Spatiotemporal variability of elemental and organic carbon in Svalbard snow during 2007–2018

Light-absorbing carbonaceous aerosols emitted by biomass or fossil fuel combustion can contribute to amplify Arctic climate 20 warming by lowering the albedo of snow. The Svalbard archipelago, being near to Europe and Russia, is particularly affected by these pollutants, and improved knowledge of their distribution in snow is needed to assess their impact. Here we present and synthesize new data obtained on Svalbard between 2007 and 2018, comprising 324 measurements of elemental (EC) and organic carbon (OC) in snow from 49 sites. We used these data, combined with meteorological and aerosol data and snowpack modelling, to investigate the variability of EC and OC deposition in Svalbard snow across latitude, longitude, elevation and 25 time. Overall, EC concentrations (CCssssssss EEEE ) ranged from <1.0 to 266.6 ng g-1, while OC concentrations (CCssssssss OOEE ) ranged from <1.0 to 9449.1 ng g-1, with the highest values observed near Ny-Ålesund. Calculated snowpack loadings (LLssssssss EC , LLssssssss OC ) in April 2016 were 0.1 to 16.2 mg m-2 and 1.7 to 320.1 mg m-2, respectively. The median CCssssssss EEEE and LLssssssss EEEE in the late 2015‒16 winter snowpack on glaciers were close to or lower than those found in earlier (2007–09), comparable surveys. Both LLssssssss EEEE and LLssssssss OOEE increased exponentially with elevation and snow accumulation, with dry deposition likely playing a minor role. Estimated 30 area-averaged snowpack loads across Svalbard were 1.8 mg EC m-2 and 71.5 mg OC m-2 in April 2016. An ~11-year long dataset of spring surface snow measurements from central Brøgger Peninsula was used to quantify the interannual variability of EC and OC deposition in snow. On average, CCssssssss EEEE and CCssssssss OOEE at Ny-Ålesund (50 m a.s.l.) were 3 and 7 times higher, respectively, than on the nearby Austre Brøggerbreen glacier (456 m a.s.l.), and the median EC/OC in Ny-Ålesund was 6 times higher, pointing to some local EC emission from Ny-Ålesund. While no long-term trends between 2011 and 2018 were found, 35 CCssssssss EEEE and CCssssssss OOEE showed synchronous variations at Ny-Ålesund and Austre Brøggerbreen. Comparing CCssssssss EEEE at Austre https://doi.org/10.5194/acp-2020-491 Preprint. Discussion started: 16 June 2020 c © Author(s) 2020. CC BY 4.0 License.

Brøggerbreen with aerosol data from Zeppelin Observatory, we found that snowfall washout ratios between 10 and 300 predict a range of in agreement with that measured in surface snow. Together, results from this study and comparable surveys confirm the existence of a longitudinal gradient in EC deposition across the Arctic and sub-Arctic, with the lowest found in the western Arctic (Alaska, Yukon) and central Greenland, and the highest in northwestern Russia and Siberia. 40

Introduction
Light-absorbing carbonaceous aerosols, such as black carbon (BC) or "brown carbon" (BrC; organic) that are transported to Arctic latitudes can lower the albedo of snow/ice-covered surfaces on which they are deposited, thereby enacting a positive feedback that amplifies climate warming (Bond et al., 2013). The Svalbard archipelago, owing to its proximity to the European and Russian mainland, is particularly affected by BC and BrC emissions from fossil fuel combustion (FF; heating, gas flaring, 45 etc.) and biomass burning (BB; e.g., agricultural or forest fires). Source attribution using carbon isotopes and atmospheric transport modelling indicates that BC associated with pollution haze events at the Zeppelin Observatory on Spitsbergen include both BB and FF contributions, the latter being proportionally more important in winter than summer (Winiger et al., 2015(Winiger et al., , 2019. Quantifying the impact of BC deposition on the Arctic surface albedo requires knowledge of its concentrations, spatial distribution and variability in snow and ice. These data may also serve to verify the efficacy of ongoing and future measures 50 to curb emissions of short-lived climate forcing aerosols, such as BC, that impact the Arctic (AMAP, 2015;Stohl et al., 2015).
