (ss) levels. The boxes and whiskers indicate 5th, 25th, 50th, 75th and 95th percentiles, respectively, of the data observed during the one-year ACE-ENA campaign. The orange dots indicate the data observed during the wildfire episodes.

5.3. Contribution to CCN concentrations and implications on aerosol 1974; Albrecht, 1989; Lu et al., 2018). Given the expected increases of indirect effects NA wildfire intensity and frequency, the high susceptibility of marine clouds albedo and precipitation to the perturbation of CCN population, The shape of number size distribution (section 5.1) and the aerosol wildfire aerosols may have important influences on the climate effects hygroscopicity (section 5.2) essentially dictate the fraction of aerosol of marine low clouds. particles that serve as CCN at given super-saturation level. Compared to fi backgound ones (Zheng et al., 2018), the wild re aerosols show higher 6. Summary number fraction and larger mode diameter in the accumulation mode κ particles, which compensated the lower values, and lead to higher The elevated injection height of North American wildfire plumes CCN-active fractions (Fig. 9). Under lower super-saturation levels of leads to extended aerosol lifetimes and geographic spread, therefore 0.1% and 0.2% which are representative for marine low clouds (Wood, enhanced influences on radiation and climate. As the properties of NA 2012), average CCN-active fractions are respectively 22% and 36% for wildfire aerosols may evolve substantially during their long transport in fi fl backgaround aerosols. In comparison, those during the wild re-in u- the atmosphere, the properties of well-aged wildfire aerosols are enced episodes almost doubled (increased to 41% and 62%, respec- needed to quantify their impact on radiation and climate given the long tively). This increase also corresponds to higher absolute CCN con- aerosol lifetime. In August 2017, extreme Canadian wildfires injected a centrations at 0.1% and 0.2% super-saturations during this wildfire – −3 large amount of aerosols into the upper troposphere lower strato- episode (302 and 459 cm , respectively), which are both ~ 3 times sphere (~12 km). Some of the plumes rose from 12 to 23 km in higher than the annual average. The tripled CCN concentrations versus 2 months due to solar heating of absorbing wildfire aerosols (Fig. 10), fi doubled CCN-active fractions are due to the higher CN during wild re extending the lifetime (i.e., up to 8 months) and geographic spread (Yu episode. et al., 2019). In contrast, we show some of the wildfire plumes aloft Given the elevated injection heights, the long-range transported travelled across the Atlantic Ocean and descended into the remote fi wild re aerosols are conventionally considered as largely separated marine boundary layer (MBL) over the eastern North Atlantic from underlying marine low clouds (Painemal et al., 2014; Rajapakshe (ENA), > 7000 km away within about two weeks. These long-range fi et al., 2017). Therefore, earlier studies of the wild re aerosol impacts transported wildfire aerosols dominated the properties of MBL aerosol on marine low clouds mainly focused on the cloud adjustment to the in the ENA from Aug. 24th to 29th, 2017. Our analysis indicates that warming caused by scattering and absorption of solar radiation by this accelerated descent is facilitated by dry intrusions associated with fi wild re aerosols above the marine low clouds (Johnson et al., 2004; mid-latitude cyclones (Fig. 10). ff Wilcox, 2012; Zhang et al., 2016), while the e ects on MBL CCN and During the episode from Aug 24 to 29, the characteristics of biomass therefore cloud microphysics have received less attention. Recent sa- burning plumes are evident from trace gases and aerosol measurements tellite and modelling studies suggested contact of biomass burning at the ENA site on Graciosa Island, Azores. These characteristics include aerosols with marine cloud decks (Costantino and Bréon, 2010; the extremely high CO mixing ratio and concentration of non-volatile Costantino and Bréon, 2013; Painemal et al., 2014; Lu et al., 2018). accumulate mode particles, linear correlations among CO, ambient and However, direct evidence of such contact is still limited. Our results non-volatile aerosol number and volume concentrations, and the fi provide direct evidence that the long-range transported wild re aero- dominance of aerosol composition by organics. As MBL aerosol was sols can descend into MBL and dominate the CCN population. The dominated by long-ranged transported plume, this provides an ex- fl fi elevated CCN concentration due to the in uence of wild re aerosols cellent opportunity to comprehensively characterize the climate-related ff can result in cooling due to increased albedo (i.e., the Twomey e ect) properties of well-aged wildfire aerosols. We show that the well-aged ff and coverage (cloud lifetime e ect) of marine low clouds (Twomey, wildfire aerosols have high single scattering albedos at 529 nm of

11 G. Zheng, et al. Environment International 139 (2020) 105680

Fig. 10. Schematic showing the transport of intense NA wildfire plumes and its impact on radiation and remote MBL CCN and clouds. Some of the wildfire plumes may rise further into stratosphere due to the solar heating of the absorbing wildfire aerosols, further extending the lifetime and geographic spread. The BrC emitted by wildfires is likely lost and/or bleached completely during the transport of the wildfire aerosols. Some wildfire plumes experience accelerated descent due to dry intrusions following mid-latitude upper-tropospheric troughs and surface cyclones. These plumes may directly descend into remote MBL or entrain from lower free troposphere into MBL, leading to an increased CCN concentration. The high CCN concentration is expected to lead to net cooling as a result of increased cloud albedo (i.e., the Twomey effect) and cloud coverage.

0.92~0.95, and very low absorption Ångström exponents (Åabs)at NA wildfires. The accelerated descent facilitated by dry intrusion is fi 464 nm / 648 nm of < 1. In contrast, Åabs of fresh wild re aerosols and expected to occur regularly following extratropical cyclones (Raveh- those aged for less than a few days typically ranges from ~1.4 to ~ 3.5. Rubin, 2017; Naud et al., 2018). Given the expected increases of NA The very low Åabs value indicates negligible contribution of brown wildfire intensity and frequency, the effects of NA wildfire on CCN and carbon to aerosol absorption, suggesting a nearly complete loss of BrC clouds in remote marine environment therefore need to be further ex- during the transport of the wildfire aerosols (Fig. 10). This nearly amined. complete loss of BrC suggests that on global average, the direct radia- − tive effect of brown carbon is likely minor (i.e., < 0.05 W m 2)(Brown et al., 2018; Wang et al., 2018). Combining the Mie calculations with Declaration of Competing Interest the measured aerosol hygroscopicity, volatility and size distributions, we show that the high ω and low Å values are best explained by an 529 abs The authors declare that they have no known competing financial external mixture of non-absorbing organic particles and absorbing interests or personal relationships that could have appeared to influ- particles that are large BC cores (> ~110 nm diameter) and thick non- ence the work reported in this paper. absorbing coatings. The accelerated descent of wildfire plumes significantly increased the CCN concentration inside MBL over the Azores (Fig. 10). The Acknowledgements number size distribution of the aged wildfire aerosols is dominated by an accumulation mode with mode diameter of 230 nm, much larger The research was conducted with funding from the Atmospheric than that of typical background marine aerosols (~157 nm (Zheng System Research (ASR) program (Award No. DE-SC0020259), Office of et al., 2018)). Hygroscopicity parameter κ of well-aged wildfire CCN Biological and Environmental Research (OBER) of the United States aerosol generally varies between 0.2~0.4, consistent with results from Department of Energy. We acknowledge additional support by the earlier studies and lower than the value of ~ 0.55 for the background Atmospheric Radiation Measurement (ARM) Climate Research Facility, marine aerosols in ENA. Despite the low κ value, aged wildfire CCN a user facility of the United States Department of Energy, Office of aerosols have higher activated fraction at representative super-satura- Science, sponsored by the Office of Biological and Environmental tion levels of marine low clouds (0.1% ~ 0.2%) due to the dominance of Research. AJS and ERL were supported by the U.S. Department of accumulation mode particles and large mode diameter. Correspond- Energy's Atmospheric System Research Program under contract DE- ingly, the CCN concentrations at 0.1% and 0.2% super-saturations − SC0012704. SRR is partially supported by the Israel Science Foundation during the wildfire episode (302 and 459 cm 3, respectively) are (grant No. 1347/18). We acknowledge the use of data and imagery both ~3 times higher than the annual average. Our results provide from LANCE FIRMS operated by the NASA/GSFC/Earth Science Data direct evidences that the long-range transported wildfire aerosols can and Information System (ESDIS) with funding provided by NASA/HQ. descend into MBL and dominate the CCN population. Because the al- We acknowledge the Israel Meteorological Service for providing access bedo (i.e., reflectivity), drizzle formation, and lifetime of marine low to ECMWF ERA-Interim data. clouds are particularly sensitive to the change in CCN concentration, the higher CCN concentrations are expected to lead to reduced droplet effective radius and drizzle formation, therefore cooling of climate as a Appendix A. Supplementary data result of increased cloud albedo (i.e., the Twomey effect) and increase cloud coverage (cloud lifetime effect) (Twomey, 1974; Albrecht, 1989; Supplementary data to this article can be found online at https:// Lu et al., 2018). This suggests significant cloud-related climate effects of doi.org/10.1016/j.envint.2020.105680.

