Analysis of aerosol composition data for western United States wildfires between 2005 and 2015: Dust emissions, chloride depletion, and most enhanced aerosol constituents.

PUBLICATIONS Journal of Geophysical Research: Atmospheres RESEARCH ARTICLE 10.1002/2017JD026547 Special Section: Quantifying the emission, properties, and diverse impacts of wildfire smoke Key Points: • Dust (i.e., fine soil and coarse mass) levels are enhanced in wildfire plumes across the western United States • Chloride levels drop significantly on peak fire influence days for nearly a half of the wildfires examined • Cl, P, K, nitrate, and Zn exhibit the highest percent change in concentration on peak fire days after carbonaceous constituents of PM2.5

Supporting Information: • Supporting Information S1 Correspondence to: A. Sorooshian, [email protected]

Citation: Schlosser, J. S., R. A. Braun, T. Bradley, H. Dadashazar, A. B. MacDonald, A. A. Aldhaif, M. A. Aghdam, A. H. Mardi, P. Xian, and A. Sorooshian (2017), Analysis of aerosol composition data for western United States wildfires between 2005 and 2015: Dust emissions, chloride depletion, and most enhanced aerosol constituents, J. Geophys. Res. Atmos., 122, 8951–8966, doi:10.1002/2017JD026547. Received 23 JAN 2017 Accepted 25 JUL 2017 Accepted article online 11 AUG 2017 Published online 29 AUG 2017

©2017. American Geophysical Union. All Rights Reserved.

SCHLOSSER ET AL.

Analysis of aerosol composition data for western United States wildfires between 2005 and 2015: Dust emissions, chloride depletion, and most enhanced aerosol constituents Joseph S. Schlosser1 , Rachel A. Braun1 , Trevor Bradley1 , Hossein Dadashazar1 , Alexander B. MacDonald1 , Abdulmonam A. Aldhaif1 , Mojtaba Azadi Aghdam1 , Ali Hossein Mardi1, Peng Xian2 , and Armin Sorooshian1,3 1

Department of Chemical and Environmental Engineering, University of Arizona, Tucson, Arizona, USA, 2United States Naval Research Laboratory, Monterey, California, USA, 3Department of Hydrology and Atmospheric Sciences, University of Arizona, Tucson, Arizona, USA

Abstract This study examines major wildfires in the western United States between 2005 and 2015 to determine which species exhibit the highest percent change in mass concentration on day of peak fire influence relative to preceding nonfire days. Forty-one fires were examined using the Environmental Protection Agency (EPA) Interagency Monitoring of Protected Visual Environments (IMPROVE) data set. Organic carbon (OC) and elemental carbon (EC) constituents exhibited the highest percent change increase. The sharpest enhancements were for the volatile (OC1) and semivolatile (OC2) OC fractions, suggestive of secondary organic aerosol formation during plume transport. Of the noncarbonaceous constituents, Cl, P, K, NO3, and Zn levels exhibited the highest percent change. Dust was significantly enhanced in wildfire plumes, based on significant enhancements in fine soil components (i.e., Si, Ca, Al, Fe, and Ti) and PMcoarse (i.e., PM10–PM2.5). A case study emphasized how transport of wildfire plumes significantly impacted downwind states, with higher levels of fine soil and PMcoarse at the downwind state (Arizona) as compared to the source of the fires (California). A global model (Navy Aerosol Analysis and Prediction System, NAAPS) did not capture the dust influence over California or Arizona during this case event because it is not designed to resolve dust dynamics in fires, which motivates improved treatment of such processes. Significant chloride depletion was observed on the peak EC day for almost a half of the fires examined. Size-resolved measurements during two specific fires at a coastal California site revealed significant chloride reductions for particle aerodynamic diameters between 1 and 10 μm. 1. Introduction The western United States is becoming increasingly vulnerable to the effects of wildfires owing to a warmer climate, drought, and fire control strategies over past decades resulting in conditions that promote larger and more frequent fires [Flannigan et al., 2000; Moritz et al., 2012; Dennison et al., 2014; Hallar et al., 2017]. Biomass burning leads to emissions of gaseous and particulate species that are complex in nature and evolve in unknown ways as they age and mix during their transport. Although wildfire plumes significantly impact regions near the fire source, they can reach high altitudes above the mixing layer and can be transported long distances [e.g., Pickering et al., 1996; Nisantzi et al., 2014] to impact global atmospheric chemistry and climate [e.g., Crutzen and Andreae, 1990; Reid et al., 2005]. Plume particles can act as both cloud condensation nuclei and ice nuclei, consequently impacting cloud properties and their precipitation potential. Wildfire emissions impact precipitation chemistry at downwind sites, in addition to affecting the radiative properties, oxidative capacity, new particle formation potential of plumes, and biogeochemical cycles [Echalar et al., 1995; Chalbot et al., 2013; Sorooshian et al., 2013]. Smoke plumes contain many nutrients (e.g., Fe) and contaminants (e.g., As) that can affect downstream ecosystems where they deposit. Chemical measurements are important both for linking air masses to fires by using tracer species and for quantifying enhancements of various constituents as a result of fires. Identification of robust tracer species is challenging as they could depend on flame condition and fuel type [Andreae and Merlet, 2001]. For example, Echalar et al. [1995] highlighted how different sets of species are enhanced for different fuel types such as savanna fires (K, Cl, Br, and Zn) and forest fires (Cr, Si, and Ca). Another example is that K, a commonly used WESTERN U.S. WILDFIRE COMPOSITION

8951

Journal of Geophysical Research: Atmospheres

10.1002/2017JD026547

tracer, is emitted more in flaming conditions as compared to smoldering fires [Cahill et al., 2008; Lee et al., 2010]. Additionally, some species such as K may be enhanced in smoke in submicrometer sizes while also having other important sources such as dust leading to elevated concentrations in coarser sizes [e.g., Eldred et al., 1997]. Levoglucosan is well known to be a marker for wood combustion as it is derived from the breakdown of cellulose and hemicellulose molecules, [Fine et al., 2001; Hays et al., 2002], but it is vulnerable to degradation after some time [Fraser and Lakshmanan, 2000; Hennigan et al., 2010; Hoffmann et al., 2010]. Unlike primarily emitted tracer species such as levoglucosan, secondarily produced species from precursors emitted by fires are more difficult to use as tracer species for fires as they require additional time for formation, have multiple sources, and may undergo additional transformations in order to be depleted. However, this does not undermine the importance of studying them since they contribute greatly to aerosol mass concentrations, including secondary organic aerosol [Spracklen et al., 2007]. As dust is one of the largest contributors to particulate matter mass concentrations in the western United States, especially the Southwest, it is important to consider how biomass burning contributes to dust concentrations in the atmosphere. Measurements at a high-altitude site in northwest Colorado (Storm Peak Laboratory) showed that the median levels of contribution from dust and biomass burning aerosols to total aerosol optical depth were comparable [Hallar et al., 2015]. Fine dust can contribute to over half of PM2.5 mass concentrations in the southwestern United States during spring months [e.g., Kavouras et al., 2007, 2009; Sorooshian et al., 2011], with the onset of the dust season being sooner in the spring based on analysis of data between 1995 and 2014 [Hand et al., 2016]. Soil dust is often associated with biomass burning plumes as a result of turbulent mixing near flames and the burn front [Palmer, 1981; Gaudichet et al., 1995; Maenhaut et al., 1996; Clements et al., 2008; Kavouras et al., 2012; Chalbot et al., 2013; Popovicheva et al., 2014; Sturtz et al., 2014; Maudlin et al., 2015]. Soil dust accounted for 75–99% of the coarse particle mass in the plumes of flaming African savanna fires [Maenhaut et al., 1996]. In another study focused on fires by the CaliforniaOregon border, Maudlin et al. [2015] observed a 513% and 408% enhancement in mass concentration for Si and Fe, respectively, in submicrometer aerosol during fire periods as compared to nonfire periods in the summer of 2013. They also observed these soil dust tracer species in stratocumulus cloud water. The goal of this work is to identify what particulate species are most enhanced in concentration in biomass burning plumes based on analysis of long-term data across the western United States from the Environmental Protection Agency (EPA) Interagency Monitoring of Protected Visual Environments (IMPROVE) network. This work builds on previous studies relying on IMPROVE data to investigate other aspects of wildfires in the same study region such as how much organic carbon (OC) and PM2.5 are produced specifically from fires in the summer season [Jaffe et al., 2008]. Particular focus is placed in this work on the extent to which fine soil and coarse mass concentrations (PMcoarse: PM10–PM2.5) are enhanced in biomass burning plumes. Unique aspects of the subsequent analysis are the broad spatial and temporal coverage of data in a region of North America vulnerable to biomass burning, the large number of species considered, treatment of different fractions of OC and elemental carbon (EC) based on thermal/optical carbon analysis, and the use of detailed size-resolved chemical data during two particular case study events to support conclusions from the IMPROVE data set. While the absolute concentration of aerosol constituents during wildfires is useful to discuss as well, this work focuses mainly on changes relative to background conditions, which has implications for both tracer species identification and for impacts on aerosol properties senstive to ratios of different aerosol constituents (e.g., hygroscopic and radiative properties). For example, if the volume fraction of low-hygroscopicity aerosol constituents is initially negligible prior to a wildfire, for the same concentration increase between low- and highhygroscopicity constituents during the fire, the overall hygroscopicity can be reduced assuming linear additive water uptake by different species. These aerosol responses to fires have implications for visibility, climate, and public health.