On Svalbard, reconnaissance surveys of BC in snow were carried out in 1984-85 by Noone and Clark (1988) and in 2007 by Doherty et al. (2010). This was followed in 2007-09 by more detailed investigations of the distribution of BC across the archipelago (Forsström et al., 2009. Localized studies have also been carried out near Longyearbyen Khan et al., 2017) and Ny-Ålesund (Sihna et al., 2018;Jacobi et al., 2019). In addition, two ice cores recovered from 55 the Lomonosovfonna and Holtedahlfonna icefields (Spitsbergen) have provided insights into longer-term variations in BC deposition on Svalbard (Ruppel et al., 2014Osmont et al., 2018).
Here we present and synthesize new observational data which document the variability of BC in snow across Svalbard in terms of latitude, longitude, altitude and time. These data were gathered through field investigations conducted between and how it relates to spatial variations in snow accumulation, (iii) describe the interannual variability of EC and OC concentration in snow on Brøgger Peninsula between 2007 and 2018, and (iv) constrain plausible estimates of the snowfall 70 washout ratio for EC. Lastly, we place our findings in a broader pan-Arctic perspective by comparison with a compilation of published data obtained between 2002 and 2018 by comparable methods. Part of the dataset presented here was produced following a comprehensive, coordinated survey of the physical, chemical and microbiological properties of the Svalbard seasonal snowpack carried out at the end of the 2015-16 winter by individuals from multiple institutions (see acknowledgments). In total, 22 sites were sampled between 4 and 29 April 2016 on 7 glaciers of Spitsbergen and Nordaustlandet (Fig. 1). Snowpits were sampled at three different elevations in the upper, middle and lower reaches of each glacier (Table S1), the snow depth increasing with altitude (e.g., Pramanik et al., 2019). Glacier sites were 80 targeted partly for logistical reasons (ease of access by snowmobile), but also because sampling supraglacial snow avoided the heterogeneities in snow properties that may arise from interactions with vegetation and/or different substrates (e.g., wet vs. dry tundra soils). In addition, the selected glacier sites were at elevations of 102 to 1193 m a.s.l., which span ~65 % of the maximum relief in Svalbard (1713 m; Fig. 2).
In advance of the field campaign, standardized protocols were developed for the measurement of important snow 85 physical properties (e.g., density, temperature) and the collection of samples for a variety of analyses, including EC, OC, and stable oxygen isotope ratios (δ 18 O). These protocols are documented in Gallet et al. (2018), and details relevant for this paper summarized hereafter. Snow sampling was performed in snowpits excavated down to the hard, icy firn surface representing the previous year's late summer ablation surface (in the accumulation zone of glaciers), or to the underlying bare ice surface (in the ablation zone). All snowpits were located well away from point sources of contamination (e.g., field camps), were 90 accessed by foot from at least 100 m, and personnel doing the sampling wore protective, non-particulating suits, gloves, and face masks, and employed pre-cleaned plastic or stainless steel tools. The snow accumulation of each snowpit (hSWE, in water equivalent; w.e.) was calculated from discrete density measurements. After recording the physical properties of snow strata, large volume snow samples (~5 L each) were collected from the top 5 cm of the snowpack, and, where snowpack depth allowed for it, at 50-cm depth intervals beneath. The near-surface samples were collected to quantify EC and OC concentrations in 95 snow ( and ) at depths where light absorption by carbonaceous particles has the largest impact on snow albedo (Marks and King, 2013 were double-bagged in sterile low-density polyethylene bags and returned frozen to a location where they were subsequently 100 melted and filtered. Depending on logistics, this was done either at the Polish Polar Station Hornsund operated by the Polish Academy of Sciences, or at the Norwegian Polar Institute (NPI) facilities in Ny-Ålesund (Sverdrup station), Longyearbyen (UNIS) or Tromsø. A total of 89 samples were obtained from all 22 sites.