12 G. Zheng, et al. Environment International 139 (2020) 105680

References Costantino, L., Bréon, F.M., 2013. Aerosol indirect effect on warm clouds over South-East Atlantic, from co-located MODIS and CALIPSO observations. Atmos. Chem. Phys. 13, 69–88. https://doi.org/10.5194/acp-13-69-2013. Abatzoglou, J.T., Williams, A.P., 2016. Impact of anthropogenic climate change on Crutzen, P.J., Andreae, M.O., 1990. Biomass Burning in the Tropics: Impact on fi wild re across western US forests. Proceedings of the National Academy of Sciences Atmospheric Chemistry and Biogeochemical Cycles. Science 250, 1669–1678. – 113, 11770 11775. https://doi.org/10.1073/pnas.1607171113. https://doi.org/10.1126/science.250.4988.1669. Adebiyi, A.A., Zuidema, P., 2016. The role of the southern African easterly jet in mod- Cubison, M.J., Ortega, A.M., Hayes, P.L., Farmer, D.K., Day, D., Lechner, M.J., Brune, ifying the southeast Atlantic aerosol and cloud environments. Q J Roy Meteor Soc W.H., Apel, E., Diskin, G.S., Fisher, J.A., Fuelberg, H.E., Hecobian, A., Knapp, D.J., – 142, 1574 1589. https://doi.org/10.1002/qj.2765. Mikoviny, T., Riemer, D., Sachse, G.W., Sessions, W., Weber, R.J., Weinheimer, A.J., Ahern, A.T., Robinson, E.S., Tkacik, D.S., Saleh, R., Hatch, L.E., Barsanti, K.C., Stockwell, Wisthaler, A., Jimenez, J.L., 2011. Eff ects of aging on organic aerosol from open C.E., Yokelson, R.J., Presto, A.A., Robinson, A.L., Sullivan, R.C., Donahue, N.M., biomass burning smoke in aircraft and laboratory studies. Atmos. Chem. Phys. 11, 2019. Production of Secondary Organic Aerosol During Aging of Biomass Burning 12049–12064. https://doi.org/10.5194/acp-11-12049-2011. Smoke From Fresh Fuels and Its Relationship to VOC Precursors. Journal of Danielsen, E.: Project Springfield report. DASA 1517, Defense Atomic Support Agency, – Geophysical Research: Atmospheres 124, 3583 3606. https://doi.org/10.1029/ Washington, DC 20301, 97 pp.[NTIS AD-607980]., 1968: Stratospheric-tropospheric 2018jd029068. exchange based on radioactivity, ozone and potential vorticity, J. Atmos. Sci, 25, Albrecht, B.A., 1989. Aerosols, Cloud Microphysics, and Fractional Cloudiness. Science 502-518, 1964. – 245, 1227 1230. https://doi.org/10.1126/science.245.4923.1227. Dee, D.P., Uppala, S.M., Simmons, A.J., Berrisford, P., Poli, P., Kobayashi, S., Andrae, U., Andreae, M.O., Merlet, P., 2001. Emission of trace gases and aerosols from biomass Balmaseda, M.A., Balsamo, G., Bauer, P., Bechtold, P., Beljaars, A.C.M., van de Berg, – burning. Global Biogeochemical Cycles 15, 955 966. https://doi.org/10.1029/ L., Bidlot, J., Bormann, N., Delsol, C., Dragani, R., Fuentes, M., Geer, A.J., 2000GB001382. Haimberger, L., Healy, S.B., Hersbach, H., Hólm, E.V., Isaksen, L., Kållberg, P., Ansmann, A., Baars, H., Chudnovsky, A., Mattis, I., Veselovskii, I., Haarig, M., Seifert, P., Köhler, M., Matricardi, M., McNally, A.P., Monge-Sanz, B.M., Morcrette, J.-J., Park, fi Engelmann, R., Wandinger, U., 2018. Extreme levels of Canadian wild re smoke in B.-K., Peubey, C., de Rosnay, P., Tavolato, C., Thépaut, J.-N., Vitart, F., 2011. The – the stratosphere over central Europe on 21 22 August 2017. Atmos. Chem. Phys. 18, ERA-Interim reanalysis: configuration and performance of the data assimilation – 11831 11845. https://doi.org/10.5194/acp-18-11831-2018. system. Q J Roy Meteor Soc 137, 553–597. https://doi.org/10.1002/qj.828. Ban-Weiss, G.A., Cao, L., Bala, G., Caldeira, K., 2012. Dependence of climate forcing and Dennison, P.E., Brewer, S.C., Arnold, J.D., Moritz, M.A., 2014. Large wildfire trends in the – response on the altitude of black carbon aerosols. Clim Dynam 38, 897 911. https:// western United States, 1984–2011. Geophysical Research Letters 41, 2928–2933. doi.org/10.1007/s00382-011-1052-y. https://doi.org/10.1002/2014gl059576. Bergstrom, R.W., Pilewskie, P., Russell, P.B., Redemann, J., Bond, T.C., Quinn, P.K., Dentener, F., Kinne, S., Bond, T., Boucher, O., Cofala, J., Generoso, S., Ginoux, P., Gong, Sierau, B., 2007. Spectral absorption properties of atmospheric aerosols. Atmos. S., Hoelzemann, J.J., Ito, A., Marelli, L., Penner, J.E., Putaud, J.P., Textor, C., Schulz, – Chem. Phys. 7, 5937 5943. https://doi.org/10.5194/acp-7-5937-2007. M., van der Werf, G.R., Wilson, J., 2006. Emissions of primary aerosol and precursor Bond, T.C., Doherty, S.J., Fahey, D.W., Forster, P.M., Berntsen, T., DeAngelo, B.J., gases in the years 2000 and 1750 prescribed data-sets for AeroCom. Atmos. Chem. Flanner, M.G., Ghan, S., Kärcher, B., Koch, D., Kinne, S., Kondo, Y., Quinn, P.K., Phys. 6, 4321–4344. https://doi.org/10.5194/acp-6-4321-2006. fi Saro m, M.C., Schultz, M.G., Schulz, M., Venkataraman, C., Zhang, H., Zhang, S., Ditas, J., Ma, N., Zhang, Y., Assmann, D., Neumaier, M., Riede, H., Karu, E., Williams, J., Bellouin, N., Guttikunda, S.K., Hopke, P.K., Jacobson, M.Z., Kaiser, J.W., Klimont, Z., Scharffe, D., Wang, Q., Saturno, J., Schwarz, J. P., Katich, J. M., McMeeking, G. R., Lohmann, U., Schwarz, J.P., Shindell, D., Storelvmo, T., Warren, S.G., Zender, C.S., Zahn, A., Hermann, M., Brenninkmeijer, C. A. M., Andreae, M. O., Pöschl, U., Su, H., fi 2013. Bounding the role of black carbon in the climate system: A scienti c assess- and Cheng, Y.: Strong impact of wildfires on the abundance and aging of black carbon – ment. Journal of Geophysical Research: Atmospheres 118, 5380 5552. https://doi. in the lowermost stratosphere, Proceedings of the National Academy of Sciences, 10. org/10.1002/jgrd.50171. 1073/pnas.1806868115, 2018. Brown, H., Liu, X., Feng, Y., Jiang, Y., Wu, M., Lu, Z., Wu, C., Murphy, S., Pokhrel, R., Duplissy, J., DeCarlo, P.F., Dommen, J., Alfarra, M.R., Metzger, A., Barmpadimos, I., ff 2018. Radiative e ect and climate impacts of brown carbon with the Community Prevot, A.S.H., Weingartner, E., Tritscher, T., Gysel, M., Aiken, A.C., Jimenez, J.L., – Atmosphere Model (CAM5). Atmos. Chem. Phys. 18, 17745 17768. https://doi.org/ Canagaratna, M.R., Worsnop, D.R., Collins, D.R., Tomlinson, J., Baltensperger, U., 10.5194/acp-18-17745-2018. 2011. Relating hygroscopicity and composition of organic aerosol particulate matter. Browning, K.A., 1997. The dry intrusion perspective of extra-tropical cyclone develop- Atmos. Chem. Phys. 11, 1155–1165. https://doi.org/10.5194/acp-11-1155-2011. – ment. Meteorological Applications 4, 317 324. https://doi.org/10.1017/ Engelhart, G.J., Hennigan, C.J., Miracolo, M.A., Robinson, A.L., Pandis, S.N., 2012. Cloud S1350482797000613. condensation nuclei activity of fresh primary and aged biomass burning aerosol. Cappa, C.D., Lack, D.A., Burkholder, J.B., Ravishankara, A.R., 2008. Bias in Filter-Based Atmos. Chem. Phys. 12, 7285–7293. https://doi.org/10.5194/acp-12-7285-2012. Aerosol Light Absorption Measurements Due to Organic Aerosol Loading: Evidence Fiebig, M., Stohl, A., Wendisch, M., Eckhardt, S., Petzold, A., 2003. Dependence of solar – from Laboratory Measurements. Aerosol Science and Technology 42, 1022 1032. radiative forcing of forest fire aerosol on ageing and state of mixture. Atmos. Chem. https://doi.org/10.1080/02786820802389285. Phys. 3, 881–891. https://doi.org/10.5194/acp-3-881-2003. Catto, J.L., Raveh-Rubin, S., 2019. Climatology and dynamics of the link between dry Field, Robert D., van der Werf, Guido R., Shen, Samuel S.P., 2009. Human amplification intrusions and cold fronts during winter. Part I: global climatology. Clim Dynam. of drought-induced biomass burning in Indonesia since 1960. Nature Geosci 2 (3), https://doi.org/10.1007/s00382-019-04745-w. 185–188. https://doi.org/10.1038/ngeo443. https://www.nature.com/articles/ Chow, J.C., Watson, J.G., Doraiswamy, P., Chen, L.