2. Methods 2.1. IMPROVE Aerosol Data This study relies on aerosol composition data from the IMPROVE network [Malm et al., 1994; http://views.cira. colostate.edu/fed/], which includes monitoring stations primarily in National Parks and Wilderness Areas.

SCHLOSSER ET AL.

WESTERN U.S. WILDFIRE COMPOSITION

8952


Figure 1. Spatial map of IMPROVE sites influenced by wildfires examined in this study.

10.1002/2017JD026547

Figure 1 shows locations of the IMPROVE stations used in the analysis, and Table 1 summarizes the site details. Particulate matter is collected on filters for 24 h every third day. Samples are analyzed for ions, metals, OC and EC, and gravimetric mass measurements of both PM2.5 and PM10. Four fractions of OC and three fractions of EC are also discussed here, which are operationally defined based on the temperatures at which they are detected using the thermal optical reflectance method of carbon analysis [e.g., Chow et al., 1993; Watson et al., 1994]. More specifically, a flame ionization detector quantifies methane produced via either (i) volatilization of particulate species at temperatures of 120°C (OC1; volatile), 250°C (OC2; semivolatile), 450°C (OC3; nonvolatile), or 550°C (OC4; nonvolatile) in pure helium or (ii) combustion at temperatures of 550°C (EC1), 700°C (EC2), or 800°C (EC3) in a 98% helium and 2% oxygen environment.

Speciated analysis is conducted for PM2.5, including ion chromatography (IC) for water-soluble ions and X-ray fluorescence (XRF) for elements. Fine soil concentrations reported in this study are calculated using the following equation [Malm et al., 1994]: Fine soil μg m3 ¼ 2:2½Al þ 2:49½Si þ 1:63½Ca þ 2:42½Fe þ 1:94½Ti (1)

The components and their coefficients in equation (1) were previously confirmed in comparisons of local resuspended soils and ambient particles in the study region [Cahill et al., 1980; Pitchford et al., 1981; Malm et al., 1994]. As this study is concerned with relative changes in fine soil levels, it is expected that this equation can successfully capture all soil-rich air masses regardless of whether minor variations exist in the factors used in equation (1). Species mass concentrations discussed in this study are from the fine fraction of aerosols (PM2.5). All species concentrations analyzed fit in the category of the “V0” status flag of the IMPROVE data set, which is defined as “Valid value.” Sampling protocols and additional details are provided elsewhere (http://vista.cira.colostate.edu/Improve/sops/). 2.2. Wildfire Identification Multiple sources of information are used to identify the presence of a wildfire. The first source of information is EC mass concentration from the IMPROVE monitoring sites. As EC concentration typically increases suddenly to a level that is far larger than its preceding and subsequent data points (Figure 2), it is possible in this way to identify the day with the strongest plume influence. As the IMPROVE measurements are every third day, it is cautioned that the peak EC day in this study for each fire may not coincide with the actual day of most smoke influence at a particular site. The EC mass concentration on the peak EC day always exceeded the average plus 2 times the standard deviation for the 10 week period centered around this day. As a secondary confirmation of fire activity, output is used from the Navy Aerosol Analysis and Prediction System (NAAPS) [Lynch et al., 2016; http://www.nrlmry.navy.mil/aerosol_web/]. NAAPS relies on global meteorological fields from the Navy Operational Global Atmospheric Prediction System (NOGAPS) [Hogan and Rosmond, 1991; Hogan and Brody, 1993] analyses and provides output at a spatial resolution of 1° × 1°,

SCHLOSSER ET AL.


8953


10.1002/2017JD026547

Table 1. Details of the 41 Wildfires Studied Between 2005 and 2015 in the Western United States Identifier A1 A2 C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 N1 N2 N3 O1 O2 O3 O4 O5 O6 O7 O8 O9 O10 O11 O12 U1 U2 U3 W1 W2 W3 W4 W5 W6 W7

State

IMPROVE Site

Site Code

Latitude (deg)

Longitude (deg)

Altitude (m)

Peak EC Date

AZ AZ CA CA CA CA CA CA CA CA CA CA CA CA CA CA NV NV NV OR OR OR OR OR OR OR OR OR OR OR OR UT UT UT WA WA WA WA WA WA WA

Mount Baldy Hence Camp-Grand Canyon NP Agua Tibia Bliss SP (TRPA) Hoover Hoover Hoover Joshua Tree NP Lava Beds NM Lava Beds NM Lava Beds NM Lava Beds NM San Gabriel San Gorgonio Wilderness Yosemite NP Yosemite NP Jarbridge Wilderness Jarbridge Wilderness Jarbridge Wilderness Crater Lake NP Crater Lake NP Hells Canyon Hells Canyon Hells Canyon Hells Canyon Kalmiopsis Mount Hood Mount Hood Starkey Starkey Three Sisters Wilderness Bryce Canyon NP Canyonlands NP Canyonlands NP Columbia Gorge #2 Columbia River Gorge North Cascades Pasayten Snoqualmie Pass White Pass White Pass

BALD GRCA AGTI BLIS HOOV HOOV HOOV JOSH LABE LABE LABE LABE SAGA SAGO YOSE YOSE JARB JARB JARB CRLA CRLA HECA HECA HECA HECA KALM MOHO MOHO STAR STAR THSI BRCA CANY CANY COGO CORI NOCA PASA SNPA WHPA WHPA

34.058 36.078 33.464 38.976 38.088 38.088 38.088 34.070 41.712 41.712 41.712 41.712 34.297 34.194 37.713 37.713 41.893 41.893 41.893 42.896 42.896 44.970 44.970 44.970 44.970 42.552 45.289 45.289 45.225 45.225 44.291 37.618 38.459 38.459 45.569 45.664 48.732 48.388 47.422 46.624 46.624

109.441 112.129 116.971 120.103 119.177 119.177 119.177 116.389 121.507 121.507 121.507 121.507 118.028 116.913 119.706 119.706 115.426 115.426 115.426 122.136 122.136 116.844 116.844 116.844 116.844 124.059 121.784 121.784 118.513 118.513 122.043 112.174 109.821 109.821 122.210 121.001 121.065 119.928 121.426 121.388 121.388

2508 1166 507 2130 2560 2560 2560 1235 1459 1459 1459 1459 1791 1726 1603 1603 1869 1869 1869 1996 1996 655 655 655 655 80 1531 1531 1259 1259 885 2481 1798 1798 230 178 568 1627 1049 1827 1827

29 Oct 2014 17 Sep 2010 5 Jul 2008 20 Sep 2014 13 Jul 2006 23 Aug 2013 8 Sep 2014 27 Oct 2007 29 Jun 2008 5 Aug 2009 25 Aug 2012 3 Aug 2014 26 Jun 2008 20 Jun 2015 29 Aug 2009 28 Jul 2014 21 Sep 2012 19 Jul 2014 30 Aug 2014 24 Aug 2006 25 Sep 2009 8 Sep 2006 29 Jul 2008 21 Sep 2012 22 Aug 2015 14 Aug 2005 18 Aug 2006 9 Sep 2011 8 Sep 2006 21 Sep 2012 19 Aug 2015 19 Aug 2015 1 Sep 2009 16 Aug 2012 22 Jul 2006 9 Sep 2011 6 Aug 2010 18 Sep 2012 22 Aug 2015 18 Sep 2008 28 Sep 2009

at 6 h intervals, and with 24 vertical levels reaching 100 mb. Smoke from biomass burning is derived from near-real-time satellite-based thermal anomaly data used to construct smoke source functions [Reid et al., 2009]. We use surface concentrations and optical depths associated with smoke to confirm that fires occurred during the time of IMPROVE EC peaks. If NAAPS did not reveal evidence of a major smoke event during the day of an IMPROVE EC peak, that particular event was omitted from the analysis. The third form of confirmation for fires included visual inspection of satellite imagery provided by NASA (fires before 2011: https://modis-atmos.gsfc.nasa.gov/IMAGES/; fires after 2011: https://worldview.earthdata.nasa. gov). Figure 2 shows an example of how the three aforementioned criteria were applied to identify a representative fire at Bliss State Park (California). Because a model and satellite data were used as criteria to identify fires, the analysis is biased toward large fires as opposed to smaller ones that these two tools could not resolve. 2.3. IMPROVE Data Analysis It was important for the subsequent discussion to define a background period by which to compare data from the peak EC day. Data were analyzed for the 5 weeks before and after the day identified with the

SCHLOSSER ET AL.