Analysis of downscaled ERA Interim climatological fields (Dee et al., 2011) over Svalbard show that surface temperatures in the 2015-16 winter exceeded the 30-year climatological normals for the 1981-2010 period by 2 to 6 °C, with 105 the largest anomalies observed in the northeastern part of the archipelago (Fig. S1a). Total winter precipitation also exceeded 1981-2010 normals by 0.2 to 0.7 m w.e. over much of central and northern Spitsbergen (Fig. S1b). These unusual conditions arose partly owing to an extreme winter warming and precipitation event associated with a southerly air intrusion over Spitsbergen that occurred in late December 2015 (Binder et al., 2017;Kim et al., 2017;Maturilli et al., 2017). The implications of these climatological circumstances for the interpretation of our snow survey data are discussed later. 110

Surface snow monitoring, Brøgger Peninsula
In addition to the April 2016 survey, we report measurements of and in surface layers sampled by NPI staff from three sites on Brøgger Peninsula between 2007 and 2018 (Fig. S2). The first of these sites is in the accumulation zone of Austre the Gruvebadet Atmospheric Laboratory (78.92° N, 11.89° E, ~50 m a.s.l.). Sampling was carried out at approximately weekly intervals by the NPI permanent staff at Sverdrup station, whenever their work schedule made it possible, and when safe snowmobile driving conditions (e.g., proper visibility, firm surface) allowed access to Austre Brøggerbreen. Because of these restrictions, the snow samples could not always be collected immediately after snow fall events. Over the ~11-year period considered, a total of 201 samples were collected between February and June, 86 % of which were taken in the spring months 120 (March-May), April being the most represented month (n = 44). Methods for field sample collection were the same as those described above for the April 2016 survey. Sample collection was limited to the top 5 cm of the snowpack (occasionally deeper). These data provide long-term estimates of the interannual variability of and in Svalbard against which results of the April 2016 survey (and others) can be compared.
We also report additional and measurements in snow collected on 6 glaciers in northwestern Spitsbergen 125 (n = 34; Table S2). These samples were collected irregularly, on an opportunistic basis, by NPI staff during other field research activities, but were handled and analyzed in the same manner as all those previously described. Altogether, the dataset presented here comprises a total of 324 measurements of and from 49 separate sites across Svalbard.

EC and OC analyses
All snow samples used in this study were processed in the same way: they were first melted at room temperature and the 130 meltwater was filtered through pre-ashed, 47-mm diameter quartz microfiber filters, following the procedure described in Forsström et al. (2009). The filters were then air-dried at room temperature overnight, stored in sterile petri dishes, and later sent to the Department of Environmental Science of Stockholm University. There, EC/OC analysis was performed using a Sunset Laboratory carbon aerosol analyzer (Sunset Laboratory Inc., Forest Grove, USA), following the European Supersites for Atmospheric Aerosol Research thermal evolution protocol (EUSAAR_2; Cavalli et al., 2010). A 1 x 1 cm 2 square section 135 was used from each filter to determine separately the particulate EC and OC mass loading on each filter ( , ), from which their mass concentrations in snow ( , ) were calculated based on the volume of meltwater filtered. Blank filters (n = 6) had particulate carbon loadings below the limit of detection (LOD) of the carbon analyzer, so no blank correction was applied to the data. The coefficient of variation (CV) on and was estimated to be ~40 % (see Supplement for details). 140 The presence of mineral dust particles in snow can lead to understimations of relative to by the TOT method (Wang et al., 2012;Lim et al., 2014). A total of 31 snow filters obtained from 7 glaciers surveyed in April 2016 (35 % of samples) were found to have faint to pronounced yellow-pink or grey-brown coloration, likely indicating the presence of k-feldspars and/or oxides which are commonly found in cryoconites, although carbonates may also be present on these filters (see notes in Supplemental dataset). The samples that produced the colored filters were typically found in snow layers near the 145 base of snowpits, suggesting windblown dust dispersion and deposition in the autumn when the ground is only partially snowcovered. In 11 of the colored filters, the measured were noticeably lower than in filters from snow layers immediately above, and in 6 filters the was < LOD, possibly due to the effect of dust on TOT measurements. Some surface snow filters obtained from Sverdrup or Gruvebadet near Ny-Ålesund between 2010 and 2018 also showed coloration indicating the probable presence of windblown dust. Correcting for the effects of dust on the OC-EC split point of individual TOT 150 thermograms is feasible, but it requires experimental data which we did not have (e.g., Wang et al., 2012). Therefore, no such corrections were applied to and data. We acknowledge, however, that the in snow samples that contained visible dust may be underestimated.
For the snowpits excavated on glaciers in April 2016, we computed mass loadings of EC ( EC ) and OC ( OC ) in the seasonal snowpack following: 155 (2) 160 where EC and OC are in mg m -2 , ρi is the mean density of snow layer i, zi its thickness, and n the number of discrete layers. For samples which yielded <1 ng g -1 we assigned a value of 0.5 ng g -1 (half the LOD) in order to compute snowpack loadings (see below). An estimated error on individual density measurements (σρ) of ± 6 % was used (Conger and McClung, 2009;Proksch et al., 2016), and the meter-scale variability of snow layer density at spatial scales of 1 to 100 m 2 was assumed to be on the order of 5 %, after Koenig et al. (2016). Combining uncertainties on and with these errors 165 yields a median CV of ~30 % for and (n = 22 snowpits).

