Smoke from biomass burning is a major source of atmospheric primary aerosol and vapors (Akagi et al., 2011; Gilman et al., 2015; Hatch et al., 2015, 2017; Jen et al., 2019; Koss et al., 2018; Reid et al., 2005; Yokelson et al., 2009), influencing air quality, local radiation budgets, cloud properties, and climate (Carrico et al., 2008; O'Dell et al., 2019; Petters et al., 2009; Ramnarine et al., 2019; Shrivastava et al., 2017) as well as the health of impacted communities (Ford et al., 2018; Gan et al., 2017; Reid et al., 2016). Dilution of a smoke plume occurs as the plume travels downwind, mixing with regional “background” air, reducing the concentrations of smoke aerosols and vapors, and potentially driving changes in the physical and chemical properties of the emissions (Adachi et al., 2019; Akagi et al., 2012; Bian et al., 2017; Cubison et al., 2011; Hecobian et al., 2011; Hodshire et al., 2019a, b; Jolleys et al., 2012, 2015; Konovalov et al., 2019; May et al., 2015; Noyes et al., 2020; Sakamoto et al., 2015, Palm et al., 2020). Fires span an immense range in size, from small agricultural burns, which may be only a few square meters in total area and last a few hours, to massive wildfires, which may burn tens of thousands of kilometers over the course of weeks (Andela et al., 2019). This range in size leads to variability in initial plume size and extent of dilution by the time of the first measurement. Plumes can dilute unevenly, with edges of the plume mixing in with surrounding air more rapidly than the core of the plume. Hence overall, these large, thick plumes dilute more slowly than small, thin plumes for similar atmospheric conditions as the cores of larger plumes are at a greater physical distance to the background air, shielding them from dilution for longer (Akagi et al., 2012; Bian et al., 2017; Cubison et al., 2011; Hecobian et al., 2011; Hodshire et al., 2019a, b; Jolleys et al., 2012, 2015; Konovalov et al., 2019; May et al., 2015; Sakamoto et al., 2015; Lee et al., 2020; Garofalo et al., 2019). Variability in dilution leads to variability in the evolution of smoke emissions as instantaneous plume aerosol concentrations will control shortwave radiative fluxes (and thus photolysis rates and oxidant concentrations), gas-particle partitioning, and particle coagulation rates (Akagi et al., 2012; Bian et al., 2017; Cubison et al., 2011; Hecobian et al., 2011; Hodshire et al., 2019a, b; Jolleys et al., 2012, 2015; Konovalov et al., 2019; May et al., 2015; Sakamoto et al., 2015; Garofalo et al., 2019; Ramnarine et al., 2019; Sakamoto et al., 2016). Thus, capturing variability in plume aerosol concentrations and dilution between fires and within fires can aid in understanding how species change within the first few hours of emission for a range of plume sizes.