8954


10.1002/2017JD026547

Figure 2. Example of how IMPROVE elemental carbon (EC) data, smoke data from the NAAPS model, and NASA satellite imagery date are used to identify cases of wildfires in this study. The peak EC value occurs on 20 September 2014 at Bliss State Park in California.

peak EC mass concentration of each fire. For cases when there were adjacent IMPROVE data points in time with comparably high EC levels at the time of the peak of fire influence, an average of the peaks is used as the peak of the fire. This was the case for 10 fires, where the EC concentration of adjacent data points were within an average of 30% ± 21%. Fires with multiple peaks that were not directly adjacent to one another in time are not included in the analysis. Fires were also omitted from the analysis in instances when concentrations were not reported for EC for any of the three sampling days before or after the day of the EC peak or when multiple species did not have the “V0” status flag indicative of “Valid value.” If there was more than one site that had the exact same EC peak date, the site was used with the most data coverage (i.e., some species concentrations are not reported) of species around the time of the peak. If data coverage was comparable, the site with the highest peak EC mass concentration was used. In order to quantify enhancements of aerosol constituents during peak fire periods (i.e., day of highest EC mass concentration), four time periods are defined that are collectively meant to capture the strength of concentration enhancements. The three data points preceeding a peak are referred to as period “1,” and the three points following were designated as period “1”; note that a 24 h data point is collected every third day and thus periods 1/1 represent over a week of time. The period extending from the start of period 1 to 5 weeks before the EC peak is denoted period “2,” and the period following period 1 until 5 weeks after the peak was designated as period “2.” Percentage change is calculated as follows: %change ¼ 100

½x peak EC day ½x period average ½x period average

(2)

where [x] is the concentration of an aerosol constituent and “period” represents any of the four periods defined above. In order to qualify as a valid fire for analysis, a fire period could have missed a maximum of only 3 days of data (i.e., when a sample had no species concentrations reported) from the 5 week periods preceeding and following a fire’s peak. Only one of these missing days could be from period 1 or 1. A total of 41 individual fires were analyzed with their dates and site details provided in Table 1. The state with the most fires was California (14 fires), followed by Oregon (12), Washington (7), Utah (3), Nevada (3), and Arizona (2). It is cautioned that the states have widely ranging site densities; for example, the number of IMPROVE sites collecting data during the study period in the various states was as follows: California (23), Arizona (17), Washington (11), Oregon (6), Utah (4), and Nevada (2). Idaho, with two active IMPROVE sites

SCHLOSSER ET AL.


8955

Journal of Geophysical Research: Atmospheres Table 2. Summary of How the Percent Change Values of 39 Different IMPROVE Aerosol Constituents Compare for the Four Different Time a Periods Used in Equation (2) Period 2 Period 2 Period 1 Period 1 Period 2

Period 1

Period 1

Period 2

0.68

0.69 1.03

1.05 1.44 1.39

0.94 0.93 0.99

1.00 0.90

0.89

10.1002/2017JD026547

during the study period, registered no fires that met the study criteria. As subsequent discussion will compare different geographic categories, a region termed Northwest (NW) is defined for convenience that includes Oregon and Washington.

Table 2 summarizes how the percent change values vary between the four time periods used in equation (2) for 39 aerosol constituents available from the IMPROVE data set. Expectedly, there are strong correlations in the percent change values when intercomparing the four time periods. The slopes are important in showing how sensitive the percent change results are to the choice of the reference time period in equation (2). The slopes closest to unity were for the intercomparison of period 2 (ordinate) versus period 2 (1.05) and period 1 (ordinate) versus period 1 (1.03). The slope furthest from unity is 1.44 when comparing period 1 (ordinate) versus period 2. Table 2 indicates that enhancements are more depressed when using periods 1 and 2 as compared to 1 and 2, most likely owing to residual influence of the wildifre emissions after the peak EC day. Subsequent analysis will focus primarily on the 2 time period as the reference period. The general conclusions of this study are preserved regardless of the time period used (or using the overall average); the only differences are that the rankings of a few components vary slightly depending on the time period used, some of which are discussed below. Results for all periods are provided in Tables S1–S6 (see supporting information). a

The top right of the matrix represents slopes of best fit lines with the time period in the left column as the y axis. The bottom left half represents 2 the correlation (r ) values.

2.4. Size-Resolved Chemical Measurements For a case study analysis with more detailed size-resolved data as compared to the IMPROVE data, sizeresolved chemical measurements are presented from two separate summertime campaigns (July–August) based in Marina, California, which is a coastal site in central California approximately 5 km from the coast (36.7°N, 121.8°W). The Nucleation in California Experiment (NiCE) took place in 2013 [Coggon et al., 2014; Crosbie et al., 2016; Wang et al., 2016], while the Fog and Stratocumulus Evolution (FASE) took place in 2016. During both campaigns, surface-based composition measurements were conducted with a MicroOrifice Uniform Deposit Impactor (MOUDI, MSP Corporation) [Marple et al., 1991] with aerodynamic cut point diameters of 0.056, 0.1, 0.18, 0.32, 0.56, 1.0, 1.8, 3.2, 5.6, 10.0, and 18.0 μm. Teflon filters were used for MOUDI sampling (PTFE membrane, 2 μm pore, 46.2 mm, Whatman). Extractions of one half of each filter was performed by using 10 mL of milli-Q water in sealed glass vials that were sonicated at 30°C for 20 min. Samples were chemically analyzed with ion chromatography (Thermo Scientific Dionex ICS—2100 system) and triple quadrupole inductively coupled plasma mass spectrometry (Agilent 8800 Series), details of which are reported elsewhere [Maudlin et al., 2015]. Table S7 summarizes the sample set details and whether they were impacted by biomass burning or not, as based on olfactory and visual evidence, and the remarkable enhancement observed in numerous species examined [i.e., Maudlin et al., 2015; Sorooshian et al., 2015; Youn et al., 2015].

3. Results and Discussion 3.1. Cumulative IMPROVE Site Analysis Figure 3 summarizes the ranking of particulate species most enhanced in concentration on the peak EC day based on all fires in Table 1, ranked from highest to lowest, relative to the 2 time period. Table S1 (supporting information) lists the percent change values of all particulate species for each time period. PM2.5 was more enhanced as compared to PM10 (673% versus 376%) owing to some potential combination of secondary production of aerosol species during plume transport, less deposition of smaller particles (Dp < 2.5 μm) during plume transport, and more primary emissions of PM2.5 as compared to PM10. 3.1.1. Carbonaceous Components Among PM2.5 constituents, carbonaceous constituents were the most enhanced as shown in past work for a variety of different fuels [e.g., Echalar et al., 1995; Schauer et al., 1996; Fine et al., 2004]. The order of the top two overall constituents are the same for the four periods, with OC1 being most enhanced

SCHLOSSER ET AL.


8956


10.1002/2017JD026547

Figure 3. Ranking of percent changes in aerosol constituent concentrations on the peak EC day relative to the preceding 2 period, based on the average of all events shown in Table 1. Note that the top y axis is on a logarithmic scale to improve presentation of all bars as OC1 is significantly higher than other constituents. FS = fine soil, CM = PMcoarse, and Chl = chloride.

(5247%) followed by OC2 (1753%). As OC1 and OC2 are the first fractions of carbon to be evolved at the lowest temperatures, the constituents comprising these fraction are volatile [e.g., Kavouras et al., 2012; Lim et al., 2012], which makes this result interesting as it would be expected that the heat of the fire source would reduce these volatile OC fractions and enhance the nonvolatile fractions (OC3OC4) [Vergnoux et al., 2011; Kavouras et al., 2012]. In fact, the volatilization of primary organic aerosol emitted from North American trees/shrubs/grasses has already been characterized extensively by May et al. [2013]. The results of this study suggest that there could be formation of OC1 and OC2 during transport of plumes via gas-to-particle conversion from precursors emitted during fires, as has been demonstrated in laboratory [Grieshop et al., 2009; Hennigan et al., 2011] and field measurements [Yokelson et al., 2009]. The next most enhanced constituent was EC1 (1431%), which represents char EC and was previously noted to be emitted more strongly in smoldering conditions as compared to flaming conditions [Lim et al., 2012]. After EC1, the percent change of other carbonaceous constituents was as follows: total EC (1209%), total OC (1172%), OC3 (834%), EC3 (627%), OC4 (435%), and EC2 (208%). In contrast to the OC fractions that decrease in percent change in order from OC1 to OC4, the order for the EC components does not follow the same pattern from EC1 to EC3. EC3 has a higher percent change as compared to EC2, where both are representative of soot EC and thought to be emitted more strongly in flaming conditions [Lim et al., 2012]. That the OC and EC components were the most enhanced on the peak EC day relative to other species suggests that the hygroscopicity of those particles consequently was reduced (i.e., reduced inorganic mass fraction), as has been confirmed in field measurements over the western United States [e.g., Hersey et al., 2013; Shingler et al., 2016a] and laboratory experiments with biomass fuels common in this region [e.g., Lewis et al., 2009; Petters et al., 2009; Carrico et al., 2010]. 3.1.2. Elements and Inorganic Ions In terms of noncarbaneceous constituents, chlorine (Cl) exhibited the highest percent change in mass concentration on the peak EC day (1162%). Biomass burning is one of the most significant sources of Clcontaining species to the atmosphere [Lobert et al., 1999]. The following four constituents were next most enhanced, all of which have been noted to be biomass burning tracers in past work [e.g., Fine et al., 2004; Mahowald et al., 2005]: NO3 (486%), P (469%), K (438%), and Zn (357%). In decreasing order of percent change of water-soluble anions after NO3 were Cl (99%) and SO42 (60%). Nitrate originates from NOx emissions in fires that are photochemically oxidized to HNO3 followed by neutralization with an alkaline species [e.g., Prabhakar et al., 2014]. While SO42 is primarily formed secondarily from gaseous precursors, mainly SO2 emissions, it typically requires high relative humidity for formation, especially in clouds [e.g., Reid et al., 2005]. One study showed that SOx is emitted most intensely during the flaming phase of fires [Ferek et al., 1998]. Another study showed that SO2 concentrations were similar between fire SCHLOSSER ET AL.