δ 18 O analyses
The stable isotope ratio of oxygen ( 16 O: 18 O) in snowpit samples collected in April 2016 was used to detect evidence of warming events associated with large autumn or winter snowfalls, that could help to interpret the and data. The measurements were made at the Institute of Geology of Tallinn's University of Technology, Estonia, using a Picarro model 170 L2120-i water isotope analyzer (Picarro Inc., Sunnyvale, USA) (Lis et al., 2008). Results are reported in the standard delta notation δ 18 O relative to the Vienna Standard Mean Ocean Water. The analytical precision was ±0.1 ‰.

Surface meteorological observations
Automated weather stations (AWS) were operated on 6 glaciers sampled during the April 2016 survey (Table S3). These 175 stations were situated close to the estimated equilibrium line altitude (ELA) of the glaciers, and provided hourly recordings of air temperature and ultrasonic soundings of snow surface height changes that were used to interpret snowpit stratigraphic data, in particular the timing of snow accumulation and of snowmelt events. Data from the AWSs were supplemented with records from Longyearbyen and the airport in Ny-Ålesund obtained from the Norwegian Meteorological Institute, and from the Polish Polar Station Hornsund (N 77.00°, E 15.11°, 9 m a.s.l.). 180

Snowpack modeling
Owing to the scarcity of direct precipitation measurements across Svalbard, reconstructing the snowpack accumulation history is challenging, and estimates from snowpits, probing and radar can only fill some of the spatial and temporal gaps. This difficulty can be partly circumvented by using the output of a snowpack model forced with meteorological observations (e.g., Jacobi et al., 2019). In this study, we use a coupled energy balance-snow model ( for the surface energy balance as well as internal snowpack processes (e.g., densification, melt-freeze events) that makes it possible to simulate the evolution of the seasonal snowpack (thickness and internal structure) for individual 1x1 km grid cells over Svalbard. The snow model routine simulates subsurface density, temperature and water content, while accounting for vertical water transport, liquid water storage, refreezing and runoff (Marchenko et al. 2017;Van Pelt et al. 2019). Fresh snow 190 density is described by a temperature-and wind-dependent function (Van Kampenhout et al. 2017), while snow densification is the sum of destructive metamorphism, compaction by overburden pressure and compaction by drifting snow (Vionnet et al. 2012). Snow scouring and redistribution by wind is not accounted for, however. Layered snow properties are modelled with a vertical resolution of 1 cm.
Here, we used the model to simulate the snowpack evolution at some of the sites sampled during the April 2016 195 survey. For the April 2016 survey snowpits, simulations were limited to those sites located close to or above the local ELA (  Reistad et al., 2009). For all modelled sites, precipitation was locally calibrated (scaled with a factor) to assure matching modelled and observed snow depths at the time of observation (April 2016). The snowpack model was also used to characterize the spatial variability of the seasonal snow cover over Brøgger 200 Peninsula for the period 2008-18 during which surface snow was sampled for EC and OC (Fig. S3).

Black carbon aerosol measurements
The atmospheric mixing ratio of BC above Svalbard follows a well-defined seasonal cycle, peaking in late winter and early spring (Eleftheriadis et al., 2009). To establish how the timing of surface snow sample collection on Brøgger Peninsula compares with this seasonal cycle, and also to estimate BC washout ratios, we obtained aerosol data from the Zeppelin 205 Observatory (N 78° 54.43', E 11° 53.20', 474 m a.s.l.), 2 km south of Ny-Ålesund, over the period 2007-18. The data used are filter-based measurements of the hourly mean aerosol light absorption coefficient (σap) made with a Particle Soot Absorption Photometer (PSAP; λ = 525 nm; Bond et al., 1999) or an aethalometer (λ = 880 nm; Eleftheriadis et al., 2009). These data were used to calculate the hourly BC mass-equivalent mixing ratio in air ( , in ng m -3 ; Petzold et al., 2013), following: where σap is in m -1 , MACλ is the wavelength-specific BC aerosol mass absorption coefficient cross-section (m 2 g -1 ), and Cf = 3.45 is a unitless correction factor accounting for light absorption in the filter matrix (Backman et al., 2017). We used an MAC525 value of 12.5 m 2 g -1 for the PSAP data, and a MAC880 value of 15.9 m 2 g -1 for the aethalometer data, after Eleftheriadis

Results
Descriptive statistics of and for all samples analyzed in this study are summarized in Table 2. The probability distributions of and both have skewness >4, therefore we use medians (̃, ̃) as measures of their central tendency, but also report arithmetic and geometric means for comparisons with other published data. As both and 220 are left-censored by the analytical LOD, the median and mean were estimated by replacing values < 1 ng g -1 with 0.5 x LOD (0.5 ng g -1 ), following Hornung and Reed (1990), while the geometric mean was estimated by the beta factor method of Ganser and Hewett (2010). Values of and < LOD are, however, included in plots (see below) to provide an as complete as possible description of our data. Overall, ranged from <1.0 to 266.6 ng g -1 , while ranged from <1.0 to 9449.1 ng g -1 . The highest (>50 ng g -1 ) were measured in spring surface snow near Ny-Ålesund (Sverdrup and Gruvebadet 225 sites). The ̃ at these two sites over the period 2007-18 (9.7 ng g -1 ) was ~4 times higher than in surface layers at glacier  Hansbreen and Werenskioldbreen also contained the most layers with >100 ng g -1 .