-W.A., Sodeman, D.A., Lowenthal, ngeo443#supplementary-information, 2009. D.H., Park, K., Arnott, W.P., Motallebi, N., 2009. Aerosol light absorption, black Forrister, H., Liu, J., Scheuer, E., Dibb, J., Ziemba, L., Thornhill, K.L., Anderson, B., carbon, and elemental carbon at the Fresno Supersite. California, Atmospheric Diskin, G., Perring, A.E., Schwarz, J.P., Campuzano-Jost, P., Day, D.A., Palm, B.B., – Research 93, 874 887. https://doi.org/10.1016/j.atmosres.2009.04.010. Jimenez, J.L., Nenes, A., Weber, R.J., 2015. Evolution of brown carbon in wildfire Clarke, A., McNaughton, C., Kapustin, V., Shinozuka, Y., Howell, S., Dibb, J., Zhou, J., plumes. Geophysical Research Letters 42, 4623–4630. https://doi.org/10.1002/ Anderson, B., Brekhovskikh, V., Turner, H., and Pinkerton, M.: Biomass burning and 2015gl063897. fl pollution aerosol over North America: Organic components and their in uence on Frank, G.P., Dusek, U., Andreae, M.O., 2006. Technical note: A method for measuring fi spectral optical properties and humidi cation response, Journal of Geophysical size-resolved CCN in the atmosphere. Atmos. Chem. Phys. Discuss. 6, 4879–4895. Research: Atmospheres, 112, n/a-n/a, 10.1029/2006JD007777, 2007. https://doi.org/10.5194/acpd-6-4879-2006. fi Clarke, A., Kapustin, V., 2010. Hemispheric Aerosol Vertical Pro les: Anthropogenic Gavin, D.G., Hallett, D.J., Hu, F.S., Lertzman, K.P., Prichard, S.J., Brown, K.J., Lynch, – Impacts on Optical Depth and Cloud Nuclei. Science 329, 1488 1492. https://doi. J.A., Bartlein, P., Peterson, D.L., 2007. Forest fire and climate change in western org/10.1126/science.1188838. North America: insights from sediment charcoal records. Frontiers in Ecology and the Clarke, A. D., Shinozuka, Y., Kapustin, V. N., Howell, S., Huebert, B., Doherty, S., Environment 5, 499–506. https://doi.org/10.1890/060161. Anderson, T., Covert, D., Anderson, J., Hua, X., Moore, K. G., McNaughton, C., Gyawali, M., Arnott, W.P., Lewis, K., Moosmüller, H., 2009. In situ aerosol optics in Reno, Carmichael, G., and Weber, R.: Size distributions and mixtures of dust and black NV, USA during and after the summer 2008 California wildfires and the influence of fl carbon aerosol in Asian out ow: Physiochemistry and optical properties, Journal of absorbing and non-absorbing organic coatings on spectral light absorption. Atmos. Geophysical Research: Atmospheres, 109, n/a-n/a, 10.1029/2003JD004378, 2004. Chem. Phys. 9, 8007–8015. https://doi.org/10.5194/acp-9-8007-2009. Clarke, A.D., Freitag, S., Simpson, R.M.C., Hudson, J.G., Howell, S.G., Brekhovskikh, V.L., Haviland, F., Jiumeng, L., Eric, S., Jack, D., Luke, Z., L., T. K., Bruce, A., Glenn, D., E., P. Campos, T., Kapustin, V.N., Zhou, J., 2013. Free troposphere as a major source of A., P., S. J., Pedro, C.-J., A., D. D., B., P. B., L., J. J., Athanasios, N., and J., W. R.: fi CCN for the equatorial paci c boundary layer: long-range transport and tele- Evolution of brown carbon in wildfire plumes, Geophysical Research Letters, 42, – connections. Atmos. Chem. Phys. 13, 7511 7529. https://doi.org/10.5194/acp-13- 4623-4630, doi:10.1002/2015GL063897, 2015. 7511-2013. Jacobson, Mark Z., 2001. Strong radiative heating due to the mixing state of black carbon fi Colarco, P.R., et al., 2004. Transport of smoke from Canadian forest res to the surface in atmospheric aerosols. Nature 409 (6821), 695–697. https://doi.org/10.1038/ near Washington, D.C.: Injection height, entrainment, and optical properties. J. 35055518. https://www.nature.com/articles/35055518#supplementary- Geophys. Res. Atmosp. 109. https://doi.org/10.1029/2003JD004248. information, 2001. ff Collier, S., Zhou, S., Onasch, T.B., Ja e, D.A., Kleinman, L., Sedlacek, A.J., Briggs, N.L., Jacobson, M.Z., 2004. The Short-Term Cooling but Long-Term Global Warming Due to Hee, J., Fortner, E., Shilling, J.E., Worsnop, D., Yokelson, R.J., Parworth, C., Ge, X., Biomass Burning. Journal of Climate 17, 2909–2926. https://doi.org/10.1175/1520- fi Xu, J., Butter eld, Z., Chand, D., Dubey, M.K., Pekour, M.S., Springston, S., Zhang, 0442(2004)017<2909:tscblg>2.0.co;2. fl fi Q., 2016. Regional In uence of Aerosol Emissions from Wild res Driven by Janhäll, S., Andreae, M.O., Pöschl, U., 2010. Biomass burning aerosol emissions from ffi Combustion E ciency: Insights from the BBOP Campaign. Environmental Science & vegetation fires: particle number and mass emission factors and size distributions. – Technology 50, 8613 8622. https://doi.org/10.1021/acs.est.6b01617. Atmos. Chem. Phys. 10, 1427–1439. https://doi.org/10.5194/acp-10-1427-2010. Costantino, L., Bréon, F.-M., 2010. Analysis of aerosol-cloud interaction from multi-sensor Jimenez, J.L., Canagaratna, M.R., Donahue, N.M., Prevot, A.S.H., Zhang, Q., Kroll, J.H., satellite observations. Geophysical Research Letters 37. https://doi.org/10.1029/ DeCarlo, P.F., Allan, J.D., Coe, H., Ng, N.L., Aiken, A.C., Docherty, K.S., Ulbrich, I.M., 2009gl041828. Grieshop, A.P., Robinson, A.L., Duplissy, J., Smith, J.D., Wilson, K.R., Lanz, V.A.,

13 G. Zheng, et al. Environment International 139 (2020) 105680

