Fossil fuel combustion and biomass burning sources of global black carbon

We identify sources (fossil fuel versus biomass burning) of black carbon (BC) in the atmosphere and in deposition using a global 3D chemical transport model GEOS-Chem. We validate the simulated sources against carbon isotope measurements of BC around the globe and find that the model reproduces biomass burning contribution (fbb, %) in various 10 regions within a factor of 2. GEOS-Chem shows that contribution from biomass burning in the Northern Hemisphere (fbb: 34%) is much less than that in the Southern Hemisphere (52%). Specifically, we find comparable contribution from biomass burning and fossil fuel in South Asia, S. America, S. Pacific, Australia and the Antarctic. fbb is the largest in Africa (64%), followed by that in East Asia (40%), Siberia (35%), the Arctic (33%), Canada (31%), the US (25%), and Europe (19%). fbb is higher in summer (59–78%) than in winter (28–32%) in the Arctic, while it is higher in winter (42–58%) and lower in 15 summer (16–42%) over the Himalayan–Tibetan plateau. The seasonal variations of fbb are relatively flat in North America, Europe, and Asia. We find that double biofuel emissions for domestic heating during cold seasons northern than 45°N increases fbb values in the Arctic and Europe in winter by ~30%, resulting in a ~20% reduction of discrepancies of fbb in the two regions. The remaining large negative discrepancies (Europe: 41%; Arctic: 46%) suggest that the biofuel emissions are probably still underestimated at high latitudes. Increasing fraction of thickly coated hydrophilic BC from 20% to 70% in 20 fresh biomass burning plumes increases the fraction of hydrophilic BC in biomass burning plumes by 0–20% (vary with seasons and regions), and thereby reduces atmospheric fbb by up to 11%. Faster aging (4 hour e-folding time versus 1.15 days of e-folding time) of BC in biomass burning plumes reduces atmospheric fbb by 7% (1– 14%), with the largest reduction in remote regions, such as the Arctic, the Antarctic and S. Pacific. Using size resolved scavenging accelerates scavenging of BC particles in both fossil fuel and biomass burning plumes, with a larger increase of the former. Thus, atmospheric fbb 25 increases in most regions by 1–14%. Overall, atmospheric fbb is determined by fbb in emissions mainly and by atmospheric processes, such as aging and scavenging, to a less extent.


Introduction
Black carbon (BC) in the atmosphere and deposited over snow and ice absorbs solar radiation, triggers positive feedbacks and exerts a large positive radiative forcing on the global climate (IPCC, 2014).Estimates of BC radiative forcing span a large range (0.2-1 W m -2 , Bond et al., 2013;IPCC, 2014).One of the uncertainties lies in the orders of magnitude different predictions of BC distributions around the globe, particularly in remote regions, by chemical transport and climate models (Samset et al., 2013;2014).To reduce the uncertainty, in addition to the widely used BC concentration observations in the troposphere, at surface and in snow, observation-based source apportionment (fossil fuel versus biomass burning) of BC provides another dimension to constrain model simulations of BC distribution.In addition, the optical properties of BC from 5 fossil fuel and biomass burning plumes are distinctively different (Bond et al., 2013), resulting in different radiative forcing from the two sources (Jacobson, 2010).Moreover, because of the relative short lifetime compared to greenhouse gases, accurate source apportionment of BC is important for short-term climate change mitigation.Thus, it is imperative to identify sources of BC globally.
Carbon isotope analysis is effective in separating emissions from fossil fuel combustion (e.g.coal, oil and natural gas) and 10 contemporary biomass burning (expressed as contribution from biomass burning, f bb , %), because fossil emissions are 14 C free and biomass emissions have a characteristic 14 C/ 12 C ratio that is proportional to atmospheric carbon dioxide at the time of carbon fixation (Reddy et al., 2002).Combining δ 13 C and Δ 14 C measurements further differentiate the contribution from coal and liquid fossil fuel combustion (oil, gasoline and diesel, Andersson et al., 2015 and references therein).Fossil fuel combustion often has an anthropogenic origin, including industrial use, domestic cooking and heating, and transport (Bond 15 et al., 2007).Contemporary biomass burning can from both anthropogenic and natural sources.The former includes mainly industrial and domestic burning of biofuels (fuelwood, charcoal, agricultural residues, and dung, Fernandes et al., 2007) and the latter involves open fires of forests, crops, grass, and peatlands (van der Werf et al., 2010).Carbon isotope measurements are widely used for source apportionment of BC in the atmosphere in South Asia (Gustafsson et al., 2009;Budhavant et al., 2015), East Asia (Chen et al., 2013;Andersson et al., 2015;Zhang et al., 2015;Li et al., 2016), Europe (Szidat et al., 2006;20 2009;Zhang et al., 2012) and the Arctic (Barrett et al., 2015;Winiger et al., 2015;2016;2017), in snow over the Himalayan-Tibetan Plateau (Li et al., 2016) and in ice core in Alpine (Jenk et al., 2006).
Previous studies (Gustafsson et al., 2009;Chen et al., 2013;Li et al., 2016) compare carbon isotope measurements directly to f bb of local bottom-up emission inventories.The assumption behind this comparison is that the major controlling factor of f bb in the atmosphere is local emissions.However, BC-containing particles in fossil fuel and biomass burning plumes have 25 distinctively different mixing states and hygroscopicities (Moteki et al., 2007;Schwarz et al., 2008;Shiraiwa et al., 2007;Akagi et al., 2012), which might further affect BC scavenging in the two kinds of plumes and thus f bb in the atmosphere and after deposition.Li et al. (2016) found smaller contribution from fossil fuel in snow than in air, suggesting that biomass burning emissions are easier to deposit compared to fossil fuel combustion emissions.Possible factors affecting f bb in the atmosphere and in deposition are mixing states and hygroscopicities in freshly emitted fossil fuel and biomass burning 30 plumes, the following aging rate and scavenging.However, as far as we are aware, no study so far has quantified the contribution of different factors to sources of global BC in the atmosphere and in deposition.
In this study, we simulate sources of BC (fossil fuel combustion versus biomass burning) using a global 3D chemical transport model GEOS-Chem.We describe the model and the carbon isotope measurements in Sections 2 and 3,