~1100 m altitude range of the 22 glacier sampling sites (Kruskal-Wallis test, p >0.1). The calculated were between 0.1 and 16.2 mg m -2 with a median of 0.8 mg m -2 (mean 2.0 mg m -2 ), while were between 1.7 and 320.1 mg m -2 , with a 250 median of 20.5 mg m -2 (mean 49.3 mg m -2 ). The median EC/OC was only marginally higher (0.06) at glacier sites <200 m a.s.l., compared to higher elevations (range: 0.03-0.04) (Fig. S7). As with ̃ and ̃, there were no discernible patterns of variation of or with respect to geographic location ( Fig. S8 and S9). On most glaciers, and/or increased with elevation along with hSWE. This was most noticeable on Kongsvegen (northwestern Spitsbergen; highest sampling site at 672 m a.s.l.) for both or . 255 : <1-1076.1 ng g -1 ) overlapped with that near Ny-Ålesund ( : <1-266.5 ng g -1 ;

Surface snow monitoring, Brøgger Peninsula
: <1-7250.3 ng g -1 ; 2 outliers 260 excluded; Fig. 5a,b). However on average (all years and months considered), ̃ near Ny-Ålesund was 3 times higher than on Austre Brøggerbreen, while ̃ was 7 times higher, but as much as 30 (Fig. 5d). However, variations in ̃ or EC/OC between 2007 and 2018 did not correlate with those in the median measured at Zeppelin Observatory, or with regional monthly snowfall anomalies (Fig. 5e,f). 270

EC and OC in the winter 2015-16 snowpack across Svalbard
The April 2016 survey showed no discernible zonal or latitudinal gradient of or across Svalbard. As noted earlier, only on Austfonna was ̃ significantly lower than in some sectors of Spitsbergen. This contrasts with findings from surveys made in the springs of 2007-09, in which ̃ on Austfonna snow was either comparable to, or larger than, that in 275 central or northeastern Spitsbergen (Forsström et al., 2009. Based on data from sites where direct comparisons with the 2007-09 surveys are possible, ̃ in the seasonal snowpack varies, from year to year, by at least one order of magnitude, and sometimes more (Fig. 3). The snowpack on glacier sites that are highest and/or further inland (e.g., Lomonosovfonna https://doi.org/10.5194/acp-2020-491 Preprint. Discussion started: 16 June 2020 c Author(s) 2020. CC BY 4.0 License. summit) had generally lower than at low-elevation, near-coastal sites. Ice-free open water areas or frost flowers on sea ice are potential sources of particulate OC aerosols (e.g., microbes, diatoms, plankton, exopolymers from biofilms) during 280 autumn and winter, some of which are likely deposited in snow by settling or through ice nucleation (Bowman and Deming, 2010;Campbell et al., 2018;Karl et al., 2019). The quantity of these aerosols deposited in Svalbard snow might be expected to decrease with inland distance and altitude, which is consistent with our observations of .
Comparing annual mean of 10 mg m -2 using snow samples and a firn core from Holtedahlfonna spanning ~8 years . The corresponding mean in the late winter (end April) snowpack could be less than half of this value (~5 mg m -2 ), but the 290 high interannual variability in net snow accumulation at this site (Pramanik et al., 2014;Van Pelt and Kohler, 2015), and the uncertainty in the chronology of the firn core makes such an estimate tentative at best. Differences between our estimates of and those from the 2007-09 surveys probably reflect, to a large extent, the variability of atmospheric EC transport and deposition between years and seasons, but also in space (local scale; Svensson et al., 2013).