Hueglin, C., Sun, Y.L., Tian, J., Laaksonen, A., Raatikainen, T., Rautiainen, J., characterization of brown carbon (BrC) chromophores in secondary organic aerosol Vaattovaara, P., Ehn, M., Kulmala, M., Tomlinson, J.M., Collins, D.R., Cubison, M.J., generated from photo-oxidation of toluene. Phys Chem Chem Phys 17, 23312–23325. Dunlea, J., Huffman, J.A., Onasch, T.B., Alfarra, M.R., Williams, P.I., Bower, K., https://doi.org/10.1039/C5CP02563J. Kondo, Y., Schneider, J., Drewnick, F., Borrmann, S., Weimer, S., Demerjian, K., Liu, C., Chung, C.E., Yin, Y., Schnaiter, M., 2018. The absorption Ångström exponent of Salcedo, D., Cottrell, L., Griffin, R., Takami, A., Miyoshi, T., Hatakeyama, S., black carbon: from numerical aspects. Atmos. Chem. Phys. 18, 6259–6273. https:// Shimono, A., Sun, J.Y., Zhang, Y.M., Dzepina, K., Kimmel, J.R., Sueper, D., Jayne, doi.org/10.5194/acp-18-6259-2018. J.T., Herndon, S.C., Trimborn, A.M., Williams, L.R., Wood, E.C., Middlebrook, A.M., Liu, J., Lin, P., Laskin, A., Laskin, J., Kathmann, S.M., Wise, M., Caylor, R., Imholt, F., Kolb, C.E., Baltensperger, U., Worsnop, D.R., 2009. Evolution of Organic Aerosols in Selimovic, V., Shilling, J.E., 2016a. Optical properties and aging of light-absorbing the Atmosphere. Science 326, 1525–1529. https://doi.org/10.1126/science. secondary organic aerosol. Atmos. Chem. Phys. 16, 12815–12827. https://doi.org/ 1180353. 10.5194/acp-16-12815-2016. Jo, D.S., Park, R.J., Lee, S., Kim, S.W., Zhang, X., 2016. A global simulation of brown Liu, P.F., Abdelmalki, N., Hung, H.M., Wang, Y., Brune, W.H., Martin, S.T., 2015. carbon: implications for photochemistry and direct radiative effect. Atmos. Chem. Ultraviolet and visible complex refractive indices of secondary organic material Phys. 16, 3413–3432. https://doi.org/10.5194/acp-16-3413-2016. produced by photooxidation of the aromatic compounds toluene and < i > m < / Johnson, B., Shine, K., Forster, P., 2004. The semi-direct aerosol effect: Impact of ab- i > -xylene. Atmos. Chem. Phys. 15, 1435–1446. https://doi.org/10.5194/acp-15- sorbing aerosols on marine stratocumulus. Q J Roy Meteor Soc 130, 1407–1422. 1435-2015. Jolly, W. M., Cochrane, M. A., Freeborn, P. H., Holden, Z. A., Brown, T. J., Williamson, G. Liu, S., Aiken, A.C., Arata, C., Dubey, M.K., Stockwell, C.E., Yokelson, R.J., Stone, E.A., J., and Bowman, D. M. J. S.: Climate-induced variations in global wildfire danger Jayarathne, T., Robinson, A.L., DeMott, P.J., Kreidenweis, S.M., 2014. Aerosol single from 1979 to 2013, Nature Communications, 6, 7537, 10.1038/ncomms8537 scattering albedo dependence on biomass combustion efficiency: Laboratory and field https://www.nature.com/articles/ncomms8537#supplementary-information, 2015. studies. Geophysical Research Letters 41, 742–748. https://doi.org/10.1002/ Kelly, R., Chipman, M.L., Higuera, P.E., Stefanova, I., Brubaker, L.B., Hu, F.S., 2013. In: 2013gl058392. Proceedings of the National Academy of Sciences, pp. 13055–13060. https://doi.org/ Liu, X., Ma, P. L., Wang, H., Tilmes, S., Singh, B., Easter, R. C., Ghan, S. J., and Rasch, P. J. 10.1073/pnas.1305069110. : Description and evaluation of a new four-mode version of the Modal Aerosol Module Khaykin, S.M., Godin-Beekmann, S., Hauchecorne, A., Pelon, J., Ravetta, F., Keckhut, P., (MAM4) within version 5.3 of the Community Atmosphere Model, Geosci. Model 2018. Stratospheric Smoke With Unprecedentedly High Backscatter Observed by Dev., 9, 505-522, 10.5194/gmd-9-505-2016, 2016b. Lidars Above Southern France. Geophysical Research Letters 45, 1639–1646. https:// Liu, X., Huey, L. G., Yokelson, R. J., Selimovic, V., Simpson, I. J., Müller, M., Jimenez, J. doi.org/10.1002/2017gl076763. L., Campuzano-Jost, P., Beyersdorf, A. J., Blake, D. R., Butterfield, Z., Choi, Y., Kloss, C., Berthet, G., Sellitto, P., Ploeger, F., Bucci, S., Khaykin, S., Jégou, F., Taha, G., Crounse, J. D., Day, D. A., Diskin, G. S., Dubey, M. K., Fortner, E., Hanisco, T. F., Hu, Thomason, L.W., Barret, B., Le Flochmoen, E., von Hobe, M., Bossolasco, A., Bègue, W., King, L. E., Kleinman, L., Meinardi, S., Mikoviny, T., Onasch, T. B., Palm, B. B., N., Legras, B., 2019. Transport of the 2017 Canadian wildfire plume to the tropics and Peischl, J., Pollack, I. B., Ryerson, T. B., Sachse, G. W., Sedlacek, A. J., Shilling, J. E., global stratosphere via the Asian monsoon circulation. Atmos. Chem. Phys. Discuss. Springston, S., St. Clair, J. M., Tanner, D. J., Teng, A. P., Wennberg, P. O., Wisthaler, 1–22, 2019. https://doi.org/10.5194/acp-2019-204. A., and Wolfe, G. M.: Airborne measurements of western U.S. wildfire emissions: Kondo, Y., Matsui, H., Moteki, N., Sahu, L., Takegawa, N., Kajino, M., Zhao, Y., Cubison, Comparison with prescribed burning and air quality implications, Journal of M., Jimenez, J., Vay, S., 2011. Emissions of black carbon, organic, and inorganic Geophysical Research: Atmospheres, 122, 6108-6129, 10.1002/2016jd026315, aerosols from biomass burning in North America and Asia in 2008. Journal of 2017. Geophysical Research: Atmospheres 116. Lu, Z., Liu, X., Zhang, Z., Zhao, C., Meyer, K., Rajapakshe, C., Wu, C., Yang, Z., Penner, Labonne, M., Bréon, F.-M., Chevallier, F., 2007. Injection height of biomass burning J.E., 2018. Biomass smoke from southern Africa can significantly enhance the aerosols as seen from a spaceborne lidar. Geophysical Research Letters 34. https:// brightness of stratocumulus over. the southeastern Atlantic Ocean, Proceedings of the doi.org/10.1029/2007gl029311. National Academy of Sciences. Lack, D.A., Cappa, C.D., Covert, D.S., Baynard, T., Massoli, P., Sierau, B., Bates, T.S., Marlon, J.R., Bartlein, P.J., Carcaillet, C., Gavin, D.G., Harrison, S.P., Higuera, P.E., Joos, Quinn, P.K., Lovejoy, E.R., Ravishankara, A.R., 2008. Bias in Filter-Based Aerosol F., Power, M.J., Prentice, I.C., 2008. Climate and human influences on global biomass Light Absorption Measurements Due to Organic Aerosol Loading: Evidence from burning over the past two millennia. Nature Geosci 1 (10), 697–702. https://doi.org/ Ambient Measurements. Aerosol Science and Technology 42, 1033–1041. https:// 10.1038/ngeo313. https://www.nature.com/articles/ngeo313#supplementary- doi.org/10.1080/02786820802389277. information, 2008. Lack, D.A., Cappa, C.D., 2010. Impact of brown and clear carbon on light absorption Marlon, J.R., Bartlein, P.J., Walsh, M.K., Harrison, S.P., Brown, K.J., Edwards, M.E., enhancement, single scatter albedo and absorption wavelength dependence of black Higuera, P.E., Power, M.J., Anderson, R.S., Briles, C., Brunelle, A., Carcaillet, C., carbon. Atmos. Chem. Phys. 10, 4207–4220. https://doi.org/10.5194/acp-10-4207- Daniels, M., Hu, F.S., Lavoie, M., Long, C., Minckley, T., Richard, P.J.H., Scott, A.C., 2010. Shafer, D.S., Tinner, W., Umbanhowar, C.E., Whitlock, C., 2009. Wildfire responses to Lack, D.A., Langridge, J.M., Bahreini, R., Cappa, C.D., Middlebrook, A.M., Schwarz, J.P., abrupt climate change in North America. Proceedings of the National Academy of 2012. Brown carbon and internal mixing in biomass burning particles. Proceedings of Sciences 106, 2519–2524. https://doi.org/10.1073/pnas.0808212106. the National Academy of Sciences 109, 14802–14807. https://doi.org/10.1073/pnas. Martin, S. T., Artaxo, P., Machado, L., Manzi, A. O., Souza, R. A. F., Schumacher, C., 1206575109. Wang, J., Biscaro, T., Brito, J., Calheiros, A., Jardine, K., Medeiros, A., Portela, B., Sá, Laing, J.R., Jaffe, D.A., Hee, J.R., 2016. Physical and optical properties of aged biomass S. S. d., Adachi, K., Aiken, A. C., Albrecht, R., Alexander, L., Andreae, M. O., Barbosa, burning aerosol from wildfires in Siberia and the Western USA at the Mt. Bachelor H. M. J., Buseck, P., Chand, D., Comstock, J. M., Day, D. A., Dubey, M., Fan, J., Fast, Observatory. Atmos. Chem. Phys. 16, 