8957


10.1002/2017JD026547

and nonfire periods during the time of a major fire in northern Alberta, Canada [Bytnerowicz et al., 2016]. Chloride is often in the form of KCl in biomass burning emissions [Reid et al., 2005], especially in young smoke [Li et al., 2003; Wonaschutz et al., 2011]. There is a large difference between Cl and Cl enhancement, and while the species are quantified via different instruments (XRF and IC, respectively), it has also been shown that Cl is the better proxy for fine-particle sea salt concentrations [White, 2008], which could at least partly explain why a difference exists. Although NO3 exhibited the largest percent change in concentration among the water-soluble anions measured, its absolute concentration was not the highest on the peak EC day for most of the 41 fires. Sulfate was the most dominant water-soluble anion in 28 of 36 wildfires, followed always by NO3; note that 5 of the 41 fires did not have SO42, NO3, nor Cl meet the “V0” status flag requirement to be included in this analysis. The remaining eight fires with valid data were characterized by NO3 having the highest mass concentration among anions on the peak EC day, followed always by SO42, and then Cl. Five of those fires were in California (C2, C4, C6, C9, and C11), with two in Oregon (O6 and O8), and one in Washington (W5). Reasons for variations in the dominant anion have been proposed to be differences in biomass density, fuel type, and amount of water present in the vegetation prior to burning [e.g., Li et al., 2003; Posfai et al., 2003; Ryu et al., 2004]. With regard to other global regions, savanna fires in the Amazon basin [Andreae et al., 1998; Yamasoe et al., 2000] and southern Africa [Gao et al., 2003] were characterized as having Cl as the dominant anion, while SO42 was most enhanced during wildfire activity around Moscow [Popovicheva et al., 2014]. The most enhanced anion, using equation (2), is sensitive to background concentrations of pollutants and their precursors that already exist. The higher percent change value for NO3 versus SO42 can be attributed most likely to its lower overall background concentrations in the study region. Based on analysis of IMPROVE data available in each state at the sites with fires in Table 1 between 2005 and 2015, the average concentrations of SO42/NO3/Cl are as follows: AZ = 0.54/0.13/0.02 μg m3; CA = 0.66/0.50/0.04 μg m3; NV = 0.35/ 0.09/0.01 μg m3; UT = 0.48/0.17/0.01 μg m3; OR = 0.35/0.15/0.05 μg m3; and WA = 0.45/0.28/0.06 μg m3. The higher frequency of nitrate being the dominant anion in fires in CA can likely be explained by how the long-term average concentration of NO3 in that state is the highest, and closest to SO42, as compared to the other areas. 3.1.3. Dust Figure 3 shows that fine soil and PMcoarse exhibit significant enhancements on peak EC days. It is assumed here that the main component of PMcoarse is dust as has been the case in many other IMPROVE studies. For example, Malm et al. [2007] investigated the composition of coarse particles across the continental United States and showed that soil was the major component (61%), with primary organic matter (POM) and ammonium nitrate contributing 24% and 8%, respectively. They also showed that the western United States generally had the highest fractional contributions of soil to PMcoarse with the exception that soil contributed only 34% to PMcoarse at Mount Rainier National Park with POM accounting for 59%. In this study, PMcoarse exhibited a lower percent change (74%) as compared to PM10 and PM2.5. The percent change in fine soil levels was similar (66%) to that of PMcoarse. Other notable species with average enhancements ≥100% included Mn (158%), Ca (148%), Br (146%), and Sr (103%). These species have origins in crustal matter [e.g., Ryu et al., 2004]. These results support the notion that turbulent mixing near flames and the burn front can lift soil dust particles, leading to the presence of soil dust with biomass burning plumes across the western United States. 3.2. Geographic IMPROVE Site Differences Table 3 summarizes percent change rankings for species separately in the states of Arizona, California, Nevada, Utah, and the Northwest (Oregon, Idaho, and Washington) based on the average for all fires in each region. Specific details about percent changes for each of the four time periods is shown in Tables S2–S6 for each of the five subregions. The various OC and EC parameters are consistently among the top ranked in terms of percent change regardless of region, with OC1 usually being the highest with the exception of Utah where EC3 exhibited the highest percent change. Of the noncarbonaceous components, Cl exhibited the highest percent change value in Arizona, California, Utah, and the Northwest, while P was highest in Nevada. NO3, Zn, and K had the next highest percent changes in each area. SCHLOSSER ET AL.


8958


10.1002/2017JD026547

Table 3. Ranking (Highest to Lowest) of Percent Change Values (Divided by 100) for Different States (NW = Northwestern States of Oregon and Washington) for the a Peak EC Day Relative to the 2 Period AZ Component OC1 OC2 EC OC EC1 OC3 PM2.5 OC4 PM10 EC2 Cl NO3 K Zn Br Chl Na Ca Ni As Mg Cr V PMcoarse EC3 Fe Fine soil Si Ti Al Sr SO4 Mn S Se Rb Cu Pb P a

Ca

NV

NW

UT

% Change

Component

% Change

Component

% Change

Component

% Change

Component

% Change

276.91 44.77 28.09 26.94 21.02 16.28 9.81 7.28 5.35 4.59 3.52 3.45 2.45 1.60 1.44 1.15 0.84 0.84 0.68 0.60 0.59 0.56 0.39 0.36 0.32 0.28 0.27 0.19 0.18 0.16 0.02 0.10 0.16 0.17 0.21 0.31 0.34 0.37 NA

OC1 OC2 Cl EC1 EC OC OC3 NO3 P PM2.5 EC3 K OC4 PM10 Zn EC2 Chl Ca Br Mn Sr Cu Pb Rb As PMcoarse SO4 Cr Se S Al V Fine soil Fe Na Ti Ni Si Mg

43.86 15.27 13.70 12.39 11.62 10.02 7.22 6.62 6.27 6.04 5.12 4.53 3.34 3.08 2.56 1.74 1.54 1.46 1.39 1.32 1.10 0.96 0.85 0.75 0.51 0.45 0.44 0.43 0.37 0.33 0.31 0.25 0.22 0.21 0.18 0.17 0.05 0.05 0.02

OC1 OC2 EC EC1 EC3 OC P OC3 Cl OC4 NO3 PM2.5 Zn K Se Pb PM10 EC2 Chl Br Ca Mn Ni V SO4 S Cr Fe Rb Ti Cu Sr Fine soil Si Al PMcoarse Mg Na As

13.91 12.41 10.80 10.63 8.19 7.78 7.08 6.63 4.29 4.27 3.75 3.46 3.10 2.38 2.28 1.94 1.38 1.25 1.23 0.92 0.68 0.58 0.32 0.32 0.31 0.28 0.19 0.17 0.16 0.15 0.14 0.07 0.04 0.00 0.05 0.19 0.19 0.22 NA

OC1 OC2 EC1 OC Cl EC OC3 PM2.5 EC3 PM10 K OC4 Zn NO3 EC2 Mn P Ca Br Fe Sr PMcoarse Fine soil Ti Si Rb As Al Cu SO4 Se Chl Mg S Pb V Cr Na Ni

46.09 18.00 15.70 12.11 11.96 11.16 8.30 7.65 5.85 4.73 4.32 4.28 4.17 3.71 2.32 2.19 1.84 1.75 1.44 1.34 1.23 1.22 1.19 1.18 1.05 0.98 0.98 0.92 0.87 0.84 0.73 0.60 0.57 0.55 0.54 0.39 0.26 0.19 0.19

EC3 OC1 Cl EC1 OC2 OC EC OC3 OC4 K Zn NO3 PM2.5 PM10 Br P EC2 Cu Sr Ca Mn Pb As SO4 S Se PMcoarse Al Fine soil Fe Ti Chl Mg Si Rb V Na Cr Ni

21.00 19.14 16.63 13.70 12.11 10.94 10.78 10.29 7.61 7.36 6.18 5.65 5.40 2.49 2.44 2.16 1.31 1.26 1.09 1.05 1.02 0.87 0.85 0.77 0.64 0.43 0.36 0.22 0.21 0.14 0.11 0.05 0.05 0.03 0.03 0.35 0.47 0.80 NA

“NA” means that sufficient data were not available for a particular species. “Chl” refers to chloride. PMcoarse = PM10–PM2.5.

Species exhibiting wide ranges in percent change among the five areas include all of the OC constituents (OC1–4) owing to their significantly higher values in Arizona versus other regions. Conversely, EC3 exhibited a percent change in Utah (2100%) that was a far larger than other regions (32–819%). Several other species exhibited significantly higher percent change values in Utah versus other regions including Cl, K, and Zn. Fine soil exhibits the greatest percent enhancement in the Northwest (119%) as compared to Arizona (27%), California (22%), Utah (21%), and Nevada (4%). While the Northwest sites exhibited a lower overall altitude than other regions, no significant trend was found between the fine soil percent changes as a function of site altitude. The more important feature of the Northwest is the lower background concentrations of fine soil. More specifically, based on analysis of IMPROVE data available in each state at the sites with fires in Table 1 between 2005 and 2015, the average concentrations of fine soil are as follows: AZ = 0.69 μg m3; CA = 0.59 μg m3; NV = 0.98 μg m3; UT = 0.64 μg m3; OR = 0.36 μg m3; and WA = 0.35 μg m3. As a result, additions of fine soil aloft in the northwestern states from fire activity yielded a higher percent change value than other regions that already have more dust in the air year-round.

SCHLOSSER ET AL.


8959


10.1002/2017JD026547

Figure 4. Graphical display of a fire in Southern California and the trajectory of its plume. (a) Image showing fire spots between 27 October and 30 October 2003 (https://earthdata.nasa.gov/firms). (b) MODIS Aqua image from 27 October 2003 showing the fires near the Agua Tibia IMPROVE station. (c) MODIS Aqua image from 30 October 2003 showing the plumes moving toward Arizona and the Saguaro West IMPROVE station. (d) Four day NOAA HYSPLIT back trajectory ending at the Saguaro West IMPROVE site.