The evolution of total particulate matter (PM) or organic aerosol (OA) mass from smoke has been the focus of many studies as PM influences both human health and climate. Secondary organic aerosol (SOA) production occurs through oxidation of gas-phase volatile organic compounds (VOCs) that can form lower-volatility products that partition to the condensed phase (Jimenez et al., 2009; Kroll and Seinfeld, 2008). SOA formation may also arise from heterogeneous and multi-phase reactions in both the organic and aqueous phases (Jimenez et al., 2009; Volkamer et al., 2009). In turn, oxidant concentrations depend on shortwave fluxes (Tang et al., 1998; Tie, 2003; Yang et al., 2009) and the composition of the plume (Yokelson et al., 2009; Akagi et al., 2012; Hobbs et al., 2003; Alvarado et al., 2015). Smoke particles contain semivolatile organic compounds (SVOCs) (Eatough et al., 2003; May et al., 2013), which may evaporate off of particles as the plume becomes more dilute (Huffman et al., 2009; May et al., 2013; Garofalo et al., 2019; Grieshop et al., 2009), leading to losses in total aerosol mass. Field observations of smoke PM and OA mass normalized for dilution (e.g., through a long-lived tracer such as CO) report that for near-field (<24 h) physical aging, net PM or OA mass can increase (Cachier et al., 1995; Formenti et al., 2003; Liu et al., 2016; Nance et al., 1993; Reid et al., 1998; Vakkari et al., 2014, 2018; Yokelson et al., 2009), decrease (Akagi et al., 2012; Hobbs et al., 2003; Jolleys et al., 2012, 2015; May et al., 2015), or remain nearly constant (Brito et al., 2014; Capes et al., 2008; Collier et al., 2016; Cubison et al., 2011; Forrister et al., 2015; Garofalo et al., 2019; Hecobian et al., 2011; Liu et al., 2016; May et al., 2015; Morgan et al., 2020; Sakamoto et al., 2015; Sedlacek et al., 2018; Zhou et al., 2017). It is theorized that both losses and gains in OA mass are likely happening concurrently in most plumes through condensation and evaporation (May et al., 2015; Hodshire et al., 2019a, 2019b; Bian et al., 2017; Palm et al., 2020), with the balance between the two determining whether net increases or decreases or no change in mass occurs during near-field aging. However, there is currently no reliable predictor of how smoke aerosol mass concentration (normalized for dilution) may change for a given fire.

Evolution of total aerosol number, size, and composition is critical for improving quantitative understanding of how biomass burning smoke plumes impact climate. These impacts include smoke aerosols' abilities to both act as cloud condensation nuclei (CCN) and to scatter or absorb solar radiation (Albrecht, 1989; Petters and Kreidenweis, 2007; Seinfeld and Pandis, 2006; Twomey, 1974; Wang et al., 2008). Particles can increase or decrease in size as well as undergo compositional changes through condensation or evaporation of more volatile compounds. In contrast, coagulation always decreases total number concentrations and increases average particle diameter. Plumes with higher aerosol number concentrations will undergo more coagulation than those with lower concentrations (Sakamoto et al., 2016).

Fires in the western United States region are predicted to increase in size, intensity, and frequency (Dennison et al., 2014; Ford et al., 2018; Spracklen et al., 2009; Yue et al., 2013). In response, several large field campaigns have taken place in the last 7 years examining wildfires in this region (Kleinman et al., 2020; Garofalo et al., 2019; Palm et al., 2020). Here, we present smoke plume observations from the Biomass Burning Observation Project (BBOP) campaign of aerosol properties from five research flights sampling wildfires downwind in seven pseudo-Lagrangian sets of transects to investigate the evolution of OA mass and oxidation state, aerosol number, and aerosol number mean diameter. A range of initial (at the time of the first plume pass in the aircraft) plume OA mass concentrations were captured within these flights, and fast (1 s) measurements of aerosols and key vapors were taken. The time resolution of the data was fast enough to segregate each transect into edge, core, or intermediate regions of the plume and examine aerosol properties within the context of both the location within the plume (edge, core, or intermediate) and the initial OA mass loading of the given location. The differences in aerosol loading serve as a proxy for differences in initial fire and plume sizes, mass fluxes, and subsequent amount of dilution. The extent to which dilution has occurred prior to the first observation is not a measurable quantity, and fire sizes and mass fluxes were not estimated as a part of the BBOP campaign. We create mathematical fits for predicting OA oxidation markers and mean particle diameter given initial plume OA mass concentration and physical age (time) of the smoke. These fits may be used to evaluate other smoke datasets and assist in building parameterizations for regional and global climate models to better predict smoke aerosol climate and health impacts.

The BBOP field campaign occurred in 2013 and included a deployment of the United States Department of Energy Gulfstream 1 (G-1) research aircraft in the Pacific Northwest region of the United States (Kleinman and Sedlacek, 2016; Sedlacek et al., 2018) from 15 June to 13 September. We analyze five cloud-free BBOP research flights that had seven total sets of across-plume transects that followed the smoke plume downwind in a Lagrangian manner (see Figs. S1–S6 for examples; Table S1) from approximately 15 min after emission to 2–4 h downwind (Kleinman and Sedlacek, 2016). The G-1 sampling setup is described in Kleinman and Sedlacek (2016), Sedlacek et al. (2018), and Kleinman et al. (2020).