15185–15197. https://doi.org/10.5194/acp- J., Fisch, G., Fortner, E., Giangrande, S., Gilles, M., Goldstein, A. H., Guenther, A., 16-15185-2016. Hubbe, J., Jensen, M., Jimenez, J. L., Keutsch, F. N., Kim, S., Kuang, C., Laskin, A., Laing, J.R., Jaffe, D.A., Sedlacek, I.I.I.A.J., 2019. Comparison of Filter-based Absorption McKinney, K., Mei, F., Miller, M., Nascimento, R., Pauliquevis, T., Pekour, M., Peres, Measurements of Biomass Burning Aerosol and Background Aerosol at the Mt. J., Petäjä, T., Pöhlker, C., Pöschl, U., Rizzo, L., Schmid, B., Shilling, J. E., Dias, M. A. Bachelor Observatory, Aerosol Air Qual Res. https://doi.org/10.4209/aaqr.2019.06. S., Smith, J. N., Tomlinson, J. M., Tóta, J., and Wendisch, M.: The Green Ocean 0298. Amazon Experiment (GoAmazon2014/5) Observes Pollution Affecting Gases, Lance, S., Raatikainen, T., Onasch, T.B., Worsnop, D.R., Yu, X.Y., Alexander, M.L., Aerosols, Clouds, and Rainfall over the Rain Forest, Bulletin of the American Stolzenburg, M.R., McMurry, P.H., Smith, J.N., Nenes, A., 2013. Aerosol mixing state, Meteorological Society, 98, 981-997, 10.1175/bams-d-15-00221.1, 2017. hygroscopic growth and cloud activation efficiency during MIRAGE 2006. Atmos. Massoli, P., Lambe, A.T., Ahern, A.T., Williams, L.R., Ehn, M., Mikkilä, J., Canagaratna, Chem. Phys. 13, 5049–5062. https://doi.org/10.5194/acp-13-5049-2013. M.R., Brune, W.H., Onasch, T.B., Jayne, J.T., Petäjä, T., Kulmala, M., Laaksonen, A., Lang, T., 2015. ORACLES (ObseRvations of Aerosols above CLouds and their Kolb, C.E., Davidovits, P., Worsnop, D.R., 2010. Relationship between aerosol oxi- intEractionS) Instrumentation and. Models. dation level and hygroscopic properties of laboratory generated secondary organic Lathem, T.L., Beyersdorf, A.J., Thornhill, K.L., Winstead, E.L., Cubison, M.J., Hecobian, aerosol (SOA) particles. Geophysical Research Letters 37. https://doi.org/10.1029/ A., Jimenez, J.L., Weber, R.J., Anderson, B.E., Nenes, A., 2013. Analysis of CCN ac- 2010gl045258. tivity of Arctic aerosol and Canadian biomass burning during summer 2008. Atmos. Mather, J.H., Voyles, J.W., 2013. The Arm Climate Research Facility: A Review of Chem. Phys. 13, 2735–2756. https://doi.org/10.5194/acp-13-2735-2013. Structure and Capabilities. Bulletin of the American Meteorological Society 94, Lelieveld, J., Bourtsoukidis, E., Brühl, C., Fischer, H., Fuchs, H., Harder, H., 377–392. https://doi.org/10.1175/bams-d-11-00218.1. Hofzumahaus, A., Holland, F., Marno, D., Neumaier, M., Pozzer, A., Schlager, H., Mei, F., Hayes, P.L., Ortega, A., Taylor, J.W., Allan, J.D., Gilman, J., Kuster, W., de Gouw, Williams, J., Zahn, A., Ziereis, H., 2018. The South Asian monsoon—Pollution pump J., Jimenez, J.L., Wang, J., 2013a. Droplet activation properties of organic aerosols and purifier. Science. https://doi.org/10.1126/science.aar2501. observed at an urban site during CalNex-LA. Journal of Geophysical Research: Levin, E.J.T., McMeeking, G.R., Carrico, C.M., Mack, L.E., Kreidenweis, S.M., Wold, C.E., Atmospheres 118, 2903–2917. https://doi.org/10.1002/jgrd.50285. Moosmüller, H., Arnott, W.P., Hao, W.M., Collett Jr., J.L., Malm, W.C., 2010. Biomass Mei, Fan, Hayes, Patrick L., Ortega, Amber, Taylor, Jonathan W., Allan, James D., burning smoke aerosol properties measured during Fire Laboratory at Missoula Gilman, Jessica, Kuster, William, de Gouw, Joost, Jimenez, Jose L., Wang, Jian, Experiments (FLAME). Journal of Geophysical Research: Atmospheres 115. https:// 2013b. Droplet activation properties of organic aerosols observed at an urban site doi.org/10.1029/2009JD013601. during CalNex-LA: CCN ACTIVITY OF ORGANIC AEROSOLS. J. Geophys. Res. Atmos. Li, J., Liu, C., Yin, Y., Kumar, K.R., 2016. Numerical investigation on the Ångström ex- 118 (7), 2903–2917. https://doi.org/10.1002/jgrd.50285. ponent of black carbon aerosol. Journal of Geophysical Research: Atmospheres 121, Moore, R.H., Nenes, A., Medina, J., 2010. Scanning Mobility CCN Analysis—A Method for 3506–3518. https://doi.org/10.1002/2015jd024718. Fast Measurements of Size-Resolved CCN Distributions and Activation Kinetics. Liang, Q., Douglass, A.R., Duncan, B.N., Stolarski, R.S., Witte, J.C., 2009. The governing Aerosol Science and Technology 44, 861–871. https://doi.org/10.1080/02786826. processes and timescales of stratosphere-to-troposphere transport and its contribution 2010.498715. to ozone in the Arctic troposphere. Atmos. Chem. Phys. 9, 3011–3025. https://doi. Naud, C.M., Booth, J.F., Lamraoui, F., 2018. Post Cold Frontal Clouds at the ARM Eastern org/10.5194/acp-9-3011-2009. North Atlantic Site: An Examination of the Relationship Between Large-Scale Lin, P., Liu, J., Shilling, J.E., Kathmann, S.M., Laskin, J., Laskin, A., 2015. Molecular Environment and Low-Level. Cloud Properties, Journal of Geophysical Research:

14 G. Zheng, et al. Environment International 139 (2020) 105680

Atmospheres 123, 12,117–112,132. https://doi.org/10.1029/2018jd029015. dependent water affinity of biogenic organic aerosol and its importance for climate. Ortega, A.M., Day, D.A., Cubison, M.J., Brune, W.H., Bon, D., de Gouw, J.A., Jimenez, Geophysical research letters 44, 5167–5177. https://doi.org/10.1002/ J.L., 2013. Secondary organic aerosol formation and primary organic aerosol oxi- 2017GL073056. dation from biomass-burning smoke in a flow reactor during FLAME-3. Atmos. Chem. Raveh-Rubin, S., 2017. Dry Intrusions: Lagrangian Climatology and Dynamical Impact on Phys. 13, 11551–11571. https://doi.org/10.5194/acp-13-11551-2013. the Planetary Boundary Layer. Journal of Climate 30, 6661–6682. https://doi.org/ Ovadnevaite, J., Ceburnis, D., Martucci, G., Bialek, J., Monahan, C., Rinaldi, M., Facchini, 10.1175/jcli-d-16-0782.1. M. C., Berresheim, H., Worsnop, D. R., and O'Dowd, C.: Primary marine organic Raveh-Rubin, S., Catto, J.L., 2019. Climatology and dynamics of the link between dry aerosol: A dichotomy of low hygroscopicity and high CCN activity, Geophysical intrusions and cold fronts during winter, Part II: Front-centred perspective. Clim Research Letters, 38, n/a-n/a, 10.1029/2011GL048869, 2011. Dynam. https://doi.org/10.1007/s00382-019-04793-2. Painemal, D., Kato, S., and Minnis, P.: Boundary layer regulation in the southeast Atlantic Reid, J.S., Eck, T.F., Christopher, S.A., Koppmann, R., Dubovik, O., Eleuterio, D.P., cloud microphysics during the biomass burning season as seen by the A-train satellite Holben, B.N., Reid, E.A., Zhang, J., 2005a. A review of biomass burning emissions constellation, Journal of Geophysical Research: Atmospheres, 119, 11,288-211,302, part III: intensive optical properties of biomass burning particles. Atmos. Chem. Phys. 10.1002/2014jd022182, 2014. 