The estimated and in the April 2016 snowpack were generally largest at higher elevations on glaciers, 295 where snow accumulation is greater ( Fig. S8 and S9). From equations (1) and (2) above, it is expected that and should increase non-linearly with hSWE, since the mass of EC in OC in the snowpack is the cumulative sum of the product of (or ) by some fraction of hSWE in discrete snow layers. For the winter 2015-16 snowpack, we modelled the relationship between and hSWE across all snowpits with an exponential function ( Fig. 6a; R 2 = 0.86; RMSE = 0.59). The goodness-of-fit was confirmed through a Kolmogorov-Smirnov (KS) test of normality on the standardized model residuals (α 300 = 0.05, p = 0.2). A similar model applied to against hSWE gave a poorer fit ( Fig. 6b; R 2 = 0.73; RMSE = 7.26; KS test: p = 0.32), owing to greater scatter in the data. The estimated by Forsström et al. (2009Forsström et al. ( , 2013 from glacier surveys in 2007-09 showed a much poorer correlation with hSWE than the April 2016 estimates (Fig. 6a). This may partly reflect methodological differences in the estimation of both and hSWE between these studies, and the fact that the estimates from Forsström et al. (2009Forsström et al. ( , 2013  We applied the exponential models for and from AWSs at several glaciers (Fig. S10) indicate that snow accumulation in the 2015-16 winter was more or less equally divided between the autumn period leading to the late December 2015 snowstorm, and the months that followed up to mid-/late April 2016, when the snowpits were sampled. The snowpack model simulations, forced with downscaled HIRLAM precipitation 335 data, gave similar results (Fig. 8). The AWSs also show that the December 2015 storm saw winter temperatures on nearly all glaciers rise above 0°C for several days, the warming being largest in southern Spitsbergen. Clear evidence for this was found in a >0.2 m thick icy snow layer at mid-depth in the snowpack on Hansbreen (site HB3). The depth of the layer is in good agreement with that predicted by the snowpack model at this site, showing that the simulation provides a reasonable estimate of local surface conditions. Icy layers also occurred in the lower half of the seasonal snowpack on other glaciers, but none of 340 these could be unambiguously ascribed to the late December 2015 storm period.
The timing of EC and OC accumulation, inferred from the snowpack model chronology, varied considerably between glaciers (Fig. 9). On Austfonna in Nordaustlandet, ~80 % of the EC and OC was found in snow layers estimated to have been deposited in or after December 2015. On glaciers of northern and central Spitsbergen (Austre Lovénbreen, Kongsvegen, Holtedahlfonna and Lomonosovfonna), the accumulation sequence was more variable and differed between EC and OC. On Hansbreen and Werenskioldbreen in southern Spitsbergen, most of the EC and OC was contained in the deeper layers of the seasonal snowpack, estimated to have been deposited prior to January 2016. Surface meteorological records from the Polish Polar Station Hornsund and from an AWS on Werenskioldbreen show that several large snowfall events occurred in this area during the autumn of 2015, as well as some thaw events (Fig. S10) Large cyclonic storms that reach Svalbard in December typically track from the south-southeast (Rinke et al., 2017), and often make their landfall on southern Spitsbergen, bringing heavy snowfall and relatively warm, moist air, as for example 355 on 19 December 2015 (Hancock et al., 2018). Such events may enhance wet deposition of EC and OC aerosols on local glaciers such as Hansbreen and Werenskioldbreen, relative to more northerly sectors of Spitsbergen that are located further along the storm track. Previous observations of acidic snowfall deposition on Hansbreen during periods of southerly polluted air advection support this interpretation (Nawrot et al., 2016). Furthermore, meltwater percolation during surface thaws (or rainon-snow) can redistribute or concentrate some of the more hydrophilic EC and OC into icy layers near the base of the snowpack 360 Xu et al., 2012). These circumstances could explain why the and were particularly high in the accumulation zones of Hansbreen and Werenskioldbreen in April 2016, and also why >65 % of the accumulated EC was found in the deepest part of the snowpack at these sites. It also suggests that a few large precipitation events could explain a large part of the interannual variability in and across Svalbard, as was previously observed for wet deposition of SO4 2-, NO3and OC on glaciers near Hornsund (Kühnel et al., 2013;Kozioł et al., 2019). local EC pollution in coastal snow within 20 km of Ny-Ålesund, but as these authors pointed out, this could have been due to 380 unfavourable wind conditions at the time of their sampling. Our data, however, suggest that that winter/spring surface snow near Ny-Ålesund is commonly enriched in EC relative to OC when compared to snow deposited higher up on Austre Brøggerbreen, as shown by ~4-fold differences in the median EC/OC between these sites ( Table 2).