Number size distributions were obtained with a fast-integrating mobility spectrometer (FIMS), providing particle size distributions nominally from approximately 20–350 nm (Kulkarni and Wang, 2006; Olfert and Wang, 2009); data were available between 20–262 nm for the flights used in this study. A soot photometer aerosol mass spectrometer (SP-AMS) provided organic and inorganic (sulfate, chlorine, nitrate, ammonium) aerosol mass concentration of PM1 (sub-micron aerosol) (Canagaratna et al., 2007), select fractional components (the fraction of the AMS OA spectra at a given mass-to-charge ratio) (Onasch et al., 2012), and elemental analysis (O/C and H/C) (Aiken et al., 2008; Canagaratna et al., 2015). Extended details on the SP-AMS are provided in Sect. S1 in the Supplement, and a brief overview is given here. The SP-AMS had its highest sensitivity between 70–500 nm, dropping to 50 % of peak sensitivity by 1000 nm (Liu et al., 2007). It was characterized to have a collection efficiency of 0.5 when the instrument's laser was off and 0.76 when the instrument's laser was on during the BBOP campaign, and these corrections have been applied to the data. There is evidence from other studies that the collection efficiency (CE) of the tungsten vaporizer (laser-off mode) (Lim et al., 2019) and the laser vaporizer (laser-on mode, run nominally at 600 ∘C) (Willis et al., 2014) changes as a function of chemical composition, refractory black carbon (rBC) coating thickness, size, and sphericity in laboratory studies (Middlebrook et al., 2012; Willis et al., 2014; Corbin et al., 2015; Massoli et al., 2015; Collier et al., 2018) and in aircraft observations (Kleinman et al., 2007). Results pertinent to changes in CE due to aging (including physical aging as well as chemical changes including oxidation, coating thickness, and sphericity) in smoke plumes are scarce (see discussion in Kleinman et al., 2020). We assume these CEs for the laser-on and laser-off modes are constant in space and time, which is a limitation of this study. We use the calculated f60 and f44 fractions (the unit mass resolution mass concentration ratios of m/z60 and 44 normalized by the total OA mass concentration) and O/C and H/C elemental ratios of OA as tracers of smoke and oxidative aging. Elevated f60 values are indicative of “levoglucosan-like” species (levoglucosan and other molecules that similarly fragment in the AMS) (Aiken et al., 2009; Cubison et al., 2011; Lee et al., 2010) that are known tracers of smoke primary organic aerosol (POA) (Cubison et al., 2011); f44, the OA fractional component observed by the SP-AMS as the high-resolution ion fragment CO2+ as well as some acid groups, is a proxy for SOA arising from oxidative aging (Alfarra et al., 2004; Cappa and Jimenez, 2010; Jimenez et al., 2009; Volkamer et al., 2006). Fractional components f60 and f44 have been shown to decrease and increase with photochemical aging, respectively, likely due to both evaporation and/or oxidation of semivolatile species that contribute to m/z60 in the SP-AMS and addition of oxidized species that contribute to m/z44 in the SP-AMS (Alfarra et al., 2004; Huffman et al., 2009). O/C tends to increase with oxidative aging (Decarlo et al., 2008), whereas H/C ranges from increasing to decreasing with oxidative aging, depending on the types of reactions occurring (Heald et al., 2010). Changes in O/C and H/C (as well as changes in total OA mass, number, f44, and f60) are also influenced by mixing of different air masses and co-oxidation of different VOC precursors (Chen et al., 2015). Tracking H/C with aging may provide clues on the types of reactions that may be occurring; however, variable oxidation timescales can make inferences of this type difficult (Chen et al., 2015). A single-particle soot photometer (SP2; Droplet Measurement Technologies) was used to measure refractory black carbon (BC) between 80–500 nm (Schwarz et al., 2010) through laser-induced incandescence (Moteki and Kondo, 2010; Schwarz et al., 2006). An off-axis integrated-cavity output spectroscopy instrument (Los Gatos, Model 907) measured CO concentrations. An SPN1 radiometer (Badosa et al., 2014; Long et al., 2010) measured total shortwave irradiance. Kleinman et al. (2020) provides extensive details for the BBOP instruments used in this work. The supporting information also includes more details on the instruments used.