Model description
GEOS-Chem is a global chemical transport model driven with assimilated meteorological fields from the Goddard Earth 5 Observing System (GEOS) of the NASA Global Modeling and Assimilation Office.We use GEOS-Chem v11.01 coupled with the TwO Moment Aerosol Section (TOMAS) microphysics scheme (Adams and Seinfeld, 2002).We use 15 size bins ranging from 3 mn to 10 µm with tracers for sulphate, sea salt, organic aerosols, BC, and dust (Pierce et al., 2007;Lee et al., 2009;D'Andrea et al., 2013;Kodros and Peirce, 2017).MERRA2 meterological data set are used to drive model simulation at 4° latitude × 5° longitude resolution and 47 vertical layers from the surface to 0.01 hPa.Global fossil fuel and biofuel 10 combustion emissions of BC are from Bond et al. (2007) andFernandes et al. (2007).We replace BC emissions in Asia by Li et al. (2017).We use daily open fire emissions from Global Fire Emissions Database version 4 (GFED4, Giglio et al., 2013) in this study.We assume 20% of the freshly emitted BC aerosols are thickly coated and are hydrophilic (Park et al., 2003).We assume hydrophobic BC is converted to hydrophilic with an e-folding time of 1.15 days (Park et al., 2005).Wet deposition follows Liu et al. (2001), with updates in Wang et al. (2011) and Qi et al. (2017a).15 GEOS-Chem captures the probability density function (PDF) of annual BC concentrations at sites in the US, Europe, China and the Arctic (see site description in Qi et al., 2017(b)) but overestimates the frequency of low BC concentrations (Fig. 1S (a)).About 30% of the simulated annual BC concentration in air is underestimated by a factor of 2 (Fig. 1S (b)).The model reproduces the PDF of BC concentration in snow preferably (Fig. 2S (a)).The simulated median BC concentrations in snow in various regions agree with observations within a factor of 2, except in region NC_Northest Border (Fig. 2S (b)), where the 20 model overestimates the observed BC concentration in snow by a factor of 3 due to the overestimate of local emissions in that region (Qi et al., 2017b).