3.3. Interstate Transport Wildfires significantly impact downind areas, and an example of this is a case from October 2003 where wildfires from California impacted Arizona. This scenario was captured at Agua Tibia in California on 27 October and Saguaro West in Arizona on 30 October. Figure 4a illustrates the high density of fires erupting in Southern California, especially near Agua Tibia, based on data from Moderate Resolution Imaging Spectroradiometer (MODIS) Collection 5 near-real-time data (https://earthdata.nasa.gov/firms). Figure 4b shows wildfire emissions concentrated near Agua Tibia on 27 October, while Figure 4c shows the movement of the plumes 3 days later to the east toward Arizona. Figure 4d shows 96 h back trajectories from the NOAA HYSPLIT model [Stein et al., 2015; Rolph, 2016] ending at the point of the Saguaro West station on its peak EC day of 30 October. The California fires influenced much of the region offshore where trajectories traveled over before reaching Saguaro West, which is confirmed by NAAPS data in Figure 5 where the progression of fire influence is shown between 27 and 30 October. Smoke enveloped much of the greater Southwest including all of Arizona (Figures 5a and 5b). A time series of numerous species, including biomass burning tracers, is shown for Agua Tibia and Saguaro West in Figure 6. PM10, PM2.5, PMcoarse, OC, EC, K, and fine soil exhibit enhancements on the peak EC day as compared to preceding and subsequent weeks, with the exception of fine soil being high on the sample day before the peak EC day at Agua Tibia. Of the two sites, the most significant increase in PMcoarse on the peak EC day was at Saguaro West (from 17 μg m3 on 27 October to 50 μg m3 on 30 October); however, this event was not identified as a local dust event with the similar IMPROVE data set in Tong et al.’s [2012] study that constructed a long-term dust record in the western United States. Two of the five different criteria used in their method to detect dust events included low PM2.5:PM10 ratios and low contributions from

SCHLOSSER ET AL.


8960


10.1002/2017JD026547

Figure 5. Evolution of surface (a, b) smoke concentrations and (c, d) dust concentrations over the western United States between 27 October and 30 October 2003, respectively, based on NAAPS model data. Figures 5a and 5c correspond to 18:00 Z on 27 October 2003, while Figures 5b and 5d correspond to 18:00 Z on 30 October 2003.

pollutants such as, Zn, Cu, Pb, SO42, NO3, OC, and EC, both of which break down to some extent when dust is emitted during a fire based on the results of our study. This is because there are high levels in wildfire plumes of both PM2.5 and the anthropogenic pollutants selected in that study.

Figure 6. Time series of selected IMPROVE aerosol chemical parameters associated with the case study event shown in Figures 4 and 5. The shaded pink day corresponds to the peak EC period at (a) Agua Tibia (California) and (b) Saguaro West (Arizona). Note that some points are missing as those data were not available. The status flag for the PM2.5 data point on 27 October 2003 is labeled as “V5” by IMPROVE, which is defined as “Valid value but qualified because of possible contamination.”

SCHLOSSER ET AL.


8961


10.1002/2017JD026547

To further motivate greater attention to fires as a source of dust, NAAPS does not show any dust influence over the Southwest during this case event (Figures 5c and 5d). This is because NAAPS is not designed to resolve dust dynamics in fires. Dust emission in NAAPS is a function of surface friction velocity, surface wetness, and surface erodibility [Westphal et al., 1988; Lynch et al., 2016]. Only when surface friction velocity or surface wind exceeds a threshold, and surface wetness is below a certain value over erodible surfaces, can dust be emitted. This type of parameterization is broadly Figure 7. Mass size distributions of chloride for MOUDI sample sets in July– used in dust models [Huneeus et al., August of 2013 and 2016 at a coastal site in central California (Marina) 2011]. Consequently, whether or not impacted by biomass burning. The vertical bars represent one standard an aerosol model can capture a dust deviation of data from sample sets with (n = 7) and without (n = 10) fire event is highly dependent on the influence from Table S7. underlying simulated surface properties, especially surface wind. Turbulence, buoyancy, downdrafts, and changes in surface wind promoted by fires are not readily represented in meteorological models, except for a limited number of emerging research models that are focused on fire-atmosphere interactions (e.g., the Weather Research and Forecasting Model WRF-Fire [Coen et al., 2013]); fire-atmosphere interactions are an ongoing research topic [Potter, 2012]. There are coexisting dust and smoke plumes in NAAPS, but they tend to be caused by separate dust and smoke processes. An improved aerosol model would ideally rely on the driving meteorology to resolve surface wind characteristics associated with fires to more accurately capture fire-promoted dust emissions; this is currently a major challenge for a global model such as NAAPS. 3.4. Chloride Depletion An important feature of Figure 6 is Cl reaching depressed levels on the peak EC day as compared to adjacent periods before and after the peak EC day. This was a common occurrence among the fires in Table 1. For specific time periods used in equation (2), the number of fires from Table 1 with negative percent change values for Cl was as follows (values in parentheses represent total number of possible cases with the “Valid value (V0)” status flag according to IMPROVE for Cl): 2 = 15 (33), 1 = 10 (33), 1 = 14 (33), 2 = 18 (34). Other work has shown that Cl is depleted in wildfire plumes, including over California [Zauscher et al., 2013; Maudlin et al., 2015]. This is due to the well-documented chloride depletion phenomenon promoted by acidic species (e.g., sulfuric and nitric acids and organic acids) [e.g., Gaudichet et al., 1995; Gao et al., 2003]. Measurements in biomass burning plumes over southern Africa suggested that KCl particles in young smoke were converted into K2SO4 and KNO3 in aged smoke via reactions with S- and N-containing species from biomass burning and other sources [Li et al., 2003]. Similar observations have been made in Southern California wildfires [Zauscher et al., 2013]. Although organic acid concentrations are not available in the IMPROVE data set, they are enhanced in wildfires [e.g., Gao et al., 2003] and can contribute appreciably to chloride depletion. To probe deeper into the chloride depletion issue, size-resolved chemical measurements are examined here that were collected during the NiCE and FASE campaigns in July–August of 2013 and 2016, respectively. During NiCE, a series of wildfires based near the California-Oregon border began near the end of July (Big Windy, Whiskey Complex, and Douglas Complex forest fires), while during FASE, the Soberanes Wildfire was present just south of Marina in Big Sur, which impacted the study region for the majority of the field campaign. Figure 7 compares Cl data in sample sets with and without biomass burning influence; fortunately, sample sets were collected in each campaign before the fires began (Table S7), to allow for a comparison between fire and nonfire periods. SCHLOSSER ET AL.


8962


10.1002/2017JD026547

The total Cl mass concentration integrated over all MOUDI stages was 1.41 ± 0.68 μg m3 and 0.33 ± 0.17 μg m3 for nonfire and fire conditions, respectively. The difference was mainly in the supermicrometer sizes, as evident in the mass size distributions, where the average Cl concentration was 1.37 ± 0.65 μg m3 and 0.31 ± 0.16 μg m3 for nonfire and fire conditions, respectively. As sea salt is abundant in this coastal area, the Cl:Na mass ratio provides an indication of the degree of chloride depletion, as the value for natural sea salt is 1.8. For the three MOUDI stages with the majority of the mass of Cl and Na (1– 5.6 μm), the Cl:Na ratio was 1.57 and 0.59 for nonfire and fire conditions (Figure 7). Interestingly, NO3 and SO42 were not enhanced in those stages during fire periods: NO3 = 0.07 μg m3 (fire) versus 0.10 μg m3 (nonfire); SO42 = 0.06 μg m3 versus 0.12 μg m3; however, organic acids, especially oxalate, were significantly enhanced (0.02 versus 0.01 μg m3) albeit at lower concentrations. Processes that alter aerosol composition and morphology during transport and aging (e.g., chloride depletion, sulfate and nitrate formation, and organic aerosol evaporation and formation) of biomass burning plumes impact the hygroscopic and radiative properties of aerosol [e.g., Lewis et al., 2009; Engelhart et al., 2012; Shingler et al., 2016b]. Therefore, it is important to understand more clearly why the sharp reduction in Cl levels only occurs in some fires and not others.

4. Conclusions This study has examined aerosol composition data for major wildfires between 2005 and 2015 across the western United States with the goal of identifying what particulate species exhibit the higher percent change in concentration on the day of the peak fire influence relative to periods before that day. After a filtering technique of isolating wildfire events using IMPROVE, NAAPS, and MODIS data, a total of 41 fires were identified and studied.

Acknowledgments All IMPROVE, NASA, and NAAPS data used can be obtained from websites provided in section 2, while FASE and NiCE data can be obtained from the corresponding author. This work was funded by grant 2 P42 ES04940 from the National Institute of Environmental Health Sciences (NIEHS) Superfund Research Program and ONR grants N00014-16-1-2567 and N00014-10-1-0811. We acknowledge the use of data and imagery from LANCE FIRMS operated by the NASA/GSFC/Earth Science Data and Information System (ESDIS) with funding provided by NASA/HQ. We acknowledge Agilent Technologies for their support and Shane Snyder’s laboratories for ICP-MS data. IMPROVE is a collaborative association of state, tribal, and federal agencies and international partners. The U.S. Environmental Protection Agency is the primary funding source, with contracting and research support from the National Park Service. The Air Quality Group at the University of California, Davis, is the central analytical laboratory, with ion analysis provided by Research Triangle Institute and carbon analysis provided by Desert Research Institute.

SCHLOSSER ET AL.