To determine the contribution to the concentration of species X from smoke emissions (ΔX), the background concentration of X is subtracted from the measured in-plume species concentrations. To correct for dilution, we normalize ΔX by background-corrected CO (ΔCO), which is inert on timescales of near-field aging (Yokelson et al., 2009). Increases or decreases in ΔX/ΔCO along the Lagrangian flight path indicate whether the total amount of X in the plume has increased or decreased (implying production or removal) since time of emission. The background concentration of X is determined as a regional average of the observed out-of-plume concentrations of X. To avoid using smoke-impacted measurements we apply a threshold of only using measurements of X that occur in regions that correspond to the lowest 10 % of CO data. We determine the lowest 10 % of CO concentrations for each flight during time periods with a similar altitude, latitude, and longitude as the smoke plume. We perform sensitivity calculations on our assumptions of background regions and discuss them in Sect. 3.

Mass concentrations of O, H, and C are calculated using the O/C and H/C and OA data from the SP-AMS (assuming all of the OA mass is from O, C, and H, and we acknowledge that omitting lower-abundance atoms, such as S and N, contributes to some errors in this assumption), allowing us to calculate the background-corrected OA atomic ratios, ΔO/ΔC, and ΔH/ΔC, following Eq. (1) (where X=O or H): 1ΔXΔC=(Xinplume-Xoutofplume)(Cinplume-Coutofplume). We note that any non-linear changes in chemistry and composition between the plume and background will not perfectly isolate the elemental factors in smoke. We also background-correct fractional f60 and f44 (using the mass concentrations of m/z60, m/z44, and OA inside and outside of the plume), but we do not normalize by CO due to these values already being normalized by OA, following Eq. (2) (where f=f60 or f44): 2Δf=(fin⋅OAin)-(foutOAout)ΔOA. We only consider data to be in-plume if the absolute CO≥150 ppbv. This threshold appears to be capturing clear plume features as seen in the number concentration while excluding background air (Figs. S7–S11). We note that we use different definitions of in-plume and background (i.e., the lowest 10 % of absolute CO measurements) in order to provide a buffer between the plume and background to ensure to the best of our abilities that we are capturing non-smoke-impacted air for the background and smoke-impacted air for in-plume cases. The regions of the lowest 10 % of CO measurements always fall under 150 ppbv (Figs. S7–S11). Similarly, we exclude the lowest 5 % of CO data in the in-plume measurements in our analyses to provide a further buffer between smoke-impacted and background air. We perform sensitivity analyses of our results to our assumptions about background and in-plume values in Sect. 3. Figures S2–S6 indicate the locations of the lowest 10 % of CO for each flight.

From the FIMS, we examine the background-corrected, normalized number concentrations of particles with mobility diameters between 40–262 nm, ΔN/ΔCO. This size range allows us to exclude potential influence of fresh nucleation upon the total number concentrations. Occasionally, the background-corrected, normalized number concentration in the FIMS size range between 20–40 nm increases by 1–2 orders of magnitude relative to typical plume conditions, indicating possible nucleation events, primarily at the edges or in between smoke plumes (Figs. S7–S11). Smoke plumes contain particles with diameters larger than 262 nm (Janhäll et al., 2010); thus, we cannot provide total number concentrations, but we can infer how ΔN/ΔCO within our observed size range evolves. We also obtain an estimate of how the number mean diameter between 40–262 nm, Dp‾, changes with aging through 3Dp‾=ΣNi⋅Dp,iΣNi, where Ni and Dp,i are the number concentration and geometric mean diameter within each FIMS size bin, respectively.