Observation data
Carbon isotope analysis of BC sources in the atmosphere is available at 41 sites across the globe in different seasons during 2002-2014 (Table S1 and Fig. S3).Generally, f bb values are larger in remote regions (54% in South Asia, 46% in the Arctic 25 and 39% over the Himalayan-Tibetan plateau) than those in urban regions (13% in North America and 29% in East Asia), indicating a larger contribution from biofuel and open fires in rural, developing and remote regions.
Isotope mass balance equation based on the Δ 14 C ( 14 C/ 12 C) data was applied to apportion the relative contributions to atmospheric BC from biomass burning of modern carbon (f bb ) and fossil fuel combustion.
Where Δ 14 C is the measured radiocarbon content of the BC component and Δ 14 C ff is -1,000‰ by definition because fossil carbon is completely depleted in radiocarbon.Δ 14 C bb end member used in this equation is usually between +70 and +225‰, depending on the type and age of the burned biomass.The former value corresponds to freshly produced biomass, such as 5 crop and grass.The latter value reflects the burn of wood, which has accumulated over the decades-to-century-long life span.
Different choice of the Δ 14 C bb end member is one of the uncertainties associated with this source apportionment method.
Uncertainty of ±25‰ translates to < 5% in the resulting f bb estimate (Winiger et al., 2016).Another uncertainty stems from the method of isolating BC from total carbon in sampled particles.Zhang et al. (2012) found that for thermal-optical method, the water-extraction treatment prior to thermal treatment and the heating temperature during thermal treatment are important 10 to the isolation of BC and organic carbon and the following isotope analysis.They found that different protocols of thermaloptical method lead to ~ 30% difference of estimated f bb values.

Contribution of biomass burning to BC in various regions
GEOS-Chem simulated f bb in the atmosphere agree with observations within a factor of 2 (Fig. 1 (a)).f bb values in Europe and 15 the Arctic are underestimated by a factor of 2, while f bb in North America is overestimated by a factor of 2. The low bias of f bb in Europe occurs in non-summer seasons (observation: 34%, model: 14%), which is partly due to the underestimate of biofuel combustion for domestic heating by current emission inventories in most of the European regions during cold seasons (Herich et al., 2011).The low bias of f bb in the Arctic can also be partly attributed to this underestimate of biofuel, since European emissions are large sources of BC in the Arctic (e.g.Qi et al., 2017c and references therein).Another reason 20 for the low bias of f bb in the Arctic is the underestimate of domestic biofuel usage in Russia as indicated by a recent study (Winiger et al., 2017).In North America, the model overestimates f bb at Salt Lake City and Mexican City by a factor of 2.
Possible reasons for the overestimate are explained in Sect.4.2.1.In E. Asia and S. Asia, the model reproduces the observed f bb values within 10%.In addition, GEOS-Chem underestimates the large variations of f bb values (error bars in Fig. 1 (a)) in the Arctic, Europe, and the Himalayan-Tibetan Plateau, due to the coarse horizontal and vertical resolutions.25 Over the Himalayan-Tibetan plateau, observations show that biomass burning dominates BC deposited in snow (64%), but its contribution in the atmosphere is much less (39%).GEOS-Chem reproduces the average f bb in snow perfectly (model: 63%) but overpredicts the average atmospheric f bb (model: 62%) by a factor of 1.6.GEOS-Chem simulated f bb values of BC deposition in snow at all sites over the Himalayan-Tibetan plateau agree with observations within 40% during both monsoon (June-August) and non-monsoon seasons (Fig. 1