Species exhibiting the highest percent change increase on the peak EC day include the various components of OC and EC with OC1 being the highest, followed by OC2 and EC1. The volatile nature of OC1 and OC2 suggests a secondary formation mechanism for these species during plume transport. Of the noncarbonaceous constituents, Cl, P, K, NO3, and Zn followed next in terms of having the highest percent change in concentration. While NO3 exhibited the greatest percent change in mass on the peak EC day among water-soluble anions (versus Cl and SO42), it was only the dominant anion on an absolute mass basis in 8 of 36 fires with valid data since SO42 was usually more abundant. Dust was significantly enhanced in study region plumes, with fine soil and PMcoarse exhibiting comparable percent change enhancements on the peak EC day. This included many trace metals associated with crustal matter (e.g., Mn, Ca, Fe, Al, Ti, and Sr). The percent change for dust was highest in the northwestern states of Oregon and Washington owing to lower background concentrations during nonfire periods as compared to the rest of the western United States. A case study emphasizes how transport of wildfire plumes can significantly impact downwind states. An important feature of the particular event examined is that there were higher levels of fine soil and PMcoarse at the downwind state (Arizona) as compared to the source of the fires (California). The results from IMPROVE, and also the inability of a global aerosol model such as NAAPS to accurately represent dust emissions during fires, warrant more attention to this issue. Significant reductions in Cl concentration are observed on the peak EC day of almost half of the fires identified in this study. Size-resolved measurements in coastal California during two different summers reveal that there is a major difference in Cl levels between 1 and 10 μm during periods with and without fire influence. While sulfate and nitrate levels were not enhanced during the fire periods, those of organic acids instead were, which highlights the potentially large importance of organic species in altering the inorganic composition of plume aerosol particles and consequently their hygroscopic and radiative properties.

References Andreae, M. O., and P. Merlet (2001), Emission of trace gases and aerosols from biomass burning, Global Biogeochem. Cycles, 15(4), 955–966, doi:10.1029/2000GB001382. Andreae, M. O., et al. (1998), Airborne studies of aerosol emissions from savanna fires in southern Africa: 2. Aerosol chemical composition, J. Geophys. Res., 103(D24), 32119–32128, doi:10.1029/98JD02280. Bytnerowicz, A., Y. M. Hsu, K. Percy, A. Legge, M. E. Fenn, S. Schilling, W. Fraczek, and D. Alexander (2016), Ground-level air pollution changes during a boreal wildland mega-fire, Sci. Total Environ., 572, 755–769, doi:10.1016/j.scitotenv.2016.07.052.


8963


10.1002/2017JD026547

Cahill, C. F., T. A. Cahill, and K. D. Perry (2008), The size- and time-resolved composition of aerosols from a sub-Arctic boreal forest prescribed burn, Atmos. Environ., 42(32), 7553–7559, doi:10.1016/j.atmosenv.2008.04.034. Cahill, T. A., L. L. Ashbaugh, R. A. Eldred, P. J. Feeney, and B. Kusko (1980), Comparisons between size-segregated resuspended soil samples and ambient aerosols in the western United-States, Abstr. Pap. Am. Chem. S., 180 (Aug), 157-Nucl. Carrico, C. M., M. D. Petters, S. M. Kreidenweis, A. P. Sullivan, G. R. McMeeking, E. J. T. Levin, G. Engling, W. C. Malm, and J. L. Collett (2010), Water uptake and chemical composition of fresh aerosols generated in open burning of biomass, Atmos. Chem. Phys., 10(11), 5165–5178, doi:10.5194/acp-10-5165-2010. Chalbot, M. C., G. Nikolich, V. Etyemezian, D. W. Dubois, J. King, D. Shafer, G. G. da Costa, J. F. Hinton, and I. G. Kavouras (2013), Soil humic-like organic compounds in prescribed fire emissions using nuclear magnetic resonance spectroscopy, Environ. Pollut., 181, 167–171, doi:10.1016/j.envpol.2013.06.008. Chow, J. C., J. G. Watson, L. C. Pritchett, W. R. Pierson, C. A. Frazier, and R. G. Purcell (1993), The DRI thermal optical reflectance carbon analysis system—Description, evaluation and applications in United-States air-quality studies, Atmos. Environ. A. Gen., 27(8), 1185–1201, doi:10.1016/0960-1686(93)90245-T. Clements, C. B., S. Y. Zhong, X. D. Bian, W. E. Heilman, and D. W. Byun (2008), First observations of turbulence generated by grass fires, J. Geophys. Res., 113, D22102, doi:10.1029/2008JD010014. Coen, J. L., M. Cameron, J. Michalakes, E. G. Patton, P. J. Riggan, and K. M. Yedinak (2013), WRF-Fire: Coupled weather-wildland fire modeling with the Weather Research and Forecasting model, J. Appl. Meteorol. Climatol., 52(1), 16–38, doi:10.1175/Jamc-D-12-023.1. Coggon, M. M., A. Sorooshian, Z. Wang, J. S. Craven, A. R. Metcalf, J. J. Lin, A. Nenes, H. H. Jonsson, R. C. Flagan, and J. H. Seinfeld (2014), Observations of continental biogenic impacts on marine aerosol and clouds off the coast of California, J. Geophys. Res. Atmos., 119, 6724–6748, doi:10.1002/2013JD021228. Crosbie, E., et al. (2016), Stratocumulus cloud clearings and notable thermodynamic and aerosol contrasts across the clear-cloudy interface, J. Atmos. Sci., 73(3), 1083–1099, doi:10.1175/Jas-D-15-0137.1. Crutzen, P. J., and M. O. Andreae (1990), Biomass burning in the tropics—Impact on atmospheric chemistry and biogeochemical cycles, Science, 250(4988), 1669–1678, doi:10.1126/science.250.4988.1669. Dennison, P. E., S. C. Brewer, J. D. Arnold, and M. A. Moritz (2014), Large wildfire trends in the western United States, 1984–2011, Geophys. Res. Lett., 41, 2928–2933, doi:10.1002/2014GL059576. Echalar, F., A. Gaudichet, H. Cachier, and P. Artaxo (1995), Aerosol emissions by tropical forest and savanna biomass burning—Characteristic trace-elements and fluxes, Geophys. Res. Lett., 22(22), 3039–3042, doi:10.1029/95GL03170. Eldred, R. A., T. A. Cahill, and R. G. Flocchini (1997), Composition of PM2.5 and PM10 aerosols in the IMPROVE network, J. Air Waste Manage. Assoc., 47(2), 194–203. Engelhart, G. J., C. J. Hennigan, M. A. Miracolo, A. L. Robinson, and S. N. Pandis (2012), Cloud condensation nuclei activity of fresh primary and aged biomass burning aerosol, Atmos. Chem. Phys., 12(15), 7285–7293, doi:10.5194/acp-12-7285-2012. Ferek, R. J., J. S. Reid, P. V. Hobbs, D. R. Blake, and C. Liousse (1998), Emission factors of hydrocarbons, halocarbons, trace gases and particles from biomass burning in Brazil, J. Geophys. Res., 103(D24), 32107–32118, doi:10.1029/98JD00692. Fine, P. M., G. R. Cass, and B. R. T. Simoneit (2001), Chemical characterization of fine particle emissions from fireplace combustion of woods grown in the northeastern United States, Environ. Sci. Technol., 35(13), 2665–2675, doi:10.1021/es001466k. Fine, P. M., G. R. Cass, and B. R. T. Simoneit (2004), Chemical characterization of fine particle emissions from the wood stove combustion of prevalent United States tree species, Environ. Eng. Sci., 21(6), 705–721, doi:10.1089/ees.2004.21.705. Flannigan, M. D., B. J. Stocks, and B. M. Wotton (2000), Climate change and forest fires, Sci. Total Environ., 262(3), 221–229, doi:10.1016/S00489697(00)00524-6. Fraser, M. P., and K. Lakshmanan (2000), Using levoglucosan as a molecular marker for the long-range transport of biomass combustion aerosols, Environ. Sci. Technol., 34(21), 4560–4564, doi:10.1021/es991229l. Gao, S., D. A. Hegg, P. V. Hobbs, T. W. Kirchstetter, B. I. Magi, and M. Sadilek (2003), Water-soluble organic components in aerosols associated with savanna fires in southern Africa: Identification, evolution, and distribution, J. Geophys. Res., 108(D13), 8491, doi:10.1029/2002JD002324. Gaudichet, A., F. Echalar, B. Chatenet, J. P. Quisefit, G. Malingre, H. Cachier, P. Buatmenard, P. Artaxo, and W. Maenhaut (1995), Trace-elements in tropical African savanna biomass burning aerosols, J. Atmos. Chem., 22(1–2), 19–39, doi:10.1007/Bf00708179. Grieshop, A. P., J. M. Logue, N. M. Donahue, and A. L. Robinson (2009), Laboratory investigation of photochemical oxidation of organic aerosol from wood fires 1: Measurement and simulation of organic aerosol evolution, Atmos. Chem. Phys., 9(4), 1263–1277. Hallar, A. G., R. Petersen, E. Andrews, J. Michalsky, I. B. McCubbin, and J. A. Ogren (2015), Contributions of dust and biomass burning to aerosols at a Colorado mountain-top site, Atmos. Chem. Phys., 15(23), 13665–13679, doi:10.5194/acp-15-13665-2015. Hallar, A. G., N. P. Molotch, J. L. Hand, B. Livneh, I. B. McCubbin, R. Petersen, J. Michalsky, D. Lowenthal, and K. E. Kunkel (2017), Impacts of increasing aridity and wildfires on aerosol loading in the intermountain western US, Environ. Res. Lett., 12(1), doi:10.1088/1748-9326/aa510a. Hand, J. L., W. H. White, K. A. Gebhart, N. P. Hyslop, T. E. Gill, and B. A. Schichtel (2016), Earlier onset of the spring fine dust season in the southwestern United States, Geophys. Res. Lett., 43, 4001–4009, doi:10.1002/2016GL068519. Hays, M. D., C. D. Geron, K. J. Linna, N. D. Smith, and J. J. Schauer (2002), Speciation of gas-phase and fine particle emissions from burning of foliar fuels, Environ. Sci. Technol., 36(11), 2281–2295, doi:10.1021/es0111683. Hennigan, C. J., A. P. Sullivan, J. L. Collett, and A. L. Robinson (2010), Levoglucosan stability in biomass burning particles exposed to hydroxyl radicals, Geophys. Res. Lett., 37, L09806, doi:10.1029/2010GL043088. Hennigan, C. J., et al. (2011), Chemical and physical transformations of organic aerosol from the photo-oxidation of open biomass burning emissions in an environmental chamber, Atmos. Chem. Phys., 11(15), 7669–7686, doi:10.5194/acp-11-7669-2011. Hersey, S. P., et al. (2013), Composition and hygroscopicity of the Los Angeles aerosol: CalNex, J. Geophys. Res. Atmos., 118, 3016–3036, doi:10.1002/jgrd.50307. Hoffmann, D., A. Tilgner, Y. Iinuma, and H. Herrmann (2010), Atmospheric stability of levoglucosan: A detailed laboratory and modeling study, Environ. Sci. Technol., 44(2), 694–699, doi:10.1021/es902476f. Hogan, T. F., and L. R. Brody (1993), Sensitivity studies of the navy global forecast model parameterizations and evaluation of improvements to NOGAPS, Mon. Weather Rev., 121(8), 2373–2395, doi:10.1175/1520-0493(1993)1212.0.Co;2. Hogan, T. F., and T. E. Rosmond (1991), The description of the Navy Operational Global Atmospheric Prediction Systems spectral forecast model, Mon. Weather Rev., 119(8), 1786–1815, doi:10.1175/1520-0493(1991)1192.0.Co;2. Huneeus, N., et al. (2011), Global dust model intercomparison in AeroCom phase I, Atmos. Chem. Phys., 11(15), 7781–7816, doi:10.5194/acp11-7781-2011. Jaffe, D., W. Hafner, D. Chand, A. Westerling, and D. Spracklen (2008), Interannual variations in PM2.5 due to wildfires in the western United States, Environ. Sci. Technol., 42(8), 2812–2818, doi:10.1021/es702755v.