The median EC/OC in snow at Ny-Ålesund and Austre Brøggerbreen was <0.10 prior to 2010, but rose to 0.43 and 0.20, respectively, in 2015, and declined after 2016. The seasons with highest median EC/OC (2013-15) were also those with 385 lowest ̃ and ̃, which implies that meteorological and/or other conditions must have prevailed which limited atmospheric deposition of OC and EC in snow, OC being more affected than EC. Possible causal factors include sea ice cover or sea-surface winds, which partly modulate emissions of marine organic aerosol (e.g., Kirpes et al., 2019), or katabatic winds from Kongsfjorden, that can affect the thermal stratification of boundary layer air in winter months (Esau and Repina, 2012;Maturilli and Kayser, 2017). 390 Between 2008 and 2018 (the years in which snow sampling was most thorough), ̃ on Brøgger Peninsula varied by up to 35 ng g -1 , and ̃ by up to 689 ng g -1 . However, there were no significant trends in either ̃ or ̃ over the whole period (Mann-Kendall test; p >> 0.05). Data from the April 2016 glacier survey (Fig. 6), as well as previous studies (Bourgeois and Bey, 2011;Browse et al., 2012) suggest that wet deposition is the predominant mode of EC deposition in Arctic snow. To control for the possible role of snowfall rate, we compared ̃ in surface layers for March, April and May 395 (MAM) 2008-18 with simulated monthly snowfall anomalies over central Brøgger Peninsula over the same period. However, as mentioned earlier, no significant correlation was found, (Fig. 5f).
A simple metric used to quantify wintertime wet deposition of BC is the snow washout ratio W (sometimes called scavenging ratio), and defined as ⁄ , which accounts for atmospheric BC removal by snowfall through both withinand below-cloud nucleation scavenging (Noone and Clark, 1988). Previous observations made during springtime snowfall 400 events on Svalbard have yielded a wide range of estimates for W of ~13 to 2000, with a median of ~380 at Zeppelin Observatory, and of ~60 to 600 in Ny-Ålesund (Hegg et al., 2011;Gogoi et al., 2018). Here, we used our data on in spring snow at Austre Brøggerbreen, combined with measured at Zeppelin Observatory, to constrain plausible estimates of W (Fig. 10). Specifically, we compared the probability distribution of in surface layers at Austre Brøggerbreen with that calculated from at Zeppelin Observatory during the same time periods, in order to identify values of W that produced 405 an optimal match. The calculation was: https://doi.org/10.5194/acp-2020-491 Preprint. Discussion started: 16 June 2020 c Author(s) 2020. CC BY 4.0 License.
where ̅ is the estimated mean concentration of eBC in the top 5 cm of the snowpack for the dates on which snow samples 410 were actually obtained, t is the number of days prior to the snow sampling date over which the computed values of were averaged, W is the washout ratio, ρair the air density and f is a precipitation weighing factor. We performed this calculation for 54 snow sampling days during the spring months of 2011-18, which are the months (years) in which the data had the fewest missing values. The simulated snowpack at Austre Brøggerbreen summit was used to estimate, for each day, the value of t (averaging interval) based on prior changes in the snow surface height. This varied between 3 and 24 days. The weighing 415 factor f was calculated from changes in hSWE for days on which the simulated snowpack surface rose by ≥ 1 mm (i.e., on days without snowfall, f was 0). Monthly mean values of ρair were used, calculated from Ny-Ålesund radiosonde data (Maturilli, 2011 and seq.).
An overlap between the calculated ̅ and measured was obtained for 10 ≤ W ≤ 300 (Fig. 10). These values agree well with those reported by Gogoi et al. (2018) for Ny-Ålesund (~13-270), but are low compared to those of Hegg et al. 420 (2011) for Zeppelin Observatory (~250-2000. There are, however, several sources of uncertainty in the calculated ̅ . First, the calculation uses measured at Zeppelin Observatory (474 m a.s.l.), whereas aerosols are likely scavenged by snowfall over a range of altitudes with variable BC mixing ratios. Second, the calculation neglects dry deposition contributions.