Temporal variation of f bb
In the Arctic at Abisko, observed f bb ranges from fall and wintertime low of 31% to summer high of 59% (Fig. 2(a)), due to 5 the large contribution from open fires in Europe in summer (Winiger et al., 2016).The model also shows a peak of f bb in summer, but the seasonal variation is relatively flat (from 23% in winter to 27% in summer).We attribute the discrepancy to two reasons.First, fossil fuel and biofuel emissions (Bond et al., 2007;Fernando et al., 2007) used in this study are annual and lack temporal variations.Second, the coarse resolution does not solve the vortex structure of the low-pressure and frontal systems, which is important for poleward transport of BC (Ma et al., 2014;Sato et al., 2016).Thus, the transport of large 10 open burning emissions in Europe to site Abisko is probably underestimated.At Barrow (Fig. 2 (b)), the seasonal variation of atmospheric f bb is even stronger, from 17% in winter to 78% in summer.The similar magnitudes and variations of f bb in local emissions and the atmosphere in summer at the site suggest that the atmospheric f bb is largely determined by local emissions.In other seasons, the atmospheric f bb values are much larger than the f bb of local emissions, indicating a large contribution from long-range transport . 15 In contrast to the seasonal cycles of f bb at sites in the Arctic, at Bode (Fig. 2(c)) over the Himalayan-Tibetan Plateau, f bb values are the lowest in summer (17%) and highest in winter (42%).Similar trend is shown at Lumbini (Fig. 2(d)), only with smaller amplitude (summer low: 42%, spring high: 58%).The lower f bb in summer is because of several reasons.First, less biofuel is consumed for domestic heating in warmer seasons (Li et al., 2016).Second, the region is barely affected by open fires.Third, biomass-sourced BC is removed more efficiently by the frequent precipitation in summer both over the 20 Himalayan-Tibetan plateau and over the surrounding source regions, such as India and East Asia (Li et al., 2016).The GEOS-Chem simulated atmospheric f bb of BC at all sites over the Himalayan-Tibetan plateau (results for Bode and Lumbini are shown and the others are not) have weak or no seasonal variations.In addition, the model does not capture the observed increasing trend of f bb along the Mustang valley and Langtang valley.Possible reasons for the discrepancies are several folds.First, the relative contributions of fossil fuel and biofuel emissions in the model have no seasonal cycle.Second, it is 25 conceivable that the coarse model resolution of global models does not reproduce the complex topography and transport pathways of BC over the Himalayan-Tibetan plateau (He et al., 2014).However, the mean modeled atmospheric f bb generally agrees with observations (within 60%) and the modeled atmospheric f bb generally follows the f bb of local emissions across the whole plateau.These comparisons suggest that the atmospheric f bb over the Himalayan-Tibetan plateau is largely determined by f bb in emissions in the region . 30 In North America (Fig. 2 loadings in winter and low loadings in summer (Mouteva et al., 2017;Yamamoto et al., 2007;Budhavant et al., 2015).In Salt Lake City in North America, the most significant local sources of PM 2.5 particles are mobile emissions, which are relatively stable through the whole year.The second most important source is area sources with solid fuel burning, which is not allowed to use when air quality forecasts predict an inversion period (Mouteva et al., 2017).This restriction limits the extra use of solid fuels in winter, and thus limited their effects on BC concentrations and f bb in the atmosphere.So the higher 5 concentration of BC in winter in Salt Lake City is largely determined by the low boundary layer height, which is proved to have limited effects on f bb (Mouteva et al., 2017).The model overestimates f bb in all seasons by a factor of 2-4.Modeled f bb in the atmosphere is much higher than the f bb values of local emissions, suggesting a large regional effect on f bb in this region.However, the observations were in urban environment with strong influence from local emissions.This mismatch of model representation and observations partly explains the large positive bias of f bb .In Tokyo, East Asia, the model 10 reproduces both the magnitude and the seasonal variations of observed f bb .The much lower f bb value in emissions than in the atmosphere also indicates a regional effect.In South Asia, GEOS-Chem reproduces the observed seasonal f bb within 30%.
Comparison of f bb in local emissions and in the atmosphere suggest that f bb at MCOH is largely affected by long-range transport, while f bb at SINH is mostly affected by local emissions.At MCOH the high f bb is probably from the large f bb in the outflow of Africa, while at SINH local burning of agricultural crop residues are the major sources (Budhavant et al., 2015).15

Spatial variation of f bb
The Southern Hemisphere has a higher contribution from biomass burning both for BC in surface air (52%) and in deposition (55%, Fig. 3 (a) and (b)).The high f bb in S. America and Australia are largely from active open fires (accounting for 48% and 81% of the total biomass burning contributions, respectively), while in Africa biofuel consumption is the major biomass burning source (64%, Fig. 3 (c) and (d)).Because of the strong seasonal variations of open fire emissions, the 20 highest f bb in Africa, S. America, S. Pacific, Australia and the Antarctic usually occur during September to November (58-71%), and the lowest values are in March-May (32-56%, Fig. S3).
In the Northern Hemisphere, the largest f bb of both BC in the atmosphere (54%) and in deposition (54%) are in South Asia, where biomass burning contribution dominates over fossil fuel emissions.East Asia has the second largest f bb (41% for BC in air and 43% for BC in deposition) due to large biofuel consumption.In other regions, such as Europe, Canada, the US, 25 Siberia and the Arctic, fossil fuel contribution (65-80%) is much larger than biomass burning.f bb of BC in air and in deposition in different regions have different seasonal variations (Figs.S4-S5).Canada, Siberia, and the Arctic have the strongest seasonal variation with a peak in summer (49-55%) because of the large contribution from open fire emissions (Fig. S6-S7).In the US, Europe, East Asia and South Asia, seasonal variation of f bb is relatively flat, consistent with observations (Fig. 2).30