All of the data are provided at 1 Hz, and all but the SP-AMS fractional component data are available on the US Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) web archive (https://www.arm.gov/research/campaigns/aaf2013bbop, last access 29 April 2021). As the plane traveled at approximately 100 m s-1 on average, the approximate spatial resolution of the data is every 100 m across the plume. The plumes spanned from approximately 5–50 km wide (Figs. S2–S6). The instruments used here had a variety of time lags (all <10 s) relative to a TSI 3563 nephelometer used as a reference. The FIMS also showed additional smearing in flushing smoky air with cleaner air when exiting the plume, with maximum observed flushing timescales around 30 s but generally less (Fig. S12). To test if these lags impact our results, we perform an additional analysis where we only consider the first half of each in-plume transect, when concentrations are generally rising with time (Figs. S12, S13), and our main conclusions are unaffected. We do not test the impacts of other time lags and do not attempt to further correct the data for any time lags. Kleinman et al. (2020) provide further information on instrument time delays during BBOP.

We use MODIS Terra and Aqua fire and thermal anomaly detection data to determine fire locations (Giglio et al., 2006, 2008). We estimate the fire center to be the approximate center of all clustered MODIS detection points for a given sampled fire (Figs. S1–S6). The true fire location at the time of sampling is likely different than the MODIS estimates, depending on the speed of the fire front. To estimate the physical age of the plume, we use the estimated fire center as well as the total FIMS number concentration to determine an approximate centerline of the plume as the smoke travels downwind (an example is provided in Fig. S1). The centerline is subjectively chosen to approximately capture the most-concentrated portion of each plume pass (as estimated using total aerosol number concentrations). We use the mean wind speed and this estimated centerline to calculate an estimated physical age for each transect, and this physical age is assumed to be constant across the transect as plume crossings took between 50–500 s; however, transects that were not perfectly tangential to the mean wind would have sampled different plume ages on the opposite sides of the plume. We did not propagate uncertainty in fire location, wind speed, or centerline through to the physical age, which is a limitation of this study.

As a case example, we examine the aging profiles of smoke from the Colockum fire during the first set of pseudo-Lagrangian transects for flight 730b (Table S1). Figure 1 provides ΔOA/ΔCO, ΔBC/ΔCO, Δf60, Δf44, ΔH/ΔC, ΔO/ΔC, ΔN/ΔCO, and Dp‾ as a function of the estimated physical age; Figs. S14–S18 provides this information for the other pseudo-Lagrangian transect flight sets studied. (Here, BC represents the refractory BC from the SP2; Sect. 2.) We have divided each transect into four regions: between the 5th–15th (edge), 15th–50th (intermediate, outer), 50th–90th (intermediate, inner), and 90th–100th (core) percentile of ΔCO within each transect. (As discussed above, we exclude the lowest 5 % in order to provide a buffer between the plume edge and background air.) Note that in Fig. 1 (and Figs. S14–S18), the points represent the mean values for each transect and percentile and do not include error bars for uncertainty in the mean or measurement uncertainty as characterization of systematic variance (within plume percentiles) with age is beyond the scope of this study. Figures S2–S6 show the locations of these CO percentile bins for each transect of individual flights. Figure 1 shows the edge and core data, both averaged per transect, and Figs. S14–S18 provide all four percentile bins for each flight. These percentile bins correspond with the thinnest (lowest CO mixing ratio) to thickest (highest CO mixing ratio) portions of the plume, respectively. If a fire has uniform emissions ratios across all regions and dilutes evenly downwind, these percentile bins would correspond to the edges, intermediate outer and intermediate inner regions, and the core of the diluting plume. We use this terminology in this study but note that uneven emissions, mixing, and/or dilution lead to the percentile bins not physically corresponding to our defined regions in some cases. We note that some plumes show more than one maxima in CO concentrations within a given plume crossing, which implies that there may be more than one fire or fire front and that these plumes from separate fires or fronts are not mixing perfectly. Multiple maxima could also imply vertical variations in the location of the core of the plumes that the flights did not capture. Additionally, in at least one of the fires (in flights “730a” and “730b”), the fuels vary between different sides of the fire, as discussed in Kleinman et al. (2020). However, the lowest two ΔCO bins tend towards the physical edges of the plume, and the highest two tend more towards the physical center of the plume (Figs. S2–S6). We do not know where the plane is vertically in the plume, which is a limitation as vertical location will also impact the amount of solar flux able to penetrate through the plume.