Third, the simulated snow surface changes at Austre Brøggerbreen that were used to define the averaging period t for do not account for the possible effect of snow drifting and mixing. Consequently, the comparison presented on Fig. 10 should be 425 considered with some caution. What the results imply is simply that it is possible to forecast a realistic range of eBC (or EC) concentrations in surface snow at Austre Brøggerbreen using at Zeppelin Observatory and W values of 10 to 300. Figure 11 shows the range of in winter/spring Svalbard snow measured in this study, compared with data from other 430 circum-Arctic or subarctic sites sampled between 2002 and 2018 (see figure caption for data sources). All EC data in this comparison were obtained by thermo-optical measurements on snow filters, but variations in the thermal protocol used between studies may account for some of the inter-site differences ( Table S4). The measurements made by Aamas et al. (2011) near Longyearbyen were excluded because of local pollution of the snow cover by coal dust (Khan et al., 2017). The pattern of EC distribution in winter/spring snow across the circum-Arctic shows the lowest ̃ in central Greenland (<1 ng g -1 ), 435 followed by the southwestern Yukon (3-5 ng g -1 ), northern and central Alaska (3-5 ng g -1 ), Svalbard (2-7 ng g -1 outside Ny-Ålesund), northern Scandinavia (~15-30 ng g -1 depending on site), and reaching maximum values of ̃ in eastern and central Siberia (35-66 ng g -1 ). This geographic pattern is in broad agreement with that observed in pan-Arctic surveys conducted in 2007-09 (Doherty et al., 2010) and between 2012 and 2016 (Mori et al., 2019) using other analytical methods.

Placing results in geographic perspective
The EC/OC (inset, Fig. 11) in the Svalbard snowpack outside Ny-Ålesund averages 0.07±0.08 (1 s.d.), which is similar, within errors, to that reported in snowpacks in central Greenland (0.08±0.05;Hagler et al., 2007), northern Alaska (0.09±0.03;Dou et al., 2017), northeastern Russia and Siberia (0.07±0.03;Evangeliou et al., 2018), and northern Finland (0.05±0.06 ;Meinander et al., 2013). The only site where a significantly lower mean EC/OC in snow has been measured is on the Eclipse Icefield in the southern Yukon (0.01±0.01; Table S5). This may be explained by the remoteness of this high elevation site (3020 m a.s.l.), which is mostly exposed in winter to relatively clean air advected from the Gulf Alaska. 445

Summary and conclusions
We have presented a large dataset of observations of atmospheric EC and OC deposited in snow on the archipelago of Svalbard, made between 2007 and 2018. The spatial snow survey conducted across 22 glacier sites in April 2016 was one of the most extensive and detailed carried out on Svalbard, and allows direct comparisons with the surveys by Forsström et al. (2009Forsström et al. ( , 2013, made nearly 10 years earlier. Across all glacier sites, in the snowpack ranged from <1.0 to 45.2 ng g -1 (median 450 2.0 ng g -1 ), while ranged from 11.9 to 3448.9 ng g -1 (median 48.6 ng g -1 ). The calculated EC were between 0.1 and 16.2 mg m -2 (median 0.8 mg m -2 ), while were between 1.7 and 320.1 mg m -2 (median 20.5 mg m -2 ). The ̃ and in 2016 were comparable or lower than those found in spring 2007-09 glacier snow, but no clear spatial gradients could be identified across the archipelago. Both and were found to increase exponentially with elevation and hSWE.
Using these relationships, we estimated the area-averaged, monthly mean EC and OC accumulation rates over the whole of 455 Svalbard to be ~0.2 mg EC m -2 mo -1 and ~8.9 mg OC m -2 mo -1 for the winter 2015-16 (September to April). The relationship between and hSWE also point to dry EC deposition in snow being minor compared to wet deposition. The accumulation of EC and OC in the snowpack was inferred to be equally distributed over the winter 2015-16 at most sites. Relatively high and were found in the accumulation zones of glaciers on southern Spitsbergen, which we attribute to enhanced wet aerosol deposition when large Atlantic cyclonic storms made landfall in this area during the autumn and mid-winter. 460  Table 2. Descriptive statistics for , and EC/OC snow in samples analyzed in this study. n* is the number of values > 1 ng g -1 . Two outliers with >1700 ng g -1 were excluded from calculations.