Uncertainty associated with biofuel emissions
Biofuel emission estimates are associated with large uncertainties (Fernandes et al., 2007).Source apportionment of BC in Europe based on multi-wavelength aethalometer measurements showed that f bb in winter (24-33%) is much higher than that in summer (2-10%), suggesting that wood burning for domestic heating increases the f bb value in the atmosphere in winter 5 significantly (Herich et al., 2011).In addition, Winiger et al. (2017) analyzed f bb based on carbon isotope measurements at Tiksi in Russia and suggested that domestic wood burning is an important source of BC, emissions of which are underestimated by current bottom-up emission inventories.Thus, we doubled biofuel emissions from domestic heating northern than 45°N during cold seasons in Exp. A. It is conceivable that the largest effects occur in the Northern four regions, including Europe, Siberia, Canada and the Arctic.Specifically, f bb values increase by ~30% in Europe, Siberia and 10 the Arctic and by 15% in Canada in winter, larger than that in spring and fall (4-13%, Fig. 4).Consequently, the model discrepancies of f bb in Europe and the Arctic are reduced from -54% and -56% to -41% and -46%, respectively.This improvement suggests that the biofuel emissions at high latitudes in the Northern Hemisphere are probably too low in current bottom-up BC emission inventories, supporting previous estimates (Herich et al., 2011;Winiger et al., 2017).

Uncertainty associated with hygroscopicity of BC in freshly emitted biomass burning plumes 15
Recent measurements find that in fossil fuel plumes the fraction of thickly coated hydrophilic BC is ~10% (Moteki et al., 2007;Schwarz et al., 2008;Shiraiwa et al., 2007), while in biomass burning plumes the fraction reaches up to 70% (Schwarz et al., 2008;Akagi et al., 2012).The higher hygroscopicity of BC in freshly emitted biomass burning plumes enhances the subsequent wet scavenging rate and thereby reduces f bb in the atmosphere.We investigate the effects of the initial hygroscopicity of BC in fresh emissions on atmospheric f bb of BC in Exp.B by assuming 70% of freshly emitted BC 20 particles from biomass burning are thickly coated and hydrophilic.The resulting fraction of hydrophilic BC in biomass burning plumes in the 12 regions increase by 0-20% (vary with seasons and regions), lowering f bb in the atmosphere by up to 11%.The largest reduction of f bb shows in June-August (-7%, Fig. 4), when open fires are frequent and active globally (Giglio et al., 2013;van der Werf et al., 2010).During this time, the largest reductions are in Canada (-11%) and Siberia (-10%), where the fraction of hydrophilic BC in biomass burning plumes increases by a large fraction (11-13%).In S. Pacific, 25 the reduction of f bb is large (-10%) as well, because large precipitation over this region removes more biomass burning BC particles in the outflow of S. America.During September-November, reduction of f bb in the Northern Hemisphere (-6%) is much larger than that in the Southern Hemisphere (-1%), because f bb values in the Southern Hemisphere are too large (Fig. S5).The changes of f bb values in other seasons are marginal.