SCHLOSSER ET AL.


8964


10.1002/2017JD026547

Kavouras, I. G., V. Etyemezian, J. Xu, D. W. DuBois, M. Green, and M. Pitchford (2007), Assessment of the local windblown component of dust in the western United States, J. Geophys. Res., 112, D08211, doi:10.1029/2006JD007832. Kavouras, I. G., V. Etyemezian, D. W. DuBois, J. Xu, and M. Pitchford (2009), Source reconciliation of atmospheric dust causing visibility impairment in class I areas of the western United States, J. Geophys. Res., 114, D02308, doi:10.1029/2008JD009923. Kavouras, I. G., G. Nikolich, V. Etyemezian, D. W. DuBois, J. King, and D. Shafer (2012), In situ observations of soil minerals and organic matter in the early phases of prescribed fires, J. Geophys. Res., 117, D18308, doi:10.1029/2012JD018771. Lee, T., et al. (2010), Chemical smoke marker emissions during flaming and smoldering phases of laboratory open burning of wildland fuels, Aerosol Sci. Technol., 44(9), I–V, doi:10.1080/02786826.2010.499884. Lewis, K. A., et al. (2009), Reduction in biomass burning aerosol light absorption upon humidification: Roles of inorganically-induced hygroscopicity, particle collapse, and photoacoustic heat and mass transfer, Atmos. Chem. Phys., 9(22), 8949–8966. Li, J., M. Posfai, P. V. Hobbs, and P. R. Buseck (2003), Individual aerosol particles from biomass burning in southern Africa: 2, Compositions and aging of inorganic particles, J. Geophys. Res., 108(D13), 8484, doi:10.1029/2002JD002310. Lim, S., M. Lee, G. Lee, S. Kim, S. Yoon, and K. Kang (2012), Ionic and carbonaceous compositions of PM10, PM2.5 and PM1.0 at Gosan ABC superstation and their ratios as source signature, Atmos. Chem. Phys., 12(4), 2007–2024, doi:10.5194/acp-12-2007-2012. Lobert, J. M., W. C. Keene, J. A. Logan, and R. Yevich (1999), Global chlorine emissions from biomass burning: Reactive chlorine emissions inventory, J. Geophys. Res., 104(D7), 8373–8389, doi:10.1029/1998JD100077. Lynch, P., et al. (2016), An 11-year global gridded aerosol optical thickness reanalysis (v1.0) for atmospheric and climate sciences, Geosci. Model Dev., 9(4), 1489–1522, doi:10.5194/gmd-9-1489-2016. Maenhaut, W., I. Salma, J. Cafmeyer, H. J. Annegarn, and M. O. Andreae (1996), Regional atmospheric aerosol composition and sources in the eastern Transvaal, South Africa, and impact of biomass burning, J. Geophys. Res., 101(D19), 23631–23650, doi:10.1029/95JD02930. Mahowald, N. M., P. Artaxo, A. R. Baker, T. D. Jickells, G. S. Okin, J. T. Randerson, and A. R. Townsend (2005), Impacts of biomass burning emissions and land use change on Amazonian atmospheric phosphorus cycling and deposition, Global Biogeochem. Cycles, 19, Gb4030, doi:10.1029/2005GB002541. Malm, W. C., J. F. Sisler, D. Huffman, R. A. Eldred, and T. A. Cahill (1994), Spatial and seasonal trends in particle concentration and optical extinction in the United-States, J. Geophys. Res., 99(D1), 1347–1370, doi:10.1029/93JD02916. Malm, W. C., M. L. Pitchford, C. McDade, and L. L. Ashbaugh (2007), Coarse particle speciation at selected locations in the rural continental United States, Atmos. Environ., 41(10), 2225–2239, doi:10.1016/j.atmosenv.2006.10.077. Marple, V. A., K. L. Rubow, and S. M. Behm (1991), A Microorifice Uniform Deposit Impactor (MOUDI)—Description, calibration, and use, Aerosol Sci. Technol., 14(4), 434–446, doi:10.1080/02786829108959504. Maudlin, L. C., Z. Wang, H. H. Jonsson, and A. Sorooshian (2015), Impact of wildfires on size-resolved aerosol composition at a coastal California site, Atmos. Environ., 119, 59–68, doi:10.1016/j.atmosenv.2015.08.039. May, A. A., E. J. T. Levin, C. J. Hennigan, I. Riipinen, T. Lee, J. L. Collett, J. L. Jimenez, S. M. Kreidenweis, and A. L. Robinson (2013), Gasparticle partitioning of primary organic aerosol emissions: 3. Biomass burning, J. Geophys. Res. Atmos., 118, 11327–11338, doi:10.1002/ jgrd.50828. Moritz, M. A., M. A. Parisien, E. Batllori, M. A. Krawchuk, J. Van Dorn, D. J. Ganz, and K. Hayhoe (2012), Climate change and disruptions to global fire activity, Ecosphere, 3(6), doi:10.1890/Es11-00345.1. Nisantzi, A., R. E. Mamouri, A. Ansmann, and D. Hadjimitsis (2014), Injection of mineral dust into the free troposphere during fire events observed with polarization lidar at Limassol, Cyprus, Atmos. Chem. Phys., 14(22), 12155–12165, doi:10.5194/acp-14-12155-2014. Palmer, T. Y. (1981), Large fire winds, gases and smoke, Atmos. Environ., 15(10–1), 2079–2090, doi:10.1016/0004-6981(81)90241-9. Petters, M. D., C. M. Carrico, S. M. Kreidenweis, A. J. Prenni, P. J. DeMott, J. L. Collett, and H. Moosmuller (2009), Cloud condensation nucleation activity of biomass burning aerosol, J. Geophys. Res., 114, D22205, doi:10.1029/2009JD012353. Pickering, K. E., et al. (1996), Convective transport of biomass burning emissions over Brazil during TRACE A, J. Geophys. Res., 101(D19), 23993–24012, doi:10.1029/96JD00346. Pitchford, M., R. G. Flocchini, R. G. Draftz, T. A. Cahill, L. L. Ashbaugh, and R. A. Eldred (1981), Silicon in sub-micron particles in the Southwest, Atmos. Environ., 15(3), 321–333, doi:10.1016/0004-6981(81)90035-4. Popovicheva, O., M. Kistler, E. Kireeva, N. Persiantseva, M. Timofeev, V. Kopeikin, and A. Kasper-Giebl (2014), Physicochemical characterization of smoke aerosol during large-scale wildfires: Extreme event of August 2010 in Moscow, Atmos. Environ., 96, 405–414, doi:10.1016/j. atmosenv.2014.03.026. Posfai, M., R. Simonics, J. Li, P. V. Hobbs, and P. R. Buseck (2003), Individual aerosol particles from biomass burning in southern Africa: 1. Compositions and size distributions of carbonaceous particles, J. Geophys. Res., 108(D13), 8483, doi:10.1029/2002JD002291. Potter, B. E. (2012), Atmospheric interactions with wildland fire behaviour—I. Basic surface interactions, vertical profiles and synoptic structures, Int. J. Wildland Fire, 21(7), 779–801, doi:10.1071/Wf11128. Prabhakar, G., B. Ervens, Z. Wang, L. C. Maudlin, M. M. Coggon, H. H. Jonsson, J. H. Seinfeld, and A. Sorooshian (2014), Sources of nitrate in stratocumulus cloud water: Airborne measurements during the 2011 E-PEACE and 2013 NiCE studies, Atmos. Environ., 97, 166–173, doi:10.1016/j.atmosenv.2014.08.019. Reid, J. S., R. Koppmann, T. F. Eck, and D. P. Eleuterio (2005), A review of biomass burning emissions part II: Intensive physical properties of biomass burning particles, Atmos. Chem. Phys., 5, 799–825. Reid, J. S., et al. (2009), Global monitoring and forecasting of biomass-burning smoke: Description of and lessons from the Fire Locating and Modeling of Burning Emissions (FLAMBE) program, IEEE J. Stars, 2(3), 144–162, doi:10.1109/Jstars.2009.2027443. Rolph, G.D. (2016), Real-time Environmental Applications and Display sYstem (READY) website (http://ready.Arl.NOAA.Gov), NOAA Air Resour. Lab., Silver Spring, Md. Ryu, S. Y., J. E. Kim, H. Zhuanshi, Y. J. Kim, and G. U. Kang (2004), Chemical composition of post-harvest biomass burning aerosols in Gwangju, Korea, J. Air Waste Manage. Assoc., 54(9), 1124–1137. Schauer, J. J., W. F. Rogge, L. M. Hildemann, M. A. Mazurek, G. R. Cass, and B. R. T. Simoneit (1996), Source apportionment of airborne particulate matter using organic compounds as tracers, Atmos. Environ., 30(22), 3837–3855, doi:10.1016/1352-2310(96)00085-4. Shingler, T., et al. (2016a), Airborne characterization of subsaturated aerosol hygroscopicity and dry refractive index from the surface to 4 6.5 km during the SEAC RS campaign, J. Geophys. Res. Atmos., 121, 4188–4210, doi:10.1002/2015JD024498. Shingler, T., et al. (2016b), Ambient observations of hygroscopic growth factor and f(RH) below 1: Case studies from surface and airborne measurements, J. Geophys. Res. Atmos., 121, 13661–13677, doi:10.1002/2016JD025471. Sorooshian, A., A. Wonaschutz, E. G. Jarjour, B. I. Hashimoto, B. A. Schichtel, and E. A. Betterton (2011), An aerosol climatology for a rapidly growing arid region (southern Arizona): Major aerosol species and remotely sensed aerosol properties, J. Geophys. Res., 116, D19205, doi:10.1029/2011JD016197.