Figure 1 shows that for this specific plume, ΔOA/ΔCO and ΔBC/ΔCO systematically vary little with age for both the 5th–15th and 90th–100th percentile of ΔCO (p values >0.5), yet both show non-systematic variability between transects. A true Lagrangian flight with the aircraft sampling the same portion of the plume and no measurement artifacts (e.g., coincidence errors at high concentrations) would have a constant ΔBC/ΔCO for each transect set. This flight and other flights studied here have variations in ΔBC/ΔCO (Figs. 1, S14–S18), which may be indicative of deviations from a Lagrangian flight path with temporal variations in emission and/or measurement uncertainties. The remaining variables plotted also show some noise and few clear trends, but it is apparent that the transect-mean values' 5th–15th and 90th–100th percentiles do show a separation for some of the individual metrics, in particular Δf44 and ΔO/ΔC. In order to determine the existence or lack of trends for these metrics, we spend the remainder of this study examining each metric from all of the pseudo-Lagrangian flights together.

The BBOP field campaign provided high-time-resolution (1 s) measurements of gas- and particle-phase smoke measurements downwind of western US wildfires along pseudo-Lagrangian transects. These flights have allowed us to examine near-field (<4 h) aging of smoke particles to provide analyses on how select species vary across a range of initial organic aerosol mass loadings (ΔOAinitial; a proxy for the relative rates at which the plume is anticipated to dilute as dilution before the first observation is not a measurable quantity). We have also examined how the species studied vary between the edges and cores of each plume. We find that although ΔOA/ΔCO does not correlate with ΔOAinitial or physical age, Δf60 (a marker for evaporation) is moderately correlated with ΔOAinitial (Spearman rank-order correlation test correlation coefficient, RΔOA,initial, of +0.43) and weakly correlated with physical age (Spearman rank-order correlation test correlation coefficient, Rage, of -0.26); Δf44 and ΔO/ΔC (markers for photochemical aging) increase with physical aging (moderate correlations of Rage of +0.5 and +0.56, respectively) and are inversely related to ΔOAinitial (moderate correlations of RΔOA,initial of -0.55 and -0.45, respectively). ΔN/ΔCO decreases with physical aging, likely through coagulation. Mean aerosol diameter increases with age primarily due to coagulation as normalized organic aerosol mass does not change significantly and is moderately correlated with physical age (Rage=+0.53). Nucleation is observed within a few of the fires and appears to occur primarily on the edges of the plumes. Differences in initial values of Δf60, Δf44, and ΔO/ΔC are evidence that evaporation and/or chemistry has occurred before the time of initial measurement and that plumes or plume regions with lower initial aerosol loading can undergo these changes more rapidly than thicker plumes. We have developed fit equations that can weakly to moderately predict Δf60, Δf44, ΔO/ΔC, and mean aerosol diameter given a known initial (at the time of first measurement) total organic aerosol mass loading and physical age. We were unable to quantify the impact on potential inter-fire variability in the emission values of the metrics studied here (such as variable emissions of species that can contribute to m/z60 and m/z44). We anticipate that being able to capture this additional source of variability may lead to stronger fits and correlation. We encourage researchers to attempt to quantify these chemical and physical changes in future studies before the initial measurement using combinations of modeling and laboratory measurements, where sampling is possible at the initial stages of the fire and smoke. We also suggest further refinement of our fit equations as additional variables (such as photolysis rates) and better quantification of inter-fire variability (such as variable emission rates) are anticipated to improve these fits. We finally suggest future near-field (<24 h) analyses of recent and future biomass burning field campaigns to include differences in initial plume mass concentrations and location within the plume as considerations for understanding chemical and physical processes in plumes.