Uncertainty associated with BC aging time
Mixing with organic and inorganic particles with larger hygroscopicity, BC particles become more hydrophilic during aging process (Bond et al., 2013).It is assumed that BC particles are converted from hydrophobic to hydrophilic with an e-folding time of 1.15 days after emission in the standard simulation (Park et al., 2005).However, observations showed that the fraction of thickly coated hydrophilic BC in urban fossil fuel plumes increases linearly with plume age (0.5-2.3% h -1 , Moteki 5 et al., 2007;Shiraiwa et al., 2007;Subramanian et al., 2010;McMeeking et al., 2011), while BC aging follows a logarithmic trend with an e-folding time of 4 hours in biomass burning plumes (Akagi et al., 2012).The aging rates differ among plumes because of different BC sizes, co-emitted hygroscopic materials and oxidation capacities of the plumes (Bond et al., 2013).
Thus, in Exp.C, we assume fossil fuel combustion generated BC ages linearly with a rate of 1% h -1 , while BC from biomass burning plumes ages with an e-folding time of 4 hours.This means that the fossil fuel plumes age slower than the standard 10 simulation and be scavenged slower, while the biomass burning plumes age much faster and are removed from the atmosphere faster in precipitation.This aging scheme leads to a 0-24% increase of fraction of hydrophilic BC in the atmosphere, which reduces f bb by up to -14%.The largest reduction of f bb is in S. Pacific in fall (MAM) and summer (DJF) in the Southern Hemisphere, followed by the Antarctic (-12%) during MAM and the Arctic (-11%) during SON.The reduction of f bb is larger in remote regions and smaller in source regions, because it takes time for the different aging rates in fossil fuel 15 and biomass burning plumes to affect the hygroscopicities of BC in the two plumes and the subsequent aging rates.

Uncertainty associated with size resolved scavenging
BC particles emitted from biomass burning plumes are usually larger in size and thicker in coating thickness (Schwarz et al., 2008;Sahu et al., 2012), suggesting an easier removal from the atmosphere.Observations show that the mass median diameter of BC particles in biomass burning plumes is 193 nm with a coating thickness of 65 nm, while in fossil fuel plumes, 20 the mass median diameter and coating thickness are 175 nm and 20 nm (Schwarz et al., 2008;Sahu et al., 2012).In addition, because of the different coating materials, hygroscopicities of BC-containing particles in the two kinds of plumes are different as well.The coating materials of BC in urban plumes are dominated by sulfate and followed by nitrate and primary and secondary organics (Shiraiwa et al., 2007), while in biomass burning plumes, the major coating materials are organics (Sahu et al., 2012).For ambient air, characteristic κ values of organics and inorganics are 0.1 (0.01-0.5) and 0.7 (0.5-1.4, 25 Petters and Kreidenweis, 2007;Gunthe et al., 2011 and references therein).Thus, the hygroscopicity and the following wet scavenging of BC particles in fossil fuel and biomass burning plumes are different.In Exp.D, we use the TOMAS microphysics scheme to process the aging and wet scavenging of BC with different sizes from fossil fuel combustion and biomass burning.The mass median diameters of fossil fuel and biomass burning BC particles are assumed to be 160 nm and 200 nm, respectively.Size resolved coagulation, condensation, nucleation and cloud processing are implemented.Coating 30 materials included are sulfate, nitrate, sea-salt, organics and mineral dust.The size-resolved aging and scavenging scheme leads to a larger increase of fraction of hydrophilic BC in fossil fuel plumes (by 16% (0-31%, vary with regions)) than in biomass burning plumes (by 12% (0-23%)), suggesting that fossil fuel BC particles are removed faster.Thus atmospheric f bb increases in most regions during MAM (by 1-14%), SON (by 0-7%) and DJF (by 1-12%).The most noticeable characteristics is that the increase of f bb in Northern Hemisphere is larger than those in Southern Hemisphere, due to the large fraction of fossil fuel emissions in the Northern Hemisphere.