SCHLOSSER ET AL.


8965


10.1002/2017JD026547

Sorooshian, A., T. Shingler, A. Harpold, C. W. Feagles, T. Meixner, and P. D. Brooks (2013), Aerosol and precipitation chemistry in the southwestern United States: Spatiotemporal trends and interrelationships, Atmos. Chem. Phys., 13(15), 7361–7379, doi:10.5194/acp-13-73612013. Sorooshian, A., E. Crosbie, L. C. Maudlin, J. S. Youn, Z. Wang, T. Shingler, A. M. Ortega, S. Hersey, and R. K. Woods (2015), Surface and airborne measurements of organosulfur and methanesulfonate over the western United States and coastal areas, J. Geophys. Res. Atmos., 120, 8535–8548, doi:10.1002/2015JD023822. Spracklen, D. V., J. A. Logan, L. J. Mickley, R. J. Park, R. Yevich, A. L. Westerling, and D. A. Jaffe (2007), Wildfires drive interannual variability of organic carbon aerosol in the western US in summer, Geophys. Res. Lett., 34, L16816, doi:10.1029/2007GL030037. Stein, A. F., R. R. Draxler, G. D. Rolph, B. J. B. Stunder, M. D. Cohen, and F. Ngan (2015), NOAA’s HYSPLIT atmospheric transport and dispersion modeling system, Bull. Am. Meteorol. Soc., 96(12), 2059–2077. Sturtz, T. M., B. A. Schichtel, and T. V. Larson (2014), Coupling chemical transport model source attributions with positive matrix factorization: Application to two IMPROVE sites impacted by wildfires, Environ. Sci. Technol., 48(19), 11389–11396, doi:10.1021/es502749r. Tong, D. Q., M. Dan, T. Wang, and P. Lee (2012), Long-term dust climatology in the western United States reconstructed from routine aerosol ground monitoring, Atmos. Chem. Phys., 12(11), 5189–5205, doi:10.5194/acp-12-5189-2012. Vergnoux, A., M. Guiliano, R. Di Rocco, M. Domeizel, F. Theraulaz, and P. Doumenq (2011), Quantitative and mid-infrared changes of humic substances from burned soils, Environ. Res., 111(2), 205–214, doi:10.1016/j.envres.2010.03.005. Wang, Z., et al. (2016), Contrasting cloud composition between coupled and decoupled marine boundary layer clouds, J. Geophys. Res. Atmos., 121, 11679–11691, doi:10.1002/2016JD025695. Watson, J. G., J. C. Chow, D. H. Lowenthal, L. C. Pritchett, C. A. Frazier, G. R. Neuroth, and R. Robbins (1994), Differences in the carbon composition of source profiles for diesel-powered and gasoline-powered vehicles, Atmos. Environ., 28(15), 2493–2505, doi:10.1016/13522310(94)90400-6. Westphal, D. L., O. B. Toon, and T. N. Carlson (1988), A case-study of mobilization and transport of Saharan dust, J. Atmos. Sci., 45(15), 2145–2175, doi:10.1175/1520-0469(1988)0452.0.Co;2. White, W. H. (2008), Chemical markers for sea salt in IMPROVE aerosol data, Atmos. Environ., 42(2), 261–274, doi:10.1016/j. atmosenv.2007.09.040. Wonaschutz, A., S. P. Hersey, A. Sorooshian, J. S. Craven, A. R. Metcalf, R. C. Flagan, and J. H. Seinfeld (2011), Impact of a large wildfire on watersoluble organic aerosol in a major urban area: The 2009 station fire in Los Angeles County, Atmos. Chem. Phys., 11(16), 8257–8270, doi:10.5194/acp-11-8257-2011. Yamasoe, M. A., P. Artaxo, A. H. Miguel, and A. G. Allen (2000), Chemical composition of aerosol particles from direct emissions of vegetation fires in the Amazon Basin: Water-soluble species and trace elements, Atmos. Environ., 34(10), 1641–1653, doi:10.1016/S1352-2310(99) 00329-5. Yokelson, R. J., et al. (2009), Emissions from biomass burning in the Yucatan, Atmos. Chem. Phys., 9(15), 5785–5812. Youn, J. S., E. Crosbie, L. C. Maudlin, Z. Wang, and A. Sorooshian (2015), Dimethylamine as a major alkyl amine species in particles and cloud water: Observations in semi-arid and coastal regions, Atmos. Environ., 122, 250–258, doi:10.1016/j.atmosenv.2015.09.061. Zauscher, M. D., Y. Wang, M. J. K. Moore, C. J. Gaston, and K. A. Prather (2013), Air quality impact and physicochemical aging of biomass burning aerosols during the 2007 San Diego wildfires, Environ. Sci. Technol., 47(14), 7633–7643, doi:10.1021/es4004137.

SCHLOSSER ET AL.


8966

Effects of anthropogenic emissions on aerosol formation from isoprene and monoterpenes in the southeastern United States.

The Association between Dust Storms and Daily Non-Accidental Mortality in the United States, 1993-2005.

Unspeciated organic emissions from combustion sources and their influence on the secondary organic aerosol budget in the United States.

Modern dust aerosol availability in northwestern China.

Enhanced Volatile Organic Compounds emissions and organic aerosol mass increase the oligomer content of atmospheric aerosols.

Secondary organic aerosol formation from photo-oxidation of unburned fuel: experimental results and implications for aerosol formation from combustion emissions.

Aerosol and precipitation chemistry in the southwestern United States: spatiotemporal trends and interrelationships.

Attribution of the United States "warming hole": aerosol indirect effect and precipitable water vapor.

Correction to Supporting Information for Xu et al., Effects of anthropogenic emissions on aerosol formation from isoprene and monoterpenes in the southeastern United States.

Frequency and Character of Extreme Aerosol Events in the Southwestern United States: A Case Study Analysis in Arizona.

Quantifying the effect of organic aerosol aging and intermediate-volatility emissions on regional-scale aerosol pollution in China.

Western Cold and Flu (WeCoF) aerosol study--preliminary results.

Dust-wind interactions can intensify aerosol pollution over eastern China.

Composition and physical properties of the Asian Tropopause Aerosol Layer and the North American Tropospheric Aerosol Layer.

Lubricating oil dominates primary organic aerosol emissions from motor vehicles.

Particulate PAH emissions from residential biomass combustion: time-resolved analysis with aerosol mass spectrometry.

Aerosol, humidity and oxygenation.

Intravenous and aerosol salbutamol.

Assessment of primary and secondary ambient particle trends using satellite aerosol optical depth and ground speciation data in the New England region, United States.

Analysis of aerosol decay during breath-holding.

Metaproteomic analysis of atmospheric aerosol samples.

Characterization of aerosol optical properties, chemical composition and mixing states in the winter season in Shanghai, China.

Light absorption properties and radiative effects of primary organic aerosol emissions.

Natural airborne dust and heavy metals: a case study for kermanshah, Western iran (2005-2011).