5, 827–849. https://doi.org/10.5194/acp-5-827-2005. Pajunoja, A., Lambe, A.T., Hakala, J., Rastak, N., Cummings, M.J., Brogan, J.F., Hao, L., Reid, J.S., Koppmann, R., Eck, T.F., Eleuterio, D.P., 2005b. A review of biomass burning Paramonov, M., Hong, J., Prisle, N.L., Malila, J., Romakkaniemi, S., Lehtinen, K.E.J., emissions part II: intensive physical properties of biomass burning particles. Atmos. Laaksonen, A., Kulmala, M., Massoli, P., Onasch, T.B., Donahue, N.M., Riipinen, I., Chem. Phys. 5, 799–825. https://doi.org/10.5194/acp-5-799-2005. Davidovits, P., Worsnop, D.R., Petäjä, T., Virtanen, A., 2015. Adsorptive uptake of Rickards, A.M.J., Miles, R.E.H., Davies, J.F., Marshall, F.H., Reid, J.P., 2013. water by semisolid secondary organic aerosols. Geophysical Research Letters 42, Measurements of the Sensitivity of Aerosol Hygroscopicity and the κ Parameter to the 3063–3068. https://doi.org/10.1002/2015gl063142. O/C Ratio. The Journal of Physical Chemistry A 117, 14120–14131. https://doi.org/ Palm, B.B., de Sá, S.S., Day, D.A., Campuzano-Jost, P., Hu, W., Seco, R., Sjostedt, S.J., 10.1021/jp407991n. Park, J.H., Guenther, A.B., Kim, S., Brito, J., Wurm, F., Artaxo, P., Thalman, R., Rogers, B.M., Randerson, J.T., Bonan, G.B., 2013. High-latitude cooling associated with Wang, J., Yee, L.D., Wernis, R., Isaacman-VanWertz, G., Goldstein, A.H., Liu, Y., landscape changes from North American boreal forest fires. Biogeosciences 10, Springston, S.R., Souza, R., Newburn, M.K., Alexander, M.L., Martin, S.T., Jimenez, 699–718. https://doi.org/10.5194/bg-10-699-2013. J.L., 2018. Secondary organic aerosol formation from ambient air in an oxidation Rogers, Brendan M., Soja, Amber J., Goulden, Michael L., Randerson, James T., 2015. flow reactor in central Amazonia. Atmos. Chem. Phys. 18, 467–493. https://doi.org/ Influence of tree species on continental differences in boreal fires and climate feed- 10.5194/acp-18-467-2018. backs. Nature Geosci 8 (3), 228–234. https://doi.org/10.1038/ngeo2352. https:// Peterson, D.A., Campbell, J.R., Hyer, E.J., Fromm, M.D., Kablick, G.P., Cossuth, J.H., www.nature.com/articles/ngeo2352#supplementary-information, 2015. DeLand, M.T., 2018. Wildfire-driven thunderstorms cause a volcano-like strato- Rose, D., Gunthe, S.S., Mikhailov, E., Frank, G.P., Dusek, U., Andreae, M.O., Pöschl, U., spheric injection of smoke, npj Climate and Atmospheric. Science 1, 30. https://doi. 2008. Calibration and measurement uncertainties of a continuous-flow cloud con- org/10.1038/s41612-018-0039-3. densation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and Petters, M.D., Kreidenweis, S.M., 2007. A single parameter representation of hygroscopic sodium chloride aerosol particles in theory and experiment. Atmos. Chem. Phys. 8, growth and cloud condensation nucleus activity. Atmos. Chem. Phys. 7, 1961–1971. 1153–1179. https://doi.org/10.5194/acp-8-1153-2008. https://doi.org/10.5194/acp-7-1961-2007. Rose, D., Gunthe, S.S., Su, H., Garland, R.M., Yang, H., Berghof, M., Cheng, Y.F., Wehner, Petters, M.D., Prenni, A.J., Kreidenweis, S.M., DeMott, P.J., 2007. On Measuring the B., Achtert, P., Nowak, A., Wiedensohler, A., Takegawa, N., Kondo, Y., Hu, M., Zhang, Critical Diameter of Cloud Condensation Nuclei Using Mobility Selected Aerosol. Y., Andreae, M.O., Pöschl, U., 2011. Cloud condensation nuclei in polluted air and Aerosol Science and Technology 41, 907–913. https://doi.org/10.1080/ biomass burning smoke near the mega-city Guangzhou, China – Part 2: Size-resolved 02786820701557214. aerosol chemical composition, diurnal cycles, and externally mixed weakly CCN-ac- Petters, M. D., Carrico, C. M., Kreidenweis, S. M., Prenni, A. J., DeMott, P. J., Collett, J. L., tive soot particles. Atmos. Chem. Phys. 11, 2817–2836. https://doi.org/10.5194/acp- and Moosmüller, H.: Cloud condensation nucleation activity of biomass burning 11-2817-2011. aerosol, Journal of Geophysical Research: Atmospheres, 114, n/a-n/a, 10.1029/ Sakaeda, N., Wood, R., Rasch, P.J., 2011. Direct and semidirect aerosol effects of southern 2009JD012353, 2009a. African biomass burning aerosol. Journal of Geophysical Research: Atmospheres 116. Petters, M.D., Wex, H., Carrico, C.M., Hallbauer, E., Massling, A., McMeeking, G.R., https://doi.org/10.1029/2010jd015540. Poulain, L., Wu, Z., Kreidenweis, S.M., Stratmann, F., 2009b. Towards closing the gap Sakamoto, K.M., Laing, J.R., Stevens, R.G., Jaffe, D.A., Pierce, J.R., 2016. The evolution of between hygroscopic growth and activation for secondary organic aerosol – Part 2: biomass-burning aerosol size distributions due to coagulation: dependence on fire Theoretical approaches. Atmos. Chem. Phys. 9, 3999–4009. https://doi.org/10. and meteorological details and parameterization. Atmos. Chem. Phys. 16, 5194/acp-9-3999-2009. 7709–7724. https://doi.org/10.5194/acp-16-7709-2016. Petzold, A., Weinzierl, B., Huntrieser, H., Stohl, A., Real, E., Cozic, J., Fiebig, M., Saleh, R., Hennigan, C.J., McMeeking, G.R., Chuang, W.K., Robinson, E.S., Coe, H., Hendricks, J., Lauer, A., Law, K., Roiger, A., Schlager, H., Weingartner, E., 2007. Donahue, N.M., Robinson, A.L., 2013. Absorptivity of brown carbon in fresh and Perturbation of the European free troposphere aerosol by North American forest fire photo-chemically aged biomass-burning emissions. Atmos. Chem. Phys. 13, plumes during the ICARTT-ITOP experiment in summer 2004. Atmos. Chem. Phys. 7, 7683–7693. https://doi.org/10.5194/acp-13-7683-2013. 5105–5127. https://doi.org/10.5194/acp-7-5105-2007. Saleh, Rawad, Robinson, Ellis S., Tkacik, Daniel S., Ahern, Adam T., Liu, Shang, Aiken, Pollet, J., Omi, P.N., 2002. Effect of thinning and prescribed burning on crown fire se- Allison C., Sullivan, Ryan C., Presto, Albert A., Dubey, Manvendra K., Yokelson, verity in ponderosa pine forests. International Journal of Wildland Fire 11, 1–10. Robert J., Donahue, Neil M., Robinson, Allen L., 2014. Brownness of organics in https://doi.org/10.1071/WF01045. aerosols from biomass burning linked to their black carbon content. Nature Geosci 7 Quinn, P. K., Coffman, D. J., Johnson, J. E., Upchurch, L. M., and Bates, T. S.: Small (9), 647–650. https://doi.org/10.1038/ngeo2220. https://www.nature.com/ fraction of marine cloud condensation nuclei made up of sea spray aerosol, Nature articles/ngeo2220#supplementary-information, 2014. Geosci, advance online publication, 10.1038/ngeo3003 http://www.nature.com/ Saleh, R., Marks, M., Heo, J., Adams, P. J., Donahue, N. M., and Robinson, A. L.: ngeo/journal/vaop/ncurrent/abs/ngeo3003.html#supplementary-information, Contribution of brown carbon and lensing to the direct radiative effect of carbo- 2017. naceous aerosols from biomass and biofuel burning emissions, Journal of Geophysical Rajapakshe, C., Zhang, Z., Yorks, J.E., Yu, H., Tan, Q., Meyer, K., Platnick, S., Winker, Research: Atmospheres, 120, 10,285-210,296, 10.1002/2015jd023697, 2015. D.M., 2017. Seasonally transported aerosol layers over southeast Atlantic are closer Samset, B.H., Myhre, G., 2011. Vertical dependence of black carbon, sulphate and bio- to underlying clouds than previously reported. Geophysical Research Letters 44, mass burning aerosol radiative forcing. Geophysical Research Letters 38. https://doi. 5818–5825. https://doi.org/10.1002/2017gl073559. org/10.1029/2011gl049697. Ramnarine, E., Kodros, J.K., Hodshire, A.L., Lonsdale, C.R., Alvarado, M.J., Pierce, J.R., Samset, B.H., Myhre, G., Schulz, M., Balkanski, Y., Bauer, S., Berntsen, T.K., Bian, H., 2018. Effects of Near-Source Coagulation of Biomass Burning Aerosols on Global Bellouin, N., Diehl, T., Easter, R.C., Ghan, S.J., Iversen, T., Kinne, S., Kirkevåg, A., Predictions of Aerosol Size Distributions and Implications for Aerosol Radiative Lamarque, J.F., Lin, G., Liu, X., Penner, J.E., Seland, Ø., Skeie, R.B., Stier, P., Effects. Atmos. Chem. Phys. Discuss. 