Conclusions 5
This study sought to understand the relative contribution of fossil fuel combustion and biomass burning to global BC.We used GEOS-Chem (version 11-01-01) driven by MERRA2 assimilated meteorological fields to simulate BC concentration from fossil fuel and biomass burning.The source apportionment results were expressed as the fraction of BC from biomass burning (f bb ).Simulated f bb was validated against carbon isotope measurements of BC in the atmosphere at 41 stations the Northern Hemisphere and 11 snow samples over the Himalayan-Tibetan plateau.We also investigated the uncertainties 10 of f bb associated with biofuel emissions, fraction of hydrophilic BC in fresh emissions, aging time and size-resolved scavenging.
The model reproduced the mean observed atmospheric f bb in various regions and in snow over the Himalayan-Tibetan plateau within a factor of 2. Generally, values of atmospheric f bb were larger in remote regions (54% in South Asia, 46% in the Arctic and 39% over the Himalayan-Tibetan plateau) than those in urban regions (13% in North America and 29% in 15 East Asia), indicating a larger contribution from biofuel and open burning sources in rural, developing and remote regions.
f bb was higher in summer (59-78%) than in winter (28-32%) in the Arctic, while it was higher in winter (42-58%) and lower in summer (16-42%) over the Himalayan-Tibetan plateau.The simulated amplitudes of the seasonal variations were much smaller in the two regions.The seasonal variation was observed to be relatively flat in North America, Europe, East and South Asia.The model reproduced the monthly mean f bb in these regions within 30%.The Southern Hemisphere had a 20 higher atmospheric f bb than the Northern Hemisphere (SH: 52%, NH: 34%) due to the large fraction of open burning emissions in S. America and Australia and large fraction of biofuel consumption in Africa.In the Northern Hemisphere, the highest f bb was in South Asia (54%), followed by East Asia (41%), due to large biofuel consumption.In other regions, such as Europe, Canada, the US, Siberia and the Arctic, f bb values are small (20-35%).
Simulated f bb was associated with uncertainties from all processes, including emissions, aging and deposition.We found that 25 doubled biofuel emissions used for domestic heating northern than 45°N resulted in a ~30% increase of f bb in Europe, Siberia and the Arctic and a 15% increase in Canada in winter.This increase reduced the discrepancy of f bb against observations by 18-23%, suggesting that the biofuel emissions at high latitudes were underestimated by the bottom-up emission inventories.
Using a higher fraction of hydrophilic BC in fresh biomass burning plumes (uncertainty simulation: 70%, standard simulation: 20%) resulted in a reduction of f bb in summer by -2 --11%, with the largest reduction in Canada and Siberia,30 where open fires were frequent.In the standard simulation, it was assumed that BC in both fossil fuel and biomass burning Atmos.Chem.Phys.Discuss., https://doi.org/10.5194/acp-2019-46Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2019 c Author(s) 2019.CC BY 4.0 License.respectively.We evaluate the model simulation of f bb in Section 4.1, analyze the spatial and temporal variations of f bb in Section 4.2, evaluate the uncertainties associated with f bb in BC emissions, BC mixing state and hycroscopicity in fresh emissions, aging rage and size-resolved scavenging in Section 4.3.
(b)), suggesting that the model captures the spatial and temporal variations 30 of f bb in BC deposition in this region.The overestimate of the atmospheric f bb is mainly from the factor of 2.3 overestimate of Atmos.Chem.Phys.Discuss., https://doi.org/10.5194/acp-2019-46Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2019 c Author(s) 2019.CC BY 4.0 License.f bb during monsoon season (observation: 29%, model: 67%).Possible reasons for the overestimate are discussed in Sect.4.2.1.
plumes aged following an e-folding time of 1.15 days.In the uncertainty simulation, we used a 4 hour e-folding life time for Atmos.Chem.Phys.Discuss., https://doi.org/10.5194/acp-2019-46Manuscript under review for journal Atmos.Chem.Phys.Discussion started: 3 April 2019 c Author(s) 2019.CC BY 4.0 License.

Figure 1 :
Figure 1: Observed and GEOS-Chem simulated fraction of biomass burning (f bb , %) of (a) BC in the atmosphere in the Arctic, South Asia, North America, Europe, East Asia, and the Himalayan-Tibetan plateau (the regions are symbol-coded) and (b) BC in snow during monsoon (red) and non-monsoon (black) seasons over the Himalayan-Tibetan plateau.Also shown in (a) are the standard deviations of observed and model simulated f bb in each region (error bars).Observations of f bb in the atmosphere in (a) 5

Figure 2 :
Figure 2: Seasonal variations of observed (lightcoral bars) and GEOS-Chem simulated (lightblue bars) f bb of BC in the atmosphere 10

Figure 4 : 5 definition
Figure 4: GEOS-Chem simulated fractional change (r) to atmospheric f bb relative to the standard simulation, as a result of doubled biofuel emissions northern than 45°N (Exp.A), 70% of hydrophilic BC of freshly emitted biomass burning BC-containing particles (Exp.B), 4 hour e-folding aging time of BC in biomass burning plumes and linear aging rate of 1% in fossil fuel plumes (Exp.C) and TOMAS microphysical aging and scavenging (Exp.D), r = ([BC] Exp.-[BC] Std.)/[BC] Std., that varies with regions (see region 5