1–28, 2018. https://doi.org/10.5194/acp-2018- Takemura, T., Tsigaridis, K., Zhang, K., 2013. Black carbon vertical profiles strongly 1084. affect its radiative forcing uncertainty. Atmos. Chem. Phys. 13, 2423–2434. https:// Randerson, J.T., Liu, H., Flanner, M.G., Chambers, S.D., Jin, Y., Hess, P.G., Pfister, G., doi.org/10.5194/acp-13-2423-2013. Mack, M.C., Treseder, K.K., Welp, L.R., Chapin, F.S., Harden, J.W., Goulden, M.L., Schmale, J., Henning, S., Decesari, S., Henzing, B., Keskinen, H., Sellegri, K., Ovadnevaite, Lyons, E., Neff, J.C., Schuur, E.A.G., Zender, C.S., 2006. The Impact of Boreal Forest J., Pöhlker, M.L., Brito, J., Bougiatioti, A., Kristensson, A., Kalivitis, N., Stavroulas, I., Fire on Climate Warming. Science 314, 1130–1132. https://doi.org/10.1126/ Carbone, S., Jefferson, A., Park, M., Schlag, P., Iwamoto, Y., Aalto, P., Äijälä, M., science.1132075. Bukowiecki, N., Ehn, M., Frank, G., Fröhlich, R., Frumau, A., Herrmann, E., Rasch, P.J., Xie, S., Ma, P.-L., Lin, W., Wang, H., Tang, Q., Burrows, S.M., Caldwell, P., Herrmann, H., Holzinger, R., Kos, G., Kulmala, M., Mihalopoulos, N., Nenes, A., Zhang, K., Easter, R.C., Cameron-Smith, P., Singh, B., Wan, H., Golaz, J.-C., Harrop, O'Dowd, C., Petäjä, T., Picard, D., Pöhlker, C., Pöschl, U., Poulain, L., Prévôt, A.S.H., B.E., Roesler, E., Bacmeister, J., Larson, V.E., Evans, K.J., Qian, Y., Taylor, M., Leung, Swietlicki, E., Andreae, M.O., Artaxo, P., Wiedensohler, A., Ogren, J., Matsuki, A., L.R., Zhang, Y., Brent, L., Branstetter, M., Hannay, C., Mahajan, S., Mametjanov, A., Yum, S.S., Stratmann, F., Baltensperger, U., Gysel, M., 2018. Long-term cloud con- Neale, R., Richter, J.H., Yoon, J.-H., Zender, C.S., Bader, D., Flanner, M., Foucar, J.G., densation nuclei number concentration, particle number size distribution and che- Jacob, R., Keen, N., Klein, S.A., Liu, X., Salinger, A.G., Shrivastava, M., Yang, Y., mical composition measurements at regionally representative observatories. Atmos. 2019. An Overview of the Atmospheric Component of the Energy Exascale Earth Chem. Phys. 18, 2853–2881. https://doi.org/10.5194/acp-18-2853-2018. System Model. J Adv Model Earth Sy. https://doi.org/10.1029/2019ms001629. Schnaiter, M., Linke, C., Möhler, O., Naumann, K.-H., Saathoff, H., Wagner, R., Schurath, Rastak, N., Pajunoja, A., Acosta Navarro, J.C., Ma, J., Song, M., Partridge, D.G., Kirkevåg, U., Wehner, B., 2005. Absorption amplification of black carbon internally mixed with A., Leong, Y., Hu, W.W., Taylor, N.F., Lambe, A., Cerully, K., Bougiatioti, A., Liu, P., secondary organic aerosol. Journal of Geophysical Research: Atmospheres 110. Krejci, R., Petäjä, T., Percival, C., Davidovits, P., Worsnop, D.R., Ekman, A.M.L., https://doi.org/10.1029/2005jd006046. Nenes, A., Martin, S., Jimenez, J.L., Collins, D.R., Topping, D.O., Bertram, A.K., Schnitzler, E.G., Abbatt, J.P.D., 2018. Heterogeneous OH oxidation of secondary brown Zuend, A., Virtanen, A., Riipinen, I., 2017. Microphysical explanation of the RH- carbon aerosol. Atmos. Chem. Phys. 18, 14539–14553. https://doi.org/10.5194/acp-

15 G. Zheng, et al. Environment International 139 (2020) 105680

18-14539-2018. Wang, X., Heald, C.L., Liu, J., Weber, R.J., Campuzano-Jost, P., Jimenez, J.L., Schwarz, Sedlacek III, A.J., Buseck, P.R., Adachi, K., Onasch, T.B., Springston, S.R., Kleinman, L., J.P., Perring, A.E., 2018. Exploring the observational constraints on the simulation of 2018. Formation and evolution of tar balls from northwestern US wildfires. Atmos. brown carbon. Atmos. Chem. Phys. 18, 635–653. https://doi.org/10.5194/acp-18- Chem. Phys. 18, 11289–11301. https://doi.org/10.5194/acp-18-11289-2018. 635-2018. Selimovic, V., Yokelson, R.J., Warneke, C., Roberts, J.M., de Gouw, J., Reardon, J., Wernli, H., 1997. A lagrangian-based analysis of extratropical cyclones. II: A detailed Griffith, D.W.T., 2018. Aerosol optical properties and trace gas emissions by PAX and case-study. Q J Roy Meteor Soc 123, 1677–1706. https://doi.org/10.1002/qj. OP-FTIR for laboratory-simulated western US wildfires during FIREX. Atmos. Chem. 49712354211. Phys. 18, 2929–2948. https://doi.org/10.5194/acp-18-2929-2018. Westerling, A.L., Hidalgo, H.G., Cayan, D.R., Swetnam, T.W., 2006. Warming and Earlier Shinozuka, Y., Clarke, A.D., Howell, S.G., Kapustin, V.N., McNaughton, C.S., Zhou, J., Spring Increase Western U.S. Forest Wildfire Activity. Science 313, 940–943. https:// Anderson, B.E., 2007. Aircraft profiles of aerosol microphysics and optical properties doi.org/10.1126/science.1128834. over North America: Aerosol optical depth and its association with PM2.5 and water Wex, H., Petters, M.D., Carrico, C.M., Hallbauer, E., Massling, A., McMeeking, G.R., uptake. Journal of Geophysical Research: Atmospheres 112. https://doi.org/10. Poulain, L., Wu, Z., Kreidenweis, S.M., Stratmann, F., 2009. Towards closing the gap 1029/2006jd007918. between hygroscopic growth and activation for secondary organic aerosol: Part 1 – Sprenger, M., Wernli, H., 2015. The LAGRANTO Lagrangian analysis tool – version 2.0. Evidence from measurements. Atmos. Chem. Phys. 9, 3987–3997. https://doi.org/ Geosci. Model Dev. 8, 2569–2586. https://doi.org/10.5194/gmd-8-2569-2015. 10.5194/acp-9-3987-2009. Stier, P., Feichter, J., Kinne, S., Kloster, S., Vignati, E., Wilson, J., Ganzeveld, L., Tegen, I., Wilcox, E.M., 2012. Direct and semi-direct radiative forcing of smoke aerosols over Werner, M., Balkanski, Y., Schulz, M., Boucher, O., Minikin, A., Petzold, A., 2005. The clouds. Atmos. Chem. Phys. 12, 139–149. https://doi.org/10.5194/acp-12-139-2012. aerosol-climate model ECHAM5-HAM. Atmos. Chem. Phys. 5, 1125–1156. https:// Williams, A.P., Abatzoglou, J.T., 2016. Recent Advances and Remaining Uncertainties in doi.org/10.5194/acp-5-1125-2005. Resolving Past and Future Climate Effects on Global Fire Activity. Current Climate Stocker, T., 2014. Climate change 2013: the physical science basis: Working Group I Change Reports 2, 1–14. https://doi.org/10.1007/s40641-016-0031-0. contribution to the Fifth assessment report of the Intergovernmental Panel on Climate Wood, R., 2012. Stratocumulus Clouds. Monthly Weather Review 140, 2373–2423. Change. Cambridge University Press. https://doi.org/10.1175/mwr-d-11-00121.1. Sumlin, B.J., Pandey, A., Walker, M.J., Pattison, R.S., Williams, B.J., Chakrabarty, R.K., Wood, R., Wyant, M., Bretherton, C. S., Rémillard, J., Kollias, P., Fletcher, J., Stemmler, 2017. Atmospheric Photooxidation Diminishes Light Absorption by Primary Brown J., Szoeke, S. d., Yuter, S., Miller, M., Mechem, D., Tselioudis, G., Chiu, J. C., Mann, J. Carbon Aerosol from Biomass Burning. Environmental Science & Technology Letters A. L., O’Connor, E. J., Hogan, R. J., Dong, X., Miller, M., Ghate, V., Jefferson, A., Min, 4, 540–545. https://doi.org/10.1021/acs.estlett.7b00393. Q., Minnis, P., Palikonda, R., Albrecht, B., Luke, E., Hannay, C., and Lin, Y.: Clouds, Taylor, J.W., Allan, J.D., Allen, G., Coe, H., Williams, P.I., Flynn, M.J., Le Breton, M., Aerosols, and Precipitation in the Marine Boundary Layer: An Arm Mobile Facility Muller, J.B.A., Percival, C.J., Oram, D., Forster, G., Lee, J.D., Rickard, A.R., Deployment, Bulletin of the American Meteorological Society, 96, 419-440, 10.1175/ Parrington, M., Palmer, P.I., 2014. Size-dependent wet removal of black carbon in bams-d-13-00180.1, 2015. Canadian biomass burning plumes. Atmos. Chem. Phys. 14, 13755–13771. https:// Wooster, M.J., Zhang, Y.H., 2004. Boreal forest fires burn less intensely in Russia than in doi.org/10.5194/acp-14-13755-2014. North America. Geophysical Research Letters 31. https://doi.org/10.1029/ Thalman, R., de Sá, S.S., Palm, B.B., Barbosa, H.M.J., Pöhlker, M.L., Alexander, M.L., 2004gl020805. Brito, J., Carbone, S., Castillo, P., Day, D.A., Kuang, C., Manzi, A., Ng, N.L., Sedlacek Yu, P., Toon, O.B., Bardeen, C.G., Zhu, Y., Rosenlof, K.H., Portmann, R.W., Thornberry, Iii, A.J., Souza, R., Springston, S., Watson, T., Pöhlker, C., Pöschl, U., Andreae, M.O., T.D., Gao, R.-S., Davis, S.M., Wolf, E.T., de Gouw, J., Peterson, D.A., Fromm, M.D., Artaxo, P., Jimenez, J.L., Martin, S.T., Wang, J., 2017. CCN activity and organic Robock, A., 2019. Black carbon lofts wildfire smoke high into the stratosphere to hygroscopicity of aerosols downwind of an urban region in central Amazonia: sea- form a persistent plume. Science 365, 587–590. https://doi.org/10.1126/science. sonal and diel variations and impact of anthropogenic emissions. Atmos. Chem. Phys. aax1748. 17, 11779–11801. https://doi.org/10.5194/acp-17-11779-2017. Zarzycki, C.M., Bond, T.C., 2010. How much can the vertical distribution of black carbon Turco, M., Llasat, M.-C., von Hardenberg, J., Provenzale, A., 2014. Climate change im- affect its global direct radiative forcing? Geophysical Research Letters 37. https:// pacts on wildfires in a Mediterranean environment. Climatic Change 125, 369–380. doi.org/10.1029/2010gl044555. https://doi.org/10.1007/s10584-014-1183-3. Zhang, A., Wang, Y., Zhang, Y., Weber, R.J., Song, Y., Ke, Z., Zou, Y., 2019. Modeling Twomey, S., 1974. Pollution and the planetary albedo. Atmospheric Environment 1967 global radiative effect of brown carbon: A larger heating source in the tropical free (8), 1251–1256. https://doi.org/10.1016/0004-6981(74)90004-3. troposphere than black carbon. Atmos. Chem. Phys. Discuss. 1–36, 2019. https://doi. Ueda, S., Nakayama, T., Taketani, F., Adachi, K., Matsuki, A., Iwamoto, Y., Sadanaga, Y., org/10.5194/acp-2019-594. Matsumi, Y., 2016. Light absorption and morphological properties of soot-containing Zhang, Z., Meyer, K., Yu, H., Platnick, S., Colarco, P., Liu, Z., Oreopoulos, L., 2016. aerosols observed at an East Asian outflow site, Noto Peninsula, Japan. Atmos. Chem. Shortwave direct radiative effects of above-cloud aerosols over global oceans derived Phys. 16, 2525–2541. https://doi.org/10.5194/acp-16-2525-2016. from 8 years of CALIOP and MODIS observations. Atmos. Chem. Phys. 16, Uin, J., 2016. 3002 Humidified Tandem Differential Mobility Analyzer Instrument 2877–2900. https://doi.org/10.5194/acp-16-2877-2016. Handbook, DOE Office of Science Atmospheric Radiation. Measurement (ARM) Zhao, R., Lee, A.K.Y., Huang, L., Li, X., Yang, F., Abbatt, J.P.D., 2015. Photochemical Program …. processing of aqueous atmospheric brown carbon. Atmos. Chem. Phys. 15, Vakkari, V., Kerminen, V.-M., Beukes, J.P., Tiitta, P., van Zyl, P.G., Josipovic, M., Venter, 6087–6100. https://doi.org/10.5194/acp-15-6087-2015. A.D., Jaars, K., Worsnop, D.R., Kulmala, M., Laakso, L., 2014. Rapid changes in Zheng, G., Wang, Y., Aiken, A.C., Gallo, F., Jensen, M.P., Kollias, P., Kuang, C., Luke, E., biomass burning aerosols by atmospheric oxidation. Geophysical Research Letters 41, Springston, S., Uin, J., Wood, R., Wang, J., 2018. Marine boundary layer aerosol in 2644–2651. https://doi.org/10.1002/2014GL059396. the eastern North Atlantic: seasonal variations and key controlling processes. Atmos. Vakkari, V., Beukes, J.P., Dal Maso, M., Aurela, M., Josipovic, M., van Zyl, P.G., 2018. Chem. Phys. 18, 17615–17635. https://doi.org/10.5194/acp-18-17615-2018. Major secondary aerosol formation in southern African open biomass burning Zhu, L., Val Martin, M., Gatti, L.V., Kahn, R., Hecobian, A., Fischer, E.V., 2018. plumes. Nature Geoscience. https://doi.org/10.1038/s41561-018-0170-0. Development and implementation of a new biomass burning emissions injection van der Werf, G.R., Randerson, J.T., Giglio, L., Collatz, G.J., Mu, M., Kasibhatla, P.S., height scheme (BBEIH v1.0) for the GEOS-Chem model (v9–01-01). Geosci. Model Morton, D.C., DeFries, R.S., Jin, Y., van Leeuwen, T.T., 2010. Global fire emissions Dev. 11, 4103–4116. https://doi.org/10.5194/gmd-11-4103-2018. and the contribution of deforestation, savanna, forest, agricultural, and peat fires Zuidema, P., Redemann, J., Haywood, J., Wood, R., Piketh, S., Hipondoka, M., Formenti, (1997–2009). Atmos. Chem. Phys. 10, 11707–11735. https://doi.org/10.5194/acp- P., 2016. Smoke and Clouds above the Southeast Atlantic: Upcoming Field Campaigns 10-11707-2010. Probe Absorbing Aerosol’s Impact on Climate. Bulletin of the American Wang, J., Dong, X., Wood, R., 2016. Aerosol and Cloud Experiments in Eastern North. Meteorological Society 97, 1131–1135. https://doi.org/10.1175/bams-d-15- Atlantic (ACE-ENA) Science Plan. 00082.1. Wang, J., Shilling, J.E., Liu, J., Zelenyuk, A., Bell, D.M., Petters, M.D., Thalman, R., Mei, Zuidema, P., Alvarado, M., Chiu, C., de Szoeke, S., Fairall, C., Feingold, G., Freedman, A., F., Zaveri, R.A., Zheng, G., 2019. Cloud droplet activation of secondary organic Ghan, S., Haywood, J., 2018. and Kollias, P.: Layered Atlantic Smoke Interactions aerosol is mainly controlled by molecular weight, not water solubility. Atmos. Chem. with Clouds (LASIC). Field Campaign Report, DOE Office of Science Atmospheric Phys. 19, 941–954. https://doi.org/10.5194/acp-19-941-2019. Radiation Measurement (ARM) Program ….